Add files using upload-large-folder tool
Browse files- README.md +57 -0
- checkpoint-142/chat_template.jinja +6 -0
- checkpoint-142/config.json +40 -0
- checkpoint-142/generation_config.json +9 -0
- checkpoint-142/model.safetensors +3 -0
- checkpoint-142/rng_state.pth +3 -0
- checkpoint-142/scheduler.pt +3 -0
- checkpoint-142/tokenizer.json +0 -0
- checkpoint-142/tokenizer_config.json +17 -0
- checkpoint-142/trainer_state.json +1465 -0
- checkpoint-142/training_args.bin +3 -0
- checkpoint-284/chat_template.jinja +6 -0
- checkpoint-284/config.json +40 -0
- checkpoint-284/generation_config.json +9 -0
- checkpoint-284/model.safetensors +3 -0
- checkpoint-284/rng_state.pth +3 -0
- checkpoint-284/scheduler.pt +3 -0
- checkpoint-284/tokenizer.json +0 -0
- checkpoint-284/tokenizer_config.json +17 -0
- checkpoint-284/trainer_state.json +2896 -0
- checkpoint-284/training_args.bin +3 -0
- checkpoint-426/chat_template.jinja +6 -0
- checkpoint-426/config.json +40 -0
- checkpoint-426/generation_config.json +9 -0
- checkpoint-426/model.safetensors +3 -0
- checkpoint-426/rng_state.pth +3 -0
- checkpoint-426/scheduler.pt +3 -0
- checkpoint-426/tokenizer.json +0 -0
- checkpoint-426/tokenizer_config.json +17 -0
- checkpoint-426/trainer_state.json +0 -0
- checkpoint-426/training_args.bin +3 -0
- checkpoint-568/chat_template.jinja +6 -0
- checkpoint-568/config.json +40 -0
- checkpoint-568/generation_config.json +9 -0
- checkpoint-568/model.safetensors +3 -0
- checkpoint-568/rng_state.pth +3 -0
- checkpoint-568/scheduler.pt +3 -0
- checkpoint-568/tokenizer.json +0 -0
- checkpoint-568/tokenizer_config.json +17 -0
- checkpoint-568/trainer_state.json +0 -0
- checkpoint-568/training_args.bin +3 -0
- checkpoint-710/chat_template.jinja +6 -0
- checkpoint-710/config.json +40 -0
- checkpoint-710/generation_config.json +9 -0
- checkpoint-710/rng_state.pth +3 -0
- checkpoint-710/scheduler.pt +3 -0
- checkpoint-710/tokenizer.json +0 -0
- checkpoint-710/tokenizer_config.json +17 -0
- checkpoint-710/trainer_state.json +0 -0
- checkpoint-710/training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
model_name: checkpoint_models
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
licence: license
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# Model Card for checkpoint_models
|
| 12 |
+
|
| 13 |
+
This model is a fine-tuned version of [None](https://huggingface.co/None).
|
| 14 |
+
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 15 |
+
|
| 16 |
+
## Quick start
|
| 17 |
+
|
| 18 |
+
```python
|
| 19 |
+
from transformers import pipeline
|
| 20 |
+
|
| 21 |
+
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
| 22 |
+
generator = pipeline("text-generation", model="None", device="cuda")
|
| 23 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 24 |
+
print(output["generated_text"])
|
| 25 |
+
```
|
| 26 |
+
|
| 27 |
+
## Training procedure
|
| 28 |
+
|
| 29 |
+
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
This model was trained with SFT.
|
| 34 |
+
|
| 35 |
+
### Framework versions
|
| 36 |
+
|
| 37 |
+
- TRL: 0.29.0
|
| 38 |
+
- Transformers: 5.2.0
|
| 39 |
+
- Pytorch: 2.10.0+cu130
|
| 40 |
+
- Datasets: 4.6.0
|
| 41 |
+
- Tokenizers: 0.22.2
|
| 42 |
+
|
| 43 |
+
## Citations
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
|
| 47 |
+
Cite TRL as:
|
| 48 |
+
|
| 49 |
+
```bibtex
|
| 50 |
+
@software{vonwerra2020trl,
|
| 51 |
+
title = {{TRL: Transformers Reinforcement Learning}},
|
| 52 |
+
author = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
|
| 53 |
+
license = {Apache-2.0},
|
| 54 |
+
url = {https://github.com/huggingface/trl},
|
| 55 |
+
year = {2020}
|
| 56 |
+
}
|
| 57 |
+
```
|
checkpoint-142/chat_template.jinja
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system
|
| 2 |
+
You are a helpful AI assistant named SmolLM, trained by Hugging Face<|im_end|>
|
| 3 |
+
' }}{% endif %}{{'<|im_start|>' + message['role'] + '
|
| 4 |
+
' + message['content'] + '<|im_end|>' + '
|
| 5 |
+
'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
|
| 6 |
+
' }}{% endif %}
|
checkpoint-142/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 576,
|
| 13 |
+
"initializer_range": 0.041666666666666664,
|
| 14 |
+
"intermediate_size": 1536,
|
| 15 |
+
"is_llama_config": true,
|
| 16 |
+
"max_position_embeddings": 8192,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"num_attention_heads": 9,
|
| 20 |
+
"num_hidden_layers": 30,
|
| 21 |
+
"num_key_value_heads": 3,
|
| 22 |
+
"pad_token_id": 2,
|
| 23 |
+
"pretraining_tp": 1,
|
| 24 |
+
"rms_norm_eps": 1e-05,
|
| 25 |
+
"rope_interleaved": false,
|
| 26 |
+
"rope_parameters": {
|
| 27 |
+
"rope_theta": 100000,
|
| 28 |
+
"rope_type": "default"
|
| 29 |
+
},
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"transformers.js_config": {
|
| 32 |
+
"kv_cache_dtype": {
|
| 33 |
+
"fp16": "float16",
|
| 34 |
+
"q4f16": "float16"
|
| 35 |
+
}
|
| 36 |
+
},
|
| 37 |
+
"transformers_version": "5.2.0",
|
| 38 |
+
"use_cache": false,
|
| 39 |
+
"vocab_size": 49152
|
| 40 |
+
}
|
checkpoint-142/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
2
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 2,
|
| 8 |
+
"transformers_version": "5.2.0"
|
| 9 |
+
}
|
checkpoint-142/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3358ae11cc733b9f79a40841204ded02bf2baf76ca8d2ce893bd67916004fe63
|
| 3 |
+
size 269060552
|
checkpoint-142/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:098b29492211804ab324a36f37466821d948280bb74fce4ba895c03f13ecd878
|
| 3 |
+
size 14645
|
checkpoint-142/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2732da9c9500b70ace14309058979c65a5e31cf20be217688967b97f7bd768e2
|
| 3 |
+
size 1465
|
checkpoint-142/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-142/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|im_start|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"extra_special_tokens": [
|
| 8 |
+
"<|im_start|>",
|
| 9 |
+
"<|im_end|>"
|
| 10 |
+
],
|
| 11 |
+
"is_local": true,
|
| 12 |
+
"model_max_length": 8192,
|
| 13 |
+
"pad_token": "<|im_end|>",
|
| 14 |
+
"tokenizer_class": "TokenizersBackend",
|
| 15 |
+
"unk_token": "<|endoftext|>",
|
| 16 |
+
"vocab_size": 49152
|
| 17 |
+
}
|
checkpoint-142/trainer_state.json
ADDED
|
@@ -0,0 +1,1465 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 142,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 1.78125,
|
| 14 |
+
"epoch": 0.007042253521126761,
|
| 15 |
+
"grad_norm": 5.375,
|
| 16 |
+
"learning_rate": 5e-05,
|
| 17 |
+
"loss": 2.342320680618286,
|
| 18 |
+
"mean_token_accuracy": 0.5398398637771606,
|
| 19 |
+
"num_tokens": 2631.0,
|
| 20 |
+
"step": 1
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.7265625,
|
| 24 |
+
"epoch": 0.014084507042253521,
|
| 25 |
+
"grad_norm": 4.28125,
|
| 26 |
+
"learning_rate": 5e-05,
|
| 27 |
+
"loss": 2.184455394744873,
|
| 28 |
+
"mean_token_accuracy": 0.5707964897155762,
|
| 29 |
+
"num_tokens": 5577.0,
|
| 30 |
+
"step": 2
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.5859375,
|
| 34 |
+
"epoch": 0.02112676056338028,
|
| 35 |
+
"grad_norm": 3.796875,
|
| 36 |
+
"learning_rate": 5e-05,
|
| 37 |
+
"loss": 1.9440301656723022,
|
| 38 |
+
"mean_token_accuracy": 0.6223160624504089,
|
| 39 |
+
"num_tokens": 8426.0,
|
| 40 |
+
"step": 3
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.75,
|
| 44 |
+
"epoch": 0.028169014084507043,
|
| 45 |
+
"grad_norm": 3.234375,
|
| 46 |
+
"learning_rate": 5e-05,
|
| 47 |
+
"loss": 2.034520149230957,
|
| 48 |
+
"mean_token_accuracy": 0.5993841886520386,
|
| 49 |
+
"num_tokens": 11357.0,
|
| 50 |
+
"step": 4
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.8984375,
|
| 54 |
+
"epoch": 0.035211267605633804,
|
| 55 |
+
"grad_norm": 3.3125,
|
| 56 |
+
"learning_rate": 5e-05,
|
| 57 |
+
"loss": 2.151127338409424,
|
| 58 |
+
"mean_token_accuracy": 0.563842236995697,
|
| 59 |
+
"num_tokens": 13926.0,
|
| 60 |
+
"step": 5
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 1.890625,
|
| 64 |
+
"epoch": 0.04225352112676056,
|
| 65 |
+
"grad_norm": 2.546875,
|
| 66 |
+
"learning_rate": 5e-05,
|
| 67 |
+
"loss": 2.066777229309082,
|
| 68 |
+
"mean_token_accuracy": 0.5754261612892151,
|
| 69 |
+
"num_tokens": 17043.0,
|
| 70 |
+
"step": 6
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 2.03125,
|
| 74 |
+
"epoch": 0.04929577464788732,
|
| 75 |
+
"grad_norm": 2.9375,
|
| 76 |
+
"learning_rate": 5e-05,
|
| 77 |
+
"loss": 2.2290213108062744,
|
| 78 |
+
"mean_token_accuracy": 0.5708707571029663,
|
| 79 |
+
"num_tokens": 19612.0,
|
| 80 |
+
"step": 7
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 1.8046875,
|
| 84 |
+
"epoch": 0.056338028169014086,
|
| 85 |
+
"grad_norm": 2.59375,
|
| 86 |
+
"learning_rate": 5e-05,
|
| 87 |
+
"loss": 1.9348505735397339,
|
| 88 |
+
"mean_token_accuracy": 0.6066350936889648,
|
| 89 |
+
"num_tokens": 22363.0,
|
| 90 |
+
"step": 8
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 1.7734375,
|
| 94 |
+
"epoch": 0.06338028169014084,
|
| 95 |
+
"grad_norm": 2.4375,
|
| 96 |
+
"learning_rate": 5e-05,
|
| 97 |
+
"loss": 1.8786349296569824,
|
| 98 |
+
"mean_token_accuracy": 0.628695011138916,
|
| 99 |
+
"num_tokens": 25145.0,
|
| 100 |
+
"step": 9
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 1.8515625,
|
| 104 |
+
"epoch": 0.07042253521126761,
|
| 105 |
+
"grad_norm": 2.296875,
|
| 106 |
+
"learning_rate": 5e-05,
|
| 107 |
+
"loss": 1.9190694093704224,
|
| 108 |
+
"mean_token_accuracy": 0.6125850081443787,
|
| 109 |
+
"num_tokens": 28093.0,
|
| 110 |
+
"step": 10
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"entropy": 1.8515625,
|
| 114 |
+
"epoch": 0.07746478873239436,
|
| 115 |
+
"grad_norm": 2.171875,
|
| 116 |
+
"learning_rate": 5e-05,
|
| 117 |
+
"loss": 1.891284704208374,
|
| 118 |
+
"mean_token_accuracy": 0.6322028040885925,
|
| 119 |
+
"num_tokens": 31138.0,
|
| 120 |
+
"step": 11
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"entropy": 1.953125,
|
| 124 |
+
"epoch": 0.08450704225352113,
|
| 125 |
+
"grad_norm": 2.421875,
|
| 126 |
+
"learning_rate": 5e-05,
|
| 127 |
+
"loss": 1.948248267173767,
|
| 128 |
+
"mean_token_accuracy": 0.6148879528045654,
|
| 129 |
+
"num_tokens": 33779.0,
|
| 130 |
+
"step": 12
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"entropy": 1.9609375,
|
| 134 |
+
"epoch": 0.09154929577464789,
|
| 135 |
+
"grad_norm": 2.671875,
|
| 136 |
+
"learning_rate": 5e-05,
|
| 137 |
+
"loss": 1.968241810798645,
|
| 138 |
+
"mean_token_accuracy": 0.6172575950622559,
|
| 139 |
+
"num_tokens": 36128.0,
|
| 140 |
+
"step": 13
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"entropy": 1.8828125,
|
| 144 |
+
"epoch": 0.09859154929577464,
|
| 145 |
+
"grad_norm": 2.015625,
|
| 146 |
+
"learning_rate": 5e-05,
|
| 147 |
+
"loss": 1.902205228805542,
|
| 148 |
+
"mean_token_accuracy": 0.6156116724014282,
|
| 149 |
+
"num_tokens": 39185.0,
|
| 150 |
+
"step": 14
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"entropy": 1.703125,
|
| 154 |
+
"epoch": 0.1056338028169014,
|
| 155 |
+
"grad_norm": 2.328125,
|
| 156 |
+
"learning_rate": 5e-05,
|
| 157 |
+
"loss": 1.6213911771774292,
|
| 158 |
+
"mean_token_accuracy": 0.6750187277793884,
|
| 159 |
+
"num_tokens": 41867.0,
|
| 160 |
+
"step": 15
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"entropy": 2.015625,
|
| 164 |
+
"epoch": 0.11267605633802817,
|
| 165 |
+
"grad_norm": 2.234375,
|
| 166 |
+
"learning_rate": 5e-05,
|
| 167 |
+
"loss": 1.9176678657531738,
|
| 168 |
+
"mean_token_accuracy": 0.6337976455688477,
|
| 169 |
+
"num_tokens": 44603.0,
|
| 170 |
+
"step": 16
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"entropy": 1.921875,
|
| 174 |
+
"epoch": 0.11971830985915492,
|
| 175 |
+
"grad_norm": 1.921875,
|
| 176 |
+
"learning_rate": 5e-05,
|
| 177 |
+
"loss": 1.8336955308914185,
|
| 178 |
+
"mean_token_accuracy": 0.6507288813591003,
|
| 179 |
+
"num_tokens": 48041.0,
|
| 180 |
+
"step": 17
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"entropy": 1.828125,
|
| 184 |
+
"epoch": 0.1267605633802817,
|
| 185 |
+
"grad_norm": 2.046875,
|
| 186 |
+
"learning_rate": 5e-05,
|
| 187 |
+
"loss": 1.7068480253219604,
|
| 188 |
+
"mean_token_accuracy": 0.6735284924507141,
|
| 189 |
+
"num_tokens": 51158.0,
|
| 190 |
+
"step": 18
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"entropy": 1.7421875,
|
| 194 |
+
"epoch": 0.13380281690140844,
|
| 195 |
+
"grad_norm": 1.8984375,
|
| 196 |
+
"learning_rate": 5e-05,
|
| 197 |
+
"loss": 1.6770858764648438,
|
| 198 |
+
"mean_token_accuracy": 0.6662946343421936,
|
| 199 |
+
"num_tokens": 54750.0,
|
| 200 |
+
"step": 19
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"entropy": 2.109375,
|
| 204 |
+
"epoch": 0.14084507042253522,
|
| 205 |
+
"grad_norm": 2.828125,
|
| 206 |
+
"learning_rate": 5e-05,
|
| 207 |
+
"loss": 1.9946776628494263,
|
| 208 |
+
"mean_token_accuracy": 0.6195476651191711,
|
| 209 |
+
"num_tokens": 57234.0,
|
| 210 |
+
"step": 20
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"entropy": 1.6328125,
|
| 214 |
+
"epoch": 0.14788732394366197,
|
| 215 |
+
"grad_norm": 1.90625,
|
| 216 |
+
"learning_rate": 5e-05,
|
| 217 |
+
"loss": 1.5617338418960571,
|
| 218 |
+
"mean_token_accuracy": 0.6951566934585571,
|
| 219 |
+
"num_tokens": 60752.0,
|
| 220 |
+
"step": 21
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"entropy": 1.78125,
|
| 224 |
+
"epoch": 0.15492957746478872,
|
| 225 |
+
"grad_norm": 2.515625,
|
| 226 |
+
"learning_rate": 5e-05,
|
| 227 |
+
"loss": 1.7242701053619385,
|
| 228 |
+
"mean_token_accuracy": 0.6592110395431519,
|
| 229 |
+
"num_tokens": 63219.0,
|
| 230 |
+
"step": 22
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"entropy": 1.6796875,
|
| 234 |
+
"epoch": 0.1619718309859155,
|
| 235 |
+
"grad_norm": 1.9375,
|
| 236 |
+
"learning_rate": 5e-05,
|
| 237 |
+
"loss": 1.6089009046554565,
|
| 238 |
+
"mean_token_accuracy": 0.6697189807891846,
|
| 239 |
+
"num_tokens": 66394.0,
|
| 240 |
+
"step": 23
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"entropy": 1.9296875,
|
| 244 |
+
"epoch": 0.16901408450704225,
|
| 245 |
+
"grad_norm": 2.125,
|
| 246 |
+
"learning_rate": 5e-05,
|
| 247 |
+
"loss": 1.8391156196594238,
|
| 248 |
+
"mean_token_accuracy": 0.6323254704475403,
|
| 249 |
+
"num_tokens": 69081.0,
|
| 250 |
+
"step": 24
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"entropy": 1.7265625,
|
| 254 |
+
"epoch": 0.176056338028169,
|
| 255 |
+
"grad_norm": 2.15625,
|
| 256 |
+
"learning_rate": 5e-05,
|
| 257 |
+
"loss": 1.6465730667114258,
|
| 258 |
+
"mean_token_accuracy": 0.654959499835968,
|
| 259 |
+
"num_tokens": 71680.0,
|
| 260 |
+
"step": 25
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"entropy": 1.9609375,
|
| 264 |
+
"epoch": 0.18309859154929578,
|
| 265 |
+
"grad_norm": 2.03125,
|
| 266 |
+
"learning_rate": 5e-05,
|
| 267 |
+
"loss": 1.873896598815918,
|
| 268 |
+
"mean_token_accuracy": 0.6338028311729431,
|
| 269 |
+
"num_tokens": 74670.0,
|
| 270 |
+
"step": 26
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"entropy": 1.6875,
|
| 274 |
+
"epoch": 0.19014084507042253,
|
| 275 |
+
"grad_norm": 1.84375,
|
| 276 |
+
"learning_rate": 5e-05,
|
| 277 |
+
"loss": 1.629098653793335,
|
| 278 |
+
"mean_token_accuracy": 0.6646864414215088,
|
| 279 |
+
"num_tokens": 77708.0,
|
| 280 |
+
"step": 27
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"entropy": 1.609375,
|
| 284 |
+
"epoch": 0.19718309859154928,
|
| 285 |
+
"grad_norm": 1.8671875,
|
| 286 |
+
"learning_rate": 5e-05,
|
| 287 |
+
"loss": 1.5465565919876099,
|
| 288 |
+
"mean_token_accuracy": 0.6922129988670349,
|
| 289 |
+
"num_tokens": 80965.0,
|
| 290 |
+
"step": 28
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"entropy": 1.59375,
|
| 294 |
+
"epoch": 0.20422535211267606,
|
| 295 |
+
"grad_norm": 2.140625,
|
| 296 |
+
"learning_rate": 5e-05,
|
| 297 |
+
"loss": 1.5382331609725952,
|
| 298 |
+
"mean_token_accuracy": 0.6933178901672363,
|
| 299 |
+
"num_tokens": 83562.0,
|
| 300 |
+
"step": 29
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"entropy": 1.5390625,
|
| 304 |
+
"epoch": 0.2112676056338028,
|
| 305 |
+
"grad_norm": 1.9921875,
|
| 306 |
+
"learning_rate": 5e-05,
|
| 307 |
+
"loss": 1.4340285062789917,
|
| 308 |
+
"mean_token_accuracy": 0.6944348216056824,
|
| 309 |
+
"num_tokens": 86463.0,
|
| 310 |
+
"step": 30
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"entropy": 1.5,
|
| 314 |
+
"epoch": 0.21830985915492956,
|
| 315 |
+
"grad_norm": 1.953125,
|
| 316 |
+
"learning_rate": 5e-05,
|
| 317 |
+
"loss": 1.41927969455719,
|
| 318 |
+
"mean_token_accuracy": 0.7242607474327087,
|
| 319 |
+
"num_tokens": 89278.0,
|
| 320 |
+
"step": 31
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"entropy": 1.484375,
|
| 324 |
+
"epoch": 0.22535211267605634,
|
| 325 |
+
"grad_norm": 2.328125,
|
| 326 |
+
"learning_rate": 5e-05,
|
| 327 |
+
"loss": 1.395577311515808,
|
| 328 |
+
"mean_token_accuracy": 0.7233359813690186,
|
| 329 |
+
"num_tokens": 91780.0,
|
| 330 |
+
"step": 32
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"entropy": 1.5234375,
|
| 334 |
+
"epoch": 0.2323943661971831,
|
| 335 |
+
"grad_norm": 1.8671875,
|
| 336 |
+
"learning_rate": 5e-05,
|
| 337 |
+
"loss": 1.4273605346679688,
|
| 338 |
+
"mean_token_accuracy": 0.7213010191917419,
|
| 339 |
+
"num_tokens": 94924.0,
|
| 340 |
+
"step": 33
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"entropy": 1.640625,
|
| 344 |
+
"epoch": 0.23943661971830985,
|
| 345 |
+
"grad_norm": 1.9375,
|
| 346 |
+
"learning_rate": 5e-05,
|
| 347 |
+
"loss": 1.549876093864441,
|
| 348 |
+
"mean_token_accuracy": 0.6991525292396545,
|
| 349 |
+
"num_tokens": 98236.0,
|
| 350 |
+
"step": 34
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"entropy": 1.3984375,
|
| 354 |
+
"epoch": 0.24647887323943662,
|
| 355 |
+
"grad_norm": 2.5625,
|
| 356 |
+
"learning_rate": 5e-05,
|
| 357 |
+
"loss": 1.2571226358413696,
|
| 358 |
+
"mean_token_accuracy": 0.7764654159545898,
|
| 359 |
+
"num_tokens": 100530.0,
|
| 360 |
+
"step": 35
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"entropy": 1.484375,
|
| 364 |
+
"epoch": 0.2535211267605634,
|
| 365 |
+
"grad_norm": 1.9921875,
|
| 366 |
+
"learning_rate": 5e-05,
|
| 367 |
+
"loss": 1.430403709411621,
|
| 368 |
+
"mean_token_accuracy": 0.7238532304763794,
|
| 369 |
+
"num_tokens": 103808.0,
|
| 370 |
+
"step": 36
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"entropy": 1.4375,
|
| 374 |
+
"epoch": 0.2605633802816901,
|
| 375 |
+
"grad_norm": 1.9296875,
|
| 376 |
+
"learning_rate": 5e-05,
|
| 377 |
+
"loss": 1.370211124420166,
|
| 378 |
+
"mean_token_accuracy": 0.7603058218955994,
|
| 379 |
+
"num_tokens": 106824.0,
|
| 380 |
+
"step": 37
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"entropy": 1.3046875,
|
| 384 |
+
"epoch": 0.2676056338028169,
|
| 385 |
+
"grad_norm": 1.9921875,
|
| 386 |
+
"learning_rate": 5e-05,
|
| 387 |
+
"loss": 1.160237431526184,
|
| 388 |
+
"mean_token_accuracy": 0.7812739610671997,
|
| 389 |
+
"num_tokens": 109438.0,
|
| 390 |
+
"step": 38
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"entropy": 1.40625,
|
| 394 |
+
"epoch": 0.2746478873239437,
|
| 395 |
+
"grad_norm": 2.265625,
|
| 396 |
+
"learning_rate": 5e-05,
|
| 397 |
+
"loss": 1.3372321128845215,
|
| 398 |
+
"mean_token_accuracy": 0.7477554678916931,
|
| 399 |
+
"num_tokens": 111785.0,
|
| 400 |
+
"step": 39
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"entropy": 1.3671875,
|
| 404 |
+
"epoch": 0.28169014084507044,
|
| 405 |
+
"grad_norm": 1.84375,
|
| 406 |
+
"learning_rate": 5e-05,
|
| 407 |
+
"loss": 1.3203842639923096,
|
| 408 |
+
"mean_token_accuracy": 0.7538297176361084,
|
| 409 |
+
"num_tokens": 114600.0,
|
| 410 |
+
"step": 40
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"entropy": 1.21875,
|
| 414 |
+
"epoch": 0.2887323943661972,
|
| 415 |
+
"grad_norm": 2.203125,
|
| 416 |
+
"learning_rate": 5e-05,
|
| 417 |
+
"loss": 1.1055949926376343,
|
| 418 |
+
"mean_token_accuracy": 0.7860545516014099,
|
| 419 |
+
"num_tokens": 116917.0,
|
| 420 |
+
"step": 41
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"entropy": 1.484375,
|
| 424 |
+
"epoch": 0.29577464788732394,
|
| 425 |
+
"grad_norm": 1.546875,
|
| 426 |
+
"learning_rate": 5e-05,
|
| 427 |
+
"loss": 1.403846025466919,
|
| 428 |
+
"mean_token_accuracy": 0.7451103925704956,
|
| 429 |
+
"num_tokens": 120095.0,
|
| 430 |
+
"step": 42
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"entropy": 1.328125,
|
| 434 |
+
"epoch": 0.3028169014084507,
|
| 435 |
+
"grad_norm": 1.65625,
|
| 436 |
+
"learning_rate": 5e-05,
|
| 437 |
+
"loss": 1.3071074485778809,
|
| 438 |
+
"mean_token_accuracy": 0.7589316964149475,
|
| 439 |
+
"num_tokens": 122986.0,
|
| 440 |
+
"step": 43
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"entropy": 1.3203125,
|
| 444 |
+
"epoch": 0.30985915492957744,
|
| 445 |
+
"grad_norm": 1.8359375,
|
| 446 |
+
"learning_rate": 5e-05,
|
| 447 |
+
"loss": 1.2948389053344727,
|
| 448 |
+
"mean_token_accuracy": 0.7675830125808716,
|
| 449 |
+
"num_tokens": 125795.0,
|
| 450 |
+
"step": 44
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"entropy": 1.3125,
|
| 454 |
+
"epoch": 0.31690140845070425,
|
| 455 |
+
"grad_norm": 1.875,
|
| 456 |
+
"learning_rate": 5e-05,
|
| 457 |
+
"loss": 1.2394049167633057,
|
| 458 |
+
"mean_token_accuracy": 0.7697368264198303,
|
| 459 |
+
"num_tokens": 128235.0,
|
| 460 |
+
"step": 45
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"entropy": 1.3203125,
|
| 464 |
+
"epoch": 0.323943661971831,
|
| 465 |
+
"grad_norm": 1.6796875,
|
| 466 |
+
"learning_rate": 5e-05,
|
| 467 |
+
"loss": 1.3117918968200684,
|
| 468 |
+
"mean_token_accuracy": 0.7587719559669495,
|
| 469 |
+
"num_tokens": 131663.0,
|
| 470 |
+
"step": 46
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"entropy": 1.3515625,
|
| 474 |
+
"epoch": 0.33098591549295775,
|
| 475 |
+
"grad_norm": 1.453125,
|
| 476 |
+
"learning_rate": 5e-05,
|
| 477 |
+
"loss": 1.3100031614303589,
|
| 478 |
+
"mean_token_accuracy": 0.7717799544334412,
|
| 479 |
+
"num_tokens": 134598.0,
|
| 480 |
+
"step": 47
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"entropy": 1.3203125,
|
| 484 |
+
"epoch": 0.3380281690140845,
|
| 485 |
+
"grad_norm": 1.6953125,
|
| 486 |
+
"learning_rate": 5e-05,
|
| 487 |
+
"loss": 1.28191077709198,
|
| 488 |
+
"mean_token_accuracy": 0.7734912037849426,
|
| 489 |
+
"num_tokens": 137224.0,
|
| 490 |
+
"step": 48
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"entropy": 1.3828125,
|
| 494 |
+
"epoch": 0.34507042253521125,
|
| 495 |
+
"grad_norm": 1.515625,
|
| 496 |
+
"learning_rate": 5e-05,
|
| 497 |
+
"loss": 1.3600674867630005,
|
| 498 |
+
"mean_token_accuracy": 0.7438323497772217,
|
| 499 |
+
"num_tokens": 140191.0,
|
| 500 |
+
"step": 49
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"entropy": 1.21875,
|
| 504 |
+
"epoch": 0.352112676056338,
|
| 505 |
+
"grad_norm": 1.46875,
|
| 506 |
+
"learning_rate": 5e-05,
|
| 507 |
+
"loss": 1.1239631175994873,
|
| 508 |
+
"mean_token_accuracy": 0.788614809513092,
|
| 509 |
+
"num_tokens": 142834.0,
|
| 510 |
+
"step": 50
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"entropy": 1.4765625,
|
| 514 |
+
"epoch": 0.3591549295774648,
|
| 515 |
+
"grad_norm": 1.46875,
|
| 516 |
+
"learning_rate": 5e-05,
|
| 517 |
+
"loss": 1.3836950063705444,
|
| 518 |
+
"mean_token_accuracy": 0.7538405060768127,
|
| 519 |
+
"num_tokens": 145576.0,
|
| 520 |
+
"step": 51
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"entropy": 1.1328125,
|
| 524 |
+
"epoch": 0.36619718309859156,
|
| 525 |
+
"grad_norm": 1.59375,
|
| 526 |
+
"learning_rate": 5e-05,
|
| 527 |
+
"loss": 1.1020089387893677,
|
| 528 |
+
"mean_token_accuracy": 0.8136998414993286,
|
| 529 |
+
"num_tokens": 147876.0,
|
| 530 |
+
"step": 52
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"entropy": 1.3671875,
|
| 534 |
+
"epoch": 0.3732394366197183,
|
| 535 |
+
"grad_norm": 1.15625,
|
| 536 |
+
"learning_rate": 5e-05,
|
| 537 |
+
"loss": 1.3127013444900513,
|
| 538 |
+
"mean_token_accuracy": 0.7502166032791138,
|
| 539 |
+
"num_tokens": 151347.0,
|
| 540 |
+
"step": 53
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"entropy": 1.0390625,
|
| 544 |
+
"epoch": 0.38028169014084506,
|
| 545 |
+
"grad_norm": 1.25,
|
| 546 |
+
"learning_rate": 5e-05,
|
| 547 |
+
"loss": 1.0258498191833496,
|
| 548 |
+
"mean_token_accuracy": 0.8231140971183777,
|
| 549 |
+
"num_tokens": 154046.0,
|
| 550 |
+
"step": 54
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"entropy": 1.296875,
|
| 554 |
+
"epoch": 0.3873239436619718,
|
| 555 |
+
"grad_norm": 1.515625,
|
| 556 |
+
"learning_rate": 5e-05,
|
| 557 |
+
"loss": 1.26340913772583,
|
| 558 |
+
"mean_token_accuracy": 0.7880998253822327,
|
| 559 |
+
"num_tokens": 156659.0,
|
| 560 |
+
"step": 55
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"entropy": 1.3828125,
|
| 564 |
+
"epoch": 0.39436619718309857,
|
| 565 |
+
"grad_norm": 1.1484375,
|
| 566 |
+
"learning_rate": 5e-05,
|
| 567 |
+
"loss": 1.3492928743362427,
|
| 568 |
+
"mean_token_accuracy": 0.7417179942131042,
|
| 569 |
+
"num_tokens": 160078.0,
|
| 570 |
+
"step": 56
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"entropy": 1.046875,
|
| 574 |
+
"epoch": 0.4014084507042254,
|
| 575 |
+
"grad_norm": 1.1171875,
|
| 576 |
+
"learning_rate": 5e-05,
|
| 577 |
+
"loss": 1.0268410444259644,
|
| 578 |
+
"mean_token_accuracy": 0.8108739256858826,
|
| 579 |
+
"num_tokens": 163084.0,
|
| 580 |
+
"step": 57
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"entropy": 1.2734375,
|
| 584 |
+
"epoch": 0.4084507042253521,
|
| 585 |
+
"grad_norm": 1.296875,
|
| 586 |
+
"learning_rate": 5e-05,
|
| 587 |
+
"loss": 1.218086838722229,
|
| 588 |
+
"mean_token_accuracy": 0.7687548398971558,
|
| 589 |
+
"num_tokens": 165678.0,
|
| 590 |
+
"step": 58
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"entropy": 1.265625,
|
| 594 |
+
"epoch": 0.4154929577464789,
|
| 595 |
+
"grad_norm": 1.078125,
|
| 596 |
+
"learning_rate": 5e-05,
|
| 597 |
+
"loss": 1.2634363174438477,
|
| 598 |
+
"mean_token_accuracy": 0.7484593987464905,
|
| 599 |
+
"num_tokens": 169256.0,
|
| 600 |
+
"step": 59
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"entropy": 1.046875,
|
| 604 |
+
"epoch": 0.4225352112676056,
|
| 605 |
+
"grad_norm": 1.1796875,
|
| 606 |
+
"learning_rate": 5e-05,
|
| 607 |
+
"loss": 0.9637977480888367,
|
| 608 |
+
"mean_token_accuracy": 0.8200221061706543,
|
| 609 |
+
"num_tokens": 171981.0,
|
| 610 |
+
"step": 60
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"entropy": 1.1015625,
|
| 614 |
+
"epoch": 0.4295774647887324,
|
| 615 |
+
"grad_norm": 1.09375,
|
| 616 |
+
"learning_rate": 5e-05,
|
| 617 |
+
"loss": 1.0381879806518555,
|
| 618 |
+
"mean_token_accuracy": 0.8146411776542664,
|
| 619 |
+
"num_tokens": 174762.0,
|
| 620 |
+
"step": 61
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"entropy": 1.4609375,
|
| 624 |
+
"epoch": 0.43661971830985913,
|
| 625 |
+
"grad_norm": 1.109375,
|
| 626 |
+
"learning_rate": 5e-05,
|
| 627 |
+
"loss": 1.442441701889038,
|
| 628 |
+
"mean_token_accuracy": 0.7236841917037964,
|
| 629 |
+
"num_tokens": 177962.0,
|
| 630 |
+
"step": 62
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"entropy": 0.9453125,
|
| 634 |
+
"epoch": 0.44366197183098594,
|
| 635 |
+
"grad_norm": 1.109375,
|
| 636 |
+
"learning_rate": 5e-05,
|
| 637 |
+
"loss": 0.9214452505111694,
|
| 638 |
+
"mean_token_accuracy": 0.8193620443344116,
|
| 639 |
+
"num_tokens": 180666.0,
|
| 640 |
+
"step": 63
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"entropy": 1.0859375,
|
| 644 |
+
"epoch": 0.4507042253521127,
|
| 645 |
+
"grad_norm": 1.1484375,
|
| 646 |
+
"learning_rate": 5e-05,
|
| 647 |
+
"loss": 1.1097257137298584,
|
| 648 |
+
"mean_token_accuracy": 0.8031823039054871,
|
| 649 |
+
"num_tokens": 183565.0,
|
| 650 |
+
"step": 64
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"entropy": 1.1640625,
|
| 654 |
+
"epoch": 0.45774647887323944,
|
| 655 |
+
"grad_norm": 1.140625,
|
| 656 |
+
"learning_rate": 5e-05,
|
| 657 |
+
"loss": 1.155249834060669,
|
| 658 |
+
"mean_token_accuracy": 0.7887681126594543,
|
| 659 |
+
"num_tokens": 186333.0,
|
| 660 |
+
"step": 65
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"entropy": 0.85546875,
|
| 664 |
+
"epoch": 0.4647887323943662,
|
| 665 |
+
"grad_norm": 1.1953125,
|
| 666 |
+
"learning_rate": 5e-05,
|
| 667 |
+
"loss": 0.8293672800064087,
|
| 668 |
+
"mean_token_accuracy": 0.8543021082878113,
|
| 669 |
+
"num_tokens": 188956.0,
|
| 670 |
+
"step": 66
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"entropy": 0.91015625,
|
| 674 |
+
"epoch": 0.47183098591549294,
|
| 675 |
+
"grad_norm": 1.1171875,
|
| 676 |
+
"learning_rate": 5e-05,
|
| 677 |
+
"loss": 0.894307553768158,
|
| 678 |
+
"mean_token_accuracy": 0.8422175049781799,
|
| 679 |
+
"num_tokens": 191309.0,
|
| 680 |
+
"step": 67
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"entropy": 1.203125,
|
| 684 |
+
"epoch": 0.4788732394366197,
|
| 685 |
+
"grad_norm": 1.0234375,
|
| 686 |
+
"learning_rate": 5e-05,
|
| 687 |
+
"loss": 1.1631497144699097,
|
| 688 |
+
"mean_token_accuracy": 0.7861995100975037,
|
| 689 |
+
"num_tokens": 194114.0,
|
| 690 |
+
"step": 68
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"entropy": 1.328125,
|
| 694 |
+
"epoch": 0.4859154929577465,
|
| 695 |
+
"grad_norm": 1.0390625,
|
| 696 |
+
"learning_rate": 5e-05,
|
| 697 |
+
"loss": 1.2809975147247314,
|
| 698 |
+
"mean_token_accuracy": 0.7543198466300964,
|
| 699 |
+
"num_tokens": 197305.0,
|
| 700 |
+
"step": 69
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"entropy": 1.015625,
|
| 704 |
+
"epoch": 0.49295774647887325,
|
| 705 |
+
"grad_norm": 1.3203125,
|
| 706 |
+
"learning_rate": 5e-05,
|
| 707 |
+
"loss": 1.0155205726623535,
|
| 708 |
+
"mean_token_accuracy": 0.834074079990387,
|
| 709 |
+
"num_tokens": 199338.0,
|
| 710 |
+
"step": 70
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"entropy": 0.83984375,
|
| 714 |
+
"epoch": 0.5,
|
| 715 |
+
"grad_norm": 1.1796875,
|
| 716 |
+
"learning_rate": 5e-05,
|
| 717 |
+
"loss": 0.7589200735092163,
|
| 718 |
+
"mean_token_accuracy": 0.8618420958518982,
|
| 719 |
+
"num_tokens": 201474.0,
|
| 720 |
+
"step": 71
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"entropy": 1.0859375,
|
| 724 |
+
"epoch": 0.5070422535211268,
|
| 725 |
+
"grad_norm": 1.2109375,
|
| 726 |
+
"learning_rate": 5e-05,
|
| 727 |
+
"loss": 1.0090736150741577,
|
| 728 |
+
"mean_token_accuracy": 0.8233413100242615,
|
| 729 |
+
"num_tokens": 203984.0,
|
| 730 |
+
"step": 72
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"entropy": 1.0703125,
|
| 734 |
+
"epoch": 0.5140845070422535,
|
| 735 |
+
"grad_norm": 0.9140625,
|
| 736 |
+
"learning_rate": 5e-05,
|
| 737 |
+
"loss": 1.046358346939087,
|
| 738 |
+
"mean_token_accuracy": 0.802097737789154,
|
| 739 |
+
"num_tokens": 207615.0,
|
| 740 |
+
"step": 73
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"entropy": 1.203125,
|
| 744 |
+
"epoch": 0.5211267605633803,
|
| 745 |
+
"grad_norm": 1.2109375,
|
| 746 |
+
"learning_rate": 5e-05,
|
| 747 |
+
"loss": 1.14384126663208,
|
| 748 |
+
"mean_token_accuracy": 0.7883082628250122,
|
| 749 |
+
"num_tokens": 209881.0,
|
| 750 |
+
"step": 74
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"entropy": 1.015625,
|
| 754 |
+
"epoch": 0.528169014084507,
|
| 755 |
+
"grad_norm": 0.98828125,
|
| 756 |
+
"learning_rate": 5e-05,
|
| 757 |
+
"loss": 0.9863561987876892,
|
| 758 |
+
"mean_token_accuracy": 0.8165745735168457,
|
| 759 |
+
"num_tokens": 212604.0,
|
| 760 |
+
"step": 75
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"entropy": 1.2421875,
|
| 764 |
+
"epoch": 0.5352112676056338,
|
| 765 |
+
"grad_norm": 1.1171875,
|
| 766 |
+
"learning_rate": 5e-05,
|
| 767 |
+
"loss": 1.1881471872329712,
|
| 768 |
+
"mean_token_accuracy": 0.7924298048019409,
|
| 769 |
+
"num_tokens": 215069.0,
|
| 770 |
+
"step": 76
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"entropy": 0.99609375,
|
| 774 |
+
"epoch": 0.5422535211267606,
|
| 775 |
+
"grad_norm": 0.86328125,
|
| 776 |
+
"learning_rate": 5e-05,
|
| 777 |
+
"loss": 0.9455610513687134,
|
| 778 |
+
"mean_token_accuracy": 0.8238698244094849,
|
| 779 |
+
"num_tokens": 217842.0,
|
| 780 |
+
"step": 77
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"entropy": 0.93359375,
|
| 784 |
+
"epoch": 0.5492957746478874,
|
| 785 |
+
"grad_norm": 1.015625,
|
| 786 |
+
"learning_rate": 5e-05,
|
| 787 |
+
"loss": 0.8940377235412598,
|
| 788 |
+
"mean_token_accuracy": 0.8325259685516357,
|
| 789 |
+
"num_tokens": 220740.0,
|
| 790 |
+
"step": 78
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"entropy": 1.078125,
|
| 794 |
+
"epoch": 0.5563380281690141,
|
| 795 |
+
"grad_norm": 1.0390625,
|
| 796 |
+
"learning_rate": 5e-05,
|
| 797 |
+
"loss": 1.072181224822998,
|
| 798 |
+
"mean_token_accuracy": 0.7974197864532471,
|
| 799 |
+
"num_tokens": 223616.0,
|
| 800 |
+
"step": 79
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"entropy": 0.9609375,
|
| 804 |
+
"epoch": 0.5633802816901409,
|
| 805 |
+
"grad_norm": 1.0625,
|
| 806 |
+
"learning_rate": 5e-05,
|
| 807 |
+
"loss": 0.968547523021698,
|
| 808 |
+
"mean_token_accuracy": 0.8254417181015015,
|
| 809 |
+
"num_tokens": 226454.0,
|
| 810 |
+
"step": 80
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"entropy": 1.0625,
|
| 814 |
+
"epoch": 0.5704225352112676,
|
| 815 |
+
"grad_norm": 1.0390625,
|
| 816 |
+
"learning_rate": 5e-05,
|
| 817 |
+
"loss": 1.029323935508728,
|
| 818 |
+
"mean_token_accuracy": 0.8091970086097717,
|
| 819 |
+
"num_tokens": 229376.0,
|
| 820 |
+
"step": 81
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"entropy": 1.296875,
|
| 824 |
+
"epoch": 0.5774647887323944,
|
| 825 |
+
"grad_norm": 1.078125,
|
| 826 |
+
"learning_rate": 5e-05,
|
| 827 |
+
"loss": 1.30299711227417,
|
| 828 |
+
"mean_token_accuracy": 0.7740401029586792,
|
| 829 |
+
"num_tokens": 232327.0,
|
| 830 |
+
"step": 82
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"entropy": 1.34375,
|
| 834 |
+
"epoch": 0.5845070422535211,
|
| 835 |
+
"grad_norm": 1.046875,
|
| 836 |
+
"learning_rate": 5e-05,
|
| 837 |
+
"loss": 1.3241432905197144,
|
| 838 |
+
"mean_token_accuracy": 0.7471368312835693,
|
| 839 |
+
"num_tokens": 235653.0,
|
| 840 |
+
"step": 83
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"entropy": 1.1953125,
|
| 844 |
+
"epoch": 0.5915492957746479,
|
| 845 |
+
"grad_norm": 1.0546875,
|
| 846 |
+
"learning_rate": 5e-05,
|
| 847 |
+
"loss": 1.1841082572937012,
|
| 848 |
+
"mean_token_accuracy": 0.7741320133209229,
|
| 849 |
+
"num_tokens": 238282.0,
|
| 850 |
+
"step": 84
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"entropy": 0.96875,
|
| 854 |
+
"epoch": 0.5985915492957746,
|
| 855 |
+
"grad_norm": 1.0,
|
| 856 |
+
"learning_rate": 5e-05,
|
| 857 |
+
"loss": 0.9051790833473206,
|
| 858 |
+
"mean_token_accuracy": 0.8358089327812195,
|
| 859 |
+
"num_tokens": 240781.0,
|
| 860 |
+
"step": 85
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"entropy": 0.81640625,
|
| 864 |
+
"epoch": 0.6056338028169014,
|
| 865 |
+
"grad_norm": 1.0625,
|
| 866 |
+
"learning_rate": 5e-05,
|
| 867 |
+
"loss": 0.8058979511260986,
|
| 868 |
+
"mean_token_accuracy": 0.8656787872314453,
|
| 869 |
+
"num_tokens": 242881.0,
|
| 870 |
+
"step": 86
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"entropy": 1.0234375,
|
| 874 |
+
"epoch": 0.6126760563380281,
|
| 875 |
+
"grad_norm": 0.89453125,
|
| 876 |
+
"learning_rate": 5e-05,
|
| 877 |
+
"loss": 0.9673476219177246,
|
| 878 |
+
"mean_token_accuracy": 0.8199172616004944,
|
| 879 |
+
"num_tokens": 246032.0,
|
| 880 |
+
"step": 87
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"entropy": 0.9609375,
|
| 884 |
+
"epoch": 0.6197183098591549,
|
| 885 |
+
"grad_norm": 1.2265625,
|
| 886 |
+
"learning_rate": 5e-05,
|
| 887 |
+
"loss": 0.9231598377227783,
|
| 888 |
+
"mean_token_accuracy": 0.833532452583313,
|
| 889 |
+
"num_tokens": 248551.0,
|
| 890 |
+
"step": 88
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"entropy": 0.9921875,
|
| 894 |
+
"epoch": 0.6267605633802817,
|
| 895 |
+
"grad_norm": 1.1015625,
|
| 896 |
+
"learning_rate": 5e-05,
|
| 897 |
+
"loss": 0.998257040977478,
|
| 898 |
+
"mean_token_accuracy": 0.8266341686248779,
|
| 899 |
+
"num_tokens": 251022.0,
|
| 900 |
+
"step": 89
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"entropy": 1.0546875,
|
| 904 |
+
"epoch": 0.6338028169014085,
|
| 905 |
+
"grad_norm": 1.015625,
|
| 906 |
+
"learning_rate": 5e-05,
|
| 907 |
+
"loss": 1.0252176523208618,
|
| 908 |
+
"mean_token_accuracy": 0.8216797709465027,
|
| 909 |
+
"num_tokens": 253447.0,
|
| 910 |
+
"step": 90
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"entropy": 1.1484375,
|
| 914 |
+
"epoch": 0.6408450704225352,
|
| 915 |
+
"grad_norm": 1.0,
|
| 916 |
+
"learning_rate": 5e-05,
|
| 917 |
+
"loss": 1.164228081703186,
|
| 918 |
+
"mean_token_accuracy": 0.7702828645706177,
|
| 919 |
+
"num_tokens": 256672.0,
|
| 920 |
+
"step": 91
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"entropy": 0.77734375,
|
| 924 |
+
"epoch": 0.647887323943662,
|
| 925 |
+
"grad_norm": 0.9609375,
|
| 926 |
+
"learning_rate": 5e-05,
|
| 927 |
+
"loss": 0.7661694288253784,
|
| 928 |
+
"mean_token_accuracy": 0.8448406457901001,
|
| 929 |
+
"num_tokens": 259316.0,
|
| 930 |
+
"step": 92
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"entropy": 1.15625,
|
| 934 |
+
"epoch": 0.6549295774647887,
|
| 935 |
+
"grad_norm": 0.98828125,
|
| 936 |
+
"learning_rate": 5e-05,
|
| 937 |
+
"loss": 1.1309537887573242,
|
| 938 |
+
"mean_token_accuracy": 0.7973990440368652,
|
| 939 |
+
"num_tokens": 262246.0,
|
| 940 |
+
"step": 93
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"entropy": 1.046875,
|
| 944 |
+
"epoch": 0.6619718309859155,
|
| 945 |
+
"grad_norm": 0.9296875,
|
| 946 |
+
"learning_rate": 5e-05,
|
| 947 |
+
"loss": 1.0406370162963867,
|
| 948 |
+
"mean_token_accuracy": 0.8099964261054993,
|
| 949 |
+
"num_tokens": 265075.0,
|
| 950 |
+
"step": 94
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"entropy": 1.0,
|
| 954 |
+
"epoch": 0.6690140845070423,
|
| 955 |
+
"grad_norm": 0.90234375,
|
| 956 |
+
"learning_rate": 5e-05,
|
| 957 |
+
"loss": 0.9478747844696045,
|
| 958 |
+
"mean_token_accuracy": 0.8167044520378113,
|
| 959 |
+
"num_tokens": 267729.0,
|
| 960 |
+
"step": 95
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"entropy": 1.125,
|
| 964 |
+
"epoch": 0.676056338028169,
|
| 965 |
+
"grad_norm": 1.0234375,
|
| 966 |
+
"learning_rate": 5e-05,
|
| 967 |
+
"loss": 1.0806983709335327,
|
| 968 |
+
"mean_token_accuracy": 0.8070116639137268,
|
| 969 |
+
"num_tokens": 270732.0,
|
| 970 |
+
"step": 96
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"entropy": 1.3359375,
|
| 974 |
+
"epoch": 0.6830985915492958,
|
| 975 |
+
"grad_norm": 0.95703125,
|
| 976 |
+
"learning_rate": 5e-05,
|
| 977 |
+
"loss": 1.2827727794647217,
|
| 978 |
+
"mean_token_accuracy": 0.7681026458740234,
|
| 979 |
+
"num_tokens": 274013.0,
|
| 980 |
+
"step": 97
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"entropy": 0.96875,
|
| 984 |
+
"epoch": 0.6901408450704225,
|
| 985 |
+
"grad_norm": 1.1484375,
|
| 986 |
+
"learning_rate": 5e-05,
|
| 987 |
+
"loss": 0.9255345463752747,
|
| 988 |
+
"mean_token_accuracy": 0.820858895778656,
|
| 989 |
+
"num_tokens": 276466.0,
|
| 990 |
+
"step": 98
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"entropy": 0.8828125,
|
| 994 |
+
"epoch": 0.6971830985915493,
|
| 995 |
+
"grad_norm": 0.828125,
|
| 996 |
+
"learning_rate": 5e-05,
|
| 997 |
+
"loss": 0.8368520736694336,
|
| 998 |
+
"mean_token_accuracy": 0.8478841185569763,
|
| 999 |
+
"num_tokens": 279097.0,
|
| 1000 |
+
"step": 99
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"entropy": 1.1875,
|
| 1004 |
+
"epoch": 0.704225352112676,
|
| 1005 |
+
"grad_norm": 0.91796875,
|
| 1006 |
+
"learning_rate": 5e-05,
|
| 1007 |
+
"loss": 1.1672570705413818,
|
| 1008 |
+
"mean_token_accuracy": 0.7689912915229797,
|
| 1009 |
+
"num_tokens": 282317.0,
|
| 1010 |
+
"step": 100
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"entropy": 1.53125,
|
| 1014 |
+
"epoch": 0.7112676056338029,
|
| 1015 |
+
"grad_norm": 1.0625,
|
| 1016 |
+
"learning_rate": 5e-05,
|
| 1017 |
+
"loss": 1.4877268075942993,
|
| 1018 |
+
"mean_token_accuracy": 0.7194092869758606,
|
| 1019 |
+
"num_tokens": 285643.0,
|
| 1020 |
+
"step": 101
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"entropy": 1.0,
|
| 1024 |
+
"epoch": 0.7183098591549296,
|
| 1025 |
+
"grad_norm": 1.0625,
|
| 1026 |
+
"learning_rate": 5e-05,
|
| 1027 |
+
"loss": 0.956112265586853,
|
| 1028 |
+
"mean_token_accuracy": 0.8256269693374634,
|
| 1029 |
+
"num_tokens": 288203.0,
|
| 1030 |
+
"step": 102
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"entropy": 1.1953125,
|
| 1034 |
+
"epoch": 0.7253521126760564,
|
| 1035 |
+
"grad_norm": 1.1328125,
|
| 1036 |
+
"learning_rate": 5e-05,
|
| 1037 |
+
"loss": 1.2514692544937134,
|
| 1038 |
+
"mean_token_accuracy": 0.771497905254364,
|
| 1039 |
+
"num_tokens": 291095.0,
|
| 1040 |
+
"step": 103
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"entropy": 1.2109375,
|
| 1044 |
+
"epoch": 0.7323943661971831,
|
| 1045 |
+
"grad_norm": 0.93359375,
|
| 1046 |
+
"learning_rate": 5e-05,
|
| 1047 |
+
"loss": 1.1472859382629395,
|
| 1048 |
+
"mean_token_accuracy": 0.7850604057312012,
|
| 1049 |
+
"num_tokens": 293834.0,
|
| 1050 |
+
"step": 104
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"entropy": 0.7734375,
|
| 1054 |
+
"epoch": 0.7394366197183099,
|
| 1055 |
+
"grad_norm": 1.28125,
|
| 1056 |
+
"learning_rate": 5e-05,
|
| 1057 |
+
"loss": 0.7254777550697327,
|
| 1058 |
+
"mean_token_accuracy": 0.8632119297981262,
|
| 1059 |
+
"num_tokens": 295984.0,
|
| 1060 |
+
"step": 105
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"entropy": 0.82421875,
|
| 1064 |
+
"epoch": 0.7464788732394366,
|
| 1065 |
+
"grad_norm": 1.078125,
|
| 1066 |
+
"learning_rate": 5e-05,
|
| 1067 |
+
"loss": 0.795278787612915,
|
| 1068 |
+
"mean_token_accuracy": 0.8470132946968079,
|
| 1069 |
+
"num_tokens": 298319.0,
|
| 1070 |
+
"step": 106
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"entropy": 1.09375,
|
| 1074 |
+
"epoch": 0.7535211267605634,
|
| 1075 |
+
"grad_norm": 0.87109375,
|
| 1076 |
+
"learning_rate": 5e-05,
|
| 1077 |
+
"loss": 1.0427790880203247,
|
| 1078 |
+
"mean_token_accuracy": 0.804815411567688,
|
| 1079 |
+
"num_tokens": 301442.0,
|
| 1080 |
+
"step": 107
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"entropy": 1.1171875,
|
| 1084 |
+
"epoch": 0.7605633802816901,
|
| 1085 |
+
"grad_norm": 0.95703125,
|
| 1086 |
+
"learning_rate": 5e-05,
|
| 1087 |
+
"loss": 1.104764699935913,
|
| 1088 |
+
"mean_token_accuracy": 0.8097593188285828,
|
| 1089 |
+
"num_tokens": 304483.0,
|
| 1090 |
+
"step": 108
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"entropy": 0.7421875,
|
| 1094 |
+
"epoch": 0.7676056338028169,
|
| 1095 |
+
"grad_norm": 1.0,
|
| 1096 |
+
"learning_rate": 5e-05,
|
| 1097 |
+
"loss": 0.6636262536048889,
|
| 1098 |
+
"mean_token_accuracy": 0.861122190952301,
|
| 1099 |
+
"num_tokens": 307004.0,
|
| 1100 |
+
"step": 109
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"entropy": 0.97265625,
|
| 1104 |
+
"epoch": 0.7746478873239436,
|
| 1105 |
+
"grad_norm": 0.90234375,
|
| 1106 |
+
"learning_rate": 5e-05,
|
| 1107 |
+
"loss": 0.9813203811645508,
|
| 1108 |
+
"mean_token_accuracy": 0.8114525079727173,
|
| 1109 |
+
"num_tokens": 309876.0,
|
| 1110 |
+
"step": 110
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"entropy": 1.09375,
|
| 1114 |
+
"epoch": 0.7816901408450704,
|
| 1115 |
+
"grad_norm": 1.078125,
|
| 1116 |
+
"learning_rate": 5e-05,
|
| 1117 |
+
"loss": 1.0708268880844116,
|
| 1118 |
+
"mean_token_accuracy": 0.7995486855506897,
|
| 1119 |
+
"num_tokens": 312543.0,
|
| 1120 |
+
"step": 111
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"entropy": 0.953125,
|
| 1124 |
+
"epoch": 0.7887323943661971,
|
| 1125 |
+
"grad_norm": 0.9453125,
|
| 1126 |
+
"learning_rate": 5e-05,
|
| 1127 |
+
"loss": 0.9298059344291687,
|
| 1128 |
+
"mean_token_accuracy": 0.8266475796699524,
|
| 1129 |
+
"num_tokens": 315343.0,
|
| 1130 |
+
"step": 112
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"entropy": 0.93359375,
|
| 1134 |
+
"epoch": 0.795774647887324,
|
| 1135 |
+
"grad_norm": 1.2109375,
|
| 1136 |
+
"learning_rate": 5e-05,
|
| 1137 |
+
"loss": 0.9146865606307983,
|
| 1138 |
+
"mean_token_accuracy": 0.8447837233543396,
|
| 1139 |
+
"num_tokens": 317709.0,
|
| 1140 |
+
"step": 113
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"entropy": 1.0390625,
|
| 1144 |
+
"epoch": 0.8028169014084507,
|
| 1145 |
+
"grad_norm": 1.015625,
|
| 1146 |
+
"learning_rate": 5e-05,
|
| 1147 |
+
"loss": 1.0023632049560547,
|
| 1148 |
+
"mean_token_accuracy": 0.8247303366661072,
|
| 1149 |
+
"num_tokens": 320313.0,
|
| 1150 |
+
"step": 114
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"entropy": 1.0234375,
|
| 1154 |
+
"epoch": 0.8098591549295775,
|
| 1155 |
+
"grad_norm": 0.9375,
|
| 1156 |
+
"learning_rate": 5e-05,
|
| 1157 |
+
"loss": 1.0119502544403076,
|
| 1158 |
+
"mean_token_accuracy": 0.8081619739532471,
|
| 1159 |
+
"num_tokens": 323433.0,
|
| 1160 |
+
"step": 115
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"entropy": 1.1484375,
|
| 1164 |
+
"epoch": 0.8169014084507042,
|
| 1165 |
+
"grad_norm": 1.09375,
|
| 1166 |
+
"learning_rate": 5e-05,
|
| 1167 |
+
"loss": 1.129537582397461,
|
| 1168 |
+
"mean_token_accuracy": 0.7915045619010925,
|
| 1169 |
+
"num_tokens": 326525.0,
|
| 1170 |
+
"step": 116
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"entropy": 0.828125,
|
| 1174 |
+
"epoch": 0.823943661971831,
|
| 1175 |
+
"grad_norm": 0.96875,
|
| 1176 |
+
"learning_rate": 5e-05,
|
| 1177 |
+
"loss": 0.8433165550231934,
|
| 1178 |
+
"mean_token_accuracy": 0.8409850001335144,
|
| 1179 |
+
"num_tokens": 328929.0,
|
| 1180 |
+
"step": 117
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"entropy": 1.0625,
|
| 1184 |
+
"epoch": 0.8309859154929577,
|
| 1185 |
+
"grad_norm": 1.0078125,
|
| 1186 |
+
"learning_rate": 5e-05,
|
| 1187 |
+
"loss": 1.0346664190292358,
|
| 1188 |
+
"mean_token_accuracy": 0.7833219170570374,
|
| 1189 |
+
"num_tokens": 331863.0,
|
| 1190 |
+
"step": 118
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"entropy": 0.98046875,
|
| 1194 |
+
"epoch": 0.8380281690140845,
|
| 1195 |
+
"grad_norm": 1.09375,
|
| 1196 |
+
"learning_rate": 5e-05,
|
| 1197 |
+
"loss": 0.950432300567627,
|
| 1198 |
+
"mean_token_accuracy": 0.8241989612579346,
|
| 1199 |
+
"num_tokens": 334243.0,
|
| 1200 |
+
"step": 119
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"entropy": 1.0546875,
|
| 1204 |
+
"epoch": 0.8450704225352113,
|
| 1205 |
+
"grad_norm": 0.8984375,
|
| 1206 |
+
"learning_rate": 5e-05,
|
| 1207 |
+
"loss": 1.0502727031707764,
|
| 1208 |
+
"mean_token_accuracy": 0.7917791604995728,
|
| 1209 |
+
"num_tokens": 337584.0,
|
| 1210 |
+
"step": 120
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"entropy": 0.89453125,
|
| 1214 |
+
"epoch": 0.852112676056338,
|
| 1215 |
+
"grad_norm": 0.9765625,
|
| 1216 |
+
"learning_rate": 5e-05,
|
| 1217 |
+
"loss": 0.9121181964874268,
|
| 1218 |
+
"mean_token_accuracy": 0.8309713006019592,
|
| 1219 |
+
"num_tokens": 340485.0,
|
| 1220 |
+
"step": 121
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"entropy": 0.88671875,
|
| 1224 |
+
"epoch": 0.8591549295774648,
|
| 1225 |
+
"grad_norm": 1.1328125,
|
| 1226 |
+
"learning_rate": 5e-05,
|
| 1227 |
+
"loss": 0.9219837188720703,
|
| 1228 |
+
"mean_token_accuracy": 0.8140540719032288,
|
| 1229 |
+
"num_tokens": 343268.0,
|
| 1230 |
+
"step": 122
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"entropy": 0.8515625,
|
| 1234 |
+
"epoch": 0.8661971830985915,
|
| 1235 |
+
"grad_norm": 1.1484375,
|
| 1236 |
+
"learning_rate": 5e-05,
|
| 1237 |
+
"loss": 0.7965174317359924,
|
| 1238 |
+
"mean_token_accuracy": 0.8607358932495117,
|
| 1239 |
+
"num_tokens": 345423.0,
|
| 1240 |
+
"step": 123
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"entropy": 0.9765625,
|
| 1244 |
+
"epoch": 0.8732394366197183,
|
| 1245 |
+
"grad_norm": 0.82421875,
|
| 1246 |
+
"learning_rate": 5e-05,
|
| 1247 |
+
"loss": 0.9578317999839783,
|
| 1248 |
+
"mean_token_accuracy": 0.8050633072853088,
|
| 1249 |
+
"num_tokens": 348591.0,
|
| 1250 |
+
"step": 124
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"entropy": 0.953125,
|
| 1254 |
+
"epoch": 0.8802816901408451,
|
| 1255 |
+
"grad_norm": 0.890625,
|
| 1256 |
+
"learning_rate": 5e-05,
|
| 1257 |
+
"loss": 0.9342446327209473,
|
| 1258 |
+
"mean_token_accuracy": 0.8261499404907227,
|
| 1259 |
+
"num_tokens": 351360.0,
|
| 1260 |
+
"step": 125
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"entropy": 1.0859375,
|
| 1264 |
+
"epoch": 0.8873239436619719,
|
| 1265 |
+
"grad_norm": 0.88671875,
|
| 1266 |
+
"learning_rate": 5e-05,
|
| 1267 |
+
"loss": 1.0417184829711914,
|
| 1268 |
+
"mean_token_accuracy": 0.7993496060371399,
|
| 1269 |
+
"num_tokens": 354443.0,
|
| 1270 |
+
"step": 126
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"entropy": 1.1875,
|
| 1274 |
+
"epoch": 0.8943661971830986,
|
| 1275 |
+
"grad_norm": 0.96484375,
|
| 1276 |
+
"learning_rate": 5e-05,
|
| 1277 |
+
"loss": 1.169641137123108,
|
| 1278 |
+
"mean_token_accuracy": 0.789621889591217,
|
| 1279 |
+
"num_tokens": 356937.0,
|
| 1280 |
+
"step": 127
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"entropy": 1.171875,
|
| 1284 |
+
"epoch": 0.9014084507042254,
|
| 1285 |
+
"grad_norm": 1.0546875,
|
| 1286 |
+
"learning_rate": 5e-05,
|
| 1287 |
+
"loss": 1.0926804542541504,
|
| 1288 |
+
"mean_token_accuracy": 0.78143709897995,
|
| 1289 |
+
"num_tokens": 359951.0,
|
| 1290 |
+
"step": 128
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"entropy": 0.90625,
|
| 1294 |
+
"epoch": 0.9084507042253521,
|
| 1295 |
+
"grad_norm": 1.046875,
|
| 1296 |
+
"learning_rate": 5e-05,
|
| 1297 |
+
"loss": 0.8831167817115784,
|
| 1298 |
+
"mean_token_accuracy": 0.8297297358512878,
|
| 1299 |
+
"num_tokens": 362549.0,
|
| 1300 |
+
"step": 129
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"entropy": 1.34375,
|
| 1304 |
+
"epoch": 0.9154929577464789,
|
| 1305 |
+
"grad_norm": 1.1640625,
|
| 1306 |
+
"learning_rate": 5e-05,
|
| 1307 |
+
"loss": 1.2982932329177856,
|
| 1308 |
+
"mean_token_accuracy": 0.7599736452102661,
|
| 1309 |
+
"num_tokens": 365590.0,
|
| 1310 |
+
"step": 130
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"entropy": 0.8515625,
|
| 1314 |
+
"epoch": 0.9225352112676056,
|
| 1315 |
+
"grad_norm": 1.140625,
|
| 1316 |
+
"learning_rate": 5e-05,
|
| 1317 |
+
"loss": 0.7926298975944519,
|
| 1318 |
+
"mean_token_accuracy": 0.8279292583465576,
|
| 1319 |
+
"num_tokens": 368312.0,
|
| 1320 |
+
"step": 131
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"entropy": 0.85546875,
|
| 1324 |
+
"epoch": 0.9295774647887324,
|
| 1325 |
+
"grad_norm": 1.0546875,
|
| 1326 |
+
"learning_rate": 5e-05,
|
| 1327 |
+
"loss": 0.876966655254364,
|
| 1328 |
+
"mean_token_accuracy": 0.8338558077812195,
|
| 1329 |
+
"num_tokens": 370872.0,
|
| 1330 |
+
"step": 132
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"entropy": 0.58203125,
|
| 1334 |
+
"epoch": 0.9366197183098591,
|
| 1335 |
+
"grad_norm": 1.09375,
|
| 1336 |
+
"learning_rate": 5e-05,
|
| 1337 |
+
"loss": 0.5232805609703064,
|
| 1338 |
+
"mean_token_accuracy": 0.9088762998580933,
|
| 1339 |
+
"num_tokens": 372998.0,
|
| 1340 |
+
"step": 133
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"entropy": 1.3125,
|
| 1344 |
+
"epoch": 0.9436619718309859,
|
| 1345 |
+
"grad_norm": 1.0859375,
|
| 1346 |
+
"learning_rate": 5e-05,
|
| 1347 |
+
"loss": 1.2858319282531738,
|
| 1348 |
+
"mean_token_accuracy": 0.764127790927887,
|
| 1349 |
+
"num_tokens": 375855.0,
|
| 1350 |
+
"step": 134
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"entropy": 0.9453125,
|
| 1354 |
+
"epoch": 0.9507042253521126,
|
| 1355 |
+
"grad_norm": 0.8125,
|
| 1356 |
+
"learning_rate": 5e-05,
|
| 1357 |
+
"loss": 0.9115528464317322,
|
| 1358 |
+
"mean_token_accuracy": 0.8201290965080261,
|
| 1359 |
+
"num_tokens": 379271.0,
|
| 1360 |
+
"step": 135
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"entropy": 1.0703125,
|
| 1364 |
+
"epoch": 0.9577464788732394,
|
| 1365 |
+
"grad_norm": 1.3828125,
|
| 1366 |
+
"learning_rate": 5e-05,
|
| 1367 |
+
"loss": 1.03813898563385,
|
| 1368 |
+
"mean_token_accuracy": 0.8119325637817383,
|
| 1369 |
+
"num_tokens": 381592.0,
|
| 1370 |
+
"step": 136
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"entropy": 1.109375,
|
| 1374 |
+
"epoch": 0.9647887323943662,
|
| 1375 |
+
"grad_norm": 0.953125,
|
| 1376 |
+
"learning_rate": 5e-05,
|
| 1377 |
+
"loss": 1.1146811246871948,
|
| 1378 |
+
"mean_token_accuracy": 0.790168285369873,
|
| 1379 |
+
"num_tokens": 384631.0,
|
| 1380 |
+
"step": 137
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"entropy": 1.1640625,
|
| 1384 |
+
"epoch": 0.971830985915493,
|
| 1385 |
+
"grad_norm": 1.0234375,
|
| 1386 |
+
"learning_rate": 5e-05,
|
| 1387 |
+
"loss": 1.173197627067566,
|
| 1388 |
+
"mean_token_accuracy": 0.7781087160110474,
|
| 1389 |
+
"num_tokens": 387325.0,
|
| 1390 |
+
"step": 138
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"entropy": 0.828125,
|
| 1394 |
+
"epoch": 0.9788732394366197,
|
| 1395 |
+
"grad_norm": 1.0546875,
|
| 1396 |
+
"learning_rate": 5e-05,
|
| 1397 |
+
"loss": 0.8895862698554993,
|
| 1398 |
+
"mean_token_accuracy": 0.8376736044883728,
|
| 1399 |
+
"num_tokens": 389637.0,
|
| 1400 |
+
"step": 139
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"entropy": 1.1953125,
|
| 1404 |
+
"epoch": 0.9859154929577465,
|
| 1405 |
+
"grad_norm": 0.8515625,
|
| 1406 |
+
"learning_rate": 5e-05,
|
| 1407 |
+
"loss": 1.2043834924697876,
|
| 1408 |
+
"mean_token_accuracy": 0.7506383061408997,
|
| 1409 |
+
"num_tokens": 393170.0,
|
| 1410 |
+
"step": 140
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"entropy": 0.87109375,
|
| 1414 |
+
"epoch": 0.9929577464788732,
|
| 1415 |
+
"grad_norm": 1.0546875,
|
| 1416 |
+
"learning_rate": 5e-05,
|
| 1417 |
+
"loss": 0.8573941588401794,
|
| 1418 |
+
"mean_token_accuracy": 0.8464516401290894,
|
| 1419 |
+
"num_tokens": 395503.0,
|
| 1420 |
+
"step": 141
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"entropy": 1.078125,
|
| 1424 |
+
"epoch": 1.0,
|
| 1425 |
+
"grad_norm": 0.90625,
|
| 1426 |
+
"learning_rate": 5e-05,
|
| 1427 |
+
"loss": 1.0620708465576172,
|
| 1428 |
+
"mean_token_accuracy": 0.7814034819602966,
|
| 1429 |
+
"num_tokens": 398361.0,
|
| 1430 |
+
"step": 142
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 1.0,
|
| 1434 |
+
"eval_entropy": 0.9779188368055556,
|
| 1435 |
+
"eval_loss": 0.9645262956619263,
|
| 1436 |
+
"eval_mean_token_accuracy": 0.8152486599153943,
|
| 1437 |
+
"eval_num_tokens": 398361.0,
|
| 1438 |
+
"eval_runtime": 3.7052,
|
| 1439 |
+
"eval_samples_per_second": 76.65,
|
| 1440 |
+
"eval_steps_per_second": 9.716,
|
| 1441 |
+
"step": 142
|
| 1442 |
+
}
|
| 1443 |
+
],
|
| 1444 |
+
"logging_steps": 1,
|
| 1445 |
+
"max_steps": 710,
|
| 1446 |
+
"num_input_tokens_seen": 0,
|
| 1447 |
+
"num_train_epochs": 5,
|
| 1448 |
+
"save_steps": 500,
|
| 1449 |
+
"stateful_callbacks": {
|
| 1450 |
+
"TrainerControl": {
|
| 1451 |
+
"args": {
|
| 1452 |
+
"should_epoch_stop": false,
|
| 1453 |
+
"should_evaluate": false,
|
| 1454 |
+
"should_log": false,
|
| 1455 |
+
"should_save": true,
|
| 1456 |
+
"should_training_stop": false
|
| 1457 |
+
},
|
| 1458 |
+
"attributes": {}
|
| 1459 |
+
}
|
| 1460 |
+
},
|
| 1461 |
+
"total_flos": 370240540913664.0,
|
| 1462 |
+
"train_batch_size": 8,
|
| 1463 |
+
"trial_name": null,
|
| 1464 |
+
"trial_params": null
|
| 1465 |
+
}
|
checkpoint-142/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fa1d74cfdb570eb6c160e915c5780a404149c8da32b2030eb89b29d13902242
|
| 3 |
+
size 5585
|
checkpoint-284/chat_template.jinja
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system
|
| 2 |
+
You are a helpful AI assistant named SmolLM, trained by Hugging Face<|im_end|>
|
| 3 |
+
' }}{% endif %}{{'<|im_start|>' + message['role'] + '
|
| 4 |
+
' + message['content'] + '<|im_end|>' + '
|
| 5 |
+
'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
|
| 6 |
+
' }}{% endif %}
|
checkpoint-284/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 576,
|
| 13 |
+
"initializer_range": 0.041666666666666664,
|
| 14 |
+
"intermediate_size": 1536,
|
| 15 |
+
"is_llama_config": true,
|
| 16 |
+
"max_position_embeddings": 8192,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"num_attention_heads": 9,
|
| 20 |
+
"num_hidden_layers": 30,
|
| 21 |
+
"num_key_value_heads": 3,
|
| 22 |
+
"pad_token_id": 2,
|
| 23 |
+
"pretraining_tp": 1,
|
| 24 |
+
"rms_norm_eps": 1e-05,
|
| 25 |
+
"rope_interleaved": false,
|
| 26 |
+
"rope_parameters": {
|
| 27 |
+
"rope_theta": 100000,
|
| 28 |
+
"rope_type": "default"
|
| 29 |
+
},
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"transformers.js_config": {
|
| 32 |
+
"kv_cache_dtype": {
|
| 33 |
+
"fp16": "float16",
|
| 34 |
+
"q4f16": "float16"
|
| 35 |
+
}
|
| 36 |
+
},
|
| 37 |
+
"transformers_version": "5.2.0",
|
| 38 |
+
"use_cache": false,
|
| 39 |
+
"vocab_size": 49152
|
| 40 |
+
}
|
checkpoint-284/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
2
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 2,
|
| 8 |
+
"transformers_version": "5.2.0"
|
| 9 |
+
}
|
checkpoint-284/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c61ebca6fe2f9381892b0596de61bd9d50929e0232e6b23a42889fa52a7bc34a
|
| 3 |
+
size 269060552
|
checkpoint-284/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8e2011629d8bed3ef560fa11175cac55684c4e12a72634bb24abf767b6c7399
|
| 3 |
+
size 14645
|
checkpoint-284/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:173f9bcfe6d98de3b547cda3c500547190199c4bb0a20a7f6ab221a288667d8b
|
| 3 |
+
size 1465
|
checkpoint-284/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-284/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|im_start|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"extra_special_tokens": [
|
| 8 |
+
"<|im_start|>",
|
| 9 |
+
"<|im_end|>"
|
| 10 |
+
],
|
| 11 |
+
"is_local": true,
|
| 12 |
+
"model_max_length": 8192,
|
| 13 |
+
"pad_token": "<|im_end|>",
|
| 14 |
+
"tokenizer_class": "TokenizersBackend",
|
| 15 |
+
"unk_token": "<|endoftext|>",
|
| 16 |
+
"vocab_size": 49152
|
| 17 |
+
}
|
checkpoint-284/trainer_state.json
ADDED
|
@@ -0,0 +1,2896 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 284,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 1.78125,
|
| 14 |
+
"epoch": 0.007042253521126761,
|
| 15 |
+
"grad_norm": 5.375,
|
| 16 |
+
"learning_rate": 5e-05,
|
| 17 |
+
"loss": 2.342320680618286,
|
| 18 |
+
"mean_token_accuracy": 0.5398398637771606,
|
| 19 |
+
"num_tokens": 2631.0,
|
| 20 |
+
"step": 1
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.7265625,
|
| 24 |
+
"epoch": 0.014084507042253521,
|
| 25 |
+
"grad_norm": 4.28125,
|
| 26 |
+
"learning_rate": 5e-05,
|
| 27 |
+
"loss": 2.184455394744873,
|
| 28 |
+
"mean_token_accuracy": 0.5707964897155762,
|
| 29 |
+
"num_tokens": 5577.0,
|
| 30 |
+
"step": 2
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.5859375,
|
| 34 |
+
"epoch": 0.02112676056338028,
|
| 35 |
+
"grad_norm": 3.796875,
|
| 36 |
+
"learning_rate": 5e-05,
|
| 37 |
+
"loss": 1.9440301656723022,
|
| 38 |
+
"mean_token_accuracy": 0.6223160624504089,
|
| 39 |
+
"num_tokens": 8426.0,
|
| 40 |
+
"step": 3
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.75,
|
| 44 |
+
"epoch": 0.028169014084507043,
|
| 45 |
+
"grad_norm": 3.234375,
|
| 46 |
+
"learning_rate": 5e-05,
|
| 47 |
+
"loss": 2.034520149230957,
|
| 48 |
+
"mean_token_accuracy": 0.5993841886520386,
|
| 49 |
+
"num_tokens": 11357.0,
|
| 50 |
+
"step": 4
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.8984375,
|
| 54 |
+
"epoch": 0.035211267605633804,
|
| 55 |
+
"grad_norm": 3.3125,
|
| 56 |
+
"learning_rate": 5e-05,
|
| 57 |
+
"loss": 2.151127338409424,
|
| 58 |
+
"mean_token_accuracy": 0.563842236995697,
|
| 59 |
+
"num_tokens": 13926.0,
|
| 60 |
+
"step": 5
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 1.890625,
|
| 64 |
+
"epoch": 0.04225352112676056,
|
| 65 |
+
"grad_norm": 2.546875,
|
| 66 |
+
"learning_rate": 5e-05,
|
| 67 |
+
"loss": 2.066777229309082,
|
| 68 |
+
"mean_token_accuracy": 0.5754261612892151,
|
| 69 |
+
"num_tokens": 17043.0,
|
| 70 |
+
"step": 6
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 2.03125,
|
| 74 |
+
"epoch": 0.04929577464788732,
|
| 75 |
+
"grad_norm": 2.9375,
|
| 76 |
+
"learning_rate": 5e-05,
|
| 77 |
+
"loss": 2.2290213108062744,
|
| 78 |
+
"mean_token_accuracy": 0.5708707571029663,
|
| 79 |
+
"num_tokens": 19612.0,
|
| 80 |
+
"step": 7
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 1.8046875,
|
| 84 |
+
"epoch": 0.056338028169014086,
|
| 85 |
+
"grad_norm": 2.59375,
|
| 86 |
+
"learning_rate": 5e-05,
|
| 87 |
+
"loss": 1.9348505735397339,
|
| 88 |
+
"mean_token_accuracy": 0.6066350936889648,
|
| 89 |
+
"num_tokens": 22363.0,
|
| 90 |
+
"step": 8
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 1.7734375,
|
| 94 |
+
"epoch": 0.06338028169014084,
|
| 95 |
+
"grad_norm": 2.4375,
|
| 96 |
+
"learning_rate": 5e-05,
|
| 97 |
+
"loss": 1.8786349296569824,
|
| 98 |
+
"mean_token_accuracy": 0.628695011138916,
|
| 99 |
+
"num_tokens": 25145.0,
|
| 100 |
+
"step": 9
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 1.8515625,
|
| 104 |
+
"epoch": 0.07042253521126761,
|
| 105 |
+
"grad_norm": 2.296875,
|
| 106 |
+
"learning_rate": 5e-05,
|
| 107 |
+
"loss": 1.9190694093704224,
|
| 108 |
+
"mean_token_accuracy": 0.6125850081443787,
|
| 109 |
+
"num_tokens": 28093.0,
|
| 110 |
+
"step": 10
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"entropy": 1.8515625,
|
| 114 |
+
"epoch": 0.07746478873239436,
|
| 115 |
+
"grad_norm": 2.171875,
|
| 116 |
+
"learning_rate": 5e-05,
|
| 117 |
+
"loss": 1.891284704208374,
|
| 118 |
+
"mean_token_accuracy": 0.6322028040885925,
|
| 119 |
+
"num_tokens": 31138.0,
|
| 120 |
+
"step": 11
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"entropy": 1.953125,
|
| 124 |
+
"epoch": 0.08450704225352113,
|
| 125 |
+
"grad_norm": 2.421875,
|
| 126 |
+
"learning_rate": 5e-05,
|
| 127 |
+
"loss": 1.948248267173767,
|
| 128 |
+
"mean_token_accuracy": 0.6148879528045654,
|
| 129 |
+
"num_tokens": 33779.0,
|
| 130 |
+
"step": 12
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"entropy": 1.9609375,
|
| 134 |
+
"epoch": 0.09154929577464789,
|
| 135 |
+
"grad_norm": 2.671875,
|
| 136 |
+
"learning_rate": 5e-05,
|
| 137 |
+
"loss": 1.968241810798645,
|
| 138 |
+
"mean_token_accuracy": 0.6172575950622559,
|
| 139 |
+
"num_tokens": 36128.0,
|
| 140 |
+
"step": 13
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"entropy": 1.8828125,
|
| 144 |
+
"epoch": 0.09859154929577464,
|
| 145 |
+
"grad_norm": 2.015625,
|
| 146 |
+
"learning_rate": 5e-05,
|
| 147 |
+
"loss": 1.902205228805542,
|
| 148 |
+
"mean_token_accuracy": 0.6156116724014282,
|
| 149 |
+
"num_tokens": 39185.0,
|
| 150 |
+
"step": 14
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"entropy": 1.703125,
|
| 154 |
+
"epoch": 0.1056338028169014,
|
| 155 |
+
"grad_norm": 2.328125,
|
| 156 |
+
"learning_rate": 5e-05,
|
| 157 |
+
"loss": 1.6213911771774292,
|
| 158 |
+
"mean_token_accuracy": 0.6750187277793884,
|
| 159 |
+
"num_tokens": 41867.0,
|
| 160 |
+
"step": 15
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"entropy": 2.015625,
|
| 164 |
+
"epoch": 0.11267605633802817,
|
| 165 |
+
"grad_norm": 2.234375,
|
| 166 |
+
"learning_rate": 5e-05,
|
| 167 |
+
"loss": 1.9176678657531738,
|
| 168 |
+
"mean_token_accuracy": 0.6337976455688477,
|
| 169 |
+
"num_tokens": 44603.0,
|
| 170 |
+
"step": 16
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"entropy": 1.921875,
|
| 174 |
+
"epoch": 0.11971830985915492,
|
| 175 |
+
"grad_norm": 1.921875,
|
| 176 |
+
"learning_rate": 5e-05,
|
| 177 |
+
"loss": 1.8336955308914185,
|
| 178 |
+
"mean_token_accuracy": 0.6507288813591003,
|
| 179 |
+
"num_tokens": 48041.0,
|
| 180 |
+
"step": 17
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"entropy": 1.828125,
|
| 184 |
+
"epoch": 0.1267605633802817,
|
| 185 |
+
"grad_norm": 2.046875,
|
| 186 |
+
"learning_rate": 5e-05,
|
| 187 |
+
"loss": 1.7068480253219604,
|
| 188 |
+
"mean_token_accuracy": 0.6735284924507141,
|
| 189 |
+
"num_tokens": 51158.0,
|
| 190 |
+
"step": 18
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"entropy": 1.7421875,
|
| 194 |
+
"epoch": 0.13380281690140844,
|
| 195 |
+
"grad_norm": 1.8984375,
|
| 196 |
+
"learning_rate": 5e-05,
|
| 197 |
+
"loss": 1.6770858764648438,
|
| 198 |
+
"mean_token_accuracy": 0.6662946343421936,
|
| 199 |
+
"num_tokens": 54750.0,
|
| 200 |
+
"step": 19
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"entropy": 2.109375,
|
| 204 |
+
"epoch": 0.14084507042253522,
|
| 205 |
+
"grad_norm": 2.828125,
|
| 206 |
+
"learning_rate": 5e-05,
|
| 207 |
+
"loss": 1.9946776628494263,
|
| 208 |
+
"mean_token_accuracy": 0.6195476651191711,
|
| 209 |
+
"num_tokens": 57234.0,
|
| 210 |
+
"step": 20
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"entropy": 1.6328125,
|
| 214 |
+
"epoch": 0.14788732394366197,
|
| 215 |
+
"grad_norm": 1.90625,
|
| 216 |
+
"learning_rate": 5e-05,
|
| 217 |
+
"loss": 1.5617338418960571,
|
| 218 |
+
"mean_token_accuracy": 0.6951566934585571,
|
| 219 |
+
"num_tokens": 60752.0,
|
| 220 |
+
"step": 21
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"entropy": 1.78125,
|
| 224 |
+
"epoch": 0.15492957746478872,
|
| 225 |
+
"grad_norm": 2.515625,
|
| 226 |
+
"learning_rate": 5e-05,
|
| 227 |
+
"loss": 1.7242701053619385,
|
| 228 |
+
"mean_token_accuracy": 0.6592110395431519,
|
| 229 |
+
"num_tokens": 63219.0,
|
| 230 |
+
"step": 22
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"entropy": 1.6796875,
|
| 234 |
+
"epoch": 0.1619718309859155,
|
| 235 |
+
"grad_norm": 1.9375,
|
| 236 |
+
"learning_rate": 5e-05,
|
| 237 |
+
"loss": 1.6089009046554565,
|
| 238 |
+
"mean_token_accuracy": 0.6697189807891846,
|
| 239 |
+
"num_tokens": 66394.0,
|
| 240 |
+
"step": 23
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"entropy": 1.9296875,
|
| 244 |
+
"epoch": 0.16901408450704225,
|
| 245 |
+
"grad_norm": 2.125,
|
| 246 |
+
"learning_rate": 5e-05,
|
| 247 |
+
"loss": 1.8391156196594238,
|
| 248 |
+
"mean_token_accuracy": 0.6323254704475403,
|
| 249 |
+
"num_tokens": 69081.0,
|
| 250 |
+
"step": 24
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"entropy": 1.7265625,
|
| 254 |
+
"epoch": 0.176056338028169,
|
| 255 |
+
"grad_norm": 2.15625,
|
| 256 |
+
"learning_rate": 5e-05,
|
| 257 |
+
"loss": 1.6465730667114258,
|
| 258 |
+
"mean_token_accuracy": 0.654959499835968,
|
| 259 |
+
"num_tokens": 71680.0,
|
| 260 |
+
"step": 25
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"entropy": 1.9609375,
|
| 264 |
+
"epoch": 0.18309859154929578,
|
| 265 |
+
"grad_norm": 2.03125,
|
| 266 |
+
"learning_rate": 5e-05,
|
| 267 |
+
"loss": 1.873896598815918,
|
| 268 |
+
"mean_token_accuracy": 0.6338028311729431,
|
| 269 |
+
"num_tokens": 74670.0,
|
| 270 |
+
"step": 26
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"entropy": 1.6875,
|
| 274 |
+
"epoch": 0.19014084507042253,
|
| 275 |
+
"grad_norm": 1.84375,
|
| 276 |
+
"learning_rate": 5e-05,
|
| 277 |
+
"loss": 1.629098653793335,
|
| 278 |
+
"mean_token_accuracy": 0.6646864414215088,
|
| 279 |
+
"num_tokens": 77708.0,
|
| 280 |
+
"step": 27
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"entropy": 1.609375,
|
| 284 |
+
"epoch": 0.19718309859154928,
|
| 285 |
+
"grad_norm": 1.8671875,
|
| 286 |
+
"learning_rate": 5e-05,
|
| 287 |
+
"loss": 1.5465565919876099,
|
| 288 |
+
"mean_token_accuracy": 0.6922129988670349,
|
| 289 |
+
"num_tokens": 80965.0,
|
| 290 |
+
"step": 28
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"entropy": 1.59375,
|
| 294 |
+
"epoch": 0.20422535211267606,
|
| 295 |
+
"grad_norm": 2.140625,
|
| 296 |
+
"learning_rate": 5e-05,
|
| 297 |
+
"loss": 1.5382331609725952,
|
| 298 |
+
"mean_token_accuracy": 0.6933178901672363,
|
| 299 |
+
"num_tokens": 83562.0,
|
| 300 |
+
"step": 29
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"entropy": 1.5390625,
|
| 304 |
+
"epoch": 0.2112676056338028,
|
| 305 |
+
"grad_norm": 1.9921875,
|
| 306 |
+
"learning_rate": 5e-05,
|
| 307 |
+
"loss": 1.4340285062789917,
|
| 308 |
+
"mean_token_accuracy": 0.6944348216056824,
|
| 309 |
+
"num_tokens": 86463.0,
|
| 310 |
+
"step": 30
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"entropy": 1.5,
|
| 314 |
+
"epoch": 0.21830985915492956,
|
| 315 |
+
"grad_norm": 1.953125,
|
| 316 |
+
"learning_rate": 5e-05,
|
| 317 |
+
"loss": 1.41927969455719,
|
| 318 |
+
"mean_token_accuracy": 0.7242607474327087,
|
| 319 |
+
"num_tokens": 89278.0,
|
| 320 |
+
"step": 31
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"entropy": 1.484375,
|
| 324 |
+
"epoch": 0.22535211267605634,
|
| 325 |
+
"grad_norm": 2.328125,
|
| 326 |
+
"learning_rate": 5e-05,
|
| 327 |
+
"loss": 1.395577311515808,
|
| 328 |
+
"mean_token_accuracy": 0.7233359813690186,
|
| 329 |
+
"num_tokens": 91780.0,
|
| 330 |
+
"step": 32
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"entropy": 1.5234375,
|
| 334 |
+
"epoch": 0.2323943661971831,
|
| 335 |
+
"grad_norm": 1.8671875,
|
| 336 |
+
"learning_rate": 5e-05,
|
| 337 |
+
"loss": 1.4273605346679688,
|
| 338 |
+
"mean_token_accuracy": 0.7213010191917419,
|
| 339 |
+
"num_tokens": 94924.0,
|
| 340 |
+
"step": 33
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"entropy": 1.640625,
|
| 344 |
+
"epoch": 0.23943661971830985,
|
| 345 |
+
"grad_norm": 1.9375,
|
| 346 |
+
"learning_rate": 5e-05,
|
| 347 |
+
"loss": 1.549876093864441,
|
| 348 |
+
"mean_token_accuracy": 0.6991525292396545,
|
| 349 |
+
"num_tokens": 98236.0,
|
| 350 |
+
"step": 34
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"entropy": 1.3984375,
|
| 354 |
+
"epoch": 0.24647887323943662,
|
| 355 |
+
"grad_norm": 2.5625,
|
| 356 |
+
"learning_rate": 5e-05,
|
| 357 |
+
"loss": 1.2571226358413696,
|
| 358 |
+
"mean_token_accuracy": 0.7764654159545898,
|
| 359 |
+
"num_tokens": 100530.0,
|
| 360 |
+
"step": 35
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"entropy": 1.484375,
|
| 364 |
+
"epoch": 0.2535211267605634,
|
| 365 |
+
"grad_norm": 1.9921875,
|
| 366 |
+
"learning_rate": 5e-05,
|
| 367 |
+
"loss": 1.430403709411621,
|
| 368 |
+
"mean_token_accuracy": 0.7238532304763794,
|
| 369 |
+
"num_tokens": 103808.0,
|
| 370 |
+
"step": 36
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"entropy": 1.4375,
|
| 374 |
+
"epoch": 0.2605633802816901,
|
| 375 |
+
"grad_norm": 1.9296875,
|
| 376 |
+
"learning_rate": 5e-05,
|
| 377 |
+
"loss": 1.370211124420166,
|
| 378 |
+
"mean_token_accuracy": 0.7603058218955994,
|
| 379 |
+
"num_tokens": 106824.0,
|
| 380 |
+
"step": 37
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"entropy": 1.3046875,
|
| 384 |
+
"epoch": 0.2676056338028169,
|
| 385 |
+
"grad_norm": 1.9921875,
|
| 386 |
+
"learning_rate": 5e-05,
|
| 387 |
+
"loss": 1.160237431526184,
|
| 388 |
+
"mean_token_accuracy": 0.7812739610671997,
|
| 389 |
+
"num_tokens": 109438.0,
|
| 390 |
+
"step": 38
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"entropy": 1.40625,
|
| 394 |
+
"epoch": 0.2746478873239437,
|
| 395 |
+
"grad_norm": 2.265625,
|
| 396 |
+
"learning_rate": 5e-05,
|
| 397 |
+
"loss": 1.3372321128845215,
|
| 398 |
+
"mean_token_accuracy": 0.7477554678916931,
|
| 399 |
+
"num_tokens": 111785.0,
|
| 400 |
+
"step": 39
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"entropy": 1.3671875,
|
| 404 |
+
"epoch": 0.28169014084507044,
|
| 405 |
+
"grad_norm": 1.84375,
|
| 406 |
+
"learning_rate": 5e-05,
|
| 407 |
+
"loss": 1.3203842639923096,
|
| 408 |
+
"mean_token_accuracy": 0.7538297176361084,
|
| 409 |
+
"num_tokens": 114600.0,
|
| 410 |
+
"step": 40
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"entropy": 1.21875,
|
| 414 |
+
"epoch": 0.2887323943661972,
|
| 415 |
+
"grad_norm": 2.203125,
|
| 416 |
+
"learning_rate": 5e-05,
|
| 417 |
+
"loss": 1.1055949926376343,
|
| 418 |
+
"mean_token_accuracy": 0.7860545516014099,
|
| 419 |
+
"num_tokens": 116917.0,
|
| 420 |
+
"step": 41
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"entropy": 1.484375,
|
| 424 |
+
"epoch": 0.29577464788732394,
|
| 425 |
+
"grad_norm": 1.546875,
|
| 426 |
+
"learning_rate": 5e-05,
|
| 427 |
+
"loss": 1.403846025466919,
|
| 428 |
+
"mean_token_accuracy": 0.7451103925704956,
|
| 429 |
+
"num_tokens": 120095.0,
|
| 430 |
+
"step": 42
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"entropy": 1.328125,
|
| 434 |
+
"epoch": 0.3028169014084507,
|
| 435 |
+
"grad_norm": 1.65625,
|
| 436 |
+
"learning_rate": 5e-05,
|
| 437 |
+
"loss": 1.3071074485778809,
|
| 438 |
+
"mean_token_accuracy": 0.7589316964149475,
|
| 439 |
+
"num_tokens": 122986.0,
|
| 440 |
+
"step": 43
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"entropy": 1.3203125,
|
| 444 |
+
"epoch": 0.30985915492957744,
|
| 445 |
+
"grad_norm": 1.8359375,
|
| 446 |
+
"learning_rate": 5e-05,
|
| 447 |
+
"loss": 1.2948389053344727,
|
| 448 |
+
"mean_token_accuracy": 0.7675830125808716,
|
| 449 |
+
"num_tokens": 125795.0,
|
| 450 |
+
"step": 44
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"entropy": 1.3125,
|
| 454 |
+
"epoch": 0.31690140845070425,
|
| 455 |
+
"grad_norm": 1.875,
|
| 456 |
+
"learning_rate": 5e-05,
|
| 457 |
+
"loss": 1.2394049167633057,
|
| 458 |
+
"mean_token_accuracy": 0.7697368264198303,
|
| 459 |
+
"num_tokens": 128235.0,
|
| 460 |
+
"step": 45
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"entropy": 1.3203125,
|
| 464 |
+
"epoch": 0.323943661971831,
|
| 465 |
+
"grad_norm": 1.6796875,
|
| 466 |
+
"learning_rate": 5e-05,
|
| 467 |
+
"loss": 1.3117918968200684,
|
| 468 |
+
"mean_token_accuracy": 0.7587719559669495,
|
| 469 |
+
"num_tokens": 131663.0,
|
| 470 |
+
"step": 46
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"entropy": 1.3515625,
|
| 474 |
+
"epoch": 0.33098591549295775,
|
| 475 |
+
"grad_norm": 1.453125,
|
| 476 |
+
"learning_rate": 5e-05,
|
| 477 |
+
"loss": 1.3100031614303589,
|
| 478 |
+
"mean_token_accuracy": 0.7717799544334412,
|
| 479 |
+
"num_tokens": 134598.0,
|
| 480 |
+
"step": 47
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"entropy": 1.3203125,
|
| 484 |
+
"epoch": 0.3380281690140845,
|
| 485 |
+
"grad_norm": 1.6953125,
|
| 486 |
+
"learning_rate": 5e-05,
|
| 487 |
+
"loss": 1.28191077709198,
|
| 488 |
+
"mean_token_accuracy": 0.7734912037849426,
|
| 489 |
+
"num_tokens": 137224.0,
|
| 490 |
+
"step": 48
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"entropy": 1.3828125,
|
| 494 |
+
"epoch": 0.34507042253521125,
|
| 495 |
+
"grad_norm": 1.515625,
|
| 496 |
+
"learning_rate": 5e-05,
|
| 497 |
+
"loss": 1.3600674867630005,
|
| 498 |
+
"mean_token_accuracy": 0.7438323497772217,
|
| 499 |
+
"num_tokens": 140191.0,
|
| 500 |
+
"step": 49
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"entropy": 1.21875,
|
| 504 |
+
"epoch": 0.352112676056338,
|
| 505 |
+
"grad_norm": 1.46875,
|
| 506 |
+
"learning_rate": 5e-05,
|
| 507 |
+
"loss": 1.1239631175994873,
|
| 508 |
+
"mean_token_accuracy": 0.788614809513092,
|
| 509 |
+
"num_tokens": 142834.0,
|
| 510 |
+
"step": 50
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"entropy": 1.4765625,
|
| 514 |
+
"epoch": 0.3591549295774648,
|
| 515 |
+
"grad_norm": 1.46875,
|
| 516 |
+
"learning_rate": 5e-05,
|
| 517 |
+
"loss": 1.3836950063705444,
|
| 518 |
+
"mean_token_accuracy": 0.7538405060768127,
|
| 519 |
+
"num_tokens": 145576.0,
|
| 520 |
+
"step": 51
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"entropy": 1.1328125,
|
| 524 |
+
"epoch": 0.36619718309859156,
|
| 525 |
+
"grad_norm": 1.59375,
|
| 526 |
+
"learning_rate": 5e-05,
|
| 527 |
+
"loss": 1.1020089387893677,
|
| 528 |
+
"mean_token_accuracy": 0.8136998414993286,
|
| 529 |
+
"num_tokens": 147876.0,
|
| 530 |
+
"step": 52
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"entropy": 1.3671875,
|
| 534 |
+
"epoch": 0.3732394366197183,
|
| 535 |
+
"grad_norm": 1.15625,
|
| 536 |
+
"learning_rate": 5e-05,
|
| 537 |
+
"loss": 1.3127013444900513,
|
| 538 |
+
"mean_token_accuracy": 0.7502166032791138,
|
| 539 |
+
"num_tokens": 151347.0,
|
| 540 |
+
"step": 53
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"entropy": 1.0390625,
|
| 544 |
+
"epoch": 0.38028169014084506,
|
| 545 |
+
"grad_norm": 1.25,
|
| 546 |
+
"learning_rate": 5e-05,
|
| 547 |
+
"loss": 1.0258498191833496,
|
| 548 |
+
"mean_token_accuracy": 0.8231140971183777,
|
| 549 |
+
"num_tokens": 154046.0,
|
| 550 |
+
"step": 54
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"entropy": 1.296875,
|
| 554 |
+
"epoch": 0.3873239436619718,
|
| 555 |
+
"grad_norm": 1.515625,
|
| 556 |
+
"learning_rate": 5e-05,
|
| 557 |
+
"loss": 1.26340913772583,
|
| 558 |
+
"mean_token_accuracy": 0.7880998253822327,
|
| 559 |
+
"num_tokens": 156659.0,
|
| 560 |
+
"step": 55
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"entropy": 1.3828125,
|
| 564 |
+
"epoch": 0.39436619718309857,
|
| 565 |
+
"grad_norm": 1.1484375,
|
| 566 |
+
"learning_rate": 5e-05,
|
| 567 |
+
"loss": 1.3492928743362427,
|
| 568 |
+
"mean_token_accuracy": 0.7417179942131042,
|
| 569 |
+
"num_tokens": 160078.0,
|
| 570 |
+
"step": 56
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"entropy": 1.046875,
|
| 574 |
+
"epoch": 0.4014084507042254,
|
| 575 |
+
"grad_norm": 1.1171875,
|
| 576 |
+
"learning_rate": 5e-05,
|
| 577 |
+
"loss": 1.0268410444259644,
|
| 578 |
+
"mean_token_accuracy": 0.8108739256858826,
|
| 579 |
+
"num_tokens": 163084.0,
|
| 580 |
+
"step": 57
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"entropy": 1.2734375,
|
| 584 |
+
"epoch": 0.4084507042253521,
|
| 585 |
+
"grad_norm": 1.296875,
|
| 586 |
+
"learning_rate": 5e-05,
|
| 587 |
+
"loss": 1.218086838722229,
|
| 588 |
+
"mean_token_accuracy": 0.7687548398971558,
|
| 589 |
+
"num_tokens": 165678.0,
|
| 590 |
+
"step": 58
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"entropy": 1.265625,
|
| 594 |
+
"epoch": 0.4154929577464789,
|
| 595 |
+
"grad_norm": 1.078125,
|
| 596 |
+
"learning_rate": 5e-05,
|
| 597 |
+
"loss": 1.2634363174438477,
|
| 598 |
+
"mean_token_accuracy": 0.7484593987464905,
|
| 599 |
+
"num_tokens": 169256.0,
|
| 600 |
+
"step": 59
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"entropy": 1.046875,
|
| 604 |
+
"epoch": 0.4225352112676056,
|
| 605 |
+
"grad_norm": 1.1796875,
|
| 606 |
+
"learning_rate": 5e-05,
|
| 607 |
+
"loss": 0.9637977480888367,
|
| 608 |
+
"mean_token_accuracy": 0.8200221061706543,
|
| 609 |
+
"num_tokens": 171981.0,
|
| 610 |
+
"step": 60
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"entropy": 1.1015625,
|
| 614 |
+
"epoch": 0.4295774647887324,
|
| 615 |
+
"grad_norm": 1.09375,
|
| 616 |
+
"learning_rate": 5e-05,
|
| 617 |
+
"loss": 1.0381879806518555,
|
| 618 |
+
"mean_token_accuracy": 0.8146411776542664,
|
| 619 |
+
"num_tokens": 174762.0,
|
| 620 |
+
"step": 61
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"entropy": 1.4609375,
|
| 624 |
+
"epoch": 0.43661971830985913,
|
| 625 |
+
"grad_norm": 1.109375,
|
| 626 |
+
"learning_rate": 5e-05,
|
| 627 |
+
"loss": 1.442441701889038,
|
| 628 |
+
"mean_token_accuracy": 0.7236841917037964,
|
| 629 |
+
"num_tokens": 177962.0,
|
| 630 |
+
"step": 62
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"entropy": 0.9453125,
|
| 634 |
+
"epoch": 0.44366197183098594,
|
| 635 |
+
"grad_norm": 1.109375,
|
| 636 |
+
"learning_rate": 5e-05,
|
| 637 |
+
"loss": 0.9214452505111694,
|
| 638 |
+
"mean_token_accuracy": 0.8193620443344116,
|
| 639 |
+
"num_tokens": 180666.0,
|
| 640 |
+
"step": 63
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"entropy": 1.0859375,
|
| 644 |
+
"epoch": 0.4507042253521127,
|
| 645 |
+
"grad_norm": 1.1484375,
|
| 646 |
+
"learning_rate": 5e-05,
|
| 647 |
+
"loss": 1.1097257137298584,
|
| 648 |
+
"mean_token_accuracy": 0.8031823039054871,
|
| 649 |
+
"num_tokens": 183565.0,
|
| 650 |
+
"step": 64
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"entropy": 1.1640625,
|
| 654 |
+
"epoch": 0.45774647887323944,
|
| 655 |
+
"grad_norm": 1.140625,
|
| 656 |
+
"learning_rate": 5e-05,
|
| 657 |
+
"loss": 1.155249834060669,
|
| 658 |
+
"mean_token_accuracy": 0.7887681126594543,
|
| 659 |
+
"num_tokens": 186333.0,
|
| 660 |
+
"step": 65
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"entropy": 0.85546875,
|
| 664 |
+
"epoch": 0.4647887323943662,
|
| 665 |
+
"grad_norm": 1.1953125,
|
| 666 |
+
"learning_rate": 5e-05,
|
| 667 |
+
"loss": 0.8293672800064087,
|
| 668 |
+
"mean_token_accuracy": 0.8543021082878113,
|
| 669 |
+
"num_tokens": 188956.0,
|
| 670 |
+
"step": 66
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"entropy": 0.91015625,
|
| 674 |
+
"epoch": 0.47183098591549294,
|
| 675 |
+
"grad_norm": 1.1171875,
|
| 676 |
+
"learning_rate": 5e-05,
|
| 677 |
+
"loss": 0.894307553768158,
|
| 678 |
+
"mean_token_accuracy": 0.8422175049781799,
|
| 679 |
+
"num_tokens": 191309.0,
|
| 680 |
+
"step": 67
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"entropy": 1.203125,
|
| 684 |
+
"epoch": 0.4788732394366197,
|
| 685 |
+
"grad_norm": 1.0234375,
|
| 686 |
+
"learning_rate": 5e-05,
|
| 687 |
+
"loss": 1.1631497144699097,
|
| 688 |
+
"mean_token_accuracy": 0.7861995100975037,
|
| 689 |
+
"num_tokens": 194114.0,
|
| 690 |
+
"step": 68
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"entropy": 1.328125,
|
| 694 |
+
"epoch": 0.4859154929577465,
|
| 695 |
+
"grad_norm": 1.0390625,
|
| 696 |
+
"learning_rate": 5e-05,
|
| 697 |
+
"loss": 1.2809975147247314,
|
| 698 |
+
"mean_token_accuracy": 0.7543198466300964,
|
| 699 |
+
"num_tokens": 197305.0,
|
| 700 |
+
"step": 69
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"entropy": 1.015625,
|
| 704 |
+
"epoch": 0.49295774647887325,
|
| 705 |
+
"grad_norm": 1.3203125,
|
| 706 |
+
"learning_rate": 5e-05,
|
| 707 |
+
"loss": 1.0155205726623535,
|
| 708 |
+
"mean_token_accuracy": 0.834074079990387,
|
| 709 |
+
"num_tokens": 199338.0,
|
| 710 |
+
"step": 70
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"entropy": 0.83984375,
|
| 714 |
+
"epoch": 0.5,
|
| 715 |
+
"grad_norm": 1.1796875,
|
| 716 |
+
"learning_rate": 5e-05,
|
| 717 |
+
"loss": 0.7589200735092163,
|
| 718 |
+
"mean_token_accuracy": 0.8618420958518982,
|
| 719 |
+
"num_tokens": 201474.0,
|
| 720 |
+
"step": 71
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"entropy": 1.0859375,
|
| 724 |
+
"epoch": 0.5070422535211268,
|
| 725 |
+
"grad_norm": 1.2109375,
|
| 726 |
+
"learning_rate": 5e-05,
|
| 727 |
+
"loss": 1.0090736150741577,
|
| 728 |
+
"mean_token_accuracy": 0.8233413100242615,
|
| 729 |
+
"num_tokens": 203984.0,
|
| 730 |
+
"step": 72
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"entropy": 1.0703125,
|
| 734 |
+
"epoch": 0.5140845070422535,
|
| 735 |
+
"grad_norm": 0.9140625,
|
| 736 |
+
"learning_rate": 5e-05,
|
| 737 |
+
"loss": 1.046358346939087,
|
| 738 |
+
"mean_token_accuracy": 0.802097737789154,
|
| 739 |
+
"num_tokens": 207615.0,
|
| 740 |
+
"step": 73
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"entropy": 1.203125,
|
| 744 |
+
"epoch": 0.5211267605633803,
|
| 745 |
+
"grad_norm": 1.2109375,
|
| 746 |
+
"learning_rate": 5e-05,
|
| 747 |
+
"loss": 1.14384126663208,
|
| 748 |
+
"mean_token_accuracy": 0.7883082628250122,
|
| 749 |
+
"num_tokens": 209881.0,
|
| 750 |
+
"step": 74
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"entropy": 1.015625,
|
| 754 |
+
"epoch": 0.528169014084507,
|
| 755 |
+
"grad_norm": 0.98828125,
|
| 756 |
+
"learning_rate": 5e-05,
|
| 757 |
+
"loss": 0.9863561987876892,
|
| 758 |
+
"mean_token_accuracy": 0.8165745735168457,
|
| 759 |
+
"num_tokens": 212604.0,
|
| 760 |
+
"step": 75
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"entropy": 1.2421875,
|
| 764 |
+
"epoch": 0.5352112676056338,
|
| 765 |
+
"grad_norm": 1.1171875,
|
| 766 |
+
"learning_rate": 5e-05,
|
| 767 |
+
"loss": 1.1881471872329712,
|
| 768 |
+
"mean_token_accuracy": 0.7924298048019409,
|
| 769 |
+
"num_tokens": 215069.0,
|
| 770 |
+
"step": 76
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"entropy": 0.99609375,
|
| 774 |
+
"epoch": 0.5422535211267606,
|
| 775 |
+
"grad_norm": 0.86328125,
|
| 776 |
+
"learning_rate": 5e-05,
|
| 777 |
+
"loss": 0.9455610513687134,
|
| 778 |
+
"mean_token_accuracy": 0.8238698244094849,
|
| 779 |
+
"num_tokens": 217842.0,
|
| 780 |
+
"step": 77
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"entropy": 0.93359375,
|
| 784 |
+
"epoch": 0.5492957746478874,
|
| 785 |
+
"grad_norm": 1.015625,
|
| 786 |
+
"learning_rate": 5e-05,
|
| 787 |
+
"loss": 0.8940377235412598,
|
| 788 |
+
"mean_token_accuracy": 0.8325259685516357,
|
| 789 |
+
"num_tokens": 220740.0,
|
| 790 |
+
"step": 78
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"entropy": 1.078125,
|
| 794 |
+
"epoch": 0.5563380281690141,
|
| 795 |
+
"grad_norm": 1.0390625,
|
| 796 |
+
"learning_rate": 5e-05,
|
| 797 |
+
"loss": 1.072181224822998,
|
| 798 |
+
"mean_token_accuracy": 0.7974197864532471,
|
| 799 |
+
"num_tokens": 223616.0,
|
| 800 |
+
"step": 79
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"entropy": 0.9609375,
|
| 804 |
+
"epoch": 0.5633802816901409,
|
| 805 |
+
"grad_norm": 1.0625,
|
| 806 |
+
"learning_rate": 5e-05,
|
| 807 |
+
"loss": 0.968547523021698,
|
| 808 |
+
"mean_token_accuracy": 0.8254417181015015,
|
| 809 |
+
"num_tokens": 226454.0,
|
| 810 |
+
"step": 80
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"entropy": 1.0625,
|
| 814 |
+
"epoch": 0.5704225352112676,
|
| 815 |
+
"grad_norm": 1.0390625,
|
| 816 |
+
"learning_rate": 5e-05,
|
| 817 |
+
"loss": 1.029323935508728,
|
| 818 |
+
"mean_token_accuracy": 0.8091970086097717,
|
| 819 |
+
"num_tokens": 229376.0,
|
| 820 |
+
"step": 81
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"entropy": 1.296875,
|
| 824 |
+
"epoch": 0.5774647887323944,
|
| 825 |
+
"grad_norm": 1.078125,
|
| 826 |
+
"learning_rate": 5e-05,
|
| 827 |
+
"loss": 1.30299711227417,
|
| 828 |
+
"mean_token_accuracy": 0.7740401029586792,
|
| 829 |
+
"num_tokens": 232327.0,
|
| 830 |
+
"step": 82
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"entropy": 1.34375,
|
| 834 |
+
"epoch": 0.5845070422535211,
|
| 835 |
+
"grad_norm": 1.046875,
|
| 836 |
+
"learning_rate": 5e-05,
|
| 837 |
+
"loss": 1.3241432905197144,
|
| 838 |
+
"mean_token_accuracy": 0.7471368312835693,
|
| 839 |
+
"num_tokens": 235653.0,
|
| 840 |
+
"step": 83
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"entropy": 1.1953125,
|
| 844 |
+
"epoch": 0.5915492957746479,
|
| 845 |
+
"grad_norm": 1.0546875,
|
| 846 |
+
"learning_rate": 5e-05,
|
| 847 |
+
"loss": 1.1841082572937012,
|
| 848 |
+
"mean_token_accuracy": 0.7741320133209229,
|
| 849 |
+
"num_tokens": 238282.0,
|
| 850 |
+
"step": 84
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"entropy": 0.96875,
|
| 854 |
+
"epoch": 0.5985915492957746,
|
| 855 |
+
"grad_norm": 1.0,
|
| 856 |
+
"learning_rate": 5e-05,
|
| 857 |
+
"loss": 0.9051790833473206,
|
| 858 |
+
"mean_token_accuracy": 0.8358089327812195,
|
| 859 |
+
"num_tokens": 240781.0,
|
| 860 |
+
"step": 85
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"entropy": 0.81640625,
|
| 864 |
+
"epoch": 0.6056338028169014,
|
| 865 |
+
"grad_norm": 1.0625,
|
| 866 |
+
"learning_rate": 5e-05,
|
| 867 |
+
"loss": 0.8058979511260986,
|
| 868 |
+
"mean_token_accuracy": 0.8656787872314453,
|
| 869 |
+
"num_tokens": 242881.0,
|
| 870 |
+
"step": 86
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"entropy": 1.0234375,
|
| 874 |
+
"epoch": 0.6126760563380281,
|
| 875 |
+
"grad_norm": 0.89453125,
|
| 876 |
+
"learning_rate": 5e-05,
|
| 877 |
+
"loss": 0.9673476219177246,
|
| 878 |
+
"mean_token_accuracy": 0.8199172616004944,
|
| 879 |
+
"num_tokens": 246032.0,
|
| 880 |
+
"step": 87
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"entropy": 0.9609375,
|
| 884 |
+
"epoch": 0.6197183098591549,
|
| 885 |
+
"grad_norm": 1.2265625,
|
| 886 |
+
"learning_rate": 5e-05,
|
| 887 |
+
"loss": 0.9231598377227783,
|
| 888 |
+
"mean_token_accuracy": 0.833532452583313,
|
| 889 |
+
"num_tokens": 248551.0,
|
| 890 |
+
"step": 88
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"entropy": 0.9921875,
|
| 894 |
+
"epoch": 0.6267605633802817,
|
| 895 |
+
"grad_norm": 1.1015625,
|
| 896 |
+
"learning_rate": 5e-05,
|
| 897 |
+
"loss": 0.998257040977478,
|
| 898 |
+
"mean_token_accuracy": 0.8266341686248779,
|
| 899 |
+
"num_tokens": 251022.0,
|
| 900 |
+
"step": 89
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"entropy": 1.0546875,
|
| 904 |
+
"epoch": 0.6338028169014085,
|
| 905 |
+
"grad_norm": 1.015625,
|
| 906 |
+
"learning_rate": 5e-05,
|
| 907 |
+
"loss": 1.0252176523208618,
|
| 908 |
+
"mean_token_accuracy": 0.8216797709465027,
|
| 909 |
+
"num_tokens": 253447.0,
|
| 910 |
+
"step": 90
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"entropy": 1.1484375,
|
| 914 |
+
"epoch": 0.6408450704225352,
|
| 915 |
+
"grad_norm": 1.0,
|
| 916 |
+
"learning_rate": 5e-05,
|
| 917 |
+
"loss": 1.164228081703186,
|
| 918 |
+
"mean_token_accuracy": 0.7702828645706177,
|
| 919 |
+
"num_tokens": 256672.0,
|
| 920 |
+
"step": 91
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"entropy": 0.77734375,
|
| 924 |
+
"epoch": 0.647887323943662,
|
| 925 |
+
"grad_norm": 0.9609375,
|
| 926 |
+
"learning_rate": 5e-05,
|
| 927 |
+
"loss": 0.7661694288253784,
|
| 928 |
+
"mean_token_accuracy": 0.8448406457901001,
|
| 929 |
+
"num_tokens": 259316.0,
|
| 930 |
+
"step": 92
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"entropy": 1.15625,
|
| 934 |
+
"epoch": 0.6549295774647887,
|
| 935 |
+
"grad_norm": 0.98828125,
|
| 936 |
+
"learning_rate": 5e-05,
|
| 937 |
+
"loss": 1.1309537887573242,
|
| 938 |
+
"mean_token_accuracy": 0.7973990440368652,
|
| 939 |
+
"num_tokens": 262246.0,
|
| 940 |
+
"step": 93
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"entropy": 1.046875,
|
| 944 |
+
"epoch": 0.6619718309859155,
|
| 945 |
+
"grad_norm": 0.9296875,
|
| 946 |
+
"learning_rate": 5e-05,
|
| 947 |
+
"loss": 1.0406370162963867,
|
| 948 |
+
"mean_token_accuracy": 0.8099964261054993,
|
| 949 |
+
"num_tokens": 265075.0,
|
| 950 |
+
"step": 94
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"entropy": 1.0,
|
| 954 |
+
"epoch": 0.6690140845070423,
|
| 955 |
+
"grad_norm": 0.90234375,
|
| 956 |
+
"learning_rate": 5e-05,
|
| 957 |
+
"loss": 0.9478747844696045,
|
| 958 |
+
"mean_token_accuracy": 0.8167044520378113,
|
| 959 |
+
"num_tokens": 267729.0,
|
| 960 |
+
"step": 95
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"entropy": 1.125,
|
| 964 |
+
"epoch": 0.676056338028169,
|
| 965 |
+
"grad_norm": 1.0234375,
|
| 966 |
+
"learning_rate": 5e-05,
|
| 967 |
+
"loss": 1.0806983709335327,
|
| 968 |
+
"mean_token_accuracy": 0.8070116639137268,
|
| 969 |
+
"num_tokens": 270732.0,
|
| 970 |
+
"step": 96
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"entropy": 1.3359375,
|
| 974 |
+
"epoch": 0.6830985915492958,
|
| 975 |
+
"grad_norm": 0.95703125,
|
| 976 |
+
"learning_rate": 5e-05,
|
| 977 |
+
"loss": 1.2827727794647217,
|
| 978 |
+
"mean_token_accuracy": 0.7681026458740234,
|
| 979 |
+
"num_tokens": 274013.0,
|
| 980 |
+
"step": 97
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"entropy": 0.96875,
|
| 984 |
+
"epoch": 0.6901408450704225,
|
| 985 |
+
"grad_norm": 1.1484375,
|
| 986 |
+
"learning_rate": 5e-05,
|
| 987 |
+
"loss": 0.9255345463752747,
|
| 988 |
+
"mean_token_accuracy": 0.820858895778656,
|
| 989 |
+
"num_tokens": 276466.0,
|
| 990 |
+
"step": 98
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"entropy": 0.8828125,
|
| 994 |
+
"epoch": 0.6971830985915493,
|
| 995 |
+
"grad_norm": 0.828125,
|
| 996 |
+
"learning_rate": 5e-05,
|
| 997 |
+
"loss": 0.8368520736694336,
|
| 998 |
+
"mean_token_accuracy": 0.8478841185569763,
|
| 999 |
+
"num_tokens": 279097.0,
|
| 1000 |
+
"step": 99
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"entropy": 1.1875,
|
| 1004 |
+
"epoch": 0.704225352112676,
|
| 1005 |
+
"grad_norm": 0.91796875,
|
| 1006 |
+
"learning_rate": 5e-05,
|
| 1007 |
+
"loss": 1.1672570705413818,
|
| 1008 |
+
"mean_token_accuracy": 0.7689912915229797,
|
| 1009 |
+
"num_tokens": 282317.0,
|
| 1010 |
+
"step": 100
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"entropy": 1.53125,
|
| 1014 |
+
"epoch": 0.7112676056338029,
|
| 1015 |
+
"grad_norm": 1.0625,
|
| 1016 |
+
"learning_rate": 5e-05,
|
| 1017 |
+
"loss": 1.4877268075942993,
|
| 1018 |
+
"mean_token_accuracy": 0.7194092869758606,
|
| 1019 |
+
"num_tokens": 285643.0,
|
| 1020 |
+
"step": 101
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"entropy": 1.0,
|
| 1024 |
+
"epoch": 0.7183098591549296,
|
| 1025 |
+
"grad_norm": 1.0625,
|
| 1026 |
+
"learning_rate": 5e-05,
|
| 1027 |
+
"loss": 0.956112265586853,
|
| 1028 |
+
"mean_token_accuracy": 0.8256269693374634,
|
| 1029 |
+
"num_tokens": 288203.0,
|
| 1030 |
+
"step": 102
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"entropy": 1.1953125,
|
| 1034 |
+
"epoch": 0.7253521126760564,
|
| 1035 |
+
"grad_norm": 1.1328125,
|
| 1036 |
+
"learning_rate": 5e-05,
|
| 1037 |
+
"loss": 1.2514692544937134,
|
| 1038 |
+
"mean_token_accuracy": 0.771497905254364,
|
| 1039 |
+
"num_tokens": 291095.0,
|
| 1040 |
+
"step": 103
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"entropy": 1.2109375,
|
| 1044 |
+
"epoch": 0.7323943661971831,
|
| 1045 |
+
"grad_norm": 0.93359375,
|
| 1046 |
+
"learning_rate": 5e-05,
|
| 1047 |
+
"loss": 1.1472859382629395,
|
| 1048 |
+
"mean_token_accuracy": 0.7850604057312012,
|
| 1049 |
+
"num_tokens": 293834.0,
|
| 1050 |
+
"step": 104
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"entropy": 0.7734375,
|
| 1054 |
+
"epoch": 0.7394366197183099,
|
| 1055 |
+
"grad_norm": 1.28125,
|
| 1056 |
+
"learning_rate": 5e-05,
|
| 1057 |
+
"loss": 0.7254777550697327,
|
| 1058 |
+
"mean_token_accuracy": 0.8632119297981262,
|
| 1059 |
+
"num_tokens": 295984.0,
|
| 1060 |
+
"step": 105
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"entropy": 0.82421875,
|
| 1064 |
+
"epoch": 0.7464788732394366,
|
| 1065 |
+
"grad_norm": 1.078125,
|
| 1066 |
+
"learning_rate": 5e-05,
|
| 1067 |
+
"loss": 0.795278787612915,
|
| 1068 |
+
"mean_token_accuracy": 0.8470132946968079,
|
| 1069 |
+
"num_tokens": 298319.0,
|
| 1070 |
+
"step": 106
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"entropy": 1.09375,
|
| 1074 |
+
"epoch": 0.7535211267605634,
|
| 1075 |
+
"grad_norm": 0.87109375,
|
| 1076 |
+
"learning_rate": 5e-05,
|
| 1077 |
+
"loss": 1.0427790880203247,
|
| 1078 |
+
"mean_token_accuracy": 0.804815411567688,
|
| 1079 |
+
"num_tokens": 301442.0,
|
| 1080 |
+
"step": 107
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"entropy": 1.1171875,
|
| 1084 |
+
"epoch": 0.7605633802816901,
|
| 1085 |
+
"grad_norm": 0.95703125,
|
| 1086 |
+
"learning_rate": 5e-05,
|
| 1087 |
+
"loss": 1.104764699935913,
|
| 1088 |
+
"mean_token_accuracy": 0.8097593188285828,
|
| 1089 |
+
"num_tokens": 304483.0,
|
| 1090 |
+
"step": 108
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"entropy": 0.7421875,
|
| 1094 |
+
"epoch": 0.7676056338028169,
|
| 1095 |
+
"grad_norm": 1.0,
|
| 1096 |
+
"learning_rate": 5e-05,
|
| 1097 |
+
"loss": 0.6636262536048889,
|
| 1098 |
+
"mean_token_accuracy": 0.861122190952301,
|
| 1099 |
+
"num_tokens": 307004.0,
|
| 1100 |
+
"step": 109
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"entropy": 0.97265625,
|
| 1104 |
+
"epoch": 0.7746478873239436,
|
| 1105 |
+
"grad_norm": 0.90234375,
|
| 1106 |
+
"learning_rate": 5e-05,
|
| 1107 |
+
"loss": 0.9813203811645508,
|
| 1108 |
+
"mean_token_accuracy": 0.8114525079727173,
|
| 1109 |
+
"num_tokens": 309876.0,
|
| 1110 |
+
"step": 110
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"entropy": 1.09375,
|
| 1114 |
+
"epoch": 0.7816901408450704,
|
| 1115 |
+
"grad_norm": 1.078125,
|
| 1116 |
+
"learning_rate": 5e-05,
|
| 1117 |
+
"loss": 1.0708268880844116,
|
| 1118 |
+
"mean_token_accuracy": 0.7995486855506897,
|
| 1119 |
+
"num_tokens": 312543.0,
|
| 1120 |
+
"step": 111
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"entropy": 0.953125,
|
| 1124 |
+
"epoch": 0.7887323943661971,
|
| 1125 |
+
"grad_norm": 0.9453125,
|
| 1126 |
+
"learning_rate": 5e-05,
|
| 1127 |
+
"loss": 0.9298059344291687,
|
| 1128 |
+
"mean_token_accuracy": 0.8266475796699524,
|
| 1129 |
+
"num_tokens": 315343.0,
|
| 1130 |
+
"step": 112
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"entropy": 0.93359375,
|
| 1134 |
+
"epoch": 0.795774647887324,
|
| 1135 |
+
"grad_norm": 1.2109375,
|
| 1136 |
+
"learning_rate": 5e-05,
|
| 1137 |
+
"loss": 0.9146865606307983,
|
| 1138 |
+
"mean_token_accuracy": 0.8447837233543396,
|
| 1139 |
+
"num_tokens": 317709.0,
|
| 1140 |
+
"step": 113
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"entropy": 1.0390625,
|
| 1144 |
+
"epoch": 0.8028169014084507,
|
| 1145 |
+
"grad_norm": 1.015625,
|
| 1146 |
+
"learning_rate": 5e-05,
|
| 1147 |
+
"loss": 1.0023632049560547,
|
| 1148 |
+
"mean_token_accuracy": 0.8247303366661072,
|
| 1149 |
+
"num_tokens": 320313.0,
|
| 1150 |
+
"step": 114
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"entropy": 1.0234375,
|
| 1154 |
+
"epoch": 0.8098591549295775,
|
| 1155 |
+
"grad_norm": 0.9375,
|
| 1156 |
+
"learning_rate": 5e-05,
|
| 1157 |
+
"loss": 1.0119502544403076,
|
| 1158 |
+
"mean_token_accuracy": 0.8081619739532471,
|
| 1159 |
+
"num_tokens": 323433.0,
|
| 1160 |
+
"step": 115
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"entropy": 1.1484375,
|
| 1164 |
+
"epoch": 0.8169014084507042,
|
| 1165 |
+
"grad_norm": 1.09375,
|
| 1166 |
+
"learning_rate": 5e-05,
|
| 1167 |
+
"loss": 1.129537582397461,
|
| 1168 |
+
"mean_token_accuracy": 0.7915045619010925,
|
| 1169 |
+
"num_tokens": 326525.0,
|
| 1170 |
+
"step": 116
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"entropy": 0.828125,
|
| 1174 |
+
"epoch": 0.823943661971831,
|
| 1175 |
+
"grad_norm": 0.96875,
|
| 1176 |
+
"learning_rate": 5e-05,
|
| 1177 |
+
"loss": 0.8433165550231934,
|
| 1178 |
+
"mean_token_accuracy": 0.8409850001335144,
|
| 1179 |
+
"num_tokens": 328929.0,
|
| 1180 |
+
"step": 117
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"entropy": 1.0625,
|
| 1184 |
+
"epoch": 0.8309859154929577,
|
| 1185 |
+
"grad_norm": 1.0078125,
|
| 1186 |
+
"learning_rate": 5e-05,
|
| 1187 |
+
"loss": 1.0346664190292358,
|
| 1188 |
+
"mean_token_accuracy": 0.7833219170570374,
|
| 1189 |
+
"num_tokens": 331863.0,
|
| 1190 |
+
"step": 118
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"entropy": 0.98046875,
|
| 1194 |
+
"epoch": 0.8380281690140845,
|
| 1195 |
+
"grad_norm": 1.09375,
|
| 1196 |
+
"learning_rate": 5e-05,
|
| 1197 |
+
"loss": 0.950432300567627,
|
| 1198 |
+
"mean_token_accuracy": 0.8241989612579346,
|
| 1199 |
+
"num_tokens": 334243.0,
|
| 1200 |
+
"step": 119
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"entropy": 1.0546875,
|
| 1204 |
+
"epoch": 0.8450704225352113,
|
| 1205 |
+
"grad_norm": 0.8984375,
|
| 1206 |
+
"learning_rate": 5e-05,
|
| 1207 |
+
"loss": 1.0502727031707764,
|
| 1208 |
+
"mean_token_accuracy": 0.7917791604995728,
|
| 1209 |
+
"num_tokens": 337584.0,
|
| 1210 |
+
"step": 120
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"entropy": 0.89453125,
|
| 1214 |
+
"epoch": 0.852112676056338,
|
| 1215 |
+
"grad_norm": 0.9765625,
|
| 1216 |
+
"learning_rate": 5e-05,
|
| 1217 |
+
"loss": 0.9121181964874268,
|
| 1218 |
+
"mean_token_accuracy": 0.8309713006019592,
|
| 1219 |
+
"num_tokens": 340485.0,
|
| 1220 |
+
"step": 121
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"entropy": 0.88671875,
|
| 1224 |
+
"epoch": 0.8591549295774648,
|
| 1225 |
+
"grad_norm": 1.1328125,
|
| 1226 |
+
"learning_rate": 5e-05,
|
| 1227 |
+
"loss": 0.9219837188720703,
|
| 1228 |
+
"mean_token_accuracy": 0.8140540719032288,
|
| 1229 |
+
"num_tokens": 343268.0,
|
| 1230 |
+
"step": 122
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"entropy": 0.8515625,
|
| 1234 |
+
"epoch": 0.8661971830985915,
|
| 1235 |
+
"grad_norm": 1.1484375,
|
| 1236 |
+
"learning_rate": 5e-05,
|
| 1237 |
+
"loss": 0.7965174317359924,
|
| 1238 |
+
"mean_token_accuracy": 0.8607358932495117,
|
| 1239 |
+
"num_tokens": 345423.0,
|
| 1240 |
+
"step": 123
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"entropy": 0.9765625,
|
| 1244 |
+
"epoch": 0.8732394366197183,
|
| 1245 |
+
"grad_norm": 0.82421875,
|
| 1246 |
+
"learning_rate": 5e-05,
|
| 1247 |
+
"loss": 0.9578317999839783,
|
| 1248 |
+
"mean_token_accuracy": 0.8050633072853088,
|
| 1249 |
+
"num_tokens": 348591.0,
|
| 1250 |
+
"step": 124
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"entropy": 0.953125,
|
| 1254 |
+
"epoch": 0.8802816901408451,
|
| 1255 |
+
"grad_norm": 0.890625,
|
| 1256 |
+
"learning_rate": 5e-05,
|
| 1257 |
+
"loss": 0.9342446327209473,
|
| 1258 |
+
"mean_token_accuracy": 0.8261499404907227,
|
| 1259 |
+
"num_tokens": 351360.0,
|
| 1260 |
+
"step": 125
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"entropy": 1.0859375,
|
| 1264 |
+
"epoch": 0.8873239436619719,
|
| 1265 |
+
"grad_norm": 0.88671875,
|
| 1266 |
+
"learning_rate": 5e-05,
|
| 1267 |
+
"loss": 1.0417184829711914,
|
| 1268 |
+
"mean_token_accuracy": 0.7993496060371399,
|
| 1269 |
+
"num_tokens": 354443.0,
|
| 1270 |
+
"step": 126
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"entropy": 1.1875,
|
| 1274 |
+
"epoch": 0.8943661971830986,
|
| 1275 |
+
"grad_norm": 0.96484375,
|
| 1276 |
+
"learning_rate": 5e-05,
|
| 1277 |
+
"loss": 1.169641137123108,
|
| 1278 |
+
"mean_token_accuracy": 0.789621889591217,
|
| 1279 |
+
"num_tokens": 356937.0,
|
| 1280 |
+
"step": 127
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"entropy": 1.171875,
|
| 1284 |
+
"epoch": 0.9014084507042254,
|
| 1285 |
+
"grad_norm": 1.0546875,
|
| 1286 |
+
"learning_rate": 5e-05,
|
| 1287 |
+
"loss": 1.0926804542541504,
|
| 1288 |
+
"mean_token_accuracy": 0.78143709897995,
|
| 1289 |
+
"num_tokens": 359951.0,
|
| 1290 |
+
"step": 128
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"entropy": 0.90625,
|
| 1294 |
+
"epoch": 0.9084507042253521,
|
| 1295 |
+
"grad_norm": 1.046875,
|
| 1296 |
+
"learning_rate": 5e-05,
|
| 1297 |
+
"loss": 0.8831167817115784,
|
| 1298 |
+
"mean_token_accuracy": 0.8297297358512878,
|
| 1299 |
+
"num_tokens": 362549.0,
|
| 1300 |
+
"step": 129
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"entropy": 1.34375,
|
| 1304 |
+
"epoch": 0.9154929577464789,
|
| 1305 |
+
"grad_norm": 1.1640625,
|
| 1306 |
+
"learning_rate": 5e-05,
|
| 1307 |
+
"loss": 1.2982932329177856,
|
| 1308 |
+
"mean_token_accuracy": 0.7599736452102661,
|
| 1309 |
+
"num_tokens": 365590.0,
|
| 1310 |
+
"step": 130
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"entropy": 0.8515625,
|
| 1314 |
+
"epoch": 0.9225352112676056,
|
| 1315 |
+
"grad_norm": 1.140625,
|
| 1316 |
+
"learning_rate": 5e-05,
|
| 1317 |
+
"loss": 0.7926298975944519,
|
| 1318 |
+
"mean_token_accuracy": 0.8279292583465576,
|
| 1319 |
+
"num_tokens": 368312.0,
|
| 1320 |
+
"step": 131
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"entropy": 0.85546875,
|
| 1324 |
+
"epoch": 0.9295774647887324,
|
| 1325 |
+
"grad_norm": 1.0546875,
|
| 1326 |
+
"learning_rate": 5e-05,
|
| 1327 |
+
"loss": 0.876966655254364,
|
| 1328 |
+
"mean_token_accuracy": 0.8338558077812195,
|
| 1329 |
+
"num_tokens": 370872.0,
|
| 1330 |
+
"step": 132
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"entropy": 0.58203125,
|
| 1334 |
+
"epoch": 0.9366197183098591,
|
| 1335 |
+
"grad_norm": 1.09375,
|
| 1336 |
+
"learning_rate": 5e-05,
|
| 1337 |
+
"loss": 0.5232805609703064,
|
| 1338 |
+
"mean_token_accuracy": 0.9088762998580933,
|
| 1339 |
+
"num_tokens": 372998.0,
|
| 1340 |
+
"step": 133
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"entropy": 1.3125,
|
| 1344 |
+
"epoch": 0.9436619718309859,
|
| 1345 |
+
"grad_norm": 1.0859375,
|
| 1346 |
+
"learning_rate": 5e-05,
|
| 1347 |
+
"loss": 1.2858319282531738,
|
| 1348 |
+
"mean_token_accuracy": 0.764127790927887,
|
| 1349 |
+
"num_tokens": 375855.0,
|
| 1350 |
+
"step": 134
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"entropy": 0.9453125,
|
| 1354 |
+
"epoch": 0.9507042253521126,
|
| 1355 |
+
"grad_norm": 0.8125,
|
| 1356 |
+
"learning_rate": 5e-05,
|
| 1357 |
+
"loss": 0.9115528464317322,
|
| 1358 |
+
"mean_token_accuracy": 0.8201290965080261,
|
| 1359 |
+
"num_tokens": 379271.0,
|
| 1360 |
+
"step": 135
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"entropy": 1.0703125,
|
| 1364 |
+
"epoch": 0.9577464788732394,
|
| 1365 |
+
"grad_norm": 1.3828125,
|
| 1366 |
+
"learning_rate": 5e-05,
|
| 1367 |
+
"loss": 1.03813898563385,
|
| 1368 |
+
"mean_token_accuracy": 0.8119325637817383,
|
| 1369 |
+
"num_tokens": 381592.0,
|
| 1370 |
+
"step": 136
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"entropy": 1.109375,
|
| 1374 |
+
"epoch": 0.9647887323943662,
|
| 1375 |
+
"grad_norm": 0.953125,
|
| 1376 |
+
"learning_rate": 5e-05,
|
| 1377 |
+
"loss": 1.1146811246871948,
|
| 1378 |
+
"mean_token_accuracy": 0.790168285369873,
|
| 1379 |
+
"num_tokens": 384631.0,
|
| 1380 |
+
"step": 137
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"entropy": 1.1640625,
|
| 1384 |
+
"epoch": 0.971830985915493,
|
| 1385 |
+
"grad_norm": 1.0234375,
|
| 1386 |
+
"learning_rate": 5e-05,
|
| 1387 |
+
"loss": 1.173197627067566,
|
| 1388 |
+
"mean_token_accuracy": 0.7781087160110474,
|
| 1389 |
+
"num_tokens": 387325.0,
|
| 1390 |
+
"step": 138
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"entropy": 0.828125,
|
| 1394 |
+
"epoch": 0.9788732394366197,
|
| 1395 |
+
"grad_norm": 1.0546875,
|
| 1396 |
+
"learning_rate": 5e-05,
|
| 1397 |
+
"loss": 0.8895862698554993,
|
| 1398 |
+
"mean_token_accuracy": 0.8376736044883728,
|
| 1399 |
+
"num_tokens": 389637.0,
|
| 1400 |
+
"step": 139
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"entropy": 1.1953125,
|
| 1404 |
+
"epoch": 0.9859154929577465,
|
| 1405 |
+
"grad_norm": 0.8515625,
|
| 1406 |
+
"learning_rate": 5e-05,
|
| 1407 |
+
"loss": 1.2043834924697876,
|
| 1408 |
+
"mean_token_accuracy": 0.7506383061408997,
|
| 1409 |
+
"num_tokens": 393170.0,
|
| 1410 |
+
"step": 140
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"entropy": 0.87109375,
|
| 1414 |
+
"epoch": 0.9929577464788732,
|
| 1415 |
+
"grad_norm": 1.0546875,
|
| 1416 |
+
"learning_rate": 5e-05,
|
| 1417 |
+
"loss": 0.8573941588401794,
|
| 1418 |
+
"mean_token_accuracy": 0.8464516401290894,
|
| 1419 |
+
"num_tokens": 395503.0,
|
| 1420 |
+
"step": 141
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"entropy": 1.078125,
|
| 1424 |
+
"epoch": 1.0,
|
| 1425 |
+
"grad_norm": 0.90625,
|
| 1426 |
+
"learning_rate": 5e-05,
|
| 1427 |
+
"loss": 1.0620708465576172,
|
| 1428 |
+
"mean_token_accuracy": 0.7814034819602966,
|
| 1429 |
+
"num_tokens": 398361.0,
|
| 1430 |
+
"step": 142
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 1.0,
|
| 1434 |
+
"eval_entropy": 0.9779188368055556,
|
| 1435 |
+
"eval_loss": 0.9645262956619263,
|
| 1436 |
+
"eval_mean_token_accuracy": 0.8152486599153943,
|
| 1437 |
+
"eval_num_tokens": 398361.0,
|
| 1438 |
+
"eval_runtime": 3.7052,
|
| 1439 |
+
"eval_samples_per_second": 76.65,
|
| 1440 |
+
"eval_steps_per_second": 9.716,
|
| 1441 |
+
"step": 142
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"entropy": 0.953125,
|
| 1445 |
+
"epoch": 1.0070422535211268,
|
| 1446 |
+
"grad_norm": 1.125,
|
| 1447 |
+
"learning_rate": 5e-05,
|
| 1448 |
+
"loss": 0.9150600433349609,
|
| 1449 |
+
"mean_token_accuracy": 0.8252961039543152,
|
| 1450 |
+
"num_tokens": 400733.0,
|
| 1451 |
+
"step": 143
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"entropy": 0.8046875,
|
| 1455 |
+
"epoch": 1.0140845070422535,
|
| 1456 |
+
"grad_norm": 1.0703125,
|
| 1457 |
+
"learning_rate": 5e-05,
|
| 1458 |
+
"loss": 0.7507684230804443,
|
| 1459 |
+
"mean_token_accuracy": 0.8355586528778076,
|
| 1460 |
+
"num_tokens": 403587.0,
|
| 1461 |
+
"step": 144
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"entropy": 0.7265625,
|
| 1465 |
+
"epoch": 1.0211267605633803,
|
| 1466 |
+
"grad_norm": 0.87890625,
|
| 1467 |
+
"learning_rate": 5e-05,
|
| 1468 |
+
"loss": 0.7104012370109558,
|
| 1469 |
+
"mean_token_accuracy": 0.8799096941947937,
|
| 1470 |
+
"num_tokens": 405810.0,
|
| 1471 |
+
"step": 145
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"entropy": 0.8515625,
|
| 1475 |
+
"epoch": 1.028169014084507,
|
| 1476 |
+
"grad_norm": 0.8671875,
|
| 1477 |
+
"learning_rate": 5e-05,
|
| 1478 |
+
"loss": 0.7941129207611084,
|
| 1479 |
+
"mean_token_accuracy": 0.8590818643569946,
|
| 1480 |
+
"num_tokens": 408323.0,
|
| 1481 |
+
"step": 146
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"entropy": 1.140625,
|
| 1485 |
+
"epoch": 1.0352112676056338,
|
| 1486 |
+
"grad_norm": 0.90625,
|
| 1487 |
+
"learning_rate": 5e-05,
|
| 1488 |
+
"loss": 1.110805630683899,
|
| 1489 |
+
"mean_token_accuracy": 0.7991207242012024,
|
| 1490 |
+
"num_tokens": 411288.0,
|
| 1491 |
+
"step": 147
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"entropy": 1.078125,
|
| 1495 |
+
"epoch": 1.0422535211267605,
|
| 1496 |
+
"grad_norm": 0.875,
|
| 1497 |
+
"learning_rate": 5e-05,
|
| 1498 |
+
"loss": 1.096816897392273,
|
| 1499 |
+
"mean_token_accuracy": 0.793763279914856,
|
| 1500 |
+
"num_tokens": 414118.0,
|
| 1501 |
+
"step": 148
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"entropy": 0.79296875,
|
| 1505 |
+
"epoch": 1.0492957746478873,
|
| 1506 |
+
"grad_norm": 0.96484375,
|
| 1507 |
+
"learning_rate": 5e-05,
|
| 1508 |
+
"loss": 0.7629069089889526,
|
| 1509 |
+
"mean_token_accuracy": 0.8661789894104004,
|
| 1510 |
+
"num_tokens": 416450.0,
|
| 1511 |
+
"step": 149
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"entropy": 0.78125,
|
| 1515 |
+
"epoch": 1.056338028169014,
|
| 1516 |
+
"grad_norm": 1.03125,
|
| 1517 |
+
"learning_rate": 5e-05,
|
| 1518 |
+
"loss": 0.7574359178543091,
|
| 1519 |
+
"mean_token_accuracy": 0.8526899814605713,
|
| 1520 |
+
"num_tokens": 418800.0,
|
| 1521 |
+
"step": 150
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"entropy": 0.71484375,
|
| 1525 |
+
"epoch": 1.0633802816901408,
|
| 1526 |
+
"grad_norm": 0.83984375,
|
| 1527 |
+
"learning_rate": 5e-05,
|
| 1528 |
+
"loss": 0.6598190069198608,
|
| 1529 |
+
"mean_token_accuracy": 0.8710424900054932,
|
| 1530 |
+
"num_tokens": 421398.0,
|
| 1531 |
+
"step": 151
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"entropy": 0.96484375,
|
| 1535 |
+
"epoch": 1.0704225352112675,
|
| 1536 |
+
"grad_norm": 1.078125,
|
| 1537 |
+
"learning_rate": 5e-05,
|
| 1538 |
+
"loss": 0.9846295118331909,
|
| 1539 |
+
"mean_token_accuracy": 0.8020523190498352,
|
| 1540 |
+
"num_tokens": 424427.0,
|
| 1541 |
+
"step": 152
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"entropy": 0.9609375,
|
| 1545 |
+
"epoch": 1.0774647887323943,
|
| 1546 |
+
"grad_norm": 0.9765625,
|
| 1547 |
+
"learning_rate": 5e-05,
|
| 1548 |
+
"loss": 0.982389509677887,
|
| 1549 |
+
"mean_token_accuracy": 0.8070852160453796,
|
| 1550 |
+
"num_tokens": 427286.0,
|
| 1551 |
+
"step": 153
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"entropy": 0.7734375,
|
| 1555 |
+
"epoch": 1.084507042253521,
|
| 1556 |
+
"grad_norm": 0.92578125,
|
| 1557 |
+
"learning_rate": 5e-05,
|
| 1558 |
+
"loss": 0.7500929236412048,
|
| 1559 |
+
"mean_token_accuracy": 0.8648427724838257,
|
| 1560 |
+
"num_tokens": 429743.0,
|
| 1561 |
+
"step": 154
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"entropy": 0.578125,
|
| 1565 |
+
"epoch": 1.091549295774648,
|
| 1566 |
+
"grad_norm": 1.1484375,
|
| 1567 |
+
"learning_rate": 5e-05,
|
| 1568 |
+
"loss": 0.5609915256500244,
|
| 1569 |
+
"mean_token_accuracy": 0.9083379507064819,
|
| 1570 |
+
"num_tokens": 431562.0,
|
| 1571 |
+
"step": 155
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"entropy": 0.84375,
|
| 1575 |
+
"epoch": 1.0985915492957747,
|
| 1576 |
+
"grad_norm": 0.8828125,
|
| 1577 |
+
"learning_rate": 5e-05,
|
| 1578 |
+
"loss": 0.8129755854606628,
|
| 1579 |
+
"mean_token_accuracy": 0.83039790391922,
|
| 1580 |
+
"num_tokens": 434636.0,
|
| 1581 |
+
"step": 156
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"entropy": 1.1640625,
|
| 1585 |
+
"epoch": 1.1056338028169015,
|
| 1586 |
+
"grad_norm": 1.0625,
|
| 1587 |
+
"learning_rate": 5e-05,
|
| 1588 |
+
"loss": 1.157644271850586,
|
| 1589 |
+
"mean_token_accuracy": 0.7780413627624512,
|
| 1590 |
+
"num_tokens": 437595.0,
|
| 1591 |
+
"step": 157
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"entropy": 0.87890625,
|
| 1595 |
+
"epoch": 1.1126760563380282,
|
| 1596 |
+
"grad_norm": 1.046875,
|
| 1597 |
+
"learning_rate": 5e-05,
|
| 1598 |
+
"loss": 0.865077793598175,
|
| 1599 |
+
"mean_token_accuracy": 0.8418839573860168,
|
| 1600 |
+
"num_tokens": 439981.0,
|
| 1601 |
+
"step": 158
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"entropy": 1.078125,
|
| 1605 |
+
"epoch": 1.119718309859155,
|
| 1606 |
+
"grad_norm": 1.03125,
|
| 1607 |
+
"learning_rate": 5e-05,
|
| 1608 |
+
"loss": 1.0671403408050537,
|
| 1609 |
+
"mean_token_accuracy": 0.7895287871360779,
|
| 1610 |
+
"num_tokens": 442854.0,
|
| 1611 |
+
"step": 159
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"entropy": 0.6875,
|
| 1615 |
+
"epoch": 1.1267605633802817,
|
| 1616 |
+
"grad_norm": 0.91796875,
|
| 1617 |
+
"learning_rate": 5e-05,
|
| 1618 |
+
"loss": 0.6512315273284912,
|
| 1619 |
+
"mean_token_accuracy": 0.8766743540763855,
|
| 1620 |
+
"num_tokens": 445027.0,
|
| 1621 |
+
"step": 160
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"entropy": 1.125,
|
| 1625 |
+
"epoch": 1.1338028169014085,
|
| 1626 |
+
"grad_norm": 1.03125,
|
| 1627 |
+
"learning_rate": 5e-05,
|
| 1628 |
+
"loss": 1.1221463680267334,
|
| 1629 |
+
"mean_token_accuracy": 0.7976190447807312,
|
| 1630 |
+
"num_tokens": 447891.0,
|
| 1631 |
+
"step": 161
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"entropy": 0.90625,
|
| 1635 |
+
"epoch": 1.1408450704225352,
|
| 1636 |
+
"grad_norm": 0.90625,
|
| 1637 |
+
"learning_rate": 5e-05,
|
| 1638 |
+
"loss": 0.8583451509475708,
|
| 1639 |
+
"mean_token_accuracy": 0.8390721082687378,
|
| 1640 |
+
"num_tokens": 450658.0,
|
| 1641 |
+
"step": 162
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"entropy": 1.0078125,
|
| 1645 |
+
"epoch": 1.147887323943662,
|
| 1646 |
+
"grad_norm": 0.890625,
|
| 1647 |
+
"learning_rate": 5e-05,
|
| 1648 |
+
"loss": 1.0121499300003052,
|
| 1649 |
+
"mean_token_accuracy": 0.8126649260520935,
|
| 1650 |
+
"num_tokens": 454077.0,
|
| 1651 |
+
"step": 163
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"entropy": 1.109375,
|
| 1655 |
+
"epoch": 1.1549295774647887,
|
| 1656 |
+
"grad_norm": 0.8515625,
|
| 1657 |
+
"learning_rate": 5e-05,
|
| 1658 |
+
"loss": 1.0865602493286133,
|
| 1659 |
+
"mean_token_accuracy": 0.8106972575187683,
|
| 1660 |
+
"num_tokens": 457656.0,
|
| 1661 |
+
"step": 164
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"entropy": 0.83984375,
|
| 1665 |
+
"epoch": 1.1619718309859155,
|
| 1666 |
+
"grad_norm": 0.9296875,
|
| 1667 |
+
"learning_rate": 5e-05,
|
| 1668 |
+
"loss": 0.784701406955719,
|
| 1669 |
+
"mean_token_accuracy": 0.8401400446891785,
|
| 1670 |
+
"num_tokens": 460235.0,
|
| 1671 |
+
"step": 165
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"entropy": 0.87890625,
|
| 1675 |
+
"epoch": 1.1690140845070423,
|
| 1676 |
+
"grad_norm": 0.8984375,
|
| 1677 |
+
"learning_rate": 5e-05,
|
| 1678 |
+
"loss": 0.8821406364440918,
|
| 1679 |
+
"mean_token_accuracy": 0.8279282450675964,
|
| 1680 |
+
"num_tokens": 462864.0,
|
| 1681 |
+
"step": 166
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"entropy": 1.09375,
|
| 1685 |
+
"epoch": 1.176056338028169,
|
| 1686 |
+
"grad_norm": 0.9296875,
|
| 1687 |
+
"learning_rate": 5e-05,
|
| 1688 |
+
"loss": 1.071103572845459,
|
| 1689 |
+
"mean_token_accuracy": 0.7772451043128967,
|
| 1690 |
+
"num_tokens": 466001.0,
|
| 1691 |
+
"step": 167
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"entropy": 1.265625,
|
| 1695 |
+
"epoch": 1.1830985915492958,
|
| 1696 |
+
"grad_norm": 1.0390625,
|
| 1697 |
+
"learning_rate": 5e-05,
|
| 1698 |
+
"loss": 1.2070709466934204,
|
| 1699 |
+
"mean_token_accuracy": 0.7771183848381042,
|
| 1700 |
+
"num_tokens": 469042.0,
|
| 1701 |
+
"step": 168
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"entropy": 1.03125,
|
| 1705 |
+
"epoch": 1.1901408450704225,
|
| 1706 |
+
"grad_norm": 0.84375,
|
| 1707 |
+
"learning_rate": 5e-05,
|
| 1708 |
+
"loss": 0.9974339604377747,
|
| 1709 |
+
"mean_token_accuracy": 0.8080986142158508,
|
| 1710 |
+
"num_tokens": 471890.0,
|
| 1711 |
+
"step": 169
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"entropy": 0.90625,
|
| 1715 |
+
"epoch": 1.1971830985915493,
|
| 1716 |
+
"grad_norm": 0.859375,
|
| 1717 |
+
"learning_rate": 5e-05,
|
| 1718 |
+
"loss": 0.8776722550392151,
|
| 1719 |
+
"mean_token_accuracy": 0.836818516254425,
|
| 1720 |
+
"num_tokens": 474576.0,
|
| 1721 |
+
"step": 170
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"entropy": 0.96484375,
|
| 1725 |
+
"epoch": 1.204225352112676,
|
| 1726 |
+
"grad_norm": 0.90625,
|
| 1727 |
+
"learning_rate": 5e-05,
|
| 1728 |
+
"loss": 0.9560971856117249,
|
| 1729 |
+
"mean_token_accuracy": 0.8277040123939514,
|
| 1730 |
+
"num_tokens": 477219.0,
|
| 1731 |
+
"step": 171
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"entropy": 1.1328125,
|
| 1735 |
+
"epoch": 1.2112676056338028,
|
| 1736 |
+
"grad_norm": 0.9375,
|
| 1737 |
+
"learning_rate": 5e-05,
|
| 1738 |
+
"loss": 1.1175223588943481,
|
| 1739 |
+
"mean_token_accuracy": 0.7786499261856079,
|
| 1740 |
+
"num_tokens": 480412.0,
|
| 1741 |
+
"step": 172
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"entropy": 1.3984375,
|
| 1745 |
+
"epoch": 1.2183098591549295,
|
| 1746 |
+
"grad_norm": 1.109375,
|
| 1747 |
+
"learning_rate": 5e-05,
|
| 1748 |
+
"loss": 1.3395131826400757,
|
| 1749 |
+
"mean_token_accuracy": 0.751716673374176,
|
| 1750 |
+
"num_tokens": 483187.0,
|
| 1751 |
+
"step": 173
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"entropy": 0.9765625,
|
| 1755 |
+
"epoch": 1.2253521126760563,
|
| 1756 |
+
"grad_norm": 0.90234375,
|
| 1757 |
+
"learning_rate": 5e-05,
|
| 1758 |
+
"loss": 1.004224419593811,
|
| 1759 |
+
"mean_token_accuracy": 0.8214024901390076,
|
| 1760 |
+
"num_tokens": 485933.0,
|
| 1761 |
+
"step": 174
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"entropy": 0.890625,
|
| 1765 |
+
"epoch": 1.232394366197183,
|
| 1766 |
+
"grad_norm": 0.7734375,
|
| 1767 |
+
"learning_rate": 5e-05,
|
| 1768 |
+
"loss": 0.8696950674057007,
|
| 1769 |
+
"mean_token_accuracy": 0.8322895169258118,
|
| 1770 |
+
"num_tokens": 488815.0,
|
| 1771 |
+
"step": 175
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"entropy": 1.109375,
|
| 1775 |
+
"epoch": 1.2394366197183098,
|
| 1776 |
+
"grad_norm": 0.91015625,
|
| 1777 |
+
"learning_rate": 5e-05,
|
| 1778 |
+
"loss": 1.0984435081481934,
|
| 1779 |
+
"mean_token_accuracy": 0.7843074202537537,
|
| 1780 |
+
"num_tokens": 491920.0,
|
| 1781 |
+
"step": 176
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"entropy": 0.95703125,
|
| 1785 |
+
"epoch": 1.2464788732394365,
|
| 1786 |
+
"grad_norm": 1.0234375,
|
| 1787 |
+
"learning_rate": 5e-05,
|
| 1788 |
+
"loss": 0.9243696331977844,
|
| 1789 |
+
"mean_token_accuracy": 0.8328244090080261,
|
| 1790 |
+
"num_tokens": 494548.0,
|
| 1791 |
+
"step": 177
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"entropy": 1.1171875,
|
| 1795 |
+
"epoch": 1.2535211267605635,
|
| 1796 |
+
"grad_norm": 1.015625,
|
| 1797 |
+
"learning_rate": 5e-05,
|
| 1798 |
+
"loss": 1.0856032371520996,
|
| 1799 |
+
"mean_token_accuracy": 0.7994632124900818,
|
| 1800 |
+
"num_tokens": 497164.0,
|
| 1801 |
+
"step": 178
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"entropy": 0.8046875,
|
| 1805 |
+
"epoch": 1.26056338028169,
|
| 1806 |
+
"grad_norm": 1.1484375,
|
| 1807 |
+
"learning_rate": 5e-05,
|
| 1808 |
+
"loss": 0.7936550378799438,
|
| 1809 |
+
"mean_token_accuracy": 0.8561053276062012,
|
| 1810 |
+
"num_tokens": 499375.0,
|
| 1811 |
+
"step": 179
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"entropy": 1.3125,
|
| 1815 |
+
"epoch": 1.267605633802817,
|
| 1816 |
+
"grad_norm": 0.94921875,
|
| 1817 |
+
"learning_rate": 5e-05,
|
| 1818 |
+
"loss": 1.3021771907806396,
|
| 1819 |
+
"mean_token_accuracy": 0.7595491409301758,
|
| 1820 |
+
"num_tokens": 502577.0,
|
| 1821 |
+
"step": 180
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"entropy": 0.83203125,
|
| 1825 |
+
"epoch": 1.2746478873239437,
|
| 1826 |
+
"grad_norm": 0.95703125,
|
| 1827 |
+
"learning_rate": 5e-05,
|
| 1828 |
+
"loss": 0.8171048760414124,
|
| 1829 |
+
"mean_token_accuracy": 0.8462969660758972,
|
| 1830 |
+
"num_tokens": 505272.0,
|
| 1831 |
+
"step": 181
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"entropy": 0.55078125,
|
| 1835 |
+
"epoch": 1.2816901408450705,
|
| 1836 |
+
"grad_norm": 1.15625,
|
| 1837 |
+
"learning_rate": 5e-05,
|
| 1838 |
+
"loss": 0.536847710609436,
|
| 1839 |
+
"mean_token_accuracy": 0.9226457476615906,
|
| 1840 |
+
"num_tokens": 507064.0,
|
| 1841 |
+
"step": 182
|
| 1842 |
+
},
|
| 1843 |
+
{
|
| 1844 |
+
"entropy": 1.0078125,
|
| 1845 |
+
"epoch": 1.2887323943661972,
|
| 1846 |
+
"grad_norm": 0.875,
|
| 1847 |
+
"learning_rate": 5e-05,
|
| 1848 |
+
"loss": 1.0221246480941772,
|
| 1849 |
+
"mean_token_accuracy": 0.8086583018302917,
|
| 1850 |
+
"num_tokens": 510098.0,
|
| 1851 |
+
"step": 183
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"entropy": 1.046875,
|
| 1855 |
+
"epoch": 1.295774647887324,
|
| 1856 |
+
"grad_norm": 0.9296875,
|
| 1857 |
+
"learning_rate": 5e-05,
|
| 1858 |
+
"loss": 1.0499778985977173,
|
| 1859 |
+
"mean_token_accuracy": 0.7971059083938599,
|
| 1860 |
+
"num_tokens": 513354.0,
|
| 1861 |
+
"step": 184
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"entropy": 0.91796875,
|
| 1865 |
+
"epoch": 1.3028169014084507,
|
| 1866 |
+
"grad_norm": 0.8046875,
|
| 1867 |
+
"learning_rate": 5e-05,
|
| 1868 |
+
"loss": 0.908235490322113,
|
| 1869 |
+
"mean_token_accuracy": 0.8296475410461426,
|
| 1870 |
+
"num_tokens": 516256.0,
|
| 1871 |
+
"step": 185
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"entropy": 1.0078125,
|
| 1875 |
+
"epoch": 1.3098591549295775,
|
| 1876 |
+
"grad_norm": 0.8359375,
|
| 1877 |
+
"learning_rate": 5e-05,
|
| 1878 |
+
"loss": 0.9736582040786743,
|
| 1879 |
+
"mean_token_accuracy": 0.8088183403015137,
|
| 1880 |
+
"num_tokens": 519099.0,
|
| 1881 |
+
"step": 186
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"entropy": 1.03125,
|
| 1885 |
+
"epoch": 1.3169014084507042,
|
| 1886 |
+
"grad_norm": 0.87890625,
|
| 1887 |
+
"learning_rate": 5e-05,
|
| 1888 |
+
"loss": 0.9836978316307068,
|
| 1889 |
+
"mean_token_accuracy": 0.8049733638763428,
|
| 1890 |
+
"num_tokens": 521922.0,
|
| 1891 |
+
"step": 187
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"entropy": 1.2265625,
|
| 1895 |
+
"epoch": 1.323943661971831,
|
| 1896 |
+
"grad_norm": 0.9609375,
|
| 1897 |
+
"learning_rate": 5e-05,
|
| 1898 |
+
"loss": 1.236617088317871,
|
| 1899 |
+
"mean_token_accuracy": 0.7495366930961609,
|
| 1900 |
+
"num_tokens": 525707.0,
|
| 1901 |
+
"step": 188
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"entropy": 0.7265625,
|
| 1905 |
+
"epoch": 1.3309859154929577,
|
| 1906 |
+
"grad_norm": 0.78515625,
|
| 1907 |
+
"learning_rate": 5e-05,
|
| 1908 |
+
"loss": 0.6896946430206299,
|
| 1909 |
+
"mean_token_accuracy": 0.8708636164665222,
|
| 1910 |
+
"num_tokens": 528193.0,
|
| 1911 |
+
"step": 189
|
| 1912 |
+
},
|
| 1913 |
+
{
|
| 1914 |
+
"entropy": 0.7890625,
|
| 1915 |
+
"epoch": 1.3380281690140845,
|
| 1916 |
+
"grad_norm": 0.8515625,
|
| 1917 |
+
"learning_rate": 5e-05,
|
| 1918 |
+
"loss": 0.7507163286209106,
|
| 1919 |
+
"mean_token_accuracy": 0.852664589881897,
|
| 1920 |
+
"num_tokens": 530753.0,
|
| 1921 |
+
"step": 190
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"entropy": 1.1328125,
|
| 1925 |
+
"epoch": 1.3450704225352113,
|
| 1926 |
+
"grad_norm": 0.859375,
|
| 1927 |
+
"learning_rate": 5e-05,
|
| 1928 |
+
"loss": 1.1465418338775635,
|
| 1929 |
+
"mean_token_accuracy": 0.7830012440681458,
|
| 1930 |
+
"num_tokens": 533973.0,
|
| 1931 |
+
"step": 191
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"entropy": 0.94140625,
|
| 1935 |
+
"epoch": 1.352112676056338,
|
| 1936 |
+
"grad_norm": 0.9140625,
|
| 1937 |
+
"learning_rate": 5e-05,
|
| 1938 |
+
"loss": 0.9713115692138672,
|
| 1939 |
+
"mean_token_accuracy": 0.8168444037437439,
|
| 1940 |
+
"num_tokens": 536700.0,
|
| 1941 |
+
"step": 192
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"entropy": 0.98046875,
|
| 1945 |
+
"epoch": 1.3591549295774648,
|
| 1946 |
+
"grad_norm": 0.91796875,
|
| 1947 |
+
"learning_rate": 5e-05,
|
| 1948 |
+
"loss": 0.9194943904876709,
|
| 1949 |
+
"mean_token_accuracy": 0.812965989112854,
|
| 1950 |
+
"num_tokens": 539793.0,
|
| 1951 |
+
"step": 193
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"entropy": 1.0234375,
|
| 1955 |
+
"epoch": 1.3661971830985915,
|
| 1956 |
+
"grad_norm": 0.83203125,
|
| 1957 |
+
"learning_rate": 5e-05,
|
| 1958 |
+
"loss": 0.9923701286315918,
|
| 1959 |
+
"mean_token_accuracy": 0.8061100840568542,
|
| 1960 |
+
"num_tokens": 543107.0,
|
| 1961 |
+
"step": 194
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"entropy": 0.77734375,
|
| 1965 |
+
"epoch": 1.3732394366197183,
|
| 1966 |
+
"grad_norm": 0.8984375,
|
| 1967 |
+
"learning_rate": 5e-05,
|
| 1968 |
+
"loss": 0.7561295032501221,
|
| 1969 |
+
"mean_token_accuracy": 0.8627303242683411,
|
| 1970 |
+
"num_tokens": 545774.0,
|
| 1971 |
+
"step": 195
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"entropy": 0.9765625,
|
| 1975 |
+
"epoch": 1.380281690140845,
|
| 1976 |
+
"grad_norm": 0.94921875,
|
| 1977 |
+
"learning_rate": 5e-05,
|
| 1978 |
+
"loss": 0.9733573198318481,
|
| 1979 |
+
"mean_token_accuracy": 0.812156617641449,
|
| 1980 |
+
"num_tokens": 548694.0,
|
| 1981 |
+
"step": 196
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"entropy": 0.8203125,
|
| 1985 |
+
"epoch": 1.3873239436619718,
|
| 1986 |
+
"grad_norm": 0.921875,
|
| 1987 |
+
"learning_rate": 5e-05,
|
| 1988 |
+
"loss": 0.7888185977935791,
|
| 1989 |
+
"mean_token_accuracy": 0.8413514494895935,
|
| 1990 |
+
"num_tokens": 551425.0,
|
| 1991 |
+
"step": 197
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"entropy": 1.046875,
|
| 1995 |
+
"epoch": 1.3943661971830985,
|
| 1996 |
+
"grad_norm": 1.1484375,
|
| 1997 |
+
"learning_rate": 5e-05,
|
| 1998 |
+
"loss": 1.1070986986160278,
|
| 1999 |
+
"mean_token_accuracy": 0.8009437918663025,
|
| 2000 |
+
"num_tokens": 553764.0,
|
| 2001 |
+
"step": 198
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"entropy": 1.0,
|
| 2005 |
+
"epoch": 1.4014084507042255,
|
| 2006 |
+
"grad_norm": 0.95703125,
|
| 2007 |
+
"learning_rate": 5e-05,
|
| 2008 |
+
"loss": 0.9817038178443909,
|
| 2009 |
+
"mean_token_accuracy": 0.8074533939361572,
|
| 2010 |
+
"num_tokens": 556348.0,
|
| 2011 |
+
"step": 199
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"entropy": 0.90234375,
|
| 2015 |
+
"epoch": 1.408450704225352,
|
| 2016 |
+
"grad_norm": 0.90234375,
|
| 2017 |
+
"learning_rate": 5e-05,
|
| 2018 |
+
"loss": 0.9156808853149414,
|
| 2019 |
+
"mean_token_accuracy": 0.8265197277069092,
|
| 2020 |
+
"num_tokens": 559169.0,
|
| 2021 |
+
"step": 200
|
| 2022 |
+
},
|
| 2023 |
+
{
|
| 2024 |
+
"entropy": 0.95703125,
|
| 2025 |
+
"epoch": 1.415492957746479,
|
| 2026 |
+
"grad_norm": 0.8984375,
|
| 2027 |
+
"learning_rate": 5e-05,
|
| 2028 |
+
"loss": 0.9911283254623413,
|
| 2029 |
+
"mean_token_accuracy": 0.8143333196640015,
|
| 2030 |
+
"num_tokens": 562177.0,
|
| 2031 |
+
"step": 201
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"entropy": 0.93359375,
|
| 2035 |
+
"epoch": 1.4225352112676055,
|
| 2036 |
+
"grad_norm": 0.85546875,
|
| 2037 |
+
"learning_rate": 5e-05,
|
| 2038 |
+
"loss": 0.9385692477226257,
|
| 2039 |
+
"mean_token_accuracy": 0.8237070441246033,
|
| 2040 |
+
"num_tokens": 564834.0,
|
| 2041 |
+
"step": 202
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"entropy": 0.98828125,
|
| 2045 |
+
"epoch": 1.4295774647887325,
|
| 2046 |
+
"grad_norm": 0.890625,
|
| 2047 |
+
"learning_rate": 5e-05,
|
| 2048 |
+
"loss": 0.9537026882171631,
|
| 2049 |
+
"mean_token_accuracy": 0.8196093440055847,
|
| 2050 |
+
"num_tokens": 567453.0,
|
| 2051 |
+
"step": 203
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"entropy": 1.1171875,
|
| 2055 |
+
"epoch": 1.436619718309859,
|
| 2056 |
+
"grad_norm": 0.88671875,
|
| 2057 |
+
"learning_rate": 5e-05,
|
| 2058 |
+
"loss": 1.1185535192489624,
|
| 2059 |
+
"mean_token_accuracy": 0.7943561673164368,
|
| 2060 |
+
"num_tokens": 570544.0,
|
| 2061 |
+
"step": 204
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"entropy": 1.2265625,
|
| 2065 |
+
"epoch": 1.443661971830986,
|
| 2066 |
+
"grad_norm": 0.87109375,
|
| 2067 |
+
"learning_rate": 5e-05,
|
| 2068 |
+
"loss": 1.2277859449386597,
|
| 2069 |
+
"mean_token_accuracy": 0.7642069458961487,
|
| 2070 |
+
"num_tokens": 574089.0,
|
| 2071 |
+
"step": 205
|
| 2072 |
+
},
|
| 2073 |
+
{
|
| 2074 |
+
"entropy": 1.046875,
|
| 2075 |
+
"epoch": 1.4507042253521127,
|
| 2076 |
+
"grad_norm": 0.92578125,
|
| 2077 |
+
"learning_rate": 5e-05,
|
| 2078 |
+
"loss": 1.0433013439178467,
|
| 2079 |
+
"mean_token_accuracy": 0.7854348421096802,
|
| 2080 |
+
"num_tokens": 577420.0,
|
| 2081 |
+
"step": 206
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"entropy": 0.8515625,
|
| 2085 |
+
"epoch": 1.4577464788732395,
|
| 2086 |
+
"grad_norm": 0.94140625,
|
| 2087 |
+
"learning_rate": 5e-05,
|
| 2088 |
+
"loss": 0.8301782011985779,
|
| 2089 |
+
"mean_token_accuracy": 0.8421682119369507,
|
| 2090 |
+
"num_tokens": 579937.0,
|
| 2091 |
+
"step": 207
|
| 2092 |
+
},
|
| 2093 |
+
{
|
| 2094 |
+
"entropy": 0.78125,
|
| 2095 |
+
"epoch": 1.4647887323943662,
|
| 2096 |
+
"grad_norm": 0.8984375,
|
| 2097 |
+
"learning_rate": 5e-05,
|
| 2098 |
+
"loss": 0.7898612022399902,
|
| 2099 |
+
"mean_token_accuracy": 0.8362168669700623,
|
| 2100 |
+
"num_tokens": 582546.0,
|
| 2101 |
+
"step": 208
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"entropy": 0.984375,
|
| 2105 |
+
"epoch": 1.471830985915493,
|
| 2106 |
+
"grad_norm": 0.82421875,
|
| 2107 |
+
"learning_rate": 5e-05,
|
| 2108 |
+
"loss": 0.9981078505516052,
|
| 2109 |
+
"mean_token_accuracy": 0.8020643591880798,
|
| 2110 |
+
"num_tokens": 585848.0,
|
| 2111 |
+
"step": 209
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"entropy": 0.94140625,
|
| 2115 |
+
"epoch": 1.4788732394366197,
|
| 2116 |
+
"grad_norm": 0.91796875,
|
| 2117 |
+
"learning_rate": 5e-05,
|
| 2118 |
+
"loss": 0.938367486000061,
|
| 2119 |
+
"mean_token_accuracy": 0.8073078393936157,
|
| 2120 |
+
"num_tokens": 588757.0,
|
| 2121 |
+
"step": 210
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"entropy": 0.96484375,
|
| 2125 |
+
"epoch": 1.4859154929577465,
|
| 2126 |
+
"grad_norm": 0.84765625,
|
| 2127 |
+
"learning_rate": 5e-05,
|
| 2128 |
+
"loss": 0.9669017791748047,
|
| 2129 |
+
"mean_token_accuracy": 0.8119603395462036,
|
| 2130 |
+
"num_tokens": 591892.0,
|
| 2131 |
+
"step": 211
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"entropy": 1.390625,
|
| 2135 |
+
"epoch": 1.4929577464788732,
|
| 2136 |
+
"grad_norm": 0.875,
|
| 2137 |
+
"learning_rate": 5e-05,
|
| 2138 |
+
"loss": 1.3462798595428467,
|
| 2139 |
+
"mean_token_accuracy": 0.7334238886833191,
|
| 2140 |
+
"num_tokens": 595580.0,
|
| 2141 |
+
"step": 212
|
| 2142 |
+
},
|
| 2143 |
+
{
|
| 2144 |
+
"entropy": 1.109375,
|
| 2145 |
+
"epoch": 1.5,
|
| 2146 |
+
"grad_norm": 0.93359375,
|
| 2147 |
+
"learning_rate": 5e-05,
|
| 2148 |
+
"loss": 1.068884015083313,
|
| 2149 |
+
"mean_token_accuracy": 0.7929913401603699,
|
| 2150 |
+
"num_tokens": 598356.0,
|
| 2151 |
+
"step": 213
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"entropy": 1.1015625,
|
| 2155 |
+
"epoch": 1.5070422535211268,
|
| 2156 |
+
"grad_norm": 0.87890625,
|
| 2157 |
+
"learning_rate": 5e-05,
|
| 2158 |
+
"loss": 1.0299583673477173,
|
| 2159 |
+
"mean_token_accuracy": 0.7973333597183228,
|
| 2160 |
+
"num_tokens": 601364.0,
|
| 2161 |
+
"step": 214
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"entropy": 1.0703125,
|
| 2165 |
+
"epoch": 1.5140845070422535,
|
| 2166 |
+
"grad_norm": 0.9140625,
|
| 2167 |
+
"learning_rate": 5e-05,
|
| 2168 |
+
"loss": 1.041951298713684,
|
| 2169 |
+
"mean_token_accuracy": 0.8040791153907776,
|
| 2170 |
+
"num_tokens": 604608.0,
|
| 2171 |
+
"step": 215
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"entropy": 0.8671875,
|
| 2175 |
+
"epoch": 1.5211267605633803,
|
| 2176 |
+
"grad_norm": 0.84375,
|
| 2177 |
+
"learning_rate": 5e-05,
|
| 2178 |
+
"loss": 0.8255084156990051,
|
| 2179 |
+
"mean_token_accuracy": 0.8424015045166016,
|
| 2180 |
+
"num_tokens": 607281.0,
|
| 2181 |
+
"step": 216
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"entropy": 0.76953125,
|
| 2185 |
+
"epoch": 1.528169014084507,
|
| 2186 |
+
"grad_norm": 0.89453125,
|
| 2187 |
+
"learning_rate": 5e-05,
|
| 2188 |
+
"loss": 0.7329130172729492,
|
| 2189 |
+
"mean_token_accuracy": 0.8575052618980408,
|
| 2190 |
+
"num_tokens": 609654.0,
|
| 2191 |
+
"step": 217
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"entropy": 1.0859375,
|
| 2195 |
+
"epoch": 1.5352112676056338,
|
| 2196 |
+
"grad_norm": 0.8359375,
|
| 2197 |
+
"learning_rate": 5e-05,
|
| 2198 |
+
"loss": 1.0870012044906616,
|
| 2199 |
+
"mean_token_accuracy": 0.7985919713973999,
|
| 2200 |
+
"num_tokens": 612929.0,
|
| 2201 |
+
"step": 218
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"entropy": 1.2421875,
|
| 2205 |
+
"epoch": 1.5422535211267605,
|
| 2206 |
+
"grad_norm": 0.90234375,
|
| 2207 |
+
"learning_rate": 5e-05,
|
| 2208 |
+
"loss": 1.22186279296875,
|
| 2209 |
+
"mean_token_accuracy": 0.7604916095733643,
|
| 2210 |
+
"num_tokens": 616273.0,
|
| 2211 |
+
"step": 219
|
| 2212 |
+
},
|
| 2213 |
+
{
|
| 2214 |
+
"entropy": 1.1640625,
|
| 2215 |
+
"epoch": 1.5492957746478875,
|
| 2216 |
+
"grad_norm": 0.87109375,
|
| 2217 |
+
"learning_rate": 5e-05,
|
| 2218 |
+
"loss": 1.1332319974899292,
|
| 2219 |
+
"mean_token_accuracy": 0.7764706015586853,
|
| 2220 |
+
"num_tokens": 619426.0,
|
| 2221 |
+
"step": 220
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"entropy": 1.0546875,
|
| 2225 |
+
"epoch": 1.556338028169014,
|
| 2226 |
+
"grad_norm": 0.8984375,
|
| 2227 |
+
"learning_rate": 5e-05,
|
| 2228 |
+
"loss": 1.0643455982208252,
|
| 2229 |
+
"mean_token_accuracy": 0.7792642116546631,
|
| 2230 |
+
"num_tokens": 622424.0,
|
| 2231 |
+
"step": 221
|
| 2232 |
+
},
|
| 2233 |
+
{
|
| 2234 |
+
"entropy": 0.6875,
|
| 2235 |
+
"epoch": 1.563380281690141,
|
| 2236 |
+
"grad_norm": 0.9765625,
|
| 2237 |
+
"learning_rate": 5e-05,
|
| 2238 |
+
"loss": 0.6475476622581482,
|
| 2239 |
+
"mean_token_accuracy": 0.8877917528152466,
|
| 2240 |
+
"num_tokens": 624660.0,
|
| 2241 |
+
"step": 222
|
| 2242 |
+
},
|
| 2243 |
+
{
|
| 2244 |
+
"entropy": 0.7890625,
|
| 2245 |
+
"epoch": 1.5704225352112675,
|
| 2246 |
+
"grad_norm": 1.0390625,
|
| 2247 |
+
"learning_rate": 5e-05,
|
| 2248 |
+
"loss": 0.7602906823158264,
|
| 2249 |
+
"mean_token_accuracy": 0.8367270231246948,
|
| 2250 |
+
"num_tokens": 627320.0,
|
| 2251 |
+
"step": 223
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"entropy": 0.9140625,
|
| 2255 |
+
"epoch": 1.5774647887323945,
|
| 2256 |
+
"grad_norm": 0.8359375,
|
| 2257 |
+
"learning_rate": 5e-05,
|
| 2258 |
+
"loss": 0.89766526222229,
|
| 2259 |
+
"mean_token_accuracy": 0.8361436724662781,
|
| 2260 |
+
"num_tokens": 630056.0,
|
| 2261 |
+
"step": 224
|
| 2262 |
+
},
|
| 2263 |
+
{
|
| 2264 |
+
"entropy": 1.046875,
|
| 2265 |
+
"epoch": 1.584507042253521,
|
| 2266 |
+
"grad_norm": 0.85546875,
|
| 2267 |
+
"learning_rate": 5e-05,
|
| 2268 |
+
"loss": 1.0017104148864746,
|
| 2269 |
+
"mean_token_accuracy": 0.800404965877533,
|
| 2270 |
+
"num_tokens": 633521.0,
|
| 2271 |
+
"step": 225
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"entropy": 0.3515625,
|
| 2275 |
+
"epoch": 1.591549295774648,
|
| 2276 |
+
"grad_norm": 0.9765625,
|
| 2277 |
+
"learning_rate": 5e-05,
|
| 2278 |
+
"loss": 0.3341764211654663,
|
| 2279 |
+
"mean_token_accuracy": 0.9567307829856873,
|
| 2280 |
+
"num_tokens": 635193.0,
|
| 2281 |
+
"step": 226
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"entropy": 0.95703125,
|
| 2285 |
+
"epoch": 1.5985915492957745,
|
| 2286 |
+
"grad_norm": 0.8515625,
|
| 2287 |
+
"learning_rate": 5e-05,
|
| 2288 |
+
"loss": 0.9153825640678406,
|
| 2289 |
+
"mean_token_accuracy": 0.8223153948783875,
|
| 2290 |
+
"num_tokens": 638302.0,
|
| 2291 |
+
"step": 227
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"entropy": 0.98046875,
|
| 2295 |
+
"epoch": 1.6056338028169015,
|
| 2296 |
+
"grad_norm": 0.87890625,
|
| 2297 |
+
"learning_rate": 5e-05,
|
| 2298 |
+
"loss": 0.9880096316337585,
|
| 2299 |
+
"mean_token_accuracy": 0.8147650957107544,
|
| 2300 |
+
"num_tokens": 641290.0,
|
| 2301 |
+
"step": 228
|
| 2302 |
+
},
|
| 2303 |
+
{
|
| 2304 |
+
"entropy": 1.140625,
|
| 2305 |
+
"epoch": 1.612676056338028,
|
| 2306 |
+
"grad_norm": 0.9140625,
|
| 2307 |
+
"learning_rate": 5e-05,
|
| 2308 |
+
"loss": 1.1274977922439575,
|
| 2309 |
+
"mean_token_accuracy": 0.7871226072311401,
|
| 2310 |
+
"num_tokens": 644544.0,
|
| 2311 |
+
"step": 229
|
| 2312 |
+
},
|
| 2313 |
+
{
|
| 2314 |
+
"entropy": 1.3828125,
|
| 2315 |
+
"epoch": 1.619718309859155,
|
| 2316 |
+
"grad_norm": 0.96875,
|
| 2317 |
+
"learning_rate": 5e-05,
|
| 2318 |
+
"loss": 1.3756309747695923,
|
| 2319 |
+
"mean_token_accuracy": 0.7243735790252686,
|
| 2320 |
+
"num_tokens": 648503.0,
|
| 2321 |
+
"step": 230
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"entropy": 1.140625,
|
| 2325 |
+
"epoch": 1.6267605633802817,
|
| 2326 |
+
"grad_norm": 0.94140625,
|
| 2327 |
+
"learning_rate": 5e-05,
|
| 2328 |
+
"loss": 1.129349946975708,
|
| 2329 |
+
"mean_token_accuracy": 0.7763996124267578,
|
| 2330 |
+
"num_tokens": 651494.0,
|
| 2331 |
+
"step": 231
|
| 2332 |
+
},
|
| 2333 |
+
{
|
| 2334 |
+
"entropy": 0.90625,
|
| 2335 |
+
"epoch": 1.6338028169014085,
|
| 2336 |
+
"grad_norm": 1.046875,
|
| 2337 |
+
"learning_rate": 5e-05,
|
| 2338 |
+
"loss": 0.8873183131217957,
|
| 2339 |
+
"mean_token_accuracy": 0.8282105326652527,
|
| 2340 |
+
"num_tokens": 653877.0,
|
| 2341 |
+
"step": 232
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"entropy": 0.80078125,
|
| 2345 |
+
"epoch": 1.6408450704225352,
|
| 2346 |
+
"grad_norm": 0.94921875,
|
| 2347 |
+
"learning_rate": 5e-05,
|
| 2348 |
+
"loss": 0.7888442277908325,
|
| 2349 |
+
"mean_token_accuracy": 0.8641618490219116,
|
| 2350 |
+
"num_tokens": 655961.0,
|
| 2351 |
+
"step": 233
|
| 2352 |
+
},
|
| 2353 |
+
{
|
| 2354 |
+
"entropy": 1.0078125,
|
| 2355 |
+
"epoch": 1.647887323943662,
|
| 2356 |
+
"grad_norm": 0.97265625,
|
| 2357 |
+
"learning_rate": 5e-05,
|
| 2358 |
+
"loss": 0.9641861319541931,
|
| 2359 |
+
"mean_token_accuracy": 0.8236775994300842,
|
| 2360 |
+
"num_tokens": 658351.0,
|
| 2361 |
+
"step": 234
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"entropy": 1.140625,
|
| 2365 |
+
"epoch": 1.6549295774647887,
|
| 2366 |
+
"grad_norm": 0.84765625,
|
| 2367 |
+
"learning_rate": 5e-05,
|
| 2368 |
+
"loss": 1.1152194738388062,
|
| 2369 |
+
"mean_token_accuracy": 0.7858839631080627,
|
| 2370 |
+
"num_tokens": 661306.0,
|
| 2371 |
+
"step": 235
|
| 2372 |
+
},
|
| 2373 |
+
{
|
| 2374 |
+
"entropy": 1.0,
|
| 2375 |
+
"epoch": 1.6619718309859155,
|
| 2376 |
+
"grad_norm": 0.9375,
|
| 2377 |
+
"learning_rate": 5e-05,
|
| 2378 |
+
"loss": 0.9575698971748352,
|
| 2379 |
+
"mean_token_accuracy": 0.8209785223007202,
|
| 2380 |
+
"num_tokens": 664012.0,
|
| 2381 |
+
"step": 236
|
| 2382 |
+
},
|
| 2383 |
+
{
|
| 2384 |
+
"entropy": 0.6171875,
|
| 2385 |
+
"epoch": 1.6690140845070423,
|
| 2386 |
+
"grad_norm": 0.90234375,
|
| 2387 |
+
"learning_rate": 5e-05,
|
| 2388 |
+
"loss": 0.5924875140190125,
|
| 2389 |
+
"mean_token_accuracy": 0.8942976593971252,
|
| 2390 |
+
"num_tokens": 666177.0,
|
| 2391 |
+
"step": 237
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"entropy": 1.1484375,
|
| 2395 |
+
"epoch": 1.676056338028169,
|
| 2396 |
+
"grad_norm": 0.8046875,
|
| 2397 |
+
"learning_rate": 5e-05,
|
| 2398 |
+
"loss": 1.1048884391784668,
|
| 2399 |
+
"mean_token_accuracy": 0.7807835936546326,
|
| 2400 |
+
"num_tokens": 669401.0,
|
| 2401 |
+
"step": 238
|
| 2402 |
+
},
|
| 2403 |
+
{
|
| 2404 |
+
"entropy": 0.8515625,
|
| 2405 |
+
"epoch": 1.6830985915492958,
|
| 2406 |
+
"grad_norm": 0.93359375,
|
| 2407 |
+
"learning_rate": 5e-05,
|
| 2408 |
+
"loss": 0.7597204446792603,
|
| 2409 |
+
"mean_token_accuracy": 0.8503086566925049,
|
| 2410 |
+
"num_tokens": 672001.0,
|
| 2411 |
+
"step": 239
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"entropy": 1.3203125,
|
| 2415 |
+
"epoch": 1.6901408450704225,
|
| 2416 |
+
"grad_norm": 0.9296875,
|
| 2417 |
+
"learning_rate": 5e-05,
|
| 2418 |
+
"loss": 1.3505604267120361,
|
| 2419 |
+
"mean_token_accuracy": 0.7473288774490356,
|
| 2420 |
+
"num_tokens": 675472.0,
|
| 2421 |
+
"step": 240
|
| 2422 |
+
},
|
| 2423 |
+
{
|
| 2424 |
+
"entropy": 1.1328125,
|
| 2425 |
+
"epoch": 1.6971830985915493,
|
| 2426 |
+
"grad_norm": 0.91015625,
|
| 2427 |
+
"learning_rate": 5e-05,
|
| 2428 |
+
"loss": 1.0587811470031738,
|
| 2429 |
+
"mean_token_accuracy": 0.801033616065979,
|
| 2430 |
+
"num_tokens": 678189.0,
|
| 2431 |
+
"step": 241
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"entropy": 0.890625,
|
| 2435 |
+
"epoch": 1.704225352112676,
|
| 2436 |
+
"grad_norm": 1.2421875,
|
| 2437 |
+
"learning_rate": 5e-05,
|
| 2438 |
+
"loss": 0.8245605230331421,
|
| 2439 |
+
"mean_token_accuracy": 0.8421264886856079,
|
| 2440 |
+
"num_tokens": 680680.0,
|
| 2441 |
+
"step": 242
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"entropy": 1.15625,
|
| 2445 |
+
"epoch": 1.711267605633803,
|
| 2446 |
+
"grad_norm": 0.9453125,
|
| 2447 |
+
"learning_rate": 5e-05,
|
| 2448 |
+
"loss": 1.1660860776901245,
|
| 2449 |
+
"mean_token_accuracy": 0.7721893787384033,
|
| 2450 |
+
"num_tokens": 683730.0,
|
| 2451 |
+
"step": 243
|
| 2452 |
+
},
|
| 2453 |
+
{
|
| 2454 |
+
"entropy": 0.87109375,
|
| 2455 |
+
"epoch": 1.7183098591549295,
|
| 2456 |
+
"grad_norm": 0.91796875,
|
| 2457 |
+
"learning_rate": 5e-05,
|
| 2458 |
+
"loss": 0.8355469703674316,
|
| 2459 |
+
"mean_token_accuracy": 0.8397717475891113,
|
| 2460 |
+
"num_tokens": 686016.0,
|
| 2461 |
+
"step": 244
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"entropy": 0.9921875,
|
| 2465 |
+
"epoch": 1.7253521126760565,
|
| 2466 |
+
"grad_norm": 0.9609375,
|
| 2467 |
+
"learning_rate": 5e-05,
|
| 2468 |
+
"loss": 0.9942226409912109,
|
| 2469 |
+
"mean_token_accuracy": 0.8151007890701294,
|
| 2470 |
+
"num_tokens": 688355.0,
|
| 2471 |
+
"step": 245
|
| 2472 |
+
},
|
| 2473 |
+
{
|
| 2474 |
+
"entropy": 0.80078125,
|
| 2475 |
+
"epoch": 1.732394366197183,
|
| 2476 |
+
"grad_norm": 0.90234375,
|
| 2477 |
+
"learning_rate": 5e-05,
|
| 2478 |
+
"loss": 0.748877763748169,
|
| 2479 |
+
"mean_token_accuracy": 0.8575118184089661,
|
| 2480 |
+
"num_tokens": 690686.0,
|
| 2481 |
+
"step": 246
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"entropy": 0.67578125,
|
| 2485 |
+
"epoch": 1.73943661971831,
|
| 2486 |
+
"grad_norm": 0.859375,
|
| 2487 |
+
"learning_rate": 5e-05,
|
| 2488 |
+
"loss": 0.6232599020004272,
|
| 2489 |
+
"mean_token_accuracy": 0.8659953474998474,
|
| 2490 |
+
"num_tokens": 693276.0,
|
| 2491 |
+
"step": 247
|
| 2492 |
+
},
|
| 2493 |
+
{
|
| 2494 |
+
"entropy": 0.89453125,
|
| 2495 |
+
"epoch": 1.7464788732394365,
|
| 2496 |
+
"grad_norm": 0.84375,
|
| 2497 |
+
"learning_rate": 5e-05,
|
| 2498 |
+
"loss": 0.8559207916259766,
|
| 2499 |
+
"mean_token_accuracy": 0.8167107105255127,
|
| 2500 |
+
"num_tokens": 696312.0,
|
| 2501 |
+
"step": 248
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"entropy": 0.98828125,
|
| 2505 |
+
"epoch": 1.7535211267605635,
|
| 2506 |
+
"grad_norm": 0.84375,
|
| 2507 |
+
"learning_rate": 5e-05,
|
| 2508 |
+
"loss": 0.9271271824836731,
|
| 2509 |
+
"mean_token_accuracy": 0.8255208134651184,
|
| 2510 |
+
"num_tokens": 699392.0,
|
| 2511 |
+
"step": 249
|
| 2512 |
+
},
|
| 2513 |
+
{
|
| 2514 |
+
"entropy": 0.67578125,
|
| 2515 |
+
"epoch": 1.76056338028169,
|
| 2516 |
+
"grad_norm": 0.859375,
|
| 2517 |
+
"learning_rate": 5e-05,
|
| 2518 |
+
"loss": 0.6688521504402161,
|
| 2519 |
+
"mean_token_accuracy": 0.8663651347160339,
|
| 2520 |
+
"num_tokens": 701832.0,
|
| 2521 |
+
"step": 250
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"entropy": 1.1015625,
|
| 2525 |
+
"epoch": 1.767605633802817,
|
| 2526 |
+
"grad_norm": 0.92578125,
|
| 2527 |
+
"learning_rate": 5e-05,
|
| 2528 |
+
"loss": 1.0740944147109985,
|
| 2529 |
+
"mean_token_accuracy": 0.7920154333114624,
|
| 2530 |
+
"num_tokens": 704946.0,
|
| 2531 |
+
"step": 251
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"entropy": 0.92578125,
|
| 2535 |
+
"epoch": 1.7746478873239435,
|
| 2536 |
+
"grad_norm": 0.86328125,
|
| 2537 |
+
"learning_rate": 5e-05,
|
| 2538 |
+
"loss": 0.9155241847038269,
|
| 2539 |
+
"mean_token_accuracy": 0.8214963674545288,
|
| 2540 |
+
"num_tokens": 707587.0,
|
| 2541 |
+
"step": 252
|
| 2542 |
+
},
|
| 2543 |
+
{
|
| 2544 |
+
"entropy": 1.0234375,
|
| 2545 |
+
"epoch": 1.7816901408450705,
|
| 2546 |
+
"grad_norm": 0.90234375,
|
| 2547 |
+
"learning_rate": 5e-05,
|
| 2548 |
+
"loss": 1.0119163990020752,
|
| 2549 |
+
"mean_token_accuracy": 0.8100035786628723,
|
| 2550 |
+
"num_tokens": 710374.0,
|
| 2551 |
+
"step": 253
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"entropy": 1.1484375,
|
| 2555 |
+
"epoch": 1.788732394366197,
|
| 2556 |
+
"grad_norm": 0.8515625,
|
| 2557 |
+
"learning_rate": 5e-05,
|
| 2558 |
+
"loss": 1.098153829574585,
|
| 2559 |
+
"mean_token_accuracy": 0.7915265560150146,
|
| 2560 |
+
"num_tokens": 713356.0,
|
| 2561 |
+
"step": 254
|
| 2562 |
+
},
|
| 2563 |
+
{
|
| 2564 |
+
"entropy": 1.21875,
|
| 2565 |
+
"epoch": 1.795774647887324,
|
| 2566 |
+
"grad_norm": 0.87109375,
|
| 2567 |
+
"learning_rate": 5e-05,
|
| 2568 |
+
"loss": 1.1987107992172241,
|
| 2569 |
+
"mean_token_accuracy": 0.77190762758255,
|
| 2570 |
+
"num_tokens": 716525.0,
|
| 2571 |
+
"step": 255
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"entropy": 1.1171875,
|
| 2575 |
+
"epoch": 1.8028169014084507,
|
| 2576 |
+
"grad_norm": 0.8359375,
|
| 2577 |
+
"learning_rate": 5e-05,
|
| 2578 |
+
"loss": 1.1219631433486938,
|
| 2579 |
+
"mean_token_accuracy": 0.7778891324996948,
|
| 2580 |
+
"num_tokens": 719527.0,
|
| 2581 |
+
"step": 256
|
| 2582 |
+
},
|
| 2583 |
+
{
|
| 2584 |
+
"entropy": 1.0390625,
|
| 2585 |
+
"epoch": 1.8098591549295775,
|
| 2586 |
+
"grad_norm": 0.953125,
|
| 2587 |
+
"learning_rate": 5e-05,
|
| 2588 |
+
"loss": 1.0219084024429321,
|
| 2589 |
+
"mean_token_accuracy": 0.8117510676383972,
|
| 2590 |
+
"num_tokens": 722122.0,
|
| 2591 |
+
"step": 257
|
| 2592 |
+
},
|
| 2593 |
+
{
|
| 2594 |
+
"entropy": 0.92578125,
|
| 2595 |
+
"epoch": 1.8169014084507042,
|
| 2596 |
+
"grad_norm": 1.09375,
|
| 2597 |
+
"learning_rate": 5e-05,
|
| 2598 |
+
"loss": 0.8954271078109741,
|
| 2599 |
+
"mean_token_accuracy": 0.8421261310577393,
|
| 2600 |
+
"num_tokens": 724651.0,
|
| 2601 |
+
"step": 258
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"entropy": 0.79296875,
|
| 2605 |
+
"epoch": 1.823943661971831,
|
| 2606 |
+
"grad_norm": 0.9140625,
|
| 2607 |
+
"learning_rate": 5e-05,
|
| 2608 |
+
"loss": 0.7704645395278931,
|
| 2609 |
+
"mean_token_accuracy": 0.8589590787887573,
|
| 2610 |
+
"num_tokens": 727176.0,
|
| 2611 |
+
"step": 259
|
| 2612 |
+
},
|
| 2613 |
+
{
|
| 2614 |
+
"entropy": 0.63671875,
|
| 2615 |
+
"epoch": 1.8309859154929577,
|
| 2616 |
+
"grad_norm": 1.03125,
|
| 2617 |
+
"learning_rate": 5e-05,
|
| 2618 |
+
"loss": 0.5992922782897949,
|
| 2619 |
+
"mean_token_accuracy": 0.8863636255264282,
|
| 2620 |
+
"num_tokens": 729516.0,
|
| 2621 |
+
"step": 260
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"entropy": 1.1484375,
|
| 2625 |
+
"epoch": 1.8380281690140845,
|
| 2626 |
+
"grad_norm": 1.171875,
|
| 2627 |
+
"learning_rate": 5e-05,
|
| 2628 |
+
"loss": 1.0654540061950684,
|
| 2629 |
+
"mean_token_accuracy": 0.8068057298660278,
|
| 2630 |
+
"num_tokens": 732257.0,
|
| 2631 |
+
"step": 261
|
| 2632 |
+
},
|
| 2633 |
+
{
|
| 2634 |
+
"entropy": 0.64453125,
|
| 2635 |
+
"epoch": 1.8450704225352113,
|
| 2636 |
+
"grad_norm": 0.828125,
|
| 2637 |
+
"learning_rate": 5e-05,
|
| 2638 |
+
"loss": 0.6201009154319763,
|
| 2639 |
+
"mean_token_accuracy": 0.8818508386611938,
|
| 2640 |
+
"num_tokens": 734491.0,
|
| 2641 |
+
"step": 262
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"entropy": 1.0,
|
| 2645 |
+
"epoch": 1.852112676056338,
|
| 2646 |
+
"grad_norm": 0.90234375,
|
| 2647 |
+
"learning_rate": 5e-05,
|
| 2648 |
+
"loss": 0.9687674045562744,
|
| 2649 |
+
"mean_token_accuracy": 0.8077198266983032,
|
| 2650 |
+
"num_tokens": 737297.0,
|
| 2651 |
+
"step": 263
|
| 2652 |
+
},
|
| 2653 |
+
{
|
| 2654 |
+
"entropy": 0.5546875,
|
| 2655 |
+
"epoch": 1.8591549295774648,
|
| 2656 |
+
"grad_norm": 0.97265625,
|
| 2657 |
+
"learning_rate": 5e-05,
|
| 2658 |
+
"loss": 0.5199809074401855,
|
| 2659 |
+
"mean_token_accuracy": 0.900547444820404,
|
| 2660 |
+
"num_tokens": 739497.0,
|
| 2661 |
+
"step": 264
|
| 2662 |
+
},
|
| 2663 |
+
{
|
| 2664 |
+
"entropy": 0.81640625,
|
| 2665 |
+
"epoch": 1.8661971830985915,
|
| 2666 |
+
"grad_norm": 0.96484375,
|
| 2667 |
+
"learning_rate": 5e-05,
|
| 2668 |
+
"loss": 0.7673903107643127,
|
| 2669 |
+
"mean_token_accuracy": 0.8505041599273682,
|
| 2670 |
+
"num_tokens": 741786.0,
|
| 2671 |
+
"step": 265
|
| 2672 |
+
},
|
| 2673 |
+
{
|
| 2674 |
+
"entropy": 1.203125,
|
| 2675 |
+
"epoch": 1.8732394366197183,
|
| 2676 |
+
"grad_norm": 0.84765625,
|
| 2677 |
+
"learning_rate": 5e-05,
|
| 2678 |
+
"loss": 1.2104980945587158,
|
| 2679 |
+
"mean_token_accuracy": 0.7729952931404114,
|
| 2680 |
+
"num_tokens": 745186.0,
|
| 2681 |
+
"step": 266
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"entropy": 0.921875,
|
| 2685 |
+
"epoch": 1.880281690140845,
|
| 2686 |
+
"grad_norm": 0.8359375,
|
| 2687 |
+
"learning_rate": 5e-05,
|
| 2688 |
+
"loss": 0.9426409006118774,
|
| 2689 |
+
"mean_token_accuracy": 0.8229434490203857,
|
| 2690 |
+
"num_tokens": 748306.0,
|
| 2691 |
+
"step": 267
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"entropy": 0.92578125,
|
| 2695 |
+
"epoch": 1.887323943661972,
|
| 2696 |
+
"grad_norm": 0.91015625,
|
| 2697 |
+
"learning_rate": 5e-05,
|
| 2698 |
+
"loss": 0.9220845103263855,
|
| 2699 |
+
"mean_token_accuracy": 0.8194718360900879,
|
| 2700 |
+
"num_tokens": 750851.0,
|
| 2701 |
+
"step": 268
|
| 2702 |
+
},
|
| 2703 |
+
{
|
| 2704 |
+
"entropy": 1.0234375,
|
| 2705 |
+
"epoch": 1.8943661971830985,
|
| 2706 |
+
"grad_norm": 0.9765625,
|
| 2707 |
+
"learning_rate": 5e-05,
|
| 2708 |
+
"loss": 1.0415750741958618,
|
| 2709 |
+
"mean_token_accuracy": 0.7877483367919922,
|
| 2710 |
+
"num_tokens": 753879.0,
|
| 2711 |
+
"step": 269
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"entropy": 0.87109375,
|
| 2715 |
+
"epoch": 1.9014084507042255,
|
| 2716 |
+
"grad_norm": 0.86328125,
|
| 2717 |
+
"learning_rate": 5e-05,
|
| 2718 |
+
"loss": 0.858406126499176,
|
| 2719 |
+
"mean_token_accuracy": 0.8365448713302612,
|
| 2720 |
+
"num_tokens": 756897.0,
|
| 2721 |
+
"step": 270
|
| 2722 |
+
},
|
| 2723 |
+
{
|
| 2724 |
+
"entropy": 1.0625,
|
| 2725 |
+
"epoch": 1.908450704225352,
|
| 2726 |
+
"grad_norm": 0.875,
|
| 2727 |
+
"learning_rate": 5e-05,
|
| 2728 |
+
"loss": 1.1015472412109375,
|
| 2729 |
+
"mean_token_accuracy": 0.7784239649772644,
|
| 2730 |
+
"num_tokens": 760344.0,
|
| 2731 |
+
"step": 271
|
| 2732 |
+
},
|
| 2733 |
+
{
|
| 2734 |
+
"entropy": 1.03125,
|
| 2735 |
+
"epoch": 1.915492957746479,
|
| 2736 |
+
"grad_norm": 0.9609375,
|
| 2737 |
+
"learning_rate": 5e-05,
|
| 2738 |
+
"loss": 1.0051850080490112,
|
| 2739 |
+
"mean_token_accuracy": 0.8070366382598877,
|
| 2740 |
+
"num_tokens": 763109.0,
|
| 2741 |
+
"step": 272
|
| 2742 |
+
},
|
| 2743 |
+
{
|
| 2744 |
+
"entropy": 1.140625,
|
| 2745 |
+
"epoch": 1.9225352112676055,
|
| 2746 |
+
"grad_norm": 0.89453125,
|
| 2747 |
+
"learning_rate": 5e-05,
|
| 2748 |
+
"loss": 1.1266851425170898,
|
| 2749 |
+
"mean_token_accuracy": 0.7730742692947388,
|
| 2750 |
+
"num_tokens": 765999.0,
|
| 2751 |
+
"step": 273
|
| 2752 |
+
},
|
| 2753 |
+
{
|
| 2754 |
+
"entropy": 0.8046875,
|
| 2755 |
+
"epoch": 1.9295774647887325,
|
| 2756 |
+
"grad_norm": 0.79296875,
|
| 2757 |
+
"learning_rate": 5e-05,
|
| 2758 |
+
"loss": 0.7725271582603455,
|
| 2759 |
+
"mean_token_accuracy": 0.855555534362793,
|
| 2760 |
+
"num_tokens": 768527.0,
|
| 2761 |
+
"step": 274
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"entropy": 0.74609375,
|
| 2765 |
+
"epoch": 1.936619718309859,
|
| 2766 |
+
"grad_norm": 0.78125,
|
| 2767 |
+
"learning_rate": 5e-05,
|
| 2768 |
+
"loss": 0.7248237133026123,
|
| 2769 |
+
"mean_token_accuracy": 0.8687350749969482,
|
| 2770 |
+
"num_tokens": 771049.0,
|
| 2771 |
+
"step": 275
|
| 2772 |
+
},
|
| 2773 |
+
{
|
| 2774 |
+
"entropy": 0.82421875,
|
| 2775 |
+
"epoch": 1.943661971830986,
|
| 2776 |
+
"grad_norm": 0.8828125,
|
| 2777 |
+
"learning_rate": 5e-05,
|
| 2778 |
+
"loss": 0.8026857972145081,
|
| 2779 |
+
"mean_token_accuracy": 0.8561615943908691,
|
| 2780 |
+
"num_tokens": 773532.0,
|
| 2781 |
+
"step": 276
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"entropy": 1.0,
|
| 2785 |
+
"epoch": 1.9507042253521125,
|
| 2786 |
+
"grad_norm": 0.9140625,
|
| 2787 |
+
"learning_rate": 5e-05,
|
| 2788 |
+
"loss": 0.9862289428710938,
|
| 2789 |
+
"mean_token_accuracy": 0.8059446811676025,
|
| 2790 |
+
"num_tokens": 776467.0,
|
| 2791 |
+
"step": 277
|
| 2792 |
+
},
|
| 2793 |
+
{
|
| 2794 |
+
"entropy": 1.1328125,
|
| 2795 |
+
"epoch": 1.9577464788732395,
|
| 2796 |
+
"grad_norm": 0.9296875,
|
| 2797 |
+
"learning_rate": 5e-05,
|
| 2798 |
+
"loss": 1.1140096187591553,
|
| 2799 |
+
"mean_token_accuracy": 0.7820383310317993,
|
| 2800 |
+
"num_tokens": 779448.0,
|
| 2801 |
+
"step": 278
|
| 2802 |
+
},
|
| 2803 |
+
{
|
| 2804 |
+
"entropy": 1.0859375,
|
| 2805 |
+
"epoch": 1.9647887323943662,
|
| 2806 |
+
"grad_norm": 0.87109375,
|
| 2807 |
+
"learning_rate": 5e-05,
|
| 2808 |
+
"loss": 1.0614746809005737,
|
| 2809 |
+
"mean_token_accuracy": 0.7838625907897949,
|
| 2810 |
+
"num_tokens": 782542.0,
|
| 2811 |
+
"step": 279
|
| 2812 |
+
},
|
| 2813 |
+
{
|
| 2814 |
+
"entropy": 1.0625,
|
| 2815 |
+
"epoch": 1.971830985915493,
|
| 2816 |
+
"grad_norm": 0.93359375,
|
| 2817 |
+
"learning_rate": 5e-05,
|
| 2818 |
+
"loss": 1.053848385810852,
|
| 2819 |
+
"mean_token_accuracy": 0.7868378758430481,
|
| 2820 |
+
"num_tokens": 785665.0,
|
| 2821 |
+
"step": 280
|
| 2822 |
+
},
|
| 2823 |
+
{
|
| 2824 |
+
"entropy": 1.0625,
|
| 2825 |
+
"epoch": 1.9788732394366197,
|
| 2826 |
+
"grad_norm": 1.015625,
|
| 2827 |
+
"learning_rate": 5e-05,
|
| 2828 |
+
"loss": 1.0552152395248413,
|
| 2829 |
+
"mean_token_accuracy": 0.7965670824050903,
|
| 2830 |
+
"num_tokens": 788819.0,
|
| 2831 |
+
"step": 281
|
| 2832 |
+
},
|
| 2833 |
+
{
|
| 2834 |
+
"entropy": 1.1171875,
|
| 2835 |
+
"epoch": 1.9859154929577465,
|
| 2836 |
+
"grad_norm": 1.0390625,
|
| 2837 |
+
"learning_rate": 5e-05,
|
| 2838 |
+
"loss": 1.146568775177002,
|
| 2839 |
+
"mean_token_accuracy": 0.7714470624923706,
|
| 2840 |
+
"num_tokens": 791916.0,
|
| 2841 |
+
"step": 282
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"entropy": 0.875,
|
| 2845 |
+
"epoch": 1.9929577464788732,
|
| 2846 |
+
"grad_norm": 1.015625,
|
| 2847 |
+
"learning_rate": 5e-05,
|
| 2848 |
+
"loss": 0.8557621240615845,
|
| 2849 |
+
"mean_token_accuracy": 0.8393825888633728,
|
| 2850 |
+
"num_tokens": 794321.0,
|
| 2851 |
+
"step": 283
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"entropy": 0.69140625,
|
| 2855 |
+
"epoch": 2.0,
|
| 2856 |
+
"grad_norm": 1.0078125,
|
| 2857 |
+
"learning_rate": 5e-05,
|
| 2858 |
+
"loss": 0.694361686706543,
|
| 2859 |
+
"mean_token_accuracy": 0.8591725826263428,
|
| 2860 |
+
"num_tokens": 796722.0,
|
| 2861 |
+
"step": 284
|
| 2862 |
+
},
|
| 2863 |
+
{
|
| 2864 |
+
"epoch": 2.0,
|
| 2865 |
+
"eval_entropy": 0.9236111111111112,
|
| 2866 |
+
"eval_loss": 0.9322065114974976,
|
| 2867 |
+
"eval_mean_token_accuracy": 0.8198160545693504,
|
| 2868 |
+
"eval_num_tokens": 796722.0,
|
| 2869 |
+
"eval_runtime": 3.7129,
|
| 2870 |
+
"eval_samples_per_second": 76.489,
|
| 2871 |
+
"eval_steps_per_second": 9.696,
|
| 2872 |
+
"step": 284
|
| 2873 |
+
}
|
| 2874 |
+
],
|
| 2875 |
+
"logging_steps": 1,
|
| 2876 |
+
"max_steps": 710,
|
| 2877 |
+
"num_input_tokens_seen": 0,
|
| 2878 |
+
"num_train_epochs": 5,
|
| 2879 |
+
"save_steps": 500,
|
| 2880 |
+
"stateful_callbacks": {
|
| 2881 |
+
"TrainerControl": {
|
| 2882 |
+
"args": {
|
| 2883 |
+
"should_epoch_stop": false,
|
| 2884 |
+
"should_evaluate": false,
|
| 2885 |
+
"should_log": false,
|
| 2886 |
+
"should_save": true,
|
| 2887 |
+
"should_training_stop": false
|
| 2888 |
+
},
|
| 2889 |
+
"attributes": {}
|
| 2890 |
+
}
|
| 2891 |
+
},
|
| 2892 |
+
"total_flos": 736331500394496.0,
|
| 2893 |
+
"train_batch_size": 8,
|
| 2894 |
+
"trial_name": null,
|
| 2895 |
+
"trial_params": null
|
| 2896 |
+
}
|
checkpoint-284/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fa1d74cfdb570eb6c160e915c5780a404149c8da32b2030eb89b29d13902242
|
| 3 |
+
size 5585
|
checkpoint-426/chat_template.jinja
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system
|
| 2 |
+
You are a helpful AI assistant named SmolLM, trained by Hugging Face<|im_end|>
|
| 3 |
+
' }}{% endif %}{{'<|im_start|>' + message['role'] + '
|
| 4 |
+
' + message['content'] + '<|im_end|>' + '
|
| 5 |
+
'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
|
| 6 |
+
' }}{% endif %}
|
checkpoint-426/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 576,
|
| 13 |
+
"initializer_range": 0.041666666666666664,
|
| 14 |
+
"intermediate_size": 1536,
|
| 15 |
+
"is_llama_config": true,
|
| 16 |
+
"max_position_embeddings": 8192,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"num_attention_heads": 9,
|
| 20 |
+
"num_hidden_layers": 30,
|
| 21 |
+
"num_key_value_heads": 3,
|
| 22 |
+
"pad_token_id": 2,
|
| 23 |
+
"pretraining_tp": 1,
|
| 24 |
+
"rms_norm_eps": 1e-05,
|
| 25 |
+
"rope_interleaved": false,
|
| 26 |
+
"rope_parameters": {
|
| 27 |
+
"rope_theta": 100000,
|
| 28 |
+
"rope_type": "default"
|
| 29 |
+
},
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"transformers.js_config": {
|
| 32 |
+
"kv_cache_dtype": {
|
| 33 |
+
"fp16": "float16",
|
| 34 |
+
"q4f16": "float16"
|
| 35 |
+
}
|
| 36 |
+
},
|
| 37 |
+
"transformers_version": "5.2.0",
|
| 38 |
+
"use_cache": false,
|
| 39 |
+
"vocab_size": 49152
|
| 40 |
+
}
|
checkpoint-426/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
2
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 2,
|
| 8 |
+
"transformers_version": "5.2.0"
|
| 9 |
+
}
|
checkpoint-426/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e28249fd7b40a2acfad0b7de39d1801c13c5eec711df23f9a6107109870a3e7
|
| 3 |
+
size 269060552
|
checkpoint-426/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:01f9a0f7843a37be87edd23f4e88aa93b38b95cc2c07503eeb1cf2e4632453a2
|
| 3 |
+
size 14645
|
checkpoint-426/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:819d56655f7abb028987362b34c5eb64d14796e7d55c61f08b0e677fcfa4f5e9
|
| 3 |
+
size 1465
|
checkpoint-426/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-426/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|im_start|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"extra_special_tokens": [
|
| 8 |
+
"<|im_start|>",
|
| 9 |
+
"<|im_end|>"
|
| 10 |
+
],
|
| 11 |
+
"is_local": true,
|
| 12 |
+
"model_max_length": 8192,
|
| 13 |
+
"pad_token": "<|im_end|>",
|
| 14 |
+
"tokenizer_class": "TokenizersBackend",
|
| 15 |
+
"unk_token": "<|endoftext|>",
|
| 16 |
+
"vocab_size": 49152
|
| 17 |
+
}
|
checkpoint-426/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-426/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fa1d74cfdb570eb6c160e915c5780a404149c8da32b2030eb89b29d13902242
|
| 3 |
+
size 5585
|
checkpoint-568/chat_template.jinja
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system
|
| 2 |
+
You are a helpful AI assistant named SmolLM, trained by Hugging Face<|im_end|>
|
| 3 |
+
' }}{% endif %}{{'<|im_start|>' + message['role'] + '
|
| 4 |
+
' + message['content'] + '<|im_end|>' + '
|
| 5 |
+
'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
|
| 6 |
+
' }}{% endif %}
|
checkpoint-568/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 576,
|
| 13 |
+
"initializer_range": 0.041666666666666664,
|
| 14 |
+
"intermediate_size": 1536,
|
| 15 |
+
"is_llama_config": true,
|
| 16 |
+
"max_position_embeddings": 8192,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"num_attention_heads": 9,
|
| 20 |
+
"num_hidden_layers": 30,
|
| 21 |
+
"num_key_value_heads": 3,
|
| 22 |
+
"pad_token_id": 2,
|
| 23 |
+
"pretraining_tp": 1,
|
| 24 |
+
"rms_norm_eps": 1e-05,
|
| 25 |
+
"rope_interleaved": false,
|
| 26 |
+
"rope_parameters": {
|
| 27 |
+
"rope_theta": 100000,
|
| 28 |
+
"rope_type": "default"
|
| 29 |
+
},
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"transformers.js_config": {
|
| 32 |
+
"kv_cache_dtype": {
|
| 33 |
+
"fp16": "float16",
|
| 34 |
+
"q4f16": "float16"
|
| 35 |
+
}
|
| 36 |
+
},
|
| 37 |
+
"transformers_version": "5.2.0",
|
| 38 |
+
"use_cache": false,
|
| 39 |
+
"vocab_size": 49152
|
| 40 |
+
}
|
checkpoint-568/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
2
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 2,
|
| 8 |
+
"transformers_version": "5.2.0"
|
| 9 |
+
}
|
checkpoint-568/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5139dfc37bea7f0984db9f6936d47139bf7d600646c0bce5a2f1e10930608fc
|
| 3 |
+
size 269060552
|
checkpoint-568/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fdc37bbd2e979f041dfbbb004a5c74bab6cdda159cb18116df728588515a9ef6
|
| 3 |
+
size 14645
|
checkpoint-568/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6c374b5a90050f0de9583fb1646b9dee9dbbe14becdd74b3cc3c95d679cad6ad
|
| 3 |
+
size 1465
|
checkpoint-568/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-568/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|im_start|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"extra_special_tokens": [
|
| 8 |
+
"<|im_start|>",
|
| 9 |
+
"<|im_end|>"
|
| 10 |
+
],
|
| 11 |
+
"is_local": true,
|
| 12 |
+
"model_max_length": 8192,
|
| 13 |
+
"pad_token": "<|im_end|>",
|
| 14 |
+
"tokenizer_class": "TokenizersBackend",
|
| 15 |
+
"unk_token": "<|endoftext|>",
|
| 16 |
+
"vocab_size": 49152
|
| 17 |
+
}
|
checkpoint-568/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-568/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fa1d74cfdb570eb6c160e915c5780a404149c8da32b2030eb89b29d13902242
|
| 3 |
+
size 5585
|
checkpoint-710/chat_template.jinja
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system
|
| 2 |
+
You are a helpful AI assistant named SmolLM, trained by Hugging Face<|im_end|>
|
| 3 |
+
' }}{% endif %}{{'<|im_start|>' + message['role'] + '
|
| 4 |
+
' + message['content'] + '<|im_end|>' + '
|
| 5 |
+
'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
|
| 6 |
+
' }}{% endif %}
|
checkpoint-710/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 576,
|
| 13 |
+
"initializer_range": 0.041666666666666664,
|
| 14 |
+
"intermediate_size": 1536,
|
| 15 |
+
"is_llama_config": true,
|
| 16 |
+
"max_position_embeddings": 8192,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"num_attention_heads": 9,
|
| 20 |
+
"num_hidden_layers": 30,
|
| 21 |
+
"num_key_value_heads": 3,
|
| 22 |
+
"pad_token_id": 2,
|
| 23 |
+
"pretraining_tp": 1,
|
| 24 |
+
"rms_norm_eps": 1e-05,
|
| 25 |
+
"rope_interleaved": false,
|
| 26 |
+
"rope_parameters": {
|
| 27 |
+
"rope_theta": 100000,
|
| 28 |
+
"rope_type": "default"
|
| 29 |
+
},
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"transformers.js_config": {
|
| 32 |
+
"kv_cache_dtype": {
|
| 33 |
+
"fp16": "float16",
|
| 34 |
+
"q4f16": "float16"
|
| 35 |
+
}
|
| 36 |
+
},
|
| 37 |
+
"transformers_version": "5.2.0",
|
| 38 |
+
"use_cache": false,
|
| 39 |
+
"vocab_size": 49152
|
| 40 |
+
}
|
checkpoint-710/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
2
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 2,
|
| 8 |
+
"transformers_version": "5.2.0"
|
| 9 |
+
}
|
checkpoint-710/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17cd930da9783ca70bad4b9cdeee6a06c0acea8f34645a333c93341f487f66a3
|
| 3 |
+
size 14645
|
checkpoint-710/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73dff87adc3c981bc959c81f548f6e14e8f23964b56dbf83945e88027f81115d
|
| 3 |
+
size 1465
|
checkpoint-710/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-710/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|im_start|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"extra_special_tokens": [
|
| 8 |
+
"<|im_start|>",
|
| 9 |
+
"<|im_end|>"
|
| 10 |
+
],
|
| 11 |
+
"is_local": true,
|
| 12 |
+
"model_max_length": 8192,
|
| 13 |
+
"pad_token": "<|im_end|>",
|
| 14 |
+
"tokenizer_class": "TokenizersBackend",
|
| 15 |
+
"unk_token": "<|endoftext|>",
|
| 16 |
+
"vocab_size": 49152
|
| 17 |
+
}
|
checkpoint-710/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-710/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fa1d74cfdb570eb6c160e915c5780a404149c8da32b2030eb89b29d13902242
|
| 3 |
+
size 5585
|