The Founder: deberta-v3-base fine-tuned on ag_news v1
Browse files- README.md +7 -7
- checkpoint-10125/config.json +51 -0
- checkpoint-10125/model.safetensors +3 -0
- checkpoint-10125/optimizer.pt +3 -0
- checkpoint-10125/rng_state.pth +3 -0
- checkpoint-10125/scheduler.pt +3 -0
- checkpoint-10125/trainer_state.json +1484 -0
- checkpoint-10125/training_args.bin +3 -0
- checkpoint-3375/optimizer.pt +1 -1
- checkpoint-3375/scheduler.pt +1 -1
- checkpoint-3375/trainer_state.json +170 -170
- checkpoint-3375/training_args.bin +1 -1
- checkpoint-6750/optimizer.pt +1 -1
- checkpoint-6750/scheduler.pt +1 -1
- checkpoint-6750/trainer_state.json +242 -242
- checkpoint-6750/training_args.bin +1 -1
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -50,12 +50,12 @@ Fine-tuned `microsoft/deberta-v3-base` on `fancyzhx/ag_news` for 4-class topic c
|
|
| 50 |
| Labels | World, Sports, Business, Sci/Tech |
|
| 51 |
| Epochs | 3 |
|
| 52 |
| Batch size | 16 |
|
| 53 |
-
| Learning rate |
|
| 54 |
| GPU | Tesla T4 (Kaggle) |
|
| 55 |
-
| Train loss | 0.
|
| 56 |
| Test loss | nan |
|
| 57 |
| Test accuracy | 0.2500 |
|
| 58 |
-
| Duration |
|
| 59 |
|
| 60 |
## How to Get Started
|
| 61 |
|
|
@@ -82,7 +82,7 @@ Not suitable for non-English text, fine-grained news categorisation, or document
|
|
| 82 |
|
| 83 |
| Hyperparameter | Value |
|
| 84 |
|---|---|
|
| 85 |
-
| Learning rate |
|
| 86 |
| Batch size | 16 |
|
| 87 |
| Epochs | 3 |
|
| 88 |
| Optimizer | AdamW |
|
|
@@ -94,15 +94,15 @@ Not suitable for non-English text, fine-grained news categorisation, or document
|
|
| 94 |
|
| 95 |
| Metric | Value |
|
| 96 |
|---|---|
|
| 97 |
-
| Train loss | 0.
|
| 98 |
| Test loss | nan |
|
| 99 |
| Test accuracy | 0.2500 |
|
| 100 |
-
| Duration |
|
| 101 |
| Hypothesis (>= 0.945) | REJECTED |
|
| 102 |
|
| 103 |
## Experiment Tracking
|
| 104 |
|
| 105 |
-
[W&B Run](https://wandb.ai/zane-ml/the-founder/runs/
|
| 106 |
|
| 107 |
## Infrastructure
|
| 108 |
|
|
|
|
| 50 |
| Labels | World, Sports, Business, Sci/Tech |
|
| 51 |
| Epochs | 3 |
|
| 52 |
| Batch size | 16 |
|
| 53 |
+
| Learning rate | 1e-05 |
|
| 54 |
| GPU | Tesla T4 (Kaggle) |
|
| 55 |
+
| Train loss | 0.6909 |
|
| 56 |
| Test loss | nan |
|
| 57 |
| Test accuracy | 0.2500 |
|
| 58 |
+
| Duration | 42.9 min |
|
| 59 |
|
| 60 |
## How to Get Started
|
| 61 |
|
|
|
|
| 82 |
|
| 83 |
| Hyperparameter | Value |
|
| 84 |
|---|---|
|
| 85 |
+
| Learning rate | 1e-05 |
|
| 86 |
| Batch size | 16 |
|
| 87 |
| Epochs | 3 |
|
| 88 |
| Optimizer | AdamW |
|
|
|
|
| 94 |
|
| 95 |
| Metric | Value |
|
| 96 |
|---|---|
|
| 97 |
+
| Train loss | 0.6909 |
|
| 98 |
| Test loss | nan |
|
| 99 |
| Test accuracy | 0.2500 |
|
| 100 |
+
| Duration | 42.9 min |
|
| 101 |
| Hypothesis (>= 0.945) | REJECTED |
|
| 102 |
|
| 103 |
## Experiment Tracking
|
| 104 |
|
| 105 |
+
[W&B Run](https://wandb.ai/zane-ml/the-founder/runs/sip1yk0c)
|
| 106 |
|
| 107 |
## Infrastructure
|
| 108 |
|
checkpoint-10125/config.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"DebertaV2ForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": null,
|
| 7 |
+
"dtype": "float16",
|
| 8 |
+
"eos_token_id": null,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "World",
|
| 14 |
+
"1": "Sports",
|
| 15 |
+
"2": "Business",
|
| 16 |
+
"3": "Sci/Tech"
|
| 17 |
+
},
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 3072,
|
| 20 |
+
"label2id": {
|
| 21 |
+
"Business": 2,
|
| 22 |
+
"Sci/Tech": 3,
|
| 23 |
+
"Sports": 1,
|
| 24 |
+
"World": 0
|
| 25 |
+
},
|
| 26 |
+
"layer_norm_eps": 1e-07,
|
| 27 |
+
"legacy": true,
|
| 28 |
+
"max_position_embeddings": 512,
|
| 29 |
+
"max_relative_positions": -1,
|
| 30 |
+
"model_type": "deberta-v2",
|
| 31 |
+
"norm_rel_ebd": "layer_norm",
|
| 32 |
+
"num_attention_heads": 12,
|
| 33 |
+
"num_hidden_layers": 12,
|
| 34 |
+
"pad_token_id": 0,
|
| 35 |
+
"pooler_dropout": 0,
|
| 36 |
+
"pooler_hidden_act": "gelu",
|
| 37 |
+
"pooler_hidden_size": 768,
|
| 38 |
+
"pos_att_type": [
|
| 39 |
+
"p2c",
|
| 40 |
+
"c2p"
|
| 41 |
+
],
|
| 42 |
+
"position_biased_input": false,
|
| 43 |
+
"position_buckets": 256,
|
| 44 |
+
"relative_attention": true,
|
| 45 |
+
"share_att_key": true,
|
| 46 |
+
"tie_word_embeddings": true,
|
| 47 |
+
"transformers_version": "5.0.0",
|
| 48 |
+
"type_vocab_size": 0,
|
| 49 |
+
"use_cache": false,
|
| 50 |
+
"vocab_size": 128100
|
| 51 |
+
}
|
checkpoint-10125/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6f7597ca6965598c1655ac83d684e92d334eac2c878f2828809e0522e3cf8fc
|
| 3 |
+
size 368874960
|
checkpoint-10125/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:33925cb4756b394e237663fea9198a58476a81a5f38b8c168f1bbfd4a871a480
|
| 3 |
+
size 737873099
|
checkpoint-10125/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8539d4301cd7b3b27c6024552cd1d7b64314fc5a95620d71497689291df7f750
|
| 3 |
+
size 14645
|
checkpoint-10125/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9031ebc229a2e8424cae2de11ede5258d8cb6d5638e01011a25d3468ade79c5b
|
| 3 |
+
size 1465
|
checkpoint-10125/trainer_state.json
ADDED
|
@@ -0,0 +1,1484 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 3375,
|
| 3 |
+
"best_metric": 0.25075,
|
| 4 |
+
"best_model_checkpoint": "./output/checkpoint-3375",
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 10125,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.014814814814814815,
|
| 14 |
+
"grad_norm": 15.2734375,
|
| 15 |
+
"learning_rate": 2.4197530864197535e-07,
|
| 16 |
+
"loss": 2.7766494750976562,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.02962962962962963,
|
| 21 |
+
"grad_norm": 14.0625,
|
| 22 |
+
"learning_rate": 4.88888888888889e-07,
|
| 23 |
+
"loss": 2.6392501831054687,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.044444444444444446,
|
| 28 |
+
"grad_norm": 7.6171875,
|
| 29 |
+
"learning_rate": 7.358024691358024e-07,
|
| 30 |
+
"loss": 2.662928161621094,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.05925925925925926,
|
| 35 |
+
"grad_norm": 13.4296875,
|
| 36 |
+
"learning_rate": 9.827160493827162e-07,
|
| 37 |
+
"loss": 3.1913311767578123,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.07407407407407407,
|
| 42 |
+
"grad_norm": 1.96484375,
|
| 43 |
+
"learning_rate": 1.2296296296296297e-06,
|
| 44 |
+
"loss": 2.97621826171875,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.08888888888888889,
|
| 49 |
+
"grad_norm": 2.583984375,
|
| 50 |
+
"learning_rate": 1.4765432098765435e-06,
|
| 51 |
+
"loss": 2.795979309082031,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.1037037037037037,
|
| 56 |
+
"grad_norm": 3.095703125,
|
| 57 |
+
"learning_rate": 1.723456790123457e-06,
|
| 58 |
+
"loss": 2.7952374267578124,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.11851851851851852,
|
| 63 |
+
"grad_norm": 2.453125,
|
| 64 |
+
"learning_rate": 1.9703703703703703e-06,
|
| 65 |
+
"loss": 2.7849395751953123,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.13333333333333333,
|
| 70 |
+
"grad_norm": 6.875,
|
| 71 |
+
"learning_rate": 2.217283950617284e-06,
|
| 72 |
+
"loss": 2.7812811279296876,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.14814814814814814,
|
| 77 |
+
"grad_norm": 2.33203125,
|
| 78 |
+
"learning_rate": 2.464197530864198e-06,
|
| 79 |
+
"loss": 2.7828436279296875,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.16296296296296298,
|
| 84 |
+
"grad_norm": 2.53125,
|
| 85 |
+
"learning_rate": 2.7111111111111116e-06,
|
| 86 |
+
"loss": 2.7977508544921874,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.17777777777777778,
|
| 91 |
+
"grad_norm": 3.587890625,
|
| 92 |
+
"learning_rate": 2.9580246913580254e-06,
|
| 93 |
+
"loss": 2.783578186035156,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1925925925925926,
|
| 98 |
+
"grad_norm": 1.9501953125,
|
| 99 |
+
"learning_rate": 3.2049382716049387e-06,
|
| 100 |
+
"loss": 2.7933184814453127,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.2074074074074074,
|
| 105 |
+
"grad_norm": 3.435546875,
|
| 106 |
+
"learning_rate": 3.4518518518518524e-06,
|
| 107 |
+
"loss": 2.80977783203125,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.2222222222222222,
|
| 112 |
+
"grad_norm": 1.8955078125,
|
| 113 |
+
"learning_rate": 3.6987654320987658e-06,
|
| 114 |
+
"loss": 2.7978311157226563,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.23703703703703705,
|
| 119 |
+
"grad_norm": 2.455078125,
|
| 120 |
+
"learning_rate": 3.94567901234568e-06,
|
| 121 |
+
"loss": 2.7951055908203126,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.2518518518518518,
|
| 126 |
+
"grad_norm": 4.1171875,
|
| 127 |
+
"learning_rate": 4.192592592592593e-06,
|
| 128 |
+
"loss": 2.785648193359375,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.26666666666666666,
|
| 133 |
+
"grad_norm": 1.84375,
|
| 134 |
+
"learning_rate": 4.439506172839507e-06,
|
| 135 |
+
"loss": 2.7951544189453124,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.2814814814814815,
|
| 140 |
+
"grad_norm": 2.359375,
|
| 141 |
+
"learning_rate": 4.68641975308642e-06,
|
| 142 |
+
"loss": 2.796750793457031,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.2962962962962963,
|
| 147 |
+
"grad_norm": 2.39453125,
|
| 148 |
+
"learning_rate": 4.933333333333334e-06,
|
| 149 |
+
"loss": 2.7995571899414062,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3111111111111111,
|
| 154 |
+
"grad_norm": 2.55078125,
|
| 155 |
+
"learning_rate": 5.180246913580247e-06,
|
| 156 |
+
"loss": 2.7804888916015624,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.32592592592592595,
|
| 161 |
+
"grad_norm": 3.67578125,
|
| 162 |
+
"learning_rate": 5.427160493827161e-06,
|
| 163 |
+
"loss": 2.792048034667969,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.34074074074074073,
|
| 168 |
+
"grad_norm": 2.130859375,
|
| 169 |
+
"learning_rate": 5.674074074074074e-06,
|
| 170 |
+
"loss": 2.789242858886719,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.35555555555555557,
|
| 175 |
+
"grad_norm": 2.412109375,
|
| 176 |
+
"learning_rate": 5.920987654320987e-06,
|
| 177 |
+
"loss": 2.788144836425781,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.37037037037037035,
|
| 182 |
+
"grad_norm": 2.72265625,
|
| 183 |
+
"learning_rate": 6.167901234567902e-06,
|
| 184 |
+
"loss": 2.8027859497070313,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.3851851851851852,
|
| 189 |
+
"grad_norm": 2.6484375,
|
| 190 |
+
"learning_rate": 6.414814814814815e-06,
|
| 191 |
+
"loss": 2.7877798461914063,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4,
|
| 196 |
+
"grad_norm": 2.3359375,
|
| 197 |
+
"learning_rate": 6.661728395061728e-06,
|
| 198 |
+
"loss": 2.7876470947265624,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.4148148148148148,
|
| 203 |
+
"grad_norm": 1.9970703125,
|
| 204 |
+
"learning_rate": 6.908641975308642e-06,
|
| 205 |
+
"loss": 2.77328857421875,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.42962962962962964,
|
| 210 |
+
"grad_norm": 1.875,
|
| 211 |
+
"learning_rate": 7.155555555555556e-06,
|
| 212 |
+
"loss": 2.79718017578125,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.4444444444444444,
|
| 217 |
+
"grad_norm": 2.248046875,
|
| 218 |
+
"learning_rate": 7.40246913580247e-06,
|
| 219 |
+
"loss": 2.7796466064453127,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.45925925925925926,
|
| 224 |
+
"grad_norm": 1.8623046875,
|
| 225 |
+
"learning_rate": 7.649382716049384e-06,
|
| 226 |
+
"loss": 2.7859207153320313,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.4740740740740741,
|
| 231 |
+
"grad_norm": 1.5302734375,
|
| 232 |
+
"learning_rate": 7.896296296296297e-06,
|
| 233 |
+
"loss": 2.781698913574219,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.4888888888888889,
|
| 238 |
+
"grad_norm": 1.3701171875,
|
| 239 |
+
"learning_rate": 8.14320987654321e-06,
|
| 240 |
+
"loss": 2.7813421630859376,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.5037037037037037,
|
| 245 |
+
"grad_norm": 1.40625,
|
| 246 |
+
"learning_rate": 8.390123456790124e-06,
|
| 247 |
+
"loss": 2.781036071777344,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.5185185185185185,
|
| 252 |
+
"grad_norm": 2.236328125,
|
| 253 |
+
"learning_rate": 8.637037037037037e-06,
|
| 254 |
+
"loss": 2.783014221191406,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.5333333333333333,
|
| 259 |
+
"grad_norm": 1.826171875,
|
| 260 |
+
"learning_rate": 8.883950617283952e-06,
|
| 261 |
+
"loss": 2.7817703247070313,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.5481481481481482,
|
| 266 |
+
"grad_norm": 1.82421875,
|
| 267 |
+
"learning_rate": 9.130864197530866e-06,
|
| 268 |
+
"loss": 2.7779510498046873,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.562962962962963,
|
| 273 |
+
"grad_norm": 2.1953125,
|
| 274 |
+
"learning_rate": 9.377777777777779e-06,
|
| 275 |
+
"loss": 2.7827484130859377,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.5777777777777777,
|
| 280 |
+
"grad_norm": 1.353515625,
|
| 281 |
+
"learning_rate": 9.624691358024692e-06,
|
| 282 |
+
"loss": 2.7817633056640627,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.5925925925925926,
|
| 287 |
+
"grad_norm": 2.69921875,
|
| 288 |
+
"learning_rate": 9.871604938271606e-06,
|
| 289 |
+
"loss": 2.7813009643554687,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.6074074074074074,
|
| 294 |
+
"grad_norm": 1.2294921875,
|
| 295 |
+
"learning_rate": 9.970370370370372e-06,
|
| 296 |
+
"loss": 2.77718994140625,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.6222222222222222,
|
| 301 |
+
"grad_norm": 1.6513671875,
|
| 302 |
+
"learning_rate": 9.908641975308642e-06,
|
| 303 |
+
"loss": 2.778031921386719,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.6370370370370371,
|
| 308 |
+
"grad_norm": 1.0498046875,
|
| 309 |
+
"learning_rate": 9.846913580246915e-06,
|
| 310 |
+
"loss": 2.778972473144531,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.6518518518518519,
|
| 315 |
+
"grad_norm": 1.7890625,
|
| 316 |
+
"learning_rate": 9.785185185185187e-06,
|
| 317 |
+
"loss": 2.7754339599609374,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.6666666666666666,
|
| 322 |
+
"grad_norm": 0.3349609375,
|
| 323 |
+
"learning_rate": 9.723456790123457e-06,
|
| 324 |
+
"loss": 2.7735556030273436,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.6814814814814815,
|
| 329 |
+
"grad_norm": 0.53271484375,
|
| 330 |
+
"learning_rate": 9.66172839506173e-06,
|
| 331 |
+
"loss": 2.7741458129882814,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.6962962962962963,
|
| 336 |
+
"grad_norm": 1.763671875,
|
| 337 |
+
"learning_rate": 9.600000000000001e-06,
|
| 338 |
+
"loss": 2.773301086425781,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.7111111111111111,
|
| 343 |
+
"grad_norm": 1.1982421875,
|
| 344 |
+
"learning_rate": 9.538271604938273e-06,
|
| 345 |
+
"loss": 2.7742254638671877,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.725925925925926,
|
| 350 |
+
"grad_norm": 2.26953125,
|
| 351 |
+
"learning_rate": 9.476543209876544e-06,
|
| 352 |
+
"loss": 2.7740939331054686,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.7407407407407407,
|
| 357 |
+
"grad_norm": NaN,
|
| 358 |
+
"learning_rate": 9.414814814814816e-06,
|
| 359 |
+
"loss": 3.09239013671875,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.7555555555555555,
|
| 364 |
+
"grad_norm": NaN,
|
| 365 |
+
"learning_rate": 9.353086419753086e-06,
|
| 366 |
+
"loss": 0.0,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.7703703703703704,
|
| 371 |
+
"grad_norm": NaN,
|
| 372 |
+
"learning_rate": 9.29135802469136e-06,
|
| 373 |
+
"loss": 0.0,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.7851851851851852,
|
| 378 |
+
"grad_norm": NaN,
|
| 379 |
+
"learning_rate": 9.22962962962963e-06,
|
| 380 |
+
"loss": 0.0,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.8,
|
| 385 |
+
"grad_norm": NaN,
|
| 386 |
+
"learning_rate": 9.167901234567903e-06,
|
| 387 |
+
"loss": 0.0,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.8148148148148148,
|
| 392 |
+
"grad_norm": NaN,
|
| 393 |
+
"learning_rate": 9.106172839506173e-06,
|
| 394 |
+
"loss": 0.0,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.8296296296296296,
|
| 399 |
+
"grad_norm": NaN,
|
| 400 |
+
"learning_rate": 9.044444444444445e-06,
|
| 401 |
+
"loss": 0.0,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.8444444444444444,
|
| 406 |
+
"grad_norm": NaN,
|
| 407 |
+
"learning_rate": 8.982716049382717e-06,
|
| 408 |
+
"loss": 0.0,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.8592592592592593,
|
| 413 |
+
"grad_norm": NaN,
|
| 414 |
+
"learning_rate": 8.92098765432099e-06,
|
| 415 |
+
"loss": 0.0,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.8740740740740741,
|
| 420 |
+
"grad_norm": NaN,
|
| 421 |
+
"learning_rate": 8.85925925925926e-06,
|
| 422 |
+
"loss": 0.0,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.8888888888888888,
|
| 427 |
+
"grad_norm": NaN,
|
| 428 |
+
"learning_rate": 8.797530864197532e-06,
|
| 429 |
+
"loss": 0.0,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.9037037037037037,
|
| 434 |
+
"grad_norm": NaN,
|
| 435 |
+
"learning_rate": 8.735802469135804e-06,
|
| 436 |
+
"loss": 0.0,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.9185185185185185,
|
| 441 |
+
"grad_norm": NaN,
|
| 442 |
+
"learning_rate": 8.674074074074074e-06,
|
| 443 |
+
"loss": 0.0,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.9333333333333333,
|
| 448 |
+
"grad_norm": NaN,
|
| 449 |
+
"learning_rate": 8.612345679012346e-06,
|
| 450 |
+
"loss": 0.0,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.9481481481481482,
|
| 455 |
+
"grad_norm": NaN,
|
| 456 |
+
"learning_rate": 8.550617283950618e-06,
|
| 457 |
+
"loss": 0.0,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.9629629629629629,
|
| 462 |
+
"grad_norm": NaN,
|
| 463 |
+
"learning_rate": 8.48888888888889e-06,
|
| 464 |
+
"loss": 0.0,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.9777777777777777,
|
| 469 |
+
"grad_norm": NaN,
|
| 470 |
+
"learning_rate": 8.42716049382716e-06,
|
| 471 |
+
"loss": 0.0,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.9925925925925926,
|
| 476 |
+
"grad_norm": NaN,
|
| 477 |
+
"learning_rate": 8.365432098765433e-06,
|
| 478 |
+
"loss": 0.0,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.0,
|
| 483 |
+
"eval_accuracy": 0.25075,
|
| 484 |
+
"eval_loss": NaN,
|
| 485 |
+
"eval_runtime": 17.5086,
|
| 486 |
+
"eval_samples_per_second": 685.376,
|
| 487 |
+
"eval_steps_per_second": 5.369,
|
| 488 |
+
"step": 3375
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.0074074074074073,
|
| 492 |
+
"grad_norm": NaN,
|
| 493 |
+
"learning_rate": 8.303703703703705e-06,
|
| 494 |
+
"loss": 0.0,
|
| 495 |
+
"step": 3400
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 1.0222222222222221,
|
| 499 |
+
"grad_norm": NaN,
|
| 500 |
+
"learning_rate": 8.241975308641975e-06,
|
| 501 |
+
"loss": 0.0,
|
| 502 |
+
"step": 3450
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 1.037037037037037,
|
| 506 |
+
"grad_norm": NaN,
|
| 507 |
+
"learning_rate": 8.180246913580248e-06,
|
| 508 |
+
"loss": 0.0,
|
| 509 |
+
"step": 3500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 1.0518518518518518,
|
| 513 |
+
"grad_norm": NaN,
|
| 514 |
+
"learning_rate": 8.11851851851852e-06,
|
| 515 |
+
"loss": 0.0,
|
| 516 |
+
"step": 3550
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 1.0666666666666667,
|
| 520 |
+
"grad_norm": NaN,
|
| 521 |
+
"learning_rate": 8.05679012345679e-06,
|
| 522 |
+
"loss": 0.0,
|
| 523 |
+
"step": 3600
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 1.0814814814814815,
|
| 527 |
+
"grad_norm": NaN,
|
| 528 |
+
"learning_rate": 7.995061728395062e-06,
|
| 529 |
+
"loss": 0.0,
|
| 530 |
+
"step": 3650
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 1.0962962962962963,
|
| 534 |
+
"grad_norm": NaN,
|
| 535 |
+
"learning_rate": 7.933333333333334e-06,
|
| 536 |
+
"loss": 0.0,
|
| 537 |
+
"step": 3700
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 1.1111111111111112,
|
| 541 |
+
"grad_norm": NaN,
|
| 542 |
+
"learning_rate": 7.871604938271605e-06,
|
| 543 |
+
"loss": 0.0,
|
| 544 |
+
"step": 3750
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 1.125925925925926,
|
| 548 |
+
"grad_norm": NaN,
|
| 549 |
+
"learning_rate": 7.809876543209877e-06,
|
| 550 |
+
"loss": 0.0,
|
| 551 |
+
"step": 3800
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 1.1407407407407408,
|
| 555 |
+
"grad_norm": NaN,
|
| 556 |
+
"learning_rate": 7.748148148148149e-06,
|
| 557 |
+
"loss": 0.0,
|
| 558 |
+
"step": 3850
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 1.1555555555555554,
|
| 562 |
+
"grad_norm": NaN,
|
| 563 |
+
"learning_rate": 7.68641975308642e-06,
|
| 564 |
+
"loss": 0.0,
|
| 565 |
+
"step": 3900
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 1.1703703703703703,
|
| 569 |
+
"grad_norm": NaN,
|
| 570 |
+
"learning_rate": 7.624691358024692e-06,
|
| 571 |
+
"loss": 0.0,
|
| 572 |
+
"step": 3950
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.1851851851851851,
|
| 576 |
+
"grad_norm": NaN,
|
| 577 |
+
"learning_rate": 7.562962962962963e-06,
|
| 578 |
+
"loss": 0.0,
|
| 579 |
+
"step": 4000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.2,
|
| 583 |
+
"grad_norm": NaN,
|
| 584 |
+
"learning_rate": 7.5012345679012354e-06,
|
| 585 |
+
"loss": 0.0,
|
| 586 |
+
"step": 4050
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 1.2148148148148148,
|
| 590 |
+
"grad_norm": NaN,
|
| 591 |
+
"learning_rate": 7.439506172839507e-06,
|
| 592 |
+
"loss": 0.0,
|
| 593 |
+
"step": 4100
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.2296296296296296,
|
| 597 |
+
"grad_norm": NaN,
|
| 598 |
+
"learning_rate": 7.377777777777778e-06,
|
| 599 |
+
"loss": 0.0,
|
| 600 |
+
"step": 4150
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 1.2444444444444445,
|
| 604 |
+
"grad_norm": NaN,
|
| 605 |
+
"learning_rate": 7.316049382716049e-06,
|
| 606 |
+
"loss": 0.0,
|
| 607 |
+
"step": 4200
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 1.2592592592592593,
|
| 611 |
+
"grad_norm": NaN,
|
| 612 |
+
"learning_rate": 7.254320987654322e-06,
|
| 613 |
+
"loss": 0.0,
|
| 614 |
+
"step": 4250
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 1.2740740740740741,
|
| 618 |
+
"grad_norm": NaN,
|
| 619 |
+
"learning_rate": 7.192592592592593e-06,
|
| 620 |
+
"loss": 0.0,
|
| 621 |
+
"step": 4300
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 1.2888888888888888,
|
| 625 |
+
"grad_norm": NaN,
|
| 626 |
+
"learning_rate": 7.130864197530865e-06,
|
| 627 |
+
"loss": 0.0,
|
| 628 |
+
"step": 4350
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 1.3037037037037038,
|
| 632 |
+
"grad_norm": NaN,
|
| 633 |
+
"learning_rate": 7.069135802469136e-06,
|
| 634 |
+
"loss": 0.0,
|
| 635 |
+
"step": 4400
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 1.3185185185185184,
|
| 639 |
+
"grad_norm": NaN,
|
| 640 |
+
"learning_rate": 7.007407407407408e-06,
|
| 641 |
+
"loss": 0.0,
|
| 642 |
+
"step": 4450
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 1.3333333333333333,
|
| 646 |
+
"grad_norm": NaN,
|
| 647 |
+
"learning_rate": 6.94567901234568e-06,
|
| 648 |
+
"loss": 0.0,
|
| 649 |
+
"step": 4500
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 1.348148148148148,
|
| 653 |
+
"grad_norm": NaN,
|
| 654 |
+
"learning_rate": 6.883950617283951e-06,
|
| 655 |
+
"loss": 0.0,
|
| 656 |
+
"step": 4550
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 1.362962962962963,
|
| 660 |
+
"grad_norm": NaN,
|
| 661 |
+
"learning_rate": 6.8222222222222225e-06,
|
| 662 |
+
"loss": 0.0,
|
| 663 |
+
"step": 4600
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 1.3777777777777778,
|
| 667 |
+
"grad_norm": NaN,
|
| 668 |
+
"learning_rate": 6.760493827160495e-06,
|
| 669 |
+
"loss": 0.0,
|
| 670 |
+
"step": 4650
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 1.3925925925925926,
|
| 674 |
+
"grad_norm": NaN,
|
| 675 |
+
"learning_rate": 6.698765432098766e-06,
|
| 676 |
+
"loss": 0.0,
|
| 677 |
+
"step": 4700
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 1.4074074074074074,
|
| 681 |
+
"grad_norm": NaN,
|
| 682 |
+
"learning_rate": 6.637037037037037e-06,
|
| 683 |
+
"loss": 0.0,
|
| 684 |
+
"step": 4750
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.4222222222222223,
|
| 688 |
+
"grad_norm": NaN,
|
| 689 |
+
"learning_rate": 6.575308641975308e-06,
|
| 690 |
+
"loss": 0.0,
|
| 691 |
+
"step": 4800
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 1.4370370370370371,
|
| 695 |
+
"grad_norm": NaN,
|
| 696 |
+
"learning_rate": 6.513580246913581e-06,
|
| 697 |
+
"loss": 0.0,
|
| 698 |
+
"step": 4850
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 1.4518518518518517,
|
| 702 |
+
"grad_norm": NaN,
|
| 703 |
+
"learning_rate": 6.4518518518518525e-06,
|
| 704 |
+
"loss": 0.0,
|
| 705 |
+
"step": 4900
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 1.4666666666666668,
|
| 709 |
+
"grad_norm": NaN,
|
| 710 |
+
"learning_rate": 6.390123456790124e-06,
|
| 711 |
+
"loss": 0.0,
|
| 712 |
+
"step": 4950
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 1.4814814814814814,
|
| 716 |
+
"grad_norm": NaN,
|
| 717 |
+
"learning_rate": 6.328395061728395e-06,
|
| 718 |
+
"loss": 0.0,
|
| 719 |
+
"step": 5000
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 1.4962962962962962,
|
| 723 |
+
"grad_norm": NaN,
|
| 724 |
+
"learning_rate": 6.266666666666668e-06,
|
| 725 |
+
"loss": 0.0,
|
| 726 |
+
"step": 5050
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 1.511111111111111,
|
| 730 |
+
"grad_norm": NaN,
|
| 731 |
+
"learning_rate": 6.204938271604939e-06,
|
| 732 |
+
"loss": 0.0,
|
| 733 |
+
"step": 5100
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 1.525925925925926,
|
| 737 |
+
"grad_norm": NaN,
|
| 738 |
+
"learning_rate": 6.1432098765432104e-06,
|
| 739 |
+
"loss": 0.0,
|
| 740 |
+
"step": 5150
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 1.5407407407407407,
|
| 744 |
+
"grad_norm": NaN,
|
| 745 |
+
"learning_rate": 6.081481481481482e-06,
|
| 746 |
+
"loss": 0.0,
|
| 747 |
+
"step": 5200
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 1.5555555555555556,
|
| 751 |
+
"grad_norm": NaN,
|
| 752 |
+
"learning_rate": 6.019753086419754e-06,
|
| 753 |
+
"loss": 0.0,
|
| 754 |
+
"step": 5250
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 1.5703703703703704,
|
| 758 |
+
"grad_norm": NaN,
|
| 759 |
+
"learning_rate": 5.958024691358025e-06,
|
| 760 |
+
"loss": 0.0,
|
| 761 |
+
"step": 5300
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 1.585185185185185,
|
| 765 |
+
"grad_norm": NaN,
|
| 766 |
+
"learning_rate": 5.896296296296296e-06,
|
| 767 |
+
"loss": 0.0,
|
| 768 |
+
"step": 5350
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.6,
|
| 772 |
+
"grad_norm": NaN,
|
| 773 |
+
"learning_rate": 5.834567901234568e-06,
|
| 774 |
+
"loss": 0.0,
|
| 775 |
+
"step": 5400
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 1.6148148148148147,
|
| 779 |
+
"grad_norm": NaN,
|
| 780 |
+
"learning_rate": 5.7728395061728404e-06,
|
| 781 |
+
"loss": 0.0,
|
| 782 |
+
"step": 5450
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 1.6296296296296298,
|
| 786 |
+
"grad_norm": NaN,
|
| 787 |
+
"learning_rate": 5.711111111111112e-06,
|
| 788 |
+
"loss": 0.0,
|
| 789 |
+
"step": 5500
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.6444444444444444,
|
| 793 |
+
"grad_norm": NaN,
|
| 794 |
+
"learning_rate": 5.649382716049383e-06,
|
| 795 |
+
"loss": 0.0,
|
| 796 |
+
"step": 5550
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 1.6592592592592592,
|
| 800 |
+
"grad_norm": NaN,
|
| 801 |
+
"learning_rate": 5.587654320987654e-06,
|
| 802 |
+
"loss": 0.0,
|
| 803 |
+
"step": 5600
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 1.674074074074074,
|
| 807 |
+
"grad_norm": NaN,
|
| 808 |
+
"learning_rate": 5.525925925925927e-06,
|
| 809 |
+
"loss": 0.0,
|
| 810 |
+
"step": 5650
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 1.6888888888888889,
|
| 814 |
+
"grad_norm": NaN,
|
| 815 |
+
"learning_rate": 5.464197530864198e-06,
|
| 816 |
+
"loss": 0.0,
|
| 817 |
+
"step": 5700
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 1.7037037037037037,
|
| 821 |
+
"grad_norm": NaN,
|
| 822 |
+
"learning_rate": 5.40246913580247e-06,
|
| 823 |
+
"loss": 0.0,
|
| 824 |
+
"step": 5750
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 1.7185185185185186,
|
| 828 |
+
"grad_norm": NaN,
|
| 829 |
+
"learning_rate": 5.340740740740741e-06,
|
| 830 |
+
"loss": 0.0,
|
| 831 |
+
"step": 5800
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 1.7333333333333334,
|
| 835 |
+
"grad_norm": NaN,
|
| 836 |
+
"learning_rate": 5.279012345679013e-06,
|
| 837 |
+
"loss": 0.0,
|
| 838 |
+
"step": 5850
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 1.748148148148148,
|
| 842 |
+
"grad_norm": NaN,
|
| 843 |
+
"learning_rate": 5.217283950617284e-06,
|
| 844 |
+
"loss": 0.0,
|
| 845 |
+
"step": 5900
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 1.762962962962963,
|
| 849 |
+
"grad_norm": NaN,
|
| 850 |
+
"learning_rate": 5.155555555555556e-06,
|
| 851 |
+
"loss": 0.0,
|
| 852 |
+
"step": 5950
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 1.7777777777777777,
|
| 856 |
+
"grad_norm": NaN,
|
| 857 |
+
"learning_rate": 5.0938271604938275e-06,
|
| 858 |
+
"loss": 0.0,
|
| 859 |
+
"step": 6000
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 1.7925925925925927,
|
| 863 |
+
"grad_norm": NaN,
|
| 864 |
+
"learning_rate": 5.0320987654321e-06,
|
| 865 |
+
"loss": 0.0,
|
| 866 |
+
"step": 6050
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 1.8074074074074074,
|
| 870 |
+
"grad_norm": NaN,
|
| 871 |
+
"learning_rate": 4.970370370370371e-06,
|
| 872 |
+
"loss": 0.0,
|
| 873 |
+
"step": 6100
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 1.8222222222222222,
|
| 877 |
+
"grad_norm": NaN,
|
| 878 |
+
"learning_rate": 4.908641975308642e-06,
|
| 879 |
+
"loss": 0.0,
|
| 880 |
+
"step": 6150
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 1.837037037037037,
|
| 884 |
+
"grad_norm": NaN,
|
| 885 |
+
"learning_rate": 4.846913580246914e-06,
|
| 886 |
+
"loss": 0.0,
|
| 887 |
+
"step": 6200
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 1.8518518518518519,
|
| 891 |
+
"grad_norm": NaN,
|
| 892 |
+
"learning_rate": 4.7851851851851854e-06,
|
| 893 |
+
"loss": 0.0,
|
| 894 |
+
"step": 6250
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 1.8666666666666667,
|
| 898 |
+
"grad_norm": NaN,
|
| 899 |
+
"learning_rate": 4.7234567901234575e-06,
|
| 900 |
+
"loss": 0.0,
|
| 901 |
+
"step": 6300
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 1.8814814814814815,
|
| 905 |
+
"grad_norm": NaN,
|
| 906 |
+
"learning_rate": 4.661728395061729e-06,
|
| 907 |
+
"loss": 0.0,
|
| 908 |
+
"step": 6350
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 1.8962962962962964,
|
| 912 |
+
"grad_norm": NaN,
|
| 913 |
+
"learning_rate": 4.600000000000001e-06,
|
| 914 |
+
"loss": 0.0,
|
| 915 |
+
"step": 6400
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 1.911111111111111,
|
| 919 |
+
"grad_norm": NaN,
|
| 920 |
+
"learning_rate": 4.538271604938272e-06,
|
| 921 |
+
"loss": 0.0,
|
| 922 |
+
"step": 6450
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 1.925925925925926,
|
| 926 |
+
"grad_norm": NaN,
|
| 927 |
+
"learning_rate": 4.476543209876544e-06,
|
| 928 |
+
"loss": 0.0,
|
| 929 |
+
"step": 6500
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 1.9407407407407407,
|
| 933 |
+
"grad_norm": NaN,
|
| 934 |
+
"learning_rate": 4.4148148148148154e-06,
|
| 935 |
+
"loss": 0.0,
|
| 936 |
+
"step": 6550
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 1.9555555555555557,
|
| 940 |
+
"grad_norm": NaN,
|
| 941 |
+
"learning_rate": 4.353086419753087e-06,
|
| 942 |
+
"loss": 0.0,
|
| 943 |
+
"step": 6600
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 1.9703703703703703,
|
| 947 |
+
"grad_norm": NaN,
|
| 948 |
+
"learning_rate": 4.291358024691358e-06,
|
| 949 |
+
"loss": 0.0,
|
| 950 |
+
"step": 6650
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 1.9851851851851852,
|
| 954 |
+
"grad_norm": NaN,
|
| 955 |
+
"learning_rate": 4.22962962962963e-06,
|
| 956 |
+
"loss": 0.0,
|
| 957 |
+
"step": 6700
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 2.0,
|
| 961 |
+
"grad_norm": NaN,
|
| 962 |
+
"learning_rate": 4.167901234567901e-06,
|
| 963 |
+
"loss": 0.0,
|
| 964 |
+
"step": 6750
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 2.0,
|
| 968 |
+
"eval_accuracy": 0.25075,
|
| 969 |
+
"eval_loss": NaN,
|
| 970 |
+
"eval_runtime": 17.7609,
|
| 971 |
+
"eval_samples_per_second": 675.641,
|
| 972 |
+
"eval_steps_per_second": 5.293,
|
| 973 |
+
"step": 6750
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 2.0148148148148146,
|
| 977 |
+
"grad_norm": NaN,
|
| 978 |
+
"learning_rate": 4.106172839506173e-06,
|
| 979 |
+
"loss": 0.0,
|
| 980 |
+
"step": 6800
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 2.0296296296296297,
|
| 984 |
+
"grad_norm": NaN,
|
| 985 |
+
"learning_rate": 4.044444444444445e-06,
|
| 986 |
+
"loss": 0.0,
|
| 987 |
+
"step": 6850
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 2.0444444444444443,
|
| 991 |
+
"grad_norm": NaN,
|
| 992 |
+
"learning_rate": 3.982716049382717e-06,
|
| 993 |
+
"loss": 0.0,
|
| 994 |
+
"step": 6900
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 2.0592592592592593,
|
| 998 |
+
"grad_norm": NaN,
|
| 999 |
+
"learning_rate": 3.920987654320988e-06,
|
| 1000 |
+
"loss": 0.0,
|
| 1001 |
+
"step": 6950
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 2.074074074074074,
|
| 1005 |
+
"grad_norm": NaN,
|
| 1006 |
+
"learning_rate": 3.85925925925926e-06,
|
| 1007 |
+
"loss": 0.0,
|
| 1008 |
+
"step": 7000
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 2.088888888888889,
|
| 1012 |
+
"grad_norm": NaN,
|
| 1013 |
+
"learning_rate": 3.7975308641975313e-06,
|
| 1014 |
+
"loss": 0.0,
|
| 1015 |
+
"step": 7050
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 2.1037037037037036,
|
| 1019 |
+
"grad_norm": NaN,
|
| 1020 |
+
"learning_rate": 3.735802469135803e-06,
|
| 1021 |
+
"loss": 0.0,
|
| 1022 |
+
"step": 7100
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 2.1185185185185187,
|
| 1026 |
+
"grad_norm": NaN,
|
| 1027 |
+
"learning_rate": 3.674074074074074e-06,
|
| 1028 |
+
"loss": 0.0,
|
| 1029 |
+
"step": 7150
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 2.1333333333333333,
|
| 1033 |
+
"grad_norm": NaN,
|
| 1034 |
+
"learning_rate": 3.6123456790123463e-06,
|
| 1035 |
+
"loss": 0.0,
|
| 1036 |
+
"step": 7200
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 2.148148148148148,
|
| 1040 |
+
"grad_norm": NaN,
|
| 1041 |
+
"learning_rate": 3.5506172839506175e-06,
|
| 1042 |
+
"loss": 0.0,
|
| 1043 |
+
"step": 7250
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 2.162962962962963,
|
| 1047 |
+
"grad_norm": NaN,
|
| 1048 |
+
"learning_rate": 3.4888888888888896e-06,
|
| 1049 |
+
"loss": 0.0,
|
| 1050 |
+
"step": 7300
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 2.1777777777777776,
|
| 1054 |
+
"grad_norm": NaN,
|
| 1055 |
+
"learning_rate": 3.427160493827161e-06,
|
| 1056 |
+
"loss": 0.0,
|
| 1057 |
+
"step": 7350
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 2.1925925925925926,
|
| 1061 |
+
"grad_norm": NaN,
|
| 1062 |
+
"learning_rate": 3.3654320987654325e-06,
|
| 1063 |
+
"loss": 0.0,
|
| 1064 |
+
"step": 7400
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 2.2074074074074073,
|
| 1068 |
+
"grad_norm": NaN,
|
| 1069 |
+
"learning_rate": 3.3037037037037038e-06,
|
| 1070 |
+
"loss": 0.0,
|
| 1071 |
+
"step": 7450
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 2.2222222222222223,
|
| 1075 |
+
"grad_norm": NaN,
|
| 1076 |
+
"learning_rate": 3.241975308641976e-06,
|
| 1077 |
+
"loss": 0.0,
|
| 1078 |
+
"step": 7500
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 2.237037037037037,
|
| 1082 |
+
"grad_norm": NaN,
|
| 1083 |
+
"learning_rate": 3.180246913580247e-06,
|
| 1084 |
+
"loss": 0.0,
|
| 1085 |
+
"step": 7550
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 2.251851851851852,
|
| 1089 |
+
"grad_norm": NaN,
|
| 1090 |
+
"learning_rate": 3.1185185185185183e-06,
|
| 1091 |
+
"loss": 0.0,
|
| 1092 |
+
"step": 7600
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 2.2666666666666666,
|
| 1096 |
+
"grad_norm": NaN,
|
| 1097 |
+
"learning_rate": 3.0567901234567904e-06,
|
| 1098 |
+
"loss": 0.0,
|
| 1099 |
+
"step": 7650
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 2.2814814814814817,
|
| 1103 |
+
"grad_norm": NaN,
|
| 1104 |
+
"learning_rate": 2.9950617283950617e-06,
|
| 1105 |
+
"loss": 0.0,
|
| 1106 |
+
"step": 7700
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 2.2962962962962963,
|
| 1110 |
+
"grad_norm": NaN,
|
| 1111 |
+
"learning_rate": 2.9333333333333338e-06,
|
| 1112 |
+
"loss": 0.0,
|
| 1113 |
+
"step": 7750
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 2.311111111111111,
|
| 1117 |
+
"grad_norm": NaN,
|
| 1118 |
+
"learning_rate": 2.871604938271605e-06,
|
| 1119 |
+
"loss": 0.0,
|
| 1120 |
+
"step": 7800
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 2.325925925925926,
|
| 1124 |
+
"grad_norm": NaN,
|
| 1125 |
+
"learning_rate": 2.8098765432098767e-06,
|
| 1126 |
+
"loss": 0.0,
|
| 1127 |
+
"step": 7850
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 2.3407407407407406,
|
| 1131 |
+
"grad_norm": NaN,
|
| 1132 |
+
"learning_rate": 2.748148148148148e-06,
|
| 1133 |
+
"loss": 0.0,
|
| 1134 |
+
"step": 7900
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 2.3555555555555556,
|
| 1138 |
+
"grad_norm": NaN,
|
| 1139 |
+
"learning_rate": 2.68641975308642e-06,
|
| 1140 |
+
"loss": 0.0,
|
| 1141 |
+
"step": 7950
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 2.3703703703703702,
|
| 1145 |
+
"grad_norm": NaN,
|
| 1146 |
+
"learning_rate": 2.6246913580246913e-06,
|
| 1147 |
+
"loss": 0.0,
|
| 1148 |
+
"step": 8000
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 2.3851851851851853,
|
| 1152 |
+
"grad_norm": NaN,
|
| 1153 |
+
"learning_rate": 2.5629629629629633e-06,
|
| 1154 |
+
"loss": 0.0,
|
| 1155 |
+
"step": 8050
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 2.4,
|
| 1159 |
+
"grad_norm": NaN,
|
| 1160 |
+
"learning_rate": 2.5012345679012346e-06,
|
| 1161 |
+
"loss": 0.0,
|
| 1162 |
+
"step": 8100
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 2.414814814814815,
|
| 1166 |
+
"grad_norm": NaN,
|
| 1167 |
+
"learning_rate": 2.4395061728395063e-06,
|
| 1168 |
+
"loss": 0.0,
|
| 1169 |
+
"step": 8150
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 2.4296296296296296,
|
| 1173 |
+
"grad_norm": NaN,
|
| 1174 |
+
"learning_rate": 2.377777777777778e-06,
|
| 1175 |
+
"loss": 0.0,
|
| 1176 |
+
"step": 8200
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 2.4444444444444446,
|
| 1180 |
+
"grad_norm": NaN,
|
| 1181 |
+
"learning_rate": 2.3160493827160496e-06,
|
| 1182 |
+
"loss": 0.0,
|
| 1183 |
+
"step": 8250
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 2.4592592592592593,
|
| 1187 |
+
"grad_norm": NaN,
|
| 1188 |
+
"learning_rate": 2.2543209876543213e-06,
|
| 1189 |
+
"loss": 0.0,
|
| 1190 |
+
"step": 8300
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 2.474074074074074,
|
| 1194 |
+
"grad_norm": NaN,
|
| 1195 |
+
"learning_rate": 2.192592592592593e-06,
|
| 1196 |
+
"loss": 0.0,
|
| 1197 |
+
"step": 8350
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 2.488888888888889,
|
| 1201 |
+
"grad_norm": NaN,
|
| 1202 |
+
"learning_rate": 2.1308641975308646e-06,
|
| 1203 |
+
"loss": 0.0,
|
| 1204 |
+
"step": 8400
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 2.5037037037037035,
|
| 1208 |
+
"grad_norm": NaN,
|
| 1209 |
+
"learning_rate": 2.069135802469136e-06,
|
| 1210 |
+
"loss": 0.0,
|
| 1211 |
+
"step": 8450
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 2.5185185185185186,
|
| 1215 |
+
"grad_norm": NaN,
|
| 1216 |
+
"learning_rate": 2.0074074074074075e-06,
|
| 1217 |
+
"loss": 0.0,
|
| 1218 |
+
"step": 8500
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 2.533333333333333,
|
| 1222 |
+
"grad_norm": NaN,
|
| 1223 |
+
"learning_rate": 1.945679012345679e-06,
|
| 1224 |
+
"loss": 0.0,
|
| 1225 |
+
"step": 8550
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 2.5481481481481483,
|
| 1229 |
+
"grad_norm": NaN,
|
| 1230 |
+
"learning_rate": 1.8839506172839508e-06,
|
| 1231 |
+
"loss": 0.0,
|
| 1232 |
+
"step": 8600
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 2.562962962962963,
|
| 1236 |
+
"grad_norm": NaN,
|
| 1237 |
+
"learning_rate": 1.8222222222222225e-06,
|
| 1238 |
+
"loss": 0.0,
|
| 1239 |
+
"step": 8650
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 2.5777777777777775,
|
| 1243 |
+
"grad_norm": NaN,
|
| 1244 |
+
"learning_rate": 1.760493827160494e-06,
|
| 1245 |
+
"loss": 0.0,
|
| 1246 |
+
"step": 8700
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 2.5925925925925926,
|
| 1250 |
+
"grad_norm": NaN,
|
| 1251 |
+
"learning_rate": 1.6987654320987656e-06,
|
| 1252 |
+
"loss": 0.0,
|
| 1253 |
+
"step": 8750
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 2.6074074074074076,
|
| 1257 |
+
"grad_norm": NaN,
|
| 1258 |
+
"learning_rate": 1.6370370370370373e-06,
|
| 1259 |
+
"loss": 0.0,
|
| 1260 |
+
"step": 8800
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 2.6222222222222222,
|
| 1264 |
+
"grad_norm": NaN,
|
| 1265 |
+
"learning_rate": 1.5753086419753088e-06,
|
| 1266 |
+
"loss": 0.0,
|
| 1267 |
+
"step": 8850
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 2.637037037037037,
|
| 1271 |
+
"grad_norm": NaN,
|
| 1272 |
+
"learning_rate": 1.5135802469135804e-06,
|
| 1273 |
+
"loss": 0.0,
|
| 1274 |
+
"step": 8900
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 2.651851851851852,
|
| 1278 |
+
"grad_norm": NaN,
|
| 1279 |
+
"learning_rate": 1.451851851851852e-06,
|
| 1280 |
+
"loss": 0.0,
|
| 1281 |
+
"step": 8950
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 2.6666666666666665,
|
| 1285 |
+
"grad_norm": NaN,
|
| 1286 |
+
"learning_rate": 1.3901234567901238e-06,
|
| 1287 |
+
"loss": 0.0,
|
| 1288 |
+
"step": 9000
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 2.6814814814814816,
|
| 1292 |
+
"grad_norm": NaN,
|
| 1293 |
+
"learning_rate": 1.3283950617283952e-06,
|
| 1294 |
+
"loss": 0.0,
|
| 1295 |
+
"step": 9050
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 2.696296296296296,
|
| 1299 |
+
"grad_norm": NaN,
|
| 1300 |
+
"learning_rate": 1.2666666666666669e-06,
|
| 1301 |
+
"loss": 0.0,
|
| 1302 |
+
"step": 9100
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 2.7111111111111112,
|
| 1306 |
+
"grad_norm": NaN,
|
| 1307 |
+
"learning_rate": 1.2049382716049383e-06,
|
| 1308 |
+
"loss": 0.0,
|
| 1309 |
+
"step": 9150
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 2.725925925925926,
|
| 1313 |
+
"grad_norm": NaN,
|
| 1314 |
+
"learning_rate": 1.14320987654321e-06,
|
| 1315 |
+
"loss": 0.0,
|
| 1316 |
+
"step": 9200
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 2.7407407407407405,
|
| 1320 |
+
"grad_norm": NaN,
|
| 1321 |
+
"learning_rate": 1.0814814814814815e-06,
|
| 1322 |
+
"loss": 0.0,
|
| 1323 |
+
"step": 9250
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 2.7555555555555555,
|
| 1327 |
+
"grad_norm": NaN,
|
| 1328 |
+
"learning_rate": 1.0197530864197531e-06,
|
| 1329 |
+
"loss": 0.0,
|
| 1330 |
+
"step": 9300
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 2.7703703703703706,
|
| 1334 |
+
"grad_norm": NaN,
|
| 1335 |
+
"learning_rate": 9.580246913580248e-07,
|
| 1336 |
+
"loss": 0.0,
|
| 1337 |
+
"step": 9350
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 2.785185185185185,
|
| 1341 |
+
"grad_norm": NaN,
|
| 1342 |
+
"learning_rate": 8.962962962962964e-07,
|
| 1343 |
+
"loss": 0.0,
|
| 1344 |
+
"step": 9400
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 2.8,
|
| 1348 |
+
"grad_norm": NaN,
|
| 1349 |
+
"learning_rate": 8.345679012345679e-07,
|
| 1350 |
+
"loss": 0.0,
|
| 1351 |
+
"step": 9450
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 2.814814814814815,
|
| 1355 |
+
"grad_norm": NaN,
|
| 1356 |
+
"learning_rate": 7.728395061728396e-07,
|
| 1357 |
+
"loss": 0.0,
|
| 1358 |
+
"step": 9500
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 2.8296296296296295,
|
| 1362 |
+
"grad_norm": NaN,
|
| 1363 |
+
"learning_rate": 7.111111111111112e-07,
|
| 1364 |
+
"loss": 0.0,
|
| 1365 |
+
"step": 9550
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 2.8444444444444446,
|
| 1369 |
+
"grad_norm": NaN,
|
| 1370 |
+
"learning_rate": 6.493827160493827e-07,
|
| 1371 |
+
"loss": 0.0,
|
| 1372 |
+
"step": 9600
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 2.859259259259259,
|
| 1376 |
+
"grad_norm": NaN,
|
| 1377 |
+
"learning_rate": 5.876543209876544e-07,
|
| 1378 |
+
"loss": 0.0,
|
| 1379 |
+
"step": 9650
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 2.8740740740740742,
|
| 1383 |
+
"grad_norm": NaN,
|
| 1384 |
+
"learning_rate": 5.25925925925926e-07,
|
| 1385 |
+
"loss": 0.0,
|
| 1386 |
+
"step": 9700
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 2.888888888888889,
|
| 1390 |
+
"grad_norm": NaN,
|
| 1391 |
+
"learning_rate": 4.6419753086419757e-07,
|
| 1392 |
+
"loss": 0.0,
|
| 1393 |
+
"step": 9750
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 2.9037037037037035,
|
| 1397 |
+
"grad_norm": NaN,
|
| 1398 |
+
"learning_rate": 4.024691358024692e-07,
|
| 1399 |
+
"loss": 0.0,
|
| 1400 |
+
"step": 9800
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 2.9185185185185185,
|
| 1404 |
+
"grad_norm": NaN,
|
| 1405 |
+
"learning_rate": 3.407407407407408e-07,
|
| 1406 |
+
"loss": 0.0,
|
| 1407 |
+
"step": 9850
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 2.9333333333333336,
|
| 1411 |
+
"grad_norm": NaN,
|
| 1412 |
+
"learning_rate": 2.7901234567901236e-07,
|
| 1413 |
+
"loss": 0.0,
|
| 1414 |
+
"step": 9900
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 2.948148148148148,
|
| 1418 |
+
"grad_norm": NaN,
|
| 1419 |
+
"learning_rate": 2.1728395061728395e-07,
|
| 1420 |
+
"loss": 0.0,
|
| 1421 |
+
"step": 9950
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 2.962962962962963,
|
| 1425 |
+
"grad_norm": NaN,
|
| 1426 |
+
"learning_rate": 1.5555555555555556e-07,
|
| 1427 |
+
"loss": 0.0,
|
| 1428 |
+
"step": 10000
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 2.977777777777778,
|
| 1432 |
+
"grad_norm": NaN,
|
| 1433 |
+
"learning_rate": 9.382716049382716e-08,
|
| 1434 |
+
"loss": 0.0,
|
| 1435 |
+
"step": 10050
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 2.9925925925925925,
|
| 1439 |
+
"grad_norm": NaN,
|
| 1440 |
+
"learning_rate": 3.2098765432098767e-08,
|
| 1441 |
+
"loss": 0.0,
|
| 1442 |
+
"step": 10100
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 3.0,
|
| 1446 |
+
"eval_accuracy": 0.25075,
|
| 1447 |
+
"eval_loss": NaN,
|
| 1448 |
+
"eval_runtime": 17.5467,
|
| 1449 |
+
"eval_samples_per_second": 683.891,
|
| 1450 |
+
"eval_steps_per_second": 5.357,
|
| 1451 |
+
"step": 10125
|
| 1452 |
+
}
|
| 1453 |
+
],
|
| 1454 |
+
"logging_steps": 50,
|
| 1455 |
+
"max_steps": 10125,
|
| 1456 |
+
"num_input_tokens_seen": 0,
|
| 1457 |
+
"num_train_epochs": 3,
|
| 1458 |
+
"save_steps": 500,
|
| 1459 |
+
"stateful_callbacks": {
|
| 1460 |
+
"EarlyStoppingCallback": {
|
| 1461 |
+
"args": {
|
| 1462 |
+
"early_stopping_patience": 2,
|
| 1463 |
+
"early_stopping_threshold": 0.0
|
| 1464 |
+
},
|
| 1465 |
+
"attributes": {
|
| 1466 |
+
"early_stopping_patience_counter": 2
|
| 1467 |
+
}
|
| 1468 |
+
},
|
| 1469 |
+
"TrainerControl": {
|
| 1470 |
+
"args": {
|
| 1471 |
+
"should_epoch_stop": false,
|
| 1472 |
+
"should_evaluate": false,
|
| 1473 |
+
"should_log": false,
|
| 1474 |
+
"should_save": true,
|
| 1475 |
+
"should_training_stop": true
|
| 1476 |
+
},
|
| 1477 |
+
"attributes": {}
|
| 1478 |
+
}
|
| 1479 |
+
},
|
| 1480 |
+
"total_flos": 2.1312760393728e+16,
|
| 1481 |
+
"train_batch_size": 32,
|
| 1482 |
+
"trial_name": null,
|
| 1483 |
+
"trial_params": null
|
| 1484 |
+
}
|
checkpoint-10125/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f2eba2f8b924060a23a0dbbe58e92be52a7eaf1410ec02a5a8c210eceb8446e
|
| 3 |
+
size 5201
|
checkpoint-3375/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 737873099
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8676f3e532480df9560eaea61e43cecaed0e47a1031a0e37dfe8f16571916b67
|
| 3 |
size 737873099
|
checkpoint-3375/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d0af5521b2caef4aa44bf07bd1f741ca9ab7bda6fd9c4edc73e94487a55becf
|
| 3 |
size 1465
|
checkpoint-3375/trainer_state.json
CHANGED
|
@@ -11,470 +11,470 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.014814814814814815,
|
| 14 |
-
"grad_norm":
|
| 15 |
-
"learning_rate":
|
| 16 |
-
"loss": 2.
|
| 17 |
"step": 50
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.02962962962962963,
|
| 21 |
-
"grad_norm":
|
| 22 |
-
"learning_rate":
|
| 23 |
-
"loss": 2.
|
| 24 |
"step": 100
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.044444444444444446,
|
| 28 |
-
"grad_norm":
|
| 29 |
-
"learning_rate":
|
| 30 |
-
"loss": 2.
|
| 31 |
"step": 150
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.05925925925925926,
|
| 35 |
-
"grad_norm":
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss": 3.
|
| 38 |
"step": 200
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.07407407407407407,
|
| 42 |
-
"grad_norm": 1.
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 2.
|
| 45 |
"step": 250
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.08888888888888889,
|
| 49 |
-
"grad_norm": 2.
|
| 50 |
-
"learning_rate":
|
| 51 |
-
"loss": 2.
|
| 52 |
"step": 300
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.1037037037037037,
|
| 56 |
-
"grad_norm":
|
| 57 |
-
"learning_rate":
|
| 58 |
-
"loss": 2.
|
| 59 |
"step": 350
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.11851851851851852,
|
| 63 |
-
"grad_norm": 2.
|
| 64 |
-
"learning_rate":
|
| 65 |
-
"loss": 2.
|
| 66 |
"step": 400
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.13333333333333333,
|
| 70 |
-
"grad_norm":
|
| 71 |
-
"learning_rate":
|
| 72 |
-
"loss": 2.
|
| 73 |
"step": 450
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.14814814814814814,
|
| 77 |
-
"grad_norm":
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 2.
|
| 80 |
"step": 500
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.16296296296296298,
|
| 84 |
-
"grad_norm": 2.
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss": 2.
|
| 87 |
"step": 550
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.17777777777777778,
|
| 91 |
-
"grad_norm":
|
| 92 |
-
"learning_rate":
|
| 93 |
-
"loss": 2.
|
| 94 |
"step": 600
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 0.1925925925925926,
|
| 98 |
-
"grad_norm": 1.
|
| 99 |
-
"learning_rate":
|
| 100 |
-
"loss": 2.
|
| 101 |
"step": 650
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 0.2074074074074074,
|
| 105 |
-
"grad_norm":
|
| 106 |
-
"learning_rate":
|
| 107 |
-
"loss": 2.
|
| 108 |
"step": 700
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 0.2222222222222222,
|
| 112 |
-
"grad_norm": 1.
|
| 113 |
-
"learning_rate":
|
| 114 |
-
"loss": 2.
|
| 115 |
"step": 750
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 0.23703703703703705,
|
| 119 |
-
"grad_norm":
|
| 120 |
-
"learning_rate":
|
| 121 |
-
"loss": 2.
|
| 122 |
"step": 800
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.2518518518518518,
|
| 126 |
-
"grad_norm":
|
| 127 |
-
"learning_rate":
|
| 128 |
-
"loss": 2.
|
| 129 |
"step": 850
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.26666666666666666,
|
| 133 |
-
"grad_norm": 1.
|
| 134 |
-
"learning_rate":
|
| 135 |
-
"loss": 2.
|
| 136 |
"step": 900
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 0.2814814814814815,
|
| 140 |
-
"grad_norm":
|
| 141 |
-
"learning_rate":
|
| 142 |
-
"loss": 2.
|
| 143 |
"step": 950
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.2962962962962963,
|
| 147 |
-
"grad_norm":
|
| 148 |
-
"learning_rate":
|
| 149 |
-
"loss": 2.
|
| 150 |
"step": 1000
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 0.3111111111111111,
|
| 154 |
-
"grad_norm": 2.
|
| 155 |
-
"learning_rate":
|
| 156 |
-
"loss": 2.
|
| 157 |
"step": 1050
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 0.32592592592592595,
|
| 161 |
-
"grad_norm":
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 2.
|
| 164 |
"step": 1100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 0.34074074074074073,
|
| 168 |
-
"grad_norm":
|
| 169 |
-
"learning_rate":
|
| 170 |
-
"loss": 2.
|
| 171 |
"step": 1150
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 0.35555555555555557,
|
| 175 |
-
"grad_norm": 2.
|
| 176 |
-
"learning_rate":
|
| 177 |
-
"loss": 2.
|
| 178 |
"step": 1200
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 0.37037037037037035,
|
| 182 |
-
"grad_norm": 2.
|
| 183 |
-
"learning_rate":
|
| 184 |
-
"loss": 2.
|
| 185 |
"step": 1250
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.3851851851851852,
|
| 189 |
-
"grad_norm": 2.
|
| 190 |
-
"learning_rate":
|
| 191 |
-
"loss": 2.
|
| 192 |
"step": 1300
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 0.4,
|
| 196 |
-
"grad_norm":
|
| 197 |
-
"learning_rate":
|
| 198 |
-
"loss": 2.
|
| 199 |
"step": 1350
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 0.4148148148148148,
|
| 203 |
-
"grad_norm": 1.
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss": 2.
|
| 206 |
"step": 1400
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 0.42962962962962964,
|
| 210 |
-
"grad_norm": 1.
|
| 211 |
-
"learning_rate":
|
| 212 |
-
"loss": 2.
|
| 213 |
"step": 1450
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 0.4444444444444444,
|
| 217 |
-
"grad_norm":
|
| 218 |
-
"learning_rate":
|
| 219 |
-
"loss":
|
| 220 |
"step": 1500
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 0.45925925925925926,
|
| 224 |
-
"grad_norm":
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss":
|
| 227 |
"step": 1550
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 0.4740740740740741,
|
| 231 |
-
"grad_norm":
|
| 232 |
-
"learning_rate":
|
| 233 |
-
"loss":
|
| 234 |
"step": 1600
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 0.4888888888888889,
|
| 238 |
-
"grad_norm":
|
| 239 |
-
"learning_rate":
|
| 240 |
-
"loss":
|
| 241 |
"step": 1650
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 0.5037037037037037,
|
| 245 |
-
"grad_norm":
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss":
|
| 248 |
"step": 1700
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 0.5185185185185185,
|
| 252 |
-
"grad_norm":
|
| 253 |
-
"learning_rate":
|
| 254 |
-
"loss":
|
| 255 |
"step": 1750
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 0.5333333333333333,
|
| 259 |
-
"grad_norm":
|
| 260 |
-
"learning_rate":
|
| 261 |
-
"loss":
|
| 262 |
"step": 1800
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 0.5481481481481482,
|
| 266 |
-
"grad_norm":
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss":
|
| 269 |
"step": 1850
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 0.562962962962963,
|
| 273 |
-
"grad_norm":
|
| 274 |
-
"learning_rate":
|
| 275 |
-
"loss":
|
| 276 |
"step": 1900
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 0.5777777777777777,
|
| 280 |
-
"grad_norm":
|
| 281 |
-
"learning_rate":
|
| 282 |
-
"loss":
|
| 283 |
"step": 1950
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 0.5925925925925926,
|
| 287 |
-
"grad_norm":
|
| 288 |
-
"learning_rate":
|
| 289 |
-
"loss":
|
| 290 |
"step": 2000
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 0.6074074074074074,
|
| 294 |
-
"grad_norm":
|
| 295 |
-
"learning_rate":
|
| 296 |
-
"loss":
|
| 297 |
"step": 2050
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 0.6222222222222222,
|
| 301 |
-
"grad_norm":
|
| 302 |
-
"learning_rate":
|
| 303 |
-
"loss":
|
| 304 |
"step": 2100
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 0.6370370370370371,
|
| 308 |
-
"grad_norm":
|
| 309 |
-
"learning_rate":
|
| 310 |
-
"loss":
|
| 311 |
"step": 2150
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 0.6518518518518519,
|
| 315 |
-
"grad_norm":
|
| 316 |
-
"learning_rate":
|
| 317 |
-
"loss":
|
| 318 |
"step": 2200
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 0.6666666666666666,
|
| 322 |
-
"grad_norm":
|
| 323 |
-
"learning_rate":
|
| 324 |
-
"loss":
|
| 325 |
"step": 2250
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 0.6814814814814815,
|
| 329 |
-
"grad_norm":
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss":
|
| 332 |
"step": 2300
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 0.6962962962962963,
|
| 336 |
-
"grad_norm":
|
| 337 |
-
"learning_rate":
|
| 338 |
-
"loss":
|
| 339 |
"step": 2350
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 0.7111111111111111,
|
| 343 |
-
"grad_norm":
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"loss":
|
| 346 |
"step": 2400
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 0.725925925925926,
|
| 350 |
-
"grad_norm":
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss":
|
| 353 |
"step": 2450
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 0.7407407407407407,
|
| 357 |
"grad_norm": NaN,
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss":
|
| 360 |
"step": 2500
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 0.7555555555555555,
|
| 364 |
"grad_norm": NaN,
|
| 365 |
-
"learning_rate":
|
| 366 |
"loss": 0.0,
|
| 367 |
"step": 2550
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 0.7703703703703704,
|
| 371 |
"grad_norm": NaN,
|
| 372 |
-
"learning_rate":
|
| 373 |
"loss": 0.0,
|
| 374 |
"step": 2600
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 0.7851851851851852,
|
| 378 |
"grad_norm": NaN,
|
| 379 |
-
"learning_rate":
|
| 380 |
"loss": 0.0,
|
| 381 |
"step": 2650
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 0.8,
|
| 385 |
"grad_norm": NaN,
|
| 386 |
-
"learning_rate":
|
| 387 |
"loss": 0.0,
|
| 388 |
"step": 2700
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 0.8148148148148148,
|
| 392 |
"grad_norm": NaN,
|
| 393 |
-
"learning_rate":
|
| 394 |
"loss": 0.0,
|
| 395 |
"step": 2750
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 0.8296296296296296,
|
| 399 |
"grad_norm": NaN,
|
| 400 |
-
"learning_rate":
|
| 401 |
"loss": 0.0,
|
| 402 |
"step": 2800
|
| 403 |
},
|
| 404 |
{
|
| 405 |
"epoch": 0.8444444444444444,
|
| 406 |
"grad_norm": NaN,
|
| 407 |
-
"learning_rate":
|
| 408 |
"loss": 0.0,
|
| 409 |
"step": 2850
|
| 410 |
},
|
| 411 |
{
|
| 412 |
"epoch": 0.8592592592592593,
|
| 413 |
"grad_norm": NaN,
|
| 414 |
-
"learning_rate":
|
| 415 |
"loss": 0.0,
|
| 416 |
"step": 2900
|
| 417 |
},
|
| 418 |
{
|
| 419 |
"epoch": 0.8740740740740741,
|
| 420 |
"grad_norm": NaN,
|
| 421 |
-
"learning_rate":
|
| 422 |
"loss": 0.0,
|
| 423 |
"step": 2950
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 0.8888888888888888,
|
| 427 |
"grad_norm": NaN,
|
| 428 |
-
"learning_rate":
|
| 429 |
"loss": 0.0,
|
| 430 |
"step": 3000
|
| 431 |
},
|
| 432 |
{
|
| 433 |
"epoch": 0.9037037037037037,
|
| 434 |
"grad_norm": NaN,
|
| 435 |
-
"learning_rate":
|
| 436 |
"loss": 0.0,
|
| 437 |
"step": 3050
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 0.9185185185185185,
|
| 441 |
"grad_norm": NaN,
|
| 442 |
-
"learning_rate":
|
| 443 |
"loss": 0.0,
|
| 444 |
"step": 3100
|
| 445 |
},
|
| 446 |
{
|
| 447 |
"epoch": 0.9333333333333333,
|
| 448 |
"grad_norm": NaN,
|
| 449 |
-
"learning_rate":
|
| 450 |
"loss": 0.0,
|
| 451 |
"step": 3150
|
| 452 |
},
|
| 453 |
{
|
| 454 |
"epoch": 0.9481481481481482,
|
| 455 |
"grad_norm": NaN,
|
| 456 |
-
"learning_rate":
|
| 457 |
"loss": 0.0,
|
| 458 |
"step": 3200
|
| 459 |
},
|
| 460 |
{
|
| 461 |
"epoch": 0.9629629629629629,
|
| 462 |
"grad_norm": NaN,
|
| 463 |
-
"learning_rate":
|
| 464 |
"loss": 0.0,
|
| 465 |
"step": 3250
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 0.9777777777777777,
|
| 469 |
"grad_norm": NaN,
|
| 470 |
-
"learning_rate":
|
| 471 |
"loss": 0.0,
|
| 472 |
"step": 3300
|
| 473 |
},
|
| 474 |
{
|
| 475 |
"epoch": 0.9925925925925926,
|
| 476 |
"grad_norm": NaN,
|
| 477 |
-
"learning_rate":
|
| 478 |
"loss": 0.0,
|
| 479 |
"step": 3350
|
| 480 |
},
|
|
@@ -482,9 +482,9 @@
|
|
| 482 |
"epoch": 1.0,
|
| 483 |
"eval_accuracy": 0.25075,
|
| 484 |
"eval_loss": NaN,
|
| 485 |
-
"eval_runtime": 17.
|
| 486 |
-
"eval_samples_per_second":
|
| 487 |
-
"eval_steps_per_second": 5.
|
| 488 |
"step": 3375
|
| 489 |
}
|
| 490 |
],
|
|
@@ -496,7 +496,7 @@
|
|
| 496 |
"stateful_callbacks": {
|
| 497 |
"EarlyStoppingCallback": {
|
| 498 |
"args": {
|
| 499 |
-
"early_stopping_patience":
|
| 500 |
"early_stopping_threshold": 0.0
|
| 501 |
},
|
| 502 |
"attributes": {
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.014814814814814815,
|
| 14 |
+
"grad_norm": 15.2734375,
|
| 15 |
+
"learning_rate": 2.4197530864197535e-07,
|
| 16 |
+
"loss": 2.7766494750976562,
|
| 17 |
"step": 50
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.02962962962962963,
|
| 21 |
+
"grad_norm": 14.0625,
|
| 22 |
+
"learning_rate": 4.88888888888889e-07,
|
| 23 |
+
"loss": 2.6392501831054687,
|
| 24 |
"step": 100
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.044444444444444446,
|
| 28 |
+
"grad_norm": 7.6171875,
|
| 29 |
+
"learning_rate": 7.358024691358024e-07,
|
| 30 |
+
"loss": 2.662928161621094,
|
| 31 |
"step": 150
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.05925925925925926,
|
| 35 |
+
"grad_norm": 13.4296875,
|
| 36 |
+
"learning_rate": 9.827160493827162e-07,
|
| 37 |
+
"loss": 3.1913311767578123,
|
| 38 |
"step": 200
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.07407407407407407,
|
| 42 |
+
"grad_norm": 1.96484375,
|
| 43 |
+
"learning_rate": 1.2296296296296297e-06,
|
| 44 |
+
"loss": 2.97621826171875,
|
| 45 |
"step": 250
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.08888888888888889,
|
| 49 |
+
"grad_norm": 2.583984375,
|
| 50 |
+
"learning_rate": 1.4765432098765435e-06,
|
| 51 |
+
"loss": 2.795979309082031,
|
| 52 |
"step": 300
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.1037037037037037,
|
| 56 |
+
"grad_norm": 3.095703125,
|
| 57 |
+
"learning_rate": 1.723456790123457e-06,
|
| 58 |
+
"loss": 2.7952374267578124,
|
| 59 |
"step": 350
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.11851851851851852,
|
| 63 |
+
"grad_norm": 2.453125,
|
| 64 |
+
"learning_rate": 1.9703703703703703e-06,
|
| 65 |
+
"loss": 2.7849395751953123,
|
| 66 |
"step": 400
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.13333333333333333,
|
| 70 |
+
"grad_norm": 6.875,
|
| 71 |
+
"learning_rate": 2.217283950617284e-06,
|
| 72 |
+
"loss": 2.7812811279296876,
|
| 73 |
"step": 450
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.14814814814814814,
|
| 77 |
+
"grad_norm": 2.33203125,
|
| 78 |
+
"learning_rate": 2.464197530864198e-06,
|
| 79 |
+
"loss": 2.7828436279296875,
|
| 80 |
"step": 500
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.16296296296296298,
|
| 84 |
+
"grad_norm": 2.53125,
|
| 85 |
+
"learning_rate": 2.7111111111111116e-06,
|
| 86 |
+
"loss": 2.7977508544921874,
|
| 87 |
"step": 550
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.17777777777777778,
|
| 91 |
+
"grad_norm": 3.587890625,
|
| 92 |
+
"learning_rate": 2.9580246913580254e-06,
|
| 93 |
+
"loss": 2.783578186035156,
|
| 94 |
"step": 600
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 0.1925925925925926,
|
| 98 |
+
"grad_norm": 1.9501953125,
|
| 99 |
+
"learning_rate": 3.2049382716049387e-06,
|
| 100 |
+
"loss": 2.7933184814453127,
|
| 101 |
"step": 650
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 0.2074074074074074,
|
| 105 |
+
"grad_norm": 3.435546875,
|
| 106 |
+
"learning_rate": 3.4518518518518524e-06,
|
| 107 |
+
"loss": 2.80977783203125,
|
| 108 |
"step": 700
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 0.2222222222222222,
|
| 112 |
+
"grad_norm": 1.8955078125,
|
| 113 |
+
"learning_rate": 3.6987654320987658e-06,
|
| 114 |
+
"loss": 2.7978311157226563,
|
| 115 |
"step": 750
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 0.23703703703703705,
|
| 119 |
+
"grad_norm": 2.455078125,
|
| 120 |
+
"learning_rate": 3.94567901234568e-06,
|
| 121 |
+
"loss": 2.7951055908203126,
|
| 122 |
"step": 800
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.2518518518518518,
|
| 126 |
+
"grad_norm": 4.1171875,
|
| 127 |
+
"learning_rate": 4.192592592592593e-06,
|
| 128 |
+
"loss": 2.785648193359375,
|
| 129 |
"step": 850
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.26666666666666666,
|
| 133 |
+
"grad_norm": 1.84375,
|
| 134 |
+
"learning_rate": 4.439506172839507e-06,
|
| 135 |
+
"loss": 2.7951544189453124,
|
| 136 |
"step": 900
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 0.2814814814814815,
|
| 140 |
+
"grad_norm": 2.359375,
|
| 141 |
+
"learning_rate": 4.68641975308642e-06,
|
| 142 |
+
"loss": 2.796750793457031,
|
| 143 |
"step": 950
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.2962962962962963,
|
| 147 |
+
"grad_norm": 2.39453125,
|
| 148 |
+
"learning_rate": 4.933333333333334e-06,
|
| 149 |
+
"loss": 2.7995571899414062,
|
| 150 |
"step": 1000
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 0.3111111111111111,
|
| 154 |
+
"grad_norm": 2.55078125,
|
| 155 |
+
"learning_rate": 5.180246913580247e-06,
|
| 156 |
+
"loss": 2.7804888916015624,
|
| 157 |
"step": 1050
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 0.32592592592592595,
|
| 161 |
+
"grad_norm": 3.67578125,
|
| 162 |
+
"learning_rate": 5.427160493827161e-06,
|
| 163 |
+
"loss": 2.792048034667969,
|
| 164 |
"step": 1100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 0.34074074074074073,
|
| 168 |
+
"grad_norm": 2.130859375,
|
| 169 |
+
"learning_rate": 5.674074074074074e-06,
|
| 170 |
+
"loss": 2.789242858886719,
|
| 171 |
"step": 1150
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 0.35555555555555557,
|
| 175 |
+
"grad_norm": 2.412109375,
|
| 176 |
+
"learning_rate": 5.920987654320987e-06,
|
| 177 |
+
"loss": 2.788144836425781,
|
| 178 |
"step": 1200
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 0.37037037037037035,
|
| 182 |
+
"grad_norm": 2.72265625,
|
| 183 |
+
"learning_rate": 6.167901234567902e-06,
|
| 184 |
+
"loss": 2.8027859497070313,
|
| 185 |
"step": 1250
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.3851851851851852,
|
| 189 |
+
"grad_norm": 2.6484375,
|
| 190 |
+
"learning_rate": 6.414814814814815e-06,
|
| 191 |
+
"loss": 2.7877798461914063,
|
| 192 |
"step": 1300
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 0.4,
|
| 196 |
+
"grad_norm": 2.3359375,
|
| 197 |
+
"learning_rate": 6.661728395061728e-06,
|
| 198 |
+
"loss": 2.7876470947265624,
|
| 199 |
"step": 1350
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 0.4148148148148148,
|
| 203 |
+
"grad_norm": 1.9970703125,
|
| 204 |
+
"learning_rate": 6.908641975308642e-06,
|
| 205 |
+
"loss": 2.77328857421875,
|
| 206 |
"step": 1400
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 0.42962962962962964,
|
| 210 |
+
"grad_norm": 1.875,
|
| 211 |
+
"learning_rate": 7.155555555555556e-06,
|
| 212 |
+
"loss": 2.79718017578125,
|
| 213 |
"step": 1450
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 0.4444444444444444,
|
| 217 |
+
"grad_norm": 2.248046875,
|
| 218 |
+
"learning_rate": 7.40246913580247e-06,
|
| 219 |
+
"loss": 2.7796466064453127,
|
| 220 |
"step": 1500
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 0.45925925925925926,
|
| 224 |
+
"grad_norm": 1.8623046875,
|
| 225 |
+
"learning_rate": 7.649382716049384e-06,
|
| 226 |
+
"loss": 2.7859207153320313,
|
| 227 |
"step": 1550
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 0.4740740740740741,
|
| 231 |
+
"grad_norm": 1.5302734375,
|
| 232 |
+
"learning_rate": 7.896296296296297e-06,
|
| 233 |
+
"loss": 2.781698913574219,
|
| 234 |
"step": 1600
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 0.4888888888888889,
|
| 238 |
+
"grad_norm": 1.3701171875,
|
| 239 |
+
"learning_rate": 8.14320987654321e-06,
|
| 240 |
+
"loss": 2.7813421630859376,
|
| 241 |
"step": 1650
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 0.5037037037037037,
|
| 245 |
+
"grad_norm": 1.40625,
|
| 246 |
+
"learning_rate": 8.390123456790124e-06,
|
| 247 |
+
"loss": 2.781036071777344,
|
| 248 |
"step": 1700
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 0.5185185185185185,
|
| 252 |
+
"grad_norm": 2.236328125,
|
| 253 |
+
"learning_rate": 8.637037037037037e-06,
|
| 254 |
+
"loss": 2.783014221191406,
|
| 255 |
"step": 1750
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 0.5333333333333333,
|
| 259 |
+
"grad_norm": 1.826171875,
|
| 260 |
+
"learning_rate": 8.883950617283952e-06,
|
| 261 |
+
"loss": 2.7817703247070313,
|
| 262 |
"step": 1800
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 0.5481481481481482,
|
| 266 |
+
"grad_norm": 1.82421875,
|
| 267 |
+
"learning_rate": 9.130864197530866e-06,
|
| 268 |
+
"loss": 2.7779510498046873,
|
| 269 |
"step": 1850
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 0.562962962962963,
|
| 273 |
+
"grad_norm": 2.1953125,
|
| 274 |
+
"learning_rate": 9.377777777777779e-06,
|
| 275 |
+
"loss": 2.7827484130859377,
|
| 276 |
"step": 1900
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 0.5777777777777777,
|
| 280 |
+
"grad_norm": 1.353515625,
|
| 281 |
+
"learning_rate": 9.624691358024692e-06,
|
| 282 |
+
"loss": 2.7817633056640627,
|
| 283 |
"step": 1950
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 0.5925925925925926,
|
| 287 |
+
"grad_norm": 2.69921875,
|
| 288 |
+
"learning_rate": 9.871604938271606e-06,
|
| 289 |
+
"loss": 2.7813009643554687,
|
| 290 |
"step": 2000
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 0.6074074074074074,
|
| 294 |
+
"grad_norm": 1.2294921875,
|
| 295 |
+
"learning_rate": 9.970370370370372e-06,
|
| 296 |
+
"loss": 2.77718994140625,
|
| 297 |
"step": 2050
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 0.6222222222222222,
|
| 301 |
+
"grad_norm": 1.6513671875,
|
| 302 |
+
"learning_rate": 9.908641975308642e-06,
|
| 303 |
+
"loss": 2.778031921386719,
|
| 304 |
"step": 2100
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 0.6370370370370371,
|
| 308 |
+
"grad_norm": 1.0498046875,
|
| 309 |
+
"learning_rate": 9.846913580246915e-06,
|
| 310 |
+
"loss": 2.778972473144531,
|
| 311 |
"step": 2150
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 0.6518518518518519,
|
| 315 |
+
"grad_norm": 1.7890625,
|
| 316 |
+
"learning_rate": 9.785185185185187e-06,
|
| 317 |
+
"loss": 2.7754339599609374,
|
| 318 |
"step": 2200
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 0.6666666666666666,
|
| 322 |
+
"grad_norm": 0.3349609375,
|
| 323 |
+
"learning_rate": 9.723456790123457e-06,
|
| 324 |
+
"loss": 2.7735556030273436,
|
| 325 |
"step": 2250
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 0.6814814814814815,
|
| 329 |
+
"grad_norm": 0.53271484375,
|
| 330 |
+
"learning_rate": 9.66172839506173e-06,
|
| 331 |
+
"loss": 2.7741458129882814,
|
| 332 |
"step": 2300
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 0.6962962962962963,
|
| 336 |
+
"grad_norm": 1.763671875,
|
| 337 |
+
"learning_rate": 9.600000000000001e-06,
|
| 338 |
+
"loss": 2.773301086425781,
|
| 339 |
"step": 2350
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 0.7111111111111111,
|
| 343 |
+
"grad_norm": 1.1982421875,
|
| 344 |
+
"learning_rate": 9.538271604938273e-06,
|
| 345 |
+
"loss": 2.7742254638671877,
|
| 346 |
"step": 2400
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 0.725925925925926,
|
| 350 |
+
"grad_norm": 2.26953125,
|
| 351 |
+
"learning_rate": 9.476543209876544e-06,
|
| 352 |
+
"loss": 2.7740939331054686,
|
| 353 |
"step": 2450
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 0.7407407407407407,
|
| 357 |
"grad_norm": NaN,
|
| 358 |
+
"learning_rate": 9.414814814814816e-06,
|
| 359 |
+
"loss": 3.09239013671875,
|
| 360 |
"step": 2500
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 0.7555555555555555,
|
| 364 |
"grad_norm": NaN,
|
| 365 |
+
"learning_rate": 9.353086419753086e-06,
|
| 366 |
"loss": 0.0,
|
| 367 |
"step": 2550
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 0.7703703703703704,
|
| 371 |
"grad_norm": NaN,
|
| 372 |
+
"learning_rate": 9.29135802469136e-06,
|
| 373 |
"loss": 0.0,
|
| 374 |
"step": 2600
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 0.7851851851851852,
|
| 378 |
"grad_norm": NaN,
|
| 379 |
+
"learning_rate": 9.22962962962963e-06,
|
| 380 |
"loss": 0.0,
|
| 381 |
"step": 2650
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 0.8,
|
| 385 |
"grad_norm": NaN,
|
| 386 |
+
"learning_rate": 9.167901234567903e-06,
|
| 387 |
"loss": 0.0,
|
| 388 |
"step": 2700
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 0.8148148148148148,
|
| 392 |
"grad_norm": NaN,
|
| 393 |
+
"learning_rate": 9.106172839506173e-06,
|
| 394 |
"loss": 0.0,
|
| 395 |
"step": 2750
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 0.8296296296296296,
|
| 399 |
"grad_norm": NaN,
|
| 400 |
+
"learning_rate": 9.044444444444445e-06,
|
| 401 |
"loss": 0.0,
|
| 402 |
"step": 2800
|
| 403 |
},
|
| 404 |
{
|
| 405 |
"epoch": 0.8444444444444444,
|
| 406 |
"grad_norm": NaN,
|
| 407 |
+
"learning_rate": 8.982716049382717e-06,
|
| 408 |
"loss": 0.0,
|
| 409 |
"step": 2850
|
| 410 |
},
|
| 411 |
{
|
| 412 |
"epoch": 0.8592592592592593,
|
| 413 |
"grad_norm": NaN,
|
| 414 |
+
"learning_rate": 8.92098765432099e-06,
|
| 415 |
"loss": 0.0,
|
| 416 |
"step": 2900
|
| 417 |
},
|
| 418 |
{
|
| 419 |
"epoch": 0.8740740740740741,
|
| 420 |
"grad_norm": NaN,
|
| 421 |
+
"learning_rate": 8.85925925925926e-06,
|
| 422 |
"loss": 0.0,
|
| 423 |
"step": 2950
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 0.8888888888888888,
|
| 427 |
"grad_norm": NaN,
|
| 428 |
+
"learning_rate": 8.797530864197532e-06,
|
| 429 |
"loss": 0.0,
|
| 430 |
"step": 3000
|
| 431 |
},
|
| 432 |
{
|
| 433 |
"epoch": 0.9037037037037037,
|
| 434 |
"grad_norm": NaN,
|
| 435 |
+
"learning_rate": 8.735802469135804e-06,
|
| 436 |
"loss": 0.0,
|
| 437 |
"step": 3050
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 0.9185185185185185,
|
| 441 |
"grad_norm": NaN,
|
| 442 |
+
"learning_rate": 8.674074074074074e-06,
|
| 443 |
"loss": 0.0,
|
| 444 |
"step": 3100
|
| 445 |
},
|
| 446 |
{
|
| 447 |
"epoch": 0.9333333333333333,
|
| 448 |
"grad_norm": NaN,
|
| 449 |
+
"learning_rate": 8.612345679012346e-06,
|
| 450 |
"loss": 0.0,
|
| 451 |
"step": 3150
|
| 452 |
},
|
| 453 |
{
|
| 454 |
"epoch": 0.9481481481481482,
|
| 455 |
"grad_norm": NaN,
|
| 456 |
+
"learning_rate": 8.550617283950618e-06,
|
| 457 |
"loss": 0.0,
|
| 458 |
"step": 3200
|
| 459 |
},
|
| 460 |
{
|
| 461 |
"epoch": 0.9629629629629629,
|
| 462 |
"grad_norm": NaN,
|
| 463 |
+
"learning_rate": 8.48888888888889e-06,
|
| 464 |
"loss": 0.0,
|
| 465 |
"step": 3250
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 0.9777777777777777,
|
| 469 |
"grad_norm": NaN,
|
| 470 |
+
"learning_rate": 8.42716049382716e-06,
|
| 471 |
"loss": 0.0,
|
| 472 |
"step": 3300
|
| 473 |
},
|
| 474 |
{
|
| 475 |
"epoch": 0.9925925925925926,
|
| 476 |
"grad_norm": NaN,
|
| 477 |
+
"learning_rate": 8.365432098765433e-06,
|
| 478 |
"loss": 0.0,
|
| 479 |
"step": 3350
|
| 480 |
},
|
|
|
|
| 482 |
"epoch": 1.0,
|
| 483 |
"eval_accuracy": 0.25075,
|
| 484 |
"eval_loss": NaN,
|
| 485 |
+
"eval_runtime": 17.5086,
|
| 486 |
+
"eval_samples_per_second": 685.376,
|
| 487 |
+
"eval_steps_per_second": 5.369,
|
| 488 |
"step": 3375
|
| 489 |
}
|
| 490 |
],
|
|
|
|
| 496 |
"stateful_callbacks": {
|
| 497 |
"EarlyStoppingCallback": {
|
| 498 |
"args": {
|
| 499 |
+
"early_stopping_patience": 2,
|
| 500 |
"early_stopping_threshold": 0.0
|
| 501 |
},
|
| 502 |
"attributes": {
|
checkpoint-3375/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5201
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f2eba2f8b924060a23a0dbbe58e92be52a7eaf1410ec02a5a8c210eceb8446e
|
| 3 |
size 5201
|
checkpoint-6750/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 737873099
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f325f5065c141d10e0de821baa10370b94c4fbc592dcf175956a060c6887827f
|
| 3 |
size 737873099
|
checkpoint-6750/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fee6859bd7aa45b0c1878c553ef1d6ea59de58d5784f3a3d959d0b3ee8630839
|
| 3 |
size 1465
|
checkpoint-6750/trainer_state.json
CHANGED
|
@@ -11,470 +11,470 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.014814814814814815,
|
| 14 |
-
"grad_norm":
|
| 15 |
-
"learning_rate":
|
| 16 |
-
"loss": 2.
|
| 17 |
"step": 50
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.02962962962962963,
|
| 21 |
-
"grad_norm":
|
| 22 |
-
"learning_rate":
|
| 23 |
-
"loss": 2.
|
| 24 |
"step": 100
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.044444444444444446,
|
| 28 |
-
"grad_norm":
|
| 29 |
-
"learning_rate":
|
| 30 |
-
"loss": 2.
|
| 31 |
"step": 150
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.05925925925925926,
|
| 35 |
-
"grad_norm":
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss": 3.
|
| 38 |
"step": 200
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.07407407407407407,
|
| 42 |
-
"grad_norm": 1.
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 2.
|
| 45 |
"step": 250
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.08888888888888889,
|
| 49 |
-
"grad_norm": 2.
|
| 50 |
-
"learning_rate":
|
| 51 |
-
"loss": 2.
|
| 52 |
"step": 300
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.1037037037037037,
|
| 56 |
-
"grad_norm":
|
| 57 |
-
"learning_rate":
|
| 58 |
-
"loss": 2.
|
| 59 |
"step": 350
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.11851851851851852,
|
| 63 |
-
"grad_norm": 2.
|
| 64 |
-
"learning_rate":
|
| 65 |
-
"loss": 2.
|
| 66 |
"step": 400
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.13333333333333333,
|
| 70 |
-
"grad_norm":
|
| 71 |
-
"learning_rate":
|
| 72 |
-
"loss": 2.
|
| 73 |
"step": 450
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.14814814814814814,
|
| 77 |
-
"grad_norm":
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 2.
|
| 80 |
"step": 500
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.16296296296296298,
|
| 84 |
-
"grad_norm": 2.
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss": 2.
|
| 87 |
"step": 550
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.17777777777777778,
|
| 91 |
-
"grad_norm":
|
| 92 |
-
"learning_rate":
|
| 93 |
-
"loss": 2.
|
| 94 |
"step": 600
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 0.1925925925925926,
|
| 98 |
-
"grad_norm": 1.
|
| 99 |
-
"learning_rate":
|
| 100 |
-
"loss": 2.
|
| 101 |
"step": 650
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 0.2074074074074074,
|
| 105 |
-
"grad_norm":
|
| 106 |
-
"learning_rate":
|
| 107 |
-
"loss": 2.
|
| 108 |
"step": 700
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 0.2222222222222222,
|
| 112 |
-
"grad_norm": 1.
|
| 113 |
-
"learning_rate":
|
| 114 |
-
"loss": 2.
|
| 115 |
"step": 750
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 0.23703703703703705,
|
| 119 |
-
"grad_norm":
|
| 120 |
-
"learning_rate":
|
| 121 |
-
"loss": 2.
|
| 122 |
"step": 800
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.2518518518518518,
|
| 126 |
-
"grad_norm":
|
| 127 |
-
"learning_rate":
|
| 128 |
-
"loss": 2.
|
| 129 |
"step": 850
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.26666666666666666,
|
| 133 |
-
"grad_norm": 1.
|
| 134 |
-
"learning_rate":
|
| 135 |
-
"loss": 2.
|
| 136 |
"step": 900
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 0.2814814814814815,
|
| 140 |
-
"grad_norm":
|
| 141 |
-
"learning_rate":
|
| 142 |
-
"loss": 2.
|
| 143 |
"step": 950
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.2962962962962963,
|
| 147 |
-
"grad_norm":
|
| 148 |
-
"learning_rate":
|
| 149 |
-
"loss": 2.
|
| 150 |
"step": 1000
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 0.3111111111111111,
|
| 154 |
-
"grad_norm": 2.
|
| 155 |
-
"learning_rate":
|
| 156 |
-
"loss": 2.
|
| 157 |
"step": 1050
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 0.32592592592592595,
|
| 161 |
-
"grad_norm":
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 2.
|
| 164 |
"step": 1100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 0.34074074074074073,
|
| 168 |
-
"grad_norm":
|
| 169 |
-
"learning_rate":
|
| 170 |
-
"loss": 2.
|
| 171 |
"step": 1150
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 0.35555555555555557,
|
| 175 |
-
"grad_norm": 2.
|
| 176 |
-
"learning_rate":
|
| 177 |
-
"loss": 2.
|
| 178 |
"step": 1200
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 0.37037037037037035,
|
| 182 |
-
"grad_norm": 2.
|
| 183 |
-
"learning_rate":
|
| 184 |
-
"loss": 2.
|
| 185 |
"step": 1250
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.3851851851851852,
|
| 189 |
-
"grad_norm": 2.
|
| 190 |
-
"learning_rate":
|
| 191 |
-
"loss": 2.
|
| 192 |
"step": 1300
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 0.4,
|
| 196 |
-
"grad_norm":
|
| 197 |
-
"learning_rate":
|
| 198 |
-
"loss": 2.
|
| 199 |
"step": 1350
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 0.4148148148148148,
|
| 203 |
-
"grad_norm": 1.
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss": 2.
|
| 206 |
"step": 1400
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 0.42962962962962964,
|
| 210 |
-
"grad_norm": 1.
|
| 211 |
-
"learning_rate":
|
| 212 |
-
"loss": 2.
|
| 213 |
"step": 1450
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 0.4444444444444444,
|
| 217 |
-
"grad_norm":
|
| 218 |
-
"learning_rate":
|
| 219 |
-
"loss":
|
| 220 |
"step": 1500
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 0.45925925925925926,
|
| 224 |
-
"grad_norm":
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss":
|
| 227 |
"step": 1550
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 0.4740740740740741,
|
| 231 |
-
"grad_norm":
|
| 232 |
-
"learning_rate":
|
| 233 |
-
"loss":
|
| 234 |
"step": 1600
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 0.4888888888888889,
|
| 238 |
-
"grad_norm":
|
| 239 |
-
"learning_rate":
|
| 240 |
-
"loss":
|
| 241 |
"step": 1650
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 0.5037037037037037,
|
| 245 |
-
"grad_norm":
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss":
|
| 248 |
"step": 1700
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 0.5185185185185185,
|
| 252 |
-
"grad_norm":
|
| 253 |
-
"learning_rate":
|
| 254 |
-
"loss":
|
| 255 |
"step": 1750
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 0.5333333333333333,
|
| 259 |
-
"grad_norm":
|
| 260 |
-
"learning_rate":
|
| 261 |
-
"loss":
|
| 262 |
"step": 1800
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 0.5481481481481482,
|
| 266 |
-
"grad_norm":
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss":
|
| 269 |
"step": 1850
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 0.562962962962963,
|
| 273 |
-
"grad_norm":
|
| 274 |
-
"learning_rate":
|
| 275 |
-
"loss":
|
| 276 |
"step": 1900
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 0.5777777777777777,
|
| 280 |
-
"grad_norm":
|
| 281 |
-
"learning_rate":
|
| 282 |
-
"loss":
|
| 283 |
"step": 1950
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 0.5925925925925926,
|
| 287 |
-
"grad_norm":
|
| 288 |
-
"learning_rate":
|
| 289 |
-
"loss":
|
| 290 |
"step": 2000
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 0.6074074074074074,
|
| 294 |
-
"grad_norm":
|
| 295 |
-
"learning_rate":
|
| 296 |
-
"loss":
|
| 297 |
"step": 2050
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 0.6222222222222222,
|
| 301 |
-
"grad_norm":
|
| 302 |
-
"learning_rate":
|
| 303 |
-
"loss":
|
| 304 |
"step": 2100
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 0.6370370370370371,
|
| 308 |
-
"grad_norm":
|
| 309 |
-
"learning_rate":
|
| 310 |
-
"loss":
|
| 311 |
"step": 2150
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 0.6518518518518519,
|
| 315 |
-
"grad_norm":
|
| 316 |
-
"learning_rate":
|
| 317 |
-
"loss":
|
| 318 |
"step": 2200
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 0.6666666666666666,
|
| 322 |
-
"grad_norm":
|
| 323 |
-
"learning_rate":
|
| 324 |
-
"loss":
|
| 325 |
"step": 2250
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 0.6814814814814815,
|
| 329 |
-
"grad_norm":
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss":
|
| 332 |
"step": 2300
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 0.6962962962962963,
|
| 336 |
-
"grad_norm":
|
| 337 |
-
"learning_rate":
|
| 338 |
-
"loss":
|
| 339 |
"step": 2350
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 0.7111111111111111,
|
| 343 |
-
"grad_norm":
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"loss":
|
| 346 |
"step": 2400
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 0.725925925925926,
|
| 350 |
-
"grad_norm":
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss":
|
| 353 |
"step": 2450
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 0.7407407407407407,
|
| 357 |
"grad_norm": NaN,
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss":
|
| 360 |
"step": 2500
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 0.7555555555555555,
|
| 364 |
"grad_norm": NaN,
|
| 365 |
-
"learning_rate":
|
| 366 |
"loss": 0.0,
|
| 367 |
"step": 2550
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 0.7703703703703704,
|
| 371 |
"grad_norm": NaN,
|
| 372 |
-
"learning_rate":
|
| 373 |
"loss": 0.0,
|
| 374 |
"step": 2600
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 0.7851851851851852,
|
| 378 |
"grad_norm": NaN,
|
| 379 |
-
"learning_rate":
|
| 380 |
"loss": 0.0,
|
| 381 |
"step": 2650
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 0.8,
|
| 385 |
"grad_norm": NaN,
|
| 386 |
-
"learning_rate":
|
| 387 |
"loss": 0.0,
|
| 388 |
"step": 2700
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 0.8148148148148148,
|
| 392 |
"grad_norm": NaN,
|
| 393 |
-
"learning_rate":
|
| 394 |
"loss": 0.0,
|
| 395 |
"step": 2750
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 0.8296296296296296,
|
| 399 |
"grad_norm": NaN,
|
| 400 |
-
"learning_rate":
|
| 401 |
"loss": 0.0,
|
| 402 |
"step": 2800
|
| 403 |
},
|
| 404 |
{
|
| 405 |
"epoch": 0.8444444444444444,
|
| 406 |
"grad_norm": NaN,
|
| 407 |
-
"learning_rate":
|
| 408 |
"loss": 0.0,
|
| 409 |
"step": 2850
|
| 410 |
},
|
| 411 |
{
|
| 412 |
"epoch": 0.8592592592592593,
|
| 413 |
"grad_norm": NaN,
|
| 414 |
-
"learning_rate":
|
| 415 |
"loss": 0.0,
|
| 416 |
"step": 2900
|
| 417 |
},
|
| 418 |
{
|
| 419 |
"epoch": 0.8740740740740741,
|
| 420 |
"grad_norm": NaN,
|
| 421 |
-
"learning_rate":
|
| 422 |
"loss": 0.0,
|
| 423 |
"step": 2950
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 0.8888888888888888,
|
| 427 |
"grad_norm": NaN,
|
| 428 |
-
"learning_rate":
|
| 429 |
"loss": 0.0,
|
| 430 |
"step": 3000
|
| 431 |
},
|
| 432 |
{
|
| 433 |
"epoch": 0.9037037037037037,
|
| 434 |
"grad_norm": NaN,
|
| 435 |
-
"learning_rate":
|
| 436 |
"loss": 0.0,
|
| 437 |
"step": 3050
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 0.9185185185185185,
|
| 441 |
"grad_norm": NaN,
|
| 442 |
-
"learning_rate":
|
| 443 |
"loss": 0.0,
|
| 444 |
"step": 3100
|
| 445 |
},
|
| 446 |
{
|
| 447 |
"epoch": 0.9333333333333333,
|
| 448 |
"grad_norm": NaN,
|
| 449 |
-
"learning_rate":
|
| 450 |
"loss": 0.0,
|
| 451 |
"step": 3150
|
| 452 |
},
|
| 453 |
{
|
| 454 |
"epoch": 0.9481481481481482,
|
| 455 |
"grad_norm": NaN,
|
| 456 |
-
"learning_rate":
|
| 457 |
"loss": 0.0,
|
| 458 |
"step": 3200
|
| 459 |
},
|
| 460 |
{
|
| 461 |
"epoch": 0.9629629629629629,
|
| 462 |
"grad_norm": NaN,
|
| 463 |
-
"learning_rate":
|
| 464 |
"loss": 0.0,
|
| 465 |
"step": 3250
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 0.9777777777777777,
|
| 469 |
"grad_norm": NaN,
|
| 470 |
-
"learning_rate":
|
| 471 |
"loss": 0.0,
|
| 472 |
"step": 3300
|
| 473 |
},
|
| 474 |
{
|
| 475 |
"epoch": 0.9925925925925926,
|
| 476 |
"grad_norm": NaN,
|
| 477 |
-
"learning_rate":
|
| 478 |
"loss": 0.0,
|
| 479 |
"step": 3350
|
| 480 |
},
|
|
@@ -482,484 +482,484 @@
|
|
| 482 |
"epoch": 1.0,
|
| 483 |
"eval_accuracy": 0.25075,
|
| 484 |
"eval_loss": NaN,
|
| 485 |
-
"eval_runtime": 17.
|
| 486 |
-
"eval_samples_per_second":
|
| 487 |
-
"eval_steps_per_second": 5.
|
| 488 |
"step": 3375
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 1.0074074074074073,
|
| 492 |
"grad_norm": NaN,
|
| 493 |
-
"learning_rate":
|
| 494 |
"loss": 0.0,
|
| 495 |
"step": 3400
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 1.0222222222222221,
|
| 499 |
"grad_norm": NaN,
|
| 500 |
-
"learning_rate":
|
| 501 |
"loss": 0.0,
|
| 502 |
"step": 3450
|
| 503 |
},
|
| 504 |
{
|
| 505 |
"epoch": 1.037037037037037,
|
| 506 |
"grad_norm": NaN,
|
| 507 |
-
"learning_rate":
|
| 508 |
"loss": 0.0,
|
| 509 |
"step": 3500
|
| 510 |
},
|
| 511 |
{
|
| 512 |
"epoch": 1.0518518518518518,
|
| 513 |
"grad_norm": NaN,
|
| 514 |
-
"learning_rate":
|
| 515 |
"loss": 0.0,
|
| 516 |
"step": 3550
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 1.0666666666666667,
|
| 520 |
"grad_norm": NaN,
|
| 521 |
-
"learning_rate":
|
| 522 |
"loss": 0.0,
|
| 523 |
"step": 3600
|
| 524 |
},
|
| 525 |
{
|
| 526 |
"epoch": 1.0814814814814815,
|
| 527 |
"grad_norm": NaN,
|
| 528 |
-
"learning_rate":
|
| 529 |
"loss": 0.0,
|
| 530 |
"step": 3650
|
| 531 |
},
|
| 532 |
{
|
| 533 |
"epoch": 1.0962962962962963,
|
| 534 |
"grad_norm": NaN,
|
| 535 |
-
"learning_rate":
|
| 536 |
"loss": 0.0,
|
| 537 |
"step": 3700
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 1.1111111111111112,
|
| 541 |
"grad_norm": NaN,
|
| 542 |
-
"learning_rate":
|
| 543 |
"loss": 0.0,
|
| 544 |
"step": 3750
|
| 545 |
},
|
| 546 |
{
|
| 547 |
"epoch": 1.125925925925926,
|
| 548 |
"grad_norm": NaN,
|
| 549 |
-
"learning_rate":
|
| 550 |
"loss": 0.0,
|
| 551 |
"step": 3800
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 1.1407407407407408,
|
| 555 |
"grad_norm": NaN,
|
| 556 |
-
"learning_rate":
|
| 557 |
"loss": 0.0,
|
| 558 |
"step": 3850
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 1.1555555555555554,
|
| 562 |
"grad_norm": NaN,
|
| 563 |
-
"learning_rate":
|
| 564 |
"loss": 0.0,
|
| 565 |
"step": 3900
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 1.1703703703703703,
|
| 569 |
"grad_norm": NaN,
|
| 570 |
-
"learning_rate":
|
| 571 |
"loss": 0.0,
|
| 572 |
"step": 3950
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 1.1851851851851851,
|
| 576 |
"grad_norm": NaN,
|
| 577 |
-
"learning_rate":
|
| 578 |
"loss": 0.0,
|
| 579 |
"step": 4000
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 1.2,
|
| 583 |
"grad_norm": NaN,
|
| 584 |
-
"learning_rate":
|
| 585 |
"loss": 0.0,
|
| 586 |
"step": 4050
|
| 587 |
},
|
| 588 |
{
|
| 589 |
"epoch": 1.2148148148148148,
|
| 590 |
"grad_norm": NaN,
|
| 591 |
-
"learning_rate":
|
| 592 |
"loss": 0.0,
|
| 593 |
"step": 4100
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 1.2296296296296296,
|
| 597 |
"grad_norm": NaN,
|
| 598 |
-
"learning_rate":
|
| 599 |
"loss": 0.0,
|
| 600 |
"step": 4150
|
| 601 |
},
|
| 602 |
{
|
| 603 |
"epoch": 1.2444444444444445,
|
| 604 |
"grad_norm": NaN,
|
| 605 |
-
"learning_rate":
|
| 606 |
"loss": 0.0,
|
| 607 |
"step": 4200
|
| 608 |
},
|
| 609 |
{
|
| 610 |
"epoch": 1.2592592592592593,
|
| 611 |
"grad_norm": NaN,
|
| 612 |
-
"learning_rate":
|
| 613 |
"loss": 0.0,
|
| 614 |
"step": 4250
|
| 615 |
},
|
| 616 |
{
|
| 617 |
"epoch": 1.2740740740740741,
|
| 618 |
"grad_norm": NaN,
|
| 619 |
-
"learning_rate":
|
| 620 |
"loss": 0.0,
|
| 621 |
"step": 4300
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 1.2888888888888888,
|
| 625 |
"grad_norm": NaN,
|
| 626 |
-
"learning_rate":
|
| 627 |
"loss": 0.0,
|
| 628 |
"step": 4350
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 1.3037037037037038,
|
| 632 |
"grad_norm": NaN,
|
| 633 |
-
"learning_rate":
|
| 634 |
"loss": 0.0,
|
| 635 |
"step": 4400
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 1.3185185185185184,
|
| 639 |
"grad_norm": NaN,
|
| 640 |
-
"learning_rate":
|
| 641 |
"loss": 0.0,
|
| 642 |
"step": 4450
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 1.3333333333333333,
|
| 646 |
"grad_norm": NaN,
|
| 647 |
-
"learning_rate":
|
| 648 |
"loss": 0.0,
|
| 649 |
"step": 4500
|
| 650 |
},
|
| 651 |
{
|
| 652 |
"epoch": 1.348148148148148,
|
| 653 |
"grad_norm": NaN,
|
| 654 |
-
"learning_rate":
|
| 655 |
"loss": 0.0,
|
| 656 |
"step": 4550
|
| 657 |
},
|
| 658 |
{
|
| 659 |
"epoch": 1.362962962962963,
|
| 660 |
"grad_norm": NaN,
|
| 661 |
-
"learning_rate":
|
| 662 |
"loss": 0.0,
|
| 663 |
"step": 4600
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 1.3777777777777778,
|
| 667 |
"grad_norm": NaN,
|
| 668 |
-
"learning_rate":
|
| 669 |
"loss": 0.0,
|
| 670 |
"step": 4650
|
| 671 |
},
|
| 672 |
{
|
| 673 |
"epoch": 1.3925925925925926,
|
| 674 |
"grad_norm": NaN,
|
| 675 |
-
"learning_rate":
|
| 676 |
"loss": 0.0,
|
| 677 |
"step": 4700
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 1.4074074074074074,
|
| 681 |
"grad_norm": NaN,
|
| 682 |
-
"learning_rate":
|
| 683 |
"loss": 0.0,
|
| 684 |
"step": 4750
|
| 685 |
},
|
| 686 |
{
|
| 687 |
"epoch": 1.4222222222222223,
|
| 688 |
"grad_norm": NaN,
|
| 689 |
-
"learning_rate":
|
| 690 |
"loss": 0.0,
|
| 691 |
"step": 4800
|
| 692 |
},
|
| 693 |
{
|
| 694 |
"epoch": 1.4370370370370371,
|
| 695 |
"grad_norm": NaN,
|
| 696 |
-
"learning_rate":
|
| 697 |
"loss": 0.0,
|
| 698 |
"step": 4850
|
| 699 |
},
|
| 700 |
{
|
| 701 |
"epoch": 1.4518518518518517,
|
| 702 |
"grad_norm": NaN,
|
| 703 |
-
"learning_rate":
|
| 704 |
"loss": 0.0,
|
| 705 |
"step": 4900
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 1.4666666666666668,
|
| 709 |
"grad_norm": NaN,
|
| 710 |
-
"learning_rate":
|
| 711 |
"loss": 0.0,
|
| 712 |
"step": 4950
|
| 713 |
},
|
| 714 |
{
|
| 715 |
"epoch": 1.4814814814814814,
|
| 716 |
"grad_norm": NaN,
|
| 717 |
-
"learning_rate":
|
| 718 |
"loss": 0.0,
|
| 719 |
"step": 5000
|
| 720 |
},
|
| 721 |
{
|
| 722 |
"epoch": 1.4962962962962962,
|
| 723 |
"grad_norm": NaN,
|
| 724 |
-
"learning_rate":
|
| 725 |
"loss": 0.0,
|
| 726 |
"step": 5050
|
| 727 |
},
|
| 728 |
{
|
| 729 |
"epoch": 1.511111111111111,
|
| 730 |
"grad_norm": NaN,
|
| 731 |
-
"learning_rate":
|
| 732 |
"loss": 0.0,
|
| 733 |
"step": 5100
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 1.525925925925926,
|
| 737 |
"grad_norm": NaN,
|
| 738 |
-
"learning_rate":
|
| 739 |
"loss": 0.0,
|
| 740 |
"step": 5150
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 1.5407407407407407,
|
| 744 |
"grad_norm": NaN,
|
| 745 |
-
"learning_rate":
|
| 746 |
"loss": 0.0,
|
| 747 |
"step": 5200
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 1.5555555555555556,
|
| 751 |
"grad_norm": NaN,
|
| 752 |
-
"learning_rate":
|
| 753 |
"loss": 0.0,
|
| 754 |
"step": 5250
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 1.5703703703703704,
|
| 758 |
"grad_norm": NaN,
|
| 759 |
-
"learning_rate":
|
| 760 |
"loss": 0.0,
|
| 761 |
"step": 5300
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 1.585185185185185,
|
| 765 |
"grad_norm": NaN,
|
| 766 |
-
"learning_rate":
|
| 767 |
"loss": 0.0,
|
| 768 |
"step": 5350
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 1.6,
|
| 772 |
"grad_norm": NaN,
|
| 773 |
-
"learning_rate":
|
| 774 |
"loss": 0.0,
|
| 775 |
"step": 5400
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 1.6148148148148147,
|
| 779 |
"grad_norm": NaN,
|
| 780 |
-
"learning_rate":
|
| 781 |
"loss": 0.0,
|
| 782 |
"step": 5450
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 1.6296296296296298,
|
| 786 |
"grad_norm": NaN,
|
| 787 |
-
"learning_rate":
|
| 788 |
"loss": 0.0,
|
| 789 |
"step": 5500
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 1.6444444444444444,
|
| 793 |
"grad_norm": NaN,
|
| 794 |
-
"learning_rate":
|
| 795 |
"loss": 0.0,
|
| 796 |
"step": 5550
|
| 797 |
},
|
| 798 |
{
|
| 799 |
"epoch": 1.6592592592592592,
|
| 800 |
"grad_norm": NaN,
|
| 801 |
-
"learning_rate":
|
| 802 |
"loss": 0.0,
|
| 803 |
"step": 5600
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 1.674074074074074,
|
| 807 |
"grad_norm": NaN,
|
| 808 |
-
"learning_rate":
|
| 809 |
"loss": 0.0,
|
| 810 |
"step": 5650
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 1.6888888888888889,
|
| 814 |
"grad_norm": NaN,
|
| 815 |
-
"learning_rate":
|
| 816 |
"loss": 0.0,
|
| 817 |
"step": 5700
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 1.7037037037037037,
|
| 821 |
"grad_norm": NaN,
|
| 822 |
-
"learning_rate":
|
| 823 |
"loss": 0.0,
|
| 824 |
"step": 5750
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 1.7185185185185186,
|
| 828 |
"grad_norm": NaN,
|
| 829 |
-
"learning_rate":
|
| 830 |
"loss": 0.0,
|
| 831 |
"step": 5800
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 1.7333333333333334,
|
| 835 |
"grad_norm": NaN,
|
| 836 |
-
"learning_rate":
|
| 837 |
"loss": 0.0,
|
| 838 |
"step": 5850
|
| 839 |
},
|
| 840 |
{
|
| 841 |
"epoch": 1.748148148148148,
|
| 842 |
"grad_norm": NaN,
|
| 843 |
-
"learning_rate":
|
| 844 |
"loss": 0.0,
|
| 845 |
"step": 5900
|
| 846 |
},
|
| 847 |
{
|
| 848 |
"epoch": 1.762962962962963,
|
| 849 |
"grad_norm": NaN,
|
| 850 |
-
"learning_rate":
|
| 851 |
"loss": 0.0,
|
| 852 |
"step": 5950
|
| 853 |
},
|
| 854 |
{
|
| 855 |
"epoch": 1.7777777777777777,
|
| 856 |
"grad_norm": NaN,
|
| 857 |
-
"learning_rate":
|
| 858 |
"loss": 0.0,
|
| 859 |
"step": 6000
|
| 860 |
},
|
| 861 |
{
|
| 862 |
"epoch": 1.7925925925925927,
|
| 863 |
"grad_norm": NaN,
|
| 864 |
-
"learning_rate":
|
| 865 |
"loss": 0.0,
|
| 866 |
"step": 6050
|
| 867 |
},
|
| 868 |
{
|
| 869 |
"epoch": 1.8074074074074074,
|
| 870 |
"grad_norm": NaN,
|
| 871 |
-
"learning_rate":
|
| 872 |
"loss": 0.0,
|
| 873 |
"step": 6100
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 1.8222222222222222,
|
| 877 |
"grad_norm": NaN,
|
| 878 |
-
"learning_rate":
|
| 879 |
"loss": 0.0,
|
| 880 |
"step": 6150
|
| 881 |
},
|
| 882 |
{
|
| 883 |
"epoch": 1.837037037037037,
|
| 884 |
"grad_norm": NaN,
|
| 885 |
-
"learning_rate":
|
| 886 |
"loss": 0.0,
|
| 887 |
"step": 6200
|
| 888 |
},
|
| 889 |
{
|
| 890 |
"epoch": 1.8518518518518519,
|
| 891 |
"grad_norm": NaN,
|
| 892 |
-
"learning_rate":
|
| 893 |
"loss": 0.0,
|
| 894 |
"step": 6250
|
| 895 |
},
|
| 896 |
{
|
| 897 |
"epoch": 1.8666666666666667,
|
| 898 |
"grad_norm": NaN,
|
| 899 |
-
"learning_rate":
|
| 900 |
"loss": 0.0,
|
| 901 |
"step": 6300
|
| 902 |
},
|
| 903 |
{
|
| 904 |
"epoch": 1.8814814814814815,
|
| 905 |
"grad_norm": NaN,
|
| 906 |
-
"learning_rate":
|
| 907 |
"loss": 0.0,
|
| 908 |
"step": 6350
|
| 909 |
},
|
| 910 |
{
|
| 911 |
"epoch": 1.8962962962962964,
|
| 912 |
"grad_norm": NaN,
|
| 913 |
-
"learning_rate":
|
| 914 |
"loss": 0.0,
|
| 915 |
"step": 6400
|
| 916 |
},
|
| 917 |
{
|
| 918 |
"epoch": 1.911111111111111,
|
| 919 |
"grad_norm": NaN,
|
| 920 |
-
"learning_rate":
|
| 921 |
"loss": 0.0,
|
| 922 |
"step": 6450
|
| 923 |
},
|
| 924 |
{
|
| 925 |
"epoch": 1.925925925925926,
|
| 926 |
"grad_norm": NaN,
|
| 927 |
-
"learning_rate":
|
| 928 |
"loss": 0.0,
|
| 929 |
"step": 6500
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 1.9407407407407407,
|
| 933 |
"grad_norm": NaN,
|
| 934 |
-
"learning_rate":
|
| 935 |
"loss": 0.0,
|
| 936 |
"step": 6550
|
| 937 |
},
|
| 938 |
{
|
| 939 |
"epoch": 1.9555555555555557,
|
| 940 |
"grad_norm": NaN,
|
| 941 |
-
"learning_rate":
|
| 942 |
"loss": 0.0,
|
| 943 |
"step": 6600
|
| 944 |
},
|
| 945 |
{
|
| 946 |
"epoch": 1.9703703703703703,
|
| 947 |
"grad_norm": NaN,
|
| 948 |
-
"learning_rate":
|
| 949 |
"loss": 0.0,
|
| 950 |
"step": 6650
|
| 951 |
},
|
| 952 |
{
|
| 953 |
"epoch": 1.9851851851851852,
|
| 954 |
"grad_norm": NaN,
|
| 955 |
-
"learning_rate":
|
| 956 |
"loss": 0.0,
|
| 957 |
"step": 6700
|
| 958 |
},
|
| 959 |
{
|
| 960 |
"epoch": 2.0,
|
| 961 |
"grad_norm": NaN,
|
| 962 |
-
"learning_rate":
|
| 963 |
"loss": 0.0,
|
| 964 |
"step": 6750
|
| 965 |
},
|
|
@@ -967,9 +967,9 @@
|
|
| 967 |
"epoch": 2.0,
|
| 968 |
"eval_accuracy": 0.25075,
|
| 969 |
"eval_loss": NaN,
|
| 970 |
-
"eval_runtime": 17.
|
| 971 |
-
"eval_samples_per_second":
|
| 972 |
-
"eval_steps_per_second": 5.
|
| 973 |
"step": 6750
|
| 974 |
}
|
| 975 |
],
|
|
@@ -981,7 +981,7 @@
|
|
| 981 |
"stateful_callbacks": {
|
| 982 |
"EarlyStoppingCallback": {
|
| 983 |
"args": {
|
| 984 |
-
"early_stopping_patience":
|
| 985 |
"early_stopping_threshold": 0.0
|
| 986 |
},
|
| 987 |
"attributes": {
|
|
@@ -994,7 +994,7 @@
|
|
| 994 |
"should_evaluate": false,
|
| 995 |
"should_log": false,
|
| 996 |
"should_save": true,
|
| 997 |
-
"should_training_stop":
|
| 998 |
},
|
| 999 |
"attributes": {}
|
| 1000 |
}
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.014814814814814815,
|
| 14 |
+
"grad_norm": 15.2734375,
|
| 15 |
+
"learning_rate": 2.4197530864197535e-07,
|
| 16 |
+
"loss": 2.7766494750976562,
|
| 17 |
"step": 50
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.02962962962962963,
|
| 21 |
+
"grad_norm": 14.0625,
|
| 22 |
+
"learning_rate": 4.88888888888889e-07,
|
| 23 |
+
"loss": 2.6392501831054687,
|
| 24 |
"step": 100
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.044444444444444446,
|
| 28 |
+
"grad_norm": 7.6171875,
|
| 29 |
+
"learning_rate": 7.358024691358024e-07,
|
| 30 |
+
"loss": 2.662928161621094,
|
| 31 |
"step": 150
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.05925925925925926,
|
| 35 |
+
"grad_norm": 13.4296875,
|
| 36 |
+
"learning_rate": 9.827160493827162e-07,
|
| 37 |
+
"loss": 3.1913311767578123,
|
| 38 |
"step": 200
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.07407407407407407,
|
| 42 |
+
"grad_norm": 1.96484375,
|
| 43 |
+
"learning_rate": 1.2296296296296297e-06,
|
| 44 |
+
"loss": 2.97621826171875,
|
| 45 |
"step": 250
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.08888888888888889,
|
| 49 |
+
"grad_norm": 2.583984375,
|
| 50 |
+
"learning_rate": 1.4765432098765435e-06,
|
| 51 |
+
"loss": 2.795979309082031,
|
| 52 |
"step": 300
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.1037037037037037,
|
| 56 |
+
"grad_norm": 3.095703125,
|
| 57 |
+
"learning_rate": 1.723456790123457e-06,
|
| 58 |
+
"loss": 2.7952374267578124,
|
| 59 |
"step": 350
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.11851851851851852,
|
| 63 |
+
"grad_norm": 2.453125,
|
| 64 |
+
"learning_rate": 1.9703703703703703e-06,
|
| 65 |
+
"loss": 2.7849395751953123,
|
| 66 |
"step": 400
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.13333333333333333,
|
| 70 |
+
"grad_norm": 6.875,
|
| 71 |
+
"learning_rate": 2.217283950617284e-06,
|
| 72 |
+
"loss": 2.7812811279296876,
|
| 73 |
"step": 450
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.14814814814814814,
|
| 77 |
+
"grad_norm": 2.33203125,
|
| 78 |
+
"learning_rate": 2.464197530864198e-06,
|
| 79 |
+
"loss": 2.7828436279296875,
|
| 80 |
"step": 500
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.16296296296296298,
|
| 84 |
+
"grad_norm": 2.53125,
|
| 85 |
+
"learning_rate": 2.7111111111111116e-06,
|
| 86 |
+
"loss": 2.7977508544921874,
|
| 87 |
"step": 550
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.17777777777777778,
|
| 91 |
+
"grad_norm": 3.587890625,
|
| 92 |
+
"learning_rate": 2.9580246913580254e-06,
|
| 93 |
+
"loss": 2.783578186035156,
|
| 94 |
"step": 600
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 0.1925925925925926,
|
| 98 |
+
"grad_norm": 1.9501953125,
|
| 99 |
+
"learning_rate": 3.2049382716049387e-06,
|
| 100 |
+
"loss": 2.7933184814453127,
|
| 101 |
"step": 650
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 0.2074074074074074,
|
| 105 |
+
"grad_norm": 3.435546875,
|
| 106 |
+
"learning_rate": 3.4518518518518524e-06,
|
| 107 |
+
"loss": 2.80977783203125,
|
| 108 |
"step": 700
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 0.2222222222222222,
|
| 112 |
+
"grad_norm": 1.8955078125,
|
| 113 |
+
"learning_rate": 3.6987654320987658e-06,
|
| 114 |
+
"loss": 2.7978311157226563,
|
| 115 |
"step": 750
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 0.23703703703703705,
|
| 119 |
+
"grad_norm": 2.455078125,
|
| 120 |
+
"learning_rate": 3.94567901234568e-06,
|
| 121 |
+
"loss": 2.7951055908203126,
|
| 122 |
"step": 800
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.2518518518518518,
|
| 126 |
+
"grad_norm": 4.1171875,
|
| 127 |
+
"learning_rate": 4.192592592592593e-06,
|
| 128 |
+
"loss": 2.785648193359375,
|
| 129 |
"step": 850
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.26666666666666666,
|
| 133 |
+
"grad_norm": 1.84375,
|
| 134 |
+
"learning_rate": 4.439506172839507e-06,
|
| 135 |
+
"loss": 2.7951544189453124,
|
| 136 |
"step": 900
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 0.2814814814814815,
|
| 140 |
+
"grad_norm": 2.359375,
|
| 141 |
+
"learning_rate": 4.68641975308642e-06,
|
| 142 |
+
"loss": 2.796750793457031,
|
| 143 |
"step": 950
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.2962962962962963,
|
| 147 |
+
"grad_norm": 2.39453125,
|
| 148 |
+
"learning_rate": 4.933333333333334e-06,
|
| 149 |
+
"loss": 2.7995571899414062,
|
| 150 |
"step": 1000
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 0.3111111111111111,
|
| 154 |
+
"grad_norm": 2.55078125,
|
| 155 |
+
"learning_rate": 5.180246913580247e-06,
|
| 156 |
+
"loss": 2.7804888916015624,
|
| 157 |
"step": 1050
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 0.32592592592592595,
|
| 161 |
+
"grad_norm": 3.67578125,
|
| 162 |
+
"learning_rate": 5.427160493827161e-06,
|
| 163 |
+
"loss": 2.792048034667969,
|
| 164 |
"step": 1100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 0.34074074074074073,
|
| 168 |
+
"grad_norm": 2.130859375,
|
| 169 |
+
"learning_rate": 5.674074074074074e-06,
|
| 170 |
+
"loss": 2.789242858886719,
|
| 171 |
"step": 1150
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 0.35555555555555557,
|
| 175 |
+
"grad_norm": 2.412109375,
|
| 176 |
+
"learning_rate": 5.920987654320987e-06,
|
| 177 |
+
"loss": 2.788144836425781,
|
| 178 |
"step": 1200
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 0.37037037037037035,
|
| 182 |
+
"grad_norm": 2.72265625,
|
| 183 |
+
"learning_rate": 6.167901234567902e-06,
|
| 184 |
+
"loss": 2.8027859497070313,
|
| 185 |
"step": 1250
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.3851851851851852,
|
| 189 |
+
"grad_norm": 2.6484375,
|
| 190 |
+
"learning_rate": 6.414814814814815e-06,
|
| 191 |
+
"loss": 2.7877798461914063,
|
| 192 |
"step": 1300
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 0.4,
|
| 196 |
+
"grad_norm": 2.3359375,
|
| 197 |
+
"learning_rate": 6.661728395061728e-06,
|
| 198 |
+
"loss": 2.7876470947265624,
|
| 199 |
"step": 1350
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 0.4148148148148148,
|
| 203 |
+
"grad_norm": 1.9970703125,
|
| 204 |
+
"learning_rate": 6.908641975308642e-06,
|
| 205 |
+
"loss": 2.77328857421875,
|
| 206 |
"step": 1400
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 0.42962962962962964,
|
| 210 |
+
"grad_norm": 1.875,
|
| 211 |
+
"learning_rate": 7.155555555555556e-06,
|
| 212 |
+
"loss": 2.79718017578125,
|
| 213 |
"step": 1450
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 0.4444444444444444,
|
| 217 |
+
"grad_norm": 2.248046875,
|
| 218 |
+
"learning_rate": 7.40246913580247e-06,
|
| 219 |
+
"loss": 2.7796466064453127,
|
| 220 |
"step": 1500
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 0.45925925925925926,
|
| 224 |
+
"grad_norm": 1.8623046875,
|
| 225 |
+
"learning_rate": 7.649382716049384e-06,
|
| 226 |
+
"loss": 2.7859207153320313,
|
| 227 |
"step": 1550
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 0.4740740740740741,
|
| 231 |
+
"grad_norm": 1.5302734375,
|
| 232 |
+
"learning_rate": 7.896296296296297e-06,
|
| 233 |
+
"loss": 2.781698913574219,
|
| 234 |
"step": 1600
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 0.4888888888888889,
|
| 238 |
+
"grad_norm": 1.3701171875,
|
| 239 |
+
"learning_rate": 8.14320987654321e-06,
|
| 240 |
+
"loss": 2.7813421630859376,
|
| 241 |
"step": 1650
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 0.5037037037037037,
|
| 245 |
+
"grad_norm": 1.40625,
|
| 246 |
+
"learning_rate": 8.390123456790124e-06,
|
| 247 |
+
"loss": 2.781036071777344,
|
| 248 |
"step": 1700
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 0.5185185185185185,
|
| 252 |
+
"grad_norm": 2.236328125,
|
| 253 |
+
"learning_rate": 8.637037037037037e-06,
|
| 254 |
+
"loss": 2.783014221191406,
|
| 255 |
"step": 1750
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 0.5333333333333333,
|
| 259 |
+
"grad_norm": 1.826171875,
|
| 260 |
+
"learning_rate": 8.883950617283952e-06,
|
| 261 |
+
"loss": 2.7817703247070313,
|
| 262 |
"step": 1800
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 0.5481481481481482,
|
| 266 |
+
"grad_norm": 1.82421875,
|
| 267 |
+
"learning_rate": 9.130864197530866e-06,
|
| 268 |
+
"loss": 2.7779510498046873,
|
| 269 |
"step": 1850
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 0.562962962962963,
|
| 273 |
+
"grad_norm": 2.1953125,
|
| 274 |
+
"learning_rate": 9.377777777777779e-06,
|
| 275 |
+
"loss": 2.7827484130859377,
|
| 276 |
"step": 1900
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 0.5777777777777777,
|
| 280 |
+
"grad_norm": 1.353515625,
|
| 281 |
+
"learning_rate": 9.624691358024692e-06,
|
| 282 |
+
"loss": 2.7817633056640627,
|
| 283 |
"step": 1950
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 0.5925925925925926,
|
| 287 |
+
"grad_norm": 2.69921875,
|
| 288 |
+
"learning_rate": 9.871604938271606e-06,
|
| 289 |
+
"loss": 2.7813009643554687,
|
| 290 |
"step": 2000
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 0.6074074074074074,
|
| 294 |
+
"grad_norm": 1.2294921875,
|
| 295 |
+
"learning_rate": 9.970370370370372e-06,
|
| 296 |
+
"loss": 2.77718994140625,
|
| 297 |
"step": 2050
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 0.6222222222222222,
|
| 301 |
+
"grad_norm": 1.6513671875,
|
| 302 |
+
"learning_rate": 9.908641975308642e-06,
|
| 303 |
+
"loss": 2.778031921386719,
|
| 304 |
"step": 2100
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 0.6370370370370371,
|
| 308 |
+
"grad_norm": 1.0498046875,
|
| 309 |
+
"learning_rate": 9.846913580246915e-06,
|
| 310 |
+
"loss": 2.778972473144531,
|
| 311 |
"step": 2150
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 0.6518518518518519,
|
| 315 |
+
"grad_norm": 1.7890625,
|
| 316 |
+
"learning_rate": 9.785185185185187e-06,
|
| 317 |
+
"loss": 2.7754339599609374,
|
| 318 |
"step": 2200
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 0.6666666666666666,
|
| 322 |
+
"grad_norm": 0.3349609375,
|
| 323 |
+
"learning_rate": 9.723456790123457e-06,
|
| 324 |
+
"loss": 2.7735556030273436,
|
| 325 |
"step": 2250
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 0.6814814814814815,
|
| 329 |
+
"grad_norm": 0.53271484375,
|
| 330 |
+
"learning_rate": 9.66172839506173e-06,
|
| 331 |
+
"loss": 2.7741458129882814,
|
| 332 |
"step": 2300
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 0.6962962962962963,
|
| 336 |
+
"grad_norm": 1.763671875,
|
| 337 |
+
"learning_rate": 9.600000000000001e-06,
|
| 338 |
+
"loss": 2.773301086425781,
|
| 339 |
"step": 2350
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 0.7111111111111111,
|
| 343 |
+
"grad_norm": 1.1982421875,
|
| 344 |
+
"learning_rate": 9.538271604938273e-06,
|
| 345 |
+
"loss": 2.7742254638671877,
|
| 346 |
"step": 2400
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 0.725925925925926,
|
| 350 |
+
"grad_norm": 2.26953125,
|
| 351 |
+
"learning_rate": 9.476543209876544e-06,
|
| 352 |
+
"loss": 2.7740939331054686,
|
| 353 |
"step": 2450
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 0.7407407407407407,
|
| 357 |
"grad_norm": NaN,
|
| 358 |
+
"learning_rate": 9.414814814814816e-06,
|
| 359 |
+
"loss": 3.09239013671875,
|
| 360 |
"step": 2500
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 0.7555555555555555,
|
| 364 |
"grad_norm": NaN,
|
| 365 |
+
"learning_rate": 9.353086419753086e-06,
|
| 366 |
"loss": 0.0,
|
| 367 |
"step": 2550
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 0.7703703703703704,
|
| 371 |
"grad_norm": NaN,
|
| 372 |
+
"learning_rate": 9.29135802469136e-06,
|
| 373 |
"loss": 0.0,
|
| 374 |
"step": 2600
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 0.7851851851851852,
|
| 378 |
"grad_norm": NaN,
|
| 379 |
+
"learning_rate": 9.22962962962963e-06,
|
| 380 |
"loss": 0.0,
|
| 381 |
"step": 2650
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 0.8,
|
| 385 |
"grad_norm": NaN,
|
| 386 |
+
"learning_rate": 9.167901234567903e-06,
|
| 387 |
"loss": 0.0,
|
| 388 |
"step": 2700
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 0.8148148148148148,
|
| 392 |
"grad_norm": NaN,
|
| 393 |
+
"learning_rate": 9.106172839506173e-06,
|
| 394 |
"loss": 0.0,
|
| 395 |
"step": 2750
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 0.8296296296296296,
|
| 399 |
"grad_norm": NaN,
|
| 400 |
+
"learning_rate": 9.044444444444445e-06,
|
| 401 |
"loss": 0.0,
|
| 402 |
"step": 2800
|
| 403 |
},
|
| 404 |
{
|
| 405 |
"epoch": 0.8444444444444444,
|
| 406 |
"grad_norm": NaN,
|
| 407 |
+
"learning_rate": 8.982716049382717e-06,
|
| 408 |
"loss": 0.0,
|
| 409 |
"step": 2850
|
| 410 |
},
|
| 411 |
{
|
| 412 |
"epoch": 0.8592592592592593,
|
| 413 |
"grad_norm": NaN,
|
| 414 |
+
"learning_rate": 8.92098765432099e-06,
|
| 415 |
"loss": 0.0,
|
| 416 |
"step": 2900
|
| 417 |
},
|
| 418 |
{
|
| 419 |
"epoch": 0.8740740740740741,
|
| 420 |
"grad_norm": NaN,
|
| 421 |
+
"learning_rate": 8.85925925925926e-06,
|
| 422 |
"loss": 0.0,
|
| 423 |
"step": 2950
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 0.8888888888888888,
|
| 427 |
"grad_norm": NaN,
|
| 428 |
+
"learning_rate": 8.797530864197532e-06,
|
| 429 |
"loss": 0.0,
|
| 430 |
"step": 3000
|
| 431 |
},
|
| 432 |
{
|
| 433 |
"epoch": 0.9037037037037037,
|
| 434 |
"grad_norm": NaN,
|
| 435 |
+
"learning_rate": 8.735802469135804e-06,
|
| 436 |
"loss": 0.0,
|
| 437 |
"step": 3050
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 0.9185185185185185,
|
| 441 |
"grad_norm": NaN,
|
| 442 |
+
"learning_rate": 8.674074074074074e-06,
|
| 443 |
"loss": 0.0,
|
| 444 |
"step": 3100
|
| 445 |
},
|
| 446 |
{
|
| 447 |
"epoch": 0.9333333333333333,
|
| 448 |
"grad_norm": NaN,
|
| 449 |
+
"learning_rate": 8.612345679012346e-06,
|
| 450 |
"loss": 0.0,
|
| 451 |
"step": 3150
|
| 452 |
},
|
| 453 |
{
|
| 454 |
"epoch": 0.9481481481481482,
|
| 455 |
"grad_norm": NaN,
|
| 456 |
+
"learning_rate": 8.550617283950618e-06,
|
| 457 |
"loss": 0.0,
|
| 458 |
"step": 3200
|
| 459 |
},
|
| 460 |
{
|
| 461 |
"epoch": 0.9629629629629629,
|
| 462 |
"grad_norm": NaN,
|
| 463 |
+
"learning_rate": 8.48888888888889e-06,
|
| 464 |
"loss": 0.0,
|
| 465 |
"step": 3250
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 0.9777777777777777,
|
| 469 |
"grad_norm": NaN,
|
| 470 |
+
"learning_rate": 8.42716049382716e-06,
|
| 471 |
"loss": 0.0,
|
| 472 |
"step": 3300
|
| 473 |
},
|
| 474 |
{
|
| 475 |
"epoch": 0.9925925925925926,
|
| 476 |
"grad_norm": NaN,
|
| 477 |
+
"learning_rate": 8.365432098765433e-06,
|
| 478 |
"loss": 0.0,
|
| 479 |
"step": 3350
|
| 480 |
},
|
|
|
|
| 482 |
"epoch": 1.0,
|
| 483 |
"eval_accuracy": 0.25075,
|
| 484 |
"eval_loss": NaN,
|
| 485 |
+
"eval_runtime": 17.5086,
|
| 486 |
+
"eval_samples_per_second": 685.376,
|
| 487 |
+
"eval_steps_per_second": 5.369,
|
| 488 |
"step": 3375
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 1.0074074074074073,
|
| 492 |
"grad_norm": NaN,
|
| 493 |
+
"learning_rate": 8.303703703703705e-06,
|
| 494 |
"loss": 0.0,
|
| 495 |
"step": 3400
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 1.0222222222222221,
|
| 499 |
"grad_norm": NaN,
|
| 500 |
+
"learning_rate": 8.241975308641975e-06,
|
| 501 |
"loss": 0.0,
|
| 502 |
"step": 3450
|
| 503 |
},
|
| 504 |
{
|
| 505 |
"epoch": 1.037037037037037,
|
| 506 |
"grad_norm": NaN,
|
| 507 |
+
"learning_rate": 8.180246913580248e-06,
|
| 508 |
"loss": 0.0,
|
| 509 |
"step": 3500
|
| 510 |
},
|
| 511 |
{
|
| 512 |
"epoch": 1.0518518518518518,
|
| 513 |
"grad_norm": NaN,
|
| 514 |
+
"learning_rate": 8.11851851851852e-06,
|
| 515 |
"loss": 0.0,
|
| 516 |
"step": 3550
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 1.0666666666666667,
|
| 520 |
"grad_norm": NaN,
|
| 521 |
+
"learning_rate": 8.05679012345679e-06,
|
| 522 |
"loss": 0.0,
|
| 523 |
"step": 3600
|
| 524 |
},
|
| 525 |
{
|
| 526 |
"epoch": 1.0814814814814815,
|
| 527 |
"grad_norm": NaN,
|
| 528 |
+
"learning_rate": 7.995061728395062e-06,
|
| 529 |
"loss": 0.0,
|
| 530 |
"step": 3650
|
| 531 |
},
|
| 532 |
{
|
| 533 |
"epoch": 1.0962962962962963,
|
| 534 |
"grad_norm": NaN,
|
| 535 |
+
"learning_rate": 7.933333333333334e-06,
|
| 536 |
"loss": 0.0,
|
| 537 |
"step": 3700
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 1.1111111111111112,
|
| 541 |
"grad_norm": NaN,
|
| 542 |
+
"learning_rate": 7.871604938271605e-06,
|
| 543 |
"loss": 0.0,
|
| 544 |
"step": 3750
|
| 545 |
},
|
| 546 |
{
|
| 547 |
"epoch": 1.125925925925926,
|
| 548 |
"grad_norm": NaN,
|
| 549 |
+
"learning_rate": 7.809876543209877e-06,
|
| 550 |
"loss": 0.0,
|
| 551 |
"step": 3800
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 1.1407407407407408,
|
| 555 |
"grad_norm": NaN,
|
| 556 |
+
"learning_rate": 7.748148148148149e-06,
|
| 557 |
"loss": 0.0,
|
| 558 |
"step": 3850
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 1.1555555555555554,
|
| 562 |
"grad_norm": NaN,
|
| 563 |
+
"learning_rate": 7.68641975308642e-06,
|
| 564 |
"loss": 0.0,
|
| 565 |
"step": 3900
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 1.1703703703703703,
|
| 569 |
"grad_norm": NaN,
|
| 570 |
+
"learning_rate": 7.624691358024692e-06,
|
| 571 |
"loss": 0.0,
|
| 572 |
"step": 3950
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 1.1851851851851851,
|
| 576 |
"grad_norm": NaN,
|
| 577 |
+
"learning_rate": 7.562962962962963e-06,
|
| 578 |
"loss": 0.0,
|
| 579 |
"step": 4000
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 1.2,
|
| 583 |
"grad_norm": NaN,
|
| 584 |
+
"learning_rate": 7.5012345679012354e-06,
|
| 585 |
"loss": 0.0,
|
| 586 |
"step": 4050
|
| 587 |
},
|
| 588 |
{
|
| 589 |
"epoch": 1.2148148148148148,
|
| 590 |
"grad_norm": NaN,
|
| 591 |
+
"learning_rate": 7.439506172839507e-06,
|
| 592 |
"loss": 0.0,
|
| 593 |
"step": 4100
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 1.2296296296296296,
|
| 597 |
"grad_norm": NaN,
|
| 598 |
+
"learning_rate": 7.377777777777778e-06,
|
| 599 |
"loss": 0.0,
|
| 600 |
"step": 4150
|
| 601 |
},
|
| 602 |
{
|
| 603 |
"epoch": 1.2444444444444445,
|
| 604 |
"grad_norm": NaN,
|
| 605 |
+
"learning_rate": 7.316049382716049e-06,
|
| 606 |
"loss": 0.0,
|
| 607 |
"step": 4200
|
| 608 |
},
|
| 609 |
{
|
| 610 |
"epoch": 1.2592592592592593,
|
| 611 |
"grad_norm": NaN,
|
| 612 |
+
"learning_rate": 7.254320987654322e-06,
|
| 613 |
"loss": 0.0,
|
| 614 |
"step": 4250
|
| 615 |
},
|
| 616 |
{
|
| 617 |
"epoch": 1.2740740740740741,
|
| 618 |
"grad_norm": NaN,
|
| 619 |
+
"learning_rate": 7.192592592592593e-06,
|
| 620 |
"loss": 0.0,
|
| 621 |
"step": 4300
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 1.2888888888888888,
|
| 625 |
"grad_norm": NaN,
|
| 626 |
+
"learning_rate": 7.130864197530865e-06,
|
| 627 |
"loss": 0.0,
|
| 628 |
"step": 4350
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 1.3037037037037038,
|
| 632 |
"grad_norm": NaN,
|
| 633 |
+
"learning_rate": 7.069135802469136e-06,
|
| 634 |
"loss": 0.0,
|
| 635 |
"step": 4400
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 1.3185185185185184,
|
| 639 |
"grad_norm": NaN,
|
| 640 |
+
"learning_rate": 7.007407407407408e-06,
|
| 641 |
"loss": 0.0,
|
| 642 |
"step": 4450
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 1.3333333333333333,
|
| 646 |
"grad_norm": NaN,
|
| 647 |
+
"learning_rate": 6.94567901234568e-06,
|
| 648 |
"loss": 0.0,
|
| 649 |
"step": 4500
|
| 650 |
},
|
| 651 |
{
|
| 652 |
"epoch": 1.348148148148148,
|
| 653 |
"grad_norm": NaN,
|
| 654 |
+
"learning_rate": 6.883950617283951e-06,
|
| 655 |
"loss": 0.0,
|
| 656 |
"step": 4550
|
| 657 |
},
|
| 658 |
{
|
| 659 |
"epoch": 1.362962962962963,
|
| 660 |
"grad_norm": NaN,
|
| 661 |
+
"learning_rate": 6.8222222222222225e-06,
|
| 662 |
"loss": 0.0,
|
| 663 |
"step": 4600
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 1.3777777777777778,
|
| 667 |
"grad_norm": NaN,
|
| 668 |
+
"learning_rate": 6.760493827160495e-06,
|
| 669 |
"loss": 0.0,
|
| 670 |
"step": 4650
|
| 671 |
},
|
| 672 |
{
|
| 673 |
"epoch": 1.3925925925925926,
|
| 674 |
"grad_norm": NaN,
|
| 675 |
+
"learning_rate": 6.698765432098766e-06,
|
| 676 |
"loss": 0.0,
|
| 677 |
"step": 4700
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 1.4074074074074074,
|
| 681 |
"grad_norm": NaN,
|
| 682 |
+
"learning_rate": 6.637037037037037e-06,
|
| 683 |
"loss": 0.0,
|
| 684 |
"step": 4750
|
| 685 |
},
|
| 686 |
{
|
| 687 |
"epoch": 1.4222222222222223,
|
| 688 |
"grad_norm": NaN,
|
| 689 |
+
"learning_rate": 6.575308641975308e-06,
|
| 690 |
"loss": 0.0,
|
| 691 |
"step": 4800
|
| 692 |
},
|
| 693 |
{
|
| 694 |
"epoch": 1.4370370370370371,
|
| 695 |
"grad_norm": NaN,
|
| 696 |
+
"learning_rate": 6.513580246913581e-06,
|
| 697 |
"loss": 0.0,
|
| 698 |
"step": 4850
|
| 699 |
},
|
| 700 |
{
|
| 701 |
"epoch": 1.4518518518518517,
|
| 702 |
"grad_norm": NaN,
|
| 703 |
+
"learning_rate": 6.4518518518518525e-06,
|
| 704 |
"loss": 0.0,
|
| 705 |
"step": 4900
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 1.4666666666666668,
|
| 709 |
"grad_norm": NaN,
|
| 710 |
+
"learning_rate": 6.390123456790124e-06,
|
| 711 |
"loss": 0.0,
|
| 712 |
"step": 4950
|
| 713 |
},
|
| 714 |
{
|
| 715 |
"epoch": 1.4814814814814814,
|
| 716 |
"grad_norm": NaN,
|
| 717 |
+
"learning_rate": 6.328395061728395e-06,
|
| 718 |
"loss": 0.0,
|
| 719 |
"step": 5000
|
| 720 |
},
|
| 721 |
{
|
| 722 |
"epoch": 1.4962962962962962,
|
| 723 |
"grad_norm": NaN,
|
| 724 |
+
"learning_rate": 6.266666666666668e-06,
|
| 725 |
"loss": 0.0,
|
| 726 |
"step": 5050
|
| 727 |
},
|
| 728 |
{
|
| 729 |
"epoch": 1.511111111111111,
|
| 730 |
"grad_norm": NaN,
|
| 731 |
+
"learning_rate": 6.204938271604939e-06,
|
| 732 |
"loss": 0.0,
|
| 733 |
"step": 5100
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 1.525925925925926,
|
| 737 |
"grad_norm": NaN,
|
| 738 |
+
"learning_rate": 6.1432098765432104e-06,
|
| 739 |
"loss": 0.0,
|
| 740 |
"step": 5150
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 1.5407407407407407,
|
| 744 |
"grad_norm": NaN,
|
| 745 |
+
"learning_rate": 6.081481481481482e-06,
|
| 746 |
"loss": 0.0,
|
| 747 |
"step": 5200
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 1.5555555555555556,
|
| 751 |
"grad_norm": NaN,
|
| 752 |
+
"learning_rate": 6.019753086419754e-06,
|
| 753 |
"loss": 0.0,
|
| 754 |
"step": 5250
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 1.5703703703703704,
|
| 758 |
"grad_norm": NaN,
|
| 759 |
+
"learning_rate": 5.958024691358025e-06,
|
| 760 |
"loss": 0.0,
|
| 761 |
"step": 5300
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 1.585185185185185,
|
| 765 |
"grad_norm": NaN,
|
| 766 |
+
"learning_rate": 5.896296296296296e-06,
|
| 767 |
"loss": 0.0,
|
| 768 |
"step": 5350
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 1.6,
|
| 772 |
"grad_norm": NaN,
|
| 773 |
+
"learning_rate": 5.834567901234568e-06,
|
| 774 |
"loss": 0.0,
|
| 775 |
"step": 5400
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 1.6148148148148147,
|
| 779 |
"grad_norm": NaN,
|
| 780 |
+
"learning_rate": 5.7728395061728404e-06,
|
| 781 |
"loss": 0.0,
|
| 782 |
"step": 5450
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 1.6296296296296298,
|
| 786 |
"grad_norm": NaN,
|
| 787 |
+
"learning_rate": 5.711111111111112e-06,
|
| 788 |
"loss": 0.0,
|
| 789 |
"step": 5500
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 1.6444444444444444,
|
| 793 |
"grad_norm": NaN,
|
| 794 |
+
"learning_rate": 5.649382716049383e-06,
|
| 795 |
"loss": 0.0,
|
| 796 |
"step": 5550
|
| 797 |
},
|
| 798 |
{
|
| 799 |
"epoch": 1.6592592592592592,
|
| 800 |
"grad_norm": NaN,
|
| 801 |
+
"learning_rate": 5.587654320987654e-06,
|
| 802 |
"loss": 0.0,
|
| 803 |
"step": 5600
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 1.674074074074074,
|
| 807 |
"grad_norm": NaN,
|
| 808 |
+
"learning_rate": 5.525925925925927e-06,
|
| 809 |
"loss": 0.0,
|
| 810 |
"step": 5650
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 1.6888888888888889,
|
| 814 |
"grad_norm": NaN,
|
| 815 |
+
"learning_rate": 5.464197530864198e-06,
|
| 816 |
"loss": 0.0,
|
| 817 |
"step": 5700
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 1.7037037037037037,
|
| 821 |
"grad_norm": NaN,
|
| 822 |
+
"learning_rate": 5.40246913580247e-06,
|
| 823 |
"loss": 0.0,
|
| 824 |
"step": 5750
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 1.7185185185185186,
|
| 828 |
"grad_norm": NaN,
|
| 829 |
+
"learning_rate": 5.340740740740741e-06,
|
| 830 |
"loss": 0.0,
|
| 831 |
"step": 5800
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 1.7333333333333334,
|
| 835 |
"grad_norm": NaN,
|
| 836 |
+
"learning_rate": 5.279012345679013e-06,
|
| 837 |
"loss": 0.0,
|
| 838 |
"step": 5850
|
| 839 |
},
|
| 840 |
{
|
| 841 |
"epoch": 1.748148148148148,
|
| 842 |
"grad_norm": NaN,
|
| 843 |
+
"learning_rate": 5.217283950617284e-06,
|
| 844 |
"loss": 0.0,
|
| 845 |
"step": 5900
|
| 846 |
},
|
| 847 |
{
|
| 848 |
"epoch": 1.762962962962963,
|
| 849 |
"grad_norm": NaN,
|
| 850 |
+
"learning_rate": 5.155555555555556e-06,
|
| 851 |
"loss": 0.0,
|
| 852 |
"step": 5950
|
| 853 |
},
|
| 854 |
{
|
| 855 |
"epoch": 1.7777777777777777,
|
| 856 |
"grad_norm": NaN,
|
| 857 |
+
"learning_rate": 5.0938271604938275e-06,
|
| 858 |
"loss": 0.0,
|
| 859 |
"step": 6000
|
| 860 |
},
|
| 861 |
{
|
| 862 |
"epoch": 1.7925925925925927,
|
| 863 |
"grad_norm": NaN,
|
| 864 |
+
"learning_rate": 5.0320987654321e-06,
|
| 865 |
"loss": 0.0,
|
| 866 |
"step": 6050
|
| 867 |
},
|
| 868 |
{
|
| 869 |
"epoch": 1.8074074074074074,
|
| 870 |
"grad_norm": NaN,
|
| 871 |
+
"learning_rate": 4.970370370370371e-06,
|
| 872 |
"loss": 0.0,
|
| 873 |
"step": 6100
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 1.8222222222222222,
|
| 877 |
"grad_norm": NaN,
|
| 878 |
+
"learning_rate": 4.908641975308642e-06,
|
| 879 |
"loss": 0.0,
|
| 880 |
"step": 6150
|
| 881 |
},
|
| 882 |
{
|
| 883 |
"epoch": 1.837037037037037,
|
| 884 |
"grad_norm": NaN,
|
| 885 |
+
"learning_rate": 4.846913580246914e-06,
|
| 886 |
"loss": 0.0,
|
| 887 |
"step": 6200
|
| 888 |
},
|
| 889 |
{
|
| 890 |
"epoch": 1.8518518518518519,
|
| 891 |
"grad_norm": NaN,
|
| 892 |
+
"learning_rate": 4.7851851851851854e-06,
|
| 893 |
"loss": 0.0,
|
| 894 |
"step": 6250
|
| 895 |
},
|
| 896 |
{
|
| 897 |
"epoch": 1.8666666666666667,
|
| 898 |
"grad_norm": NaN,
|
| 899 |
+
"learning_rate": 4.7234567901234575e-06,
|
| 900 |
"loss": 0.0,
|
| 901 |
"step": 6300
|
| 902 |
},
|
| 903 |
{
|
| 904 |
"epoch": 1.8814814814814815,
|
| 905 |
"grad_norm": NaN,
|
| 906 |
+
"learning_rate": 4.661728395061729e-06,
|
| 907 |
"loss": 0.0,
|
| 908 |
"step": 6350
|
| 909 |
},
|
| 910 |
{
|
| 911 |
"epoch": 1.8962962962962964,
|
| 912 |
"grad_norm": NaN,
|
| 913 |
+
"learning_rate": 4.600000000000001e-06,
|
| 914 |
"loss": 0.0,
|
| 915 |
"step": 6400
|
| 916 |
},
|
| 917 |
{
|
| 918 |
"epoch": 1.911111111111111,
|
| 919 |
"grad_norm": NaN,
|
| 920 |
+
"learning_rate": 4.538271604938272e-06,
|
| 921 |
"loss": 0.0,
|
| 922 |
"step": 6450
|
| 923 |
},
|
| 924 |
{
|
| 925 |
"epoch": 1.925925925925926,
|
| 926 |
"grad_norm": NaN,
|
| 927 |
+
"learning_rate": 4.476543209876544e-06,
|
| 928 |
"loss": 0.0,
|
| 929 |
"step": 6500
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 1.9407407407407407,
|
| 933 |
"grad_norm": NaN,
|
| 934 |
+
"learning_rate": 4.4148148148148154e-06,
|
| 935 |
"loss": 0.0,
|
| 936 |
"step": 6550
|
| 937 |
},
|
| 938 |
{
|
| 939 |
"epoch": 1.9555555555555557,
|
| 940 |
"grad_norm": NaN,
|
| 941 |
+
"learning_rate": 4.353086419753087e-06,
|
| 942 |
"loss": 0.0,
|
| 943 |
"step": 6600
|
| 944 |
},
|
| 945 |
{
|
| 946 |
"epoch": 1.9703703703703703,
|
| 947 |
"grad_norm": NaN,
|
| 948 |
+
"learning_rate": 4.291358024691358e-06,
|
| 949 |
"loss": 0.0,
|
| 950 |
"step": 6650
|
| 951 |
},
|
| 952 |
{
|
| 953 |
"epoch": 1.9851851851851852,
|
| 954 |
"grad_norm": NaN,
|
| 955 |
+
"learning_rate": 4.22962962962963e-06,
|
| 956 |
"loss": 0.0,
|
| 957 |
"step": 6700
|
| 958 |
},
|
| 959 |
{
|
| 960 |
"epoch": 2.0,
|
| 961 |
"grad_norm": NaN,
|
| 962 |
+
"learning_rate": 4.167901234567901e-06,
|
| 963 |
"loss": 0.0,
|
| 964 |
"step": 6750
|
| 965 |
},
|
|
|
|
| 967 |
"epoch": 2.0,
|
| 968 |
"eval_accuracy": 0.25075,
|
| 969 |
"eval_loss": NaN,
|
| 970 |
+
"eval_runtime": 17.7609,
|
| 971 |
+
"eval_samples_per_second": 675.641,
|
| 972 |
+
"eval_steps_per_second": 5.293,
|
| 973 |
"step": 6750
|
| 974 |
}
|
| 975 |
],
|
|
|
|
| 981 |
"stateful_callbacks": {
|
| 982 |
"EarlyStoppingCallback": {
|
| 983 |
"args": {
|
| 984 |
+
"early_stopping_patience": 2,
|
| 985 |
"early_stopping_threshold": 0.0
|
| 986 |
},
|
| 987 |
"attributes": {
|
|
|
|
| 994 |
"should_evaluate": false,
|
| 995 |
"should_log": false,
|
| 996 |
"should_save": true,
|
| 997 |
+
"should_training_stop": false
|
| 998 |
},
|
| 999 |
"attributes": {}
|
| 1000 |
}
|
checkpoint-6750/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5201
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f2eba2f8b924060a23a0dbbe58e92be52a7eaf1410ec02a5a8c210eceb8446e
|
| 3 |
size 5201
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5201
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f2eba2f8b924060a23a0dbbe58e92be52a7eaf1410ec02a5a8c210eceb8446e
|
| 3 |
size 5201
|