End of training
Browse files
README.md
CHANGED
|
@@ -7,16 +7,18 @@ tags:
|
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
-
- name:
|
| 11 |
results: []
|
| 12 |
---
|
| 13 |
|
| 14 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
|
| 17 |
-
#
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [ahxt/LiteLlama-460M-1T](https://huggingface.co/ahxt/LiteLlama-460M-1T) on the None dataset.
|
|
|
|
|
|
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
@@ -36,18 +38,26 @@ More information needed
|
|
| 36 |
|
| 37 |
The following hyperparameters were used during training:
|
| 38 |
- learning_rate: 0.0002
|
| 39 |
-
- train_batch_size:
|
| 40 |
-
- eval_batch_size:
|
| 41 |
- seed: 42
|
| 42 |
-
- gradient_accumulation_steps:
|
| 43 |
-
- total_train_batch_size:
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
-
- lr_scheduler_type:
|
| 46 |
- lr_scheduler_warmup_ratio: 0.03
|
| 47 |
- num_epochs: 5
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
|
| 52 |
|
| 53 |
### Framework versions
|
|
|
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
+
- name: ColdLLamaLite
|
| 11 |
results: []
|
| 12 |
---
|
| 13 |
|
| 14 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
|
| 17 |
+
# ColdLLamaLite
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [ahxt/LiteLlama-460M-1T](https://huggingface.co/ahxt/LiteLlama-460M-1T) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 2.3021
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
- learning_rate: 0.0002
|
| 41 |
+
- train_batch_size: 32
|
| 42 |
+
- eval_batch_size: 32
|
| 43 |
- seed: 42
|
| 44 |
+
- gradient_accumulation_steps: 8
|
| 45 |
+
- total_train_batch_size: 256
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
- lr_scheduler_warmup_ratio: 0.03
|
| 49 |
- num_epochs: 5
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 54 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 55 |
+
| 4.1436 | 0.8 | 25 | 3.8815 |
|
| 56 |
+
| 3.6028 | 1.6 | 50 | 3.2639 |
|
| 57 |
+
| 2.9395 | 2.4 | 75 | 2.5905 |
|
| 58 |
+
| 2.4548 | 3.2 | 100 | 2.3582 |
|
| 59 |
+
| 2.337 | 4.0 | 125 | 2.3102 |
|
| 60 |
+
| 2.3125 | 4.8 | 150 | 2.3024 |
|
| 61 |
|
| 62 |
|
| 63 |
### Framework versions
|
runs/Aug02_13-07-01_fastgpuserv/events.out.tfevents.1722602007.fastgpuserv.3714303.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6890949d3abe45db8d962b80fd220ed72f34e906ba435db853a4ad9a61faf49
|
| 3 |
+
size 359
|