Model save
Browse files- README.md +67 -0
- adapter_config.json +3 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +966 -0
- training_args.bin +1 -1
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: 01-ai/Yi-6B
|
| 9 |
+
model-index:
|
| 10 |
+
- name: Yi-6B-ruozhiba-5e-5-50
|
| 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 |
+
# Yi-6B-ruozhiba-5e-5-50
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [01-ai/Yi-6B](https://huggingface.co/01-ai/Yi-6B) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 3.1352
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 5e-05
|
| 41 |
+
- train_batch_size: 4
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: cosine
|
| 46 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 47 |
+
- num_epochs: 10
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 53 |
+
| 1.6668 | 2.0 | 110 | 1.8270 |
|
| 54 |
+
| 1.5865 | 3.0 | 165 | 1.8256 |
|
| 55 |
+
| 1.264 | 4.0 | 220 | 1.9826 |
|
| 56 |
+
| 0.99 | 5.0 | 275 | 2.2128 |
|
| 57 |
+
| 0.7218 | 6.0 | 330 | 2.5388 |
|
| 58 |
+
| 0.5291 | 7.0 | 385 | 2.7745 |
|
| 59 |
+
|
| 60 |
+
|
| 61 |
+
### Framework versions
|
| 62 |
+
|
| 63 |
+
- PEFT 0.7.1
|
| 64 |
+
- Transformers 4.36.2
|
| 65 |
+
- Pytorch 2.3.0+cu121
|
| 66 |
+
- Datasets 2.14.6
|
| 67 |
+
- Tokenizers 0.15.2
|
adapter_config.json
CHANGED
|
@@ -19,11 +19,11 @@
|
|
| 19 |
"rank_pattern": {},
|
| 20 |
"revision": null,
|
| 21 |
"target_modules": [
|
| 22 |
-
"down_proj",
|
| 23 |
-
"o_proj",
|
| 24 |
"k_proj",
|
| 25 |
-
"
|
| 26 |
"v_proj",
|
|
|
|
|
|
|
| 27 |
"q_proj",
|
| 28 |
"up_proj"
|
| 29 |
],
|
|
|
|
| 19 |
"rank_pattern": {},
|
| 20 |
"revision": null,
|
| 21 |
"target_modules": [
|
|
|
|
|
|
|
| 22 |
"k_proj",
|
| 23 |
+
"o_proj",
|
| 24 |
"v_proj",
|
| 25 |
+
"gate_proj",
|
| 26 |
+
"down_proj",
|
| 27 |
"q_proj",
|
| 28 |
"up_proj"
|
| 29 |
],
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 3.135209083557129,
|
| 4 |
+
"eval_runtime": 6.6706,
|
| 5 |
+
"eval_samples": 23,
|
| 6 |
+
"eval_samples_per_second": 3.448,
|
| 7 |
+
"eval_steps_per_second": 0.899,
|
| 8 |
+
"train_loss": 0.0,
|
| 9 |
+
"train_runtime": 12.8072,
|
| 10 |
+
"train_samples": 217,
|
| 11 |
+
"train_samples_per_second": 169.436,
|
| 12 |
+
"train_steps_per_second": 42.945
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 3.135209083557129,
|
| 4 |
+
"eval_runtime": 6.6706,
|
| 5 |
+
"eval_samples": 23,
|
| 6 |
+
"eval_samples_per_second": 3.448,
|
| 7 |
+
"eval_steps_per_second": 0.899
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 0.0,
|
| 4 |
+
"train_runtime": 12.8072,
|
| 5 |
+
"train_samples": 217,
|
| 6 |
+
"train_samples_per_second": 169.436,
|
| 7 |
+
"train_steps_per_second": 42.945
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,966 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 550,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 9.09090909090909e-07,
|
| 14 |
+
"loss": 2.3833,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.07,
|
| 19 |
+
"learning_rate": 3.636363636363636e-06,
|
| 20 |
+
"loss": 2.4779,
|
| 21 |
+
"step": 4
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.15,
|
| 25 |
+
"learning_rate": 7.272727272727272e-06,
|
| 26 |
+
"loss": 2.3099,
|
| 27 |
+
"step": 8
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.22,
|
| 31 |
+
"learning_rate": 1.0909090909090909e-05,
|
| 32 |
+
"loss": 2.3051,
|
| 33 |
+
"step": 12
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.29,
|
| 37 |
+
"learning_rate": 1.4545454545454545e-05,
|
| 38 |
+
"loss": 2.2475,
|
| 39 |
+
"step": 16
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.36,
|
| 43 |
+
"learning_rate": 1.8181818181818182e-05,
|
| 44 |
+
"loss": 2.2982,
|
| 45 |
+
"step": 20
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.44,
|
| 49 |
+
"learning_rate": 2.1818181818181818e-05,
|
| 50 |
+
"loss": 2.2629,
|
| 51 |
+
"step": 24
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.51,
|
| 55 |
+
"learning_rate": 2.5454545454545454e-05,
|
| 56 |
+
"loss": 2.4193,
|
| 57 |
+
"step": 28
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.58,
|
| 61 |
+
"learning_rate": 2.909090909090909e-05,
|
| 62 |
+
"loss": 2.2926,
|
| 63 |
+
"step": 32
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.65,
|
| 67 |
+
"learning_rate": 3.272727272727273e-05,
|
| 68 |
+
"loss": 2.1356,
|
| 69 |
+
"step": 36
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.73,
|
| 73 |
+
"learning_rate": 3.6363636363636364e-05,
|
| 74 |
+
"loss": 1.9494,
|
| 75 |
+
"step": 40
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.8,
|
| 79 |
+
"learning_rate": 4e-05,
|
| 80 |
+
"loss": 2.0152,
|
| 81 |
+
"step": 44
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.87,
|
| 85 |
+
"learning_rate": 4.3636363636363636e-05,
|
| 86 |
+
"loss": 2.1044,
|
| 87 |
+
"step": 48
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.95,
|
| 91 |
+
"learning_rate": 4.7272727272727275e-05,
|
| 92 |
+
"loss": 1.9755,
|
| 93 |
+
"step": 52
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.02,
|
| 97 |
+
"learning_rate": 4.999949650182266e-05,
|
| 98 |
+
"loss": 1.991,
|
| 99 |
+
"step": 56
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.09,
|
| 103 |
+
"learning_rate": 4.9987413559579636e-05,
|
| 104 |
+
"loss": 1.8347,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.16,
|
| 109 |
+
"learning_rate": 4.995922759815339e-05,
|
| 110 |
+
"loss": 1.9198,
|
| 111 |
+
"step": 64
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.24,
|
| 115 |
+
"learning_rate": 4.991495678185202e-05,
|
| 116 |
+
"loss": 1.8634,
|
| 117 |
+
"step": 68
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.31,
|
| 121 |
+
"learning_rate": 4.985462964079137e-05,
|
| 122 |
+
"loss": 1.8104,
|
| 123 |
+
"step": 72
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.38,
|
| 127 |
+
"learning_rate": 4.977828505250903e-05,
|
| 128 |
+
"loss": 1.7029,
|
| 129 |
+
"step": 76
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.45,
|
| 133 |
+
"learning_rate": 4.968597221690986e-05,
|
| 134 |
+
"loss": 1.7826,
|
| 135 |
+
"step": 80
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.53,
|
| 139 |
+
"learning_rate": 4.957775062455933e-05,
|
| 140 |
+
"loss": 1.7827,
|
| 141 |
+
"step": 84
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.6,
|
| 145 |
+
"learning_rate": 4.9453690018345144e-05,
|
| 146 |
+
"loss": 1.806,
|
| 147 |
+
"step": 88
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.67,
|
| 151 |
+
"learning_rate": 4.931387034853173e-05,
|
| 152 |
+
"loss": 1.7537,
|
| 153 |
+
"step": 92
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.75,
|
| 157 |
+
"learning_rate": 4.915838172123671e-05,
|
| 158 |
+
"loss": 1.7506,
|
| 159 |
+
"step": 96
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.82,
|
| 163 |
+
"learning_rate": 4.898732434036244e-05,
|
| 164 |
+
"loss": 1.7881,
|
| 165 |
+
"step": 100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.89,
|
| 169 |
+
"learning_rate": 4.880080844302004e-05,
|
| 170 |
+
"loss": 1.736,
|
| 171 |
+
"step": 104
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.96,
|
| 175 |
+
"learning_rate": 4.859895422848767e-05,
|
| 176 |
+
"loss": 1.6668,
|
| 177 |
+
"step": 108
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.0,
|
| 181 |
+
"gpt4_scores": 0.4759999999999999,
|
| 182 |
+
"step": 110
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 2.0,
|
| 186 |
+
"std": 0.0573103830034314,
|
| 187 |
+
"step": 110
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 2.0,
|
| 191 |
+
"eval_loss": 1.8269920349121094,
|
| 192 |
+
"eval_runtime": 4.9635,
|
| 193 |
+
"eval_samples_per_second": 4.634,
|
| 194 |
+
"eval_steps_per_second": 1.209,
|
| 195 |
+
"step": 110
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.04,
|
| 199 |
+
"learning_rate": 4.838189178074867e-05,
|
| 200 |
+
"loss": 1.8895,
|
| 201 |
+
"step": 112
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.11,
|
| 205 |
+
"learning_rate": 4.8149760984659506e-05,
|
| 206 |
+
"loss": 1.6187,
|
| 207 |
+
"step": 116
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.18,
|
| 211 |
+
"learning_rate": 4.790271143580174e-05,
|
| 212 |
+
"loss": 1.7413,
|
| 213 |
+
"step": 120
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.25,
|
| 217 |
+
"learning_rate": 4.764090234407577e-05,
|
| 218 |
+
"loss": 1.6013,
|
| 219 |
+
"step": 124
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.33,
|
| 223 |
+
"learning_rate": 4.7364502431098844e-05,
|
| 224 |
+
"loss": 1.5573,
|
| 225 |
+
"step": 128
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.4,
|
| 229 |
+
"learning_rate": 4.707368982147318e-05,
|
| 230 |
+
"loss": 1.6328,
|
| 231 |
+
"step": 132
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.47,
|
| 235 |
+
"learning_rate": 4.6768651927994434e-05,
|
| 236 |
+
"loss": 1.592,
|
| 237 |
+
"step": 136
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.55,
|
| 241 |
+
"learning_rate": 4.644958533087443e-05,
|
| 242 |
+
"loss": 1.5389,
|
| 243 |
+
"step": 140
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.62,
|
| 247 |
+
"learning_rate": 4.611669565105596e-05,
|
| 248 |
+
"loss": 1.5301,
|
| 249 |
+
"step": 144
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.69,
|
| 253 |
+
"learning_rate": 4.5770197417701365e-05,
|
| 254 |
+
"loss": 1.5516,
|
| 255 |
+
"step": 148
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.76,
|
| 259 |
+
"learning_rate": 4.5410313929940244e-05,
|
| 260 |
+
"loss": 1.5279,
|
| 261 |
+
"step": 152
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.84,
|
| 265 |
+
"learning_rate": 4.503727711296538e-05,
|
| 266 |
+
"loss": 1.6698,
|
| 267 |
+
"step": 156
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.91,
|
| 271 |
+
"learning_rate": 4.465132736856969e-05,
|
| 272 |
+
"loss": 1.7181,
|
| 273 |
+
"step": 160
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.98,
|
| 277 |
+
"learning_rate": 4.425271342022039e-05,
|
| 278 |
+
"loss": 1.5865,
|
| 279 |
+
"step": 164
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 3.0,
|
| 283 |
+
"gpt4_scores": 0.6679999999999999,
|
| 284 |
+
"step": 165
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 3.0,
|
| 288 |
+
"std": 0.05110303317808053,
|
| 289 |
+
"step": 165
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 3.0,
|
| 293 |
+
"eval_loss": 1.8256371021270752,
|
| 294 |
+
"eval_runtime": 4.9731,
|
| 295 |
+
"eval_samples_per_second": 4.625,
|
| 296 |
+
"eval_steps_per_second": 1.206,
|
| 297 |
+
"step": 165
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.05,
|
| 301 |
+
"learning_rate": 4.384169215277041e-05,
|
| 302 |
+
"loss": 1.3406,
|
| 303 |
+
"step": 168
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 3.13,
|
| 307 |
+
"learning_rate": 4.341852844691012e-05,
|
| 308 |
+
"loss": 1.4171,
|
| 309 |
+
"step": 172
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.2,
|
| 313 |
+
"learning_rate": 4.2983495008466276e-05,
|
| 314 |
+
"loss": 1.3461,
|
| 315 |
+
"step": 176
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 3.27,
|
| 319 |
+
"learning_rate": 4.2536872192658036e-05,
|
| 320 |
+
"loss": 1.408,
|
| 321 |
+
"step": 180
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.35,
|
| 325 |
+
"learning_rate": 4.2078947823423364e-05,
|
| 326 |
+
"loss": 1.3575,
|
| 327 |
+
"step": 184
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 3.42,
|
| 331 |
+
"learning_rate": 4.161001700793231e-05,
|
| 332 |
+
"loss": 1.3114,
|
| 333 |
+
"step": 188
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 3.49,
|
| 337 |
+
"learning_rate": 4.113038194640658e-05,
|
| 338 |
+
"loss": 1.2919,
|
| 339 |
+
"step": 192
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.56,
|
| 343 |
+
"learning_rate": 4.064035173736804e-05,
|
| 344 |
+
"loss": 1.3044,
|
| 345 |
+
"step": 196
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.64,
|
| 349 |
+
"learning_rate": 4.014024217844167e-05,
|
| 350 |
+
"loss": 1.3425,
|
| 351 |
+
"step": 200
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.71,
|
| 355 |
+
"learning_rate": 3.9630375562841295e-05,
|
| 356 |
+
"loss": 1.3524,
|
| 357 |
+
"step": 204
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 3.78,
|
| 361 |
+
"learning_rate": 3.911108047166924e-05,
|
| 362 |
+
"loss": 1.3514,
|
| 363 |
+
"step": 208
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 3.85,
|
| 367 |
+
"learning_rate": 3.858269156216383e-05,
|
| 368 |
+
"loss": 1.2055,
|
| 369 |
+
"step": 212
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 3.93,
|
| 373 |
+
"learning_rate": 3.804554935203115e-05,
|
| 374 |
+
"loss": 1.3618,
|
| 375 |
+
"step": 216
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 4.0,
|
| 379 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 380 |
+
"loss": 1.264,
|
| 381 |
+
"step": 220
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 4.0,
|
| 385 |
+
"gpt4_scores": 0.6719999999999999,
|
| 386 |
+
"step": 220
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 4.0,
|
| 390 |
+
"std": 0.049561275205547324,
|
| 391 |
+
"step": 220
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 4.0,
|
| 395 |
+
"eval_loss": 1.9825626611709595,
|
| 396 |
+
"eval_runtime": 4.9575,
|
| 397 |
+
"eval_samples_per_second": 4.639,
|
| 398 |
+
"eval_steps_per_second": 1.21,
|
| 399 |
+
"step": 220
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.07,
|
| 403 |
+
"learning_rate": 3.694639508274158e-05,
|
| 404 |
+
"loss": 1.0472,
|
| 405 |
+
"step": 224
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 4.15,
|
| 409 |
+
"learning_rate": 3.638509136829758e-05,
|
| 410 |
+
"loss": 1.0674,
|
| 411 |
+
"step": 228
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 4.22,
|
| 415 |
+
"learning_rate": 3.581645058616271e-05,
|
| 416 |
+
"loss": 0.9661,
|
| 417 |
+
"step": 232
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.29,
|
| 421 |
+
"learning_rate": 3.5240839194169885e-05,
|
| 422 |
+
"loss": 1.0761,
|
| 423 |
+
"step": 236
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.36,
|
| 427 |
+
"learning_rate": 3.465862814232822e-05,
|
| 428 |
+
"loss": 1.0675,
|
| 429 |
+
"step": 240
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 4.44,
|
| 433 |
+
"learning_rate": 3.4070192633766025e-05,
|
| 434 |
+
"loss": 0.9969,
|
| 435 |
+
"step": 244
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.51,
|
| 439 |
+
"learning_rate": 3.3475911882933015e-05,
|
| 440 |
+
"loss": 1.0195,
|
| 441 |
+
"step": 248
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.58,
|
| 445 |
+
"learning_rate": 3.2876168871217325e-05,
|
| 446 |
+
"loss": 1.0246,
|
| 447 |
+
"step": 252
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 4.65,
|
| 451 |
+
"learning_rate": 3.2271350100134975e-05,
|
| 452 |
+
"loss": 1.0308,
|
| 453 |
+
"step": 256
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 4.73,
|
| 457 |
+
"learning_rate": 3.166184534225087e-05,
|
| 458 |
+
"loss": 0.9465,
|
| 459 |
+
"step": 260
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 4.8,
|
| 463 |
+
"learning_rate": 3.104804738999169e-05,
|
| 464 |
+
"loss": 1.0121,
|
| 465 |
+
"step": 264
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.87,
|
| 469 |
+
"learning_rate": 3.0430351802512698e-05,
|
| 470 |
+
"loss": 0.9884,
|
| 471 |
+
"step": 268
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 4.95,
|
| 475 |
+
"learning_rate": 2.9809156650781528e-05,
|
| 476 |
+
"loss": 0.99,
|
| 477 |
+
"step": 272
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 5.0,
|
| 481 |
+
"gpt4_scores": 0.6259999999999999,
|
| 482 |
+
"step": 275
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 5.0,
|
| 486 |
+
"std": 0.05452045487704591,
|
| 487 |
+
"step": 275
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 5.0,
|
| 491 |
+
"eval_loss": 2.212796449661255,
|
| 492 |
+
"eval_runtime": 4.9762,
|
| 493 |
+
"eval_samples_per_second": 4.622,
|
| 494 |
+
"eval_steps_per_second": 1.206,
|
| 495 |
+
"step": 275
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 5.02,
|
| 499 |
+
"learning_rate": 2.918486226104327e-05,
|
| 500 |
+
"loss": 0.9253,
|
| 501 |
+
"step": 276
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 5.09,
|
| 505 |
+
"learning_rate": 2.8557870956832132e-05,
|
| 506 |
+
"loss": 0.7838,
|
| 507 |
+
"step": 280
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 5.16,
|
| 511 |
+
"learning_rate": 2.792858679969596e-05,
|
| 512 |
+
"loss": 0.8005,
|
| 513 |
+
"step": 284
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 5.24,
|
| 517 |
+
"learning_rate": 2.7297415328800692e-05,
|
| 518 |
+
"loss": 0.7363,
|
| 519 |
+
"step": 288
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.31,
|
| 523 |
+
"learning_rate": 2.6664763299582602e-05,
|
| 524 |
+
"loss": 0.6807,
|
| 525 |
+
"step": 292
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 5.38,
|
| 529 |
+
"learning_rate": 2.6031038421616683e-05,
|
| 530 |
+
"loss": 0.7882,
|
| 531 |
+
"step": 296
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 5.45,
|
| 535 |
+
"learning_rate": 2.5396649095870202e-05,
|
| 536 |
+
"loss": 0.6869,
|
| 537 |
+
"step": 300
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 5.53,
|
| 541 |
+
"learning_rate": 2.4762004151510584e-05,
|
| 542 |
+
"loss": 0.7016,
|
| 543 |
+
"step": 304
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 5.6,
|
| 547 |
+
"learning_rate": 2.4127512582437485e-05,
|
| 548 |
+
"loss": 0.6709,
|
| 549 |
+
"step": 308
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 5.67,
|
| 553 |
+
"learning_rate": 2.349358328370854e-05,
|
| 554 |
+
"loss": 0.7557,
|
| 555 |
+
"step": 312
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 5.75,
|
| 559 |
+
"learning_rate": 2.2860624788029013e-05,
|
| 560 |
+
"loss": 0.7442,
|
| 561 |
+
"step": 316
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 5.82,
|
| 565 |
+
"learning_rate": 2.222904500247473e-05,
|
| 566 |
+
"loss": 0.7328,
|
| 567 |
+
"step": 320
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 5.89,
|
| 571 |
+
"learning_rate": 2.1599250945618402e-05,
|
| 572 |
+
"loss": 0.663,
|
| 573 |
+
"step": 324
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 5.96,
|
| 577 |
+
"learning_rate": 2.09716484852284e-05,
|
| 578 |
+
"loss": 0.7218,
|
| 579 |
+
"step": 328
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 6.0,
|
| 583 |
+
"gpt4_scores": 0.54,
|
| 584 |
+
"step": 330
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 6.0,
|
| 588 |
+
"std": 0.06151422599691879,
|
| 589 |
+
"step": 330
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 6.0,
|
| 593 |
+
"eval_loss": 2.538802146911621,
|
| 594 |
+
"eval_runtime": 4.9512,
|
| 595 |
+
"eval_samples_per_second": 4.645,
|
| 596 |
+
"eval_steps_per_second": 1.212,
|
| 597 |
+
"step": 330
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 6.04,
|
| 601 |
+
"learning_rate": 2.034664207670925e-05,
|
| 602 |
+
"loss": 0.61,
|
| 603 |
+
"step": 332
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 6.11,
|
| 607 |
+
"learning_rate": 1.972463450245226e-05,
|
| 608 |
+
"loss": 0.4884,
|
| 609 |
+
"step": 336
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 6.18,
|
| 613 |
+
"learning_rate": 1.9106026612264316e-05,
|
| 614 |
+
"loss": 0.5136,
|
| 615 |
+
"step": 340
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 6.25,
|
| 619 |
+
"learning_rate": 1.84912170650422e-05,
|
| 620 |
+
"loss": 0.5489,
|
| 621 |
+
"step": 344
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 6.33,
|
| 625 |
+
"learning_rate": 1.7880602071858692e-05,
|
| 626 |
+
"loss": 0.51,
|
| 627 |
+
"step": 348
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 6.4,
|
| 631 |
+
"learning_rate": 1.7274575140626318e-05,
|
| 632 |
+
"loss": 0.4517,
|
| 633 |
+
"step": 352
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 6.47,
|
| 637 |
+
"learning_rate": 1.667352682250298e-05,
|
| 638 |
+
"loss": 0.5259,
|
| 639 |
+
"step": 356
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 6.55,
|
| 643 |
+
"learning_rate": 1.6077844460203206e-05,
|
| 644 |
+
"loss": 0.5205,
|
| 645 |
+
"step": 360
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 6.62,
|
| 649 |
+
"learning_rate": 1.5487911938376924e-05,
|
| 650 |
+
"loss": 0.5436,
|
| 651 |
+
"step": 364
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 6.69,
|
| 655 |
+
"learning_rate": 1.4904109436216884e-05,
|
| 656 |
+
"loss": 0.5362,
|
| 657 |
+
"step": 368
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 6.76,
|
| 661 |
+
"learning_rate": 1.4326813182453958e-05,
|
| 662 |
+
"loss": 0.4788,
|
| 663 |
+
"step": 372
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 6.84,
|
| 667 |
+
"learning_rate": 1.3756395212898359e-05,
|
| 668 |
+
"loss": 0.4858,
|
| 669 |
+
"step": 376
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 6.91,
|
| 673 |
+
"learning_rate": 1.3193223130682936e-05,
|
| 674 |
+
"loss": 0.5685,
|
| 675 |
+
"step": 380
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 6.98,
|
| 679 |
+
"learning_rate": 1.2637659869363083e-05,
|
| 680 |
+
"loss": 0.5291,
|
| 681 |
+
"step": 384
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 7.0,
|
| 685 |
+
"gpt4_scores": 0.5159999999999999,
|
| 686 |
+
"step": 385
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 7.0,
|
| 690 |
+
"std": 0.058265598769771526,
|
| 691 |
+
"step": 385
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 7.0,
|
| 695 |
+
"eval_loss": 2.774492025375366,
|
| 696 |
+
"eval_runtime": 4.9875,
|
| 697 |
+
"eval_samples_per_second": 4.612,
|
| 698 |
+
"eval_steps_per_second": 1.203,
|
| 699 |
+
"step": 385
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 7.05,
|
| 703 |
+
"learning_rate": 1.2090063459025955e-05,
|
| 704 |
+
"loss": 0.3004,
|
| 705 |
+
"step": 388
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 7.13,
|
| 709 |
+
"learning_rate": 1.155078679555969e-05,
|
| 710 |
+
"loss": 0.3865,
|
| 711 |
+
"step": 392
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 7.2,
|
| 715 |
+
"learning_rate": 1.1020177413231334e-05,
|
| 716 |
+
"loss": 0.3657,
|
| 717 |
+
"step": 396
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 7.27,
|
| 721 |
+
"learning_rate": 1.049857726072005e-05,
|
| 722 |
+
"loss": 0.347,
|
| 723 |
+
"step": 400
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 7.35,
|
| 727 |
+
"learning_rate": 9.986322480749927e-06,
|
| 728 |
+
"loss": 0.3479,
|
| 729 |
+
"step": 404
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 7.42,
|
| 733 |
+
"learning_rate": 9.483743193464408e-06,
|
| 734 |
+
"loss": 0.4031,
|
| 735 |
+
"step": 408
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 7.49,
|
| 739 |
+
"learning_rate": 8.991163283681944e-06,
|
| 740 |
+
"loss": 0.4777,
|
| 741 |
+
"step": 412
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 7.56,
|
| 745 |
+
"learning_rate": 8.508900192169964e-06,
|
| 746 |
+
"loss": 0.3905,
|
| 747 |
+
"step": 416
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 7.64,
|
| 751 |
+
"learning_rate": 8.0372647110717e-06,
|
| 752 |
+
"loss": 0.3871,
|
| 753 |
+
"step": 420
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 7.71,
|
| 757 |
+
"learning_rate": 7.576560783617668e-06,
|
| 758 |
+
"loss": 0.3674,
|
| 759 |
+
"step": 424
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 7.78,
|
| 763 |
+
"learning_rate": 7.127085308250914e-06,
|
| 764 |
+
"loss": 0.3878,
|
| 765 |
+
"step": 428
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 7.85,
|
| 769 |
+
"learning_rate": 6.689127947292231e-06,
|
| 770 |
+
"loss": 0.4137,
|
| 771 |
+
"step": 432
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 7.93,
|
| 775 |
+
"learning_rate": 6.2629709402686535e-06,
|
| 776 |
+
"loss": 0.4006,
|
| 777 |
+
"step": 436
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 8.0,
|
| 781 |
+
"learning_rate": 5.848888922025553e-06,
|
| 782 |
+
"loss": 0.3732,
|
| 783 |
+
"step": 440
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 8.07,
|
| 787 |
+
"learning_rate": 5.4471487457395225e-06,
|
| 788 |
+
"loss": 1.2569,
|
| 789 |
+
"step": 444
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 8.15,
|
| 793 |
+
"learning_rate": 5.058009310946119e-06,
|
| 794 |
+
"loss": 1.2022,
|
| 795 |
+
"step": 448
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 8.22,
|
| 799 |
+
"learning_rate": 4.681721396693303e-06,
|
| 800 |
+
"loss": 1.0283,
|
| 801 |
+
"step": 452
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 8.29,
|
| 805 |
+
"learning_rate": 4.318527499928074e-06,
|
| 806 |
+
"loss": 0.8692,
|
| 807 |
+
"step": 456
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 8.36,
|
| 811 |
+
"learning_rate": 3.968661679220468e-06,
|
| 812 |
+
"loss": 0.9156,
|
| 813 |
+
"step": 460
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 8.44,
|
| 817 |
+
"learning_rate": 3.632349403925664e-06,
|
| 818 |
+
"loss": 0.9772,
|
| 819 |
+
"step": 464
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 8.51,
|
| 823 |
+
"learning_rate": 3.3098074088812686e-06,
|
| 824 |
+
"loss": 0.9159,
|
| 825 |
+
"step": 468
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 8.58,
|
| 829 |
+
"learning_rate": 3.0012435547336737e-06,
|
| 830 |
+
"loss": 0.7694,
|
| 831 |
+
"step": 472
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 8.65,
|
| 835 |
+
"learning_rate": 2.7068566939831645e-06,
|
| 836 |
+
"loss": 0.8599,
|
| 837 |
+
"step": 476
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 8.73,
|
| 841 |
+
"learning_rate": 2.4268365428344736e-06,
|
| 842 |
+
"loss": 0.8493,
|
| 843 |
+
"step": 480
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 8.8,
|
| 847 |
+
"learning_rate": 2.1613635589349756e-06,
|
| 848 |
+
"loss": 0.9502,
|
| 849 |
+
"step": 484
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 8.87,
|
| 853 |
+
"learning_rate": 1.9106088250797267e-06,
|
| 854 |
+
"loss": 0.8163,
|
| 855 |
+
"step": 488
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 8.95,
|
| 859 |
+
"learning_rate": 1.674733938957873e-06,
|
| 860 |
+
"loss": 0.791,
|
| 861 |
+
"step": 492
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 9.02,
|
| 865 |
+
"learning_rate": 1.4538909090118846e-06,
|
| 866 |
+
"loss": 0.8613,
|
| 867 |
+
"step": 496
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 9.09,
|
| 871 |
+
"learning_rate": 1.248222056476367e-06,
|
| 872 |
+
"loss": 0.7612,
|
| 873 |
+
"step": 500
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 9.16,
|
| 877 |
+
"learning_rate": 1.0578599236598707e-06,
|
| 878 |
+
"loss": 0.7711,
|
| 879 |
+
"step": 504
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 9.24,
|
| 883 |
+
"learning_rate": 8.829271885286094e-07,
|
| 884 |
+
"loss": 0.7856,
|
| 885 |
+
"step": 508
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 9.31,
|
| 889 |
+
"learning_rate": 7.235365856472442e-07,
|
| 890 |
+
"loss": 0.8414,
|
| 891 |
+
"step": 512
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 9.38,
|
| 895 |
+
"learning_rate": 5.797908335276214e-07,
|
| 896 |
+
"loss": 0.7963,
|
| 897 |
+
"step": 516
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 9.45,
|
| 901 |
+
"learning_rate": 4.517825684323324e-07,
|
| 902 |
+
"loss": 0.7588,
|
| 903 |
+
"step": 520
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 9.53,
|
| 907 |
+
"learning_rate": 3.395942846757066e-07,
|
| 908 |
+
"loss": 0.7257,
|
| 909 |
+
"step": 524
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 9.6,
|
| 913 |
+
"learning_rate": 2.4329828146074095e-07,
|
| 914 |
+
"loss": 0.8102,
|
| 915 |
+
"step": 528
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 9.67,
|
| 919 |
+
"learning_rate": 1.6295661628624447e-07,
|
| 920 |
+
"loss": 0.8791,
|
| 921 |
+
"step": 532
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 9.75,
|
| 925 |
+
"learning_rate": 9.862106495415469e-08,
|
| 926 |
+
"loss": 0.7275,
|
| 927 |
+
"step": 536
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 9.82,
|
| 931 |
+
"learning_rate": 5.033308820289184e-08,
|
| 932 |
+
"loss": 0.8194,
|
| 933 |
+
"step": 540
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 9.89,
|
| 937 |
+
"learning_rate": 1.812380498815991e-08,
|
| 938 |
+
"loss": 0.892,
|
| 939 |
+
"step": 544
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 9.96,
|
| 943 |
+
"learning_rate": 2.0139724285161977e-09,
|
| 944 |
+
"loss": 0.8365,
|
| 945 |
+
"step": 548
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 10.0,
|
| 949 |
+
"step": 550,
|
| 950 |
+
"total_flos": 1.881214300967731e+16,
|
| 951 |
+
"train_loss": 0.0,
|
| 952 |
+
"train_runtime": 12.8072,
|
| 953 |
+
"train_samples_per_second": 169.436,
|
| 954 |
+
"train_steps_per_second": 42.945
|
| 955 |
+
}
|
| 956 |
+
],
|
| 957 |
+
"logging_steps": 4,
|
| 958 |
+
"max_steps": 550,
|
| 959 |
+
"num_input_tokens_seen": 0,
|
| 960 |
+
"num_train_epochs": 10,
|
| 961 |
+
"save_steps": 55,
|
| 962 |
+
"total_flos": 1.881214300967731e+16,
|
| 963 |
+
"train_batch_size": 4,
|
| 964 |
+
"trial_name": null,
|
| 965 |
+
"trial_params": null
|
| 966 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4792
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2ac1a75e32f09f33c03788de3c043f3d4c6f3120526cf79ded4b1d581c116179
|
| 3 |
size 4792
|