Model save
Browse files- README.md +81 -0
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +2046 -0
README.md
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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-4-50
|
| 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-4-50
|
| 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: 3.5394
|
| 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: 0.0005
|
| 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: 20
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 53 |
+
| 1.8768 | 1.0 | 55 | 1.8434 |
|
| 54 |
+
| 1.5109 | 2.0 | 110 | 1.8907 |
|
| 55 |
+
| 1.1215 | 3.0 | 165 | 2.1649 |
|
| 56 |
+
| 0.4876 | 4.0 | 220 | 2.6900 |
|
| 57 |
+
| 0.2916 | 5.0 | 275 | 2.8212 |
|
| 58 |
+
| 0.1983 | 6.0 | 330 | 2.8984 |
|
| 59 |
+
| 0.1199 | 7.0 | 385 | 2.9498 |
|
| 60 |
+
| 0.0743 | 8.0 | 440 | 3.2507 |
|
| 61 |
+
| 0.062 | 9.0 | 495 | 3.1474 |
|
| 62 |
+
| 0.0404 | 10.0 | 550 | 3.3666 |
|
| 63 |
+
| 0.038 | 11.0 | 605 | 3.3349 |
|
| 64 |
+
| 0.0374 | 12.0 | 660 | 3.4456 |
|
| 65 |
+
| 0.0384 | 13.0 | 715 | 3.4822 |
|
| 66 |
+
| 0.0408 | 14.0 | 770 | 3.4718 |
|
| 67 |
+
| 0.0347 | 15.0 | 825 | 3.5028 |
|
| 68 |
+
| 0.0377 | 16.0 | 880 | 3.5218 |
|
| 69 |
+
| 0.0395 | 17.0 | 935 | 3.5320 |
|
| 70 |
+
| 0.0408 | 18.0 | 990 | 3.5371 |
|
| 71 |
+
| 0.0468 | 19.0 | 1045 | 3.5391 |
|
| 72 |
+
| 0.0358 | 20.0 | 1100 | 3.5394 |
|
| 73 |
+
|
| 74 |
+
|
| 75 |
+
### Framework versions
|
| 76 |
+
|
| 77 |
+
- PEFT 0.7.1
|
| 78 |
+
- Transformers 4.36.2
|
| 79 |
+
- Pytorch 2.2.2+cu118
|
| 80 |
+
- Datasets 2.14.6
|
| 81 |
+
- Tokenizers 0.15.2
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 3.5393826961517334,
|
| 4 |
+
"eval_runtime": 4.9183,
|
| 5 |
+
"eval_samples": 23,
|
| 6 |
+
"eval_samples_per_second": 4.676,
|
| 7 |
+
"eval_steps_per_second": 1.22,
|
| 8 |
+
"train_loss": 0.31520286625081845,
|
| 9 |
+
"train_runtime": 10333.7168,
|
| 10 |
+
"train_samples": 217,
|
| 11 |
+
"train_samples_per_second": 0.42,
|
| 12 |
+
"train_steps_per_second": 0.106
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 3.5393826961517334,
|
| 4 |
+
"eval_runtime": 4.9183,
|
| 5 |
+
"eval_samples": 23,
|
| 6 |
+
"eval_samples_per_second": 4.676,
|
| 7 |
+
"eval_steps_per_second": 1.22
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"train_loss": 0.31520286625081845,
|
| 4 |
+
"train_runtime": 10333.7168,
|
| 5 |
+
"train_samples": 217,
|
| 6 |
+
"train_samples_per_second": 0.42,
|
| 7 |
+
"train_steps_per_second": 0.106
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2046 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1100,
|
| 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": 4.5454545454545455e-06,
|
| 14 |
+
"loss": 2.3833,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.07,
|
| 19 |
+
"learning_rate": 1.8181818181818182e-05,
|
| 20 |
+
"loss": 2.4734,
|
| 21 |
+
"step": 4
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.15,
|
| 25 |
+
"learning_rate": 3.6363636363636364e-05,
|
| 26 |
+
"loss": 2.2655,
|
| 27 |
+
"step": 8
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.22,
|
| 31 |
+
"learning_rate": 5.4545454545454546e-05,
|
| 32 |
+
"loss": 2.2091,
|
| 33 |
+
"step": 12
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.29,
|
| 37 |
+
"learning_rate": 7.272727272727273e-05,
|
| 38 |
+
"loss": 2.1358,
|
| 39 |
+
"step": 16
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.36,
|
| 43 |
+
"learning_rate": 9.090909090909092e-05,
|
| 44 |
+
"loss": 2.0997,
|
| 45 |
+
"step": 20
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.44,
|
| 49 |
+
"learning_rate": 0.00010909090909090909,
|
| 50 |
+
"loss": 1.931,
|
| 51 |
+
"step": 24
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.51,
|
| 55 |
+
"learning_rate": 0.00012727272727272725,
|
| 56 |
+
"loss": 2.0453,
|
| 57 |
+
"step": 28
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.58,
|
| 61 |
+
"learning_rate": 0.00014545454545454546,
|
| 62 |
+
"loss": 1.9392,
|
| 63 |
+
"step": 32
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.65,
|
| 67 |
+
"learning_rate": 0.00016363636363636363,
|
| 68 |
+
"loss": 1.8909,
|
| 69 |
+
"step": 36
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.73,
|
| 73 |
+
"learning_rate": 0.00018181818181818183,
|
| 74 |
+
"loss": 1.7724,
|
| 75 |
+
"step": 40
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.8,
|
| 79 |
+
"learning_rate": 0.0002,
|
| 80 |
+
"loss": 1.854,
|
| 81 |
+
"step": 44
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.87,
|
| 85 |
+
"learning_rate": 0.00021818181818181818,
|
| 86 |
+
"loss": 1.9588,
|
| 87 |
+
"step": 48
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.95,
|
| 91 |
+
"learning_rate": 0.00023636363636363636,
|
| 92 |
+
"loss": 1.8768,
|
| 93 |
+
"step": 52
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"gpt4_scores": 0.67,
|
| 98 |
+
"step": 55
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.0,
|
| 102 |
+
"std": 0.12413702106946178,
|
| 103 |
+
"step": 55
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 1.0,
|
| 107 |
+
"eval_loss": 1.84342360496521,
|
| 108 |
+
"eval_runtime": 4.9467,
|
| 109 |
+
"eval_samples_per_second": 4.65,
|
| 110 |
+
"eval_steps_per_second": 1.213,
|
| 111 |
+
"step": 55
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.02,
|
| 115 |
+
"learning_rate": 0.0002545454545454545,
|
| 116 |
+
"loss": 1.7451,
|
| 117 |
+
"step": 56
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.09,
|
| 121 |
+
"learning_rate": 0.00027272727272727274,
|
| 122 |
+
"loss": 1.7394,
|
| 123 |
+
"step": 60
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.16,
|
| 127 |
+
"learning_rate": 0.0002909090909090909,
|
| 128 |
+
"loss": 1.6971,
|
| 129 |
+
"step": 64
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.24,
|
| 133 |
+
"learning_rate": 0.0003090909090909091,
|
| 134 |
+
"loss": 1.6419,
|
| 135 |
+
"step": 68
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.31,
|
| 139 |
+
"learning_rate": 0.00032727272727272726,
|
| 140 |
+
"loss": 1.65,
|
| 141 |
+
"step": 72
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.38,
|
| 145 |
+
"learning_rate": 0.00034545454545454544,
|
| 146 |
+
"loss": 1.594,
|
| 147 |
+
"step": 76
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.45,
|
| 151 |
+
"learning_rate": 0.00036363636363636367,
|
| 152 |
+
"loss": 1.5985,
|
| 153 |
+
"step": 80
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.53,
|
| 157 |
+
"learning_rate": 0.00038181818181818184,
|
| 158 |
+
"loss": 1.719,
|
| 159 |
+
"step": 84
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.6,
|
| 163 |
+
"learning_rate": 0.0004,
|
| 164 |
+
"loss": 1.655,
|
| 165 |
+
"step": 88
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.67,
|
| 169 |
+
"learning_rate": 0.00041818181818181814,
|
| 170 |
+
"loss": 1.6052,
|
| 171 |
+
"step": 92
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.75,
|
| 175 |
+
"learning_rate": 0.00043636363636363637,
|
| 176 |
+
"loss": 1.6288,
|
| 177 |
+
"step": 96
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.82,
|
| 181 |
+
"learning_rate": 0.00045454545454545455,
|
| 182 |
+
"loss": 1.6272,
|
| 183 |
+
"step": 100
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.89,
|
| 187 |
+
"learning_rate": 0.0004727272727272727,
|
| 188 |
+
"loss": 1.6124,
|
| 189 |
+
"step": 104
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.96,
|
| 193 |
+
"learning_rate": 0.0004909090909090909,
|
| 194 |
+
"loss": 1.5109,
|
| 195 |
+
"step": 108
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.0,
|
| 199 |
+
"gpt4_scores": 0.675,
|
| 200 |
+
"step": 110
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.0,
|
| 204 |
+
"std": 0.11688669727560959,
|
| 205 |
+
"step": 110
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 2.0,
|
| 209 |
+
"eval_loss": 1.89065682888031,
|
| 210 |
+
"eval_runtime": 4.9546,
|
| 211 |
+
"eval_samples_per_second": 4.642,
|
| 212 |
+
"eval_steps_per_second": 1.211,
|
| 213 |
+
"step": 110
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.04,
|
| 217 |
+
"learning_rate": 0.0004999949650182266,
|
| 218 |
+
"loss": 1.3323,
|
| 219 |
+
"step": 112
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.11,
|
| 223 |
+
"learning_rate": 0.0004999546863808815,
|
| 224 |
+
"loss": 1.0715,
|
| 225 |
+
"step": 116
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.18,
|
| 229 |
+
"learning_rate": 0.0004998741355957963,
|
| 230 |
+
"loss": 0.8812,
|
| 231 |
+
"step": 120
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.25,
|
| 235 |
+
"learning_rate": 0.0004997533256411359,
|
| 236 |
+
"loss": 0.9464,
|
| 237 |
+
"step": 124
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.33,
|
| 241 |
+
"learning_rate": 0.0004995922759815339,
|
| 242 |
+
"loss": 0.9049,
|
| 243 |
+
"step": 128
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.4,
|
| 247 |
+
"learning_rate": 0.0004993910125649561,
|
| 248 |
+
"loss": 0.9246,
|
| 249 |
+
"step": 132
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.47,
|
| 253 |
+
"learning_rate": 0.0004991495678185201,
|
| 254 |
+
"loss": 1.0853,
|
| 255 |
+
"step": 136
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.55,
|
| 259 |
+
"learning_rate": 0.0004988679806432712,
|
| 260 |
+
"loss": 1.1289,
|
| 261 |
+
"step": 140
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.62,
|
| 265 |
+
"learning_rate": 0.0004985462964079136,
|
| 266 |
+
"loss": 0.9598,
|
| 267 |
+
"step": 144
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.69,
|
| 271 |
+
"learning_rate": 0.0004981845669415021,
|
| 272 |
+
"loss": 0.8825,
|
| 273 |
+
"step": 148
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.76,
|
| 277 |
+
"learning_rate": 0.0004977828505250904,
|
| 278 |
+
"loss": 0.985,
|
| 279 |
+
"step": 152
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.84,
|
| 283 |
+
"learning_rate": 0.0004973412118823412,
|
| 284 |
+
"loss": 0.9869,
|
| 285 |
+
"step": 156
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.91,
|
| 289 |
+
"learning_rate": 0.0004968597221690986,
|
| 290 |
+
"loss": 0.985,
|
| 291 |
+
"step": 160
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.98,
|
| 295 |
+
"learning_rate": 0.0004963384589619233,
|
| 296 |
+
"loss": 1.1215,
|
| 297 |
+
"step": 164
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.0,
|
| 301 |
+
"gpt4_scores": 0.67,
|
| 302 |
+
"step": 165
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.0,
|
| 306 |
+
"std": 0.11920570456148481,
|
| 307 |
+
"step": 165
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 3.0,
|
| 311 |
+
"eval_loss": 2.164942741394043,
|
| 312 |
+
"eval_runtime": 4.9498,
|
| 313 |
+
"eval_samples_per_second": 4.647,
|
| 314 |
+
"eval_steps_per_second": 1.212,
|
| 315 |
+
"step": 165
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 3.05,
|
| 319 |
+
"learning_rate": 0.0004957775062455933,
|
| 320 |
+
"loss": 0.8362,
|
| 321 |
+
"step": 168
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.13,
|
| 325 |
+
"learning_rate": 0.0004951769543995731,
|
| 326 |
+
"loss": 0.3812,
|
| 327 |
+
"step": 172
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 3.2,
|
| 331 |
+
"learning_rate": 0.0004945369001834514,
|
| 332 |
+
"loss": 0.4813,
|
| 333 |
+
"step": 176
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 3.27,
|
| 337 |
+
"learning_rate": 0.0004938574467213517,
|
| 338 |
+
"loss": 0.4693,
|
| 339 |
+
"step": 180
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.35,
|
| 343 |
+
"learning_rate": 0.0004931387034853173,
|
| 344 |
+
"loss": 0.447,
|
| 345 |
+
"step": 184
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.42,
|
| 349 |
+
"learning_rate": 0.0004923807862776728,
|
| 350 |
+
"loss": 0.6197,
|
| 351 |
+
"step": 188
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.49,
|
| 355 |
+
"learning_rate": 0.0004915838172123671,
|
| 356 |
+
"loss": 0.5158,
|
| 357 |
+
"step": 192
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 3.56,
|
| 361 |
+
"learning_rate": 0.0004907479246952981,
|
| 362 |
+
"loss": 0.5785,
|
| 363 |
+
"step": 196
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 3.64,
|
| 367 |
+
"learning_rate": 0.0004898732434036243,
|
| 368 |
+
"loss": 0.5258,
|
| 369 |
+
"step": 200
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 3.71,
|
| 373 |
+
"learning_rate": 0.0004889599142640663,
|
| 374 |
+
"loss": 0.4577,
|
| 375 |
+
"step": 204
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.78,
|
| 379 |
+
"learning_rate": 0.0004880080844302004,
|
| 380 |
+
"loss": 0.5965,
|
| 381 |
+
"step": 208
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.85,
|
| 385 |
+
"learning_rate": 0.0004870179072587499,
|
| 386 |
+
"loss": 0.5373,
|
| 387 |
+
"step": 212
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.93,
|
| 391 |
+
"learning_rate": 0.0004859895422848767,
|
| 392 |
+
"loss": 0.4731,
|
| 393 |
+
"step": 216
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 4.0,
|
| 397 |
+
"learning_rate": 0.0004849231551964771,
|
| 398 |
+
"loss": 0.4876,
|
| 399 |
+
"step": 220
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.0,
|
| 403 |
+
"gpt4_scores": 0.61,
|
| 404 |
+
"step": 220
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.0,
|
| 408 |
+
"std": 0.12445882853377657,
|
| 409 |
+
"step": 220
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.0,
|
| 413 |
+
"eval_loss": 2.6900253295898438,
|
| 414 |
+
"eval_runtime": 4.9381,
|
| 415 |
+
"eval_samples_per_second": 4.658,
|
| 416 |
+
"eval_steps_per_second": 1.215,
|
| 417 |
+
"step": 220
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.07,
|
| 421 |
+
"learning_rate": 0.00048381891780748665,
|
| 422 |
+
"loss": 0.288,
|
| 423 |
+
"step": 224
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.15,
|
| 427 |
+
"learning_rate": 0.00048267700803019775,
|
| 428 |
+
"loss": 0.2556,
|
| 429 |
+
"step": 228
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 4.22,
|
| 433 |
+
"learning_rate": 0.0004814976098465951,
|
| 434 |
+
"loss": 0.2897,
|
| 435 |
+
"step": 232
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.29,
|
| 439 |
+
"learning_rate": 0.00048028091327871256,
|
| 440 |
+
"loss": 0.2375,
|
| 441 |
+
"step": 236
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.36,
|
| 445 |
+
"learning_rate": 0.0004790271143580174,
|
| 446 |
+
"loss": 0.2367,
|
| 447 |
+
"step": 240
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 4.44,
|
| 451 |
+
"learning_rate": 0.00047773641509382626,
|
| 452 |
+
"loss": 0.2878,
|
| 453 |
+
"step": 244
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 4.51,
|
| 457 |
+
"learning_rate": 0.0004764090234407577,
|
| 458 |
+
"loss": 0.3138,
|
| 459 |
+
"step": 248
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 4.58,
|
| 463 |
+
"learning_rate": 0.00047504515326522696,
|
| 464 |
+
"loss": 0.277,
|
| 465 |
+
"step": 252
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.65,
|
| 469 |
+
"learning_rate": 0.0004736450243109884,
|
| 470 |
+
"loss": 0.2624,
|
| 471 |
+
"step": 256
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 4.73,
|
| 475 |
+
"learning_rate": 0.0004722088621637309,
|
| 476 |
+
"loss": 0.2888,
|
| 477 |
+
"step": 260
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 4.8,
|
| 481 |
+
"learning_rate": 0.00047073689821473173,
|
| 482 |
+
"loss": 0.2256,
|
| 483 |
+
"step": 264
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 4.87,
|
| 487 |
+
"learning_rate": 0.00046922936962357577,
|
| 488 |
+
"loss": 0.2243,
|
| 489 |
+
"step": 268
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 4.95,
|
| 493 |
+
"learning_rate": 0.00046768651927994433,
|
| 494 |
+
"loss": 0.2916,
|
| 495 |
+
"step": 272
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 5.0,
|
| 499 |
+
"gpt4_scores": 0.63,
|
| 500 |
+
"step": 275
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 5.0,
|
| 504 |
+
"std": 0.12573782247199924,
|
| 505 |
+
"step": 275
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 5.0,
|
| 509 |
+
"eval_loss": 2.821211099624634,
|
| 510 |
+
"eval_runtime": 4.9351,
|
| 511 |
+
"eval_samples_per_second": 4.661,
|
| 512 |
+
"eval_steps_per_second": 1.216,
|
| 513 |
+
"step": 275
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 5.02,
|
| 517 |
+
"learning_rate": 0.0004661085957644817,
|
| 518 |
+
"loss": 0.233,
|
| 519 |
+
"step": 276
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.09,
|
| 523 |
+
"learning_rate": 0.0004644958533087443,
|
| 524 |
+
"loss": 0.1537,
|
| 525 |
+
"step": 280
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 5.16,
|
| 529 |
+
"learning_rate": 0.0004628485517542392,
|
| 530 |
+
"loss": 0.1743,
|
| 531 |
+
"step": 284
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 5.24,
|
| 535 |
+
"learning_rate": 0.0004611669565105596,
|
| 536 |
+
"loss": 0.1827,
|
| 537 |
+
"step": 288
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 5.31,
|
| 541 |
+
"learning_rate": 0.00045945133851262184,
|
| 542 |
+
"loss": 0.1928,
|
| 543 |
+
"step": 292
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 5.38,
|
| 547 |
+
"learning_rate": 0.00045770197417701366,
|
| 548 |
+
"loss": 0.1718,
|
| 549 |
+
"step": 296
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 5.45,
|
| 553 |
+
"learning_rate": 0.0004559191453574582,
|
| 554 |
+
"loss": 0.166,
|
| 555 |
+
"step": 300
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 5.53,
|
| 559 |
+
"learning_rate": 0.00045410313929940244,
|
| 560 |
+
"loss": 0.1484,
|
| 561 |
+
"step": 304
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 5.6,
|
| 565 |
+
"learning_rate": 0.0004522542485937369,
|
| 566 |
+
"loss": 0.1521,
|
| 567 |
+
"step": 308
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 5.67,
|
| 571 |
+
"learning_rate": 0.00045037277112965383,
|
| 572 |
+
"loss": 0.195,
|
| 573 |
+
"step": 312
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 5.75,
|
| 577 |
+
"learning_rate": 0.0004484590100466523,
|
| 578 |
+
"loss": 0.153,
|
| 579 |
+
"step": 316
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 5.82,
|
| 583 |
+
"learning_rate": 0.0004465132736856969,
|
| 584 |
+
"loss": 0.1751,
|
| 585 |
+
"step": 320
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 5.89,
|
| 589 |
+
"learning_rate": 0.0004445358755395382,
|
| 590 |
+
"loss": 0.2153,
|
| 591 |
+
"step": 324
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 5.96,
|
| 595 |
+
"learning_rate": 0.00044252713420220394,
|
| 596 |
+
"loss": 0.1983,
|
| 597 |
+
"step": 328
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 6.0,
|
| 601 |
+
"gpt4_scores": 0.7500000000000001,
|
| 602 |
+
"step": 330
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 6.0,
|
| 606 |
+
"std": 0.0972111104761179,
|
| 607 |
+
"step": 330
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 6.0,
|
| 611 |
+
"eval_loss": 2.8983585834503174,
|
| 612 |
+
"eval_runtime": 4.9684,
|
| 613 |
+
"eval_samples_per_second": 4.629,
|
| 614 |
+
"eval_steps_per_second": 1.208,
|
| 615 |
+
"step": 330
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 6.04,
|
| 619 |
+
"learning_rate": 0.0004404873733176677,
|
| 620 |
+
"loss": 0.1353,
|
| 621 |
+
"step": 332
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 6.11,
|
| 625 |
+
"learning_rate": 0.00043841692152770415,
|
| 626 |
+
"loss": 0.102,
|
| 627 |
+
"step": 336
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 6.18,
|
| 631 |
+
"learning_rate": 0.0004363161124189387,
|
| 632 |
+
"loss": 0.1055,
|
| 633 |
+
"step": 340
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 6.25,
|
| 637 |
+
"learning_rate": 0.00043418528446910123,
|
| 638 |
+
"loss": 0.0998,
|
| 639 |
+
"step": 344
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 6.33,
|
| 643 |
+
"learning_rate": 0.00043202478099249104,
|
| 644 |
+
"loss": 0.1168,
|
| 645 |
+
"step": 348
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 6.4,
|
| 649 |
+
"learning_rate": 0.0004298349500846628,
|
| 650 |
+
"loss": 0.1156,
|
| 651 |
+
"step": 352
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 6.47,
|
| 655 |
+
"learning_rate": 0.00042761614456634226,
|
| 656 |
+
"loss": 0.1345,
|
| 657 |
+
"step": 356
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 6.55,
|
| 661 |
+
"learning_rate": 0.00042536872192658034,
|
| 662 |
+
"loss": 0.0968,
|
| 663 |
+
"step": 360
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 6.62,
|
| 667 |
+
"learning_rate": 0.0004230930442651557,
|
| 668 |
+
"loss": 0.1179,
|
| 669 |
+
"step": 364
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 6.69,
|
| 673 |
+
"learning_rate": 0.00042078947823423365,
|
| 674 |
+
"loss": 0.0959,
|
| 675 |
+
"step": 368
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 6.76,
|
| 679 |
+
"learning_rate": 0.00041845839497929203,
|
| 680 |
+
"loss": 0.1078,
|
| 681 |
+
"step": 372
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 6.84,
|
| 685 |
+
"learning_rate": 0.0004161001700793231,
|
| 686 |
+
"loss": 0.1175,
|
| 687 |
+
"step": 376
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 6.91,
|
| 691 |
+
"learning_rate": 0.0004137151834863213,
|
| 692 |
+
"loss": 0.1095,
|
| 693 |
+
"step": 380
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 6.98,
|
| 697 |
+
"learning_rate": 0.00041130381946406574,
|
| 698 |
+
"loss": 0.1199,
|
| 699 |
+
"step": 384
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 7.0,
|
| 703 |
+
"gpt4_scores": 0.49000000000000005,
|
| 704 |
+
"step": 385
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 7.0,
|
| 708 |
+
"std": 0.12525973016097391,
|
| 709 |
+
"step": 385
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 7.0,
|
| 713 |
+
"eval_loss": 2.949794292449951,
|
| 714 |
+
"eval_runtime": 4.9646,
|
| 715 |
+
"eval_samples_per_second": 4.633,
|
| 716 |
+
"eval_steps_per_second": 1.209,
|
| 717 |
+
"step": 385
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 7.05,
|
| 721 |
+
"learning_rate": 0.0004088664665262091,
|
| 722 |
+
"loss": 0.076,
|
| 723 |
+
"step": 388
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 7.13,
|
| 727 |
+
"learning_rate": 0.0004064035173736804,
|
| 728 |
+
"loss": 0.0709,
|
| 729 |
+
"step": 392
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 7.2,
|
| 733 |
+
"learning_rate": 0.00040391536883141455,
|
| 734 |
+
"loss": 0.073,
|
| 735 |
+
"step": 396
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 7.27,
|
| 739 |
+
"learning_rate": 0.00040140242178441667,
|
| 740 |
+
"loss": 0.0819,
|
| 741 |
+
"step": 400
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 7.35,
|
| 745 |
+
"learning_rate": 0.000398865081113172,
|
| 746 |
+
"loss": 0.0643,
|
| 747 |
+
"step": 404
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 7.42,
|
| 751 |
+
"learning_rate": 0.0003963037556284129,
|
| 752 |
+
"loss": 0.1603,
|
| 753 |
+
"step": 408
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 7.49,
|
| 757 |
+
"learning_rate": 0.0003937188580052518,
|
| 758 |
+
"loss": 0.119,
|
| 759 |
+
"step": 412
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 7.56,
|
| 763 |
+
"learning_rate": 0.0003911108047166924,
|
| 764 |
+
"loss": 0.0754,
|
| 765 |
+
"step": 416
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 7.64,
|
| 769 |
+
"learning_rate": 0.0003884800159665276,
|
| 770 |
+
"loss": 0.0622,
|
| 771 |
+
"step": 420
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 7.71,
|
| 775 |
+
"learning_rate": 0.00038582691562163827,
|
| 776 |
+
"loss": 0.0711,
|
| 777 |
+
"step": 424
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 7.78,
|
| 781 |
+
"learning_rate": 0.00038315193114369994,
|
| 782 |
+
"loss": 0.0643,
|
| 783 |
+
"step": 428
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 7.85,
|
| 787 |
+
"learning_rate": 0.0003804554935203115,
|
| 788 |
+
"loss": 0.0672,
|
| 789 |
+
"step": 432
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 7.93,
|
| 793 |
+
"learning_rate": 0.00037773803719555514,
|
| 794 |
+
"loss": 0.0688,
|
| 795 |
+
"step": 436
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 8.0,
|
| 799 |
+
"learning_rate": 0.000375,
|
| 800 |
+
"loss": 0.0743,
|
| 801 |
+
"step": 440
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 8.0,
|
| 805 |
+
"gpt4_scores": 0.51,
|
| 806 |
+
"step": 440
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 8.0,
|
| 810 |
+
"std": 0.12763228431709586,
|
| 811 |
+
"step": 440
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 8.0,
|
| 815 |
+
"eval_loss": 3.250723123550415,
|
| 816 |
+
"eval_runtime": 4.9396,
|
| 817 |
+
"eval_samples_per_second": 4.656,
|
| 818 |
+
"eval_steps_per_second": 1.215,
|
| 819 |
+
"step": 440
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 8.07,
|
| 823 |
+
"learning_rate": 0.00037224182308015974,
|
| 824 |
+
"loss": 0.0528,
|
| 825 |
+
"step": 444
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 8.15,
|
| 829 |
+
"learning_rate": 0.0003694639508274158,
|
| 830 |
+
"loss": 0.0502,
|
| 831 |
+
"step": 448
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 8.22,
|
| 835 |
+
"learning_rate": 0.00036666683080641843,
|
| 836 |
+
"loss": 0.0659,
|
| 837 |
+
"step": 452
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 8.29,
|
| 841 |
+
"learning_rate": 0.0003638509136829758,
|
| 842 |
+
"loss": 0.065,
|
| 843 |
+
"step": 456
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 8.36,
|
| 847 |
+
"learning_rate": 0.00036101665315144355,
|
| 848 |
+
"loss": 0.0524,
|
| 849 |
+
"step": 460
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 8.44,
|
| 853 |
+
"learning_rate": 0.00035816450586162706,
|
| 854 |
+
"loss": 0.0565,
|
| 855 |
+
"step": 464
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 8.51,
|
| 859 |
+
"learning_rate": 0.00035529493134520666,
|
| 860 |
+
"loss": 0.0432,
|
| 861 |
+
"step": 468
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 8.58,
|
| 865 |
+
"learning_rate": 0.00035240839194169884,
|
| 866 |
+
"loss": 0.0582,
|
| 867 |
+
"step": 472
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 8.65,
|
| 871 |
+
"learning_rate": 0.0003495053527239656,
|
| 872 |
+
"loss": 0.0507,
|
| 873 |
+
"step": 476
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 8.73,
|
| 877 |
+
"learning_rate": 0.00034658628142328216,
|
| 878 |
+
"loss": 0.0571,
|
| 879 |
+
"step": 480
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 8.8,
|
| 883 |
+
"learning_rate": 0.00034365164835397803,
|
| 884 |
+
"loss": 0.0925,
|
| 885 |
+
"step": 484
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 8.87,
|
| 889 |
+
"learning_rate": 0.00034070192633766023,
|
| 890 |
+
"loss": 0.0601,
|
| 891 |
+
"step": 488
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 8.95,
|
| 895 |
+
"learning_rate": 0.00033773759062703394,
|
| 896 |
+
"loss": 0.062,
|
| 897 |
+
"step": 492
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 9.0,
|
| 901 |
+
"gpt4_scores": 0.53,
|
| 902 |
+
"step": 495
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 9.0,
|
| 906 |
+
"std": 0.1192057045614848,
|
| 907 |
+
"step": 495
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 9.0,
|
| 911 |
+
"eval_loss": 3.1474077701568604,
|
| 912 |
+
"eval_runtime": 4.9704,
|
| 913 |
+
"eval_samples_per_second": 4.627,
|
| 914 |
+
"eval_steps_per_second": 1.207,
|
| 915 |
+
"step": 495
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 9.02,
|
| 919 |
+
"learning_rate": 0.0003347591188293301,
|
| 920 |
+
"loss": 0.0535,
|
| 921 |
+
"step": 496
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 9.09,
|
| 925 |
+
"learning_rate": 0.00033176699082935546,
|
| 926 |
+
"loss": 0.0484,
|
| 927 |
+
"step": 500
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 9.16,
|
| 931 |
+
"learning_rate": 0.00032876168871217323,
|
| 932 |
+
"loss": 0.0541,
|
| 933 |
+
"step": 504
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 9.24,
|
| 937 |
+
"learning_rate": 0.00032574369668543187,
|
| 938 |
+
"loss": 0.0438,
|
| 939 |
+
"step": 508
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 9.31,
|
| 943 |
+
"learning_rate": 0.00032271350100134975,
|
| 944 |
+
"loss": 0.0451,
|
| 945 |
+
"step": 512
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 9.38,
|
| 949 |
+
"learning_rate": 0.00031967158987837195,
|
| 950 |
+
"loss": 0.0403,
|
| 951 |
+
"step": 516
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 9.45,
|
| 955 |
+
"learning_rate": 0.0003166184534225087,
|
| 956 |
+
"loss": 0.0517,
|
| 957 |
+
"step": 520
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 9.53,
|
| 961 |
+
"learning_rate": 0.0003135545835483718,
|
| 962 |
+
"loss": 0.0498,
|
| 963 |
+
"step": 524
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 9.6,
|
| 967 |
+
"learning_rate": 0.0003104804738999169,
|
| 968 |
+
"loss": 0.0516,
|
| 969 |
+
"step": 528
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 9.67,
|
| 973 |
+
"learning_rate": 0.00030739661977091025,
|
| 974 |
+
"loss": 0.0487,
|
| 975 |
+
"step": 532
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 9.75,
|
| 979 |
+
"learning_rate": 0.00030430351802512693,
|
| 980 |
+
"loss": 0.045,
|
| 981 |
+
"step": 536
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 9.82,
|
| 985 |
+
"learning_rate": 0.0003012016670162977,
|
| 986 |
+
"loss": 0.0397,
|
| 987 |
+
"step": 540
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 9.89,
|
| 991 |
+
"learning_rate": 0.00029809156650781527,
|
| 992 |
+
"loss": 0.0484,
|
| 993 |
+
"step": 544
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 9.96,
|
| 997 |
+
"learning_rate": 0.0002949737175922135,
|
| 998 |
+
"loss": 0.0404,
|
| 999 |
+
"step": 548
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 10.0,
|
| 1003 |
+
"gpt4_scores": 0.6749999999999999,
|
| 1004 |
+
"step": 550
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 10.0,
|
| 1008 |
+
"std": 0.11602801385872293,
|
| 1009 |
+
"step": 550
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 10.0,
|
| 1013 |
+
"eval_loss": 3.3666398525238037,
|
| 1014 |
+
"eval_runtime": 4.9475,
|
| 1015 |
+
"eval_samples_per_second": 4.649,
|
| 1016 |
+
"eval_steps_per_second": 1.213,
|
| 1017 |
+
"step": 550
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 10.04,
|
| 1021 |
+
"learning_rate": 0.0002918486226104327,
|
| 1022 |
+
"loss": 0.0403,
|
| 1023 |
+
"step": 552
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 10.11,
|
| 1027 |
+
"learning_rate": 0.0002887167850708831,
|
| 1028 |
+
"loss": 0.0401,
|
| 1029 |
+
"step": 556
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 10.18,
|
| 1033 |
+
"learning_rate": 0.00028557870956832135,
|
| 1034 |
+
"loss": 0.0442,
|
| 1035 |
+
"step": 560
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 10.25,
|
| 1039 |
+
"learning_rate": 0.00028243490170255044,
|
| 1040 |
+
"loss": 0.041,
|
| 1041 |
+
"step": 564
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 10.33,
|
| 1045 |
+
"learning_rate": 0.0002792858679969596,
|
| 1046 |
+
"loss": 0.036,
|
| 1047 |
+
"step": 568
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 10.4,
|
| 1051 |
+
"learning_rate": 0.0002761321158169134,
|
| 1052 |
+
"loss": 0.0533,
|
| 1053 |
+
"step": 572
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 10.47,
|
| 1057 |
+
"learning_rate": 0.0002729741532880069,
|
| 1058 |
+
"loss": 0.0506,
|
| 1059 |
+
"step": 576
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 10.55,
|
| 1063 |
+
"learning_rate": 0.0002698124892141971,
|
| 1064 |
+
"loss": 0.0532,
|
| 1065 |
+
"step": 580
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 10.62,
|
| 1069 |
+
"learning_rate": 0.000266647632995826,
|
| 1070 |
+
"loss": 0.0468,
|
| 1071 |
+
"step": 584
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 10.69,
|
| 1075 |
+
"learning_rate": 0.0002634800945475465,
|
| 1076 |
+
"loss": 0.0396,
|
| 1077 |
+
"step": 588
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 10.76,
|
| 1081 |
+
"learning_rate": 0.00026031038421616684,
|
| 1082 |
+
"loss": 0.0383,
|
| 1083 |
+
"step": 592
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 10.84,
|
| 1087 |
+
"learning_rate": 0.00025713901269842405,
|
| 1088 |
+
"loss": 0.0475,
|
| 1089 |
+
"step": 596
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 10.91,
|
| 1093 |
+
"learning_rate": 0.000253966490958702,
|
| 1094 |
+
"loss": 0.0491,
|
| 1095 |
+
"step": 600
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 10.98,
|
| 1099 |
+
"learning_rate": 0.00025079333014670557,
|
| 1100 |
+
"loss": 0.038,
|
| 1101 |
+
"step": 604
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 11.0,
|
| 1105 |
+
"gpt4_scores": 0.63,
|
| 1106 |
+
"step": 605
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 11.0,
|
| 1110 |
+
"std": 0.11580155439371269,
|
| 1111 |
+
"step": 605
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 11.0,
|
| 1115 |
+
"eval_loss": 3.3349289894104004,
|
| 1116 |
+
"eval_runtime": 4.9384,
|
| 1117 |
+
"eval_samples_per_second": 4.657,
|
| 1118 |
+
"eval_steps_per_second": 1.215,
|
| 1119 |
+
"step": 605
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 11.05,
|
| 1123 |
+
"learning_rate": 0.00024762004151510585,
|
| 1124 |
+
"loss": 0.0482,
|
| 1125 |
+
"step": 608
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 11.13,
|
| 1129 |
+
"learning_rate": 0.00024444713633716764,
|
| 1130 |
+
"loss": 0.0356,
|
| 1131 |
+
"step": 612
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 11.2,
|
| 1135 |
+
"learning_rate": 0.00024127512582437484,
|
| 1136 |
+
"loss": 0.0444,
|
| 1137 |
+
"step": 616
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 11.27,
|
| 1141 |
+
"learning_rate": 0.00023810452104406444,
|
| 1142 |
+
"loss": 0.0411,
|
| 1143 |
+
"step": 620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 11.35,
|
| 1147 |
+
"learning_rate": 0.00023493583283708543,
|
| 1148 |
+
"loss": 0.0441,
|
| 1149 |
+
"step": 624
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 11.42,
|
| 1153 |
+
"learning_rate": 0.00023176957173549233,
|
| 1154 |
+
"loss": 0.0393,
|
| 1155 |
+
"step": 628
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 11.49,
|
| 1159 |
+
"learning_rate": 0.00022860624788029015,
|
| 1160 |
+
"loss": 0.0363,
|
| 1161 |
+
"step": 632
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 11.56,
|
| 1165 |
+
"learning_rate": 0.00022544637093924072,
|
| 1166 |
+
"loss": 0.0393,
|
| 1167 |
+
"step": 636
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 11.64,
|
| 1171 |
+
"learning_rate": 0.00022229045002474727,
|
| 1172 |
+
"loss": 0.0452,
|
| 1173 |
+
"step": 640
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 11.71,
|
| 1177 |
+
"learning_rate": 0.00021913899361182632,
|
| 1178 |
+
"loss": 0.0417,
|
| 1179 |
+
"step": 644
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 11.78,
|
| 1183 |
+
"learning_rate": 0.000215992509456184,
|
| 1184 |
+
"loss": 0.0366,
|
| 1185 |
+
"step": 648
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 11.85,
|
| 1189 |
+
"learning_rate": 0.00021285150451240712,
|
| 1190 |
+
"loss": 0.0443,
|
| 1191 |
+
"step": 652
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 11.93,
|
| 1195 |
+
"learning_rate": 0.000209716484852284,
|
| 1196 |
+
"loss": 0.0447,
|
| 1197 |
+
"step": 656
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 12.0,
|
| 1201 |
+
"learning_rate": 0.00020658795558326743,
|
| 1202 |
+
"loss": 0.0374,
|
| 1203 |
+
"step": 660
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 12.0,
|
| 1207 |
+
"gpt4_scores": 0.6100000000000001,
|
| 1208 |
+
"step": 660
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 12.0,
|
| 1212 |
+
"std": 0.12841339493993606,
|
| 1213 |
+
"step": 660
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 12.0,
|
| 1217 |
+
"eval_loss": 3.4455862045288086,
|
| 1218 |
+
"eval_runtime": 4.9572,
|
| 1219 |
+
"eval_samples_per_second": 4.64,
|
| 1220 |
+
"eval_steps_per_second": 1.21,
|
| 1221 |
+
"step": 660
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 12.07,
|
| 1225 |
+
"learning_rate": 0.0002034664207670925,
|
| 1226 |
+
"loss": 0.0442,
|
| 1227 |
+
"step": 664
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 12.15,
|
| 1231 |
+
"learning_rate": 0.00020035238333856371,
|
| 1232 |
+
"loss": 0.0391,
|
| 1233 |
+
"step": 668
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 12.22,
|
| 1237 |
+
"learning_rate": 0.0001972463450245226,
|
| 1238 |
+
"loss": 0.0392,
|
| 1239 |
+
"step": 672
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 12.29,
|
| 1243 |
+
"learning_rate": 0.00019414880626301146,
|
| 1244 |
+
"loss": 0.0426,
|
| 1245 |
+
"step": 676
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 12.36,
|
| 1249 |
+
"learning_rate": 0.00019106026612264316,
|
| 1250 |
+
"loss": 0.043,
|
| 1251 |
+
"step": 680
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 12.44,
|
| 1255 |
+
"learning_rate": 0.0001879812222221929,
|
| 1256 |
+
"loss": 0.0395,
|
| 1257 |
+
"step": 684
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 12.51,
|
| 1261 |
+
"learning_rate": 0.00018491217065042198,
|
| 1262 |
+
"loss": 0.0396,
|
| 1263 |
+
"step": 688
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 12.58,
|
| 1267 |
+
"learning_rate": 0.00018185360588615057,
|
| 1268 |
+
"loss": 0.0394,
|
| 1269 |
+
"step": 692
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 12.65,
|
| 1273 |
+
"learning_rate": 0.00017880602071858692,
|
| 1274 |
+
"loss": 0.0403,
|
| 1275 |
+
"step": 696
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 12.73,
|
| 1279 |
+
"learning_rate": 0.00017576990616793137,
|
| 1280 |
+
"loss": 0.0349,
|
| 1281 |
+
"step": 700
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 12.8,
|
| 1285 |
+
"learning_rate": 0.00017274575140626317,
|
| 1286 |
+
"loss": 0.0404,
|
| 1287 |
+
"step": 704
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 12.87,
|
| 1291 |
+
"learning_rate": 0.0001697340436787273,
|
| 1292 |
+
"loss": 0.0344,
|
| 1293 |
+
"step": 708
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 12.95,
|
| 1297 |
+
"learning_rate": 0.00016673526822502983,
|
| 1298 |
+
"loss": 0.0384,
|
| 1299 |
+
"step": 712
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 13.0,
|
| 1303 |
+
"gpt4_scores": 0.53,
|
| 1304 |
+
"step": 715
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 13.0,
|
| 1308 |
+
"std": 0.13787675656179324,
|
| 1309 |
+
"step": 715
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 13.0,
|
| 1313 |
+
"eval_loss": 3.4822309017181396,
|
| 1314 |
+
"eval_runtime": 4.946,
|
| 1315 |
+
"eval_samples_per_second": 4.65,
|
| 1316 |
+
"eval_steps_per_second": 1.213,
|
| 1317 |
+
"step": 715
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 13.02,
|
| 1321 |
+
"learning_rate": 0.0001637499082012574,
|
| 1322 |
+
"loss": 0.0522,
|
| 1323 |
+
"step": 716
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 13.09,
|
| 1327 |
+
"learning_rate": 0.00016077844460203207,
|
| 1328 |
+
"loss": 0.0384,
|
| 1329 |
+
"step": 720
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 13.16,
|
| 1333 |
+
"learning_rate": 0.00015782135618301485,
|
| 1334 |
+
"loss": 0.0336,
|
| 1335 |
+
"step": 724
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 13.24,
|
| 1339 |
+
"learning_rate": 0.00015487911938376925,
|
| 1340 |
+
"loss": 0.0401,
|
| 1341 |
+
"step": 728
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 13.31,
|
| 1345 |
+
"learning_rate": 0.00015195220825099862,
|
| 1346 |
+
"loss": 0.0395,
|
| 1347 |
+
"step": 732
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 13.38,
|
| 1351 |
+
"learning_rate": 0.00014904109436216883,
|
| 1352 |
+
"loss": 0.0414,
|
| 1353 |
+
"step": 736
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 13.45,
|
| 1357 |
+
"learning_rate": 0.0001461462467495284,
|
| 1358 |
+
"loss": 0.0393,
|
| 1359 |
+
"step": 740
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 13.53,
|
| 1363 |
+
"learning_rate": 0.00014326813182453956,
|
| 1364 |
+
"loss": 0.0383,
|
| 1365 |
+
"step": 744
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 13.6,
|
| 1369 |
+
"learning_rate": 0.00014040721330273062,
|
| 1370 |
+
"loss": 0.038,
|
| 1371 |
+
"step": 748
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 13.67,
|
| 1375 |
+
"learning_rate": 0.0001375639521289836,
|
| 1376 |
+
"loss": 0.0383,
|
| 1377 |
+
"step": 752
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 13.75,
|
| 1381 |
+
"learning_rate": 0.00013473880640326724,
|
| 1382 |
+
"loss": 0.0405,
|
| 1383 |
+
"step": 756
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 13.82,
|
| 1387 |
+
"learning_rate": 0.00013193223130682935,
|
| 1388 |
+
"loss": 0.0422,
|
| 1389 |
+
"step": 760
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 13.89,
|
| 1393 |
+
"learning_rate": 0.000129144679028859,
|
| 1394 |
+
"loss": 0.0408,
|
| 1395 |
+
"step": 764
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 13.96,
|
| 1399 |
+
"learning_rate": 0.00012637659869363084,
|
| 1400 |
+
"loss": 0.0408,
|
| 1401 |
+
"step": 768
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 14.0,
|
| 1405 |
+
"gpt4_scores": 0.6500000000000001,
|
| 1406 |
+
"step": 770
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 14.0,
|
| 1410 |
+
"std": 0.11423659658795862,
|
| 1411 |
+
"step": 770
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 14.0,
|
| 1415 |
+
"eval_loss": 3.471827983856201,
|
| 1416 |
+
"eval_runtime": 4.9416,
|
| 1417 |
+
"eval_samples_per_second": 4.654,
|
| 1418 |
+
"eval_steps_per_second": 1.214,
|
| 1419 |
+
"step": 770
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 14.04,
|
| 1423 |
+
"learning_rate": 0.00012362843628814266,
|
| 1424 |
+
"loss": 0.0343,
|
| 1425 |
+
"step": 772
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 14.11,
|
| 1429 |
+
"learning_rate": 0.00012090063459025954,
|
| 1430 |
+
"loss": 0.0386,
|
| 1431 |
+
"step": 776
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 14.18,
|
| 1435 |
+
"learning_rate": 0.00011819363309737438,
|
| 1436 |
+
"loss": 0.0412,
|
| 1437 |
+
"step": 780
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 14.25,
|
| 1441 |
+
"learning_rate": 0.0001155078679555969,
|
| 1442 |
+
"loss": 0.0402,
|
| 1443 |
+
"step": 784
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 14.33,
|
| 1447 |
+
"learning_rate": 0.00011284377188948258,
|
| 1448 |
+
"loss": 0.038,
|
| 1449 |
+
"step": 788
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 14.4,
|
| 1453 |
+
"learning_rate": 0.00011020177413231333,
|
| 1454 |
+
"loss": 0.0367,
|
| 1455 |
+
"step": 792
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 14.47,
|
| 1459 |
+
"learning_rate": 0.0001075823003569403,
|
| 1460 |
+
"loss": 0.0449,
|
| 1461 |
+
"step": 796
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 14.55,
|
| 1465 |
+
"learning_rate": 0.00010498577260720049,
|
| 1466 |
+
"loss": 0.036,
|
| 1467 |
+
"step": 800
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 14.62,
|
| 1471 |
+
"learning_rate": 0.00010241260922991761,
|
| 1472 |
+
"loss": 0.0385,
|
| 1473 |
+
"step": 804
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 14.69,
|
| 1477 |
+
"learning_rate": 9.986322480749927e-05,
|
| 1478 |
+
"loss": 0.0428,
|
| 1479 |
+
"step": 808
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 14.76,
|
| 1483 |
+
"learning_rate": 9.733803009114044e-05,
|
| 1484 |
+
"loss": 0.0375,
|
| 1485 |
+
"step": 812
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 14.84,
|
| 1489 |
+
"learning_rate": 9.483743193464408e-05,
|
| 1490 |
+
"loss": 0.0435,
|
| 1491 |
+
"step": 816
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 14.91,
|
| 1495 |
+
"learning_rate": 9.236183322886945e-05,
|
| 1496 |
+
"loss": 0.0394,
|
| 1497 |
+
"step": 820
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 14.98,
|
| 1501 |
+
"learning_rate": 8.991163283681945e-05,
|
| 1502 |
+
"loss": 0.0347,
|
| 1503 |
+
"step": 824
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 15.0,
|
| 1507 |
+
"gpt4_scores": 0.67,
|
| 1508 |
+
"step": 825
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 15.0,
|
| 1512 |
+
"std": 0.11644741302407709,
|
| 1513 |
+
"step": 825
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 15.0,
|
| 1517 |
+
"eval_loss": 3.502847671508789,
|
| 1518 |
+
"eval_runtime": 4.9425,
|
| 1519 |
+
"eval_samples_per_second": 4.654,
|
| 1520 |
+
"eval_steps_per_second": 1.214,
|
| 1521 |
+
"step": 825
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 15.05,
|
| 1525 |
+
"learning_rate": 8.748722552937688e-05,
|
| 1526 |
+
"loss": 0.0395,
|
| 1527 |
+
"step": 828
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 15.13,
|
| 1531 |
+
"learning_rate": 8.508900192169963e-05,
|
| 1532 |
+
"loss": 0.0389,
|
| 1533 |
+
"step": 832
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 15.2,
|
| 1537 |
+
"learning_rate": 8.271734841028553e-05,
|
| 1538 |
+
"loss": 0.0425,
|
| 1539 |
+
"step": 836
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 15.27,
|
| 1543 |
+
"learning_rate": 8.037264711071699e-05,
|
| 1544 |
+
"loss": 0.0403,
|
| 1545 |
+
"step": 840
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 15.35,
|
| 1549 |
+
"learning_rate": 7.805527579609576e-05,
|
| 1550 |
+
"loss": 0.0405,
|
| 1551 |
+
"step": 844
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 15.42,
|
| 1555 |
+
"learning_rate": 7.576560783617667e-05,
|
| 1556 |
+
"loss": 0.0392,
|
| 1557 |
+
"step": 848
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 15.49,
|
| 1561 |
+
"learning_rate": 7.35040121372109e-05,
|
| 1562 |
+
"loss": 0.0383,
|
| 1563 |
+
"step": 852
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 15.56,
|
| 1567 |
+
"learning_rate": 7.127085308250913e-05,
|
| 1568 |
+
"loss": 0.0404,
|
| 1569 |
+
"step": 856
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 15.64,
|
| 1573 |
+
"learning_rate": 6.906649047373245e-05,
|
| 1574 |
+
"loss": 0.0347,
|
| 1575 |
+
"step": 860
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 15.71,
|
| 1579 |
+
"learning_rate": 6.689127947292231e-05,
|
| 1580 |
+
"loss": 0.0363,
|
| 1581 |
+
"step": 864
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 15.78,
|
| 1585 |
+
"learning_rate": 6.474557054527707e-05,
|
| 1586 |
+
"loss": 0.0341,
|
| 1587 |
+
"step": 868
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 15.85,
|
| 1591 |
+
"learning_rate": 6.262970940268654e-05,
|
| 1592 |
+
"loss": 0.0377,
|
| 1593 |
+
"step": 872
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 15.93,
|
| 1597 |
+
"learning_rate": 6.054403694803079e-05,
|
| 1598 |
+
"loss": 0.0442,
|
| 1599 |
+
"step": 876
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 16.0,
|
| 1603 |
+
"learning_rate": 5.848888922025553e-05,
|
| 1604 |
+
"loss": 0.0377,
|
| 1605 |
+
"step": 880
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 16.0,
|
| 1609 |
+
"gpt4_scores": 0.5700000000000001,
|
| 1610 |
+
"step": 880
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 16.0,
|
| 1614 |
+
"std": 0.1319469590403659,
|
| 1615 |
+
"step": 880
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 16.0,
|
| 1619 |
+
"eval_loss": 3.5217506885528564,
|
| 1620 |
+
"eval_runtime": 4.9355,
|
| 1621 |
+
"eval_samples_per_second": 4.66,
|
| 1622 |
+
"eval_steps_per_second": 1.216,
|
| 1623 |
+
"step": 880
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 16.07,
|
| 1627 |
+
"learning_rate": 5.646459734022938e-05,
|
| 1628 |
+
"loss": 0.041,
|
| 1629 |
+
"step": 884
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 16.15,
|
| 1633 |
+
"learning_rate": 5.4471487457395216e-05,
|
| 1634 |
+
"loss": 0.0333,
|
| 1635 |
+
"step": 888
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 16.22,
|
| 1639 |
+
"learning_rate": 5.2509880697220956e-05,
|
| 1640 |
+
"loss": 0.0407,
|
| 1641 |
+
"step": 892
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 16.29,
|
| 1645 |
+
"learning_rate": 5.058009310946118e-05,
|
| 1646 |
+
"loss": 0.0391,
|
| 1647 |
+
"step": 896
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 16.36,
|
| 1651 |
+
"learning_rate": 4.8682435617235344e-05,
|
| 1652 |
+
"loss": 0.0365,
|
| 1653 |
+
"step": 900
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 16.44,
|
| 1657 |
+
"learning_rate": 4.6817213966933034e-05,
|
| 1658 |
+
"loss": 0.0425,
|
| 1659 |
+
"step": 904
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 16.51,
|
| 1663 |
+
"learning_rate": 4.498472867895223e-05,
|
| 1664 |
+
"loss": 0.0363,
|
| 1665 |
+
"step": 908
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 16.58,
|
| 1669 |
+
"learning_rate": 4.318527499928074e-05,
|
| 1670 |
+
"loss": 0.0391,
|
| 1671 |
+
"step": 912
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 16.65,
|
| 1675 |
+
"learning_rate": 4.141914285192619e-05,
|
| 1676 |
+
"loss": 0.0375,
|
| 1677 |
+
"step": 916
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 16.73,
|
| 1681 |
+
"learning_rate": 3.968661679220467e-05,
|
| 1682 |
+
"loss": 0.0388,
|
| 1683 |
+
"step": 920
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 16.8,
|
| 1687 |
+
"learning_rate": 3.798797596089351e-05,
|
| 1688 |
+
"loss": 0.0417,
|
| 1689 |
+
"step": 924
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 16.87,
|
| 1693 |
+
"learning_rate": 3.632349403925664e-05,
|
| 1694 |
+
"loss": 0.0378,
|
| 1695 |
+
"step": 928
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 16.95,
|
| 1699 |
+
"learning_rate": 3.4693439204949856e-05,
|
| 1700 |
+
"loss": 0.0395,
|
| 1701 |
+
"step": 932
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"epoch": 17.0,
|
| 1705 |
+
"gpt4_scores": 0.595,
|
| 1706 |
+
"step": 935
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 17.0,
|
| 1710 |
+
"std": 0.1132364782214636,
|
| 1711 |
+
"step": 935
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 17.0,
|
| 1715 |
+
"eval_loss": 3.5319790840148926,
|
| 1716 |
+
"eval_runtime": 4.9408,
|
| 1717 |
+
"eval_samples_per_second": 4.655,
|
| 1718 |
+
"eval_steps_per_second": 1.214,
|
| 1719 |
+
"step": 935
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 17.02,
|
| 1723 |
+
"learning_rate": 3.309807408881269e-05,
|
| 1724 |
+
"loss": 0.0416,
|
| 1725 |
+
"step": 936
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 17.09,
|
| 1729 |
+
"learning_rate": 3.1537655732553766e-05,
|
| 1730 |
+
"loss": 0.0409,
|
| 1731 |
+
"step": 940
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 17.16,
|
| 1735 |
+
"learning_rate": 3.0012435547336736e-05,
|
| 1736 |
+
"loss": 0.0412,
|
| 1737 |
+
"step": 944
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 17.24,
|
| 1741 |
+
"learning_rate": 2.8522659273273603e-05,
|
| 1742 |
+
"loss": 0.0411,
|
| 1743 |
+
"step": 948
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 17.31,
|
| 1747 |
+
"learning_rate": 2.7068566939831645e-05,
|
| 1748 |
+
"loss": 0.0332,
|
| 1749 |
+
"step": 952
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 17.38,
|
| 1753 |
+
"learning_rate": 2.5650392827160445e-05,
|
| 1754 |
+
"loss": 0.0315,
|
| 1755 |
+
"step": 956
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 17.45,
|
| 1759 |
+
"learning_rate": 2.4268365428344735e-05,
|
| 1760 |
+
"loss": 0.0448,
|
| 1761 |
+
"step": 960
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 17.53,
|
| 1765 |
+
"learning_rate": 2.29227074125907e-05,
|
| 1766 |
+
"loss": 0.0349,
|
| 1767 |
+
"step": 964
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 17.6,
|
| 1771 |
+
"learning_rate": 2.1613635589349755e-05,
|
| 1772 |
+
"loss": 0.0367,
|
| 1773 |
+
"step": 968
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 17.67,
|
| 1777 |
+
"learning_rate": 2.0341360873386672e-05,
|
| 1778 |
+
"loss": 0.0386,
|
| 1779 |
+
"step": 972
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 17.75,
|
| 1783 |
+
"learning_rate": 1.9106088250797264e-05,
|
| 1784 |
+
"loss": 0.0371,
|
| 1785 |
+
"step": 976
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 17.82,
|
| 1789 |
+
"learning_rate": 1.7908016745981858e-05,
|
| 1790 |
+
"loss": 0.0369,
|
| 1791 |
+
"step": 980
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 17.89,
|
| 1795 |
+
"learning_rate": 1.674733938957873e-05,
|
| 1796 |
+
"loss": 0.0464,
|
| 1797 |
+
"step": 984
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 17.96,
|
| 1801 |
+
"learning_rate": 1.562424318736344e-05,
|
| 1802 |
+
"loss": 0.0408,
|
| 1803 |
+
"step": 988
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 18.0,
|
| 1807 |
+
"gpt4_scores": 0.59,
|
| 1808 |
+
"step": 990
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 18.0,
|
| 1812 |
+
"std": 0.13224976370489286,
|
| 1813 |
+
"step": 990
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 18.0,
|
| 1817 |
+
"eval_loss": 3.537071466445923,
|
| 1818 |
+
"eval_runtime": 4.9404,
|
| 1819 |
+
"eval_samples_per_second": 4.656,
|
| 1820 |
+
"eval_steps_per_second": 1.214,
|
| 1821 |
+
"step": 990
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 18.04,
|
| 1825 |
+
"learning_rate": 1.4538909090118846e-05,
|
| 1826 |
+
"loss": 0.0424,
|
| 1827 |
+
"step": 992
|
| 1828 |
+
},
|
| 1829 |
+
{
|
| 1830 |
+
"epoch": 18.11,
|
| 1831 |
+
"learning_rate": 1.3491511964480702e-05,
|
| 1832 |
+
"loss": 0.0406,
|
| 1833 |
+
"step": 996
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 18.18,
|
| 1837 |
+
"learning_rate": 1.2482220564763668e-05,
|
| 1838 |
+
"loss": 0.0404,
|
| 1839 |
+
"step": 1000
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 18.25,
|
| 1843 |
+
"learning_rate": 1.1511197505771842e-05,
|
| 1844 |
+
"loss": 0.039,
|
| 1845 |
+
"step": 1004
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 18.33,
|
| 1849 |
+
"learning_rate": 1.0578599236598707e-05,
|
| 1850 |
+
"loss": 0.0398,
|
| 1851 |
+
"step": 1008
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 18.4,
|
| 1855 |
+
"learning_rate": 9.684576015420277e-06,
|
| 1856 |
+
"loss": 0.0347,
|
| 1857 |
+
"step": 1012
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 18.47,
|
| 1861 |
+
"learning_rate": 8.829271885286095e-06,
|
| 1862 |
+
"loss": 0.0386,
|
| 1863 |
+
"step": 1016
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 18.55,
|
| 1867 |
+
"learning_rate": 8.012824650910938e-06,
|
| 1868 |
+
"loss": 0.034,
|
| 1869 |
+
"step": 1020
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 18.62,
|
| 1873 |
+
"learning_rate": 7.235365856472442e-06,
|
| 1874 |
+
"loss": 0.0349,
|
| 1875 |
+
"step": 1024
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 18.69,
|
| 1879 |
+
"learning_rate": 6.497020764416634e-06,
|
| 1880 |
+
"loss": 0.0376,
|
| 1881 |
+
"step": 1028
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 18.76,
|
| 1885 |
+
"learning_rate": 5.797908335276214e-06,
|
| 1886 |
+
"loss": 0.0422,
|
| 1887 |
+
"step": 1032
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 18.84,
|
| 1891 |
+
"learning_rate": 5.1381412085036995e-06,
|
| 1892 |
+
"loss": 0.0404,
|
| 1893 |
+
"step": 1036
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 18.91,
|
| 1897 |
+
"learning_rate": 4.517825684323323e-06,
|
| 1898 |
+
"loss": 0.0356,
|
| 1899 |
+
"step": 1040
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 18.98,
|
| 1903 |
+
"learning_rate": 3.937061706604072e-06,
|
| 1904 |
+
"loss": 0.0468,
|
| 1905 |
+
"step": 1044
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 19.0,
|
| 1909 |
+
"gpt4_scores": 0.575,
|
| 1910 |
+
"step": 1045
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"epoch": 19.0,
|
| 1914 |
+
"std": 0.12791598805466028,
|
| 1915 |
+
"step": 1045
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 19.0,
|
| 1919 |
+
"eval_loss": 3.5391297340393066,
|
| 1920 |
+
"eval_runtime": 4.9485,
|
| 1921 |
+
"eval_samples_per_second": 4.648,
|
| 1922 |
+
"eval_steps_per_second": 1.212,
|
| 1923 |
+
"step": 1045
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 19.05,
|
| 1927 |
+
"learning_rate": 3.3959428467570664e-06,
|
| 1928 |
+
"loss": 0.0362,
|
| 1929 |
+
"step": 1048
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 19.13,
|
| 1933 |
+
"learning_rate": 2.8945562886593944e-06,
|
| 1934 |
+
"loss": 0.0408,
|
| 1935 |
+
"step": 1052
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 19.2,
|
| 1939 |
+
"learning_rate": 2.4329828146074094e-06,
|
| 1940 |
+
"loss": 0.0391,
|
| 1941 |
+
"step": 1056
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 19.27,
|
| 1945 |
+
"learning_rate": 2.011296792301165e-06,
|
| 1946 |
+
"loss": 0.0376,
|
| 1947 |
+
"step": 1060
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 19.35,
|
| 1951 |
+
"learning_rate": 1.6295661628624448e-06,
|
| 1952 |
+
"loss": 0.0446,
|
| 1953 |
+
"step": 1064
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 19.42,
|
| 1957 |
+
"learning_rate": 1.2878524298882698e-06,
|
| 1958 |
+
"loss": 0.0332,
|
| 1959 |
+
"step": 1068
|
| 1960 |
+
},
|
| 1961 |
+
{
|
| 1962 |
+
"epoch": 19.49,
|
| 1963 |
+
"learning_rate": 9.862106495415469e-07,
|
| 1964 |
+
"loss": 0.0375,
|
| 1965 |
+
"step": 1072
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 19.56,
|
| 1969 |
+
"learning_rate": 7.246894216806354e-07,
|
| 1970 |
+
"loss": 0.0451,
|
| 1971 |
+
"step": 1076
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 19.64,
|
| 1975 |
+
"learning_rate": 5.033308820289185e-07,
|
| 1976 |
+
"loss": 0.0451,
|
| 1977 |
+
"step": 1080
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 19.71,
|
| 1981 |
+
"learning_rate": 3.221706953860093e-07,
|
| 1982 |
+
"loss": 0.0354,
|
| 1983 |
+
"step": 1084
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 19.78,
|
| 1987 |
+
"learning_rate": 1.8123804988159908e-07,
|
| 1988 |
+
"loss": 0.0397,
|
| 1989 |
+
"step": 1088
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 19.85,
|
| 1993 |
+
"learning_rate": 8.0555652272718e-08,
|
| 1994 |
+
"loss": 0.0394,
|
| 1995 |
+
"step": 1092
|
| 1996 |
+
},
|
| 1997 |
+
{
|
| 1998 |
+
"epoch": 19.93,
|
| 1999 |
+
"learning_rate": 2.0139724285161975e-08,
|
| 2000 |
+
"loss": 0.0321,
|
| 2001 |
+
"step": 1096
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 20.0,
|
| 2005 |
+
"learning_rate": 0.0,
|
| 2006 |
+
"loss": 0.0358,
|
| 2007 |
+
"step": 1100
|
| 2008 |
+
},
|
| 2009 |
+
{
|
| 2010 |
+
"epoch": 20.0,
|
| 2011 |
+
"gpt4_scores": 0.58,
|
| 2012 |
+
"step": 1100
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 20.0,
|
| 2016 |
+
"std": 0.12692517480783708,
|
| 2017 |
+
"step": 1100
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 20.0,
|
| 2021 |
+
"eval_loss": 3.5393826961517334,
|
| 2022 |
+
"eval_runtime": 4.9612,
|
| 2023 |
+
"eval_samples_per_second": 4.636,
|
| 2024 |
+
"eval_steps_per_second": 1.209,
|
| 2025 |
+
"step": 1100
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 20.0,
|
| 2029 |
+
"step": 1100,
|
| 2030 |
+
"total_flos": 3.792163606268314e+16,
|
| 2031 |
+
"train_loss": 0.31520286625081845,
|
| 2032 |
+
"train_runtime": 10333.7168,
|
| 2033 |
+
"train_samples_per_second": 0.42,
|
| 2034 |
+
"train_steps_per_second": 0.106
|
| 2035 |
+
}
|
| 2036 |
+
],
|
| 2037 |
+
"logging_steps": 4,
|
| 2038 |
+
"max_steps": 1100,
|
| 2039 |
+
"num_input_tokens_seen": 0,
|
| 2040 |
+
"num_train_epochs": 20,
|
| 2041 |
+
"save_steps": 55,
|
| 2042 |
+
"total_flos": 3.792163606268314e+16,
|
| 2043 |
+
"train_batch_size": 4,
|
| 2044 |
+
"trial_name": null,
|
| 2045 |
+
"trial_params": null
|
| 2046 |
+
}
|