Commit
•
1d6e4da
1
Parent(s):
a5fd3bc
Model save
Browse files- README.md +100 -0
- adapter_model.safetensors +1 -1
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +1417 -0
README.md
ADDED
|
@@ -0,0 +1,100 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: llama3.1
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: Llama-31-8B_task-2_180-samples_config-4
|
| 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 |
+
# Llama-31-8B_task-2_180-samples_config-4
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.7611
|
| 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: 1e-05
|
| 41 |
+
- train_batch_size: 1
|
| 42 |
+
- eval_batch_size: 1
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- gradient_accumulation_steps: 16
|
| 46 |
+
- total_train_batch_size: 16
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: cosine
|
| 49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 50 |
+
- num_epochs: 150
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-------:|:----:|:---------------:|
|
| 56 |
+
| 1.0917 | 0.9412 | 8 | 1.1326 |
|
| 57 |
+
| 1.1392 | 2.0 | 17 | 1.1271 |
|
| 58 |
+
| 1.1572 | 2.9412 | 25 | 1.1161 |
|
| 59 |
+
| 1.0856 | 4.0 | 34 | 1.0970 |
|
| 60 |
+
| 1.0789 | 4.9412 | 42 | 1.0719 |
|
| 61 |
+
| 1.0546 | 6.0 | 51 | 1.0337 |
|
| 62 |
+
| 0.9806 | 6.9412 | 59 | 0.9923 |
|
| 63 |
+
| 0.9617 | 8.0 | 68 | 0.9472 |
|
| 64 |
+
| 0.933 | 8.9412 | 76 | 0.9129 |
|
| 65 |
+
| 0.9005 | 10.0 | 85 | 0.8825 |
|
| 66 |
+
| 0.9183 | 10.9412 | 93 | 0.8578 |
|
| 67 |
+
| 0.7739 | 12.0 | 102 | 0.8335 |
|
| 68 |
+
| 0.7649 | 12.9412 | 110 | 0.8165 |
|
| 69 |
+
| 0.8197 | 14.0 | 119 | 0.8009 |
|
| 70 |
+
| 0.7488 | 14.9412 | 127 | 0.7889 |
|
| 71 |
+
| 0.7651 | 16.0 | 136 | 0.7770 |
|
| 72 |
+
| 0.6992 | 16.9412 | 144 | 0.7679 |
|
| 73 |
+
| 0.7609 | 18.0 | 153 | 0.7580 |
|
| 74 |
+
| 0.6868 | 18.9412 | 161 | 0.7510 |
|
| 75 |
+
| 0.7077 | 20.0 | 170 | 0.7437 |
|
| 76 |
+
| 0.6862 | 20.9412 | 178 | 0.7379 |
|
| 77 |
+
| 0.6939 | 22.0 | 187 | 0.7319 |
|
| 78 |
+
| 0.6564 | 22.9412 | 195 | 0.7275 |
|
| 79 |
+
| 0.6446 | 24.0 | 204 | 0.7236 |
|
| 80 |
+
| 0.6304 | 24.9412 | 212 | 0.7200 |
|
| 81 |
+
| 0.6583 | 26.0 | 221 | 0.7178 |
|
| 82 |
+
| 0.5974 | 26.9412 | 229 | 0.7159 |
|
| 83 |
+
| 0.6266 | 28.0 | 238 | 0.7159 |
|
| 84 |
+
| 0.6081 | 28.9412 | 246 | 0.7156 |
|
| 85 |
+
| 0.5853 | 30.0 | 255 | 0.7165 |
|
| 86 |
+
| 0.54 | 30.9412 | 263 | 0.7200 |
|
| 87 |
+
| 0.5438 | 32.0 | 272 | 0.7231 |
|
| 88 |
+
| 0.5163 | 32.9412 | 280 | 0.7268 |
|
| 89 |
+
| 0.5302 | 34.0 | 289 | 0.7345 |
|
| 90 |
+
| 0.4865 | 34.9412 | 297 | 0.7542 |
|
| 91 |
+
| 0.4774 | 36.0 | 306 | 0.7611 |
|
| 92 |
+
|
| 93 |
+
|
| 94 |
+
### Framework versions
|
| 95 |
+
|
| 96 |
+
- PEFT 0.12.0
|
| 97 |
+
- Transformers 4.44.0
|
| 98 |
+
- Pytorch 2.1.2+cu121
|
| 99 |
+
- Datasets 2.20.0
|
| 100 |
+
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 167832240
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f013ff6736a3c56cf3f2e1ce91c4ca7b6dec22499303b59a71f45a80eada264c
|
| 3 |
size 167832240
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.0,
|
| 3 |
+
"total_flos": 4.686344037995643e+17,
|
| 4 |
+
"train_loss": 0.7647335636459924,
|
| 5 |
+
"train_runtime": 16723.6429,
|
| 6 |
+
"train_samples": 136,
|
| 7 |
+
"train_samples_per_second": 1.22,
|
| 8 |
+
"train_steps_per_second": 0.072
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.0,
|
| 3 |
+
"total_flos": 4.686344037995643e+17,
|
| 4 |
+
"train_loss": 0.7647335636459924,
|
| 5 |
+
"train_runtime": 16723.6429,
|
| 6 |
+
"train_samples": 136,
|
| 7 |
+
"train_samples_per_second": 1.22,
|
| 8 |
+
"train_steps_per_second": 0.072
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1417 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7155756950378418,
|
| 3 |
+
"best_model_checkpoint": "data/Llama-31-8B_task-2_180-samples_config-4/checkpoint-246",
|
| 4 |
+
"epoch": 36.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 306,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11764705882352941,
|
| 13 |
+
"grad_norm": 0.6938014030456543,
|
| 14 |
+
"learning_rate": 8.333333333333334e-08,
|
| 15 |
+
"loss": 1.1573,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.23529411764705882,
|
| 20 |
+
"grad_norm": 0.7594343423843384,
|
| 21 |
+
"learning_rate": 1.6666666666666668e-07,
|
| 22 |
+
"loss": 1.1671,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.47058823529411764,
|
| 27 |
+
"grad_norm": 0.7093557715415955,
|
| 28 |
+
"learning_rate": 3.3333333333333335e-07,
|
| 29 |
+
"loss": 1.1466,
|
| 30 |
+
"step": 4
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.7058823529411765,
|
| 34 |
+
"grad_norm": 0.6927754878997803,
|
| 35 |
+
"learning_rate": 5.000000000000001e-07,
|
| 36 |
+
"loss": 1.1113,
|
| 37 |
+
"step": 6
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.9411764705882353,
|
| 41 |
+
"grad_norm": 0.6476206183433533,
|
| 42 |
+
"learning_rate": 6.666666666666667e-07,
|
| 43 |
+
"loss": 1.0917,
|
| 44 |
+
"step": 8
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.9411764705882353,
|
| 48 |
+
"eval_loss": 1.1325923204421997,
|
| 49 |
+
"eval_runtime": 37.009,
|
| 50 |
+
"eval_samples_per_second": 0.973,
|
| 51 |
+
"eval_steps_per_second": 0.973,
|
| 52 |
+
"step": 8
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.1764705882352942,
|
| 56 |
+
"grad_norm": 0.6894296407699585,
|
| 57 |
+
"learning_rate": 8.333333333333333e-07,
|
| 58 |
+
"loss": 1.1424,
|
| 59 |
+
"step": 10
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.4117647058823528,
|
| 63 |
+
"grad_norm": 0.6622559428215027,
|
| 64 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 65 |
+
"loss": 1.1225,
|
| 66 |
+
"step": 12
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.6470588235294117,
|
| 70 |
+
"grad_norm": 0.6495747566223145,
|
| 71 |
+
"learning_rate": 1.1666666666666668e-06,
|
| 72 |
+
"loss": 1.0675,
|
| 73 |
+
"step": 14
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.8823529411764706,
|
| 77 |
+
"grad_norm": 0.7559149861335754,
|
| 78 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 79 |
+
"loss": 1.1392,
|
| 80 |
+
"step": 16
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.0,
|
| 84 |
+
"eval_loss": 1.1271247863769531,
|
| 85 |
+
"eval_runtime": 36.9985,
|
| 86 |
+
"eval_samples_per_second": 0.973,
|
| 87 |
+
"eval_steps_per_second": 0.973,
|
| 88 |
+
"step": 17
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.1176470588235294,
|
| 92 |
+
"grad_norm": 0.6588976979255676,
|
| 93 |
+
"learning_rate": 1.5e-06,
|
| 94 |
+
"loss": 1.1406,
|
| 95 |
+
"step": 18
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.3529411764705883,
|
| 99 |
+
"grad_norm": 0.6868743300437927,
|
| 100 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 101 |
+
"loss": 1.0912,
|
| 102 |
+
"step": 20
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.588235294117647,
|
| 106 |
+
"grad_norm": 0.699292778968811,
|
| 107 |
+
"learning_rate": 1.8333333333333333e-06,
|
| 108 |
+
"loss": 1.1073,
|
| 109 |
+
"step": 22
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.8235294117647056,
|
| 113 |
+
"grad_norm": 0.7510226964950562,
|
| 114 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 115 |
+
"loss": 1.1572,
|
| 116 |
+
"step": 24
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.9411764705882355,
|
| 120 |
+
"eval_loss": 1.1161284446716309,
|
| 121 |
+
"eval_runtime": 37.0046,
|
| 122 |
+
"eval_samples_per_second": 0.973,
|
| 123 |
+
"eval_steps_per_second": 0.973,
|
| 124 |
+
"step": 25
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 3.0588235294117645,
|
| 128 |
+
"grad_norm": 0.7101291418075562,
|
| 129 |
+
"learning_rate": 2.166666666666667e-06,
|
| 130 |
+
"loss": 1.0988,
|
| 131 |
+
"step": 26
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.2941176470588234,
|
| 135 |
+
"grad_norm": 0.5981112718582153,
|
| 136 |
+
"learning_rate": 2.3333333333333336e-06,
|
| 137 |
+
"loss": 1.0899,
|
| 138 |
+
"step": 28
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 3.5294117647058822,
|
| 142 |
+
"grad_norm": 0.7123465538024902,
|
| 143 |
+
"learning_rate": 2.5e-06,
|
| 144 |
+
"loss": 1.1263,
|
| 145 |
+
"step": 30
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 3.764705882352941,
|
| 149 |
+
"grad_norm": 0.6823212504386902,
|
| 150 |
+
"learning_rate": 2.666666666666667e-06,
|
| 151 |
+
"loss": 1.1119,
|
| 152 |
+
"step": 32
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 4.0,
|
| 156 |
+
"grad_norm": 0.7324620485305786,
|
| 157 |
+
"learning_rate": 2.8333333333333335e-06,
|
| 158 |
+
"loss": 1.0856,
|
| 159 |
+
"step": 34
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 4.0,
|
| 163 |
+
"eval_loss": 1.0970431566238403,
|
| 164 |
+
"eval_runtime": 37.0081,
|
| 165 |
+
"eval_samples_per_second": 0.973,
|
| 166 |
+
"eval_steps_per_second": 0.973,
|
| 167 |
+
"step": 34
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 4.235294117647059,
|
| 171 |
+
"grad_norm": 0.7198031544685364,
|
| 172 |
+
"learning_rate": 3e-06,
|
| 173 |
+
"loss": 1.1012,
|
| 174 |
+
"step": 36
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 4.470588235294118,
|
| 178 |
+
"grad_norm": 0.7194771766662598,
|
| 179 |
+
"learning_rate": 3.1666666666666667e-06,
|
| 180 |
+
"loss": 1.0752,
|
| 181 |
+
"step": 38
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 4.705882352941177,
|
| 185 |
+
"grad_norm": 0.702494740486145,
|
| 186 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 187 |
+
"loss": 1.0602,
|
| 188 |
+
"step": 40
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.9411764705882355,
|
| 192 |
+
"grad_norm": 0.6754999160766602,
|
| 193 |
+
"learning_rate": 3.5e-06,
|
| 194 |
+
"loss": 1.0789,
|
| 195 |
+
"step": 42
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 4.9411764705882355,
|
| 199 |
+
"eval_loss": 1.0719497203826904,
|
| 200 |
+
"eval_runtime": 37.0094,
|
| 201 |
+
"eval_samples_per_second": 0.973,
|
| 202 |
+
"eval_steps_per_second": 0.973,
|
| 203 |
+
"step": 42
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 5.176470588235294,
|
| 207 |
+
"grad_norm": 0.7691217660903931,
|
| 208 |
+
"learning_rate": 3.6666666666666666e-06,
|
| 209 |
+
"loss": 1.0709,
|
| 210 |
+
"step": 44
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.411764705882353,
|
| 214 |
+
"grad_norm": 0.6451730132102966,
|
| 215 |
+
"learning_rate": 3.833333333333334e-06,
|
| 216 |
+
"loss": 1.0372,
|
| 217 |
+
"step": 46
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 5.647058823529412,
|
| 221 |
+
"grad_norm": 0.6900641322135925,
|
| 222 |
+
"learning_rate": 4.000000000000001e-06,
|
| 223 |
+
"loss": 0.9947,
|
| 224 |
+
"step": 48
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 5.882352941176471,
|
| 228 |
+
"grad_norm": 0.6281608939170837,
|
| 229 |
+
"learning_rate": 4.166666666666667e-06,
|
| 230 |
+
"loss": 1.0546,
|
| 231 |
+
"step": 50
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 6.0,
|
| 235 |
+
"eval_loss": 1.0336723327636719,
|
| 236 |
+
"eval_runtime": 37.0003,
|
| 237 |
+
"eval_samples_per_second": 0.973,
|
| 238 |
+
"eval_steps_per_second": 0.973,
|
| 239 |
+
"step": 51
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 6.117647058823529,
|
| 243 |
+
"grad_norm": 0.6548106074333191,
|
| 244 |
+
"learning_rate": 4.333333333333334e-06,
|
| 245 |
+
"loss": 1.0604,
|
| 246 |
+
"step": 52
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 6.352941176470588,
|
| 250 |
+
"grad_norm": 0.6569029688835144,
|
| 251 |
+
"learning_rate": 4.5e-06,
|
| 252 |
+
"loss": 0.9777,
|
| 253 |
+
"step": 54
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 6.588235294117647,
|
| 257 |
+
"grad_norm": 0.6138703227043152,
|
| 258 |
+
"learning_rate": 4.666666666666667e-06,
|
| 259 |
+
"loss": 1.0248,
|
| 260 |
+
"step": 56
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 6.823529411764706,
|
| 264 |
+
"grad_norm": 0.5528715252876282,
|
| 265 |
+
"learning_rate": 4.833333333333333e-06,
|
| 266 |
+
"loss": 0.9806,
|
| 267 |
+
"step": 58
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 6.9411764705882355,
|
| 271 |
+
"eval_loss": 0.9922827482223511,
|
| 272 |
+
"eval_runtime": 37.0043,
|
| 273 |
+
"eval_samples_per_second": 0.973,
|
| 274 |
+
"eval_steps_per_second": 0.973,
|
| 275 |
+
"step": 59
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.0588235294117645,
|
| 279 |
+
"grad_norm": 0.5543631315231323,
|
| 280 |
+
"learning_rate": 5e-06,
|
| 281 |
+
"loss": 1.0264,
|
| 282 |
+
"step": 60
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.294117647058823,
|
| 286 |
+
"grad_norm": 0.5159863233566284,
|
| 287 |
+
"learning_rate": 5.1666666666666675e-06,
|
| 288 |
+
"loss": 0.9702,
|
| 289 |
+
"step": 62
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 7.529411764705882,
|
| 293 |
+
"grad_norm": 0.5009753704071045,
|
| 294 |
+
"learning_rate": 5.333333333333334e-06,
|
| 295 |
+
"loss": 0.9452,
|
| 296 |
+
"step": 64
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.764705882352941,
|
| 300 |
+
"grad_norm": 0.49283164739608765,
|
| 301 |
+
"learning_rate": 5.500000000000001e-06,
|
| 302 |
+
"loss": 0.9367,
|
| 303 |
+
"step": 66
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 8.0,
|
| 307 |
+
"grad_norm": 0.47310400009155273,
|
| 308 |
+
"learning_rate": 5.666666666666667e-06,
|
| 309 |
+
"loss": 0.9617,
|
| 310 |
+
"step": 68
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 8.0,
|
| 314 |
+
"eval_loss": 0.9472079873085022,
|
| 315 |
+
"eval_runtime": 37.0171,
|
| 316 |
+
"eval_samples_per_second": 0.973,
|
| 317 |
+
"eval_steps_per_second": 0.973,
|
| 318 |
+
"step": 68
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 8.235294117647058,
|
| 322 |
+
"grad_norm": 0.3864011764526367,
|
| 323 |
+
"learning_rate": 5.833333333333334e-06,
|
| 324 |
+
"loss": 0.9326,
|
| 325 |
+
"step": 70
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 8.470588235294118,
|
| 329 |
+
"grad_norm": 0.37244734168052673,
|
| 330 |
+
"learning_rate": 6e-06,
|
| 331 |
+
"loss": 0.948,
|
| 332 |
+
"step": 72
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 8.705882352941176,
|
| 336 |
+
"grad_norm": 0.3857339322566986,
|
| 337 |
+
"learning_rate": 6.166666666666667e-06,
|
| 338 |
+
"loss": 0.8741,
|
| 339 |
+
"step": 74
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 8.941176470588236,
|
| 343 |
+
"grad_norm": 0.39583349227905273,
|
| 344 |
+
"learning_rate": 6.333333333333333e-06,
|
| 345 |
+
"loss": 0.933,
|
| 346 |
+
"step": 76
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 8.941176470588236,
|
| 350 |
+
"eval_loss": 0.912886381149292,
|
| 351 |
+
"eval_runtime": 37.0055,
|
| 352 |
+
"eval_samples_per_second": 0.973,
|
| 353 |
+
"eval_steps_per_second": 0.973,
|
| 354 |
+
"step": 76
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 9.176470588235293,
|
| 358 |
+
"grad_norm": 0.30479201674461365,
|
| 359 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 360 |
+
"loss": 0.8891,
|
| 361 |
+
"step": 78
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 9.411764705882353,
|
| 365 |
+
"grad_norm": 0.30881431698799133,
|
| 366 |
+
"learning_rate": 6.666666666666667e-06,
|
| 367 |
+
"loss": 0.848,
|
| 368 |
+
"step": 80
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 9.647058823529411,
|
| 372 |
+
"grad_norm": 0.3065626919269562,
|
| 373 |
+
"learning_rate": 6.833333333333334e-06,
|
| 374 |
+
"loss": 0.8886,
|
| 375 |
+
"step": 82
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 9.882352941176471,
|
| 379 |
+
"grad_norm": 0.3278353214263916,
|
| 380 |
+
"learning_rate": 7e-06,
|
| 381 |
+
"loss": 0.9005,
|
| 382 |
+
"step": 84
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 10.0,
|
| 386 |
+
"eval_loss": 0.8825066089630127,
|
| 387 |
+
"eval_runtime": 36.9987,
|
| 388 |
+
"eval_samples_per_second": 0.973,
|
| 389 |
+
"eval_steps_per_second": 0.973,
|
| 390 |
+
"step": 85
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 10.117647058823529,
|
| 394 |
+
"grad_norm": 0.3112061023712158,
|
| 395 |
+
"learning_rate": 7.166666666666667e-06,
|
| 396 |
+
"loss": 0.8631,
|
| 397 |
+
"step": 86
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 10.352941176470589,
|
| 401 |
+
"grad_norm": 0.2803957164287567,
|
| 402 |
+
"learning_rate": 7.333333333333333e-06,
|
| 403 |
+
"loss": 0.8166,
|
| 404 |
+
"step": 88
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 10.588235294117647,
|
| 408 |
+
"grad_norm": 0.2763586938381195,
|
| 409 |
+
"learning_rate": 7.500000000000001e-06,
|
| 410 |
+
"loss": 0.8208,
|
| 411 |
+
"step": 90
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 10.823529411764707,
|
| 415 |
+
"grad_norm": 0.29917630553245544,
|
| 416 |
+
"learning_rate": 7.666666666666667e-06,
|
| 417 |
+
"loss": 0.9183,
|
| 418 |
+
"step": 92
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 10.941176470588236,
|
| 422 |
+
"eval_loss": 0.8577808141708374,
|
| 423 |
+
"eval_runtime": 37.0008,
|
| 424 |
+
"eval_samples_per_second": 0.973,
|
| 425 |
+
"eval_steps_per_second": 0.973,
|
| 426 |
+
"step": 93
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 11.058823529411764,
|
| 430 |
+
"grad_norm": 0.2617398202419281,
|
| 431 |
+
"learning_rate": 7.833333333333333e-06,
|
| 432 |
+
"loss": 0.8786,
|
| 433 |
+
"step": 94
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 11.294117647058824,
|
| 437 |
+
"grad_norm": 0.251274436712265,
|
| 438 |
+
"learning_rate": 8.000000000000001e-06,
|
| 439 |
+
"loss": 0.8257,
|
| 440 |
+
"step": 96
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 11.529411764705882,
|
| 444 |
+
"grad_norm": 0.27396726608276367,
|
| 445 |
+
"learning_rate": 8.166666666666668e-06,
|
| 446 |
+
"loss": 0.8913,
|
| 447 |
+
"step": 98
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 11.764705882352942,
|
| 451 |
+
"grad_norm": 0.27446839213371277,
|
| 452 |
+
"learning_rate": 8.333333333333334e-06,
|
| 453 |
+
"loss": 0.7921,
|
| 454 |
+
"step": 100
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 12.0,
|
| 458 |
+
"grad_norm": 0.24932074546813965,
|
| 459 |
+
"learning_rate": 8.5e-06,
|
| 460 |
+
"loss": 0.7739,
|
| 461 |
+
"step": 102
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 12.0,
|
| 465 |
+
"eval_loss": 0.8334917426109314,
|
| 466 |
+
"eval_runtime": 37.0127,
|
| 467 |
+
"eval_samples_per_second": 0.973,
|
| 468 |
+
"eval_steps_per_second": 0.973,
|
| 469 |
+
"step": 102
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 12.235294117647058,
|
| 473 |
+
"grad_norm": 0.297094464302063,
|
| 474 |
+
"learning_rate": 8.666666666666668e-06,
|
| 475 |
+
"loss": 0.8754,
|
| 476 |
+
"step": 104
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 12.470588235294118,
|
| 480 |
+
"grad_norm": 0.23659107089042664,
|
| 481 |
+
"learning_rate": 8.833333333333334e-06,
|
| 482 |
+
"loss": 0.764,
|
| 483 |
+
"step": 106
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 12.705882352941176,
|
| 487 |
+
"grad_norm": 0.23405446112155914,
|
| 488 |
+
"learning_rate": 9e-06,
|
| 489 |
+
"loss": 0.8218,
|
| 490 |
+
"step": 108
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 12.941176470588236,
|
| 494 |
+
"grad_norm": 0.23257797956466675,
|
| 495 |
+
"learning_rate": 9.166666666666666e-06,
|
| 496 |
+
"loss": 0.7649,
|
| 497 |
+
"step": 110
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 12.941176470588236,
|
| 501 |
+
"eval_loss": 0.8164898157119751,
|
| 502 |
+
"eval_runtime": 37.01,
|
| 503 |
+
"eval_samples_per_second": 0.973,
|
| 504 |
+
"eval_steps_per_second": 0.973,
|
| 505 |
+
"step": 110
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 13.176470588235293,
|
| 509 |
+
"grad_norm": 0.23545639216899872,
|
| 510 |
+
"learning_rate": 9.333333333333334e-06,
|
| 511 |
+
"loss": 0.7674,
|
| 512 |
+
"step": 112
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 13.411764705882353,
|
| 516 |
+
"grad_norm": 0.26143553853034973,
|
| 517 |
+
"learning_rate": 9.5e-06,
|
| 518 |
+
"loss": 0.7662,
|
| 519 |
+
"step": 114
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 13.647058823529411,
|
| 523 |
+
"grad_norm": 0.21934638917446136,
|
| 524 |
+
"learning_rate": 9.666666666666667e-06,
|
| 525 |
+
"loss": 0.7928,
|
| 526 |
+
"step": 116
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 13.882352941176471,
|
| 530 |
+
"grad_norm": 0.2396870255470276,
|
| 531 |
+
"learning_rate": 9.833333333333333e-06,
|
| 532 |
+
"loss": 0.8197,
|
| 533 |
+
"step": 118
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 14.0,
|
| 537 |
+
"eval_loss": 0.8008810877799988,
|
| 538 |
+
"eval_runtime": 36.9979,
|
| 539 |
+
"eval_samples_per_second": 0.973,
|
| 540 |
+
"eval_steps_per_second": 0.973,
|
| 541 |
+
"step": 119
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 14.117647058823529,
|
| 545 |
+
"grad_norm": 0.24245895445346832,
|
| 546 |
+
"learning_rate": 1e-05,
|
| 547 |
+
"loss": 0.8115,
|
| 548 |
+
"step": 120
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 14.352941176470589,
|
| 552 |
+
"grad_norm": 0.23691204190254211,
|
| 553 |
+
"learning_rate": 9.999915384288723e-06,
|
| 554 |
+
"loss": 0.7565,
|
| 555 |
+
"step": 122
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 14.588235294117647,
|
| 559 |
+
"grad_norm": 0.21797290444374084,
|
| 560 |
+
"learning_rate": 9.999661540018812e-06,
|
| 561 |
+
"loss": 0.7858,
|
| 562 |
+
"step": 124
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 14.823529411764707,
|
| 566 |
+
"grad_norm": 0.23253759741783142,
|
| 567 |
+
"learning_rate": 9.999238475781957e-06,
|
| 568 |
+
"loss": 0.7488,
|
| 569 |
+
"step": 126
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 14.941176470588236,
|
| 573 |
+
"eval_loss": 0.7888637185096741,
|
| 574 |
+
"eval_runtime": 36.999,
|
| 575 |
+
"eval_samples_per_second": 0.973,
|
| 576 |
+
"eval_steps_per_second": 0.973,
|
| 577 |
+
"step": 127
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 15.058823529411764,
|
| 581 |
+
"grad_norm": 0.2564374804496765,
|
| 582 |
+
"learning_rate": 9.99864620589731e-06,
|
| 583 |
+
"loss": 0.7646,
|
| 584 |
+
"step": 128
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 15.294117647058824,
|
| 588 |
+
"grad_norm": 0.23032395541667938,
|
| 589 |
+
"learning_rate": 9.997884750411004e-06,
|
| 590 |
+
"loss": 0.7598,
|
| 591 |
+
"step": 130
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 15.529411764705882,
|
| 595 |
+
"grad_norm": 0.22293731570243835,
|
| 596 |
+
"learning_rate": 9.99695413509548e-06,
|
| 597 |
+
"loss": 0.733,
|
| 598 |
+
"step": 132
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 15.764705882352942,
|
| 602 |
+
"grad_norm": 0.2295634150505066,
|
| 603 |
+
"learning_rate": 9.995854391448607e-06,
|
| 604 |
+
"loss": 0.7611,
|
| 605 |
+
"step": 134
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 16.0,
|
| 609 |
+
"grad_norm": 0.24928726255893707,
|
| 610 |
+
"learning_rate": 9.994585556692624e-06,
|
| 611 |
+
"loss": 0.7651,
|
| 612 |
+
"step": 136
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 16.0,
|
| 616 |
+
"eval_loss": 0.7769894003868103,
|
| 617 |
+
"eval_runtime": 36.9969,
|
| 618 |
+
"eval_samples_per_second": 0.973,
|
| 619 |
+
"eval_steps_per_second": 0.973,
|
| 620 |
+
"step": 136
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 16.235294117647058,
|
| 624 |
+
"grad_norm": 0.21673335134983063,
|
| 625 |
+
"learning_rate": 9.993147673772869e-06,
|
| 626 |
+
"loss": 0.7225,
|
| 627 |
+
"step": 138
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 16.470588235294116,
|
| 631 |
+
"grad_norm": 0.22104425728321075,
|
| 632 |
+
"learning_rate": 9.991540791356342e-06,
|
| 633 |
+
"loss": 0.7727,
|
| 634 |
+
"step": 140
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 16.705882352941178,
|
| 638 |
+
"grad_norm": 0.23531343042850494,
|
| 639 |
+
"learning_rate": 9.989764963830038e-06,
|
| 640 |
+
"loss": 0.7589,
|
| 641 |
+
"step": 142
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 16.941176470588236,
|
| 645 |
+
"grad_norm": 0.21621249616146088,
|
| 646 |
+
"learning_rate": 9.987820251299121e-06,
|
| 647 |
+
"loss": 0.6992,
|
| 648 |
+
"step": 144
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 16.941176470588236,
|
| 652 |
+
"eval_loss": 0.7678544521331787,
|
| 653 |
+
"eval_runtime": 36.9951,
|
| 654 |
+
"eval_samples_per_second": 0.973,
|
| 655 |
+
"eval_steps_per_second": 0.973,
|
| 656 |
+
"step": 144
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 17.176470588235293,
|
| 660 |
+
"grad_norm": 0.242046520113945,
|
| 661 |
+
"learning_rate": 9.985706719584888e-06,
|
| 662 |
+
"loss": 0.7617,
|
| 663 |
+
"step": 146
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 17.41176470588235,
|
| 667 |
+
"grad_norm": 0.21592232584953308,
|
| 668 |
+
"learning_rate": 9.98342444022253e-06,
|
| 669 |
+
"loss": 0.6826,
|
| 670 |
+
"step": 148
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 17.647058823529413,
|
| 674 |
+
"grad_norm": 0.2592144310474396,
|
| 675 |
+
"learning_rate": 9.980973490458728e-06,
|
| 676 |
+
"loss": 0.7452,
|
| 677 |
+
"step": 150
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 17.88235294117647,
|
| 681 |
+
"grad_norm": 0.2321804314851761,
|
| 682 |
+
"learning_rate": 9.978353953249023e-06,
|
| 683 |
+
"loss": 0.7609,
|
| 684 |
+
"step": 152
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 18.0,
|
| 688 |
+
"eval_loss": 0.7580164670944214,
|
| 689 |
+
"eval_runtime": 37.0023,
|
| 690 |
+
"eval_samples_per_second": 0.973,
|
| 691 |
+
"eval_steps_per_second": 0.973,
|
| 692 |
+
"step": 153
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 18.11764705882353,
|
| 696 |
+
"grad_norm": 0.2417711317539215,
|
| 697 |
+
"learning_rate": 9.975565917255017e-06,
|
| 698 |
+
"loss": 0.6975,
|
| 699 |
+
"step": 154
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 18.352941176470587,
|
| 703 |
+
"grad_norm": 0.2503942549228668,
|
| 704 |
+
"learning_rate": 9.972609476841368e-06,
|
| 705 |
+
"loss": 0.7259,
|
| 706 |
+
"step": 156
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 18.58823529411765,
|
| 710 |
+
"grad_norm": 0.23440955579280853,
|
| 711 |
+
"learning_rate": 9.9694847320726e-06,
|
| 712 |
+
"loss": 0.7242,
|
| 713 |
+
"step": 158
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 18.823529411764707,
|
| 717 |
+
"grad_norm": 0.2478140890598297,
|
| 718 |
+
"learning_rate": 9.966191788709716e-06,
|
| 719 |
+
"loss": 0.6868,
|
| 720 |
+
"step": 160
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 18.941176470588236,
|
| 724 |
+
"eval_loss": 0.7510344386100769,
|
| 725 |
+
"eval_runtime": 37.0057,
|
| 726 |
+
"eval_samples_per_second": 0.973,
|
| 727 |
+
"eval_steps_per_second": 0.973,
|
| 728 |
+
"step": 161
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 19.058823529411764,
|
| 732 |
+
"grad_norm": 0.2523338496685028,
|
| 733 |
+
"learning_rate": 9.962730758206612e-06,
|
| 734 |
+
"loss": 0.7132,
|
| 735 |
+
"step": 162
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 19.294117647058822,
|
| 739 |
+
"grad_norm": 0.2554718255996704,
|
| 740 |
+
"learning_rate": 9.959101757706308e-06,
|
| 741 |
+
"loss": 0.7306,
|
| 742 |
+
"step": 164
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 19.529411764705884,
|
| 746 |
+
"grad_norm": 0.2540837526321411,
|
| 747 |
+
"learning_rate": 9.955304910036993e-06,
|
| 748 |
+
"loss": 0.6959,
|
| 749 |
+
"step": 166
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 19.764705882352942,
|
| 753 |
+
"grad_norm": 0.23473863303661346,
|
| 754 |
+
"learning_rate": 9.951340343707852e-06,
|
| 755 |
+
"loss": 0.677,
|
| 756 |
+
"step": 168
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 20.0,
|
| 760 |
+
"grad_norm": 0.26500314474105835,
|
| 761 |
+
"learning_rate": 9.947208192904722e-06,
|
| 762 |
+
"loss": 0.7077,
|
| 763 |
+
"step": 170
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 20.0,
|
| 767 |
+
"eval_loss": 0.7436654567718506,
|
| 768 |
+
"eval_runtime": 37.0063,
|
| 769 |
+
"eval_samples_per_second": 0.973,
|
| 770 |
+
"eval_steps_per_second": 0.973,
|
| 771 |
+
"step": 170
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 20.235294117647058,
|
| 775 |
+
"grad_norm": 0.2688054144382477,
|
| 776 |
+
"learning_rate": 9.942908597485558e-06,
|
| 777 |
+
"loss": 0.7092,
|
| 778 |
+
"step": 172
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 20.470588235294116,
|
| 782 |
+
"grad_norm": 0.24925930798053741,
|
| 783 |
+
"learning_rate": 9.938441702975689e-06,
|
| 784 |
+
"loss": 0.7176,
|
| 785 |
+
"step": 174
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 20.705882352941178,
|
| 789 |
+
"grad_norm": 0.2611039876937866,
|
| 790 |
+
"learning_rate": 9.933807660562898e-06,
|
| 791 |
+
"loss": 0.6457,
|
| 792 |
+
"step": 176
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 20.941176470588236,
|
| 796 |
+
"grad_norm": 0.2844160497188568,
|
| 797 |
+
"learning_rate": 9.929006627092298e-06,
|
| 798 |
+
"loss": 0.6862,
|
| 799 |
+
"step": 178
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 20.941176470588236,
|
| 803 |
+
"eval_loss": 0.7379278540611267,
|
| 804 |
+
"eval_runtime": 36.9999,
|
| 805 |
+
"eval_samples_per_second": 0.973,
|
| 806 |
+
"eval_steps_per_second": 0.973,
|
| 807 |
+
"step": 178
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 21.176470588235293,
|
| 811 |
+
"grad_norm": 0.27488094568252563,
|
| 812 |
+
"learning_rate": 9.924038765061042e-06,
|
| 813 |
+
"loss": 0.7067,
|
| 814 |
+
"step": 180
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 21.41176470588235,
|
| 818 |
+
"grad_norm": 0.24594415724277496,
|
| 819 |
+
"learning_rate": 9.918904242612794e-06,
|
| 820 |
+
"loss": 0.6842,
|
| 821 |
+
"step": 182
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 21.647058823529413,
|
| 825 |
+
"grad_norm": 0.2662781774997711,
|
| 826 |
+
"learning_rate": 9.913603233532067e-06,
|
| 827 |
+
"loss": 0.6661,
|
| 828 |
+
"step": 184
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 21.88235294117647,
|
| 832 |
+
"grad_norm": 0.2941681146621704,
|
| 833 |
+
"learning_rate": 9.908135917238321e-06,
|
| 834 |
+
"loss": 0.6939,
|
| 835 |
+
"step": 186
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 22.0,
|
| 839 |
+
"eval_loss": 0.7319392561912537,
|
| 840 |
+
"eval_runtime": 37.0166,
|
| 841 |
+
"eval_samples_per_second": 0.973,
|
| 842 |
+
"eval_steps_per_second": 0.973,
|
| 843 |
+
"step": 187
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 22.11764705882353,
|
| 847 |
+
"grad_norm": 0.3132288157939911,
|
| 848 |
+
"learning_rate": 9.902502478779897e-06,
|
| 849 |
+
"loss": 0.6804,
|
| 850 |
+
"step": 188
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 22.352941176470587,
|
| 854 |
+
"grad_norm": 0.2927018105983734,
|
| 855 |
+
"learning_rate": 9.896703108827758e-06,
|
| 856 |
+
"loss": 0.649,
|
| 857 |
+
"step": 190
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 22.58823529411765,
|
| 861 |
+
"grad_norm": 0.28771597146987915,
|
| 862 |
+
"learning_rate": 9.890738003669029e-06,
|
| 863 |
+
"loss": 0.6525,
|
| 864 |
+
"step": 192
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 22.823529411764707,
|
| 868 |
+
"grad_norm": 0.31475573778152466,
|
| 869 |
+
"learning_rate": 9.884607365200355e-06,
|
| 870 |
+
"loss": 0.6564,
|
| 871 |
+
"step": 194
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 22.941176470588236,
|
| 875 |
+
"eval_loss": 0.7275494933128357,
|
| 876 |
+
"eval_runtime": 37.002,
|
| 877 |
+
"eval_samples_per_second": 0.973,
|
| 878 |
+
"eval_steps_per_second": 0.973,
|
| 879 |
+
"step": 195
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 23.058823529411764,
|
| 883 |
+
"grad_norm": 0.29836949706077576,
|
| 884 |
+
"learning_rate": 9.878311400921072e-06,
|
| 885 |
+
"loss": 0.6561,
|
| 886 |
+
"step": 196
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 23.294117647058822,
|
| 890 |
+
"grad_norm": 0.30295827984809875,
|
| 891 |
+
"learning_rate": 9.871850323926178e-06,
|
| 892 |
+
"loss": 0.6815,
|
| 893 |
+
"step": 198
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 23.529411764705884,
|
| 897 |
+
"grad_norm": 0.2973706126213074,
|
| 898 |
+
"learning_rate": 9.86522435289912e-06,
|
| 899 |
+
"loss": 0.6261,
|
| 900 |
+
"step": 200
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 23.764705882352942,
|
| 904 |
+
"grad_norm": 0.32414522767066956,
|
| 905 |
+
"learning_rate": 9.858433712104403e-06,
|
| 906 |
+
"loss": 0.6752,
|
| 907 |
+
"step": 202
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 24.0,
|
| 911 |
+
"grad_norm": 0.30349239706993103,
|
| 912 |
+
"learning_rate": 9.851478631379982e-06,
|
| 913 |
+
"loss": 0.6446,
|
| 914 |
+
"step": 204
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 24.0,
|
| 918 |
+
"eval_loss": 0.7236446738243103,
|
| 919 |
+
"eval_runtime": 37.004,
|
| 920 |
+
"eval_samples_per_second": 0.973,
|
| 921 |
+
"eval_steps_per_second": 0.973,
|
| 922 |
+
"step": 204
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 24.235294117647058,
|
| 926 |
+
"grad_norm": 0.30559128522872925,
|
| 927 |
+
"learning_rate": 9.844359346129504e-06,
|
| 928 |
+
"loss": 0.6672,
|
| 929 |
+
"step": 206
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 24.470588235294116,
|
| 933 |
+
"grad_norm": 0.2979726195335388,
|
| 934 |
+
"learning_rate": 9.83707609731432e-06,
|
| 935 |
+
"loss": 0.6283,
|
| 936 |
+
"step": 208
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 24.705882352941178,
|
| 940 |
+
"grad_norm": 0.3105686604976654,
|
| 941 |
+
"learning_rate": 9.829629131445342e-06,
|
| 942 |
+
"loss": 0.6412,
|
| 943 |
+
"step": 210
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 24.941176470588236,
|
| 947 |
+
"grad_norm": 0.3595413863658905,
|
| 948 |
+
"learning_rate": 9.822018700574696e-06,
|
| 949 |
+
"loss": 0.6304,
|
| 950 |
+
"step": 212
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 24.941176470588236,
|
| 954 |
+
"eval_loss": 0.7200143337249756,
|
| 955 |
+
"eval_runtime": 37.0035,
|
| 956 |
+
"eval_samples_per_second": 0.973,
|
| 957 |
+
"eval_steps_per_second": 0.973,
|
| 958 |
+
"step": 212
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 25.176470588235293,
|
| 962 |
+
"grad_norm": 0.3050694763660431,
|
| 963 |
+
"learning_rate": 9.81424506228719e-06,
|
| 964 |
+
"loss": 0.6179,
|
| 965 |
+
"step": 214
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 25.41176470588235,
|
| 969 |
+
"grad_norm": 0.3428696095943451,
|
| 970 |
+
"learning_rate": 9.806308479691595e-06,
|
| 971 |
+
"loss": 0.6223,
|
| 972 |
+
"step": 216
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 25.647058823529413,
|
| 976 |
+
"grad_norm": 0.3474690616130829,
|
| 977 |
+
"learning_rate": 9.798209221411748e-06,
|
| 978 |
+
"loss": 0.652,
|
| 979 |
+
"step": 218
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 25.88235294117647,
|
| 983 |
+
"grad_norm": 0.38103392720222473,
|
| 984 |
+
"learning_rate": 9.789947561577445e-06,
|
| 985 |
+
"loss": 0.6583,
|
| 986 |
+
"step": 220
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 26.0,
|
| 990 |
+
"eval_loss": 0.7177853584289551,
|
| 991 |
+
"eval_runtime": 37.0015,
|
| 992 |
+
"eval_samples_per_second": 0.973,
|
| 993 |
+
"eval_steps_per_second": 0.973,
|
| 994 |
+
"step": 221
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 26.11764705882353,
|
| 998 |
+
"grad_norm": 0.3361453413963318,
|
| 999 |
+
"learning_rate": 9.781523779815178e-06,
|
| 1000 |
+
"loss": 0.6239,
|
| 1001 |
+
"step": 222
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 26.352941176470587,
|
| 1005 |
+
"grad_norm": 0.38879382610321045,
|
| 1006 |
+
"learning_rate": 9.77293816123866e-06,
|
| 1007 |
+
"loss": 0.6229,
|
| 1008 |
+
"step": 224
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 26.58823529411765,
|
| 1012 |
+
"grad_norm": 0.3591279685497284,
|
| 1013 |
+
"learning_rate": 9.764190996439181e-06,
|
| 1014 |
+
"loss": 0.6374,
|
| 1015 |
+
"step": 226
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 26.823529411764707,
|
| 1019 |
+
"grad_norm": 0.3179728388786316,
|
| 1020 |
+
"learning_rate": 9.755282581475769e-06,
|
| 1021 |
+
"loss": 0.5974,
|
| 1022 |
+
"step": 228
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 26.941176470588236,
|
| 1026 |
+
"eval_loss": 0.7159412503242493,
|
| 1027 |
+
"eval_runtime": 36.9995,
|
| 1028 |
+
"eval_samples_per_second": 0.973,
|
| 1029 |
+
"eval_steps_per_second": 0.973,
|
| 1030 |
+
"step": 229
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 27.058823529411764,
|
| 1034 |
+
"grad_norm": 0.3596520721912384,
|
| 1035 |
+
"learning_rate": 9.74621321786517e-06,
|
| 1036 |
+
"loss": 0.6042,
|
| 1037 |
+
"step": 230
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 27.294117647058822,
|
| 1041 |
+
"grad_norm": 0.36043640971183777,
|
| 1042 |
+
"learning_rate": 9.736983212571646e-06,
|
| 1043 |
+
"loss": 0.6058,
|
| 1044 |
+
"step": 232
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 27.529411764705884,
|
| 1048 |
+
"grad_norm": 0.35985061526298523,
|
| 1049 |
+
"learning_rate": 9.727592877996585e-06,
|
| 1050 |
+
"loss": 0.5959,
|
| 1051 |
+
"step": 234
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 27.764705882352942,
|
| 1055 |
+
"grad_norm": 0.3502049446105957,
|
| 1056 |
+
"learning_rate": 9.718042531967918e-06,
|
| 1057 |
+
"loss": 0.5894,
|
| 1058 |
+
"step": 236
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 28.0,
|
| 1062 |
+
"grad_norm": 0.3679797053337097,
|
| 1063 |
+
"learning_rate": 9.708332497729378e-06,
|
| 1064 |
+
"loss": 0.6266,
|
| 1065 |
+
"step": 238
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 28.0,
|
| 1069 |
+
"eval_loss": 0.7158757448196411,
|
| 1070 |
+
"eval_runtime": 37.0093,
|
| 1071 |
+
"eval_samples_per_second": 0.973,
|
| 1072 |
+
"eval_steps_per_second": 0.973,
|
| 1073 |
+
"step": 238
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 28.235294117647058,
|
| 1077 |
+
"grad_norm": 0.35959428548812866,
|
| 1078 |
+
"learning_rate": 9.698463103929542e-06,
|
| 1079 |
+
"loss": 0.5731,
|
| 1080 |
+
"step": 240
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 28.470588235294116,
|
| 1084 |
+
"grad_norm": 0.3938836455345154,
|
| 1085 |
+
"learning_rate": 9.688434684610725e-06,
|
| 1086 |
+
"loss": 0.5868,
|
| 1087 |
+
"step": 242
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 28.705882352941178,
|
| 1091 |
+
"grad_norm": 0.6573047637939453,
|
| 1092 |
+
"learning_rate": 9.678247579197658e-06,
|
| 1093 |
+
"loss": 0.6126,
|
| 1094 |
+
"step": 244
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 28.941176470588236,
|
| 1098 |
+
"grad_norm": 0.3609618842601776,
|
| 1099 |
+
"learning_rate": 9.667902132486009e-06,
|
| 1100 |
+
"loss": 0.6081,
|
| 1101 |
+
"step": 246
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 28.941176470588236,
|
| 1105 |
+
"eval_loss": 0.7155756950378418,
|
| 1106 |
+
"eval_runtime": 37.01,
|
| 1107 |
+
"eval_samples_per_second": 0.973,
|
| 1108 |
+
"eval_steps_per_second": 0.973,
|
| 1109 |
+
"step": 246
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 29.176470588235293,
|
| 1113 |
+
"grad_norm": 0.38799017667770386,
|
| 1114 |
+
"learning_rate": 9.657398694630713e-06,
|
| 1115 |
+
"loss": 0.5564,
|
| 1116 |
+
"step": 248
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 29.41176470588235,
|
| 1120 |
+
"grad_norm": 0.4135487675666809,
|
| 1121 |
+
"learning_rate": 9.646737621134112e-06,
|
| 1122 |
+
"loss": 0.6149,
|
| 1123 |
+
"step": 250
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 29.647058823529413,
|
| 1127 |
+
"grad_norm": 0.3742540180683136,
|
| 1128 |
+
"learning_rate": 9.635919272833938e-06,
|
| 1129 |
+
"loss": 0.5747,
|
| 1130 |
+
"step": 252
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 29.88235294117647,
|
| 1134 |
+
"grad_norm": 0.40949445962905884,
|
| 1135 |
+
"learning_rate": 9.62494401589108e-06,
|
| 1136 |
+
"loss": 0.5853,
|
| 1137 |
+
"step": 254
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 30.0,
|
| 1141 |
+
"eval_loss": 0.7164749503135681,
|
| 1142 |
+
"eval_runtime": 37.0011,
|
| 1143 |
+
"eval_samples_per_second": 0.973,
|
| 1144 |
+
"eval_steps_per_second": 0.973,
|
| 1145 |
+
"step": 255
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 30.11764705882353,
|
| 1149 |
+
"grad_norm": 0.4069327414035797,
|
| 1150 |
+
"learning_rate": 9.613812221777212e-06,
|
| 1151 |
+
"loss": 0.5611,
|
| 1152 |
+
"step": 256
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 30.352941176470587,
|
| 1156 |
+
"grad_norm": 0.4262481927871704,
|
| 1157 |
+
"learning_rate": 9.602524267262202e-06,
|
| 1158 |
+
"loss": 0.5845,
|
| 1159 |
+
"step": 258
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 30.58823529411765,
|
| 1163 |
+
"grad_norm": 0.42745286226272583,
|
| 1164 |
+
"learning_rate": 9.591080534401371e-06,
|
| 1165 |
+
"loss": 0.5742,
|
| 1166 |
+
"step": 260
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 30.823529411764707,
|
| 1170 |
+
"grad_norm": 0.4548822343349457,
|
| 1171 |
+
"learning_rate": 9.579481410522556e-06,
|
| 1172 |
+
"loss": 0.54,
|
| 1173 |
+
"step": 262
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 30.941176470588236,
|
| 1177 |
+
"eval_loss": 0.7200043201446533,
|
| 1178 |
+
"eval_runtime": 37.0013,
|
| 1179 |
+
"eval_samples_per_second": 0.973,
|
| 1180 |
+
"eval_steps_per_second": 0.973,
|
| 1181 |
+
"step": 263
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 31.058823529411764,
|
| 1185 |
+
"grad_norm": 0.41799986362457275,
|
| 1186 |
+
"learning_rate": 9.567727288213005e-06,
|
| 1187 |
+
"loss": 0.5325,
|
| 1188 |
+
"step": 264
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 31.294117647058822,
|
| 1192 |
+
"grad_norm": 0.4568859338760376,
|
| 1193 |
+
"learning_rate": 9.555818565306086e-06,
|
| 1194 |
+
"loss": 0.5688,
|
| 1195 |
+
"step": 266
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 31.529411764705884,
|
| 1199 |
+
"grad_norm": 0.441699355840683,
|
| 1200 |
+
"learning_rate": 9.543755644867823e-06,
|
| 1201 |
+
"loss": 0.5628,
|
| 1202 |
+
"step": 268
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 31.764705882352942,
|
| 1206 |
+
"grad_norm": 0.45061326026916504,
|
| 1207 |
+
"learning_rate": 9.531538935183252e-06,
|
| 1208 |
+
"loss": 0.5419,
|
| 1209 |
+
"step": 270
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 32.0,
|
| 1213 |
+
"grad_norm": 0.4943307638168335,
|
| 1214 |
+
"learning_rate": 9.519168849742603e-06,
|
| 1215 |
+
"loss": 0.5438,
|
| 1216 |
+
"step": 272
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 32.0,
|
| 1220 |
+
"eval_loss": 0.7231061458587646,
|
| 1221 |
+
"eval_runtime": 37.0086,
|
| 1222 |
+
"eval_samples_per_second": 0.973,
|
| 1223 |
+
"eval_steps_per_second": 0.973,
|
| 1224 |
+
"step": 272
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 32.23529411764706,
|
| 1228 |
+
"grad_norm": 0.46770891547203064,
|
| 1229 |
+
"learning_rate": 9.506645807227311e-06,
|
| 1230 |
+
"loss": 0.5162,
|
| 1231 |
+
"step": 274
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 32.470588235294116,
|
| 1235 |
+
"grad_norm": 0.48194897174835205,
|
| 1236 |
+
"learning_rate": 9.493970231495836e-06,
|
| 1237 |
+
"loss": 0.5715,
|
| 1238 |
+
"step": 276
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 32.705882352941174,
|
| 1242 |
+
"grad_norm": 0.47527700662612915,
|
| 1243 |
+
"learning_rate": 9.481142551569318e-06,
|
| 1244 |
+
"loss": 0.5233,
|
| 1245 |
+
"step": 278
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 32.94117647058823,
|
| 1249 |
+
"grad_norm": 0.5583000779151917,
|
| 1250 |
+
"learning_rate": 9.468163201617063e-06,
|
| 1251 |
+
"loss": 0.5163,
|
| 1252 |
+
"step": 280
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 32.94117647058823,
|
| 1256 |
+
"eval_loss": 0.7267724275588989,
|
| 1257 |
+
"eval_runtime": 37.0063,
|
| 1258 |
+
"eval_samples_per_second": 0.973,
|
| 1259 |
+
"eval_steps_per_second": 0.973,
|
| 1260 |
+
"step": 280
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 33.1764705882353,
|
| 1264 |
+
"grad_norm": 0.5567420125007629,
|
| 1265 |
+
"learning_rate": 9.45503262094184e-06,
|
| 1266 |
+
"loss": 0.5146,
|
| 1267 |
+
"step": 282
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 33.411764705882355,
|
| 1271 |
+
"grad_norm": 0.5160362124443054,
|
| 1272 |
+
"learning_rate": 9.441751253965022e-06,
|
| 1273 |
+
"loss": 0.5227,
|
| 1274 |
+
"step": 284
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 33.64705882352941,
|
| 1278 |
+
"grad_norm": 0.5851587057113647,
|
| 1279 |
+
"learning_rate": 9.428319550211531e-06,
|
| 1280 |
+
"loss": 0.4936,
|
| 1281 |
+
"step": 286
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 33.88235294117647,
|
| 1285 |
+
"grad_norm": 0.564764142036438,
|
| 1286 |
+
"learning_rate": 9.414737964294636e-06,
|
| 1287 |
+
"loss": 0.5302,
|
| 1288 |
+
"step": 288
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 34.0,
|
| 1292 |
+
"eval_loss": 0.7345450520515442,
|
| 1293 |
+
"eval_runtime": 37.0031,
|
| 1294 |
+
"eval_samples_per_second": 0.973,
|
| 1295 |
+
"eval_steps_per_second": 0.973,
|
| 1296 |
+
"step": 289
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 34.11764705882353,
|
| 1300 |
+
"grad_norm": 0.5420270562171936,
|
| 1301 |
+
"learning_rate": 9.401006955900555e-06,
|
| 1302 |
+
"loss": 0.5254,
|
| 1303 |
+
"step": 290
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 34.35294117647059,
|
| 1307 |
+
"grad_norm": 0.5466279983520508,
|
| 1308 |
+
"learning_rate": 9.38712698977291e-06,
|
| 1309 |
+
"loss": 0.5236,
|
| 1310 |
+
"step": 292
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 34.588235294117645,
|
| 1314 |
+
"grad_norm": 0.5974857807159424,
|
| 1315 |
+
"learning_rate": 9.37309853569698e-06,
|
| 1316 |
+
"loss": 0.4812,
|
| 1317 |
+
"step": 294
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 34.8235294117647,
|
| 1321 |
+
"grad_norm": 0.6072189807891846,
|
| 1322 |
+
"learning_rate": 9.358922068483813e-06,
|
| 1323 |
+
"loss": 0.4865,
|
| 1324 |
+
"step": 296
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 34.94117647058823,
|
| 1328 |
+
"eval_loss": 0.7542312145233154,
|
| 1329 |
+
"eval_runtime": 37.0073,
|
| 1330 |
+
"eval_samples_per_second": 0.973,
|
| 1331 |
+
"eval_steps_per_second": 0.973,
|
| 1332 |
+
"step": 297
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 35.05882352941177,
|
| 1336 |
+
"grad_norm": 0.7493066191673279,
|
| 1337 |
+
"learning_rate": 9.344598067954151e-06,
|
| 1338 |
+
"loss": 0.4651,
|
| 1339 |
+
"step": 298
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 35.294117647058826,
|
| 1343 |
+
"grad_norm": 0.6359327435493469,
|
| 1344 |
+
"learning_rate": 9.330127018922195e-06,
|
| 1345 |
+
"loss": 0.5043,
|
| 1346 |
+
"step": 300
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 35.529411764705884,
|
| 1350 |
+
"grad_norm": 0.5997797846794128,
|
| 1351 |
+
"learning_rate": 9.315509411179182e-06,
|
| 1352 |
+
"loss": 0.4598,
|
| 1353 |
+
"step": 302
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 35.76470588235294,
|
| 1357 |
+
"grad_norm": 0.586499810218811,
|
| 1358 |
+
"learning_rate": 9.30074573947683e-06,
|
| 1359 |
+
"loss": 0.4543,
|
| 1360 |
+
"step": 304
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 36.0,
|
| 1364 |
+
"grad_norm": 0.6805587410926819,
|
| 1365 |
+
"learning_rate": 9.285836503510562e-06,
|
| 1366 |
+
"loss": 0.4774,
|
| 1367 |
+
"step": 306
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 36.0,
|
| 1371 |
+
"eval_loss": 0.7611451148986816,
|
| 1372 |
+
"eval_runtime": 37.0054,
|
| 1373 |
+
"eval_samples_per_second": 0.973,
|
| 1374 |
+
"eval_steps_per_second": 0.973,
|
| 1375 |
+
"step": 306
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 36.0,
|
| 1379 |
+
"step": 306,
|
| 1380 |
+
"total_flos": 4.686344037995643e+17,
|
| 1381 |
+
"train_loss": 0.7647335636459924,
|
| 1382 |
+
"train_runtime": 16723.6429,
|
| 1383 |
+
"train_samples_per_second": 1.22,
|
| 1384 |
+
"train_steps_per_second": 0.072
|
| 1385 |
+
}
|
| 1386 |
+
],
|
| 1387 |
+
"logging_steps": 2,
|
| 1388 |
+
"max_steps": 1200,
|
| 1389 |
+
"num_input_tokens_seen": 0,
|
| 1390 |
+
"num_train_epochs": 150,
|
| 1391 |
+
"save_steps": 25,
|
| 1392 |
+
"stateful_callbacks": {
|
| 1393 |
+
"EarlyStoppingCallback": {
|
| 1394 |
+
"args": {
|
| 1395 |
+
"early_stopping_patience": 7,
|
| 1396 |
+
"early_stopping_threshold": 0.0
|
| 1397 |
+
},
|
| 1398 |
+
"attributes": {
|
| 1399 |
+
"early_stopping_patience_counter": 0
|
| 1400 |
+
}
|
| 1401 |
+
},
|
| 1402 |
+
"TrainerControl": {
|
| 1403 |
+
"args": {
|
| 1404 |
+
"should_epoch_stop": false,
|
| 1405 |
+
"should_evaluate": false,
|
| 1406 |
+
"should_log": false,
|
| 1407 |
+
"should_save": true,
|
| 1408 |
+
"should_training_stop": true
|
| 1409 |
+
},
|
| 1410 |
+
"attributes": {}
|
| 1411 |
+
}
|
| 1412 |
+
},
|
| 1413 |
+
"total_flos": 4.686344037995643e+17,
|
| 1414 |
+
"train_batch_size": 1,
|
| 1415 |
+
"trial_name": null,
|
| 1416 |
+
"trial_params": null
|
| 1417 |
+
}
|