Commit
•
f44f856
1
Parent(s):
775e6f9
Model save
Browse files- README.md +117 -0
- adapter_model.safetensors +1 -1
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +2057 -0
README.md
ADDED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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-1_180-samples_config-4_full
|
| 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-1_180-samples_config-4_full
|
| 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: 0.9228
|
| 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 |
+
| 2.4745 | 0.9412 | 8 | 2.4335 |
|
| 57 |
+
| 2.4286 | 2.0 | 17 | 2.4114 |
|
| 58 |
+
| 2.419 | 2.9412 | 25 | 2.3814 |
|
| 59 |
+
| 2.3475 | 4.0 | 34 | 2.3262 |
|
| 60 |
+
| 2.3147 | 4.9412 | 42 | 2.2541 |
|
| 61 |
+
| 2.2214 | 6.0 | 51 | 2.1716 |
|
| 62 |
+
| 2.1097 | 6.9412 | 59 | 2.0745 |
|
| 63 |
+
| 1.9617 | 8.0 | 68 | 1.9479 |
|
| 64 |
+
| 1.908 | 8.9412 | 76 | 1.8375 |
|
| 65 |
+
| 1.7669 | 10.0 | 85 | 1.6953 |
|
| 66 |
+
| 1.6325 | 10.9412 | 93 | 1.5461 |
|
| 67 |
+
| 1.3201 | 12.0 | 102 | 1.3739 |
|
| 68 |
+
| 1.2477 | 12.9412 | 110 | 1.2331 |
|
| 69 |
+
| 1.163 | 14.0 | 119 | 1.1330 |
|
| 70 |
+
| 1.0579 | 14.9412 | 127 | 1.0861 |
|
| 71 |
+
| 1.0655 | 16.0 | 136 | 1.0611 |
|
| 72 |
+
| 0.9976 | 16.9412 | 144 | 1.0455 |
|
| 73 |
+
| 1.0285 | 18.0 | 153 | 1.0318 |
|
| 74 |
+
| 0.998 | 18.9412 | 161 | 1.0205 |
|
| 75 |
+
| 1.0038 | 20.0 | 170 | 1.0102 |
|
| 76 |
+
| 0.9907 | 20.9412 | 178 | 1.0020 |
|
| 77 |
+
| 0.9673 | 22.0 | 187 | 0.9929 |
|
| 78 |
+
| 0.95 | 22.9412 | 195 | 0.9870 |
|
| 79 |
+
| 0.9467 | 24.0 | 204 | 0.9801 |
|
| 80 |
+
| 0.9423 | 24.9412 | 212 | 0.9737 |
|
| 81 |
+
| 0.937 | 26.0 | 221 | 0.9675 |
|
| 82 |
+
| 0.9035 | 26.9412 | 229 | 0.9626 |
|
| 83 |
+
| 0.9074 | 28.0 | 238 | 0.9582 |
|
| 84 |
+
| 0.8944 | 28.9412 | 246 | 0.9534 |
|
| 85 |
+
| 0.8785 | 30.0 | 255 | 0.9493 |
|
| 86 |
+
| 0.8797 | 30.9412 | 263 | 0.9451 |
|
| 87 |
+
| 0.8764 | 32.0 | 272 | 0.9422 |
|
| 88 |
+
| 0.8903 | 32.9412 | 280 | 0.9389 |
|
| 89 |
+
| 0.8835 | 34.0 | 289 | 0.9377 |
|
| 90 |
+
| 0.8452 | 34.9412 | 297 | 0.9332 |
|
| 91 |
+
| 0.8777 | 36.0 | 306 | 0.9272 |
|
| 92 |
+
| 0.8101 | 36.9412 | 314 | 0.9257 |
|
| 93 |
+
| 0.8526 | 38.0 | 323 | 0.9229 |
|
| 94 |
+
| 0.8228 | 38.9412 | 331 | 0.9197 |
|
| 95 |
+
| 0.8066 | 40.0 | 340 | 0.9176 |
|
| 96 |
+
| 0.7701 | 40.9412 | 348 | 0.9199 |
|
| 97 |
+
| 0.8132 | 42.0 | 357 | 0.9162 |
|
| 98 |
+
| 0.7804 | 42.9412 | 365 | 0.9104 |
|
| 99 |
+
| 0.7508 | 44.0 | 374 | 0.9083 |
|
| 100 |
+
| 0.7192 | 44.9412 | 382 | 0.9052 |
|
| 101 |
+
| 0.7633 | 46.0 | 391 | 0.9048 |
|
| 102 |
+
| 0.7534 | 46.9412 | 399 | 0.9052 |
|
| 103 |
+
| 0.666 | 48.0 | 408 | 0.9151 |
|
| 104 |
+
| 0.7298 | 48.9412 | 416 | 0.9143 |
|
| 105 |
+
| 0.6815 | 50.0 | 425 | 0.9157 |
|
| 106 |
+
| 0.6845 | 50.9412 | 433 | 0.9170 |
|
| 107 |
+
| 0.6524 | 52.0 | 442 | 0.9216 |
|
| 108 |
+
| 0.6397 | 52.9412 | 450 | 0.9228 |
|
| 109 |
+
|
| 110 |
+
|
| 111 |
+
### Framework versions
|
| 112 |
+
|
| 113 |
+
- PEFT 0.12.0
|
| 114 |
+
- Transformers 4.44.0
|
| 115 |
+
- Pytorch 2.1.2+cu121
|
| 116 |
+
- Datasets 2.20.0
|
| 117 |
+
- 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 sha256:226ca2d58af1e3e6c8926ddc3c47ebc0befbe10fe909bbaee1c4264be2674ec3
|
| 3 |
size 167832240
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 52.94117647058823,
|
| 3 |
+
"total_flos": 1.1473460706410496e+17,
|
| 4 |
+
"train_loss": 1.1512780372301739,
|
| 5 |
+
"train_runtime": 8506.7232,
|
| 6 |
+
"train_samples": 136,
|
| 7 |
+
"train_samples_per_second": 2.398,
|
| 8 |
+
"train_steps_per_second": 0.141
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 52.94117647058823,
|
| 3 |
+
"total_flos": 1.1473460706410496e+17,
|
| 4 |
+
"train_loss": 1.1512780372301739,
|
| 5 |
+
"train_runtime": 8506.7232,
|
| 6 |
+
"train_samples": 136,
|
| 7 |
+
"train_samples_per_second": 2.398,
|
| 8 |
+
"train_steps_per_second": 0.141
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2057 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9047689437866211,
|
| 3 |
+
"best_model_checkpoint": "data/Llama-31-8B_task-1_180-samples_config-4_full/checkpoint-391",
|
| 4 |
+
"epoch": 52.94117647058823,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 450,
|
| 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": 1.835170865058899,
|
| 14 |
+
"learning_rate": 8.333333333333334e-08,
|
| 15 |
+
"loss": 2.4537,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.23529411764705882,
|
| 20 |
+
"grad_norm": 1.8771226406097412,
|
| 21 |
+
"learning_rate": 1.6666666666666668e-07,
|
| 22 |
+
"loss": 2.3789,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.47058823529411764,
|
| 27 |
+
"grad_norm": 2.0154590606689453,
|
| 28 |
+
"learning_rate": 3.3333333333333335e-07,
|
| 29 |
+
"loss": 2.4445,
|
| 30 |
+
"step": 4
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.7058823529411765,
|
| 34 |
+
"grad_norm": 2.0653038024902344,
|
| 35 |
+
"learning_rate": 5.000000000000001e-07,
|
| 36 |
+
"loss": 2.4809,
|
| 37 |
+
"step": 6
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.9411764705882353,
|
| 41 |
+
"grad_norm": 1.925260305404663,
|
| 42 |
+
"learning_rate": 6.666666666666667e-07,
|
| 43 |
+
"loss": 2.4745,
|
| 44 |
+
"step": 8
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.9411764705882353,
|
| 48 |
+
"eval_loss": 2.43351149559021,
|
| 49 |
+
"eval_runtime": 14.4922,
|
| 50 |
+
"eval_samples_per_second": 2.484,
|
| 51 |
+
"eval_steps_per_second": 2.484,
|
| 52 |
+
"step": 8
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.1764705882352942,
|
| 56 |
+
"grad_norm": 1.789984107017517,
|
| 57 |
+
"learning_rate": 8.333333333333333e-07,
|
| 58 |
+
"loss": 2.4155,
|
| 59 |
+
"step": 10
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.4117647058823528,
|
| 63 |
+
"grad_norm": 1.7350174188613892,
|
| 64 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 65 |
+
"loss": 2.4448,
|
| 66 |
+
"step": 12
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.6470588235294117,
|
| 70 |
+
"grad_norm": 1.7722196578979492,
|
| 71 |
+
"learning_rate": 1.1666666666666668e-06,
|
| 72 |
+
"loss": 2.4888,
|
| 73 |
+
"step": 14
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.8823529411764706,
|
| 77 |
+
"grad_norm": 1.6953096389770508,
|
| 78 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 79 |
+
"loss": 2.4286,
|
| 80 |
+
"step": 16
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.0,
|
| 84 |
+
"eval_loss": 2.411425828933716,
|
| 85 |
+
"eval_runtime": 14.4913,
|
| 86 |
+
"eval_samples_per_second": 2.484,
|
| 87 |
+
"eval_steps_per_second": 2.484,
|
| 88 |
+
"step": 17
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.1176470588235294,
|
| 92 |
+
"grad_norm": 1.5511802434921265,
|
| 93 |
+
"learning_rate": 1.5e-06,
|
| 94 |
+
"loss": 2.4245,
|
| 95 |
+
"step": 18
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.3529411764705883,
|
| 99 |
+
"grad_norm": 1.547842025756836,
|
| 100 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 101 |
+
"loss": 2.4254,
|
| 102 |
+
"step": 20
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.588235294117647,
|
| 106 |
+
"grad_norm": 1.5457754135131836,
|
| 107 |
+
"learning_rate": 1.8333333333333333e-06,
|
| 108 |
+
"loss": 2.4147,
|
| 109 |
+
"step": 22
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.8235294117647056,
|
| 113 |
+
"grad_norm": 1.5168875455856323,
|
| 114 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 115 |
+
"loss": 2.419,
|
| 116 |
+
"step": 24
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.9411764705882355,
|
| 120 |
+
"eval_loss": 2.381402015686035,
|
| 121 |
+
"eval_runtime": 14.4968,
|
| 122 |
+
"eval_samples_per_second": 2.483,
|
| 123 |
+
"eval_steps_per_second": 2.483,
|
| 124 |
+
"step": 25
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 3.0588235294117645,
|
| 128 |
+
"grad_norm": 1.573783278465271,
|
| 129 |
+
"learning_rate": 2.166666666666667e-06,
|
| 130 |
+
"loss": 2.375,
|
| 131 |
+
"step": 26
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.2941176470588234,
|
| 135 |
+
"grad_norm": 1.4476714134216309,
|
| 136 |
+
"learning_rate": 2.3333333333333336e-06,
|
| 137 |
+
"loss": 2.3949,
|
| 138 |
+
"step": 28
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 3.5294117647058822,
|
| 142 |
+
"grad_norm": 1.569410800933838,
|
| 143 |
+
"learning_rate": 2.5e-06,
|
| 144 |
+
"loss": 2.3609,
|
| 145 |
+
"step": 30
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 3.764705882352941,
|
| 149 |
+
"grad_norm": 1.604489803314209,
|
| 150 |
+
"learning_rate": 2.666666666666667e-06,
|
| 151 |
+
"loss": 2.3778,
|
| 152 |
+
"step": 32
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 4.0,
|
| 156 |
+
"grad_norm": 1.7640857696533203,
|
| 157 |
+
"learning_rate": 2.8333333333333335e-06,
|
| 158 |
+
"loss": 2.3475,
|
| 159 |
+
"step": 34
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 4.0,
|
| 163 |
+
"eval_loss": 2.326162099838257,
|
| 164 |
+
"eval_runtime": 14.494,
|
| 165 |
+
"eval_samples_per_second": 2.484,
|
| 166 |
+
"eval_steps_per_second": 2.484,
|
| 167 |
+
"step": 34
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 4.235294117647059,
|
| 171 |
+
"grad_norm": 1.7499988079071045,
|
| 172 |
+
"learning_rate": 3e-06,
|
| 173 |
+
"loss": 2.3113,
|
| 174 |
+
"step": 36
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 4.470588235294118,
|
| 178 |
+
"grad_norm": 1.8631538152694702,
|
| 179 |
+
"learning_rate": 3.1666666666666667e-06,
|
| 180 |
+
"loss": 2.3479,
|
| 181 |
+
"step": 38
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 4.705882352941177,
|
| 185 |
+
"grad_norm": 1.75290846824646,
|
| 186 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 187 |
+
"loss": 2.2702,
|
| 188 |
+
"step": 40
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.9411764705882355,
|
| 192 |
+
"grad_norm": 1.6694003343582153,
|
| 193 |
+
"learning_rate": 3.5e-06,
|
| 194 |
+
"loss": 2.3147,
|
| 195 |
+
"step": 42
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 4.9411764705882355,
|
| 199 |
+
"eval_loss": 2.254105567932129,
|
| 200 |
+
"eval_runtime": 14.4974,
|
| 201 |
+
"eval_samples_per_second": 2.483,
|
| 202 |
+
"eval_steps_per_second": 2.483,
|
| 203 |
+
"step": 42
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 5.176470588235294,
|
| 207 |
+
"grad_norm": 1.4519027471542358,
|
| 208 |
+
"learning_rate": 3.6666666666666666e-06,
|
| 209 |
+
"loss": 2.2606,
|
| 210 |
+
"step": 44
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.411764705882353,
|
| 214 |
+
"grad_norm": 1.507379174232483,
|
| 215 |
+
"learning_rate": 3.833333333333334e-06,
|
| 216 |
+
"loss": 2.1865,
|
| 217 |
+
"step": 46
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 5.647058823529412,
|
| 221 |
+
"grad_norm": 1.6610608100891113,
|
| 222 |
+
"learning_rate": 4.000000000000001e-06,
|
| 223 |
+
"loss": 2.2445,
|
| 224 |
+
"step": 48
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 5.882352941176471,
|
| 228 |
+
"grad_norm": 1.8665122985839844,
|
| 229 |
+
"learning_rate": 4.166666666666667e-06,
|
| 230 |
+
"loss": 2.2214,
|
| 231 |
+
"step": 50
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 6.0,
|
| 235 |
+
"eval_loss": 2.1715917587280273,
|
| 236 |
+
"eval_runtime": 14.4909,
|
| 237 |
+
"eval_samples_per_second": 2.484,
|
| 238 |
+
"eval_steps_per_second": 2.484,
|
| 239 |
+
"step": 51
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 6.117647058823529,
|
| 243 |
+
"grad_norm": 2.0021536350250244,
|
| 244 |
+
"learning_rate": 4.333333333333334e-06,
|
| 245 |
+
"loss": 2.1689,
|
| 246 |
+
"step": 52
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 6.352941176470588,
|
| 250 |
+
"grad_norm": 1.3459073305130005,
|
| 251 |
+
"learning_rate": 4.5e-06,
|
| 252 |
+
"loss": 2.1659,
|
| 253 |
+
"step": 54
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 6.588235294117647,
|
| 257 |
+
"grad_norm": 1.1446928977966309,
|
| 258 |
+
"learning_rate": 4.666666666666667e-06,
|
| 259 |
+
"loss": 2.1603,
|
| 260 |
+
"step": 56
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 6.823529411764706,
|
| 264 |
+
"grad_norm": 1.0907998085021973,
|
| 265 |
+
"learning_rate": 4.833333333333333e-06,
|
| 266 |
+
"loss": 2.1097,
|
| 267 |
+
"step": 58
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 6.9411764705882355,
|
| 271 |
+
"eval_loss": 2.074471950531006,
|
| 272 |
+
"eval_runtime": 14.4917,
|
| 273 |
+
"eval_samples_per_second": 2.484,
|
| 274 |
+
"eval_steps_per_second": 2.484,
|
| 275 |
+
"step": 59
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.0588235294117645,
|
| 279 |
+
"grad_norm": 1.0119956731796265,
|
| 280 |
+
"learning_rate": 5e-06,
|
| 281 |
+
"loss": 2.1078,
|
| 282 |
+
"step": 60
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.294117647058823,
|
| 286 |
+
"grad_norm": 1.069096326828003,
|
| 287 |
+
"learning_rate": 5.1666666666666675e-06,
|
| 288 |
+
"loss": 2.0802,
|
| 289 |
+
"step": 62
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 7.529411764705882,
|
| 293 |
+
"grad_norm": 1.0303503274917603,
|
| 294 |
+
"learning_rate": 5.333333333333334e-06,
|
| 295 |
+
"loss": 2.0106,
|
| 296 |
+
"step": 64
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.764705882352941,
|
| 300 |
+
"grad_norm": 1.0196495056152344,
|
| 301 |
+
"learning_rate": 5.500000000000001e-06,
|
| 302 |
+
"loss": 2.0143,
|
| 303 |
+
"step": 66
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 8.0,
|
| 307 |
+
"grad_norm": 0.940999448299408,
|
| 308 |
+
"learning_rate": 5.666666666666667e-06,
|
| 309 |
+
"loss": 1.9617,
|
| 310 |
+
"step": 68
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 8.0,
|
| 314 |
+
"eval_loss": 1.9479138851165771,
|
| 315 |
+
"eval_runtime": 14.5012,
|
| 316 |
+
"eval_samples_per_second": 2.483,
|
| 317 |
+
"eval_steps_per_second": 2.483,
|
| 318 |
+
"step": 68
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 8.235294117647058,
|
| 322 |
+
"grad_norm": 0.9313414692878723,
|
| 323 |
+
"learning_rate": 5.833333333333334e-06,
|
| 324 |
+
"loss": 1.9513,
|
| 325 |
+
"step": 70
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 8.470588235294118,
|
| 329 |
+
"grad_norm": 0.8379154801368713,
|
| 330 |
+
"learning_rate": 6e-06,
|
| 331 |
+
"loss": 1.9401,
|
| 332 |
+
"step": 72
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 8.705882352941176,
|
| 336 |
+
"grad_norm": 0.8604364395141602,
|
| 337 |
+
"learning_rate": 6.166666666666667e-06,
|
| 338 |
+
"loss": 1.8323,
|
| 339 |
+
"step": 74
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 8.941176470588236,
|
| 343 |
+
"grad_norm": 0.895465612411499,
|
| 344 |
+
"learning_rate": 6.333333333333333e-06,
|
| 345 |
+
"loss": 1.908,
|
| 346 |
+
"step": 76
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 8.941176470588236,
|
| 350 |
+
"eval_loss": 1.8374704122543335,
|
| 351 |
+
"eval_runtime": 14.5083,
|
| 352 |
+
"eval_samples_per_second": 2.481,
|
| 353 |
+
"eval_steps_per_second": 2.481,
|
| 354 |
+
"step": 76
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 9.176470588235293,
|
| 358 |
+
"grad_norm": 0.7938605546951294,
|
| 359 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 360 |
+
"loss": 1.811,
|
| 361 |
+
"step": 78
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 9.411764705882353,
|
| 365 |
+
"grad_norm": 0.8125358819961548,
|
| 366 |
+
"learning_rate": 6.666666666666667e-06,
|
| 367 |
+
"loss": 1.7889,
|
| 368 |
+
"step": 80
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 9.647058823529411,
|
| 372 |
+
"grad_norm": 0.8654666543006897,
|
| 373 |
+
"learning_rate": 6.833333333333334e-06,
|
| 374 |
+
"loss": 1.756,
|
| 375 |
+
"step": 82
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 9.882352941176471,
|
| 379 |
+
"grad_norm": 0.9033392071723938,
|
| 380 |
+
"learning_rate": 7e-06,
|
| 381 |
+
"loss": 1.7669,
|
| 382 |
+
"step": 84
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 10.0,
|
| 386 |
+
"eval_loss": 1.6952563524246216,
|
| 387 |
+
"eval_runtime": 14.4915,
|
| 388 |
+
"eval_samples_per_second": 2.484,
|
| 389 |
+
"eval_steps_per_second": 2.484,
|
| 390 |
+
"step": 85
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 10.117647058823529,
|
| 394 |
+
"grad_norm": 0.8959240317344666,
|
| 395 |
+
"learning_rate": 7.166666666666667e-06,
|
| 396 |
+
"loss": 1.6923,
|
| 397 |
+
"step": 86
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 10.352941176470589,
|
| 401 |
+
"grad_norm": 0.9546213150024414,
|
| 402 |
+
"learning_rate": 7.333333333333333e-06,
|
| 403 |
+
"loss": 1.6253,
|
| 404 |
+
"step": 88
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 10.588235294117647,
|
| 408 |
+
"grad_norm": 1.0652508735656738,
|
| 409 |
+
"learning_rate": 7.500000000000001e-06,
|
| 410 |
+
"loss": 1.6249,
|
| 411 |
+
"step": 90
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 10.823529411764707,
|
| 415 |
+
"grad_norm": 1.0107028484344482,
|
| 416 |
+
"learning_rate": 7.666666666666667e-06,
|
| 417 |
+
"loss": 1.6325,
|
| 418 |
+
"step": 92
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 10.941176470588236,
|
| 422 |
+
"eval_loss": 1.5460655689239502,
|
| 423 |
+
"eval_runtime": 14.5077,
|
| 424 |
+
"eval_samples_per_second": 2.481,
|
| 425 |
+
"eval_steps_per_second": 2.481,
|
| 426 |
+
"step": 93
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 11.058823529411764,
|
| 430 |
+
"grad_norm": 0.8419122099876404,
|
| 431 |
+
"learning_rate": 7.833333333333333e-06,
|
| 432 |
+
"loss": 1.5955,
|
| 433 |
+
"step": 94
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 11.294117647058824,
|
| 437 |
+
"grad_norm": 0.8873438835144043,
|
| 438 |
+
"learning_rate": 8.000000000000001e-06,
|
| 439 |
+
"loss": 1.5361,
|
| 440 |
+
"step": 96
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 11.529411764705882,
|
| 444 |
+
"grad_norm": 0.8289064168930054,
|
| 445 |
+
"learning_rate": 8.166666666666668e-06,
|
| 446 |
+
"loss": 1.4957,
|
| 447 |
+
"step": 98
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 11.764705882352942,
|
| 451 |
+
"grad_norm": 0.890578031539917,
|
| 452 |
+
"learning_rate": 8.333333333333334e-06,
|
| 453 |
+
"loss": 1.4318,
|
| 454 |
+
"step": 100
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 12.0,
|
| 458 |
+
"grad_norm": 0.8572049140930176,
|
| 459 |
+
"learning_rate": 8.5e-06,
|
| 460 |
+
"loss": 1.3201,
|
| 461 |
+
"step": 102
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 12.0,
|
| 465 |
+
"eval_loss": 1.3738850355148315,
|
| 466 |
+
"eval_runtime": 14.5098,
|
| 467 |
+
"eval_samples_per_second": 2.481,
|
| 468 |
+
"eval_steps_per_second": 2.481,
|
| 469 |
+
"step": 102
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 12.235294117647058,
|
| 473 |
+
"grad_norm": 0.9629900455474854,
|
| 474 |
+
"learning_rate": 8.666666666666668e-06,
|
| 475 |
+
"loss": 1.3875,
|
| 476 |
+
"step": 104
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 12.470588235294118,
|
| 480 |
+
"grad_norm": 0.845135509967804,
|
| 481 |
+
"learning_rate": 8.833333333333334e-06,
|
| 482 |
+
"loss": 1.2982,
|
| 483 |
+
"step": 106
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 12.705882352941176,
|
| 487 |
+
"grad_norm": 0.8272332549095154,
|
| 488 |
+
"learning_rate": 9e-06,
|
| 489 |
+
"loss": 1.2623,
|
| 490 |
+
"step": 108
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 12.941176470588236,
|
| 494 |
+
"grad_norm": 0.777883768081665,
|
| 495 |
+
"learning_rate": 9.166666666666666e-06,
|
| 496 |
+
"loss": 1.2477,
|
| 497 |
+
"step": 110
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 12.941176470588236,
|
| 501 |
+
"eval_loss": 1.233130931854248,
|
| 502 |
+
"eval_runtime": 14.5015,
|
| 503 |
+
"eval_samples_per_second": 2.482,
|
| 504 |
+
"eval_steps_per_second": 2.482,
|
| 505 |
+
"step": 110
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 13.176470588235293,
|
| 509 |
+
"grad_norm": 0.6998254656791687,
|
| 510 |
+
"learning_rate": 9.333333333333334e-06,
|
| 511 |
+
"loss": 1.209,
|
| 512 |
+
"step": 112
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 13.411764705882353,
|
| 516 |
+
"grad_norm": 0.6446437835693359,
|
| 517 |
+
"learning_rate": 9.5e-06,
|
| 518 |
+
"loss": 1.165,
|
| 519 |
+
"step": 114
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 13.647058823529411,
|
| 523 |
+
"grad_norm": 0.6399714946746826,
|
| 524 |
+
"learning_rate": 9.666666666666667e-06,
|
| 525 |
+
"loss": 1.1563,
|
| 526 |
+
"step": 116
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 13.882352941176471,
|
| 530 |
+
"grad_norm": 0.6025314331054688,
|
| 531 |
+
"learning_rate": 9.833333333333333e-06,
|
| 532 |
+
"loss": 1.163,
|
| 533 |
+
"step": 118
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 14.0,
|
| 537 |
+
"eval_loss": 1.1329914331436157,
|
| 538 |
+
"eval_runtime": 14.4965,
|
| 539 |
+
"eval_samples_per_second": 2.483,
|
| 540 |
+
"eval_steps_per_second": 2.483,
|
| 541 |
+
"step": 119
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 14.117647058823529,
|
| 545 |
+
"grad_norm": 0.5692251920700073,
|
| 546 |
+
"learning_rate": 1e-05,
|
| 547 |
+
"loss": 1.1526,
|
| 548 |
+
"step": 120
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 14.352941176470589,
|
| 552 |
+
"grad_norm": 0.5997884273529053,
|
| 553 |
+
"learning_rate": 9.999915384288723e-06,
|
| 554 |
+
"loss": 1.1087,
|
| 555 |
+
"step": 122
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 14.588235294117647,
|
| 559 |
+
"grad_norm": 0.575445294380188,
|
| 560 |
+
"learning_rate": 9.999661540018812e-06,
|
| 561 |
+
"loss": 1.1228,
|
| 562 |
+
"step": 124
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 14.823529411764707,
|
| 566 |
+
"grad_norm": 0.5134881734848022,
|
| 567 |
+
"learning_rate": 9.999238475781957e-06,
|
| 568 |
+
"loss": 1.0579,
|
| 569 |
+
"step": 126
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 14.941176470588236,
|
| 573 |
+
"eval_loss": 1.086132526397705,
|
| 574 |
+
"eval_runtime": 14.5111,
|
| 575 |
+
"eval_samples_per_second": 2.481,
|
| 576 |
+
"eval_steps_per_second": 2.481,
|
| 577 |
+
"step": 127
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 15.058823529411764,
|
| 581 |
+
"grad_norm": 0.49025943875312805,
|
| 582 |
+
"learning_rate": 9.99864620589731e-06,
|
| 583 |
+
"loss": 1.0658,
|
| 584 |
+
"step": 128
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 15.294117647058824,
|
| 588 |
+
"grad_norm": 0.41751477122306824,
|
| 589 |
+
"learning_rate": 9.997884750411004e-06,
|
| 590 |
+
"loss": 1.0543,
|
| 591 |
+
"step": 130
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 15.529411764705882,
|
| 595 |
+
"grad_norm": 0.4086260199546814,
|
| 596 |
+
"learning_rate": 9.99695413509548e-06,
|
| 597 |
+
"loss": 1.0696,
|
| 598 |
+
"step": 132
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 15.764705882352942,
|
| 602 |
+
"grad_norm": 0.38625603914260864,
|
| 603 |
+
"learning_rate": 9.995854391448607e-06,
|
| 604 |
+
"loss": 1.0581,
|
| 605 |
+
"step": 134
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 16.0,
|
| 609 |
+
"grad_norm": 0.3801828622817993,
|
| 610 |
+
"learning_rate": 9.994585556692624e-06,
|
| 611 |
+
"loss": 1.0655,
|
| 612 |
+
"step": 136
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 16.0,
|
| 616 |
+
"eval_loss": 1.0610604286193848,
|
| 617 |
+
"eval_runtime": 14.4857,
|
| 618 |
+
"eval_samples_per_second": 2.485,
|
| 619 |
+
"eval_steps_per_second": 2.485,
|
| 620 |
+
"step": 136
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 16.235294117647058,
|
| 624 |
+
"grad_norm": 0.3603919744491577,
|
| 625 |
+
"learning_rate": 9.993147673772869e-06,
|
| 626 |
+
"loss": 1.0141,
|
| 627 |
+
"step": 138
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 16.470588235294116,
|
| 631 |
+
"grad_norm": 0.3525612950325012,
|
| 632 |
+
"learning_rate": 9.991540791356342e-06,
|
| 633 |
+
"loss": 1.0913,
|
| 634 |
+
"step": 140
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 16.705882352941178,
|
| 638 |
+
"grad_norm": 0.34835246205329895,
|
| 639 |
+
"learning_rate": 9.989764963830038e-06,
|
| 640 |
+
"loss": 1.0581,
|
| 641 |
+
"step": 142
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 16.941176470588236,
|
| 645 |
+
"grad_norm": 0.3293631076812744,
|
| 646 |
+
"learning_rate": 9.987820251299121e-06,
|
| 647 |
+
"loss": 0.9976,
|
| 648 |
+
"step": 144
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 16.941176470588236,
|
| 652 |
+
"eval_loss": 1.0454792976379395,
|
| 653 |
+
"eval_runtime": 14.4924,
|
| 654 |
+
"eval_samples_per_second": 2.484,
|
| 655 |
+
"eval_steps_per_second": 2.484,
|
| 656 |
+
"step": 144
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 17.176470588235293,
|
| 660 |
+
"grad_norm": 0.3257598578929901,
|
| 661 |
+
"learning_rate": 9.985706719584888e-06,
|
| 662 |
+
"loss": 1.0318,
|
| 663 |
+
"step": 146
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 17.41176470588235,
|
| 667 |
+
"grad_norm": 0.351471483707428,
|
| 668 |
+
"learning_rate": 9.98342444022253e-06,
|
| 669 |
+
"loss": 0.9863,
|
| 670 |
+
"step": 148
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 17.647058823529413,
|
| 674 |
+
"grad_norm": 0.35920023918151855,
|
| 675 |
+
"learning_rate": 9.980973490458728e-06,
|
| 676 |
+
"loss": 1.0559,
|
| 677 |
+
"step": 150
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 17.88235294117647,
|
| 681 |
+
"grad_norm": 0.3804563283920288,
|
| 682 |
+
"learning_rate": 9.978353953249023e-06,
|
| 683 |
+
"loss": 1.0285,
|
| 684 |
+
"step": 152
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 18.0,
|
| 688 |
+
"eval_loss": 1.0318020582199097,
|
| 689 |
+
"eval_runtime": 14.4883,
|
| 690 |
+
"eval_samples_per_second": 2.485,
|
| 691 |
+
"eval_steps_per_second": 2.485,
|
| 692 |
+
"step": 153
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 18.11764705882353,
|
| 696 |
+
"grad_norm": 0.3365037739276886,
|
| 697 |
+
"learning_rate": 9.975565917255017e-06,
|
| 698 |
+
"loss": 0.9899,
|
| 699 |
+
"step": 154
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 18.352941176470587,
|
| 703 |
+
"grad_norm": 0.33106836676597595,
|
| 704 |
+
"learning_rate": 9.972609476841368e-06,
|
| 705 |
+
"loss": 1.0316,
|
| 706 |
+
"step": 156
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 18.58823529411765,
|
| 710 |
+
"grad_norm": 0.34729650616645813,
|
| 711 |
+
"learning_rate": 9.9694847320726e-06,
|
| 712 |
+
"loss": 1.0052,
|
| 713 |
+
"step": 158
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 18.823529411764707,
|
| 717 |
+
"grad_norm": 0.3627675473690033,
|
| 718 |
+
"learning_rate": 9.966191788709716e-06,
|
| 719 |
+
"loss": 0.998,
|
| 720 |
+
"step": 160
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 18.941176470588236,
|
| 724 |
+
"eval_loss": 1.0204837322235107,
|
| 725 |
+
"eval_runtime": 14.4913,
|
| 726 |
+
"eval_samples_per_second": 2.484,
|
| 727 |
+
"eval_steps_per_second": 2.484,
|
| 728 |
+
"step": 161
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 19.058823529411764,
|
| 732 |
+
"grad_norm": 0.37889716029167175,
|
| 733 |
+
"learning_rate": 9.962730758206612e-06,
|
| 734 |
+
"loss": 0.9902,
|
| 735 |
+
"step": 162
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 19.294117647058822,
|
| 739 |
+
"grad_norm": 0.36187246441841125,
|
| 740 |
+
"learning_rate": 9.959101757706308e-06,
|
| 741 |
+
"loss": 1.0316,
|
| 742 |
+
"step": 164
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 19.529411764705884,
|
| 746 |
+
"grad_norm": 0.4003053903579712,
|
| 747 |
+
"learning_rate": 9.955304910036993e-06,
|
| 748 |
+
"loss": 0.9643,
|
| 749 |
+
"step": 166
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 19.764705882352942,
|
| 753 |
+
"grad_norm": 0.351055771112442,
|
| 754 |
+
"learning_rate": 9.951340343707852e-06,
|
| 755 |
+
"loss": 0.9944,
|
| 756 |
+
"step": 168
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 20.0,
|
| 760 |
+
"grad_norm": 0.3519394099712372,
|
| 761 |
+
"learning_rate": 9.947208192904722e-06,
|
| 762 |
+
"loss": 1.0038,
|
| 763 |
+
"step": 170
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 20.0,
|
| 767 |
+
"eval_loss": 1.01017427444458,
|
| 768 |
+
"eval_runtime": 14.4899,
|
| 769 |
+
"eval_samples_per_second": 2.484,
|
| 770 |
+
"eval_steps_per_second": 2.484,
|
| 771 |
+
"step": 170
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 20.235294117647058,
|
| 775 |
+
"grad_norm": 0.3706846237182617,
|
| 776 |
+
"learning_rate": 9.942908597485558e-06,
|
| 777 |
+
"loss": 0.9619,
|
| 778 |
+
"step": 172
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 20.470588235294116,
|
| 782 |
+
"grad_norm": 0.39562636613845825,
|
| 783 |
+
"learning_rate": 9.938441702975689e-06,
|
| 784 |
+
"loss": 1.0219,
|
| 785 |
+
"step": 174
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 20.705882352941178,
|
| 789 |
+
"grad_norm": 0.48325368762016296,
|
| 790 |
+
"learning_rate": 9.933807660562898e-06,
|
| 791 |
+
"loss": 0.9589,
|
| 792 |
+
"step": 176
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 20.941176470588236,
|
| 796 |
+
"grad_norm": 0.4090341627597809,
|
| 797 |
+
"learning_rate": 9.929006627092298e-06,
|
| 798 |
+
"loss": 0.9907,
|
| 799 |
+
"step": 178
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 20.941176470588236,
|
| 803 |
+
"eval_loss": 1.0019959211349487,
|
| 804 |
+
"eval_runtime": 14.4938,
|
| 805 |
+
"eval_samples_per_second": 2.484,
|
| 806 |
+
"eval_steps_per_second": 2.484,
|
| 807 |
+
"step": 178
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 21.176470588235293,
|
| 811 |
+
"grad_norm": 0.3393973708152771,
|
| 812 |
+
"learning_rate": 9.924038765061042e-06,
|
| 813 |
+
"loss": 0.9841,
|
| 814 |
+
"step": 180
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 21.41176470588235,
|
| 818 |
+
"grad_norm": 0.47041693329811096,
|
| 819 |
+
"learning_rate": 9.918904242612794e-06,
|
| 820 |
+
"loss": 1.0023,
|
| 821 |
+
"step": 182
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 21.647058823529413,
|
| 825 |
+
"grad_norm": 0.4426097273826599,
|
| 826 |
+
"learning_rate": 9.913603233532067e-06,
|
| 827 |
+
"loss": 0.9654,
|
| 828 |
+
"step": 184
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 21.88235294117647,
|
| 832 |
+
"grad_norm": 0.4066382646560669,
|
| 833 |
+
"learning_rate": 9.908135917238321e-06,
|
| 834 |
+
"loss": 0.9673,
|
| 835 |
+
"step": 186
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 22.0,
|
| 839 |
+
"eval_loss": 0.9929290413856506,
|
| 840 |
+
"eval_runtime": 14.5036,
|
| 841 |
+
"eval_samples_per_second": 2.482,
|
| 842 |
+
"eval_steps_per_second": 2.482,
|
| 843 |
+
"step": 187
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 22.11764705882353,
|
| 847 |
+
"grad_norm": 0.4103560447692871,
|
| 848 |
+
"learning_rate": 9.902502478779897e-06,
|
| 849 |
+
"loss": 0.9763,
|
| 850 |
+
"step": 188
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 22.352941176470587,
|
| 854 |
+
"grad_norm": 0.3923400938510895,
|
| 855 |
+
"learning_rate": 9.896703108827758e-06,
|
| 856 |
+
"loss": 0.9583,
|
| 857 |
+
"step": 190
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 22.58823529411765,
|
| 861 |
+
"grad_norm": 0.43239229917526245,
|
| 862 |
+
"learning_rate": 9.890738003669029e-06,
|
| 863 |
+
"loss": 0.9429,
|
| 864 |
+
"step": 192
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 22.823529411764707,
|
| 868 |
+
"grad_norm": 0.45320191979408264,
|
| 869 |
+
"learning_rate": 9.884607365200355e-06,
|
| 870 |
+
"loss": 0.95,
|
| 871 |
+
"step": 194
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 22.941176470588236,
|
| 875 |
+
"eval_loss": 0.9870390295982361,
|
| 876 |
+
"eval_runtime": 14.4888,
|
| 877 |
+
"eval_samples_per_second": 2.485,
|
| 878 |
+
"eval_steps_per_second": 2.485,
|
| 879 |
+
"step": 195
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 23.058823529411764,
|
| 883 |
+
"grad_norm": 0.48725754022598267,
|
| 884 |
+
"learning_rate": 9.878311400921072e-06,
|
| 885 |
+
"loss": 0.9369,
|
| 886 |
+
"step": 196
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 23.294117647058822,
|
| 890 |
+
"grad_norm": 0.4515678286552429,
|
| 891 |
+
"learning_rate": 9.871850323926178e-06,
|
| 892 |
+
"loss": 0.9978,
|
| 893 |
+
"step": 198
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 23.529411764705884,
|
| 897 |
+
"grad_norm": 0.44924628734588623,
|
| 898 |
+
"learning_rate": 9.86522435289912e-06,
|
| 899 |
+
"loss": 0.9126,
|
| 900 |
+
"step": 200
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 23.764705882352942,
|
| 904 |
+
"grad_norm": 0.4258614182472229,
|
| 905 |
+
"learning_rate": 9.858433712104403e-06,
|
| 906 |
+
"loss": 0.9716,
|
| 907 |
+
"step": 202
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 24.0,
|
| 911 |
+
"grad_norm": 0.46707141399383545,
|
| 912 |
+
"learning_rate": 9.851478631379982e-06,
|
| 913 |
+
"loss": 0.9467,
|
| 914 |
+
"step": 204
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 24.0,
|
| 918 |
+
"eval_loss": 0.980078399181366,
|
| 919 |
+
"eval_runtime": 14.4952,
|
| 920 |
+
"eval_samples_per_second": 2.484,
|
| 921 |
+
"eval_steps_per_second": 2.484,
|
| 922 |
+
"step": 204
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 24.235294117647058,
|
| 926 |
+
"grad_norm": 0.5438560843467712,
|
| 927 |
+
"learning_rate": 9.844359346129504e-06,
|
| 928 |
+
"loss": 0.9756,
|
| 929 |
+
"step": 206
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 24.470588235294116,
|
| 933 |
+
"grad_norm": 0.47229835391044617,
|
| 934 |
+
"learning_rate": 9.83707609731432e-06,
|
| 935 |
+
"loss": 0.9285,
|
| 936 |
+
"step": 208
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 24.705882352941178,
|
| 940 |
+
"grad_norm": 0.4578736126422882,
|
| 941 |
+
"learning_rate": 9.829629131445342e-06,
|
| 942 |
+
"loss": 0.9288,
|
| 943 |
+
"step": 210
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 24.941176470588236,
|
| 947 |
+
"grad_norm": 0.45513561367988586,
|
| 948 |
+
"learning_rate": 9.822018700574696e-06,
|
| 949 |
+
"loss": 0.9423,
|
| 950 |
+
"step": 212
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 24.941176470588236,
|
| 954 |
+
"eval_loss": 0.9737484455108643,
|
| 955 |
+
"eval_runtime": 14.4981,
|
| 956 |
+
"eval_samples_per_second": 2.483,
|
| 957 |
+
"eval_steps_per_second": 2.483,
|
| 958 |
+
"step": 212
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 25.176470588235293,
|
| 962 |
+
"grad_norm": 0.4210945963859558,
|
| 963 |
+
"learning_rate": 9.81424506228719e-06,
|
| 964 |
+
"loss": 0.9093,
|
| 965 |
+
"step": 214
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 25.41176470588235,
|
| 969 |
+
"grad_norm": 0.5261120796203613,
|
| 970 |
+
"learning_rate": 9.806308479691595e-06,
|
| 971 |
+
"loss": 0.9351,
|
| 972 |
+
"step": 216
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 25.647058823529413,
|
| 976 |
+
"grad_norm": 0.5155240893363953,
|
| 977 |
+
"learning_rate": 9.798209221411748e-06,
|
| 978 |
+
"loss": 0.9548,
|
| 979 |
+
"step": 218
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 25.88235294117647,
|
| 983 |
+
"grad_norm": 0.5193476676940918,
|
| 984 |
+
"learning_rate": 9.789947561577445e-06,
|
| 985 |
+
"loss": 0.937,
|
| 986 |
+
"step": 220
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 26.0,
|
| 990 |
+
"eval_loss": 0.9675381183624268,
|
| 991 |
+
"eval_runtime": 14.5154,
|
| 992 |
+
"eval_samples_per_second": 2.48,
|
| 993 |
+
"eval_steps_per_second": 2.48,
|
| 994 |
+
"step": 221
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 26.11764705882353,
|
| 998 |
+
"grad_norm": 0.4255479574203491,
|
| 999 |
+
"learning_rate": 9.781523779815178e-06,
|
| 1000 |
+
"loss": 0.9061,
|
| 1001 |
+
"step": 222
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 26.352941176470587,
|
| 1005 |
+
"grad_norm": 0.5112140774726868,
|
| 1006 |
+
"learning_rate": 9.77293816123866e-06,
|
| 1007 |
+
"loss": 0.9409,
|
| 1008 |
+
"step": 224
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 26.58823529411765,
|
| 1012 |
+
"grad_norm": 0.5464062094688416,
|
| 1013 |
+
"learning_rate": 9.764190996439181e-06,
|
| 1014 |
+
"loss": 0.9274,
|
| 1015 |
+
"step": 226
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 26.823529411764707,
|
| 1019 |
+
"grad_norm": 0.473880797624588,
|
| 1020 |
+
"learning_rate": 9.755282581475769e-06,
|
| 1021 |
+
"loss": 0.9035,
|
| 1022 |
+
"step": 228
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 26.941176470588236,
|
| 1026 |
+
"eval_loss": 0.9625819325447083,
|
| 1027 |
+
"eval_runtime": 14.5037,
|
| 1028 |
+
"eval_samples_per_second": 2.482,
|
| 1029 |
+
"eval_steps_per_second": 2.482,
|
| 1030 |
+
"step": 229
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 27.058823529411764,
|
| 1034 |
+
"grad_norm": 0.5459398627281189,
|
| 1035 |
+
"learning_rate": 9.74621321786517e-06,
|
| 1036 |
+
"loss": 0.9274,
|
| 1037 |
+
"step": 230
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 27.294117647058822,
|
| 1041 |
+
"grad_norm": 0.5077070593833923,
|
| 1042 |
+
"learning_rate": 9.736983212571646e-06,
|
| 1043 |
+
"loss": 0.911,
|
| 1044 |
+
"step": 232
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 27.529411764705884,
|
| 1048 |
+
"grad_norm": 0.510081946849823,
|
| 1049 |
+
"learning_rate": 9.727592877996585e-06,
|
| 1050 |
+
"loss": 0.8977,
|
| 1051 |
+
"step": 234
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 27.764705882352942,
|
| 1055 |
+
"grad_norm": 0.5523853898048401,
|
| 1056 |
+
"learning_rate": 9.718042531967918e-06,
|
| 1057 |
+
"loss": 0.9165,
|
| 1058 |
+
"step": 236
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 28.0,
|
| 1062 |
+
"grad_norm": 0.587631344795227,
|
| 1063 |
+
"learning_rate": 9.708332497729378e-06,
|
| 1064 |
+
"loss": 0.9074,
|
| 1065 |
+
"step": 238
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 28.0,
|
| 1069 |
+
"eval_loss": 0.9581733345985413,
|
| 1070 |
+
"eval_runtime": 14.489,
|
| 1071 |
+
"eval_samples_per_second": 2.485,
|
| 1072 |
+
"eval_steps_per_second": 2.485,
|
| 1073 |
+
"step": 238
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 28.235294117647058,
|
| 1077 |
+
"grad_norm": 0.5389757752418518,
|
| 1078 |
+
"learning_rate": 9.698463103929542e-06,
|
| 1079 |
+
"loss": 0.8634,
|
| 1080 |
+
"step": 240
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 28.470588235294116,
|
| 1084 |
+
"grad_norm": 0.5718576908111572,
|
| 1085 |
+
"learning_rate": 9.688434684610725e-06,
|
| 1086 |
+
"loss": 0.941,
|
| 1087 |
+
"step": 242
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 28.705882352941178,
|
| 1091 |
+
"grad_norm": 0.5238634943962097,
|
| 1092 |
+
"learning_rate": 9.678247579197658e-06,
|
| 1093 |
+
"loss": 0.9235,
|
| 1094 |
+
"step": 244
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 28.941176470588236,
|
| 1098 |
+
"grad_norm": 0.5517392754554749,
|
| 1099 |
+
"learning_rate": 9.667902132486009e-06,
|
| 1100 |
+
"loss": 0.8944,
|
| 1101 |
+
"step": 246
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 28.941176470588236,
|
| 1105 |
+
"eval_loss": 0.9534149169921875,
|
| 1106 |
+
"eval_runtime": 14.4915,
|
| 1107 |
+
"eval_samples_per_second": 2.484,
|
| 1108 |
+
"eval_steps_per_second": 2.484,
|
| 1109 |
+
"step": 246
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 29.176470588235293,
|
| 1113 |
+
"grad_norm": 0.6139256954193115,
|
| 1114 |
+
"learning_rate": 9.657398694630713e-06,
|
| 1115 |
+
"loss": 0.8854,
|
| 1116 |
+
"step": 248
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 29.41176470588235,
|
| 1120 |
+
"grad_norm": 0.5577236413955688,
|
| 1121 |
+
"learning_rate": 9.646737621134112e-06,
|
| 1122 |
+
"loss": 0.9253,
|
| 1123 |
+
"step": 250
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 29.647058823529413,
|
| 1127 |
+
"grad_norm": 0.638213038444519,
|
| 1128 |
+
"learning_rate": 9.635919272833938e-06,
|
| 1129 |
+
"loss": 0.9007,
|
| 1130 |
+
"step": 252
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 29.88235294117647,
|
| 1134 |
+
"grad_norm": 0.5903518795967102,
|
| 1135 |
+
"learning_rate": 9.62494401589108e-06,
|
| 1136 |
+
"loss": 0.8785,
|
| 1137 |
+
"step": 254
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 30.0,
|
| 1141 |
+
"eval_loss": 0.9493101835250854,
|
| 1142 |
+
"eval_runtime": 14.4864,
|
| 1143 |
+
"eval_samples_per_second": 2.485,
|
| 1144 |
+
"eval_steps_per_second": 2.485,
|
| 1145 |
+
"step": 255
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 30.11764705882353,
|
| 1149 |
+
"grad_norm": 0.6490582823753357,
|
| 1150 |
+
"learning_rate": 9.613812221777212e-06,
|
| 1151 |
+
"loss": 0.8782,
|
| 1152 |
+
"step": 256
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 30.352941176470587,
|
| 1156 |
+
"grad_norm": 0.5926429033279419,
|
| 1157 |
+
"learning_rate": 9.602524267262202e-06,
|
| 1158 |
+
"loss": 0.8326,
|
| 1159 |
+
"step": 258
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 30.58823529411765,
|
| 1163 |
+
"grad_norm": 0.6460428237915039,
|
| 1164 |
+
"learning_rate": 9.591080534401371e-06,
|
| 1165 |
+
"loss": 0.9207,
|
| 1166 |
+
"step": 260
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 30.823529411764707,
|
| 1170 |
+
"grad_norm": 0.59434574842453,
|
| 1171 |
+
"learning_rate": 9.579481410522556e-06,
|
| 1172 |
+
"loss": 0.8797,
|
| 1173 |
+
"step": 262
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 30.941176470588236,
|
| 1177 |
+
"eval_loss": 0.9451322555541992,
|
| 1178 |
+
"eval_runtime": 14.513,
|
| 1179 |
+
"eval_samples_per_second": 2.481,
|
| 1180 |
+
"eval_steps_per_second": 2.481,
|
| 1181 |
+
"step": 263
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 31.058823529411764,
|
| 1185 |
+
"grad_norm": 0.5878000855445862,
|
| 1186 |
+
"learning_rate": 9.567727288213005e-06,
|
| 1187 |
+
"loss": 0.864,
|
| 1188 |
+
"step": 264
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 31.294117647058822,
|
| 1192 |
+
"grad_norm": 0.5548307299613953,
|
| 1193 |
+
"learning_rate": 9.555818565306086e-06,
|
| 1194 |
+
"loss": 0.8766,
|
| 1195 |
+
"step": 266
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 31.529411764705884,
|
| 1199 |
+
"grad_norm": 0.6302839517593384,
|
| 1200 |
+
"learning_rate": 9.543755644867823e-06,
|
| 1201 |
+
"loss": 0.8702,
|
| 1202 |
+
"step": 268
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 31.764705882352942,
|
| 1206 |
+
"grad_norm": 0.6953015923500061,
|
| 1207 |
+
"learning_rate": 9.531538935183252e-06,
|
| 1208 |
+
"loss": 0.8888,
|
| 1209 |
+
"step": 270
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 32.0,
|
| 1213 |
+
"grad_norm": 0.8207558989524841,
|
| 1214 |
+
"learning_rate": 9.519168849742603e-06,
|
| 1215 |
+
"loss": 0.8764,
|
| 1216 |
+
"step": 272
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 32.0,
|
| 1220 |
+
"eval_loss": 0.9421913623809814,
|
| 1221 |
+
"eval_runtime": 14.4916,
|
| 1222 |
+
"eval_samples_per_second": 2.484,
|
| 1223 |
+
"eval_steps_per_second": 2.484,
|
| 1224 |
+
"step": 272
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 32.23529411764706,
|
| 1228 |
+
"grad_norm": 0.699301540851593,
|
| 1229 |
+
"learning_rate": 9.506645807227311e-06,
|
| 1230 |
+
"loss": 0.8338,
|
| 1231 |
+
"step": 274
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 32.470588235294116,
|
| 1235 |
+
"grad_norm": 0.7847335934638977,
|
| 1236 |
+
"learning_rate": 9.493970231495836e-06,
|
| 1237 |
+
"loss": 0.9099,
|
| 1238 |
+
"step": 276
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 32.705882352941174,
|
| 1242 |
+
"grad_norm": 0.7237752079963684,
|
| 1243 |
+
"learning_rate": 9.481142551569318e-06,
|
| 1244 |
+
"loss": 0.8442,
|
| 1245 |
+
"step": 278
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 32.94117647058823,
|
| 1249 |
+
"grad_norm": 0.7358724474906921,
|
| 1250 |
+
"learning_rate": 9.468163201617063e-06,
|
| 1251 |
+
"loss": 0.8903,
|
| 1252 |
+
"step": 280
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 32.94117647058823,
|
| 1256 |
+
"eval_loss": 0.9388971328735352,
|
| 1257 |
+
"eval_runtime": 14.4936,
|
| 1258 |
+
"eval_samples_per_second": 2.484,
|
| 1259 |
+
"eval_steps_per_second": 2.484,
|
| 1260 |
+
"step": 280
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 33.1764705882353,
|
| 1264 |
+
"grad_norm": 0.7364740371704102,
|
| 1265 |
+
"learning_rate": 9.45503262094184e-06,
|
| 1266 |
+
"loss": 0.8319,
|
| 1267 |
+
"step": 282
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 33.411764705882355,
|
| 1271 |
+
"grad_norm": 0.6712108254432678,
|
| 1272 |
+
"learning_rate": 9.441751253965022e-06,
|
| 1273 |
+
"loss": 0.8394,
|
| 1274 |
+
"step": 284
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 33.64705882352941,
|
| 1278 |
+
"grad_norm": 0.9382649660110474,
|
| 1279 |
+
"learning_rate": 9.428319550211531e-06,
|
| 1280 |
+
"loss": 0.8431,
|
| 1281 |
+
"step": 286
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 33.88235294117647,
|
| 1285 |
+
"grad_norm": 0.6663911938667297,
|
| 1286 |
+
"learning_rate": 9.414737964294636e-06,
|
| 1287 |
+
"loss": 0.8835,
|
| 1288 |
+
"step": 288
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 34.0,
|
| 1292 |
+
"eval_loss": 0.9377151131629944,
|
| 1293 |
+
"eval_runtime": 14.4983,
|
| 1294 |
+
"eval_samples_per_second": 2.483,
|
| 1295 |
+
"eval_steps_per_second": 2.483,
|
| 1296 |
+
"step": 289
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 34.11764705882353,
|
| 1300 |
+
"grad_norm": 0.9108101725578308,
|
| 1301 |
+
"learning_rate": 9.401006955900555e-06,
|
| 1302 |
+
"loss": 0.8886,
|
| 1303 |
+
"step": 290
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 34.35294117647059,
|
| 1307 |
+
"grad_norm": 0.6387782096862793,
|
| 1308 |
+
"learning_rate": 9.38712698977291e-06,
|
| 1309 |
+
"loss": 0.8418,
|
| 1310 |
+
"step": 292
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 34.588235294117645,
|
| 1314 |
+
"grad_norm": 0.7615544199943542,
|
| 1315 |
+
"learning_rate": 9.37309853569698e-06,
|
| 1316 |
+
"loss": 0.833,
|
| 1317 |
+
"step": 294
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 34.8235294117647,
|
| 1321 |
+
"grad_norm": 0.7540785074234009,
|
| 1322 |
+
"learning_rate": 9.358922068483813e-06,
|
| 1323 |
+
"loss": 0.8452,
|
| 1324 |
+
"step": 296
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 34.94117647058823,
|
| 1328 |
+
"eval_loss": 0.9331609010696411,
|
| 1329 |
+
"eval_runtime": 14.4883,
|
| 1330 |
+
"eval_samples_per_second": 2.485,
|
| 1331 |
+
"eval_steps_per_second": 2.485,
|
| 1332 |
+
"step": 297
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 35.05882352941177,
|
| 1336 |
+
"grad_norm": 0.9316972494125366,
|
| 1337 |
+
"learning_rate": 9.344598067954151e-06,
|
| 1338 |
+
"loss": 0.8427,
|
| 1339 |
+
"step": 298
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 35.294117647058826,
|
| 1343 |
+
"grad_norm": 0.8066386580467224,
|
| 1344 |
+
"learning_rate": 9.330127018922195e-06,
|
| 1345 |
+
"loss": 0.8404,
|
| 1346 |
+
"step": 300
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 35.529411764705884,
|
| 1350 |
+
"grad_norm": 0.7952571511268616,
|
| 1351 |
+
"learning_rate": 9.315509411179182e-06,
|
| 1352 |
+
"loss": 0.7953,
|
| 1353 |
+
"step": 302
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 35.76470588235294,
|
| 1357 |
+
"grad_norm": 0.7538934946060181,
|
| 1358 |
+
"learning_rate": 9.30074573947683e-06,
|
| 1359 |
+
"loss": 0.8435,
|
| 1360 |
+
"step": 304
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 36.0,
|
| 1364 |
+
"grad_norm": 0.7602026462554932,
|
| 1365 |
+
"learning_rate": 9.285836503510562e-06,
|
| 1366 |
+
"loss": 0.8777,
|
| 1367 |
+
"step": 306
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 36.0,
|
| 1371 |
+
"eval_loss": 0.9272398352622986,
|
| 1372 |
+
"eval_runtime": 14.4824,
|
| 1373 |
+
"eval_samples_per_second": 2.486,
|
| 1374 |
+
"eval_steps_per_second": 2.486,
|
| 1375 |
+
"step": 306
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 36.23529411764706,
|
| 1379 |
+
"grad_norm": 0.7411664128303528,
|
| 1380 |
+
"learning_rate": 9.27078220790263e-06,
|
| 1381 |
+
"loss": 0.8382,
|
| 1382 |
+
"step": 308
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 36.470588235294116,
|
| 1386 |
+
"grad_norm": 0.7392826676368713,
|
| 1387 |
+
"learning_rate": 9.255583362184998e-06,
|
| 1388 |
+
"loss": 0.8206,
|
| 1389 |
+
"step": 310
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 36.705882352941174,
|
| 1393 |
+
"grad_norm": 0.8133856058120728,
|
| 1394 |
+
"learning_rate": 9.24024048078213e-06,
|
| 1395 |
+
"loss": 0.8384,
|
| 1396 |
+
"step": 312
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 36.94117647058823,
|
| 1400 |
+
"grad_norm": 0.8298240900039673,
|
| 1401 |
+
"learning_rate": 9.224754082993553e-06,
|
| 1402 |
+
"loss": 0.8101,
|
| 1403 |
+
"step": 314
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 36.94117647058823,
|
| 1407 |
+
"eval_loss": 0.9257263541221619,
|
| 1408 |
+
"eval_runtime": 14.5049,
|
| 1409 |
+
"eval_samples_per_second": 2.482,
|
| 1410 |
+
"eval_steps_per_second": 2.482,
|
| 1411 |
+
"step": 314
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 37.1764705882353,
|
| 1415 |
+
"grad_norm": 0.756746768951416,
|
| 1416 |
+
"learning_rate": 9.209124692976287e-06,
|
| 1417 |
+
"loss": 0.8068,
|
| 1418 |
+
"step": 316
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 37.411764705882355,
|
| 1422 |
+
"grad_norm": 0.7719102501869202,
|
| 1423 |
+
"learning_rate": 9.193352839727122e-06,
|
| 1424 |
+
"loss": 0.8067,
|
| 1425 |
+
"step": 318
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 37.64705882352941,
|
| 1429 |
+
"grad_norm": 0.7538597583770752,
|
| 1430 |
+
"learning_rate": 9.177439057064684e-06,
|
| 1431 |
+
"loss": 0.821,
|
| 1432 |
+
"step": 320
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 37.88235294117647,
|
| 1436 |
+
"grad_norm": 0.8648045063018799,
|
| 1437 |
+
"learning_rate": 9.16138388361139e-06,
|
| 1438 |
+
"loss": 0.8526,
|
| 1439 |
+
"step": 322
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 38.0,
|
| 1443 |
+
"eval_loss": 0.9229225516319275,
|
| 1444 |
+
"eval_runtime": 14.498,
|
| 1445 |
+
"eval_samples_per_second": 2.483,
|
| 1446 |
+
"eval_steps_per_second": 2.483,
|
| 1447 |
+
"step": 323
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 38.11764705882353,
|
| 1451 |
+
"grad_norm": 0.7806123495101929,
|
| 1452 |
+
"learning_rate": 9.145187862775208e-06,
|
| 1453 |
+
"loss": 0.8019,
|
| 1454 |
+
"step": 324
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 38.35294117647059,
|
| 1458 |
+
"grad_norm": 0.8803067803382874,
|
| 1459 |
+
"learning_rate": 9.128851542731271e-06,
|
| 1460 |
+
"loss": 0.7752,
|
| 1461 |
+
"step": 326
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 38.588235294117645,
|
| 1465 |
+
"grad_norm": 0.9083964824676514,
|
| 1466 |
+
"learning_rate": 9.112375476403313e-06,
|
| 1467 |
+
"loss": 0.8176,
|
| 1468 |
+
"step": 328
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 38.8235294117647,
|
| 1472 |
+
"grad_norm": 0.8712770938873291,
|
| 1473 |
+
"learning_rate": 9.09576022144496e-06,
|
| 1474 |
+
"loss": 0.8228,
|
| 1475 |
+
"step": 330
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 38.94117647058823,
|
| 1479 |
+
"eval_loss": 0.9196635484695435,
|
| 1480 |
+
"eval_runtime": 14.5068,
|
| 1481 |
+
"eval_samples_per_second": 2.482,
|
| 1482 |
+
"eval_steps_per_second": 2.482,
|
| 1483 |
+
"step": 331
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 39.05882352941177,
|
| 1487 |
+
"grad_norm": 0.8734506964683533,
|
| 1488 |
+
"learning_rate": 9.079006340220862e-06,
|
| 1489 |
+
"loss": 0.8059,
|
| 1490 |
+
"step": 332
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 39.294117647058826,
|
| 1494 |
+
"grad_norm": 0.8067296147346497,
|
| 1495 |
+
"learning_rate": 9.062114399787648e-06,
|
| 1496 |
+
"loss": 0.7936,
|
| 1497 |
+
"step": 334
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 39.529411764705884,
|
| 1501 |
+
"grad_norm": 0.825115978717804,
|
| 1502 |
+
"learning_rate": 9.045084971874738e-06,
|
| 1503 |
+
"loss": 0.837,
|
| 1504 |
+
"step": 336
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 39.76470588235294,
|
| 1508 |
+
"grad_norm": 0.9862772226333618,
|
| 1509 |
+
"learning_rate": 9.027918632864998e-06,
|
| 1510 |
+
"loss": 0.756,
|
| 1511 |
+
"step": 338
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 40.0,
|
| 1515 |
+
"grad_norm": 0.9583683609962463,
|
| 1516 |
+
"learning_rate": 9.01061596377522e-06,
|
| 1517 |
+
"loss": 0.8066,
|
| 1518 |
+
"step": 340
|
| 1519 |
+
},
|
| 1520 |
+
{
|
| 1521 |
+
"epoch": 40.0,
|
| 1522 |
+
"eval_loss": 0.9175994396209717,
|
| 1523 |
+
"eval_runtime": 14.4912,
|
| 1524 |
+
"eval_samples_per_second": 2.484,
|
| 1525 |
+
"eval_steps_per_second": 2.484,
|
| 1526 |
+
"step": 340
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 40.23529411764706,
|
| 1530 |
+
"grad_norm": 0.9128953814506531,
|
| 1531 |
+
"learning_rate": 8.993177550236464e-06,
|
| 1532 |
+
"loss": 0.8042,
|
| 1533 |
+
"step": 342
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 40.470588235294116,
|
| 1537 |
+
"grad_norm": 0.9705334901809692,
|
| 1538 |
+
"learning_rate": 8.97560398247424e-06,
|
| 1539 |
+
"loss": 0.8072,
|
| 1540 |
+
"step": 344
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 40.705882352941174,
|
| 1544 |
+
"grad_norm": 0.9346606731414795,
|
| 1545 |
+
"learning_rate": 8.957895855288517e-06,
|
| 1546 |
+
"loss": 0.7423,
|
| 1547 |
+
"step": 346
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 40.94117647058823,
|
| 1551 |
+
"grad_norm": 1.2344577312469482,
|
| 1552 |
+
"learning_rate": 8.94005376803361e-06,
|
| 1553 |
+
"loss": 0.7701,
|
| 1554 |
+
"step": 348
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 40.94117647058823,
|
| 1558 |
+
"eval_loss": 0.9198606014251709,
|
| 1559 |
+
"eval_runtime": 14.4977,
|
| 1560 |
+
"eval_samples_per_second": 2.483,
|
| 1561 |
+
"eval_steps_per_second": 2.483,
|
| 1562 |
+
"step": 348
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 41.1764705882353,
|
| 1566 |
+
"grad_norm": 0.9365864992141724,
|
| 1567 |
+
"learning_rate": 8.92207832459788e-06,
|
| 1568 |
+
"loss": 0.7861,
|
| 1569 |
+
"step": 350
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 41.411764705882355,
|
| 1573 |
+
"grad_norm": 0.8833426833152771,
|
| 1574 |
+
"learning_rate": 8.903970133383297e-06,
|
| 1575 |
+
"loss": 0.7686,
|
| 1576 |
+
"step": 352
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 41.64705882352941,
|
| 1580 |
+
"grad_norm": 0.9196489453315735,
|
| 1581 |
+
"learning_rate": 8.885729807284855e-06,
|
| 1582 |
+
"loss": 0.773,
|
| 1583 |
+
"step": 354
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 41.88235294117647,
|
| 1587 |
+
"grad_norm": 1.0096192359924316,
|
| 1588 |
+
"learning_rate": 8.867357963669821e-06,
|
| 1589 |
+
"loss": 0.8132,
|
| 1590 |
+
"step": 356
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 42.0,
|
| 1594 |
+
"eval_loss": 0.916193962097168,
|
| 1595 |
+
"eval_runtime": 14.4918,
|
| 1596 |
+
"eval_samples_per_second": 2.484,
|
| 1597 |
+
"eval_steps_per_second": 2.484,
|
| 1598 |
+
"step": 357
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 42.11764705882353,
|
| 1602 |
+
"grad_norm": 0.9623170495033264,
|
| 1603 |
+
"learning_rate": 8.84885522435684e-06,
|
| 1604 |
+
"loss": 0.7446,
|
| 1605 |
+
"step": 358
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 42.35294117647059,
|
| 1609 |
+
"grad_norm": 0.9639574885368347,
|
| 1610 |
+
"learning_rate": 8.83022221559489e-06,
|
| 1611 |
+
"loss": 0.8012,
|
| 1612 |
+
"step": 360
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 42.588235294117645,
|
| 1616 |
+
"grad_norm": 1.0695879459381104,
|
| 1617 |
+
"learning_rate": 8.811459568042092e-06,
|
| 1618 |
+
"loss": 0.7134,
|
| 1619 |
+
"step": 362
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 42.8235294117647,
|
| 1623 |
+
"grad_norm": 1.0046355724334717,
|
| 1624 |
+
"learning_rate": 8.792567916744346e-06,
|
| 1625 |
+
"loss": 0.7804,
|
| 1626 |
+
"step": 364
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 42.94117647058823,
|
| 1630 |
+
"eval_loss": 0.9103953838348389,
|
| 1631 |
+
"eval_runtime": 14.5212,
|
| 1632 |
+
"eval_samples_per_second": 2.479,
|
| 1633 |
+
"eval_steps_per_second": 2.479,
|
| 1634 |
+
"step": 365
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 43.05882352941177,
|
| 1638 |
+
"grad_norm": 0.9536579251289368,
|
| 1639 |
+
"learning_rate": 8.773547901113862e-06,
|
| 1640 |
+
"loss": 0.7675,
|
| 1641 |
+
"step": 366
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 43.294117647058826,
|
| 1645 |
+
"grad_norm": 0.9319316148757935,
|
| 1646 |
+
"learning_rate": 8.754400164907496e-06,
|
| 1647 |
+
"loss": 0.7311,
|
| 1648 |
+
"step": 368
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 43.529411764705884,
|
| 1652 |
+
"grad_norm": 1.153613805770874,
|
| 1653 |
+
"learning_rate": 8.735125356204982e-06,
|
| 1654 |
+
"loss": 0.7777,
|
| 1655 |
+
"step": 370
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 43.76470588235294,
|
| 1659 |
+
"grad_norm": 1.0393705368041992,
|
| 1660 |
+
"learning_rate": 8.715724127386971e-06,
|
| 1661 |
+
"loss": 0.7322,
|
| 1662 |
+
"step": 372
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 44.0,
|
| 1666 |
+
"grad_norm": 0.9913296103477478,
|
| 1667 |
+
"learning_rate": 8.69619713511298e-06,
|
| 1668 |
+
"loss": 0.7508,
|
| 1669 |
+
"step": 374
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 44.0,
|
| 1673 |
+
"eval_loss": 0.908278226852417,
|
| 1674 |
+
"eval_runtime": 14.4883,
|
| 1675 |
+
"eval_samples_per_second": 2.485,
|
| 1676 |
+
"eval_steps_per_second": 2.485,
|
| 1677 |
+
"step": 374
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 44.23529411764706,
|
| 1681 |
+
"grad_norm": 0.9832679033279419,
|
| 1682 |
+
"learning_rate": 8.676545040299145e-06,
|
| 1683 |
+
"loss": 0.7355,
|
| 1684 |
+
"step": 376
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 44.470588235294116,
|
| 1688 |
+
"grad_norm": 1.2460023164749146,
|
| 1689 |
+
"learning_rate": 8.656768508095853e-06,
|
| 1690 |
+
"loss": 0.7238,
|
| 1691 |
+
"step": 378
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 44.705882352941174,
|
| 1695 |
+
"grad_norm": 0.964077889919281,
|
| 1696 |
+
"learning_rate": 8.636868207865244e-06,
|
| 1697 |
+
"loss": 0.7564,
|
| 1698 |
+
"step": 380
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 44.94117647058823,
|
| 1702 |
+
"grad_norm": 0.9776307344436646,
|
| 1703 |
+
"learning_rate": 8.61684481315854e-06,
|
| 1704 |
+
"loss": 0.7192,
|
| 1705 |
+
"step": 382
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 44.94117647058823,
|
| 1709 |
+
"eval_loss": 0.9051859378814697,
|
| 1710 |
+
"eval_runtime": 14.5008,
|
| 1711 |
+
"eval_samples_per_second": 2.483,
|
| 1712 |
+
"eval_steps_per_second": 2.483,
|
| 1713 |
+
"step": 382
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 45.1764705882353,
|
| 1717 |
+
"grad_norm": 0.9675242304801941,
|
| 1718 |
+
"learning_rate": 8.596699001693257e-06,
|
| 1719 |
+
"loss": 0.754,
|
| 1720 |
+
"step": 384
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 45.411764705882355,
|
| 1724 |
+
"grad_norm": 1.1183017492294312,
|
| 1725 |
+
"learning_rate": 8.576431455330258e-06,
|
| 1726 |
+
"loss": 0.6805,
|
| 1727 |
+
"step": 386
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 45.64705882352941,
|
| 1731 |
+
"grad_norm": 0.9894379377365112,
|
| 1732 |
+
"learning_rate": 8.556042860050686e-06,
|
| 1733 |
+
"loss": 0.7691,
|
| 1734 |
+
"step": 388
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 45.88235294117647,
|
| 1738 |
+
"grad_norm": 0.8682206869125366,
|
| 1739 |
+
"learning_rate": 8.535533905932739e-06,
|
| 1740 |
+
"loss": 0.7633,
|
| 1741 |
+
"step": 390
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 46.0,
|
| 1745 |
+
"eval_loss": 0.9047689437866211,
|
| 1746 |
+
"eval_runtime": 14.5123,
|
| 1747 |
+
"eval_samples_per_second": 2.481,
|
| 1748 |
+
"eval_steps_per_second": 2.481,
|
| 1749 |
+
"step": 391
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 46.11764705882353,
|
| 1753 |
+
"grad_norm": 0.9697660803794861,
|
| 1754 |
+
"learning_rate": 8.51490528712831e-06,
|
| 1755 |
+
"loss": 0.7131,
|
| 1756 |
+
"step": 392
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 46.35294117647059,
|
| 1760 |
+
"grad_norm": 0.9910651445388794,
|
| 1761 |
+
"learning_rate": 8.4941577018395e-06,
|
| 1762 |
+
"loss": 0.7183,
|
| 1763 |
+
"step": 394
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 46.588235294117645,
|
| 1767 |
+
"grad_norm": 1.670358419418335,
|
| 1768 |
+
"learning_rate": 8.473291852294986e-06,
|
| 1769 |
+
"loss": 0.6982,
|
| 1770 |
+
"step": 396
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 46.8235294117647,
|
| 1774 |
+
"grad_norm": 1.1389248371124268,
|
| 1775 |
+
"learning_rate": 8.452308444726249e-06,
|
| 1776 |
+
"loss": 0.7534,
|
| 1777 |
+
"step": 398
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"epoch": 46.94117647058823,
|
| 1781 |
+
"eval_loss": 0.9052470326423645,
|
| 1782 |
+
"eval_runtime": 14.5068,
|
| 1783 |
+
"eval_samples_per_second": 2.482,
|
| 1784 |
+
"eval_steps_per_second": 2.482,
|
| 1785 |
+
"step": 399
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 47.05882352941177,
|
| 1789 |
+
"grad_norm": 0.963392972946167,
|
| 1790 |
+
"learning_rate": 8.43120818934367e-06,
|
| 1791 |
+
"loss": 0.7282,
|
| 1792 |
+
"step": 400
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 47.294117647058826,
|
| 1796 |
+
"grad_norm": 1.0950130224227905,
|
| 1797 |
+
"learning_rate": 8.409991800312493e-06,
|
| 1798 |
+
"loss": 0.7273,
|
| 1799 |
+
"step": 402
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 47.529411764705884,
|
| 1803 |
+
"grad_norm": 1.0003873109817505,
|
| 1804 |
+
"learning_rate": 8.388659995728662e-06,
|
| 1805 |
+
"loss": 0.7113,
|
| 1806 |
+
"step": 404
|
| 1807 |
+
},
|
| 1808 |
+
{
|
| 1809 |
+
"epoch": 47.76470588235294,
|
| 1810 |
+
"grad_norm": 1.1168714761734009,
|
| 1811 |
+
"learning_rate": 8.367213497594501e-06,
|
| 1812 |
+
"loss": 0.7019,
|
| 1813 |
+
"step": 406
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 48.0,
|
| 1817 |
+
"grad_norm": 1.1041070222854614,
|
| 1818 |
+
"learning_rate": 8.345653031794292e-06,
|
| 1819 |
+
"loss": 0.666,
|
| 1820 |
+
"step": 408
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 48.0,
|
| 1824 |
+
"eval_loss": 0.915124237537384,
|
| 1825 |
+
"eval_runtime": 14.4863,
|
| 1826 |
+
"eval_samples_per_second": 2.485,
|
| 1827 |
+
"eval_steps_per_second": 2.485,
|
| 1828 |
+
"step": 408
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 48.23529411764706,
|
| 1832 |
+
"grad_norm": 0.8982572555541992,
|
| 1833 |
+
"learning_rate": 8.323979328069689e-06,
|
| 1834 |
+
"loss": 0.6762,
|
| 1835 |
+
"step": 410
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 48.470588235294116,
|
| 1839 |
+
"grad_norm": 1.0614545345306396,
|
| 1840 |
+
"learning_rate": 8.302193119995038e-06,
|
| 1841 |
+
"loss": 0.6779,
|
| 1842 |
+
"step": 412
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 48.705882352941174,
|
| 1846 |
+
"grad_norm": 0.9689596891403198,
|
| 1847 |
+
"learning_rate": 8.280295144952537e-06,
|
| 1848 |
+
"loss": 0.7313,
|
| 1849 |
+
"step": 414
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 48.94117647058823,
|
| 1853 |
+
"grad_norm": 0.9231712818145752,
|
| 1854 |
+
"learning_rate": 8.258286144107277e-06,
|
| 1855 |
+
"loss": 0.7298,
|
| 1856 |
+
"step": 416
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 48.94117647058823,
|
| 1860 |
+
"eval_loss": 0.9143383502960205,
|
| 1861 |
+
"eval_runtime": 14.4951,
|
| 1862 |
+
"eval_samples_per_second": 2.484,
|
| 1863 |
+
"eval_steps_per_second": 2.484,
|
| 1864 |
+
"step": 416
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 49.1764705882353,
|
| 1868 |
+
"grad_norm": 1.089706301689148,
|
| 1869 |
+
"learning_rate": 8.236166862382163e-06,
|
| 1870 |
+
"loss": 0.6328,
|
| 1871 |
+
"step": 418
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 49.411764705882355,
|
| 1875 |
+
"grad_norm": 1.1155279874801636,
|
| 1876 |
+
"learning_rate": 8.213938048432697e-06,
|
| 1877 |
+
"loss": 0.6914,
|
| 1878 |
+
"step": 420
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 49.64705882352941,
|
| 1882 |
+
"grad_norm": 0.9705954194068909,
|
| 1883 |
+
"learning_rate": 8.191600454621642e-06,
|
| 1884 |
+
"loss": 0.7003,
|
| 1885 |
+
"step": 422
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 49.88235294117647,
|
| 1889 |
+
"grad_norm": 1.0692474842071533,
|
| 1890 |
+
"learning_rate": 8.16915483699355e-06,
|
| 1891 |
+
"loss": 0.6815,
|
| 1892 |
+
"step": 424
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 50.0,
|
| 1896 |
+
"eval_loss": 0.9157248735427856,
|
| 1897 |
+
"eval_runtime": 14.4868,
|
| 1898 |
+
"eval_samples_per_second": 2.485,
|
| 1899 |
+
"eval_steps_per_second": 2.485,
|
| 1900 |
+
"step": 425
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 50.11764705882353,
|
| 1904 |
+
"grad_norm": 1.0241312980651855,
|
| 1905 |
+
"learning_rate": 8.146601955249187e-06,
|
| 1906 |
+
"loss": 0.6999,
|
| 1907 |
+
"step": 426
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 50.35294117647059,
|
| 1911 |
+
"grad_norm": 1.097902774810791,
|
| 1912 |
+
"learning_rate": 8.123942572719801e-06,
|
| 1913 |
+
"loss": 0.6549,
|
| 1914 |
+
"step": 428
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 50.588235294117645,
|
| 1918 |
+
"grad_norm": 1.1850600242614746,
|
| 1919 |
+
"learning_rate": 8.101177456341301e-06,
|
| 1920 |
+
"loss": 0.6675,
|
| 1921 |
+
"step": 430
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 50.8235294117647,
|
| 1925 |
+
"grad_norm": 1.080406904220581,
|
| 1926 |
+
"learning_rate": 8.078307376628292e-06,
|
| 1927 |
+
"loss": 0.6845,
|
| 1928 |
+
"step": 432
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 50.94117647058823,
|
| 1932 |
+
"eval_loss": 0.9170003533363342,
|
| 1933 |
+
"eval_runtime": 14.5105,
|
| 1934 |
+
"eval_samples_per_second": 2.481,
|
| 1935 |
+
"eval_steps_per_second": 2.481,
|
| 1936 |
+
"step": 433
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 51.05882352941177,
|
| 1940 |
+
"grad_norm": 1.128921389579773,
|
| 1941 |
+
"learning_rate": 8.055333107648e-06,
|
| 1942 |
+
"loss": 0.7022,
|
| 1943 |
+
"step": 434
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 51.294117647058826,
|
| 1947 |
+
"grad_norm": 1.2979844808578491,
|
| 1948 |
+
"learning_rate": 8.032255426994069e-06,
|
| 1949 |
+
"loss": 0.6501,
|
| 1950 |
+
"step": 436
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 51.529411764705884,
|
| 1954 |
+
"grad_norm": 1.0280219316482544,
|
| 1955 |
+
"learning_rate": 8.009075115760243e-06,
|
| 1956 |
+
"loss": 0.6824,
|
| 1957 |
+
"step": 438
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 51.76470588235294,
|
| 1961 |
+
"grad_norm": 1.2425734996795654,
|
| 1962 |
+
"learning_rate": 7.985792958513932e-06,
|
| 1963 |
+
"loss": 0.6787,
|
| 1964 |
+
"step": 440
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 52.0,
|
| 1968 |
+
"grad_norm": 1.114044427871704,
|
| 1969 |
+
"learning_rate": 7.962409743269654e-06,
|
| 1970 |
+
"loss": 0.6524,
|
| 1971 |
+
"step": 442
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 52.0,
|
| 1975 |
+
"eval_loss": 0.9215981960296631,
|
| 1976 |
+
"eval_runtime": 14.5021,
|
| 1977 |
+
"eval_samples_per_second": 2.482,
|
| 1978 |
+
"eval_steps_per_second": 2.482,
|
| 1979 |
+
"step": 442
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 52.23529411764706,
|
| 1983 |
+
"grad_norm": 1.1072558164596558,
|
| 1984 |
+
"learning_rate": 7.938926261462366e-06,
|
| 1985 |
+
"loss": 0.6602,
|
| 1986 |
+
"step": 444
|
| 1987 |
+
},
|
| 1988 |
+
{
|
| 1989 |
+
"epoch": 52.470588235294116,
|
| 1990 |
+
"grad_norm": 1.1433347463607788,
|
| 1991 |
+
"learning_rate": 7.915343307920674e-06,
|
| 1992 |
+
"loss": 0.6649,
|
| 1993 |
+
"step": 446
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 52.705882352941174,
|
| 1997 |
+
"grad_norm": 1.1459075212478638,
|
| 1998 |
+
"learning_rate": 7.891661680839932e-06,
|
| 1999 |
+
"loss": 0.6407,
|
| 2000 |
+
"step": 448
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 52.94117647058823,
|
| 2004 |
+
"grad_norm": 1.146519422531128,
|
| 2005 |
+
"learning_rate": 7.86788218175523e-06,
|
| 2006 |
+
"loss": 0.6397,
|
| 2007 |
+
"step": 450
|
| 2008 |
+
},
|
| 2009 |
+
{
|
| 2010 |
+
"epoch": 52.94117647058823,
|
| 2011 |
+
"eval_loss": 0.9228353500366211,
|
| 2012 |
+
"eval_runtime": 14.5275,
|
| 2013 |
+
"eval_samples_per_second": 2.478,
|
| 2014 |
+
"eval_steps_per_second": 2.478,
|
| 2015 |
+
"step": 450
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 52.94117647058823,
|
| 2019 |
+
"step": 450,
|
| 2020 |
+
"total_flos": 1.1473460706410496e+17,
|
| 2021 |
+
"train_loss": 1.1512780372301739,
|
| 2022 |
+
"train_runtime": 8506.7232,
|
| 2023 |
+
"train_samples_per_second": 2.398,
|
| 2024 |
+
"train_steps_per_second": 0.141
|
| 2025 |
+
}
|
| 2026 |
+
],
|
| 2027 |
+
"logging_steps": 2,
|
| 2028 |
+
"max_steps": 1200,
|
| 2029 |
+
"num_input_tokens_seen": 0,
|
| 2030 |
+
"num_train_epochs": 150,
|
| 2031 |
+
"save_steps": 25,
|
| 2032 |
+
"stateful_callbacks": {
|
| 2033 |
+
"EarlyStoppingCallback": {
|
| 2034 |
+
"args": {
|
| 2035 |
+
"early_stopping_patience": 7,
|
| 2036 |
+
"early_stopping_threshold": 0.0
|
| 2037 |
+
},
|
| 2038 |
+
"attributes": {
|
| 2039 |
+
"early_stopping_patience_counter": 0
|
| 2040 |
+
}
|
| 2041 |
+
},
|
| 2042 |
+
"TrainerControl": {
|
| 2043 |
+
"args": {
|
| 2044 |
+
"should_epoch_stop": false,
|
| 2045 |
+
"should_evaluate": false,
|
| 2046 |
+
"should_log": false,
|
| 2047 |
+
"should_save": true,
|
| 2048 |
+
"should_training_stop": true
|
| 2049 |
+
},
|
| 2050 |
+
"attributes": {}
|
| 2051 |
+
}
|
| 2052 |
+
},
|
| 2053 |
+
"total_flos": 1.1473460706410496e+17,
|
| 2054 |
+
"train_batch_size": 1,
|
| 2055 |
+
"trial_name": null,
|
| 2056 |
+
"trial_params": null
|
| 2057 |
+
}
|