Commit
•
c7de38b
1
Parent(s):
d867a42
Model save
Browse files- README.md +76 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- train_results.json +9 -0
- trainer_state.json +381 -0
README.md
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: alignment-handbook/zephyr-7b-sft-full
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: zephyr-7b-math-case-7
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# zephyr-7b-math-case-7
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.0013
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 1e-05
|
| 43 |
+
- train_batch_size: 8
|
| 44 |
+
- eval_batch_size: 8
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 4
|
| 48 |
+
- total_train_batch_size: 32
|
| 49 |
+
- total_eval_batch_size: 32
|
| 50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
+
- lr_scheduler_type: cosine
|
| 52 |
+
- lr_scheduler_warmup_ratio: 0.03
|
| 53 |
+
- num_epochs: 10
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 58 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 59 |
+
| 0.8719 | 1.0 | 18 | 0.5943 |
|
| 60 |
+
| 0.4801 | 2.0 | 36 | 0.2952 |
|
| 61 |
+
| 0.2452 | 3.0 | 54 | 0.1275 |
|
| 62 |
+
| 0.098 | 4.0 | 72 | 0.0448 |
|
| 63 |
+
| 0.0439 | 5.0 | 90 | 0.0259 |
|
| 64 |
+
| 0.0306 | 6.0 | 108 | 0.0149 |
|
| 65 |
+
| 0.017 | 7.0 | 126 | 0.0068 |
|
| 66 |
+
| 0.006 | 8.0 | 144 | 0.0025 |
|
| 67 |
+
| 0.0025 | 9.0 | 162 | 0.0014 |
|
| 68 |
+
| 0.0012 | 10.0 | 180 | 0.0013 |
|
| 69 |
+
|
| 70 |
+
|
| 71 |
+
### Framework versions
|
| 72 |
+
|
| 73 |
+
- Transformers 4.44.2
|
| 74 |
+
- Pytorch 2.4.1+cu121
|
| 75 |
+
- Datasets 3.0.0
|
| 76 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 37688338022400.0,
|
| 4 |
+
"train_loss": 0.18709226546602117,
|
| 5 |
+
"train_runtime": 2578.874,
|
| 6 |
+
"train_samples": 3999,
|
| 7 |
+
"train_samples_per_second": 2.148,
|
| 8 |
+
"train_steps_per_second": 0.07
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.44.2"
|
| 6 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 37688338022400.0,
|
| 4 |
+
"train_loss": 0.18709226546602117,
|
| 5 |
+
"train_runtime": 2578.874,
|
| 6 |
+
"train_samples": 3999,
|
| 7 |
+
"train_samples_per_second": 2.148,
|
| 8 |
+
"train_steps_per_second": 0.07
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,381 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 180,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05555555555555555,
|
| 13 |
+
"grad_norm": 7.355470178103522,
|
| 14 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 15 |
+
"loss": 0.9936,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.2777777777777778,
|
| 20 |
+
"grad_norm": 9.150698239786399,
|
| 21 |
+
"learning_rate": 8.333333333333334e-06,
|
| 22 |
+
"loss": 0.9935,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.5555555555555556,
|
| 27 |
+
"grad_norm": 3.0603212678780776,
|
| 28 |
+
"learning_rate": 9.986966157589751e-06,
|
| 29 |
+
"loss": 0.918,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.8333333333333334,
|
| 34 |
+
"grad_norm": 2.5047426466184493,
|
| 35 |
+
"learning_rate": 9.934132612707631e-06,
|
| 36 |
+
"loss": 0.8719,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_loss": 0.5942633748054504,
|
| 42 |
+
"eval_runtime": 48.0382,
|
| 43 |
+
"eval_samples_per_second": 11.532,
|
| 44 |
+
"eval_steps_per_second": 0.375,
|
| 45 |
+
"step": 18
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.1111111111111112,
|
| 49 |
+
"grad_norm": 2.854633191400044,
|
| 50 |
+
"learning_rate": 9.841114703012817e-06,
|
| 51 |
+
"loss": 0.711,
|
| 52 |
+
"step": 20
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.3888888888888888,
|
| 56 |
+
"grad_norm": 3.0542285095627815,
|
| 57 |
+
"learning_rate": 9.708669981232542e-06,
|
| 58 |
+
"loss": 0.5253,
|
| 59 |
+
"step": 25
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.6666666666666665,
|
| 63 |
+
"grad_norm": 2.9849351058447633,
|
| 64 |
+
"learning_rate": 9.537877098354787e-06,
|
| 65 |
+
"loss": 0.5337,
|
| 66 |
+
"step": 30
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.9444444444444444,
|
| 70 |
+
"grad_norm": 2.6887676315185307,
|
| 71 |
+
"learning_rate": 9.330127018922195e-06,
|
| 72 |
+
"loss": 0.4801,
|
| 73 |
+
"step": 35
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 2.0,
|
| 77 |
+
"eval_loss": 0.2952410876750946,
|
| 78 |
+
"eval_runtime": 47.9311,
|
| 79 |
+
"eval_samples_per_second": 11.558,
|
| 80 |
+
"eval_steps_per_second": 0.376,
|
| 81 |
+
"step": 36
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.2222222222222223,
|
| 85 |
+
"grad_norm": 5.591116487375825,
|
| 86 |
+
"learning_rate": 9.08711169279446e-06,
|
| 87 |
+
"loss": 0.2937,
|
| 88 |
+
"step": 40
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.5,
|
| 92 |
+
"grad_norm": 2.583058931916133,
|
| 93 |
+
"learning_rate": 8.810810275638183e-06,
|
| 94 |
+
"loss": 0.2401,
|
| 95 |
+
"step": 45
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.7777777777777777,
|
| 99 |
+
"grad_norm": 3.0860918384845974,
|
| 100 |
+
"learning_rate": 8.503473010366713e-06,
|
| 101 |
+
"loss": 0.2452,
|
| 102 |
+
"step": 50
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 3.0,
|
| 106 |
+
"eval_loss": 0.12752698361873627,
|
| 107 |
+
"eval_runtime": 47.9454,
|
| 108 |
+
"eval_samples_per_second": 11.555,
|
| 109 |
+
"eval_steps_per_second": 0.375,
|
| 110 |
+
"step": 54
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 3.0555555555555554,
|
| 114 |
+
"grad_norm": 10.045301056606435,
|
| 115 |
+
"learning_rate": 8.167602900802121e-06,
|
| 116 |
+
"loss": 0.2086,
|
| 117 |
+
"step": 55
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 3.3333333333333335,
|
| 121 |
+
"grad_norm": 3.9568837117288234,
|
| 122 |
+
"learning_rate": 7.805935326811913e-06,
|
| 123 |
+
"loss": 0.1144,
|
| 124 |
+
"step": 60
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 3.611111111111111,
|
| 128 |
+
"grad_norm": 2.319433762920475,
|
| 129 |
+
"learning_rate": 7.421415766938098e-06,
|
| 130 |
+
"loss": 0.1038,
|
| 131 |
+
"step": 65
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.888888888888889,
|
| 135 |
+
"grad_norm": 2.1464957865789307,
|
| 136 |
+
"learning_rate": 7.017175809949044e-06,
|
| 137 |
+
"loss": 0.098,
|
| 138 |
+
"step": 70
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 4.0,
|
| 142 |
+
"eval_loss": 0.04484181106090546,
|
| 143 |
+
"eval_runtime": 48.0587,
|
| 144 |
+
"eval_samples_per_second": 11.528,
|
| 145 |
+
"eval_steps_per_second": 0.375,
|
| 146 |
+
"step": 72
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 4.166666666666667,
|
| 150 |
+
"grad_norm": 1.7528155962578806,
|
| 151 |
+
"learning_rate": 6.5965076506799e-06,
|
| 152 |
+
"loss": 0.0637,
|
| 153 |
+
"step": 75
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 4.444444444444445,
|
| 157 |
+
"grad_norm": 1.8998235047238219,
|
| 158 |
+
"learning_rate": 6.162837277871553e-06,
|
| 159 |
+
"loss": 0.0492,
|
| 160 |
+
"step": 80
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 4.722222222222222,
|
| 164 |
+
"grad_norm": 1.727143217884404,
|
| 165 |
+
"learning_rate": 5.719696572370596e-06,
|
| 166 |
+
"loss": 0.0494,
|
| 167 |
+
"step": 85
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 5.0,
|
| 171 |
+
"grad_norm": 1.419982551702697,
|
| 172 |
+
"learning_rate": 5.270694542927089e-06,
|
| 173 |
+
"loss": 0.0439,
|
| 174 |
+
"step": 90
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 5.0,
|
| 178 |
+
"eval_loss": 0.025857795029878616,
|
| 179 |
+
"eval_runtime": 47.9969,
|
| 180 |
+
"eval_samples_per_second": 11.542,
|
| 181 |
+
"eval_steps_per_second": 0.375,
|
| 182 |
+
"step": 90
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 5.277777777777778,
|
| 186 |
+
"grad_norm": 1.5318777607270062,
|
| 187 |
+
"learning_rate": 4.819487933850319e-06,
|
| 188 |
+
"loss": 0.0278,
|
| 189 |
+
"step": 95
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 5.555555555555555,
|
| 193 |
+
"grad_norm": 1.5662579234139284,
|
| 194 |
+
"learning_rate": 4.369751443898554e-06,
|
| 195 |
+
"loss": 0.0273,
|
| 196 |
+
"step": 100
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 5.833333333333333,
|
| 200 |
+
"grad_norm": 1.2926901021879944,
|
| 201 |
+
"learning_rate": 3.92514779894488e-06,
|
| 202 |
+
"loss": 0.0306,
|
| 203 |
+
"step": 105
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 6.0,
|
| 207 |
+
"eval_loss": 0.014916397631168365,
|
| 208 |
+
"eval_runtime": 48.0425,
|
| 209 |
+
"eval_samples_per_second": 11.531,
|
| 210 |
+
"eval_steps_per_second": 0.375,
|
| 211 |
+
"step": 108
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 6.111111111111111,
|
| 215 |
+
"grad_norm": 0.859749869255068,
|
| 216 |
+
"learning_rate": 3.489297922152136e-06,
|
| 217 |
+
"loss": 0.0222,
|
| 218 |
+
"step": 110
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 6.388888888888889,
|
| 222 |
+
"grad_norm": 1.0464247096475507,
|
| 223 |
+
"learning_rate": 3.0657514445958055e-06,
|
| 224 |
+
"loss": 0.0136,
|
| 225 |
+
"step": 115
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 6.666666666666667,
|
| 229 |
+
"grad_norm": 1.355335206119406,
|
| 230 |
+
"learning_rate": 2.65795779650105e-06,
|
| 231 |
+
"loss": 0.0152,
|
| 232 |
+
"step": 120
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 6.944444444444445,
|
| 236 |
+
"grad_norm": 1.0250847159574037,
|
| 237 |
+
"learning_rate": 2.269238114531515e-06,
|
| 238 |
+
"loss": 0.017,
|
| 239 |
+
"step": 125
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 7.0,
|
| 243 |
+
"eval_loss": 0.006789184175431728,
|
| 244 |
+
"eval_runtime": 47.9791,
|
| 245 |
+
"eval_samples_per_second": 11.547,
|
| 246 |
+
"eval_steps_per_second": 0.375,
|
| 247 |
+
"step": 126
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 7.222222222222222,
|
| 251 |
+
"grad_norm": 0.5699491687319692,
|
| 252 |
+
"learning_rate": 1.9027581939213852e-06,
|
| 253 |
+
"loss": 0.0078,
|
| 254 |
+
"step": 130
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 7.5,
|
| 258 |
+
"grad_norm": 0.663596820501962,
|
| 259 |
+
"learning_rate": 1.561502705732883e-06,
|
| 260 |
+
"loss": 0.0053,
|
| 261 |
+
"step": 135
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.777777777777778,
|
| 265 |
+
"grad_norm": 0.8863531225772922,
|
| 266 |
+
"learning_rate": 1.2482508892179884e-06,
|
| 267 |
+
"loss": 0.006,
|
| 268 |
+
"step": 140
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 8.0,
|
| 272 |
+
"eval_loss": 0.002534537808969617,
|
| 273 |
+
"eval_runtime": 48.0729,
|
| 274 |
+
"eval_samples_per_second": 11.524,
|
| 275 |
+
"eval_steps_per_second": 0.374,
|
| 276 |
+
"step": 144
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 8.055555555555555,
|
| 280 |
+
"grad_norm": 0.4661094231461166,
|
| 281 |
+
"learning_rate": 9.65553917249648e-07,
|
| 282 |
+
"loss": 0.0065,
|
| 283 |
+
"step": 145
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 8.333333333333334,
|
| 287 |
+
"grad_norm": 0.39663216492684406,
|
| 288 |
+
"learning_rate": 7.157141191620548e-07,
|
| 289 |
+
"loss": 0.0021,
|
| 290 |
+
"step": 150
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 8.61111111111111,
|
| 294 |
+
"grad_norm": 0.3373551351268207,
|
| 295 |
+
"learning_rate": 5.007662302124671e-07,
|
| 296 |
+
"loss": 0.0019,
|
| 297 |
+
"step": 155
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 8.88888888888889,
|
| 301 |
+
"grad_norm": 0.3425926240340406,
|
| 302 |
+
"learning_rate": 3.224608203719953e-07,
|
| 303 |
+
"loss": 0.0025,
|
| 304 |
+
"step": 160
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 9.0,
|
| 308 |
+
"eval_loss": 0.0013950355350971222,
|
| 309 |
+
"eval_runtime": 47.9782,
|
| 310 |
+
"eval_samples_per_second": 11.547,
|
| 311 |
+
"eval_steps_per_second": 0.375,
|
| 312 |
+
"step": 162
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 9.166666666666666,
|
| 316 |
+
"grad_norm": 0.2182536824166498,
|
| 317 |
+
"learning_rate": 1.8225003740388546e-07,
|
| 318 |
+
"loss": 0.0021,
|
| 319 |
+
"step": 165
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 9.444444444444445,
|
| 323 |
+
"grad_norm": 0.1196015045593264,
|
| 324 |
+
"learning_rate": 8.127578033998663e-08,
|
| 325 |
+
"loss": 0.0012,
|
| 326 |
+
"step": 170
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 9.722222222222221,
|
| 330 |
+
"grad_norm": 0.24750584137098056,
|
| 331 |
+
"learning_rate": 2.0360399672478826e-08,
|
| 332 |
+
"loss": 0.0013,
|
| 333 |
+
"step": 175
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 10.0,
|
| 337 |
+
"grad_norm": 0.1571976207810524,
|
| 338 |
+
"learning_rate": 0.0,
|
| 339 |
+
"loss": 0.0012,
|
| 340 |
+
"step": 180
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 10.0,
|
| 344 |
+
"eval_loss": 0.001282889163121581,
|
| 345 |
+
"eval_runtime": 47.8096,
|
| 346 |
+
"eval_samples_per_second": 11.588,
|
| 347 |
+
"eval_steps_per_second": 0.376,
|
| 348 |
+
"step": 180
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 10.0,
|
| 352 |
+
"step": 180,
|
| 353 |
+
"total_flos": 37688338022400.0,
|
| 354 |
+
"train_loss": 0.18709226546602117,
|
| 355 |
+
"train_runtime": 2578.874,
|
| 356 |
+
"train_samples_per_second": 2.148,
|
| 357 |
+
"train_steps_per_second": 0.07
|
| 358 |
+
}
|
| 359 |
+
],
|
| 360 |
+
"logging_steps": 5,
|
| 361 |
+
"max_steps": 180,
|
| 362 |
+
"num_input_tokens_seen": 0,
|
| 363 |
+
"num_train_epochs": 10,
|
| 364 |
+
"save_steps": 100,
|
| 365 |
+
"stateful_callbacks": {
|
| 366 |
+
"TrainerControl": {
|
| 367 |
+
"args": {
|
| 368 |
+
"should_epoch_stop": false,
|
| 369 |
+
"should_evaluate": false,
|
| 370 |
+
"should_log": false,
|
| 371 |
+
"should_save": true,
|
| 372 |
+
"should_training_stop": true
|
| 373 |
+
},
|
| 374 |
+
"attributes": {}
|
| 375 |
+
}
|
| 376 |
+
},
|
| 377 |
+
"total_flos": 37688338022400.0,
|
| 378 |
+
"train_batch_size": 8,
|
| 379 |
+
"trial_name": null,
|
| 380 |
+
"trial_params": null
|
| 381 |
+
}
|