Model save
Browse files- README.md +78 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +340 -0
README.md
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
license: llama3.2
|
| 4 |
+
base_model: meta-llama/Llama-3.2-1B
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: llama3-1b-classification-gpt4o-100k
|
| 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 |
+
# llama3-1b-classification-gpt4o-100k
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [meta-llama/Llama-3.2-1B](https://huggingface.co/meta-llama/Llama-3.2-1B) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 2.3686
|
| 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: 0.0002
|
| 43 |
+
- train_batch_size: 32
|
| 44 |
+
- eval_batch_size: 32
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 8
|
| 48 |
+
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 512
|
| 50 |
+
- total_eval_batch_size: 256
|
| 51 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 52 |
+
- lr_scheduler_type: cosine
|
| 53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 54 |
+
- num_epochs: 10
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 60 |
+
| 2.759 | 1.0 | 15 | 2.6168 |
|
| 61 |
+
| 2.3698 | 2.0 | 30 | 2.4602 |
|
| 62 |
+
| 2.2521 | 3.0 | 45 | 2.3993 |
|
| 63 |
+
| 2.1865 | 4.0 | 60 | 2.3780 |
|
| 64 |
+
| 2.1518 | 5.0 | 75 | 2.3706 |
|
| 65 |
+
| 2.1331 | 6.0 | 90 | 2.3701 |
|
| 66 |
+
| 2.1168 | 7.0 | 105 | 2.3705 |
|
| 67 |
+
| 2.1121 | 8.0 | 120 | 2.3697 |
|
| 68 |
+
| 2.1081 | 9.0 | 135 | 2.3696 |
|
| 69 |
+
| 2.1077 | 10.0 | 150 | 2.3686 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
### Framework versions
|
| 73 |
+
|
| 74 |
+
- PEFT 0.15.1
|
| 75 |
+
- Transformers 4.50.3
|
| 76 |
+
- Pytorch 2.6.0+cu124
|
| 77 |
+
- Datasets 3.5.0
|
| 78 |
+
- Tokenizers 0.21.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 4.607959816470528e+17,
|
| 4 |
+
"train_loss": 2.2735363483428954,
|
| 5 |
+
"train_runtime": 707.5313,
|
| 6 |
+
"train_samples": 92634,
|
| 7 |
+
"train_samples_per_second": 106.978,
|
| 8 |
+
"train_steps_per_second": 0.212
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 4.607959816470528e+17,
|
| 4 |
+
"train_loss": 2.2735363483428954,
|
| 5 |
+
"train_runtime": 707.5313,
|
| 6 |
+
"train_samples": 92634,
|
| 7 |
+
"train_samples_per_second": 106.978,
|
| 8 |
+
"train_steps_per_second": 0.212
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,340 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 10.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 150,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.06666666666666667,
|
| 14 |
+
"grad_norm": 6.182748794555664,
|
| 15 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 16 |
+
"loss": 3.3242,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.3333333333333333,
|
| 21 |
+
"grad_norm": 5.428755283355713,
|
| 22 |
+
"learning_rate": 6.666666666666667e-05,
|
| 23 |
+
"loss": 3.3024,
|
| 24 |
+
"step": 5
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.6666666666666666,
|
| 28 |
+
"grad_norm": 1.8031461238861084,
|
| 29 |
+
"learning_rate": 0.00013333333333333334,
|
| 30 |
+
"loss": 3.0512,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.0,
|
| 35 |
+
"grad_norm": 1.0242935419082642,
|
| 36 |
+
"learning_rate": 0.0002,
|
| 37 |
+
"loss": 2.759,
|
| 38 |
+
"step": 15
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.0,
|
| 42 |
+
"eval_loss": 2.6168200969696045,
|
| 43 |
+
"eval_runtime": 0.8249,
|
| 44 |
+
"eval_samples_per_second": 2.425,
|
| 45 |
+
"eval_steps_per_second": 1.212,
|
| 46 |
+
"step": 15
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.3333333333333333,
|
| 50 |
+
"grad_norm": 0.654896080493927,
|
| 51 |
+
"learning_rate": 0.00019932383577419432,
|
| 52 |
+
"loss": 2.5777,
|
| 53 |
+
"step": 20
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.6666666666666665,
|
| 57 |
+
"grad_norm": 0.4863899052143097,
|
| 58 |
+
"learning_rate": 0.00019730448705798239,
|
| 59 |
+
"loss": 2.451,
|
| 60 |
+
"step": 25
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 2.0,
|
| 64 |
+
"grad_norm": 0.3055514991283417,
|
| 65 |
+
"learning_rate": 0.00019396926207859084,
|
| 66 |
+
"loss": 2.3698,
|
| 67 |
+
"step": 30
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.0,
|
| 71 |
+
"eval_loss": 2.460226058959961,
|
| 72 |
+
"eval_runtime": 0.8278,
|
| 73 |
+
"eval_samples_per_second": 2.416,
|
| 74 |
+
"eval_steps_per_second": 1.208,
|
| 75 |
+
"step": 30
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 2.3333333333333335,
|
| 79 |
+
"grad_norm": 0.2144581824541092,
|
| 80 |
+
"learning_rate": 0.00018936326403234125,
|
| 81 |
+
"loss": 2.3071,
|
| 82 |
+
"step": 35
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 2.6666666666666665,
|
| 86 |
+
"grad_norm": 0.18541619181632996,
|
| 87 |
+
"learning_rate": 0.00018354878114129367,
|
| 88 |
+
"loss": 2.2806,
|
| 89 |
+
"step": 40
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 3.0,
|
| 93 |
+
"grad_norm": 0.16438105702400208,
|
| 94 |
+
"learning_rate": 0.0001766044443118978,
|
| 95 |
+
"loss": 2.2521,
|
| 96 |
+
"step": 45
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.0,
|
| 100 |
+
"eval_loss": 2.399292230606079,
|
| 101 |
+
"eval_runtime": 0.8291,
|
| 102 |
+
"eval_samples_per_second": 2.412,
|
| 103 |
+
"eval_steps_per_second": 1.206,
|
| 104 |
+
"step": 45
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 3.3333333333333335,
|
| 108 |
+
"grad_norm": 0.1444542109966278,
|
| 109 |
+
"learning_rate": 0.0001686241637868734,
|
| 110 |
+
"loss": 2.2256,
|
| 111 |
+
"step": 50
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 3.6666666666666665,
|
| 115 |
+
"grad_norm": 0.12350399047136307,
|
| 116 |
+
"learning_rate": 0.00015971585917027862,
|
| 117 |
+
"loss": 2.2069,
|
| 118 |
+
"step": 55
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 4.0,
|
| 122 |
+
"grad_norm": 0.12858685851097107,
|
| 123 |
+
"learning_rate": 0.00015000000000000001,
|
| 124 |
+
"loss": 2.1865,
|
| 125 |
+
"step": 60
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.0,
|
| 129 |
+
"eval_loss": 2.3779759407043457,
|
| 130 |
+
"eval_runtime": 0.8275,
|
| 131 |
+
"eval_samples_per_second": 2.417,
|
| 132 |
+
"eval_steps_per_second": 1.208,
|
| 133 |
+
"step": 60
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 4.333333333333333,
|
| 137 |
+
"grad_norm": 0.11426849663257599,
|
| 138 |
+
"learning_rate": 0.0001396079766039157,
|
| 139 |
+
"loss": 2.1796,
|
| 140 |
+
"step": 65
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 4.666666666666667,
|
| 144 |
+
"grad_norm": 0.11002852022647858,
|
| 145 |
+
"learning_rate": 0.00012868032327110904,
|
| 146 |
+
"loss": 2.1632,
|
| 147 |
+
"step": 70
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 5.0,
|
| 151 |
+
"grad_norm": 0.11463355273008347,
|
| 152 |
+
"learning_rate": 0.00011736481776669306,
|
| 153 |
+
"loss": 2.1518,
|
| 154 |
+
"step": 75
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 5.0,
|
| 158 |
+
"eval_loss": 2.3705692291259766,
|
| 159 |
+
"eval_runtime": 0.8277,
|
| 160 |
+
"eval_samples_per_second": 2.416,
|
| 161 |
+
"eval_steps_per_second": 1.208,
|
| 162 |
+
"step": 75
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 5.333333333333333,
|
| 166 |
+
"grad_norm": 0.10701094567775726,
|
| 167 |
+
"learning_rate": 0.00010581448289104758,
|
| 168 |
+
"loss": 2.1438,
|
| 169 |
+
"step": 80
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 5.666666666666667,
|
| 173 |
+
"grad_norm": 0.10402818769216537,
|
| 174 |
+
"learning_rate": 9.418551710895243e-05,
|
| 175 |
+
"loss": 2.1353,
|
| 176 |
+
"step": 85
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.0,
|
| 180 |
+
"grad_norm": 0.10872997343540192,
|
| 181 |
+
"learning_rate": 8.263518223330697e-05,
|
| 182 |
+
"loss": 2.1331,
|
| 183 |
+
"step": 90
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 6.0,
|
| 187 |
+
"eval_loss": 2.370138645172119,
|
| 188 |
+
"eval_runtime": 0.8279,
|
| 189 |
+
"eval_samples_per_second": 2.416,
|
| 190 |
+
"eval_steps_per_second": 1.208,
|
| 191 |
+
"step": 90
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 6.333333333333333,
|
| 195 |
+
"grad_norm": 0.11619067192077637,
|
| 196 |
+
"learning_rate": 7.131967672889101e-05,
|
| 197 |
+
"loss": 2.1258,
|
| 198 |
+
"step": 95
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 6.666666666666667,
|
| 202 |
+
"grad_norm": 0.12228544801473618,
|
| 203 |
+
"learning_rate": 6.039202339608432e-05,
|
| 204 |
+
"loss": 2.1196,
|
| 205 |
+
"step": 100
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.0,
|
| 209 |
+
"grad_norm": 0.11531439423561096,
|
| 210 |
+
"learning_rate": 5.000000000000002e-05,
|
| 211 |
+
"loss": 2.1168,
|
| 212 |
+
"step": 105
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 7.0,
|
| 216 |
+
"eval_loss": 2.3705224990844727,
|
| 217 |
+
"eval_runtime": 0.8289,
|
| 218 |
+
"eval_samples_per_second": 2.413,
|
| 219 |
+
"eval_steps_per_second": 1.206,
|
| 220 |
+
"step": 105
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 7.333333333333333,
|
| 224 |
+
"grad_norm": 0.10404527932405472,
|
| 225 |
+
"learning_rate": 4.028414082972141e-05,
|
| 226 |
+
"loss": 2.1119,
|
| 227 |
+
"step": 110
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 7.666666666666667,
|
| 231 |
+
"grad_norm": 0.10795953124761581,
|
| 232 |
+
"learning_rate": 3.137583621312665e-05,
|
| 233 |
+
"loss": 2.1085,
|
| 234 |
+
"step": 115
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 8.0,
|
| 238 |
+
"grad_norm": 0.11347240209579468,
|
| 239 |
+
"learning_rate": 2.339555568810221e-05,
|
| 240 |
+
"loss": 2.1121,
|
| 241 |
+
"step": 120
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 8.0,
|
| 245 |
+
"eval_loss": 2.369734048843384,
|
| 246 |
+
"eval_runtime": 0.8296,
|
| 247 |
+
"eval_samples_per_second": 2.411,
|
| 248 |
+
"eval_steps_per_second": 1.205,
|
| 249 |
+
"step": 120
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 8.333333333333334,
|
| 253 |
+
"grad_norm": 0.10211199522018433,
|
| 254 |
+
"learning_rate": 1.6451218858706374e-05,
|
| 255 |
+
"loss": 2.1088,
|
| 256 |
+
"step": 125
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 8.666666666666666,
|
| 260 |
+
"grad_norm": 0.0971965491771698,
|
| 261 |
+
"learning_rate": 1.0636735967658784e-05,
|
| 262 |
+
"loss": 2.101,
|
| 263 |
+
"step": 130
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 9.0,
|
| 267 |
+
"grad_norm": 0.0981559231877327,
|
| 268 |
+
"learning_rate": 6.030737921409169e-06,
|
| 269 |
+
"loss": 2.1081,
|
| 270 |
+
"step": 135
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 9.0,
|
| 274 |
+
"eval_loss": 2.3695523738861084,
|
| 275 |
+
"eval_runtime": 0.8287,
|
| 276 |
+
"eval_samples_per_second": 2.413,
|
| 277 |
+
"eval_steps_per_second": 1.207,
|
| 278 |
+
"step": 135
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 9.333333333333334,
|
| 282 |
+
"grad_norm": 0.09640078991651535,
|
| 283 |
+
"learning_rate": 2.6955129420176196e-06,
|
| 284 |
+
"loss": 2.1009,
|
| 285 |
+
"step": 140
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 9.666666666666666,
|
| 289 |
+
"grad_norm": 0.09775780886411667,
|
| 290 |
+
"learning_rate": 6.761642258056978e-07,
|
| 291 |
+
"loss": 2.1037,
|
| 292 |
+
"step": 145
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 10.0,
|
| 296 |
+
"grad_norm": 0.09380096942186356,
|
| 297 |
+
"learning_rate": 0.0,
|
| 298 |
+
"loss": 2.1077,
|
| 299 |
+
"step": 150
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 10.0,
|
| 303 |
+
"eval_loss": 2.368610143661499,
|
| 304 |
+
"eval_runtime": 0.8302,
|
| 305 |
+
"eval_samples_per_second": 2.409,
|
| 306 |
+
"eval_steps_per_second": 1.205,
|
| 307 |
+
"step": 150
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 10.0,
|
| 311 |
+
"step": 150,
|
| 312 |
+
"total_flos": 4.607959816470528e+17,
|
| 313 |
+
"train_loss": 2.2735363483428954,
|
| 314 |
+
"train_runtime": 707.5313,
|
| 315 |
+
"train_samples_per_second": 106.978,
|
| 316 |
+
"train_steps_per_second": 0.212
|
| 317 |
+
}
|
| 318 |
+
],
|
| 319 |
+
"logging_steps": 5,
|
| 320 |
+
"max_steps": 150,
|
| 321 |
+
"num_input_tokens_seen": 0,
|
| 322 |
+
"num_train_epochs": 10,
|
| 323 |
+
"save_steps": 100,
|
| 324 |
+
"stateful_callbacks": {
|
| 325 |
+
"TrainerControl": {
|
| 326 |
+
"args": {
|
| 327 |
+
"should_epoch_stop": false,
|
| 328 |
+
"should_evaluate": false,
|
| 329 |
+
"should_log": false,
|
| 330 |
+
"should_save": true,
|
| 331 |
+
"should_training_stop": true
|
| 332 |
+
},
|
| 333 |
+
"attributes": {}
|
| 334 |
+
}
|
| 335 |
+
},
|
| 336 |
+
"total_flos": 4.607959816470528e+17,
|
| 337 |
+
"train_batch_size": 32,
|
| 338 |
+
"trial_name": null,
|
| 339 |
+
"trial_params": null
|
| 340 |
+
}
|