End of training
Browse files- README.md +2 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +416 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: mistralai/Mistral-7B-v0.1
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: hp_ablations_grid_mistral_bsz2048_lr5e-6_scheduler-cosine-warmup0.15
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# hp_ablations_grid_mistral_bsz2048_lr5e-6_scheduler-cosine-warmup0.15
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.0580
|
| 21 |
|
|
|
|
| 4 |
base_model: mistralai/Mistral-7B-v0.1
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: hp_ablations_grid_mistral_bsz2048_lr5e-6_scheduler-cosine-warmup0.15
|
|
|
|
| 16 |
|
| 17 |
# hp_ablations_grid_mistral_bsz2048_lr5e-6_scheduler-cosine-warmup0.15
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the mlfoundations-dev/oh-dcft-v3-llama3.1-nemotron-70b_shareGPT_format dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.0580
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.057990334928035736,
|
| 4 |
+
"eval_runtime": 175.703,
|
| 5 |
+
"eval_samples_per_second": 103.032,
|
| 6 |
+
"eval_steps_per_second": 0.404,
|
| 7 |
+
"total_flos": 3376037568184320.0,
|
| 8 |
+
"train_loss": 0.46887724030585515,
|
| 9 |
+
"train_runtime": 29847.5634,
|
| 10 |
+
"train_samples_per_second": 34.571,
|
| 11 |
+
"train_steps_per_second": 0.017
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.057990334928035736,
|
| 4 |
+
"eval_runtime": 175.703,
|
| 5 |
+
"eval_samples_per_second": 103.032,
|
| 6 |
+
"eval_steps_per_second": 0.404
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 3376037568184320.0,
|
| 4 |
+
"train_loss": 0.46887724030585515,
|
| 5 |
+
"train_runtime": 29847.5634,
|
| 6 |
+
"train_samples_per_second": 34.571,
|
| 7 |
+
"train_steps_per_second": 0.017
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,416 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 504,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05952380952380952,
|
| 13 |
+
"grad_norm": 5.302074700564879,
|
| 14 |
+
"learning_rate": 6.578947368421053e-07,
|
| 15 |
+
"loss": 0.8677,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.11904761904761904,
|
| 20 |
+
"grad_norm": 2.3539338952660676,
|
| 21 |
+
"learning_rate": 1.3157894736842106e-06,
|
| 22 |
+
"loss": 0.7536,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.17857142857142858,
|
| 27 |
+
"grad_norm": 1.3061427287792207,
|
| 28 |
+
"learning_rate": 1.973684210526316e-06,
|
| 29 |
+
"loss": 0.6719,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.23809523809523808,
|
| 34 |
+
"grad_norm": 1.7706136132278263,
|
| 35 |
+
"learning_rate": 2.631578947368421e-06,
|
| 36 |
+
"loss": 0.6226,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.2976190476190476,
|
| 41 |
+
"grad_norm": 2.31915144001752,
|
| 42 |
+
"learning_rate": 3.289473684210527e-06,
|
| 43 |
+
"loss": 0.5929,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.35714285714285715,
|
| 48 |
+
"grad_norm": 3.6318279241243174,
|
| 49 |
+
"learning_rate": 3.947368421052632e-06,
|
| 50 |
+
"loss": 0.5717,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.4166666666666667,
|
| 55 |
+
"grad_norm": 5.466002436425688,
|
| 56 |
+
"learning_rate": 4.605263157894737e-06,
|
| 57 |
+
"loss": 0.56,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.47619047619047616,
|
| 62 |
+
"grad_norm": 2.658157430170319,
|
| 63 |
+
"learning_rate": 4.998922515567496e-06,
|
| 64 |
+
"loss": 0.5529,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.5357142857142857,
|
| 69 |
+
"grad_norm": 2.958620225945063,
|
| 70 |
+
"learning_rate": 4.986811479610213e-06,
|
| 71 |
+
"loss": 0.5459,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.5952380952380952,
|
| 76 |
+
"grad_norm": 2.3621141521204074,
|
| 77 |
+
"learning_rate": 4.9613079925074435e-06,
|
| 78 |
+
"loss": 0.5336,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.6547619047619048,
|
| 83 |
+
"grad_norm": 2.132170449736533,
|
| 84 |
+
"learning_rate": 4.922549400498685e-06,
|
| 85 |
+
"loss": 0.5296,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.7142857142857143,
|
| 90 |
+
"grad_norm": 2.422127656789715,
|
| 91 |
+
"learning_rate": 4.870744433740688e-06,
|
| 92 |
+
"loss": 0.52,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.7738095238095238,
|
| 97 |
+
"grad_norm": 3.209901466327175,
|
| 98 |
+
"learning_rate": 4.806172082214032e-06,
|
| 99 |
+
"loss": 0.5164,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.8333333333333334,
|
| 104 |
+
"grad_norm": 2.3530816407534223,
|
| 105 |
+
"learning_rate": 4.729180093253106e-06,
|
| 106 |
+
"loss": 0.5092,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.8928571428571429,
|
| 111 |
+
"grad_norm": 2.0107372006602007,
|
| 112 |
+
"learning_rate": 4.640183098790896e-06,
|
| 113 |
+
"loss": 0.5046,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.9523809523809523,
|
| 118 |
+
"grad_norm": 1.6986015532338576,
|
| 119 |
+
"learning_rate": 4.53966038240406e-06,
|
| 120 |
+
"loss": 0.4983,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.0,
|
| 125 |
+
"eval_loss": 0.061965156346559525,
|
| 126 |
+
"eval_runtime": 175.8558,
|
| 127 |
+
"eval_samples_per_second": 102.942,
|
| 128 |
+
"eval_steps_per_second": 0.404,
|
| 129 |
+
"step": 168
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.0119047619047619,
|
| 133 |
+
"grad_norm": 2.0617265758036574,
|
| 134 |
+
"learning_rate": 4.428153298183676e-06,
|
| 135 |
+
"loss": 0.4899,
|
| 136 |
+
"step": 170
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.0714285714285714,
|
| 140 |
+
"grad_norm": 2.461751865882228,
|
| 141 |
+
"learning_rate": 4.306262355332006e-06,
|
| 142 |
+
"loss": 0.4566,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.130952380952381,
|
| 147 |
+
"grad_norm": 2.540542988750059,
|
| 148 |
+
"learning_rate": 4.17464398418589e-06,
|
| 149 |
+
"loss": 0.4541,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.1904761904761905,
|
| 154 |
+
"grad_norm": 1.908700820605462,
|
| 155 |
+
"learning_rate": 4.034007001082985e-06,
|
| 156 |
+
"loss": 0.4525,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.25,
|
| 161 |
+
"grad_norm": 1.3119343985161747,
|
| 162 |
+
"learning_rate": 3.8851087911089315e-06,
|
| 163 |
+
"loss": 0.4533,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.3095238095238095,
|
| 168 |
+
"grad_norm": 2.004423567844585,
|
| 169 |
+
"learning_rate": 3.7287512292828364e-06,
|
| 170 |
+
"loss": 0.4491,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.369047619047619,
|
| 175 |
+
"grad_norm": 2.5928234091324764,
|
| 176 |
+
"learning_rate": 3.5657763621470905e-06,
|
| 177 |
+
"loss": 0.4493,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.4285714285714286,
|
| 182 |
+
"grad_norm": 1.8913706584793621,
|
| 183 |
+
"learning_rate": 3.39706187301784e-06,
|
| 184 |
+
"loss": 0.4476,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.4880952380952381,
|
| 189 |
+
"grad_norm": 2.058386268253094,
|
| 190 |
+
"learning_rate": 3.2235163553175174e-06,
|
| 191 |
+
"loss": 0.4437,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.5476190476190477,
|
| 196 |
+
"grad_norm": 1.9521413891863681,
|
| 197 |
+
"learning_rate": 3.0460744194443658e-06,
|
| 198 |
+
"loss": 0.4423,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.6071428571428572,
|
| 203 |
+
"grad_norm": 1.652934076293377,
|
| 204 |
+
"learning_rate": 2.8656916595304026e-06,
|
| 205 |
+
"loss": 0.44,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.6666666666666665,
|
| 210 |
+
"grad_norm": 1.4254067692233685,
|
| 211 |
+
"learning_rate": 2.68333950719376e-06,
|
| 212 |
+
"loss": 0.4404,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.7261904761904763,
|
| 217 |
+
"grad_norm": 1.9096049494562852,
|
| 218 |
+
"learning_rate": 2.5e-06,
|
| 219 |
+
"loss": 0.4397,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.7857142857142856,
|
| 224 |
+
"grad_norm": 1.2870553552525479,
|
| 225 |
+
"learning_rate": 2.3166604928062407e-06,
|
| 226 |
+
"loss": 0.4351,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.8452380952380953,
|
| 231 |
+
"grad_norm": 1.2328177499933723,
|
| 232 |
+
"learning_rate": 2.1343083404695987e-06,
|
| 233 |
+
"loss": 0.4341,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.9047619047619047,
|
| 238 |
+
"grad_norm": 1.3465694131057355,
|
| 239 |
+
"learning_rate": 1.9539255805556346e-06,
|
| 240 |
+
"loss": 0.4338,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.9642857142857144,
|
| 245 |
+
"grad_norm": 1.2032285531530662,
|
| 246 |
+
"learning_rate": 1.7764836446824835e-06,
|
| 247 |
+
"loss": 0.4333,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.0,
|
| 252 |
+
"eval_loss": 0.05795557424426079,
|
| 253 |
+
"eval_runtime": 175.972,
|
| 254 |
+
"eval_samples_per_second": 102.874,
|
| 255 |
+
"eval_steps_per_second": 0.403,
|
| 256 |
+
"step": 336
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.0238095238095237,
|
| 260 |
+
"grad_norm": 1.789131444902922,
|
| 261 |
+
"learning_rate": 1.6029381269821607e-06,
|
| 262 |
+
"loss": 0.414,
|
| 263 |
+
"step": 340
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.0833333333333335,
|
| 267 |
+
"grad_norm": 1.3420020277085403,
|
| 268 |
+
"learning_rate": 1.4342236378529106e-06,
|
| 269 |
+
"loss": 0.3888,
|
| 270 |
+
"step": 350
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.142857142857143,
|
| 274 |
+
"grad_norm": 1.1852540907855833,
|
| 275 |
+
"learning_rate": 1.2712487707171645e-06,
|
| 276 |
+
"loss": 0.3841,
|
| 277 |
+
"step": 360
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.2023809523809526,
|
| 281 |
+
"grad_norm": 1.6133303890307713,
|
| 282 |
+
"learning_rate": 1.1148912088910687e-06,
|
| 283 |
+
"loss": 0.3837,
|
| 284 |
+
"step": 370
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.261904761904762,
|
| 288 |
+
"grad_norm": 0.892698173344749,
|
| 289 |
+
"learning_rate": 9.659929989170156e-07,
|
| 290 |
+
"loss": 0.3863,
|
| 291 |
+
"step": 380
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.3214285714285716,
|
| 295 |
+
"grad_norm": 0.8511252404619682,
|
| 296 |
+
"learning_rate": 8.253560158141111e-07,
|
| 297 |
+
"loss": 0.3825,
|
| 298 |
+
"step": 390
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.380952380952381,
|
| 302 |
+
"grad_norm": 0.7995163038995144,
|
| 303 |
+
"learning_rate": 6.93737644667995e-07,
|
| 304 |
+
"loss": 0.3817,
|
| 305 |
+
"step": 400
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.4404761904761907,
|
| 309 |
+
"grad_norm": 0.6938195822302593,
|
| 310 |
+
"learning_rate": 5.718467018163243e-07,
|
| 311 |
+
"loss": 0.3816,
|
| 312 |
+
"step": 410
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.5,
|
| 316 |
+
"grad_norm": 0.7359496640818899,
|
| 317 |
+
"learning_rate": 4.6033961759594045e-07,
|
| 318 |
+
"loss": 0.3815,
|
| 319 |
+
"step": 420
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.5595238095238093,
|
| 323 |
+
"grad_norm": 0.7334287317942382,
|
| 324 |
+
"learning_rate": 3.598169012091049e-07,
|
| 325 |
+
"loss": 0.3829,
|
| 326 |
+
"step": 430
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.619047619047619,
|
| 330 |
+
"grad_norm": 0.6722577072763187,
|
| 331 |
+
"learning_rate": 2.708199067468939e-07,
|
| 332 |
+
"loss": 0.3837,
|
| 333 |
+
"step": 440
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.678571428571429,
|
| 337 |
+
"grad_norm": 0.7090527130304488,
|
| 338 |
+
"learning_rate": 1.9382791778596864e-07,
|
| 339 |
+
"loss": 0.3818,
|
| 340 |
+
"step": 450
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.738095238095238,
|
| 344 |
+
"grad_norm": 0.6664042341300833,
|
| 345 |
+
"learning_rate": 1.2925556625931174e-07,
|
| 346 |
+
"loss": 0.3828,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.7976190476190474,
|
| 351 |
+
"grad_norm": 0.6629290902040299,
|
| 352 |
+
"learning_rate": 7.7450599501315e-08,
|
| 353 |
+
"loss": 0.3819,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.857142857142857,
|
| 358 |
+
"grad_norm": 0.6540655845446479,
|
| 359 |
+
"learning_rate": 3.869200749255703e-08,
|
| 360 |
+
"loss": 0.3815,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.9166666666666665,
|
| 365 |
+
"grad_norm": 0.6419630459088013,
|
| 366 |
+
"learning_rate": 1.3188520389787462e-08,
|
| 367 |
+
"loss": 0.3791,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.9761904761904763,
|
| 372 |
+
"grad_norm": 0.6685967829333805,
|
| 373 |
+
"learning_rate": 1.0774844325039946e-09,
|
| 374 |
+
"loss": 0.3811,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.0,
|
| 379 |
+
"eval_loss": 0.057990334928035736,
|
| 380 |
+
"eval_runtime": 175.1942,
|
| 381 |
+
"eval_samples_per_second": 103.331,
|
| 382 |
+
"eval_steps_per_second": 0.405,
|
| 383 |
+
"step": 504
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 3.0,
|
| 387 |
+
"step": 504,
|
| 388 |
+
"total_flos": 3376037568184320.0,
|
| 389 |
+
"train_loss": 0.46887724030585515,
|
| 390 |
+
"train_runtime": 29847.5634,
|
| 391 |
+
"train_samples_per_second": 34.571,
|
| 392 |
+
"train_steps_per_second": 0.017
|
| 393 |
+
}
|
| 394 |
+
],
|
| 395 |
+
"logging_steps": 10,
|
| 396 |
+
"max_steps": 504,
|
| 397 |
+
"num_input_tokens_seen": 0,
|
| 398 |
+
"num_train_epochs": 3,
|
| 399 |
+
"save_steps": 500,
|
| 400 |
+
"stateful_callbacks": {
|
| 401 |
+
"TrainerControl": {
|
| 402 |
+
"args": {
|
| 403 |
+
"should_epoch_stop": false,
|
| 404 |
+
"should_evaluate": false,
|
| 405 |
+
"should_log": false,
|
| 406 |
+
"should_save": true,
|
| 407 |
+
"should_training_stop": true
|
| 408 |
+
},
|
| 409 |
+
"attributes": {}
|
| 410 |
+
}
|
| 411 |
+
},
|
| 412 |
+
"total_flos": 3376037568184320.0,
|
| 413 |
+
"train_batch_size": 8,
|
| 414 |
+
"trial_name": null,
|
| 415 |
+
"trial_params": null
|
| 416 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|