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-minlr5e-7
|
|
@@ -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-minlr5e-7
|
| 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: 0.0577
|
| 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-minlr5e-7
|
|
|
|
| 16 |
|
| 17 |
# hp_ablations_grid_mistral_bsz2048_lr5e-6_scheduler-cosine-warmup0.15-minlr5e-7
|
| 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: 0.0577
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.05768963322043419,
|
| 4 |
+
"eval_runtime": 115.3652,
|
| 5 |
+
"eval_samples_per_second": 156.919,
|
| 6 |
+
"eval_steps_per_second": 0.615,
|
| 7 |
+
"total_flos": 3376037568184320.0,
|
| 8 |
+
"train_loss": 0.4685976881356466,
|
| 9 |
+
"train_runtime": 17114.7356,
|
| 10 |
+
"train_samples_per_second": 60.291,
|
| 11 |
+
"train_steps_per_second": 0.029
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.05768963322043419,
|
| 4 |
+
"eval_runtime": 115.3652,
|
| 5 |
+
"eval_samples_per_second": 156.919,
|
| 6 |
+
"eval_steps_per_second": 0.615
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 3376037568184320.0,
|
| 4 |
+
"train_loss": 0.4685976881356466,
|
| 5 |
+
"train_runtime": 17114.7356,
|
| 6 |
+
"train_samples_per_second": 60.291,
|
| 7 |
+
"train_steps_per_second": 0.029
|
| 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": 5.299262070423908,
|
| 14 |
+
"learning_rate": 6.578947368421053e-07,
|
| 15 |
+
"loss": 0.8677,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.11904761904761904,
|
| 20 |
+
"grad_norm": 2.380389310720017,
|
| 21 |
+
"learning_rate": 1.3157894736842106e-06,
|
| 22 |
+
"loss": 0.7536,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.17857142857142858,
|
| 27 |
+
"grad_norm": 1.3714299753599561,
|
| 28 |
+
"learning_rate": 1.973684210526316e-06,
|
| 29 |
+
"loss": 0.672,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.23809523809523808,
|
| 34 |
+
"grad_norm": 2.440712010584784,
|
| 35 |
+
"learning_rate": 2.631578947368421e-06,
|
| 36 |
+
"loss": 0.6227,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.2976190476190476,
|
| 41 |
+
"grad_norm": 3.672454000950744,
|
| 42 |
+
"learning_rate": 3.289473684210527e-06,
|
| 43 |
+
"loss": 0.5956,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.35714285714285715,
|
| 48 |
+
"grad_norm": 3.8923742001385255,
|
| 49 |
+
"learning_rate": 3.947368421052632e-06,
|
| 50 |
+
"loss": 0.5737,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.4166666666666667,
|
| 55 |
+
"grad_norm": 2.7164238633471167,
|
| 56 |
+
"learning_rate": 4.605263157894737e-06,
|
| 57 |
+
"loss": 0.5614,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.47619047619047616,
|
| 62 |
+
"grad_norm": 3.3873414602992855,
|
| 63 |
+
"learning_rate": 4.999030264010747e-06,
|
| 64 |
+
"loss": 0.5528,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.5357142857142857,
|
| 69 |
+
"grad_norm": 2.968687348637466,
|
| 70 |
+
"learning_rate": 4.988130331649192e-06,
|
| 71 |
+
"loss": 0.5453,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.5952380952380952,
|
| 76 |
+
"grad_norm": 1.9568832636419962,
|
| 77 |
+
"learning_rate": 4.965177193256699e-06,
|
| 78 |
+
"loss": 0.5341,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.6547619047619048,
|
| 83 |
+
"grad_norm": 2.440397495262987,
|
| 84 |
+
"learning_rate": 4.930294460448816e-06,
|
| 85 |
+
"loss": 0.5308,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.7142857142857143,
|
| 90 |
+
"grad_norm": 2.9778062943530497,
|
| 91 |
+
"learning_rate": 4.88366999036662e-06,
|
| 92 |
+
"loss": 0.5197,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.7738095238095238,
|
| 97 |
+
"grad_norm": 1.8637684628832845,
|
| 98 |
+
"learning_rate": 4.825554873992629e-06,
|
| 99 |
+
"loss": 0.516,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.8333333333333334,
|
| 104 |
+
"grad_norm": 1.3121883792571427,
|
| 105 |
+
"learning_rate": 4.756262083927795e-06,
|
| 106 |
+
"loss": 0.5079,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.8928571428571429,
|
| 111 |
+
"grad_norm": 2.4383628804182473,
|
| 112 |
+
"learning_rate": 4.676164788911806e-06,
|
| 113 |
+
"loss": 0.505,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.9523809523809523,
|
| 118 |
+
"grad_norm": 2.0463100870594877,
|
| 119 |
+
"learning_rate": 4.585694344163653e-06,
|
| 120 |
+
"loss": 0.4986,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.0,
|
| 125 |
+
"eval_loss": 0.06195693090558052,
|
| 126 |
+
"eval_runtime": 116.3718,
|
| 127 |
+
"eval_samples_per_second": 155.562,
|
| 128 |
+
"eval_steps_per_second": 0.61,
|
| 129 |
+
"step": 168
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.0119047619047619,
|
| 133 |
+
"grad_norm": 2.515098432986105,
|
| 134 |
+
"learning_rate": 4.485337968365309e-06,
|
| 135 |
+
"loss": 0.4896,
|
| 136 |
+
"step": 170
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.0714285714285714,
|
| 140 |
+
"grad_norm": 2.13915818649895,
|
| 141 |
+
"learning_rate": 4.375636119798806e-06,
|
| 142 |
+
"loss": 0.4561,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.130952380952381,
|
| 147 |
+
"grad_norm": 2.186912611298388,
|
| 148 |
+
"learning_rate": 4.257179585767301e-06,
|
| 149 |
+
"loss": 0.4533,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.1904761904761905,
|
| 154 |
+
"grad_norm": 1.779282800982749,
|
| 155 |
+
"learning_rate": 4.130606300974687e-06,
|
| 156 |
+
"loss": 0.4518,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.25,
|
| 161 |
+
"grad_norm": 2.25475165594455,
|
| 162 |
+
"learning_rate": 3.996597911998038e-06,
|
| 163 |
+
"loss": 0.4525,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.3095238095238095,
|
| 168 |
+
"grad_norm": 1.557469577660601,
|
| 169 |
+
"learning_rate": 3.855876106354553e-06,
|
| 170 |
+
"loss": 0.4487,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.369047619047619,
|
| 175 |
+
"grad_norm": 1.9579669101720405,
|
| 176 |
+
"learning_rate": 3.7091987259323816e-06,
|
| 177 |
+
"loss": 0.4485,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.4285714285714286,
|
| 182 |
+
"grad_norm": 2.0812279589952936,
|
| 183 |
+
"learning_rate": 3.5573556857160563e-06,
|
| 184 |
+
"loss": 0.4459,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.4880952380952381,
|
| 189 |
+
"grad_norm": 1.4997277173521812,
|
| 190 |
+
"learning_rate": 3.4011647197857655e-06,
|
| 191 |
+
"loss": 0.4426,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.5476190476190477,
|
| 196 |
+
"grad_norm": 1.354646988285002,
|
| 197 |
+
"learning_rate": 3.241466977499929e-06,
|
| 198 |
+
"loss": 0.4414,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.6071428571428572,
|
| 203 |
+
"grad_norm": 1.9258480779001574,
|
| 204 |
+
"learning_rate": 3.0791224935773623e-06,
|
| 205 |
+
"loss": 0.4396,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.6666666666666665,
|
| 210 |
+
"grad_norm": 1.988708522981766,
|
| 211 |
+
"learning_rate": 2.915005556474384e-06,
|
| 212 |
+
"loss": 0.4401,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.7261904761904763,
|
| 217 |
+
"grad_norm": 2.4323525910012975,
|
| 218 |
+
"learning_rate": 2.7500000000000004e-06,
|
| 219 |
+
"loss": 0.4398,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.7857142857142856,
|
| 224 |
+
"grad_norm": 1.678639667621983,
|
| 225 |
+
"learning_rate": 2.584994443525617e-06,
|
| 226 |
+
"loss": 0.4345,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.8452380952380953,
|
| 231 |
+
"grad_norm": 1.1702436863538217,
|
| 232 |
+
"learning_rate": 2.4208775064226384e-06,
|
| 233 |
+
"loss": 0.4333,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.9047619047619047,
|
| 238 |
+
"grad_norm": 1.4024476470543301,
|
| 239 |
+
"learning_rate": 2.258533022500071e-06,
|
| 240 |
+
"loss": 0.4325,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.9642857142857144,
|
| 245 |
+
"grad_norm": 1.1603090253830357,
|
| 246 |
+
"learning_rate": 2.098835280214235e-06,
|
| 247 |
+
"loss": 0.4321,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.0,
|
| 252 |
+
"eval_loss": 0.057941731065511703,
|
| 253 |
+
"eval_runtime": 115.9916,
|
| 254 |
+
"eval_samples_per_second": 156.072,
|
| 255 |
+
"eval_steps_per_second": 0.612,
|
| 256 |
+
"step": 336
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.0238095238095237,
|
| 260 |
+
"grad_norm": 1.9978348550406826,
|
| 261 |
+
"learning_rate": 1.9426443142839448e-06,
|
| 262 |
+
"loss": 0.4132,
|
| 263 |
+
"step": 340
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.0833333333333335,
|
| 267 |
+
"grad_norm": 1.986804137597318,
|
| 268 |
+
"learning_rate": 1.7908012740676195e-06,
|
| 269 |
+
"loss": 0.388,
|
| 270 |
+
"step": 350
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.142857142857143,
|
| 274 |
+
"grad_norm": 1.731363227328417,
|
| 275 |
+
"learning_rate": 1.6441238936454482e-06,
|
| 276 |
+
"loss": 0.3835,
|
| 277 |
+
"step": 360
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.2023809523809526,
|
| 281 |
+
"grad_norm": 2.0119447131912582,
|
| 282 |
+
"learning_rate": 1.503402088001962e-06,
|
| 283 |
+
"loss": 0.3832,
|
| 284 |
+
"step": 370
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.261904761904762,
|
| 288 |
+
"grad_norm": 1.5941336039873037,
|
| 289 |
+
"learning_rate": 1.369393699025314e-06,
|
| 290 |
+
"loss": 0.3858,
|
| 291 |
+
"step": 380
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.3214285714285716,
|
| 295 |
+
"grad_norm": 1.0359242123124663,
|
| 296 |
+
"learning_rate": 1.2428204142327e-06,
|
| 297 |
+
"loss": 0.3821,
|
| 298 |
+
"step": 390
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.380952380952381,
|
| 302 |
+
"grad_norm": 0.8376201687570947,
|
| 303 |
+
"learning_rate": 1.1243638802011956e-06,
|
| 304 |
+
"loss": 0.381,
|
| 305 |
+
"step": 400
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.4404761904761907,
|
| 309 |
+
"grad_norm": 1.0050355147182548,
|
| 310 |
+
"learning_rate": 1.014662031634692e-06,
|
| 311 |
+
"loss": 0.3811,
|
| 312 |
+
"step": 410
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.5,
|
| 316 |
+
"grad_norm": 0.806787956864186,
|
| 317 |
+
"learning_rate": 9.143056558363464e-07,
|
| 318 |
+
"loss": 0.3816,
|
| 319 |
+
"step": 420
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.5595238095238093,
|
| 323 |
+
"grad_norm": 0.7603098782531363,
|
| 324 |
+
"learning_rate": 8.238352110881944e-07,
|
| 325 |
+
"loss": 0.383,
|
| 326 |
+
"step": 430
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.619047619047619,
|
| 330 |
+
"grad_norm": 0.774908703628573,
|
| 331 |
+
"learning_rate": 7.437379160722046e-07,
|
| 332 |
+
"loss": 0.3835,
|
| 333 |
+
"step": 440
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.678571428571429,
|
| 337 |
+
"grad_norm": 0.7922659662804896,
|
| 338 |
+
"learning_rate": 6.744451260073717e-07,
|
| 339 |
+
"loss": 0.3816,
|
| 340 |
+
"step": 450
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.738095238095238,
|
| 344 |
+
"grad_norm": 0.8064438610431734,
|
| 345 |
+
"learning_rate": 6.163300096333806e-07,
|
| 346 |
+
"loss": 0.3827,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.7976190476190474,
|
| 351 |
+
"grad_norm": 0.7992242840549578,
|
| 352 |
+
"learning_rate": 5.697055395511835e-07,
|
| 353 |
+
"loss": 0.3815,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.857142857142857,
|
| 358 |
+
"grad_norm": 0.7341170669407623,
|
| 359 |
+
"learning_rate": 5.348228067433014e-07,
|
| 360 |
+
"loss": 0.3811,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.9166666666666665,
|
| 365 |
+
"grad_norm": 0.7372523390864286,
|
| 366 |
+
"learning_rate": 5.118696683508087e-07,
|
| 367 |
+
"loss": 0.3783,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.9761904761904763,
|
| 372 |
+
"grad_norm": 0.7078290682973725,
|
| 373 |
+
"learning_rate": 5.009697359892537e-07,
|
| 374 |
+
"loss": 0.3803,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.0,
|
| 379 |
+
"eval_loss": 0.05768963322043419,
|
| 380 |
+
"eval_runtime": 114.2584,
|
| 381 |
+
"eval_samples_per_second": 158.439,
|
| 382 |
+
"eval_steps_per_second": 0.621,
|
| 383 |
+
"step": 504
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 3.0,
|
| 387 |
+
"step": 504,
|
| 388 |
+
"total_flos": 3376037568184320.0,
|
| 389 |
+
"train_loss": 0.4685976881356466,
|
| 390 |
+
"train_runtime": 17114.7356,
|
| 391 |
+
"train_samples_per_second": 60.291,
|
| 392 |
+
"train_steps_per_second": 0.029
|
| 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
|