Commit
•
f400082
1
Parent(s):
fca16bb
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 +479 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: llama3.1
|
|
| 4 |
base_model: meta-llama/Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: OH_original_wo_slimorca_550k
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# OH_original_wo_slimorca_550k
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.5386
|
| 21 |
|
|
|
|
| 4 |
base_model: meta-llama/Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: OH_original_wo_slimorca_550k
|
|
|
|
| 16 |
|
| 17 |
# OH_original_wo_slimorca_550k
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on the mlfoundations-dev/OH_original_wo_slimorca_550k dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5386
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.5386124849319458,
|
| 4 |
+
"eval_runtime": 67.4227,
|
| 5 |
+
"eval_samples_per_second": 78.712,
|
| 6 |
+
"eval_steps_per_second": 0.623,
|
| 7 |
+
"total_flos": 989528252743680.0,
|
| 8 |
+
"train_loss": 0.5134775242224563,
|
| 9 |
+
"train_runtime": 10543.3353,
|
| 10 |
+
"train_samples_per_second": 28.688,
|
| 11 |
+
"train_steps_per_second": 0.056
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.5386124849319458,
|
| 4 |
+
"eval_runtime": 67.4227,
|
| 5 |
+
"eval_samples_per_second": 78.712,
|
| 6 |
+
"eval_steps_per_second": 0.623
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 989528252743680.0,
|
| 4 |
+
"train_loss": 0.5134775242224563,
|
| 5 |
+
"train_runtime": 10543.3353,
|
| 6 |
+
"train_samples_per_second": 28.688,
|
| 7 |
+
"train_steps_per_second": 0.056
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,479 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 591,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.050761421319796954,
|
| 13 |
+
"grad_norm": 2.884176452669992,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.7058,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.10152284263959391,
|
| 20 |
+
"grad_norm": 1.1031748057601416,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.6396,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.15228426395939088,
|
| 27 |
+
"grad_norm": 0.8763770464242356,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.6263,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.20304568527918782,
|
| 34 |
+
"grad_norm": 0.8637694735287882,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.6163,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.25380710659898476,
|
| 41 |
+
"grad_norm": 0.6443562464039193,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.5983,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.30456852791878175,
|
| 48 |
+
"grad_norm": 0.7181436435893084,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.5841,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.3553299492385787,
|
| 55 |
+
"grad_norm": 0.5736245789251513,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.5851,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.40609137055837563,
|
| 62 |
+
"grad_norm": 0.5370831215793392,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.5743,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.45685279187817257,
|
| 69 |
+
"grad_norm": 0.7868112154827747,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.5631,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.5076142131979695,
|
| 76 |
+
"grad_norm": 0.6384967672565564,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.5621,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.5583756345177665,
|
| 83 |
+
"grad_norm": 0.6105632833081082,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.5626,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.6091370558375635,
|
| 90 |
+
"grad_norm": 0.4980885433595067,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.5576,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.6598984771573604,
|
| 97 |
+
"grad_norm": 0.5567494710307067,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.5634,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.7106598984771574,
|
| 104 |
+
"grad_norm": 0.5747955173184647,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.5556,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.7614213197969543,
|
| 111 |
+
"grad_norm": 0.7422760458895346,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.5485,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.8121827411167513,
|
| 118 |
+
"grad_norm": 0.5519407925537647,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.5555,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.8629441624365483,
|
| 125 |
+
"grad_norm": 0.3938384364399605,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.5534,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.9137055837563451,
|
| 132 |
+
"grad_norm": 0.4274758585586587,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.5504,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.9644670050761421,
|
| 139 |
+
"grad_norm": 0.4623946226615239,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.5428,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.0,
|
| 146 |
+
"eval_loss": 0.545667827129364,
|
| 147 |
+
"eval_runtime": 69.9223,
|
| 148 |
+
"eval_samples_per_second": 75.899,
|
| 149 |
+
"eval_steps_per_second": 0.601,
|
| 150 |
+
"step": 197
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.015228426395939,
|
| 154 |
+
"grad_norm": 0.4892463181375632,
|
| 155 |
+
"learning_rate": 5e-06,
|
| 156 |
+
"loss": 0.538,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.0659898477157361,
|
| 161 |
+
"grad_norm": 0.5258103265854674,
|
| 162 |
+
"learning_rate": 5e-06,
|
| 163 |
+
"loss": 0.5005,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.116751269035533,
|
| 168 |
+
"grad_norm": 0.43953346933881265,
|
| 169 |
+
"learning_rate": 5e-06,
|
| 170 |
+
"loss": 0.5116,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.16751269035533,
|
| 175 |
+
"grad_norm": 0.555350254923011,
|
| 176 |
+
"learning_rate": 5e-06,
|
| 177 |
+
"loss": 0.5039,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.218274111675127,
|
| 182 |
+
"grad_norm": 0.4240258545882722,
|
| 183 |
+
"learning_rate": 5e-06,
|
| 184 |
+
"loss": 0.4979,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.2690355329949239,
|
| 189 |
+
"grad_norm": 0.6055626365057643,
|
| 190 |
+
"learning_rate": 5e-06,
|
| 191 |
+
"loss": 0.4977,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.3197969543147208,
|
| 196 |
+
"grad_norm": 0.5183814019968731,
|
| 197 |
+
"learning_rate": 5e-06,
|
| 198 |
+
"loss": 0.5015,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.3705583756345177,
|
| 203 |
+
"grad_norm": 0.44198994392173363,
|
| 204 |
+
"learning_rate": 5e-06,
|
| 205 |
+
"loss": 0.5052,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.4213197969543148,
|
| 210 |
+
"grad_norm": 0.4597615345223239,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.4974,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.4720812182741116,
|
| 217 |
+
"grad_norm": 0.47036560762175594,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.5044,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.5228426395939088,
|
| 224 |
+
"grad_norm": 0.4356873877920564,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.5134,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.5736040609137056,
|
| 231 |
+
"grad_norm": 0.4547653150275499,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.5022,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.6243654822335025,
|
| 238 |
+
"grad_norm": 0.5243750306692014,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.5001,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.6751269035532994,
|
| 245 |
+
"grad_norm": 0.5072947429791255,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.504,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.7258883248730963,
|
| 252 |
+
"grad_norm": 0.4591167012346128,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.5008,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.7766497461928934,
|
| 259 |
+
"grad_norm": 0.559953933092174,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.5058,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.8274111675126905,
|
| 266 |
+
"grad_norm": 0.4615691289798925,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.5013,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.8781725888324874,
|
| 273 |
+
"grad_norm": 0.4673102096867509,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.505,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.9289340101522843,
|
| 280 |
+
"grad_norm": 0.41382969409293313,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.5029,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.9796954314720812,
|
| 287 |
+
"grad_norm": 0.4786556881881671,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.4991,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 2.0,
|
| 294 |
+
"eval_loss": 0.5367357730865479,
|
| 295 |
+
"eval_runtime": 69.7466,
|
| 296 |
+
"eval_samples_per_second": 76.09,
|
| 297 |
+
"eval_steps_per_second": 0.602,
|
| 298 |
+
"step": 394
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.030456852791878,
|
| 302 |
+
"grad_norm": 0.7058322019162009,
|
| 303 |
+
"learning_rate": 5e-06,
|
| 304 |
+
"loss": 0.4785,
|
| 305 |
+
"step": 400
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.081218274111675,
|
| 309 |
+
"grad_norm": 0.45765334889032916,
|
| 310 |
+
"learning_rate": 5e-06,
|
| 311 |
+
"loss": 0.4563,
|
| 312 |
+
"step": 410
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.1319796954314723,
|
| 316 |
+
"grad_norm": 0.4288566534513188,
|
| 317 |
+
"learning_rate": 5e-06,
|
| 318 |
+
"loss": 0.4564,
|
| 319 |
+
"step": 420
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.182741116751269,
|
| 323 |
+
"grad_norm": 0.4543531173101596,
|
| 324 |
+
"learning_rate": 5e-06,
|
| 325 |
+
"loss": 0.4559,
|
| 326 |
+
"step": 430
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.233502538071066,
|
| 330 |
+
"grad_norm": 0.5677467898296061,
|
| 331 |
+
"learning_rate": 5e-06,
|
| 332 |
+
"loss": 0.454,
|
| 333 |
+
"step": 440
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.284263959390863,
|
| 337 |
+
"grad_norm": 0.5213901371480834,
|
| 338 |
+
"learning_rate": 5e-06,
|
| 339 |
+
"loss": 0.4584,
|
| 340 |
+
"step": 450
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.33502538071066,
|
| 344 |
+
"grad_norm": 0.5106205098926287,
|
| 345 |
+
"learning_rate": 5e-06,
|
| 346 |
+
"loss": 0.4603,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.3857868020304567,
|
| 351 |
+
"grad_norm": 0.48058806334750254,
|
| 352 |
+
"learning_rate": 5e-06,
|
| 353 |
+
"loss": 0.4586,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.436548223350254,
|
| 358 |
+
"grad_norm": 0.4462504100857437,
|
| 359 |
+
"learning_rate": 5e-06,
|
| 360 |
+
"loss": 0.4547,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.487309644670051,
|
| 365 |
+
"grad_norm": 0.47373720391622604,
|
| 366 |
+
"learning_rate": 5e-06,
|
| 367 |
+
"loss": 0.4553,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.5380710659898478,
|
| 372 |
+
"grad_norm": 0.44268127361884035,
|
| 373 |
+
"learning_rate": 5e-06,
|
| 374 |
+
"loss": 0.4552,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.5888324873096447,
|
| 379 |
+
"grad_norm": 0.5168789659092264,
|
| 380 |
+
"learning_rate": 5e-06,
|
| 381 |
+
"loss": 0.4495,
|
| 382 |
+
"step": 510
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.6395939086294415,
|
| 386 |
+
"grad_norm": 0.4759213683260235,
|
| 387 |
+
"learning_rate": 5e-06,
|
| 388 |
+
"loss": 0.4601,
|
| 389 |
+
"step": 520
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.6903553299492384,
|
| 393 |
+
"grad_norm": 0.5405150525609788,
|
| 394 |
+
"learning_rate": 5e-06,
|
| 395 |
+
"loss": 0.4639,
|
| 396 |
+
"step": 530
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.7411167512690353,
|
| 400 |
+
"grad_norm": 0.48554448131156325,
|
| 401 |
+
"learning_rate": 5e-06,
|
| 402 |
+
"loss": 0.4561,
|
| 403 |
+
"step": 540
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.7918781725888326,
|
| 407 |
+
"grad_norm": 0.5715450026884347,
|
| 408 |
+
"learning_rate": 5e-06,
|
| 409 |
+
"loss": 0.458,
|
| 410 |
+
"step": 550
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.8426395939086295,
|
| 414 |
+
"grad_norm": 0.45709891570844524,
|
| 415 |
+
"learning_rate": 5e-06,
|
| 416 |
+
"loss": 0.4582,
|
| 417 |
+
"step": 560
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.8934010152284264,
|
| 421 |
+
"grad_norm": 0.4329230281361173,
|
| 422 |
+
"learning_rate": 5e-06,
|
| 423 |
+
"loss": 0.4535,
|
| 424 |
+
"step": 570
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.9441624365482233,
|
| 428 |
+
"grad_norm": 0.4707895543051601,
|
| 429 |
+
"learning_rate": 5e-06,
|
| 430 |
+
"loss": 0.4609,
|
| 431 |
+
"step": 580
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.99492385786802,
|
| 435 |
+
"grad_norm": 0.4717664390758382,
|
| 436 |
+
"learning_rate": 5e-06,
|
| 437 |
+
"loss": 0.4636,
|
| 438 |
+
"step": 590
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 3.0,
|
| 442 |
+
"eval_loss": 0.5386124849319458,
|
| 443 |
+
"eval_runtime": 67.8364,
|
| 444 |
+
"eval_samples_per_second": 78.232,
|
| 445 |
+
"eval_steps_per_second": 0.619,
|
| 446 |
+
"step": 591
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 3.0,
|
| 450 |
+
"step": 591,
|
| 451 |
+
"total_flos": 989528252743680.0,
|
| 452 |
+
"train_loss": 0.5134775242224563,
|
| 453 |
+
"train_runtime": 10543.3353,
|
| 454 |
+
"train_samples_per_second": 28.688,
|
| 455 |
+
"train_steps_per_second": 0.056
|
| 456 |
+
}
|
| 457 |
+
],
|
| 458 |
+
"logging_steps": 10,
|
| 459 |
+
"max_steps": 591,
|
| 460 |
+
"num_input_tokens_seen": 0,
|
| 461 |
+
"num_train_epochs": 3,
|
| 462 |
+
"save_steps": 500,
|
| 463 |
+
"stateful_callbacks": {
|
| 464 |
+
"TrainerControl": {
|
| 465 |
+
"args": {
|
| 466 |
+
"should_epoch_stop": false,
|
| 467 |
+
"should_evaluate": false,
|
| 468 |
+
"should_log": false,
|
| 469 |
+
"should_save": true,
|
| 470 |
+
"should_training_stop": true
|
| 471 |
+
},
|
| 472 |
+
"attributes": {}
|
| 473 |
+
}
|
| 474 |
+
},
|
| 475 |
+
"total_flos": 989528252743680.0,
|
| 476 |
+
"train_batch_size": 8,
|
| 477 |
+
"trial_name": null,
|
| 478 |
+
"trial_params": null
|
| 479 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|