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 +402 -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/Meta-Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: oh_teknium_scaling_down_random_0.5
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# oh_teknium_scaling_down_random_0.5
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.5205
|
| 21 |
|
|
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: oh_teknium_scaling_down_random_0.5
|
|
|
|
| 16 |
|
| 17 |
# oh_teknium_scaling_down_random_0.5
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/oh_teknium_scaling_down_random_0.5 dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5205
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9976580796252925,
|
| 3 |
+
"eval_loss": 0.5204988121986389,
|
| 4 |
+
"eval_runtime": 109.268,
|
| 5 |
+
"eval_samples_per_second": 39.472,
|
| 6 |
+
"eval_steps_per_second": 0.622,
|
| 7 |
+
"total_flos": 803808498155520.0,
|
| 8 |
+
"train_loss": 0.5004926055669785,
|
| 9 |
+
"train_runtime": 16013.1849,
|
| 10 |
+
"train_samples_per_second": 15.351,
|
| 11 |
+
"train_steps_per_second": 0.03
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9976580796252925,
|
| 3 |
+
"eval_loss": 0.5204988121986389,
|
| 4 |
+
"eval_runtime": 109.268,
|
| 5 |
+
"eval_samples_per_second": 39.472,
|
| 6 |
+
"eval_steps_per_second": 0.622
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9976580796252925,
|
| 3 |
+
"total_flos": 803808498155520.0,
|
| 4 |
+
"train_loss": 0.5004926055669785,
|
| 5 |
+
"train_runtime": 16013.1849,
|
| 6 |
+
"train_samples_per_second": 15.351,
|
| 7 |
+
"train_steps_per_second": 0.03
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,402 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9976580796252925,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 480,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0624512099921936,
|
| 13 |
+
"grad_norm": 5.695734079803832,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.7013,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1249024199843872,
|
| 20 |
+
"grad_norm": 1.6672905613542388,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.6275,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.1873536299765808,
|
| 27 |
+
"grad_norm": 1.231437836117211,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.6011,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2498048399687744,
|
| 34 |
+
"grad_norm": 0.6660382816468988,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.5905,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.312256049960968,
|
| 41 |
+
"grad_norm": 0.6878662625368892,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.5739,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.3747072599531616,
|
| 48 |
+
"grad_norm": 0.8735325294404173,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.5545,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.4371584699453552,
|
| 55 |
+
"grad_norm": 0.865826564748345,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.5532,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.4996096799375488,
|
| 62 |
+
"grad_norm": 0.6254769035827942,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.5474,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.5620608899297423,
|
| 69 |
+
"grad_norm": 0.633780736055842,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.5477,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.624512099921936,
|
| 76 |
+
"grad_norm": 0.429646086801005,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.5536,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.6869633099141296,
|
| 83 |
+
"grad_norm": 0.47226853053932544,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.547,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.7494145199063232,
|
| 90 |
+
"grad_norm": 0.5108448433850594,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.5456,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.8118657298985168,
|
| 97 |
+
"grad_norm": 0.5120890633578824,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.5338,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.8743169398907104,
|
| 104 |
+
"grad_norm": 0.5054688976629029,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.5415,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.936768149882904,
|
| 111 |
+
"grad_norm": 0.4546611078615034,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.5266,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.9992193598750976,
|
| 118 |
+
"grad_norm": 0.5777441256544157,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.5327,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.9992193598750976,
|
| 125 |
+
"eval_loss": 0.5309467315673828,
|
| 126 |
+
"eval_runtime": 112.9263,
|
| 127 |
+
"eval_samples_per_second": 38.193,
|
| 128 |
+
"eval_steps_per_second": 0.602,
|
| 129 |
+
"step": 160
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.0616705698672912,
|
| 133 |
+
"grad_norm": 0.56357097034926,
|
| 134 |
+
"learning_rate": 5e-06,
|
| 135 |
+
"loss": 0.5284,
|
| 136 |
+
"step": 170
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.1241217798594847,
|
| 140 |
+
"grad_norm": 0.6406606004904944,
|
| 141 |
+
"learning_rate": 5e-06,
|
| 142 |
+
"loss": 0.491,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.1865729898516784,
|
| 147 |
+
"grad_norm": 0.4470486896482656,
|
| 148 |
+
"learning_rate": 5e-06,
|
| 149 |
+
"loss": 0.4924,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.249024199843872,
|
| 154 |
+
"grad_norm": 0.43090473351959857,
|
| 155 |
+
"learning_rate": 5e-06,
|
| 156 |
+
"loss": 0.4824,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.3114754098360657,
|
| 161 |
+
"grad_norm": 0.43197317838800425,
|
| 162 |
+
"learning_rate": 5e-06,
|
| 163 |
+
"loss": 0.4911,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.3739266198282591,
|
| 168 |
+
"grad_norm": 0.49491142532222065,
|
| 169 |
+
"learning_rate": 5e-06,
|
| 170 |
+
"loss": 0.4888,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.4363778298204528,
|
| 175 |
+
"grad_norm": 0.4582690739604794,
|
| 176 |
+
"learning_rate": 5e-06,
|
| 177 |
+
"loss": 0.4922,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.4988290398126463,
|
| 182 |
+
"grad_norm": 0.5179775405720543,
|
| 183 |
+
"learning_rate": 5e-06,
|
| 184 |
+
"loss": 0.4902,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.56128024980484,
|
| 189 |
+
"grad_norm": 0.46044165550988314,
|
| 190 |
+
"learning_rate": 5e-06,
|
| 191 |
+
"loss": 0.4853,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.6237314597970336,
|
| 196 |
+
"grad_norm": 0.49784346733707135,
|
| 197 |
+
"learning_rate": 5e-06,
|
| 198 |
+
"loss": 0.4885,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.6861826697892273,
|
| 203 |
+
"grad_norm": 0.45069349680457893,
|
| 204 |
+
"learning_rate": 5e-06,
|
| 205 |
+
"loss": 0.486,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.748633879781421,
|
| 210 |
+
"grad_norm": 0.44312447239988956,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.4865,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.8110850897736144,
|
| 217 |
+
"grad_norm": 0.5540661328732004,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.4917,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.8735362997658078,
|
| 224 |
+
"grad_norm": 0.4234046253558461,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.4828,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.9359875097580015,
|
| 231 |
+
"grad_norm": 0.4388004925712562,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.4789,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.9984387197501952,
|
| 238 |
+
"grad_norm": 0.5610659198792445,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.4787,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.9984387197501952,
|
| 245 |
+
"eval_loss": 0.5198609828948975,
|
| 246 |
+
"eval_runtime": 110.8922,
|
| 247 |
+
"eval_samples_per_second": 38.894,
|
| 248 |
+
"eval_steps_per_second": 0.613,
|
| 249 |
+
"step": 320
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.060889929742389,
|
| 253 |
+
"grad_norm": 0.5484717697155285,
|
| 254 |
+
"learning_rate": 5e-06,
|
| 255 |
+
"loss": 0.4873,
|
| 256 |
+
"step": 330
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.1233411397345825,
|
| 260 |
+
"grad_norm": 0.45228929433011156,
|
| 261 |
+
"learning_rate": 5e-06,
|
| 262 |
+
"loss": 0.4362,
|
| 263 |
+
"step": 340
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.185792349726776,
|
| 267 |
+
"grad_norm": 0.5293854629631661,
|
| 268 |
+
"learning_rate": 5e-06,
|
| 269 |
+
"loss": 0.4425,
|
| 270 |
+
"step": 350
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.2482435597189694,
|
| 274 |
+
"grad_norm": 0.6002239962428959,
|
| 275 |
+
"learning_rate": 5e-06,
|
| 276 |
+
"loss": 0.4355,
|
| 277 |
+
"step": 360
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.310694769711163,
|
| 281 |
+
"grad_norm": 0.5258587782431947,
|
| 282 |
+
"learning_rate": 5e-06,
|
| 283 |
+
"loss": 0.4384,
|
| 284 |
+
"step": 370
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.3731459797033567,
|
| 288 |
+
"grad_norm": 0.5243329962737817,
|
| 289 |
+
"learning_rate": 5e-06,
|
| 290 |
+
"loss": 0.4381,
|
| 291 |
+
"step": 380
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.4355971896955504,
|
| 295 |
+
"grad_norm": 0.5890309295737689,
|
| 296 |
+
"learning_rate": 5e-06,
|
| 297 |
+
"loss": 0.4425,
|
| 298 |
+
"step": 390
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.498048399687744,
|
| 302 |
+
"grad_norm": 0.5358514650116517,
|
| 303 |
+
"learning_rate": 5e-06,
|
| 304 |
+
"loss": 0.4439,
|
| 305 |
+
"step": 400
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.5604996096799377,
|
| 309 |
+
"grad_norm": 0.4710746244266771,
|
| 310 |
+
"learning_rate": 5e-06,
|
| 311 |
+
"loss": 0.4434,
|
| 312 |
+
"step": 410
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.6229508196721314,
|
| 316 |
+
"grad_norm": 0.4783994299645445,
|
| 317 |
+
"learning_rate": 5e-06,
|
| 318 |
+
"loss": 0.4376,
|
| 319 |
+
"step": 420
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.6854020296643246,
|
| 323 |
+
"grad_norm": 0.5532060358770908,
|
| 324 |
+
"learning_rate": 5e-06,
|
| 325 |
+
"loss": 0.4418,
|
| 326 |
+
"step": 430
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.7478532396565183,
|
| 330 |
+
"grad_norm": 0.5357167310213664,
|
| 331 |
+
"learning_rate": 5e-06,
|
| 332 |
+
"loss": 0.4489,
|
| 333 |
+
"step": 440
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.810304449648712,
|
| 337 |
+
"grad_norm": 0.49033181773774204,
|
| 338 |
+
"learning_rate": 5e-06,
|
| 339 |
+
"loss": 0.4446,
|
| 340 |
+
"step": 450
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.8727556596409056,
|
| 344 |
+
"grad_norm": 0.4295027030082963,
|
| 345 |
+
"learning_rate": 5e-06,
|
| 346 |
+
"loss": 0.4427,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.9352068696330993,
|
| 351 |
+
"grad_norm": 0.5261820799303164,
|
| 352 |
+
"learning_rate": 5e-06,
|
| 353 |
+
"loss": 0.4448,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.9976580796252925,
|
| 358 |
+
"grad_norm": 0.45577050194466817,
|
| 359 |
+
"learning_rate": 5e-06,
|
| 360 |
+
"loss": 0.443,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.9976580796252925,
|
| 365 |
+
"eval_loss": 0.5204988121986389,
|
| 366 |
+
"eval_runtime": 106.3066,
|
| 367 |
+
"eval_samples_per_second": 40.571,
|
| 368 |
+
"eval_steps_per_second": 0.64,
|
| 369 |
+
"step": 480
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.9976580796252925,
|
| 373 |
+
"step": 480,
|
| 374 |
+
"total_flos": 803808498155520.0,
|
| 375 |
+
"train_loss": 0.5004926055669785,
|
| 376 |
+
"train_runtime": 16013.1849,
|
| 377 |
+
"train_samples_per_second": 15.351,
|
| 378 |
+
"train_steps_per_second": 0.03
|
| 379 |
+
}
|
| 380 |
+
],
|
| 381 |
+
"logging_steps": 10,
|
| 382 |
+
"max_steps": 480,
|
| 383 |
+
"num_input_tokens_seen": 0,
|
| 384 |
+
"num_train_epochs": 3,
|
| 385 |
+
"save_steps": 500,
|
| 386 |
+
"stateful_callbacks": {
|
| 387 |
+
"TrainerControl": {
|
| 388 |
+
"args": {
|
| 389 |
+
"should_epoch_stop": false,
|
| 390 |
+
"should_evaluate": false,
|
| 391 |
+
"should_log": false,
|
| 392 |
+
"should_save": true,
|
| 393 |
+
"should_training_stop": true
|
| 394 |
+
},
|
| 395 |
+
"attributes": {}
|
| 396 |
+
}
|
| 397 |
+
},
|
| 398 |
+
"total_flos": 803808498155520.0,
|
| 399 |
+
"train_batch_size": 8,
|
| 400 |
+
"trial_name": null,
|
| 401 |
+
"trial_params": null
|
| 402 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|