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 +288 -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: Qwen/Qwen2.5-7B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: hp_ablations_qwen_epoch1_dcftv1.2
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# hp_ablations_qwen_epoch1_dcftv1.2
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.6407
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: hp_ablations_qwen_epoch1_dcftv1.2
|
|
|
|
| 16 |
|
| 17 |
# hp_ablations_qwen_epoch1_dcftv1.2
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on the mlfoundations-dev/oh-dcft-v1.2_no-curation_gpt-4o-mini dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.6407
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9978054133138259,
|
| 3 |
+
"eval_loss": 0.6406828165054321,
|
| 4 |
+
"eval_runtime": 345.7045,
|
| 5 |
+
"eval_samples_per_second": 26.638,
|
| 6 |
+
"eval_steps_per_second": 0.417,
|
| 7 |
+
"total_flos": 714820936531968.0,
|
| 8 |
+
"train_loss": 0.6610767383379671,
|
| 9 |
+
"train_runtime": 18392.4604,
|
| 10 |
+
"train_samples_per_second": 9.512,
|
| 11 |
+
"train_steps_per_second": 0.019
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9978054133138259,
|
| 3 |
+
"eval_loss": 0.6406828165054321,
|
| 4 |
+
"eval_runtime": 345.7045,
|
| 5 |
+
"eval_samples_per_second": 26.638,
|
| 6 |
+
"eval_steps_per_second": 0.417
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9978054133138259,
|
| 3 |
+
"total_flos": 714820936531968.0,
|
| 4 |
+
"train_loss": 0.6610767383379671,
|
| 5 |
+
"train_runtime": 18392.4604,
|
| 6 |
+
"train_samples_per_second": 9.512,
|
| 7 |
+
"train_steps_per_second": 0.019
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,288 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9978054133138259,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 341,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.029261155815654718,
|
| 13 |
+
"grad_norm": 1.3814841088477259,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.7983,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.058522311631309436,
|
| 20 |
+
"grad_norm": 1.2887254498462366,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.7383,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.08778346744696415,
|
| 27 |
+
"grad_norm": 1.2838337638318262,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.7071,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.11704462326261887,
|
| 34 |
+
"grad_norm": 1.0922966245519412,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.6966,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.14630577907827358,
|
| 41 |
+
"grad_norm": 1.1509702467959955,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6871,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1755669348939283,
|
| 48 |
+
"grad_norm": 0.9131326378339343,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6766,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.20482809070958302,
|
| 55 |
+
"grad_norm": 0.6023605382305695,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.67,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.23408924652523774,
|
| 62 |
+
"grad_norm": 0.4161189143299673,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.6647,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.26335040234089246,
|
| 69 |
+
"grad_norm": 0.4292872692725048,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.662,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.29261155815654716,
|
| 76 |
+
"grad_norm": 0.4514981368981087,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.6544,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.3218727139722019,
|
| 83 |
+
"grad_norm": 0.42808475020683995,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6632,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.3511338697878566,
|
| 90 |
+
"grad_norm": 0.37727288970274103,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.6681,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.38039502560351135,
|
| 97 |
+
"grad_norm": 0.3822819962955356,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.6519,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.40965618141916604,
|
| 104 |
+
"grad_norm": 0.36899229568727576,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6526,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.4389173372348208,
|
| 111 |
+
"grad_norm": 0.3468922138887062,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.648,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.4681784930504755,
|
| 118 |
+
"grad_norm": 0.3795489753723712,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.6499,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.49743964886613024,
|
| 125 |
+
"grad_norm": 0.36716418800909245,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.655,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5267008046817849,
|
| 132 |
+
"grad_norm": 0.33393389585281086,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.6546,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.5559619604974396,
|
| 139 |
+
"grad_norm": 0.3581250031024453,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.6431,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5852231163130943,
|
| 146 |
+
"grad_norm": 0.34112480920310834,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.6443,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.6144842721287491,
|
| 153 |
+
"grad_norm": 0.3396720319535308,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.6509,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.6437454279444038,
|
| 160 |
+
"grad_norm": 0.354125295809357,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.6389,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.6730065837600585,
|
| 167 |
+
"grad_norm": 0.3606633109240175,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.6388,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.7022677395757132,
|
| 174 |
+
"grad_norm": 0.3251299701359152,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.6457,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.731528895391368,
|
| 181 |
+
"grad_norm": 0.33939812328375596,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.6441,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.7607900512070227,
|
| 188 |
+
"grad_norm": 0.3366111360440969,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.6467,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.7900512070226774,
|
| 195 |
+
"grad_norm": 0.37753830325225896,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.649,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.8193123628383321,
|
| 202 |
+
"grad_norm": 0.34478553790521904,
|
| 203 |
+
"learning_rate": 5e-06,
|
| 204 |
+
"loss": 0.6477,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.8485735186539868,
|
| 209 |
+
"grad_norm": 0.33926125432670784,
|
| 210 |
+
"learning_rate": 5e-06,
|
| 211 |
+
"loss": 0.64,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.8778346744696416,
|
| 216 |
+
"grad_norm": 0.3502233770134374,
|
| 217 |
+
"learning_rate": 5e-06,
|
| 218 |
+
"loss": 0.6318,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.9070958302852963,
|
| 223 |
+
"grad_norm": 0.34993547260225705,
|
| 224 |
+
"learning_rate": 5e-06,
|
| 225 |
+
"loss": 0.6449,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.936356986100951,
|
| 230 |
+
"grad_norm": 0.3358780908708201,
|
| 231 |
+
"learning_rate": 5e-06,
|
| 232 |
+
"loss": 0.6369,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.9656181419166057,
|
| 237 |
+
"grad_norm": 0.37650022189557825,
|
| 238 |
+
"learning_rate": 5e-06,
|
| 239 |
+
"loss": 0.6402,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.9948792977322605,
|
| 244 |
+
"grad_norm": 0.3494056493360394,
|
| 245 |
+
"learning_rate": 5e-06,
|
| 246 |
+
"loss": 0.6362,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.9978054133138259,
|
| 251 |
+
"eval_loss": 0.6406828165054321,
|
| 252 |
+
"eval_runtime": 343.5643,
|
| 253 |
+
"eval_samples_per_second": 26.804,
|
| 254 |
+
"eval_steps_per_second": 0.419,
|
| 255 |
+
"step": 341
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.9978054133138259,
|
| 259 |
+
"step": 341,
|
| 260 |
+
"total_flos": 714820936531968.0,
|
| 261 |
+
"train_loss": 0.6610767383379671,
|
| 262 |
+
"train_runtime": 18392.4604,
|
| 263 |
+
"train_samples_per_second": 9.512,
|
| 264 |
+
"train_steps_per_second": 0.019
|
| 265 |
+
}
|
| 266 |
+
],
|
| 267 |
+
"logging_steps": 10,
|
| 268 |
+
"max_steps": 341,
|
| 269 |
+
"num_input_tokens_seen": 0,
|
| 270 |
+
"num_train_epochs": 1,
|
| 271 |
+
"save_steps": 500,
|
| 272 |
+
"stateful_callbacks": {
|
| 273 |
+
"TrainerControl": {
|
| 274 |
+
"args": {
|
| 275 |
+
"should_epoch_stop": false,
|
| 276 |
+
"should_evaluate": false,
|
| 277 |
+
"should_log": false,
|
| 278 |
+
"should_save": true,
|
| 279 |
+
"should_training_stop": true
|
| 280 |
+
},
|
| 281 |
+
"attributes": {}
|
| 282 |
+
}
|
| 283 |
+
},
|
| 284 |
+
"total_flos": 714820936531968.0,
|
| 285 |
+
"train_batch_size": 8,
|
| 286 |
+
"trial_name": null,
|
| 287 |
+
"trial_params": null
|
| 288 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|