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 +318 -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: Bespoke-Stratos-17k
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# Bespoke-Stratos-17k
|
| 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: 0.5591
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: Bespoke-Stratos-17k
|
|
|
|
| 16 |
|
| 17 |
# Bespoke-Stratos-17k
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on the bespokelabs/Bespoke-Stratos-17k dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5591
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9876543209876543,
|
| 3 |
+
"eval_loss": 0.55907142162323,
|
| 4 |
+
"eval_runtime": 31.1269,
|
| 5 |
+
"eval_samples_per_second": 26.215,
|
| 6 |
+
"eval_steps_per_second": 0.418,
|
| 7 |
+
"total_flos": 190042236518400.0,
|
| 8 |
+
"train_loss": 0.506568502460301,
|
| 9 |
+
"train_runtime": 5156.7482,
|
| 10 |
+
"train_samples_per_second": 9.019,
|
| 11 |
+
"train_steps_per_second": 0.07
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9876543209876543,
|
| 3 |
+
"eval_loss": 0.55907142162323,
|
| 4 |
+
"eval_runtime": 31.1269,
|
| 5 |
+
"eval_samples_per_second": 26.215,
|
| 6 |
+
"eval_steps_per_second": 0.418
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9876543209876543,
|
| 3 |
+
"total_flos": 190042236518400.0,
|
| 4 |
+
"train_loss": 0.506568502460301,
|
| 5 |
+
"train_runtime": 5156.7482,
|
| 6 |
+
"train_samples_per_second": 9.019,
|
| 7 |
+
"train_steps_per_second": 0.07
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,318 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9876543209876543,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 363,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0823045267489712,
|
| 13 |
+
"grad_norm": 2.3707054376650794,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.8018,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1646090534979424,
|
| 20 |
+
"grad_norm": 0.9189518749553744,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.6615,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.24691358024691357,
|
| 27 |
+
"grad_norm": 0.7181231203254296,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.6226,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.3292181069958848,
|
| 34 |
+
"grad_norm": 0.8098373994551952,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.5947,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.411522633744856,
|
| 41 |
+
"grad_norm": 0.6752478716308398,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.5731,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.49382716049382713,
|
| 48 |
+
"grad_norm": 0.6563536956259921,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.5748,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.5761316872427984,
|
| 55 |
+
"grad_norm": 0.7032292694552905,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.5696,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.6584362139917695,
|
| 62 |
+
"grad_norm": 0.7003547749235107,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.5579,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.7407407407407407,
|
| 69 |
+
"grad_norm": 0.6613096075940125,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.563,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.823045267489712,
|
| 76 |
+
"grad_norm": 0.5810433314696346,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.5506,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.9053497942386831,
|
| 83 |
+
"grad_norm": 0.6666365832464547,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.5526,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.9876543209876543,
|
| 90 |
+
"grad_norm": 0.7058806830798787,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.5415,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.9958847736625515,
|
| 97 |
+
"eval_loss": 0.5658594369888306,
|
| 98 |
+
"eval_runtime": 31.1396,
|
| 99 |
+
"eval_samples_per_second": 26.205,
|
| 100 |
+
"eval_steps_per_second": 0.417,
|
| 101 |
+
"step": 121
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.0699588477366255,
|
| 105 |
+
"grad_norm": 0.6935479068001943,
|
| 106 |
+
"learning_rate": 5e-06,
|
| 107 |
+
"loss": 0.5334,
|
| 108 |
+
"step": 130
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.1522633744855968,
|
| 112 |
+
"grad_norm": 0.760550165297717,
|
| 113 |
+
"learning_rate": 5e-06,
|
| 114 |
+
"loss": 0.5067,
|
| 115 |
+
"step": 140
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.2345679012345678,
|
| 119 |
+
"grad_norm": 0.643311514826356,
|
| 120 |
+
"learning_rate": 5e-06,
|
| 121 |
+
"loss": 0.4975,
|
| 122 |
+
"step": 150
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.316872427983539,
|
| 126 |
+
"grad_norm": 0.7180631408695722,
|
| 127 |
+
"learning_rate": 5e-06,
|
| 128 |
+
"loss": 0.4907,
|
| 129 |
+
"step": 160
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.3991769547325104,
|
| 133 |
+
"grad_norm": 0.6268855948646844,
|
| 134 |
+
"learning_rate": 5e-06,
|
| 135 |
+
"loss": 0.4846,
|
| 136 |
+
"step": 170
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.4814814814814814,
|
| 140 |
+
"grad_norm": 0.6205279314065159,
|
| 141 |
+
"learning_rate": 5e-06,
|
| 142 |
+
"loss": 0.4857,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.5637860082304527,
|
| 147 |
+
"grad_norm": 0.6432594466375224,
|
| 148 |
+
"learning_rate": 5e-06,
|
| 149 |
+
"loss": 0.4922,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.646090534979424,
|
| 154 |
+
"grad_norm": 0.5837924660934569,
|
| 155 |
+
"learning_rate": 5e-06,
|
| 156 |
+
"loss": 0.485,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.7283950617283952,
|
| 161 |
+
"grad_norm": 0.5800173415658518,
|
| 162 |
+
"learning_rate": 5e-06,
|
| 163 |
+
"loss": 0.4899,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.8106995884773662,
|
| 168 |
+
"grad_norm": 0.5452696697421119,
|
| 169 |
+
"learning_rate": 5e-06,
|
| 170 |
+
"loss": 0.4794,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.8930041152263375,
|
| 175 |
+
"grad_norm": 0.6589276092411267,
|
| 176 |
+
"learning_rate": 5e-06,
|
| 177 |
+
"loss": 0.4925,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.9753086419753085,
|
| 182 |
+
"grad_norm": 0.7099238180212905,
|
| 183 |
+
"learning_rate": 5e-06,
|
| 184 |
+
"loss": 0.486,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 2.0,
|
| 189 |
+
"eval_loss": 0.552381157875061,
|
| 190 |
+
"eval_runtime": 31.0974,
|
| 191 |
+
"eval_samples_per_second": 26.24,
|
| 192 |
+
"eval_steps_per_second": 0.418,
|
| 193 |
+
"step": 243
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.05761316872428,
|
| 197 |
+
"grad_norm": 0.6173968011853433,
|
| 198 |
+
"learning_rate": 5e-06,
|
| 199 |
+
"loss": 0.4643,
|
| 200 |
+
"step": 250
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.139917695473251,
|
| 204 |
+
"grad_norm": 0.6515589754644828,
|
| 205 |
+
"learning_rate": 5e-06,
|
| 206 |
+
"loss": 0.4264,
|
| 207 |
+
"step": 260
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.2222222222222223,
|
| 211 |
+
"grad_norm": 0.6329003562540839,
|
| 212 |
+
"learning_rate": 5e-06,
|
| 213 |
+
"loss": 0.427,
|
| 214 |
+
"step": 270
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 2.3045267489711936,
|
| 218 |
+
"grad_norm": 0.61321586370113,
|
| 219 |
+
"learning_rate": 5e-06,
|
| 220 |
+
"loss": 0.4284,
|
| 221 |
+
"step": 280
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 2.386831275720165,
|
| 225 |
+
"grad_norm": 0.5986219957694312,
|
| 226 |
+
"learning_rate": 5e-06,
|
| 227 |
+
"loss": 0.4298,
|
| 228 |
+
"step": 290
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.4691358024691357,
|
| 232 |
+
"grad_norm": 0.5797625936229965,
|
| 233 |
+
"learning_rate": 5e-06,
|
| 234 |
+
"loss": 0.4332,
|
| 235 |
+
"step": 300
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 2.551440329218107,
|
| 239 |
+
"grad_norm": 0.6606165599990808,
|
| 240 |
+
"learning_rate": 5e-06,
|
| 241 |
+
"loss": 0.432,
|
| 242 |
+
"step": 310
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.633744855967078,
|
| 246 |
+
"grad_norm": 0.6055776178246806,
|
| 247 |
+
"learning_rate": 5e-06,
|
| 248 |
+
"loss": 0.4297,
|
| 249 |
+
"step": 320
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.7160493827160495,
|
| 253 |
+
"grad_norm": 0.5792227650214353,
|
| 254 |
+
"learning_rate": 5e-06,
|
| 255 |
+
"loss": 0.4301,
|
| 256 |
+
"step": 330
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.7983539094650207,
|
| 260 |
+
"grad_norm": 0.5678659977310005,
|
| 261 |
+
"learning_rate": 5e-06,
|
| 262 |
+
"loss": 0.4256,
|
| 263 |
+
"step": 340
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.8806584362139915,
|
| 267 |
+
"grad_norm": 0.5959535146183862,
|
| 268 |
+
"learning_rate": 5e-06,
|
| 269 |
+
"loss": 0.4252,
|
| 270 |
+
"step": 350
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.962962962962963,
|
| 274 |
+
"grad_norm": 0.624214333713647,
|
| 275 |
+
"learning_rate": 5e-06,
|
| 276 |
+
"loss": 0.4216,
|
| 277 |
+
"step": 360
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.9876543209876543,
|
| 281 |
+
"eval_loss": 0.55907142162323,
|
| 282 |
+
"eval_runtime": 30.0336,
|
| 283 |
+
"eval_samples_per_second": 27.17,
|
| 284 |
+
"eval_steps_per_second": 0.433,
|
| 285 |
+
"step": 363
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.9876543209876543,
|
| 289 |
+
"step": 363,
|
| 290 |
+
"total_flos": 190042236518400.0,
|
| 291 |
+
"train_loss": 0.506568502460301,
|
| 292 |
+
"train_runtime": 5156.7482,
|
| 293 |
+
"train_samples_per_second": 9.019,
|
| 294 |
+
"train_steps_per_second": 0.07
|
| 295 |
+
}
|
| 296 |
+
],
|
| 297 |
+
"logging_steps": 10,
|
| 298 |
+
"max_steps": 363,
|
| 299 |
+
"num_input_tokens_seen": 0,
|
| 300 |
+
"num_train_epochs": 3,
|
| 301 |
+
"save_steps": 500,
|
| 302 |
+
"stateful_callbacks": {
|
| 303 |
+
"TrainerControl": {
|
| 304 |
+
"args": {
|
| 305 |
+
"should_epoch_stop": false,
|
| 306 |
+
"should_evaluate": false,
|
| 307 |
+
"should_log": false,
|
| 308 |
+
"should_save": true,
|
| 309 |
+
"should_training_stop": true
|
| 310 |
+
},
|
| 311 |
+
"attributes": {}
|
| 312 |
+
}
|
| 313 |
+
},
|
| 314 |
+
"total_flos": 190042236518400.0,
|
| 315 |
+
"train_batch_size": 8,
|
| 316 |
+
"trial_name": null,
|
| 317 |
+
"trial_params": null
|
| 318 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|