End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 29.16,
|
| 3 |
+
"eval_accuracy": 0.05126700710676839,
|
| 4 |
+
"eval_loss": 7.343224048614502,
|
| 5 |
+
"eval_runtime": 145.4722,
|
| 6 |
+
"eval_samples": 24055,
|
| 7 |
+
"eval_samples_per_second": 165.358,
|
| 8 |
+
"eval_steps_per_second": 5.169,
|
| 9 |
+
"perplexity": 1545.687460491957,
|
| 10 |
+
"train_loss": 7.568343920026507,
|
| 11 |
+
"train_runtime": 10701.7058,
|
| 12 |
+
"train_samples": 24910,
|
| 13 |
+
"train_samples_per_second": 232.767,
|
| 14 |
+
"train_steps_per_second": 0.112
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 29.16,
|
| 3 |
+
"eval_accuracy": 0.05126700710676839,
|
| 4 |
+
"eval_loss": 7.343224048614502,
|
| 5 |
+
"eval_runtime": 145.4722,
|
| 6 |
+
"eval_samples": 24055,
|
| 7 |
+
"eval_samples_per_second": 165.358,
|
| 8 |
+
"eval_steps_per_second": 5.169,
|
| 9 |
+
"perplexity": 1545.687460491957
|
| 10 |
+
}
|
runs/Feb15_23-43-00_monica.sm.unibocconi.it/events.out.tfevents.1676511884.monica.sm.unibocconi.it.1476666.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60e838b6f6b2297035e7510ccebdc969ba4811352956dd96ca299885477197a6
|
| 3 |
+
size 363
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 29.16,
|
| 3 |
+
"train_loss": 7.568343920026507,
|
| 4 |
+
"train_runtime": 10701.7058,
|
| 5 |
+
"train_samples": 24910,
|
| 6 |
+
"train_samples_per_second": 232.767,
|
| 7 |
+
"train_steps_per_second": 0.112
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,298 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 7.343978404998779,
|
| 3 |
+
"best_model_checkpoint": "/data1/attanasiog/babylm/roberta-tiny-4l-10M/checkpoint-200",
|
| 4 |
+
"epoch": 29.16441875401413,
|
| 5 |
+
"global_step": 350,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.82,
|
| 12 |
+
"learning_rate": 0.00014000000000000001,
|
| 13 |
+
"loss": 10.2422,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.66,
|
| 18 |
+
"learning_rate": 0.00028000000000000003,
|
| 19 |
+
"loss": 8.631,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 2.49,
|
| 24 |
+
"learning_rate": 0.00041999999999999996,
|
| 25 |
+
"loss": 7.5098,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 3.33,
|
| 30 |
+
"learning_rate": 0.0005600000000000001,
|
| 31 |
+
"loss": 9.1052,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 4.16,
|
| 36 |
+
"learning_rate": 0.0007,
|
| 37 |
+
"loss": 7.4785,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.16,
|
| 42 |
+
"eval_accuracy": 0.051358309226408695,
|
| 43 |
+
"eval_loss": 7.383351802825928,
|
| 44 |
+
"eval_runtime": 145.6157,
|
| 45 |
+
"eval_samples_per_second": 165.195,
|
| 46 |
+
"eval_steps_per_second": 5.164,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 4.99,
|
| 51 |
+
"learning_rate": 0.0006998694084416687,
|
| 52 |
+
"loss": 7.3182,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 5.82,
|
| 57 |
+
"learning_rate": 0.0006994777312189898,
|
| 58 |
+
"loss": 7.4319,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 6.66,
|
| 63 |
+
"learning_rate": 0.0006988252606161853,
|
| 64 |
+
"loss": 7.4297,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 7.49,
|
| 69 |
+
"learning_rate": 0.0006979124835312712,
|
| 70 |
+
"loss": 7.426,
|
| 71 |
+
"step": 90
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 8.33,
|
| 75 |
+
"learning_rate": 0.0006967400811127157,
|
| 76 |
+
"loss": 7.425,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 8.33,
|
| 81 |
+
"eval_accuracy": 0.05140231728427503,
|
| 82 |
+
"eval_loss": 7.355892181396484,
|
| 83 |
+
"eval_runtime": 145.5988,
|
| 84 |
+
"eval_samples_per_second": 165.214,
|
| 85 |
+
"eval_steps_per_second": 5.165,
|
| 86 |
+
"step": 100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 9.16,
|
| 90 |
+
"learning_rate": 0.0006953089282511407,
|
| 91 |
+
"loss": 7.427,
|
| 92 |
+
"step": 110
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 9.99,
|
| 96 |
+
"learning_rate": 0.0006936200929264454,
|
| 97 |
+
"loss": 7.2955,
|
| 98 |
+
"step": 120
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 10.82,
|
| 102 |
+
"learning_rate": 0.0006916748354108406,
|
| 103 |
+
"loss": 7.4214,
|
| 104 |
+
"step": 130
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 11.66,
|
| 108 |
+
"learning_rate": 0.0006894746073283849,
|
| 109 |
+
"loss": 7.4198,
|
| 110 |
+
"step": 140
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 12.49,
|
| 114 |
+
"learning_rate": 0.0006870210505717298,
|
| 115 |
+
"loss": 7.4187,
|
| 116 |
+
"step": 150
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 12.49,
|
| 120 |
+
"eval_accuracy": 0.051204356500776864,
|
| 121 |
+
"eval_loss": 7.351686477661133,
|
| 122 |
+
"eval_runtime": 309.0562,
|
| 123 |
+
"eval_samples_per_second": 77.834,
|
| 124 |
+
"eval_steps_per_second": 2.433,
|
| 125 |
+
"step": 150
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 13.33,
|
| 129 |
+
"learning_rate": 0.0006843159960768768,
|
| 130 |
+
"loss": 7.4248,
|
| 131 |
+
"step": 160
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 14.16,
|
| 135 |
+
"learning_rate": 0.0006813614624568652,
|
| 136 |
+
"loss": 7.418,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 14.99,
|
| 141 |
+
"learning_rate": 0.0006781596544954071,
|
| 142 |
+
"loss": 7.295,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 15.82,
|
| 147 |
+
"learning_rate": 0.0006747129615015945,
|
| 148 |
+
"loss": 7.413,
|
| 149 |
+
"step": 190
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 16.66,
|
| 153 |
+
"learning_rate": 0.0006710239555269085,
|
| 154 |
+
"loss": 7.4204,
|
| 155 |
+
"step": 200
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 16.66,
|
| 159 |
+
"eval_accuracy": 0.05138556381472711,
|
| 160 |
+
"eval_loss": 7.343978404998779,
|
| 161 |
+
"eval_runtime": 145.4825,
|
| 162 |
+
"eval_samples_per_second": 165.346,
|
| 163 |
+
"eval_steps_per_second": 5.169,
|
| 164 |
+
"step": 200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 17.49,
|
| 168 |
+
"learning_rate": 0.0006670953894458573,
|
| 169 |
+
"loss": 7.4064,
|
| 170 |
+
"step": 210
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 18.33,
|
| 174 |
+
"learning_rate": 0.0006629301949016779,
|
| 175 |
+
"loss": 7.4197,
|
| 176 |
+
"step": 220
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 19.16,
|
| 180 |
+
"learning_rate": 0.000658531480118635,
|
| 181 |
+
"loss": 7.4168,
|
| 182 |
+
"step": 230
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 19.99,
|
| 186 |
+
"learning_rate": 0.0006539025275825468,
|
| 187 |
+
"loss": 7.2923,
|
| 188 |
+
"step": 240
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 20.82,
|
| 192 |
+
"learning_rate": 0.000649046791591271,
|
| 193 |
+
"loss": 7.4099,
|
| 194 |
+
"step": 250
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 20.82,
|
| 198 |
+
"eval_accuracy": 0.051534592005405604,
|
| 199 |
+
"eval_loss": 7.345396041870117,
|
| 200 |
+
"eval_runtime": 145.5459,
|
| 201 |
+
"eval_samples_per_second": 165.274,
|
| 202 |
+
"eval_steps_per_second": 5.167,
|
| 203 |
+
"step": 250
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 21.66,
|
| 207 |
+
"learning_rate": 0.0006439678956769787,
|
| 208 |
+
"loss": 7.4172,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 22.49,
|
| 213 |
+
"learning_rate": 0.0006386696299021389,
|
| 214 |
+
"loss": 7.4133,
|
| 215 |
+
"step": 270
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 23.33,
|
| 219 |
+
"learning_rate": 0.0006331559480312316,
|
| 220 |
+
"loss": 7.4126,
|
| 221 |
+
"step": 280
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 24.16,
|
| 225 |
+
"learning_rate": 0.0006274309645803004,
|
| 226 |
+
"loss": 7.4111,
|
| 227 |
+
"step": 290
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 24.99,
|
| 231 |
+
"learning_rate": 0.000621498951746547,
|
| 232 |
+
"loss": 7.2916,
|
| 233 |
+
"step": 300
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 24.99,
|
| 237 |
+
"eval_accuracy": 0.05147383671825258,
|
| 238 |
+
"eval_loss": 7.34423303604126,
|
| 239 |
+
"eval_runtime": 145.4168,
|
| 240 |
+
"eval_samples_per_second": 165.421,
|
| 241 |
+
"eval_steps_per_second": 5.171,
|
| 242 |
+
"step": 300
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 25.82,
|
| 246 |
+
"learning_rate": 0.0006153643362202569,
|
| 247 |
+
"loss": 7.408,
|
| 248 |
+
"step": 310
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 26.66,
|
| 252 |
+
"learning_rate": 0.0006090316958814382,
|
| 253 |
+
"loss": 7.4066,
|
| 254 |
+
"step": 320
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 27.49,
|
| 258 |
+
"learning_rate": 0.0006025057563836345,
|
| 259 |
+
"loss": 7.4151,
|
| 260 |
+
"step": 330
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 28.33,
|
| 264 |
+
"learning_rate": 0.0005957913876274665,
|
| 265 |
+
"loss": 7.4088,
|
| 266 |
+
"step": 340
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 29.16,
|
| 270 |
+
"learning_rate": 0.0005888936001265289,
|
| 271 |
+
"loss": 7.4117,
|
| 272 |
+
"step": 350
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 29.16,
|
| 276 |
+
"eval_accuracy": 0.05125750394432031,
|
| 277 |
+
"eval_loss": 7.344011306762695,
|
| 278 |
+
"eval_runtime": 145.4424,
|
| 279 |
+
"eval_samples_per_second": 165.392,
|
| 280 |
+
"eval_steps_per_second": 5.17,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 29.16,
|
| 285 |
+
"step": 350,
|
| 286 |
+
"total_flos": 9.618048555527578e+16,
|
| 287 |
+
"train_loss": 7.568343920026507,
|
| 288 |
+
"train_runtime": 10701.7058,
|
| 289 |
+
"train_samples_per_second": 232.767,
|
| 290 |
+
"train_steps_per_second": 0.112
|
| 291 |
+
}
|
| 292 |
+
],
|
| 293 |
+
"max_steps": 1200,
|
| 294 |
+
"num_train_epochs": 100,
|
| 295 |
+
"total_flos": 9.618048555527578e+16,
|
| 296 |
+
"trial_name": null,
|
| 297 |
+
"trial_params": null
|
| 298 |
+
}
|