Upload folder using huggingface_hub
Browse files- adapter_config.json +4 -4
- adapter_model.safetensors +2 -2
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +34 -448
- training_args.bin +1 -1
adapter_config.json
CHANGED
|
@@ -12,19 +12,19 @@
|
|
| 12 |
"layers_pattern": null,
|
| 13 |
"layers_to_transform": null,
|
| 14 |
"loftq_config": {},
|
| 15 |
-
"lora_alpha":
|
| 16 |
"lora_bias": false,
|
| 17 |
-
"lora_dropout": 0.
|
| 18 |
"megatron_config": null,
|
| 19 |
"megatron_core": "megatron.core",
|
| 20 |
"modules_to_save": null,
|
| 21 |
"peft_type": "LORA",
|
| 22 |
-
"r":
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
-
"o_proj",
|
| 27 |
"down_proj",
|
|
|
|
| 28 |
"gate_up_proj",
|
| 29 |
"qkv_proj"
|
| 30 |
],
|
|
|
|
| 12 |
"layers_pattern": null,
|
| 13 |
"layers_to_transform": null,
|
| 14 |
"loftq_config": {},
|
| 15 |
+
"lora_alpha": 256,
|
| 16 |
"lora_bias": false,
|
| 17 |
+
"lora_dropout": 0.0001,
|
| 18 |
"megatron_config": null,
|
| 19 |
"megatron_core": "megatron.core",
|
| 20 |
"modules_to_save": null,
|
| 21 |
"peft_type": "LORA",
|
| 22 |
+
"r": 128,
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
|
|
|
| 26 |
"down_proj",
|
| 27 |
+
"o_proj",
|
| 28 |
"gate_up_proj",
|
| 29 |
"qkv_proj"
|
| 30 |
],
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d64777516b5921ef8d744b5ebd508a3756f6ad1854cd4f846edc778151c52cea
|
| 3 |
+
size 805341552
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b026fc64ffcd926472bc17624cbfe986d3a3db1c0cb76db198da37defead598a
|
| 3 |
+
size 1610828538
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3e64dac40ba173bacca43da1c2f8d06bd604cd010c45a36fa0bb45c3bacdc23
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fe414f5049a6513e25a7549fe53863c7bf47a5a8dee1e296c2165e39af671b8
|
| 3 |
+
size 1064
|
trainer_state.json
CHANGED
|
@@ -1,492 +1,78 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": 27,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": "//outputs/task7_microsoft/Phi-3.5-mini-instruct/checkpoint-27",
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.5882352941176471,
|
| 14 |
-
"grad_norm": 0.
|
| 15 |
-
"learning_rate": 0.
|
| 16 |
-
"loss": 0.
|
| 17 |
"step": 5
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 1.0,
|
| 21 |
-
"eval_loss": 0.
|
| 22 |
-
"eval_runtime": 3.
|
| 23 |
-
"eval_samples_per_second": 4.
|
| 24 |
-
"eval_steps_per_second": 0.
|
| 25 |
"step": 9
|
| 26 |
},
|
| 27 |
{
|
| 28 |
"epoch": 1.1176470588235294,
|
| 29 |
-
"grad_norm": 0.
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss": 0.
|
| 32 |
"step": 10
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 1.7058823529411766,
|
| 36 |
-
"grad_norm": 0.
|
| 37 |
-
"learning_rate":
|
| 38 |
-
"loss": 0.
|
| 39 |
"step": 15
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 2.0,
|
| 43 |
-
"eval_loss": 0.
|
| 44 |
-
"eval_runtime": 3.
|
| 45 |
-
"eval_samples_per_second": 4.
|
| 46 |
-
"eval_steps_per_second": 0.
|
| 47 |
"step": 18
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 2.235294117647059,
|
| 51 |
-
"grad_norm": 0.
|
| 52 |
-
"learning_rate":
|
| 53 |
-
"loss": 0.
|
| 54 |
"step": 20
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 2.8235294117647056,
|
| 58 |
-
"grad_norm": 0.
|
| 59 |
-
"learning_rate":
|
| 60 |
-
"loss": 0.
|
| 61 |
"step": 25
|
| 62 |
},
|
| 63 |
{
|
| 64 |
"epoch": 3.0,
|
| 65 |
-
"eval_loss": 0.
|
| 66 |
-
"eval_runtime": 3.
|
| 67 |
-
"eval_samples_per_second": 4.
|
| 68 |
-
"eval_steps_per_second": 0.
|
| 69 |
"step": 27
|
| 70 |
-
},
|
| 71 |
-
{
|
| 72 |
-
"epoch": 3.3529411764705883,
|
| 73 |
-
"grad_norm": 0.23658457398414612,
|
| 74 |
-
"learning_rate": 0.00029187258625509513,
|
| 75 |
-
"loss": 0.4257,
|
| 76 |
-
"step": 30
|
| 77 |
-
},
|
| 78 |
-
{
|
| 79 |
-
"epoch": 3.9411764705882355,
|
| 80 |
-
"grad_norm": 0.27489331364631653,
|
| 81 |
-
"learning_rate": 0.00028736599899825856,
|
| 82 |
-
"loss": 0.3254,
|
| 83 |
-
"step": 35
|
| 84 |
-
},
|
| 85 |
-
{
|
| 86 |
-
"epoch": 4.0,
|
| 87 |
-
"eval_loss": 0.5997567772865295,
|
| 88 |
-
"eval_runtime": 3.3651,
|
| 89 |
-
"eval_samples_per_second": 4.458,
|
| 90 |
-
"eval_steps_per_second": 0.594,
|
| 91 |
-
"step": 36
|
| 92 |
-
},
|
| 93 |
-
{
|
| 94 |
-
"epoch": 4.470588235294118,
|
| 95 |
-
"grad_norm": 0.3510420620441437,
|
| 96 |
-
"learning_rate": 0.00028192106268097334,
|
| 97 |
-
"loss": 0.2985,
|
| 98 |
-
"step": 40
|
| 99 |
-
},
|
| 100 |
-
{
|
| 101 |
-
"epoch": 5.0,
|
| 102 |
-
"grad_norm": 0.4653756320476532,
|
| 103 |
-
"learning_rate": 0.00027557497173937923,
|
| 104 |
-
"loss": 0.2828,
|
| 105 |
-
"step": 45
|
| 106 |
-
},
|
| 107 |
-
{
|
| 108 |
-
"epoch": 5.0,
|
| 109 |
-
"eval_loss": 0.6584295034408569,
|
| 110 |
-
"eval_runtime": 3.3668,
|
| 111 |
-
"eval_samples_per_second": 4.455,
|
| 112 |
-
"eval_steps_per_second": 0.594,
|
| 113 |
-
"step": 45
|
| 114 |
-
},
|
| 115 |
-
{
|
| 116 |
-
"epoch": 5.588235294117647,
|
| 117 |
-
"grad_norm": 0.3469710052013397,
|
| 118 |
-
"learning_rate": 0.00026837107640945905,
|
| 119 |
-
"loss": 0.2038,
|
| 120 |
-
"step": 50
|
| 121 |
-
},
|
| 122 |
-
{
|
| 123 |
-
"epoch": 6.0,
|
| 124 |
-
"eval_loss": 0.751620888710022,
|
| 125 |
-
"eval_runtime": 3.3651,
|
| 126 |
-
"eval_samples_per_second": 4.458,
|
| 127 |
-
"eval_steps_per_second": 0.594,
|
| 128 |
-
"step": 54
|
| 129 |
-
},
|
| 130 |
-
{
|
| 131 |
-
"epoch": 6.117647058823529,
|
| 132 |
-
"grad_norm": 0.3903103470802307,
|
| 133 |
-
"learning_rate": 0.0002603585866009697,
|
| 134 |
-
"loss": 0.1656,
|
| 135 |
-
"step": 55
|
| 136 |
-
},
|
| 137 |
-
{
|
| 138 |
-
"epoch": 6.705882352941177,
|
| 139 |
-
"grad_norm": 0.38518330454826355,
|
| 140 |
-
"learning_rate": 0.00025159223574386114,
|
| 141 |
-
"loss": 0.1296,
|
| 142 |
-
"step": 60
|
| 143 |
-
},
|
| 144 |
-
{
|
| 145 |
-
"epoch": 7.0,
|
| 146 |
-
"eval_loss": 0.8753386735916138,
|
| 147 |
-
"eval_runtime": 3.3654,
|
| 148 |
-
"eval_samples_per_second": 4.457,
|
| 149 |
-
"eval_steps_per_second": 0.594,
|
| 150 |
-
"step": 63
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 7.235294117647059,
|
| 154 |
-
"grad_norm": 0.24312527477741241,
|
| 155 |
-
"learning_rate": 0.00024213190690345018,
|
| 156 |
-
"loss": 0.085,
|
| 157 |
-
"step": 65
|
| 158 |
-
},
|
| 159 |
-
{
|
| 160 |
-
"epoch": 7.823529411764706,
|
| 161 |
-
"grad_norm": 0.3267911970615387,
|
| 162 |
-
"learning_rate": 0.00023204222371836405,
|
| 163 |
-
"loss": 0.0751,
|
| 164 |
-
"step": 70
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 8.0,
|
| 168 |
-
"eval_loss": 0.8750011920928955,
|
| 169 |
-
"eval_runtime": 3.3678,
|
| 170 |
-
"eval_samples_per_second": 4.454,
|
| 171 |
-
"eval_steps_per_second": 0.594,
|
| 172 |
-
"step": 72
|
| 173 |
-
},
|
| 174 |
-
{
|
| 175 |
-
"epoch": 8.352941176470589,
|
| 176 |
-
"grad_norm": 0.24549448490142822,
|
| 177 |
-
"learning_rate": 0.00022139210895556104,
|
| 178 |
-
"loss": 0.0576,
|
| 179 |
-
"step": 75
|
| 180 |
-
},
|
| 181 |
-
{
|
| 182 |
-
"epoch": 8.941176470588236,
|
| 183 |
-
"grad_norm": 0.3071894347667694,
|
| 184 |
-
"learning_rate": 0.0002102543136979454,
|
| 185 |
-
"loss": 0.0475,
|
| 186 |
-
"step": 80
|
| 187 |
-
},
|
| 188 |
-
{
|
| 189 |
-
"epoch": 9.0,
|
| 190 |
-
"eval_loss": 0.9990332126617432,
|
| 191 |
-
"eval_runtime": 3.3671,
|
| 192 |
-
"eval_samples_per_second": 4.455,
|
| 193 |
-
"eval_steps_per_second": 0.594,
|
| 194 |
-
"step": 81
|
| 195 |
-
},
|
| 196 |
-
{
|
| 197 |
-
"epoch": 9.470588235294118,
|
| 198 |
-
"grad_norm": 0.17012739181518555,
|
| 199 |
-
"learning_rate": 0.00019870492038070252,
|
| 200 |
-
"loss": 0.0282,
|
| 201 |
-
"step": 85
|
| 202 |
-
},
|
| 203 |
-
{
|
| 204 |
-
"epoch": 10.0,
|
| 205 |
-
"grad_norm": 0.2809393107891083,
|
| 206 |
-
"learning_rate": 0.00018682282307111987,
|
| 207 |
-
"loss": 0.0249,
|
| 208 |
-
"step": 90
|
| 209 |
-
},
|
| 210 |
-
{
|
| 211 |
-
"epoch": 10.0,
|
| 212 |
-
"eval_loss": 1.022998571395874,
|
| 213 |
-
"eval_runtime": 3.3665,
|
| 214 |
-
"eval_samples_per_second": 4.456,
|
| 215 |
-
"eval_steps_per_second": 0.594,
|
| 216 |
-
"step": 90
|
| 217 |
-
},
|
| 218 |
-
{
|
| 219 |
-
"epoch": 10.588235294117647,
|
| 220 |
-
"grad_norm": 0.16731834411621094,
|
| 221 |
-
"learning_rate": 0.00017468918854211007,
|
| 222 |
-
"loss": 0.0151,
|
| 223 |
-
"step": 95
|
| 224 |
-
},
|
| 225 |
-
{
|
| 226 |
-
"epoch": 11.0,
|
| 227 |
-
"eval_loss": 1.0694772005081177,
|
| 228 |
-
"eval_runtime": 3.3662,
|
| 229 |
-
"eval_samples_per_second": 4.456,
|
| 230 |
-
"eval_steps_per_second": 0.594,
|
| 231 |
-
"step": 99
|
| 232 |
-
},
|
| 233 |
-
{
|
| 234 |
-
"epoch": 11.117647058823529,
|
| 235 |
-
"grad_norm": 0.149811789393425,
|
| 236 |
-
"learning_rate": 0.00016238690182084986,
|
| 237 |
-
"loss": 0.0166,
|
| 238 |
-
"step": 100
|
| 239 |
-
},
|
| 240 |
-
{
|
| 241 |
-
"epoch": 11.705882352941176,
|
| 242 |
-
"grad_norm": 0.13291925191879272,
|
| 243 |
-
"learning_rate": 0.00015,
|
| 244 |
-
"loss": 0.0095,
|
| 245 |
-
"step": 105
|
| 246 |
-
},
|
| 247 |
-
{
|
| 248 |
-
"epoch": 12.0,
|
| 249 |
-
"eval_loss": 1.0867458581924438,
|
| 250 |
-
"eval_runtime": 3.3664,
|
| 251 |
-
"eval_samples_per_second": 4.456,
|
| 252 |
-
"eval_steps_per_second": 0.594,
|
| 253 |
-
"step": 108
|
| 254 |
-
},
|
| 255 |
-
{
|
| 256 |
-
"epoch": 12.235294117647058,
|
| 257 |
-
"grad_norm": 0.08576802164316177,
|
| 258 |
-
"learning_rate": 0.00013761309817915014,
|
| 259 |
-
"loss": 0.009,
|
| 260 |
-
"step": 110
|
| 261 |
-
},
|
| 262 |
-
{
|
| 263 |
-
"epoch": 12.823529411764707,
|
| 264 |
-
"grad_norm": 0.07071960717439651,
|
| 265 |
-
"learning_rate": 0.00012531081145788987,
|
| 266 |
-
"loss": 0.0075,
|
| 267 |
-
"step": 115
|
| 268 |
-
},
|
| 269 |
-
{
|
| 270 |
-
"epoch": 13.0,
|
| 271 |
-
"eval_loss": 1.1065537929534912,
|
| 272 |
-
"eval_runtime": 3.3672,
|
| 273 |
-
"eval_samples_per_second": 4.455,
|
| 274 |
-
"eval_steps_per_second": 0.594,
|
| 275 |
-
"step": 117
|
| 276 |
-
},
|
| 277 |
-
{
|
| 278 |
-
"epoch": 13.352941176470589,
|
| 279 |
-
"grad_norm": 0.1268472969532013,
|
| 280 |
-
"learning_rate": 0.00011317717692888012,
|
| 281 |
-
"loss": 0.0125,
|
| 282 |
-
"step": 120
|
| 283 |
-
},
|
| 284 |
-
{
|
| 285 |
-
"epoch": 13.941176470588236,
|
| 286 |
-
"grad_norm": 0.08666448295116425,
|
| 287 |
-
"learning_rate": 0.00010129507961929748,
|
| 288 |
-
"loss": 0.0058,
|
| 289 |
-
"step": 125
|
| 290 |
-
},
|
| 291 |
-
{
|
| 292 |
-
"epoch": 14.0,
|
| 293 |
-
"eval_loss": 1.0826694965362549,
|
| 294 |
-
"eval_runtime": 3.3663,
|
| 295 |
-
"eval_samples_per_second": 4.456,
|
| 296 |
-
"eval_steps_per_second": 0.594,
|
| 297 |
-
"step": 126
|
| 298 |
-
},
|
| 299 |
-
{
|
| 300 |
-
"epoch": 14.470588235294118,
|
| 301 |
-
"grad_norm": 0.20298424363136292,
|
| 302 |
-
"learning_rate": 8.97456863020546e-05,
|
| 303 |
-
"loss": 0.0063,
|
| 304 |
-
"step": 130
|
| 305 |
-
},
|
| 306 |
-
{
|
| 307 |
-
"epoch": 15.0,
|
| 308 |
-
"grad_norm": 0.07050978392362595,
|
| 309 |
-
"learning_rate": 7.860789104443896e-05,
|
| 310 |
-
"loss": 0.0059,
|
| 311 |
-
"step": 135
|
| 312 |
-
},
|
| 313 |
-
{
|
| 314 |
-
"epoch": 15.0,
|
| 315 |
-
"eval_loss": 1.107211947441101,
|
| 316 |
-
"eval_runtime": 3.3663,
|
| 317 |
-
"eval_samples_per_second": 4.456,
|
| 318 |
-
"eval_steps_per_second": 0.594,
|
| 319 |
-
"step": 135
|
| 320 |
-
},
|
| 321 |
-
{
|
| 322 |
-
"epoch": 15.588235294117647,
|
| 323 |
-
"grad_norm": 0.042294692248106,
|
| 324 |
-
"learning_rate": 6.795777628163599e-05,
|
| 325 |
-
"loss": 0.0045,
|
| 326 |
-
"step": 140
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 16.0,
|
| 330 |
-
"eval_loss": 1.1277925968170166,
|
| 331 |
-
"eval_runtime": 3.3661,
|
| 332 |
-
"eval_samples_per_second": 4.456,
|
| 333 |
-
"eval_steps_per_second": 0.594,
|
| 334 |
-
"step": 144
|
| 335 |
-
},
|
| 336 |
-
{
|
| 337 |
-
"epoch": 16.11764705882353,
|
| 338 |
-
"grad_norm": 0.03719251602888107,
|
| 339 |
-
"learning_rate": 5.786809309654982e-05,
|
| 340 |
-
"loss": 0.007,
|
| 341 |
-
"step": 145
|
| 342 |
-
},
|
| 343 |
-
{
|
| 344 |
-
"epoch": 16.705882352941178,
|
| 345 |
-
"grad_norm": 0.03469066694378853,
|
| 346 |
-
"learning_rate": 4.840776425613886e-05,
|
| 347 |
-
"loss": 0.0051,
|
| 348 |
-
"step": 150
|
| 349 |
-
},
|
| 350 |
-
{
|
| 351 |
-
"epoch": 17.0,
|
| 352 |
-
"eval_loss": 1.1320844888687134,
|
| 353 |
-
"eval_runtime": 3.3665,
|
| 354 |
-
"eval_samples_per_second": 4.456,
|
| 355 |
-
"eval_steps_per_second": 0.594,
|
| 356 |
-
"step": 153
|
| 357 |
-
},
|
| 358 |
-
{
|
| 359 |
-
"epoch": 17.235294117647058,
|
| 360 |
-
"grad_norm": 0.04715495929121971,
|
| 361 |
-
"learning_rate": 3.964141339903026e-05,
|
| 362 |
-
"loss": 0.0039,
|
| 363 |
-
"step": 155
|
| 364 |
-
},
|
| 365 |
-
{
|
| 366 |
-
"epoch": 17.823529411764707,
|
| 367 |
-
"grad_norm": 0.10318444669246674,
|
| 368 |
-
"learning_rate": 3.162892359054098e-05,
|
| 369 |
-
"loss": 0.0031,
|
| 370 |
-
"step": 160
|
| 371 |
-
},
|
| 372 |
-
{
|
| 373 |
-
"epoch": 18.0,
|
| 374 |
-
"eval_loss": 1.1403193473815918,
|
| 375 |
-
"eval_runtime": 3.3665,
|
| 376 |
-
"eval_samples_per_second": 4.456,
|
| 377 |
-
"eval_steps_per_second": 0.594,
|
| 378 |
-
"step": 162
|
| 379 |
-
},
|
| 380 |
-
{
|
| 381 |
-
"epoch": 18.352941176470587,
|
| 382 |
-
"grad_norm": 0.030727876350283623,
|
| 383 |
-
"learning_rate": 2.4425028260620715e-05,
|
| 384 |
-
"loss": 0.0028,
|
| 385 |
-
"step": 165
|
| 386 |
-
},
|
| 387 |
-
{
|
| 388 |
-
"epoch": 18.941176470588236,
|
| 389 |
-
"grad_norm": 0.019685346633195877,
|
| 390 |
-
"learning_rate": 1.8078937319026654e-05,
|
| 391 |
-
"loss": 0.0028,
|
| 392 |
-
"step": 170
|
| 393 |
-
},
|
| 394 |
-
{
|
| 395 |
-
"epoch": 19.0,
|
| 396 |
-
"eval_loss": 1.1482117176055908,
|
| 397 |
-
"eval_runtime": 3.3668,
|
| 398 |
-
"eval_samples_per_second": 4.455,
|
| 399 |
-
"eval_steps_per_second": 0.594,
|
| 400 |
-
"step": 171
|
| 401 |
-
},
|
| 402 |
-
{
|
| 403 |
-
"epoch": 19.470588235294116,
|
| 404 |
-
"grad_norm": 0.014024076983332634,
|
| 405 |
-
"learning_rate": 1.2634001001741373e-05,
|
| 406 |
-
"loss": 0.0028,
|
| 407 |
-
"step": 175
|
| 408 |
-
},
|
| 409 |
-
{
|
| 410 |
-
"epoch": 20.0,
|
| 411 |
-
"grad_norm": 0.029733629897236824,
|
| 412 |
-
"learning_rate": 8.127413744904804e-06,
|
| 413 |
-
"loss": 0.0026,
|
| 414 |
-
"step": 180
|
| 415 |
-
},
|
| 416 |
-
{
|
| 417 |
-
"epoch": 20.0,
|
| 418 |
-
"eval_loss": 1.1509085893630981,
|
| 419 |
-
"eval_runtime": 3.3653,
|
| 420 |
-
"eval_samples_per_second": 4.457,
|
| 421 |
-
"eval_steps_per_second": 0.594,
|
| 422 |
-
"step": 180
|
| 423 |
-
},
|
| 424 |
-
{
|
| 425 |
-
"epoch": 20.58823529411765,
|
| 426 |
-
"grad_norm": 0.014151590876281261,
|
| 427 |
-
"learning_rate": 4.589960109100444e-06,
|
| 428 |
-
"loss": 0.0031,
|
| 429 |
-
"step": 185
|
| 430 |
-
},
|
| 431 |
-
{
|
| 432 |
-
"epoch": 21.0,
|
| 433 |
-
"eval_loss": 1.1523716449737549,
|
| 434 |
-
"eval_runtime": 3.3671,
|
| 435 |
-
"eval_samples_per_second": 4.455,
|
| 436 |
-
"eval_steps_per_second": 0.594,
|
| 437 |
-
"step": 189
|
| 438 |
-
},
|
| 439 |
-
{
|
| 440 |
-
"epoch": 21.11764705882353,
|
| 441 |
-
"grad_norm": 0.21795187890529633,
|
| 442 |
-
"learning_rate": 2.0458044895916513e-06,
|
| 443 |
-
"loss": 0.004,
|
| 444 |
-
"step": 190
|
| 445 |
-
},
|
| 446 |
-
{
|
| 447 |
-
"epoch": 21.705882352941178,
|
| 448 |
-
"grad_norm": 0.02381719835102558,
|
| 449 |
-
"learning_rate": 5.123260489995229e-07,
|
| 450 |
-
"loss": 0.0049,
|
| 451 |
-
"step": 195
|
| 452 |
-
},
|
| 453 |
-
{
|
| 454 |
-
"epoch": 22.0,
|
| 455 |
-
"eval_loss": 1.1526206731796265,
|
| 456 |
-
"eval_runtime": 3.3658,
|
| 457 |
-
"eval_samples_per_second": 4.457,
|
| 458 |
-
"eval_steps_per_second": 0.594,
|
| 459 |
-
"step": 198
|
| 460 |
-
},
|
| 461 |
-
{
|
| 462 |
-
"epoch": 22.235294117647058,
|
| 463 |
-
"grad_norm": 0.10794735699892044,
|
| 464 |
-
"learning_rate": 0.0,
|
| 465 |
-
"loss": 0.0028,
|
| 466 |
-
"step": 200
|
| 467 |
-
},
|
| 468 |
-
{
|
| 469 |
-
"epoch": 22.235294117647058,
|
| 470 |
-
"eval_loss": 1.1525808572769165,
|
| 471 |
-
"eval_runtime": 3.3658,
|
| 472 |
-
"eval_samples_per_second": 4.457,
|
| 473 |
-
"eval_steps_per_second": 0.594,
|
| 474 |
-
"step": 200
|
| 475 |
-
},
|
| 476 |
-
{
|
| 477 |
-
"epoch": 22.235294117647058,
|
| 478 |
-
"step": 200,
|
| 479 |
-
"total_flos": 1.040122623832658e+17,
|
| 480 |
-
"train_loss": 0.13381972634699196,
|
| 481 |
-
"train_runtime": 1987.8506,
|
| 482 |
-
"train_samples_per_second": 1.61,
|
| 483 |
-
"train_steps_per_second": 0.101
|
| 484 |
}
|
| 485 |
],
|
| 486 |
"logging_steps": 5,
|
| 487 |
-
"max_steps":
|
| 488 |
"num_input_tokens_seen": 0,
|
| 489 |
-
"num_train_epochs":
|
| 490 |
"save_steps": 500,
|
| 491 |
"stateful_callbacks": {
|
| 492 |
"TrainerControl": {
|
|
@@ -495,12 +81,12 @@
|
|
| 495 |
"should_evaluate": false,
|
| 496 |
"should_log": false,
|
| 497 |
"should_save": true,
|
| 498 |
-
"should_training_stop":
|
| 499 |
},
|
| 500 |
"attributes": {}
|
| 501 |
}
|
| 502 |
},
|
| 503 |
-
"total_flos": 1.
|
| 504 |
"train_batch_size": 2,
|
| 505 |
"trial_name": null,
|
| 506 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": 27,
|
| 3 |
+
"best_metric": 0.5784786939620972,
|
| 4 |
"best_model_checkpoint": "//outputs/task7_microsoft/Phi-3.5-mini-instruct/checkpoint-27",
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 27,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.5882352941176471,
|
| 14 |
+
"grad_norm": 0.38448742032051086,
|
| 15 |
+
"learning_rate": 0.0001,
|
| 16 |
+
"loss": 0.7574,
|
| 17 |
"step": 5
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 1.0,
|
| 21 |
+
"eval_loss": 0.6299339532852173,
|
| 22 |
+
"eval_runtime": 3.415,
|
| 23 |
+
"eval_samples_per_second": 4.392,
|
| 24 |
+
"eval_steps_per_second": 0.586,
|
| 25 |
"step": 9
|
| 26 |
},
|
| 27 |
{
|
| 28 |
"epoch": 1.1176470588235294,
|
| 29 |
+
"grad_norm": 0.339433491230011,
|
| 30 |
+
"learning_rate": 9.931806517013612e-05,
|
| 31 |
+
"loss": 0.6323,
|
| 32 |
"step": 10
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 1.7058823529411766,
|
| 36 |
+
"grad_norm": 0.32222846150398254,
|
| 37 |
+
"learning_rate": 9.729086208503174e-05,
|
| 38 |
+
"loss": 0.5438,
|
| 39 |
"step": 15
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 2.0,
|
| 43 |
+
"eval_loss": 0.5842064619064331,
|
| 44 |
+
"eval_runtime": 3.4146,
|
| 45 |
+
"eval_samples_per_second": 4.393,
|
| 46 |
+
"eval_steps_per_second": 0.586,
|
| 47 |
"step": 18
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 2.235294117647059,
|
| 51 |
+
"grad_norm": 0.26292550563812256,
|
| 52 |
+
"learning_rate": 9.397368756032445e-05,
|
| 53 |
+
"loss": 0.4495,
|
| 54 |
"step": 20
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 2.8235294117647056,
|
| 58 |
+
"grad_norm": 0.28129199147224426,
|
| 59 |
+
"learning_rate": 8.945702546981969e-05,
|
| 60 |
+
"loss": 0.3702,
|
| 61 |
"step": 25
|
| 62 |
},
|
| 63 |
{
|
| 64 |
"epoch": 3.0,
|
| 65 |
+
"eval_loss": 0.5784786939620972,
|
| 66 |
+
"eval_runtime": 3.4121,
|
| 67 |
+
"eval_samples_per_second": 4.396,
|
| 68 |
+
"eval_steps_per_second": 0.586,
|
| 69 |
"step": 27
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
}
|
| 71 |
],
|
| 72 |
"logging_steps": 5,
|
| 73 |
+
"max_steps": 100,
|
| 74 |
"num_input_tokens_seen": 0,
|
| 75 |
+
"num_train_epochs": 13,
|
| 76 |
"save_steps": 500,
|
| 77 |
"stateful_callbacks": {
|
| 78 |
"TrainerControl": {
|
|
|
|
| 81 |
"should_evaluate": false,
|
| 82 |
"should_log": false,
|
| 83 |
"should_save": true,
|
| 84 |
+
"should_training_stop": false
|
| 85 |
},
|
| 86 |
"attributes": {}
|
| 87 |
}
|
| 88 |
},
|
| 89 |
+
"total_flos": 1.4777106488322048e+16,
|
| 90 |
"train_batch_size": 2,
|
| 91 |
"trial_name": null,
|
| 92 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5624
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d8619b2675a3aadf91623d77096e075400d78b3a6660267fb11439bce8f9586
|
| 3 |
size 5624
|