Training in progress, step 100, checkpoint
Browse files
last-checkpoint/README.md
CHANGED
|
@@ -216,4 +216,42 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 216 |
### Framework versions
|
| 217 |
|
| 218 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 219 |
- PEFT 0.7.0.dev0
|
|
|
|
| 216 |
### Framework versions
|
| 217 |
|
| 218 |
|
| 219 |
+
- PEFT 0.7.0.dev0
|
| 220 |
+
## Training procedure
|
| 221 |
+
|
| 222 |
+
|
| 223 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 224 |
+
- quant_method: bitsandbytes
|
| 225 |
+
- load_in_8bit: True
|
| 226 |
+
- load_in_4bit: False
|
| 227 |
+
- llm_int8_threshold: 6.0
|
| 228 |
+
- llm_int8_skip_modules: None
|
| 229 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 230 |
+
- llm_int8_has_fp16_weight: False
|
| 231 |
+
- bnb_4bit_quant_type: fp4
|
| 232 |
+
- bnb_4bit_use_double_quant: False
|
| 233 |
+
- bnb_4bit_compute_dtype: float32
|
| 234 |
+
|
| 235 |
+
### Framework versions
|
| 236 |
+
|
| 237 |
+
|
| 238 |
+
- PEFT 0.7.0.dev0
|
| 239 |
+
## Training procedure
|
| 240 |
+
|
| 241 |
+
|
| 242 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 243 |
+
- quant_method: bitsandbytes
|
| 244 |
+
- load_in_8bit: True
|
| 245 |
+
- load_in_4bit: False
|
| 246 |
+
- llm_int8_threshold: 6.0
|
| 247 |
+
- llm_int8_skip_modules: None
|
| 248 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 249 |
+
- llm_int8_has_fp16_weight: False
|
| 250 |
+
- bnb_4bit_quant_type: fp4
|
| 251 |
+
- bnb_4bit_use_double_quant: False
|
| 252 |
+
- bnb_4bit_compute_dtype: float32
|
| 253 |
+
|
| 254 |
+
### Framework versions
|
| 255 |
+
|
| 256 |
+
|
| 257 |
- PEFT 0.7.0.dev0
|
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 9444296
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:436ac47f550180c481cd11c0ce4b4e5b4b28617cba70726682d529c5278ee00f
|
| 3 |
size 9444296
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 18902665
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64cc9defec0add6325a218977493d532d8056c94a70a3e0dc85e33d4d306115d
|
| 3 |
size 18902665
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7393dbde4bffa4ea759a39a2e6dd5d0164b7e91c9e8ab3bfffc0ca38d5daac71
|
| 3 |
size 14575
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:216f76b8039f833c337db298c81f13b12082d5fd4f9d866cecd34b2ca7550b37
|
| 3 |
size 627
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "./outputs/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -11,583 +11,23 @@
|
|
| 11 |
{
|
| 12 |
"epoch": 0.07,
|
| 13 |
"learning_rate": 0.0002,
|
| 14 |
-
"loss": 2.
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.07,
|
| 19 |
-
"eval_loss": 2.
|
| 20 |
-
"eval_runtime":
|
| 21 |
-
"eval_samples_per_second": 30.
|
| 22 |
-
"eval_steps_per_second": 3.
|
| 23 |
"step": 100
|
| 24 |
-
},
|
| 25 |
-
{
|
| 26 |
-
"epoch": 0.15,
|
| 27 |
-
"learning_rate": 0.0002,
|
| 28 |
-
"loss": 2.6046,
|
| 29 |
-
"step": 200
|
| 30 |
-
},
|
| 31 |
-
{
|
| 32 |
-
"epoch": 0.15,
|
| 33 |
-
"eval_loss": 2.592717409133911,
|
| 34 |
-
"eval_runtime": 207.0706,
|
| 35 |
-
"eval_samples_per_second": 30.299,
|
| 36 |
-
"eval_steps_per_second": 3.791,
|
| 37 |
-
"step": 200
|
| 38 |
-
},
|
| 39 |
-
{
|
| 40 |
-
"epoch": 0.22,
|
| 41 |
-
"learning_rate": 0.0002,
|
| 42 |
-
"loss": 2.5628,
|
| 43 |
-
"step": 300
|
| 44 |
-
},
|
| 45 |
-
{
|
| 46 |
-
"epoch": 0.22,
|
| 47 |
-
"eval_loss": 2.5575265884399414,
|
| 48 |
-
"eval_runtime": 206.3145,
|
| 49 |
-
"eval_samples_per_second": 30.41,
|
| 50 |
-
"eval_steps_per_second": 3.805,
|
| 51 |
-
"step": 300
|
| 52 |
-
},
|
| 53 |
-
{
|
| 54 |
-
"epoch": 0.29,
|
| 55 |
-
"learning_rate": 0.0002,
|
| 56 |
-
"loss": 2.5372,
|
| 57 |
-
"step": 400
|
| 58 |
-
},
|
| 59 |
-
{
|
| 60 |
-
"epoch": 0.29,
|
| 61 |
-
"eval_loss": 2.525071859359741,
|
| 62 |
-
"eval_runtime": 206.5249,
|
| 63 |
-
"eval_samples_per_second": 30.379,
|
| 64 |
-
"eval_steps_per_second": 3.801,
|
| 65 |
-
"step": 400
|
| 66 |
-
},
|
| 67 |
-
{
|
| 68 |
-
"epoch": 0.36,
|
| 69 |
-
"learning_rate": 0.0002,
|
| 70 |
-
"loss": 2.4945,
|
| 71 |
-
"step": 500
|
| 72 |
-
},
|
| 73 |
-
{
|
| 74 |
-
"epoch": 0.36,
|
| 75 |
-
"eval_loss": 2.4999899864196777,
|
| 76 |
-
"eval_runtime": 206.4967,
|
| 77 |
-
"eval_samples_per_second": 30.383,
|
| 78 |
-
"eval_steps_per_second": 3.802,
|
| 79 |
-
"step": 500
|
| 80 |
-
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 0.44,
|
| 83 |
-
"learning_rate": 0.0002,
|
| 84 |
-
"loss": 2.4681,
|
| 85 |
-
"step": 600
|
| 86 |
-
},
|
| 87 |
-
{
|
| 88 |
-
"epoch": 0.44,
|
| 89 |
-
"eval_loss": 2.474062919616699,
|
| 90 |
-
"eval_runtime": 206.9377,
|
| 91 |
-
"eval_samples_per_second": 30.318,
|
| 92 |
-
"eval_steps_per_second": 3.793,
|
| 93 |
-
"step": 600
|
| 94 |
-
},
|
| 95 |
-
{
|
| 96 |
-
"epoch": 0.51,
|
| 97 |
-
"learning_rate": 0.0002,
|
| 98 |
-
"loss": 2.4511,
|
| 99 |
-
"step": 700
|
| 100 |
-
},
|
| 101 |
-
{
|
| 102 |
-
"epoch": 0.51,
|
| 103 |
-
"eval_loss": 2.4490628242492676,
|
| 104 |
-
"eval_runtime": 206.1477,
|
| 105 |
-
"eval_samples_per_second": 30.434,
|
| 106 |
-
"eval_steps_per_second": 3.808,
|
| 107 |
-
"step": 700
|
| 108 |
-
},
|
| 109 |
-
{
|
| 110 |
-
"epoch": 0.58,
|
| 111 |
-
"learning_rate": 0.0002,
|
| 112 |
-
"loss": 2.428,
|
| 113 |
-
"step": 800
|
| 114 |
-
},
|
| 115 |
-
{
|
| 116 |
-
"epoch": 0.58,
|
| 117 |
-
"eval_loss": 2.4292285442352295,
|
| 118 |
-
"eval_runtime": 206.472,
|
| 119 |
-
"eval_samples_per_second": 30.387,
|
| 120 |
-
"eval_steps_per_second": 3.802,
|
| 121 |
-
"step": 800
|
| 122 |
-
},
|
| 123 |
-
{
|
| 124 |
-
"epoch": 0.66,
|
| 125 |
-
"learning_rate": 0.0002,
|
| 126 |
-
"loss": 2.3917,
|
| 127 |
-
"step": 900
|
| 128 |
-
},
|
| 129 |
-
{
|
| 130 |
-
"epoch": 0.66,
|
| 131 |
-
"eval_loss": 2.4080374240875244,
|
| 132 |
-
"eval_runtime": 206.5411,
|
| 133 |
-
"eval_samples_per_second": 30.377,
|
| 134 |
-
"eval_steps_per_second": 3.801,
|
| 135 |
-
"step": 900
|
| 136 |
-
},
|
| 137 |
-
{
|
| 138 |
-
"epoch": 0.73,
|
| 139 |
-
"learning_rate": 0.0002,
|
| 140 |
-
"loss": 2.3881,
|
| 141 |
-
"step": 1000
|
| 142 |
-
},
|
| 143 |
-
{
|
| 144 |
-
"epoch": 0.73,
|
| 145 |
-
"eval_loss": 2.3876078128814697,
|
| 146 |
-
"eval_runtime": 206.4718,
|
| 147 |
-
"eval_samples_per_second": 30.387,
|
| 148 |
-
"eval_steps_per_second": 3.802,
|
| 149 |
-
"step": 1000
|
| 150 |
-
},
|
| 151 |
-
{
|
| 152 |
-
"epoch": 0.8,
|
| 153 |
-
"learning_rate": 0.0002,
|
| 154 |
-
"loss": 2.3815,
|
| 155 |
-
"step": 1100
|
| 156 |
-
},
|
| 157 |
-
{
|
| 158 |
-
"epoch": 0.8,
|
| 159 |
-
"eval_loss": 2.3660354614257812,
|
| 160 |
-
"eval_runtime": 206.9685,
|
| 161 |
-
"eval_samples_per_second": 30.314,
|
| 162 |
-
"eval_steps_per_second": 3.793,
|
| 163 |
-
"step": 1100
|
| 164 |
-
},
|
| 165 |
-
{
|
| 166 |
-
"epoch": 0.87,
|
| 167 |
-
"learning_rate": 0.0002,
|
| 168 |
-
"loss": 2.3383,
|
| 169 |
-
"step": 1200
|
| 170 |
-
},
|
| 171 |
-
{
|
| 172 |
-
"epoch": 0.87,
|
| 173 |
-
"eval_loss": 2.3476908206939697,
|
| 174 |
-
"eval_runtime": 206.7525,
|
| 175 |
-
"eval_samples_per_second": 30.345,
|
| 176 |
-
"eval_steps_per_second": 3.797,
|
| 177 |
-
"step": 1200
|
| 178 |
-
},
|
| 179 |
-
{
|
| 180 |
-
"epoch": 0.95,
|
| 181 |
-
"learning_rate": 0.0002,
|
| 182 |
-
"loss": 2.3498,
|
| 183 |
-
"step": 1300
|
| 184 |
-
},
|
| 185 |
-
{
|
| 186 |
-
"epoch": 0.95,
|
| 187 |
-
"eval_loss": 2.3309385776519775,
|
| 188 |
-
"eval_runtime": 206.5818,
|
| 189 |
-
"eval_samples_per_second": 30.371,
|
| 190 |
-
"eval_steps_per_second": 3.8,
|
| 191 |
-
"step": 1300
|
| 192 |
-
},
|
| 193 |
-
{
|
| 194 |
-
"epoch": 1.02,
|
| 195 |
-
"learning_rate": 0.0002,
|
| 196 |
-
"loss": 2.2994,
|
| 197 |
-
"step": 1400
|
| 198 |
-
},
|
| 199 |
-
{
|
| 200 |
-
"epoch": 1.02,
|
| 201 |
-
"eval_loss": 2.3077964782714844,
|
| 202 |
-
"eval_runtime": 206.4915,
|
| 203 |
-
"eval_samples_per_second": 30.384,
|
| 204 |
-
"eval_steps_per_second": 3.802,
|
| 205 |
-
"step": 1400
|
| 206 |
-
},
|
| 207 |
-
{
|
| 208 |
-
"epoch": 1.09,
|
| 209 |
-
"learning_rate": 0.0002,
|
| 210 |
-
"loss": 2.2611,
|
| 211 |
-
"step": 1500
|
| 212 |
-
},
|
| 213 |
-
{
|
| 214 |
-
"epoch": 1.09,
|
| 215 |
-
"eval_loss": 2.290616989135742,
|
| 216 |
-
"eval_runtime": 206.9291,
|
| 217 |
-
"eval_samples_per_second": 30.32,
|
| 218 |
-
"eval_steps_per_second": 3.794,
|
| 219 |
-
"step": 1500
|
| 220 |
-
},
|
| 221 |
-
{
|
| 222 |
-
"epoch": 1.17,
|
| 223 |
-
"learning_rate": 0.0002,
|
| 224 |
-
"loss": 2.2529,
|
| 225 |
-
"step": 1600
|
| 226 |
-
},
|
| 227 |
-
{
|
| 228 |
-
"epoch": 1.17,
|
| 229 |
-
"eval_loss": 2.2742502689361572,
|
| 230 |
-
"eval_runtime": 207.033,
|
| 231 |
-
"eval_samples_per_second": 30.304,
|
| 232 |
-
"eval_steps_per_second": 3.792,
|
| 233 |
-
"step": 1600
|
| 234 |
-
},
|
| 235 |
-
{
|
| 236 |
-
"epoch": 1.24,
|
| 237 |
-
"learning_rate": 0.0002,
|
| 238 |
-
"loss": 2.2533,
|
| 239 |
-
"step": 1700
|
| 240 |
-
},
|
| 241 |
-
{
|
| 242 |
-
"epoch": 1.24,
|
| 243 |
-
"eval_loss": 2.2574470043182373,
|
| 244 |
-
"eval_runtime": 206.7407,
|
| 245 |
-
"eval_samples_per_second": 30.347,
|
| 246 |
-
"eval_steps_per_second": 3.797,
|
| 247 |
-
"step": 1700
|
| 248 |
-
},
|
| 249 |
-
{
|
| 250 |
-
"epoch": 1.31,
|
| 251 |
-
"learning_rate": 0.0002,
|
| 252 |
-
"loss": 2.214,
|
| 253 |
-
"step": 1800
|
| 254 |
-
},
|
| 255 |
-
{
|
| 256 |
-
"epoch": 1.31,
|
| 257 |
-
"eval_loss": 2.241929769515991,
|
| 258 |
-
"eval_runtime": 206.7574,
|
| 259 |
-
"eval_samples_per_second": 30.345,
|
| 260 |
-
"eval_steps_per_second": 3.797,
|
| 261 |
-
"step": 1800
|
| 262 |
-
},
|
| 263 |
-
{
|
| 264 |
-
"epoch": 1.38,
|
| 265 |
-
"learning_rate": 0.0002,
|
| 266 |
-
"loss": 2.199,
|
| 267 |
-
"step": 1900
|
| 268 |
-
},
|
| 269 |
-
{
|
| 270 |
-
"epoch": 1.38,
|
| 271 |
-
"eval_loss": 2.2262773513793945,
|
| 272 |
-
"eval_runtime": 206.8241,
|
| 273 |
-
"eval_samples_per_second": 30.335,
|
| 274 |
-
"eval_steps_per_second": 3.795,
|
| 275 |
-
"step": 1900
|
| 276 |
-
},
|
| 277 |
-
{
|
| 278 |
-
"epoch": 1.46,
|
| 279 |
-
"learning_rate": 0.0002,
|
| 280 |
-
"loss": 2.1938,
|
| 281 |
-
"step": 2000
|
| 282 |
-
},
|
| 283 |
-
{
|
| 284 |
-
"epoch": 1.46,
|
| 285 |
-
"eval_loss": 2.2117421627044678,
|
| 286 |
-
"eval_runtime": 207.1013,
|
| 287 |
-
"eval_samples_per_second": 30.294,
|
| 288 |
-
"eval_steps_per_second": 3.79,
|
| 289 |
-
"step": 2000
|
| 290 |
-
},
|
| 291 |
-
{
|
| 292 |
-
"epoch": 1.53,
|
| 293 |
-
"learning_rate": 0.0002,
|
| 294 |
-
"loss": 2.1761,
|
| 295 |
-
"step": 2100
|
| 296 |
-
},
|
| 297 |
-
{
|
| 298 |
-
"epoch": 1.53,
|
| 299 |
-
"eval_loss": 2.1993796825408936,
|
| 300 |
-
"eval_runtime": 207.1774,
|
| 301 |
-
"eval_samples_per_second": 30.283,
|
| 302 |
-
"eval_steps_per_second": 3.789,
|
| 303 |
-
"step": 2100
|
| 304 |
-
},
|
| 305 |
-
{
|
| 306 |
-
"epoch": 1.6,
|
| 307 |
-
"learning_rate": 0.0002,
|
| 308 |
-
"loss": 2.159,
|
| 309 |
-
"step": 2200
|
| 310 |
-
},
|
| 311 |
-
{
|
| 312 |
-
"epoch": 1.6,
|
| 313 |
-
"eval_loss": 2.1830873489379883,
|
| 314 |
-
"eval_runtime": 207.1307,
|
| 315 |
-
"eval_samples_per_second": 30.29,
|
| 316 |
-
"eval_steps_per_second": 3.79,
|
| 317 |
-
"step": 2200
|
| 318 |
-
},
|
| 319 |
-
{
|
| 320 |
-
"epoch": 1.68,
|
| 321 |
-
"learning_rate": 0.0002,
|
| 322 |
-
"loss": 2.1497,
|
| 323 |
-
"step": 2300
|
| 324 |
-
},
|
| 325 |
-
{
|
| 326 |
-
"epoch": 1.68,
|
| 327 |
-
"eval_loss": 2.1709437370300293,
|
| 328 |
-
"eval_runtime": 207.0264,
|
| 329 |
-
"eval_samples_per_second": 30.305,
|
| 330 |
-
"eval_steps_per_second": 3.792,
|
| 331 |
-
"step": 2300
|
| 332 |
-
},
|
| 333 |
-
{
|
| 334 |
-
"epoch": 1.75,
|
| 335 |
-
"learning_rate": 0.0002,
|
| 336 |
-
"loss": 2.1548,
|
| 337 |
-
"step": 2400
|
| 338 |
-
},
|
| 339 |
-
{
|
| 340 |
-
"epoch": 1.75,
|
| 341 |
-
"eval_loss": 2.1570258140563965,
|
| 342 |
-
"eval_runtime": 207.0802,
|
| 343 |
-
"eval_samples_per_second": 30.297,
|
| 344 |
-
"eval_steps_per_second": 3.791,
|
| 345 |
-
"step": 2400
|
| 346 |
-
},
|
| 347 |
-
{
|
| 348 |
-
"epoch": 1.82,
|
| 349 |
-
"learning_rate": 0.0002,
|
| 350 |
-
"loss": 2.1218,
|
| 351 |
-
"step": 2500
|
| 352 |
-
},
|
| 353 |
-
{
|
| 354 |
-
"epoch": 1.82,
|
| 355 |
-
"eval_loss": 2.1442573070526123,
|
| 356 |
-
"eval_runtime": 207.3061,
|
| 357 |
-
"eval_samples_per_second": 30.264,
|
| 358 |
-
"eval_steps_per_second": 3.787,
|
| 359 |
-
"step": 2500
|
| 360 |
-
},
|
| 361 |
-
{
|
| 362 |
-
"epoch": 1.89,
|
| 363 |
-
"learning_rate": 0.0002,
|
| 364 |
-
"loss": 2.1246,
|
| 365 |
-
"step": 2600
|
| 366 |
-
},
|
| 367 |
-
{
|
| 368 |
-
"epoch": 1.89,
|
| 369 |
-
"eval_loss": 2.1329336166381836,
|
| 370 |
-
"eval_runtime": 206.7598,
|
| 371 |
-
"eval_samples_per_second": 30.344,
|
| 372 |
-
"eval_steps_per_second": 3.797,
|
| 373 |
-
"step": 2600
|
| 374 |
-
},
|
| 375 |
-
{
|
| 376 |
-
"epoch": 1.97,
|
| 377 |
-
"learning_rate": 0.0002,
|
| 378 |
-
"loss": 2.1067,
|
| 379 |
-
"step": 2700
|
| 380 |
-
},
|
| 381 |
-
{
|
| 382 |
-
"epoch": 1.97,
|
| 383 |
-
"eval_loss": 2.117797374725342,
|
| 384 |
-
"eval_runtime": 207.3313,
|
| 385 |
-
"eval_samples_per_second": 30.261,
|
| 386 |
-
"eval_steps_per_second": 3.786,
|
| 387 |
-
"step": 2700
|
| 388 |
-
},
|
| 389 |
-
{
|
| 390 |
-
"epoch": 2.04,
|
| 391 |
-
"learning_rate": 0.0002,
|
| 392 |
-
"loss": 2.0643,
|
| 393 |
-
"step": 2800
|
| 394 |
-
},
|
| 395 |
-
{
|
| 396 |
-
"epoch": 2.04,
|
| 397 |
-
"eval_loss": 2.111072540283203,
|
| 398 |
-
"eval_runtime": 312.3448,
|
| 399 |
-
"eval_samples_per_second": 20.087,
|
| 400 |
-
"eval_steps_per_second": 2.513,
|
| 401 |
-
"step": 2800
|
| 402 |
-
},
|
| 403 |
-
{
|
| 404 |
-
"epoch": 2.11,
|
| 405 |
-
"learning_rate": 0.0002,
|
| 406 |
-
"loss": 2.0379,
|
| 407 |
-
"step": 2900
|
| 408 |
-
},
|
| 409 |
-
{
|
| 410 |
-
"epoch": 2.11,
|
| 411 |
-
"eval_loss": 2.097459554672241,
|
| 412 |
-
"eval_runtime": 207.1063,
|
| 413 |
-
"eval_samples_per_second": 30.294,
|
| 414 |
-
"eval_steps_per_second": 3.79,
|
| 415 |
-
"step": 2900
|
| 416 |
-
},
|
| 417 |
-
{
|
| 418 |
-
"epoch": 2.19,
|
| 419 |
-
"learning_rate": 0.0002,
|
| 420 |
-
"loss": 2.0567,
|
| 421 |
-
"step": 3000
|
| 422 |
-
},
|
| 423 |
-
{
|
| 424 |
-
"epoch": 2.19,
|
| 425 |
-
"eval_loss": 2.0867855548858643,
|
| 426 |
-
"eval_runtime": 207.096,
|
| 427 |
-
"eval_samples_per_second": 30.295,
|
| 428 |
-
"eval_steps_per_second": 3.791,
|
| 429 |
-
"step": 3000
|
| 430 |
-
},
|
| 431 |
-
{
|
| 432 |
-
"epoch": 2.26,
|
| 433 |
-
"learning_rate": 0.0002,
|
| 434 |
-
"loss": 2.0243,
|
| 435 |
-
"step": 3100
|
| 436 |
-
},
|
| 437 |
-
{
|
| 438 |
-
"epoch": 2.26,
|
| 439 |
-
"eval_loss": 2.0787196159362793,
|
| 440 |
-
"eval_runtime": 207.0741,
|
| 441 |
-
"eval_samples_per_second": 30.298,
|
| 442 |
-
"eval_steps_per_second": 3.791,
|
| 443 |
-
"step": 3100
|
| 444 |
-
},
|
| 445 |
-
{
|
| 446 |
-
"epoch": 2.33,
|
| 447 |
-
"learning_rate": 0.0002,
|
| 448 |
-
"loss": 2.0407,
|
| 449 |
-
"step": 3200
|
| 450 |
-
},
|
| 451 |
-
{
|
| 452 |
-
"epoch": 2.33,
|
| 453 |
-
"eval_loss": 2.066490888595581,
|
| 454 |
-
"eval_runtime": 207.3107,
|
| 455 |
-
"eval_samples_per_second": 30.264,
|
| 456 |
-
"eval_steps_per_second": 3.787,
|
| 457 |
-
"step": 3200
|
| 458 |
-
},
|
| 459 |
-
{
|
| 460 |
-
"epoch": 2.4,
|
| 461 |
-
"learning_rate": 0.0002,
|
| 462 |
-
"loss": 2.009,
|
| 463 |
-
"step": 3300
|
| 464 |
-
},
|
| 465 |
-
{
|
| 466 |
-
"epoch": 2.4,
|
| 467 |
-
"eval_loss": 2.0566089153289795,
|
| 468 |
-
"eval_runtime": 207.0074,
|
| 469 |
-
"eval_samples_per_second": 30.308,
|
| 470 |
-
"eval_steps_per_second": 3.792,
|
| 471 |
-
"step": 3300
|
| 472 |
-
},
|
| 473 |
-
{
|
| 474 |
-
"epoch": 2.48,
|
| 475 |
-
"learning_rate": 0.0002,
|
| 476 |
-
"loss": 2.002,
|
| 477 |
-
"step": 3400
|
| 478 |
-
},
|
| 479 |
-
{
|
| 480 |
-
"epoch": 2.48,
|
| 481 |
-
"eval_loss": 2.0491139888763428,
|
| 482 |
-
"eval_runtime": 207.1382,
|
| 483 |
-
"eval_samples_per_second": 30.289,
|
| 484 |
-
"eval_steps_per_second": 3.79,
|
| 485 |
-
"step": 3400
|
| 486 |
-
},
|
| 487 |
-
{
|
| 488 |
-
"epoch": 2.55,
|
| 489 |
-
"learning_rate": 0.0002,
|
| 490 |
-
"loss": 2.0002,
|
| 491 |
-
"step": 3500
|
| 492 |
-
},
|
| 493 |
-
{
|
| 494 |
-
"epoch": 2.55,
|
| 495 |
-
"eval_loss": 2.0374956130981445,
|
| 496 |
-
"eval_runtime": 207.2444,
|
| 497 |
-
"eval_samples_per_second": 30.273,
|
| 498 |
-
"eval_steps_per_second": 3.788,
|
| 499 |
-
"step": 3500
|
| 500 |
-
},
|
| 501 |
-
{
|
| 502 |
-
"epoch": 2.62,
|
| 503 |
-
"learning_rate": 0.0002,
|
| 504 |
-
"loss": 1.9945,
|
| 505 |
-
"step": 3600
|
| 506 |
-
},
|
| 507 |
-
{
|
| 508 |
-
"epoch": 2.62,
|
| 509 |
-
"eval_loss": 2.027383804321289,
|
| 510 |
-
"eval_runtime": 207.1782,
|
| 511 |
-
"eval_samples_per_second": 30.283,
|
| 512 |
-
"eval_steps_per_second": 3.789,
|
| 513 |
-
"step": 3600
|
| 514 |
-
},
|
| 515 |
-
{
|
| 516 |
-
"epoch": 2.7,
|
| 517 |
-
"learning_rate": 0.0002,
|
| 518 |
-
"loss": 1.9759,
|
| 519 |
-
"step": 3700
|
| 520 |
-
},
|
| 521 |
-
{
|
| 522 |
-
"epoch": 2.7,
|
| 523 |
-
"eval_loss": 2.0199601650238037,
|
| 524 |
-
"eval_runtime": 207.4478,
|
| 525 |
-
"eval_samples_per_second": 30.244,
|
| 526 |
-
"eval_steps_per_second": 3.784,
|
| 527 |
-
"step": 3700
|
| 528 |
-
},
|
| 529 |
-
{
|
| 530 |
-
"epoch": 2.77,
|
| 531 |
-
"learning_rate": 0.0002,
|
| 532 |
-
"loss": 1.9675,
|
| 533 |
-
"step": 3800
|
| 534 |
-
},
|
| 535 |
-
{
|
| 536 |
-
"epoch": 2.77,
|
| 537 |
-
"eval_loss": 2.0125296115875244,
|
| 538 |
-
"eval_runtime": 207.2437,
|
| 539 |
-
"eval_samples_per_second": 30.274,
|
| 540 |
-
"eval_steps_per_second": 3.788,
|
| 541 |
-
"step": 3800
|
| 542 |
-
},
|
| 543 |
-
{
|
| 544 |
-
"epoch": 2.84,
|
| 545 |
-
"learning_rate": 0.0002,
|
| 546 |
-
"loss": 1.9542,
|
| 547 |
-
"step": 3900
|
| 548 |
-
},
|
| 549 |
-
{
|
| 550 |
-
"epoch": 2.84,
|
| 551 |
-
"eval_loss": 1.9999363422393799,
|
| 552 |
-
"eval_runtime": 206.9139,
|
| 553 |
-
"eval_samples_per_second": 30.322,
|
| 554 |
-
"eval_steps_per_second": 3.794,
|
| 555 |
-
"step": 3900
|
| 556 |
-
},
|
| 557 |
-
{
|
| 558 |
-
"epoch": 2.91,
|
| 559 |
-
"learning_rate": 0.0002,
|
| 560 |
-
"loss": 1.9555,
|
| 561 |
-
"step": 4000
|
| 562 |
-
},
|
| 563 |
-
{
|
| 564 |
-
"epoch": 2.91,
|
| 565 |
-
"eval_loss": 1.9915796518325806,
|
| 566 |
-
"eval_runtime": 208.2968,
|
| 567 |
-
"eval_samples_per_second": 30.12,
|
| 568 |
-
"eval_steps_per_second": 3.769,
|
| 569 |
-
"step": 4000
|
| 570 |
-
},
|
| 571 |
-
{
|
| 572 |
-
"epoch": 2.99,
|
| 573 |
-
"learning_rate": 0.0002,
|
| 574 |
-
"loss": 1.9575,
|
| 575 |
-
"step": 4100
|
| 576 |
-
},
|
| 577 |
-
{
|
| 578 |
-
"epoch": 2.99,
|
| 579 |
-
"eval_loss": 1.9820196628570557,
|
| 580 |
-
"eval_runtime": 208.7546,
|
| 581 |
-
"eval_samples_per_second": 30.054,
|
| 582 |
-
"eval_steps_per_second": 3.76,
|
| 583 |
-
"step": 4100
|
| 584 |
}
|
| 585 |
],
|
| 586 |
"logging_steps": 100,
|
| 587 |
"max_steps": 4116,
|
| 588 |
"num_train_epochs": 3,
|
| 589 |
"save_steps": 100,
|
| 590 |
-
"total_flos":
|
| 591 |
"trial_name": null,
|
| 592 |
"trial_params": null
|
| 593 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 2.6418025493621826,
|
| 3 |
+
"best_model_checkpoint": "./outputs/checkpoint-100",
|
| 4 |
+
"epoch": 0.07285974499089254,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 100,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 0.07,
|
| 13 |
"learning_rate": 0.0002,
|
| 14 |
+
"loss": 2.7388,
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.07,
|
| 19 |
+
"eval_loss": 2.6418025493621826,
|
| 20 |
+
"eval_runtime": 204.724,
|
| 21 |
+
"eval_samples_per_second": 30.646,
|
| 22 |
+
"eval_steps_per_second": 3.834,
|
| 23 |
"step": 100
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
}
|
| 25 |
],
|
| 26 |
"logging_steps": 100,
|
| 27 |
"max_steps": 4116,
|
| 28 |
"num_train_epochs": 3,
|
| 29 |
"save_steps": 100,
|
| 30 |
+
"total_flos": 2917794121482240.0,
|
| 31 |
"trial_name": null,
|
| 32 |
"trial_params": null
|
| 33 |
}
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4219
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c66c4c8ecfd84aad650ec729a9571031304ebef290f19f46a950adc124725eb2
|
| 3 |
size 4219
|