Commit
•
35d2898
1
Parent(s):
560a4d5
Training in progress, step 50000
Browse files- adapter_model.bin +1 -1
- {checkpoint-46000 → checkpoint-49000/adapter_model}/README.md +0 -0
- {checkpoint-46000 → checkpoint-49000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-46000 → checkpoint-49000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-46000/adapter_model → checkpoint-50000}/README.md +0 -0
- {checkpoint-46000/adapter_model → checkpoint-50000}/adapter_config.json +0 -0
- {checkpoint-46000/adapter_model → checkpoint-50000}/adapter_model.bin +1 -1
- {checkpoint-46000 → checkpoint-50000}/optimizer.pt +1 -1
- {checkpoint-46000 → checkpoint-50000}/rng_state.pth +1 -1
- {checkpoint-46000 → checkpoint-50000}/scheduler.pt +1 -1
- {checkpoint-46000 → checkpoint-50000}/trainer_state.json +277 -5
- {checkpoint-46000 → checkpoint-50000}/training_args.bin +0 -0
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32be7061508650a7b44573a520d3c128102014dd59e5edcb15fba389f54f54b2
|
| 3 |
size 500897101
|
{checkpoint-46000 → checkpoint-49000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-46000 → checkpoint-49000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-46000 → checkpoint-49000/adapter_model}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7f22ce6ae60ccd99df75571e4088978fc9ea69fb44c013def638981af59f74d
|
| 3 |
size 500897101
|
{checkpoint-46000/adapter_model → checkpoint-50000}/README.md
RENAMED
|
File without changes
|
{checkpoint-46000/adapter_model → checkpoint-50000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-46000/adapter_model → checkpoint-50000}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32be7061508650a7b44573a520d3c128102014dd59e5edcb15fba389f54f54b2
|
| 3 |
size 500897101
|
{checkpoint-46000 → checkpoint-50000}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1001723453
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3087a179049987e8bdb2eef11934226e684e5d3c25576f5f628b38468571253b
|
| 3 |
size 1001723453
|
{checkpoint-46000 → checkpoint-50000}/rng_state.pth
RENAMED
|
@@ -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 sha256:820be5b55e6b5f5d702c191a5edda95aa32948ce2c589072c579566ed1905628
|
| 3 |
size 14575
|
{checkpoint-46000 → checkpoint-50000}/scheduler.pt
RENAMED
|
@@ -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:bd8fc03bd8adaa68647e06a4aa474a2806173acac7d29895eca100c8286e4e71
|
| 3 |
size 627
|
{checkpoint-46000 → checkpoint-50000}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./qlora-out/checkpoint-
|
| 4 |
-
"epoch": 1.
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -3134,11 +3134,283 @@
|
|
| 3134 |
"eval_samples_per_second": 0.429,
|
| 3135 |
"eval_steps_per_second": 0.429,
|
| 3136 |
"step": 46000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3137 |
}
|
| 3138 |
],
|
| 3139 |
"max_steps": 80463,
|
| 3140 |
"num_train_epochs": 3,
|
| 3141 |
-
"total_flos": 1.
|
| 3142 |
"trial_name": null,
|
| 3143 |
"trial_params": null
|
| 3144 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.507805347442627,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-50000",
|
| 4 |
+
"epoch": 1.864210879534693,
|
| 5 |
+
"global_step": 50000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 3134 |
"eval_samples_per_second": 0.429,
|
| 3135 |
"eval_steps_per_second": 0.429,
|
| 3136 |
"step": 46000
|
| 3137 |
+
},
|
| 3138 |
+
{
|
| 3139 |
+
"epoch": 1.72,
|
| 3140 |
+
"learning_rate": 7.730326748289895e-05,
|
| 3141 |
+
"loss": 0.4525,
|
| 3142 |
+
"step": 46100
|
| 3143 |
+
},
|
| 3144 |
+
{
|
| 3145 |
+
"epoch": 1.72,
|
| 3146 |
+
"learning_rate": 7.6923144354732e-05,
|
| 3147 |
+
"loss": 0.483,
|
| 3148 |
+
"step": 46200
|
| 3149 |
+
},
|
| 3150 |
+
{
|
| 3151 |
+
"epoch": 1.73,
|
| 3152 |
+
"learning_rate": 7.654337310393787e-05,
|
| 3153 |
+
"loss": 0.4941,
|
| 3154 |
+
"step": 46300
|
| 3155 |
+
},
|
| 3156 |
+
{
|
| 3157 |
+
"epoch": 1.73,
|
| 3158 |
+
"learning_rate": 7.616395952129287e-05,
|
| 3159 |
+
"loss": 0.4637,
|
| 3160 |
+
"step": 46400
|
| 3161 |
+
},
|
| 3162 |
+
{
|
| 3163 |
+
"epoch": 1.73,
|
| 3164 |
+
"learning_rate": 7.578490939211965e-05,
|
| 3165 |
+
"loss": 0.4874,
|
| 3166 |
+
"step": 46500
|
| 3167 |
+
},
|
| 3168 |
+
{
|
| 3169 |
+
"epoch": 1.74,
|
| 3170 |
+
"learning_rate": 7.540622849619883e-05,
|
| 3171 |
+
"loss": 0.4815,
|
| 3172 |
+
"step": 46600
|
| 3173 |
+
},
|
| 3174 |
+
{
|
| 3175 |
+
"epoch": 1.74,
|
| 3176 |
+
"learning_rate": 7.5027922607681e-05,
|
| 3177 |
+
"loss": 0.4973,
|
| 3178 |
+
"step": 46700
|
| 3179 |
+
},
|
| 3180 |
+
{
|
| 3181 |
+
"epoch": 1.74,
|
| 3182 |
+
"learning_rate": 7.46499974949985e-05,
|
| 3183 |
+
"loss": 0.5003,
|
| 3184 |
+
"step": 46800
|
| 3185 |
+
},
|
| 3186 |
+
{
|
| 3187 |
+
"epoch": 1.75,
|
| 3188 |
+
"learning_rate": 7.427245892077775e-05,
|
| 3189 |
+
"loss": 0.4331,
|
| 3190 |
+
"step": 46900
|
| 3191 |
+
},
|
| 3192 |
+
{
|
| 3193 |
+
"epoch": 1.75,
|
| 3194 |
+
"learning_rate": 7.389531264175103e-05,
|
| 3195 |
+
"loss": 0.4691,
|
| 3196 |
+
"step": 47000
|
| 3197 |
+
},
|
| 3198 |
+
{
|
| 3199 |
+
"epoch": 1.75,
|
| 3200 |
+
"eval_loss": 0.5184645652770996,
|
| 3201 |
+
"eval_runtime": 1306.5029,
|
| 3202 |
+
"eval_samples_per_second": 0.415,
|
| 3203 |
+
"eval_steps_per_second": 0.415,
|
| 3204 |
+
"step": 47000
|
| 3205 |
+
},
|
| 3206 |
+
{
|
| 3207 |
+
"epoch": 1.76,
|
| 3208 |
+
"learning_rate": 7.351856440866895e-05,
|
| 3209 |
+
"loss": 0.4875,
|
| 3210 |
+
"step": 47100
|
| 3211 |
+
},
|
| 3212 |
+
{
|
| 3213 |
+
"epoch": 1.76,
|
| 3214 |
+
"learning_rate": 7.314221996621279e-05,
|
| 3215 |
+
"loss": 0.4615,
|
| 3216 |
+
"step": 47200
|
| 3217 |
+
},
|
| 3218 |
+
{
|
| 3219 |
+
"epoch": 1.76,
|
| 3220 |
+
"learning_rate": 7.276628505290663e-05,
|
| 3221 |
+
"loss": 0.4522,
|
| 3222 |
+
"step": 47300
|
| 3223 |
+
},
|
| 3224 |
+
{
|
| 3225 |
+
"epoch": 1.77,
|
| 3226 |
+
"learning_rate": 7.239076540103013e-05,
|
| 3227 |
+
"loss": 0.4861,
|
| 3228 |
+
"step": 47400
|
| 3229 |
+
},
|
| 3230 |
+
{
|
| 3231 |
+
"epoch": 1.77,
|
| 3232 |
+
"learning_rate": 7.2015666736531e-05,
|
| 3233 |
+
"loss": 0.4676,
|
| 3234 |
+
"step": 47500
|
| 3235 |
+
},
|
| 3236 |
+
{
|
| 3237 |
+
"epoch": 1.77,
|
| 3238 |
+
"learning_rate": 7.164099477893768e-05,
|
| 3239 |
+
"loss": 0.4451,
|
| 3240 |
+
"step": 47600
|
| 3241 |
+
},
|
| 3242 |
+
{
|
| 3243 |
+
"epoch": 1.78,
|
| 3244 |
+
"learning_rate": 7.126675524127217e-05,
|
| 3245 |
+
"loss": 0.456,
|
| 3246 |
+
"step": 47700
|
| 3247 |
+
},
|
| 3248 |
+
{
|
| 3249 |
+
"epoch": 1.78,
|
| 3250 |
+
"learning_rate": 7.089295382996294e-05,
|
| 3251 |
+
"loss": 0.4676,
|
| 3252 |
+
"step": 47800
|
| 3253 |
+
},
|
| 3254 |
+
{
|
| 3255 |
+
"epoch": 1.79,
|
| 3256 |
+
"learning_rate": 7.05195962447578e-05,
|
| 3257 |
+
"loss": 0.4687,
|
| 3258 |
+
"step": 47900
|
| 3259 |
+
},
|
| 3260 |
+
{
|
| 3261 |
+
"epoch": 1.79,
|
| 3262 |
+
"learning_rate": 7.014668817863719e-05,
|
| 3263 |
+
"loss": 0.4415,
|
| 3264 |
+
"step": 48000
|
| 3265 |
+
},
|
| 3266 |
+
{
|
| 3267 |
+
"epoch": 1.79,
|
| 3268 |
+
"eval_loss": 0.5149570107460022,
|
| 3269 |
+
"eval_runtime": 1238.5109,
|
| 3270 |
+
"eval_samples_per_second": 0.438,
|
| 3271 |
+
"eval_steps_per_second": 0.438,
|
| 3272 |
+
"step": 48000
|
| 3273 |
+
},
|
| 3274 |
+
{
|
| 3275 |
+
"epoch": 1.79,
|
| 3276 |
+
"learning_rate": 6.977423531772711e-05,
|
| 3277 |
+
"loss": 0.5093,
|
| 3278 |
+
"step": 48100
|
| 3279 |
+
},
|
| 3280 |
+
{
|
| 3281 |
+
"epoch": 1.8,
|
| 3282 |
+
"learning_rate": 6.940224334121264e-05,
|
| 3283 |
+
"loss": 0.4556,
|
| 3284 |
+
"step": 48200
|
| 3285 |
+
},
|
| 3286 |
+
{
|
| 3287 |
+
"epoch": 1.8,
|
| 3288 |
+
"learning_rate": 6.903071792125136e-05,
|
| 3289 |
+
"loss": 0.4554,
|
| 3290 |
+
"step": 48300
|
| 3291 |
+
},
|
| 3292 |
+
{
|
| 3293 |
+
"epoch": 1.8,
|
| 3294 |
+
"learning_rate": 6.865966472288655e-05,
|
| 3295 |
+
"loss": 0.4442,
|
| 3296 |
+
"step": 48400
|
| 3297 |
+
},
|
| 3298 |
+
{
|
| 3299 |
+
"epoch": 1.81,
|
| 3300 |
+
"learning_rate": 6.828908940396123e-05,
|
| 3301 |
+
"loss": 0.4583,
|
| 3302 |
+
"step": 48500
|
| 3303 |
+
},
|
| 3304 |
+
{
|
| 3305 |
+
"epoch": 1.81,
|
| 3306 |
+
"learning_rate": 6.791899761503153e-05,
|
| 3307 |
+
"loss": 0.4598,
|
| 3308 |
+
"step": 48600
|
| 3309 |
+
},
|
| 3310 |
+
{
|
| 3311 |
+
"epoch": 1.82,
|
| 3312 |
+
"learning_rate": 6.754939499928079e-05,
|
| 3313 |
+
"loss": 0.4541,
|
| 3314 |
+
"step": 48700
|
| 3315 |
+
},
|
| 3316 |
+
{
|
| 3317 |
+
"epoch": 1.82,
|
| 3318 |
+
"learning_rate": 6.718028719243335e-05,
|
| 3319 |
+
"loss": 0.4474,
|
| 3320 |
+
"step": 48800
|
| 3321 |
+
},
|
| 3322 |
+
{
|
| 3323 |
+
"epoch": 1.82,
|
| 3324 |
+
"learning_rate": 6.68116798226687e-05,
|
| 3325 |
+
"loss": 0.4531,
|
| 3326 |
+
"step": 48900
|
| 3327 |
+
},
|
| 3328 |
+
{
|
| 3329 |
+
"epoch": 1.83,
|
| 3330 |
+
"learning_rate": 6.644357851053562e-05,
|
| 3331 |
+
"loss": 0.4748,
|
| 3332 |
+
"step": 49000
|
| 3333 |
+
},
|
| 3334 |
+
{
|
| 3335 |
+
"epoch": 1.83,
|
| 3336 |
+
"eval_loss": 0.5114006400108337,
|
| 3337 |
+
"eval_runtime": 1274.2814,
|
| 3338 |
+
"eval_samples_per_second": 0.425,
|
| 3339 |
+
"eval_steps_per_second": 0.425,
|
| 3340 |
+
"step": 49000
|
| 3341 |
+
},
|
| 3342 |
+
{
|
| 3343 |
+
"epoch": 1.83,
|
| 3344 |
+
"learning_rate": 6.607598886886645e-05,
|
| 3345 |
+
"loss": 0.4803,
|
| 3346 |
+
"step": 49100
|
| 3347 |
+
},
|
| 3348 |
+
{
|
| 3349 |
+
"epoch": 1.83,
|
| 3350 |
+
"learning_rate": 6.57089165026916e-05,
|
| 3351 |
+
"loss": 0.4703,
|
| 3352 |
+
"step": 49200
|
| 3353 |
+
},
|
| 3354 |
+
{
|
| 3355 |
+
"epoch": 1.84,
|
| 3356 |
+
"learning_rate": 6.534236700915406e-05,
|
| 3357 |
+
"loss": 0.4963,
|
| 3358 |
+
"step": 49300
|
| 3359 |
+
},
|
| 3360 |
+
{
|
| 3361 |
+
"epoch": 1.84,
|
| 3362 |
+
"learning_rate": 6.497634597742399e-05,
|
| 3363 |
+
"loss": 0.4694,
|
| 3364 |
+
"step": 49400
|
| 3365 |
+
},
|
| 3366 |
+
{
|
| 3367 |
+
"epoch": 1.85,
|
| 3368 |
+
"learning_rate": 6.46108589886135e-05,
|
| 3369 |
+
"loss": 0.4612,
|
| 3370 |
+
"step": 49500
|
| 3371 |
+
},
|
| 3372 |
+
{
|
| 3373 |
+
"epoch": 1.85,
|
| 3374 |
+
"learning_rate": 6.424591161569158e-05,
|
| 3375 |
+
"loss": 0.4361,
|
| 3376 |
+
"step": 49600
|
| 3377 |
+
},
|
| 3378 |
+
{
|
| 3379 |
+
"epoch": 1.85,
|
| 3380 |
+
"learning_rate": 6.388150942339917e-05,
|
| 3381 |
+
"loss": 0.4652,
|
| 3382 |
+
"step": 49700
|
| 3383 |
+
},
|
| 3384 |
+
{
|
| 3385 |
+
"epoch": 1.86,
|
| 3386 |
+
"learning_rate": 6.35176579681642e-05,
|
| 3387 |
+
"loss": 0.4741,
|
| 3388 |
+
"step": 49800
|
| 3389 |
+
},
|
| 3390 |
+
{
|
| 3391 |
+
"epoch": 1.86,
|
| 3392 |
+
"learning_rate": 6.315436279801704e-05,
|
| 3393 |
+
"loss": 0.4447,
|
| 3394 |
+
"step": 49900
|
| 3395 |
+
},
|
| 3396 |
+
{
|
| 3397 |
+
"epoch": 1.86,
|
| 3398 |
+
"learning_rate": 6.279162945250561e-05,
|
| 3399 |
+
"loss": 0.4542,
|
| 3400 |
+
"step": 50000
|
| 3401 |
+
},
|
| 3402 |
+
{
|
| 3403 |
+
"epoch": 1.86,
|
| 3404 |
+
"eval_loss": 0.507805347442627,
|
| 3405 |
+
"eval_runtime": 1229.0064,
|
| 3406 |
+
"eval_samples_per_second": 0.441,
|
| 3407 |
+
"eval_steps_per_second": 0.441,
|
| 3408 |
+
"step": 50000
|
| 3409 |
}
|
| 3410 |
],
|
| 3411 |
"max_steps": 80463,
|
| 3412 |
"num_train_epochs": 3,
|
| 3413 |
+
"total_flos": 1.4020960472087839e+19,
|
| 3414 |
"trial_name": null,
|
| 3415 |
"trial_params": null
|
| 3416 |
}
|
{checkpoint-46000 → checkpoint-50000}/training_args.bin
RENAMED
|
File without changes
|