Training in progress, step 24000, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1153648
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36e7c8e59f44842b339fac2ef000aea78811dbb794e276f14fed4bd6df5cea78
|
| 3 |
size 1153648
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2334411
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c2f6bdce455b9c98bcb1cd1f8e6fbbebff571aca2b3a9401a1dd0a554a62a5ac
|
| 3 |
size 2334411
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14645
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe23bac147ae4f270ef388d0372bf9f6ef6c0243d969a719ae9d2e248641e942
|
| 3 |
size 14645
|
last-checkpoint/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1383
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e49b9119cfe53a5f9361daba9a41e626c2662be41e03d5f6df7991b02a776fc9
|
| 3 |
size 1383
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a43cee2ab7b75b1990ef803da3443a3aebff596ad01f8bbf4e0371ee6d1c8d3
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.
|
| 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,
|
|
@@ -3088,6 +3088,286 @@
|
|
| 3088 |
"learning_rate": 0.000577392607415337,
|
| 3089 |
"loss": 5.046322631835937,
|
| 3090 |
"step": 22000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3091 |
}
|
| 3092 |
],
|
| 3093 |
"logging_steps": 50,
|
|
@@ -3107,7 +3387,7 @@
|
|
| 3107 |
"attributes": {}
|
| 3108 |
}
|
| 3109 |
},
|
| 3110 |
-
"total_flos":
|
| 3111 |
"train_batch_size": 16,
|
| 3112 |
"trial_name": null,
|
| 3113 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.4915111921194372,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 24000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 3088 |
"learning_rate": 0.000577392607415337,
|
| 3089 |
"loss": 5.046322631835937,
|
| 3090 |
"step": 22000
|
| 3091 |
+
},
|
| 3092 |
+
{
|
| 3093 |
+
"epoch": 0.45157590775973294,
|
| 3094 |
+
"grad_norm": 0.9067041277885437,
|
| 3095 |
+
"learning_rate": 0.000575803128093074,
|
| 3096 |
+
"loss": 5.041178894042969,
|
| 3097 |
+
"step": 22050
|
| 3098 |
+
},
|
| 3099 |
+
{
|
| 3100 |
+
"epoch": 0.45259988940998175,
|
| 3101 |
+
"grad_norm": 0.9458829164505005,
|
| 3102 |
+
"learning_rate": 0.0005742128643107327,
|
| 3103 |
+
"loss": 5.038160705566407,
|
| 3104 |
+
"step": 22100
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 0.4536238710602306,
|
| 3108 |
+
"grad_norm": 0.8975867629051208,
|
| 3109 |
+
"learning_rate": 0.0005726218325253978,
|
| 3110 |
+
"loss": 5.018882446289062,
|
| 3111 |
+
"step": 22150
|
| 3112 |
+
},
|
| 3113 |
+
{
|
| 3114 |
+
"epoch": 0.4546478527104794,
|
| 3115 |
+
"grad_norm": 0.6916885375976562,
|
| 3116 |
+
"learning_rate": 0.0005710300492021006,
|
| 3117 |
+
"loss": 5.032647705078125,
|
| 3118 |
+
"step": 22200
|
| 3119 |
+
},
|
| 3120 |
+
{
|
| 3121 |
+
"epoch": 0.4556718343607283,
|
| 3122 |
+
"grad_norm": 0.948345959186554,
|
| 3123 |
+
"learning_rate": 0.0005694375308136506,
|
| 3124 |
+
"loss": 5.031848449707031,
|
| 3125 |
+
"step": 22250
|
| 3126 |
+
},
|
| 3127 |
+
{
|
| 3128 |
+
"epoch": 0.4566958160109771,
|
| 3129 |
+
"grad_norm": 0.8653504252433777,
|
| 3130 |
+
"learning_rate": 0.0005678442938404642,
|
| 3131 |
+
"loss": 5.028753662109375,
|
| 3132 |
+
"step": 22300
|
| 3133 |
+
},
|
| 3134 |
+
{
|
| 3135 |
+
"epoch": 0.4577197976612259,
|
| 3136 |
+
"grad_norm": 0.8769348859786987,
|
| 3137 |
+
"learning_rate": 0.0005662503547703934,
|
| 3138 |
+
"loss": 5.036540222167969,
|
| 3139 |
+
"step": 22350
|
| 3140 |
+
},
|
| 3141 |
+
{
|
| 3142 |
+
"epoch": 0.45874377931147475,
|
| 3143 |
+
"grad_norm": 1.1168607473373413,
|
| 3144 |
+
"learning_rate": 0.0005646557300985572,
|
| 3145 |
+
"loss": 5.023377380371094,
|
| 3146 |
+
"step": 22400
|
| 3147 |
+
},
|
| 3148 |
+
{
|
| 3149 |
+
"epoch": 0.45976776096172356,
|
| 3150 |
+
"grad_norm": 0.7529086470603943,
|
| 3151 |
+
"learning_rate": 0.0005630604363271685,
|
| 3152 |
+
"loss": 5.049166564941406,
|
| 3153 |
+
"step": 22450
|
| 3154 |
+
},
|
| 3155 |
+
{
|
| 3156 |
+
"epoch": 0.46079174261197237,
|
| 3157 |
+
"grad_norm": 0.7045357823371887,
|
| 3158 |
+
"learning_rate": 0.0005614644899653652,
|
| 3159 |
+
"loss": 5.023310241699218,
|
| 3160 |
+
"step": 22500
|
| 3161 |
+
},
|
| 3162 |
+
{
|
| 3163 |
+
"epoch": 0.4618157242622212,
|
| 3164 |
+
"grad_norm": 0.8497369289398193,
|
| 3165 |
+
"learning_rate": 0.0005598679075290381,
|
| 3166 |
+
"loss": 5.041716003417969,
|
| 3167 |
+
"step": 22550
|
| 3168 |
+
},
|
| 3169 |
+
{
|
| 3170 |
+
"epoch": 0.46283970591247003,
|
| 3171 |
+
"grad_norm": 0.8301413655281067,
|
| 3172 |
+
"learning_rate": 0.000558270705540661,
|
| 3173 |
+
"loss": 5.019425659179688,
|
| 3174 |
+
"step": 22600
|
| 3175 |
+
},
|
| 3176 |
+
{
|
| 3177 |
+
"epoch": 0.4638636875627189,
|
| 3178 |
+
"grad_norm": 1.038455843925476,
|
| 3179 |
+
"learning_rate": 0.000556672900529119,
|
| 3180 |
+
"loss": 5.0140682983398435,
|
| 3181 |
+
"step": 22650
|
| 3182 |
+
},
|
| 3183 |
+
{
|
| 3184 |
+
"epoch": 0.4648876692129677,
|
| 3185 |
+
"grad_norm": 0.7119101881980896,
|
| 3186 |
+
"learning_rate": 0.0005550745090295378,
|
| 3187 |
+
"loss": 5.0490634155273435,
|
| 3188 |
+
"step": 22700
|
| 3189 |
+
},
|
| 3190 |
+
{
|
| 3191 |
+
"epoch": 0.4659116508632165,
|
| 3192 |
+
"grad_norm": 0.8264387249946594,
|
| 3193 |
+
"learning_rate": 0.0005534755475831121,
|
| 3194 |
+
"loss": 5.018643188476562,
|
| 3195 |
+
"step": 22750
|
| 3196 |
+
},
|
| 3197 |
+
{
|
| 3198 |
+
"epoch": 0.46693563251346537,
|
| 3199 |
+
"grad_norm": 0.6974812150001526,
|
| 3200 |
+
"learning_rate": 0.0005518760327369353,
|
| 3201 |
+
"loss": 5.034531860351563,
|
| 3202 |
+
"step": 22800
|
| 3203 |
+
},
|
| 3204 |
+
{
|
| 3205 |
+
"epoch": 0.4679596141637142,
|
| 3206 |
+
"grad_norm": 0.6481902599334717,
|
| 3207 |
+
"learning_rate": 0.0005502759810438275,
|
| 3208 |
+
"loss": 5.030080261230469,
|
| 3209 |
+
"step": 22850
|
| 3210 |
+
},
|
| 3211 |
+
{
|
| 3212 |
+
"epoch": 0.46898359581396304,
|
| 3213 |
+
"grad_norm": 0.8283994197845459,
|
| 3214 |
+
"learning_rate": 0.0005486754090621642,
|
| 3215 |
+
"loss": 5.034128723144531,
|
| 3216 |
+
"step": 22900
|
| 3217 |
+
},
|
| 3218 |
+
{
|
| 3219 |
+
"epoch": 0.47000757746421185,
|
| 3220 |
+
"grad_norm": 0.7362484335899353,
|
| 3221 |
+
"learning_rate": 0.0005470743333557055,
|
| 3222 |
+
"loss": 5.063320617675782,
|
| 3223 |
+
"step": 22950
|
| 3224 |
+
},
|
| 3225 |
+
{
|
| 3226 |
+
"epoch": 0.47103155911446065,
|
| 3227 |
+
"grad_norm": 0.7104576230049133,
|
| 3228 |
+
"learning_rate": 0.0005454727704934241,
|
| 3229 |
+
"loss": 5.0402743530273435,
|
| 3230 |
+
"step": 23000
|
| 3231 |
+
},
|
| 3232 |
+
{
|
| 3233 |
+
"epoch": 0.4720555407647095,
|
| 3234 |
+
"grad_norm": 0.8112474679946899,
|
| 3235 |
+
"learning_rate": 0.0005438707370493348,
|
| 3236 |
+
"loss": 5.014774475097656,
|
| 3237 |
+
"step": 23050
|
| 3238 |
+
},
|
| 3239 |
+
{
|
| 3240 |
+
"epoch": 0.4730795224149583,
|
| 3241 |
+
"grad_norm": 1.010627269744873,
|
| 3242 |
+
"learning_rate": 0.0005422682496023214,
|
| 3243 |
+
"loss": 5.022322082519532,
|
| 3244 |
+
"step": 23100
|
| 3245 |
+
},
|
| 3246 |
+
{
|
| 3247 |
+
"epoch": 0.4741035040652071,
|
| 3248 |
+
"grad_norm": 0.8355703949928284,
|
| 3249 |
+
"learning_rate": 0.0005406653247359662,
|
| 3250 |
+
"loss": 5.047538146972657,
|
| 3251 |
+
"step": 23150
|
| 3252 |
+
},
|
| 3253 |
+
{
|
| 3254 |
+
"epoch": 0.475127485715456,
|
| 3255 |
+
"grad_norm": 0.9319697618484497,
|
| 3256 |
+
"learning_rate": 0.0005390619790383789,
|
| 3257 |
+
"loss": 5.015643005371094,
|
| 3258 |
+
"step": 23200
|
| 3259 |
+
},
|
| 3260 |
+
{
|
| 3261 |
+
"epoch": 0.4761514673657048,
|
| 3262 |
+
"grad_norm": 0.8085825443267822,
|
| 3263 |
+
"learning_rate": 0.0005374582291020233,
|
| 3264 |
+
"loss": 5.049990234375,
|
| 3265 |
+
"step": 23250
|
| 3266 |
+
},
|
| 3267 |
+
{
|
| 3268 |
+
"epoch": 0.47717544901595366,
|
| 3269 |
+
"grad_norm": 0.9724649786949158,
|
| 3270 |
+
"learning_rate": 0.000535854091523547,
|
| 3271 |
+
"loss": 5.015569152832032,
|
| 3272 |
+
"step": 23300
|
| 3273 |
+
},
|
| 3274 |
+
{
|
| 3275 |
+
"epoch": 0.47819943066620246,
|
| 3276 |
+
"grad_norm": 0.7908557653427124,
|
| 3277 |
+
"learning_rate": 0.0005342495829036095,
|
| 3278 |
+
"loss": 5.022587890625,
|
| 3279 |
+
"step": 23350
|
| 3280 |
+
},
|
| 3281 |
+
{
|
| 3282 |
+
"epoch": 0.47922341231645127,
|
| 3283 |
+
"grad_norm": 0.6975404024124146,
|
| 3284 |
+
"learning_rate": 0.0005326447198467092,
|
| 3285 |
+
"loss": 4.996109313964844,
|
| 3286 |
+
"step": 23400
|
| 3287 |
+
},
|
| 3288 |
+
{
|
| 3289 |
+
"epoch": 0.48024739396670013,
|
| 3290 |
+
"grad_norm": 0.7565339803695679,
|
| 3291 |
+
"learning_rate": 0.0005310395189610133,
|
| 3292 |
+
"loss": 5.0324838256835935,
|
| 3293 |
+
"step": 23450
|
| 3294 |
+
},
|
| 3295 |
+
{
|
| 3296 |
+
"epoch": 0.48127137561694894,
|
| 3297 |
+
"grad_norm": 0.6791459918022156,
|
| 3298 |
+
"learning_rate": 0.0005294339968581845,
|
| 3299 |
+
"loss": 5.0436093139648435,
|
| 3300 |
+
"step": 23500
|
| 3301 |
+
},
|
| 3302 |
+
{
|
| 3303 |
+
"epoch": 0.4822953572671978,
|
| 3304 |
+
"grad_norm": 0.7872998118400574,
|
| 3305 |
+
"learning_rate": 0.0005278281701532099,
|
| 3306 |
+
"loss": 5.04802734375,
|
| 3307 |
+
"step": 23550
|
| 3308 |
+
},
|
| 3309 |
+
{
|
| 3310 |
+
"epoch": 0.4833193389174466,
|
| 3311 |
+
"grad_norm": 0.7677999138832092,
|
| 3312 |
+
"learning_rate": 0.000526222055464229,
|
| 3313 |
+
"loss": 5.04433349609375,
|
| 3314 |
+
"step": 23600
|
| 3315 |
+
},
|
| 3316 |
+
{
|
| 3317 |
+
"epoch": 0.4843433205676954,
|
| 3318 |
+
"grad_norm": 0.6659406423568726,
|
| 3319 |
+
"learning_rate": 0.0005246156694123611,
|
| 3320 |
+
"loss": 5.021984558105469,
|
| 3321 |
+
"step": 23650
|
| 3322 |
+
},
|
| 3323 |
+
{
|
| 3324 |
+
"epoch": 0.4853673022179443,
|
| 3325 |
+
"grad_norm": 2.627041816711426,
|
| 3326 |
+
"learning_rate": 0.0005230090286215343,
|
| 3327 |
+
"loss": 5.054336853027344,
|
| 3328 |
+
"step": 23700
|
| 3329 |
+
},
|
| 3330 |
+
{
|
| 3331 |
+
"epoch": 0.4863912838681931,
|
| 3332 |
+
"grad_norm": 0.718858540058136,
|
| 3333 |
+
"learning_rate": 0.0005214021497183123,
|
| 3334 |
+
"loss": 5.05353515625,
|
| 3335 |
+
"step": 23750
|
| 3336 |
+
},
|
| 3337 |
+
{
|
| 3338 |
+
"epoch": 0.4874152655184419,
|
| 3339 |
+
"grad_norm": 0.7855371236801147,
|
| 3340 |
+
"learning_rate": 0.0005197950493317233,
|
| 3341 |
+
"loss": 5.015243835449219,
|
| 3342 |
+
"step": 23800
|
| 3343 |
+
},
|
| 3344 |
+
{
|
| 3345 |
+
"epoch": 0.48843924716869075,
|
| 3346 |
+
"grad_norm": 1.190674901008606,
|
| 3347 |
+
"learning_rate": 0.0005181877440930875,
|
| 3348 |
+
"loss": 5.060979309082032,
|
| 3349 |
+
"step": 23850
|
| 3350 |
+
},
|
| 3351 |
+
{
|
| 3352 |
+
"epoch": 0.48946322881893956,
|
| 3353 |
+
"grad_norm": 0.8439124822616577,
|
| 3354 |
+
"learning_rate": 0.0005165802506358451,
|
| 3355 |
+
"loss": 5.124266967773438,
|
| 3356 |
+
"step": 23900
|
| 3357 |
+
},
|
| 3358 |
+
{
|
| 3359 |
+
"epoch": 0.4904872104691884,
|
| 3360 |
+
"grad_norm": 0.7516206502914429,
|
| 3361 |
+
"learning_rate": 0.0005149725855953841,
|
| 3362 |
+
"loss": 5.03356201171875,
|
| 3363 |
+
"step": 23950
|
| 3364 |
+
},
|
| 3365 |
+
{
|
| 3366 |
+
"epoch": 0.4915111921194372,
|
| 3367 |
+
"grad_norm": 0.8173161149024963,
|
| 3368 |
+
"learning_rate": 0.0005133647656088678,
|
| 3369 |
+
"loss": 5.039023742675782,
|
| 3370 |
+
"step": 24000
|
| 3371 |
}
|
| 3372 |
],
|
| 3373 |
"logging_steps": 50,
|
|
|
|
| 3387 |
"attributes": {}
|
| 3388 |
}
|
| 3389 |
},
|
| 3390 |
+
"total_flos": 312786026496000.0,
|
| 3391 |
"train_batch_size": 16,
|
| 3392 |
"trial_name": null,
|
| 3393 |
"trial_params": null
|