Training in progress, step 4884, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
|
@@ -33,13 +33,13 @@
|
|
| 33 |
"rank_pattern": {},
|
| 34 |
"revision": null,
|
| 35 |
"target_modules": [
|
| 36 |
-
"
|
| 37 |
"k_proj",
|
| 38 |
-
"down_proj",
|
| 39 |
-
"v_proj",
|
| 40 |
"o_proj",
|
|
|
|
|
|
|
| 41 |
"gate_proj",
|
| 42 |
-
"
|
| 43 |
],
|
| 44 |
"target_parameters": null,
|
| 45 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 33 |
"rank_pattern": {},
|
| 34 |
"revision": null,
|
| 35 |
"target_modules": [
|
| 36 |
+
"up_proj",
|
| 37 |
"k_proj",
|
|
|
|
|
|
|
| 38 |
"o_proj",
|
| 39 |
+
"q_proj",
|
| 40 |
+
"down_proj",
|
| 41 |
"gate_proj",
|
| 42 |
+
"v_proj"
|
| 43 |
],
|
| 44 |
"target_parameters": null,
|
| 45 |
"task_type": "CAUSAL_LM",
|
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 44871152
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d5b422ac5f04a7a4177cdba90fc93d851d10998551e2497f0041fbeb1727c03
|
| 3 |
size 44871152
|
last-checkpoint/optimizer.pt
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 sha256:282915048ac85bc5b6a15d87a72d32807a901c579b2b093a344d4b64fb53752e
|
| 3 |
+
size 24034627
|
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:f93653f1c06ce15c05df15b4d2b5f8a54bf434de9d23e07e064d47da48f70882
|
| 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 sha256:6ca82344486f1216c74a1c8be0e3c3136dd89f87a4958ff66f8be7174bafd1f9
|
| 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":
|
| 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,
|
|
@@ -3060,6 +3060,370 @@
|
|
| 3060 |
"learning_rate": 6.00053082583324e-06,
|
| 3061 |
"loss": 0.4025435924530029,
|
| 3062 |
"step": 4360
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3063 |
}
|
| 3064 |
],
|
| 3065 |
"logging_steps": 10,
|
|
@@ -3074,12 +3438,12 @@
|
|
| 3074 |
"should_evaluate": false,
|
| 3075 |
"should_log": false,
|
| 3076 |
"should_save": true,
|
| 3077 |
-
"should_training_stop":
|
| 3078 |
},
|
| 3079 |
"attributes": {}
|
| 3080 |
}
|
| 3081 |
},
|
| 3082 |
-
"total_flos":
|
| 3083 |
"train_batch_size": 4,
|
| 3084 |
"trial_name": null,
|
| 3085 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 4884,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 3060 |
"learning_rate": 6.00053082583324e-06,
|
| 3061 |
"loss": 0.4025435924530029,
|
| 3062 |
"step": 4360
|
| 3063 |
+
},
|
| 3064 |
+
{
|
| 3065 |
+
"epoch": 1.7895976246544487,
|
| 3066 |
+
"grad_norm": 0.21897482872009277,
|
| 3067 |
+
"learning_rate": 5.776322089099384e-06,
|
| 3068 |
+
"loss": 0.4200548529624939,
|
| 3069 |
+
"step": 4370
|
| 3070 |
+
},
|
| 3071 |
+
{
|
| 3072 |
+
"epoch": 1.7936930480188389,
|
| 3073 |
+
"grad_norm": 0.1700364649295807,
|
| 3074 |
+
"learning_rate": 5.556257654757624e-06,
|
| 3075 |
+
"loss": 0.34192743301391604,
|
| 3076 |
+
"step": 4380
|
| 3077 |
+
},
|
| 3078 |
+
{
|
| 3079 |
+
"epoch": 1.7977884713832293,
|
| 3080 |
+
"grad_norm": 0.2201203852891922,
|
| 3081 |
+
"learning_rate": 5.340347202047669e-06,
|
| 3082 |
+
"loss": 0.3917534828186035,
|
| 3083 |
+
"step": 4390
|
| 3084 |
+
},
|
| 3085 |
+
{
|
| 3086 |
+
"epoch": 1.8018838947476197,
|
| 3087 |
+
"grad_norm": 0.2158016562461853,
|
| 3088 |
+
"learning_rate": 5.128600227501834e-06,
|
| 3089 |
+
"loss": 0.3703617572784424,
|
| 3090 |
+
"step": 4400
|
| 3091 |
+
},
|
| 3092 |
+
{
|
| 3093 |
+
"epoch": 1.8059793181120098,
|
| 3094 |
+
"grad_norm": 0.2428576648235321,
|
| 3095 |
+
"learning_rate": 4.921026044527466e-06,
|
| 3096 |
+
"loss": 0.4067427635192871,
|
| 3097 |
+
"step": 4410
|
| 3098 |
+
},
|
| 3099 |
+
{
|
| 3100 |
+
"epoch": 1.8100747414764,
|
| 3101 |
+
"grad_norm": 0.186968132853508,
|
| 3102 |
+
"learning_rate": 4.7176337829972884e-06,
|
| 3103 |
+
"loss": 0.377484655380249,
|
| 3104 |
+
"step": 4420
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 1.8141701648407904,
|
| 3108 |
+
"grad_norm": 0.21294040977954865,
|
| 3109 |
+
"learning_rate": 4.518432388847705e-06,
|
| 3110 |
+
"loss": 0.4246979713439941,
|
| 3111 |
+
"step": 4430
|
| 3112 |
+
},
|
| 3113 |
+
{
|
| 3114 |
+
"epoch": 1.8182655882051808,
|
| 3115 |
+
"grad_norm": 0.20785194635391235,
|
| 3116 |
+
"learning_rate": 4.3234306236855315e-06,
|
| 3117 |
+
"loss": 0.39151136875152587,
|
| 3118 |
+
"step": 4440
|
| 3119 |
+
},
|
| 3120 |
+
{
|
| 3121 |
+
"epoch": 1.822361011569571,
|
| 3122 |
+
"grad_norm": 0.2211901694536209,
|
| 3123 |
+
"learning_rate": 4.132637064402445e-06,
|
| 3124 |
+
"loss": 0.40515637397766113,
|
| 3125 |
+
"step": 4450
|
| 3126 |
+
},
|
| 3127 |
+
{
|
| 3128 |
+
"epoch": 1.8264564349339611,
|
| 3129 |
+
"grad_norm": 0.1828216165304184,
|
| 3130 |
+
"learning_rate": 3.946060102797877e-06,
|
| 3131 |
+
"loss": 0.37872354984283446,
|
| 3132 |
+
"step": 4460
|
| 3133 |
+
},
|
| 3134 |
+
{
|
| 3135 |
+
"epoch": 1.8305518582983515,
|
| 3136 |
+
"grad_norm": 0.19599486887454987,
|
| 3137 |
+
"learning_rate": 3.7637079452098045e-06,
|
| 3138 |
+
"loss": 0.3668804407119751,
|
| 3139 |
+
"step": 4470
|
| 3140 |
+
},
|
| 3141 |
+
{
|
| 3142 |
+
"epoch": 1.834647281662742,
|
| 3143 |
+
"grad_norm": 0.24288585782051086,
|
| 3144 |
+
"learning_rate": 3.585588612153867e-06,
|
| 3145 |
+
"loss": 0.4075489521026611,
|
| 3146 |
+
"step": 4480
|
| 3147 |
+
},
|
| 3148 |
+
{
|
| 3149 |
+
"epoch": 1.838742705027132,
|
| 3150 |
+
"grad_norm": 0.16427487134933472,
|
| 3151 |
+
"learning_rate": 3.4117099379706195e-06,
|
| 3152 |
+
"loss": 0.3669139385223389,
|
| 3153 |
+
"step": 4490
|
| 3154 |
+
},
|
| 3155 |
+
{
|
| 3156 |
+
"epoch": 1.8428381283915225,
|
| 3157 |
+
"grad_norm": 0.20215265452861786,
|
| 3158 |
+
"learning_rate": 3.2420795704808492e-06,
|
| 3159 |
+
"loss": 0.40285863876342776,
|
| 3160 |
+
"step": 4500
|
| 3161 |
+
},
|
| 3162 |
+
{
|
| 3163 |
+
"epoch": 1.846933551755913,
|
| 3164 |
+
"grad_norm": 0.1883266270160675,
|
| 3165 |
+
"learning_rate": 3.0767049706493136e-06,
|
| 3166 |
+
"loss": 0.39141478538513186,
|
| 3167 |
+
"step": 4510
|
| 3168 |
+
},
|
| 3169 |
+
{
|
| 3170 |
+
"epoch": 1.851028975120303,
|
| 3171 |
+
"grad_norm": 0.2351822555065155,
|
| 3172 |
+
"learning_rate": 2.915593412256479e-06,
|
| 3173 |
+
"loss": 0.3928338289260864,
|
| 3174 |
+
"step": 4520
|
| 3175 |
+
},
|
| 3176 |
+
{
|
| 3177 |
+
"epoch": 1.8551243984846932,
|
| 3178 |
+
"grad_norm": 0.21443790197372437,
|
| 3179 |
+
"learning_rate": 2.7587519815786336e-06,
|
| 3180 |
+
"loss": 0.37882561683654786,
|
| 3181 |
+
"step": 4530
|
| 3182 |
+
},
|
| 3183 |
+
{
|
| 3184 |
+
"epoch": 1.8592198218490836,
|
| 3185 |
+
"grad_norm": 0.26582658290863037,
|
| 3186 |
+
"learning_rate": 2.6061875770762913e-06,
|
| 3187 |
+
"loss": 0.40909738540649415,
|
| 3188 |
+
"step": 4540
|
| 3189 |
+
},
|
| 3190 |
+
{
|
| 3191 |
+
"epoch": 1.863315245213474,
|
| 3192 |
+
"grad_norm": 0.1906818002462387,
|
| 3193 |
+
"learning_rate": 2.457906909090613e-06,
|
| 3194 |
+
"loss": 0.3933555603027344,
|
| 3195 |
+
"step": 4550
|
| 3196 |
+
},
|
| 3197 |
+
{
|
| 3198 |
+
"epoch": 1.8674106685778642,
|
| 3199 |
+
"grad_norm": 0.20051346719264984,
|
| 3200 |
+
"learning_rate": 2.313916499548352e-06,
|
| 3201 |
+
"loss": 0.4398690700531006,
|
| 3202 |
+
"step": 4560
|
| 3203 |
+
},
|
| 3204 |
+
{
|
| 3205 |
+
"epoch": 1.8715060919422544,
|
| 3206 |
+
"grad_norm": 0.1904095709323883,
|
| 3207 |
+
"learning_rate": 2.174222681675009e-06,
|
| 3208 |
+
"loss": 0.3948297739028931,
|
| 3209 |
+
"step": 4570
|
| 3210 |
+
},
|
| 3211 |
+
{
|
| 3212 |
+
"epoch": 1.875601515306645,
|
| 3213 |
+
"grad_norm": 0.21440091729164124,
|
| 3214 |
+
"learning_rate": 2.03883159971624e-06,
|
| 3215 |
+
"loss": 0.40198678970336915,
|
| 3216 |
+
"step": 4580
|
| 3217 |
+
},
|
| 3218 |
+
{
|
| 3219 |
+
"epoch": 1.8796969386710352,
|
| 3220 |
+
"grad_norm": 0.16143283247947693,
|
| 3221 |
+
"learning_rate": 1.9077492086676087e-06,
|
| 3222 |
+
"loss": 0.35410706996917723,
|
| 3223 |
+
"step": 4590
|
| 3224 |
+
},
|
| 3225 |
+
{
|
| 3226 |
+
"epoch": 1.8837923620354253,
|
| 3227 |
+
"grad_norm": 0.20751744508743286,
|
| 3228 |
+
"learning_rate": 1.7809812740127053e-06,
|
| 3229 |
+
"loss": 0.3728046894073486,
|
| 3230 |
+
"step": 4600
|
| 3231 |
+
},
|
| 3232 |
+
{
|
| 3233 |
+
"epoch": 1.8878877853998157,
|
| 3234 |
+
"grad_norm": 0.2542204260826111,
|
| 3235 |
+
"learning_rate": 1.658533371469495e-06,
|
| 3236 |
+
"loss": 0.3811838626861572,
|
| 3237 |
+
"step": 4610
|
| 3238 |
+
},
|
| 3239 |
+
{
|
| 3240 |
+
"epoch": 1.8919832087642061,
|
| 3241 |
+
"grad_norm": 0.17141331732273102,
|
| 3242 |
+
"learning_rate": 1.540410886745114e-06,
|
| 3243 |
+
"loss": 0.3526181221008301,
|
| 3244 |
+
"step": 4620
|
| 3245 |
+
},
|
| 3246 |
+
{
|
| 3247 |
+
"epoch": 1.8960786321285963,
|
| 3248 |
+
"grad_norm": 0.23039455711841583,
|
| 3249 |
+
"learning_rate": 1.4266190152990023e-06,
|
| 3250 |
+
"loss": 0.38018968105316164,
|
| 3251 |
+
"step": 4630
|
| 3252 |
+
},
|
| 3253 |
+
{
|
| 3254 |
+
"epoch": 1.9001740554929865,
|
| 3255 |
+
"grad_norm": 0.2467181533575058,
|
| 3256 |
+
"learning_rate": 1.3171627621143656e-06,
|
| 3257 |
+
"loss": 0.3932963848114014,
|
| 3258 |
+
"step": 4640
|
| 3259 |
+
},
|
| 3260 |
+
{
|
| 3261 |
+
"epoch": 1.9042694788573769,
|
| 3262 |
+
"grad_norm": 0.181559756398201,
|
| 3263 |
+
"learning_rate": 1.2120469414780289e-06,
|
| 3264 |
+
"loss": 0.4113152503967285,
|
| 3265 |
+
"step": 4650
|
| 3266 |
+
},
|
| 3267 |
+
{
|
| 3268 |
+
"epoch": 1.9083649022217672,
|
| 3269 |
+
"grad_norm": 0.19314563274383545,
|
| 3270 |
+
"learning_rate": 1.1112761767687053e-06,
|
| 3271 |
+
"loss": 0.34855678081512453,
|
| 3272 |
+
"step": 4660
|
| 3273 |
+
},
|
| 3274 |
+
{
|
| 3275 |
+
"epoch": 1.9124603255861574,
|
| 3276 |
+
"grad_norm": 0.2201295793056488,
|
| 3277 |
+
"learning_rate": 1.0148549002536477e-06,
|
| 3278 |
+
"loss": 0.40614027976989747,
|
| 3279 |
+
"step": 4670
|
| 3280 |
+
},
|
| 3281 |
+
{
|
| 3282 |
+
"epoch": 1.9165557489505478,
|
| 3283 |
+
"grad_norm": 0.18508629500865936,
|
| 3284 |
+
"learning_rate": 9.227873528936725e-07,
|
| 3285 |
+
"loss": 0.35499498844146726,
|
| 3286 |
+
"step": 4680
|
| 3287 |
+
},
|
| 3288 |
+
{
|
| 3289 |
+
"epoch": 1.9206511723149382,
|
| 3290 |
+
"grad_norm": 0.23364585638046265,
|
| 3291 |
+
"learning_rate": 8.350775841566627e-07,
|
| 3292 |
+
"loss": 0.40017199516296387,
|
| 3293 |
+
"step": 4690
|
| 3294 |
+
},
|
| 3295 |
+
{
|
| 3296 |
+
"epoch": 1.9247465956793284,
|
| 3297 |
+
"grad_norm": 0.19473063945770264,
|
| 3298 |
+
"learning_rate": 7.517294518394335e-07,
|
| 3299 |
+
"loss": 0.36209428310394287,
|
| 3300 |
+
"step": 4700
|
| 3301 |
+
},
|
| 3302 |
+
{
|
| 3303 |
+
"epoch": 1.9288420190437185,
|
| 3304 |
+
"grad_norm": 0.2109328657388687,
|
| 3305 |
+
"learning_rate": 6.727466218980461e-07,
|
| 3306 |
+
"loss": 0.377748966217041,
|
| 3307 |
+
"step": 4710
|
| 3308 |
+
},
|
| 3309 |
+
{
|
| 3310 |
+
"epoch": 1.932937442408109,
|
| 3311 |
+
"grad_norm": 0.1606670320034027,
|
| 3312 |
+
"learning_rate": 5.981325682865913e-07,
|
| 3313 |
+
"loss": 0.3436415195465088,
|
| 3314 |
+
"step": 4720
|
| 3315 |
+
},
|
| 3316 |
+
{
|
| 3317 |
+
"epoch": 1.9370328657724993,
|
| 3318 |
+
"grad_norm": 0.188730388879776,
|
| 3319 |
+
"learning_rate": 5.278905728043793e-07,
|
| 3320 |
+
"loss": 0.40039310455322263,
|
| 3321 |
+
"step": 4730
|
| 3322 |
+
},
|
| 3323 |
+
{
|
| 3324 |
+
"epoch": 1.9411282891368895,
|
| 3325 |
+
"grad_norm": 0.22509531676769257,
|
| 3326 |
+
"learning_rate": 4.6202372495159907e-07,
|
| 3327 |
+
"loss": 0.37746610641479494,
|
| 3328 |
+
"step": 4740
|
| 3329 |
+
},
|
| 3330 |
+
{
|
| 3331 |
+
"epoch": 1.9452237125012797,
|
| 3332 |
+
"grad_norm": 0.22731053829193115,
|
| 3333 |
+
"learning_rate": 4.0053492179340513e-07,
|
| 3334 |
+
"loss": 0.3691138982772827,
|
| 3335 |
+
"step": 4750
|
| 3336 |
+
},
|
| 3337 |
+
{
|
| 3338 |
+
"epoch": 1.94931913586567,
|
| 3339 |
+
"grad_norm": 0.2142152190208435,
|
| 3340 |
+
"learning_rate": 3.434268678325192e-07,
|
| 3341 |
+
"loss": 0.3509155035018921,
|
| 3342 |
+
"step": 4760
|
| 3343 |
+
},
|
| 3344 |
+
{
|
| 3345 |
+
"epoch": 1.9534145592300605,
|
| 3346 |
+
"grad_norm": 0.2099379301071167,
|
| 3347 |
+
"learning_rate": 2.9070207489030334e-07,
|
| 3348 |
+
"loss": 0.38893296718597414,
|
| 3349 |
+
"step": 4770
|
| 3350 |
+
},
|
| 3351 |
+
{
|
| 3352 |
+
"epoch": 1.9575099825944506,
|
| 3353 |
+
"grad_norm": 0.19143544137477875,
|
| 3354 |
+
"learning_rate": 2.423628619961926e-07,
|
| 3355 |
+
"loss": 0.34151449203491213,
|
| 3356 |
+
"step": 4780
|
| 3357 |
+
},
|
| 3358 |
+
{
|
| 3359 |
+
"epoch": 1.961605405958841,
|
| 3360 |
+
"grad_norm": 0.22415593266487122,
|
| 3361 |
+
"learning_rate": 1.9841135528581023e-07,
|
| 3362 |
+
"loss": 0.410687255859375,
|
| 3363 |
+
"step": 4790
|
| 3364 |
+
},
|
| 3365 |
+
{
|
| 3366 |
+
"epoch": 1.9657008293232314,
|
| 3367 |
+
"grad_norm": 0.21912957727909088,
|
| 3368 |
+
"learning_rate": 1.5884948790735322e-07,
|
| 3369 |
+
"loss": 0.4230350017547607,
|
| 3370 |
+
"step": 4800
|
| 3371 |
+
},
|
| 3372 |
+
{
|
| 3373 |
+
"epoch": 1.9697962526876216,
|
| 3374 |
+
"grad_norm": 0.19697561860084534,
|
| 3375 |
+
"learning_rate": 1.2367899993662725e-07,
|
| 3376 |
+
"loss": 0.3369002103805542,
|
| 3377 |
+
"step": 4810
|
| 3378 |
+
},
|
| 3379 |
+
{
|
| 3380 |
+
"epoch": 1.9738916760520118,
|
| 3381 |
+
"grad_norm": 0.17736215889453888,
|
| 3382 |
+
"learning_rate": 9.290143830048559e-08,
|
| 3383 |
+
"loss": 0.38332953453063967,
|
| 3384 |
+
"step": 4820
|
| 3385 |
+
},
|
| 3386 |
+
{
|
| 3387 |
+
"epoch": 1.9779870994164022,
|
| 3388 |
+
"grad_norm": 0.22228537499904633,
|
| 3389 |
+
"learning_rate": 6.651815670877248e-08,
|
| 3390 |
+
"loss": 0.38815762996673586,
|
| 3391 |
+
"step": 4830
|
| 3392 |
+
},
|
| 3393 |
+
{
|
| 3394 |
+
"epoch": 1.9820825227807926,
|
| 3395 |
+
"grad_norm": 0.2782864570617676,
|
| 3396 |
+
"learning_rate": 4.453031559485954e-08,
|
| 3397 |
+
"loss": 0.35874664783477783,
|
| 3398 |
+
"step": 4840
|
| 3399 |
+
},
|
| 3400 |
+
{
|
| 3401 |
+
"epoch": 1.9861779461451827,
|
| 3402 |
+
"grad_norm": 0.2207091599702835,
|
| 3403 |
+
"learning_rate": 2.6938882064486692e-08,
|
| 3404 |
+
"loss": 0.40201964378356936,
|
| 3405 |
+
"step": 4850
|
| 3406 |
+
},
|
| 3407 |
+
{
|
| 3408 |
+
"epoch": 1.990273369509573,
|
| 3409 |
+
"grad_norm": 0.270383358001709,
|
| 3410 |
+
"learning_rate": 1.3744629853329472e-08,
|
| 3411 |
+
"loss": 0.38563265800476076,
|
| 3412 |
+
"step": 4860
|
| 3413 |
+
},
|
| 3414 |
+
{
|
| 3415 |
+
"epoch": 1.9943687928739633,
|
| 3416 |
+
"grad_norm": 0.1851816177368164,
|
| 3417 |
+
"learning_rate": 4.948139292926257e-09,
|
| 3418 |
+
"loss": 0.35580291748046877,
|
| 3419 |
+
"step": 4870
|
| 3420 |
+
},
|
| 3421 |
+
{
|
| 3422 |
+
"epoch": 1.9984642162383537,
|
| 3423 |
+
"grad_norm": 0.1871575564146042,
|
| 3424 |
+
"learning_rate": 5.497972851209454e-10,
|
| 3425 |
+
"loss": 0.3794358253479004,
|
| 3426 |
+
"step": 4880
|
| 3427 |
}
|
| 3428 |
],
|
| 3429 |
"logging_steps": 10,
|
|
|
|
| 3438 |
"should_evaluate": false,
|
| 3439 |
"should_log": false,
|
| 3440 |
"should_save": true,
|
| 3441 |
+
"should_training_stop": true
|
| 3442 |
},
|
| 3443 |
"attributes": {}
|
| 3444 |
}
|
| 3445 |
},
|
| 3446 |
+
"total_flos": 8.475349355694029e+17,
|
| 3447 |
"train_batch_size": 4,
|
| 3448 |
"trial_name": null,
|
| 3449 |
"trial_params": null
|