|
{
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 25.6,
|
|
"eval_steps": 500,
|
|
"global_step": 800,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 0.16,
|
|
"grad_norm": 2.2721662521362305,
|
|
"learning_rate": 4.999643406399275e-05,
|
|
"loss": 4.0376,
|
|
"num_input_tokens_seen": 6208,
|
|
"step": 5
|
|
},
|
|
{
|
|
"epoch": 0.32,
|
|
"grad_norm": 2.1794309616088867,
|
|
"learning_rate": 4.998573727324295e-05,
|
|
"loss": 3.9633,
|
|
"num_input_tokens_seen": 11968,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 0.48,
|
|
"grad_norm": 1.3466308116912842,
|
|
"learning_rate": 4.9967912679276316e-05,
|
|
"loss": 3.7175,
|
|
"num_input_tokens_seen": 18016,
|
|
"step": 15
|
|
},
|
|
{
|
|
"epoch": 0.64,
|
|
"grad_norm": 1.3052715063095093,
|
|
"learning_rate": 4.994296536700177e-05,
|
|
"loss": 3.6465,
|
|
"num_input_tokens_seen": 23760,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 0.8,
|
|
"grad_norm": 1.289433479309082,
|
|
"learning_rate": 4.9910902453260824e-05,
|
|
"loss": 3.6384,
|
|
"num_input_tokens_seen": 29648,
|
|
"step": 25
|
|
},
|
|
{
|
|
"epoch": 0.96,
|
|
"grad_norm": 1.0926741361618042,
|
|
"learning_rate": 4.987173308479738e-05,
|
|
"loss": 3.5703,
|
|
"num_input_tokens_seen": 35968,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 1.12,
|
|
"grad_norm": 1.3285070657730103,
|
|
"learning_rate": 4.982546843564834e-05,
|
|
"loss": 3.5096,
|
|
"num_input_tokens_seen": 42112,
|
|
"step": 35
|
|
},
|
|
{
|
|
"epoch": 1.28,
|
|
"grad_norm": 1.691308856010437,
|
|
"learning_rate": 4.977212170395598e-05,
|
|
"loss": 3.2806,
|
|
"num_input_tokens_seen": 48352,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 1.44,
|
|
"grad_norm": 1.3265893459320068,
|
|
"learning_rate": 4.971170810820279e-05,
|
|
"loss": 3.4431,
|
|
"num_input_tokens_seen": 54144,
|
|
"step": 45
|
|
},
|
|
{
|
|
"epoch": 1.6,
|
|
"grad_norm": 1.2742599248886108,
|
|
"learning_rate": 4.964424488287009e-05,
|
|
"loss": 3.2597,
|
|
"num_input_tokens_seen": 60224,
|
|
"step": 50
|
|
},
|
|
{
|
|
"epoch": 1.76,
|
|
"grad_norm": 1.2702339887619019,
|
|
"learning_rate": 4.9569751273521454e-05,
|
|
"loss": 3.275,
|
|
"num_input_tokens_seen": 66512,
|
|
"step": 55
|
|
},
|
|
{
|
|
"epoch": 1.92,
|
|
"grad_norm": 1.6329398155212402,
|
|
"learning_rate": 4.948824853131236e-05,
|
|
"loss": 3.1563,
|
|
"num_input_tokens_seen": 72208,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 2.08,
|
|
"grad_norm": 1.348823070526123,
|
|
"learning_rate": 4.939975990692789e-05,
|
|
"loss": 3.1883,
|
|
"num_input_tokens_seen": 78368,
|
|
"step": 65
|
|
},
|
|
{
|
|
"epoch": 2.24,
|
|
"grad_norm": 1.441171646118164,
|
|
"learning_rate": 4.930431064394977e-05,
|
|
"loss": 3.2454,
|
|
"num_input_tokens_seen": 84288,
|
|
"step": 70
|
|
},
|
|
{
|
|
"epoch": 2.4,
|
|
"grad_norm": 1.3091288805007935,
|
|
"learning_rate": 4.920192797165511e-05,
|
|
"loss": 3.1905,
|
|
"num_input_tokens_seen": 90464,
|
|
"step": 75
|
|
},
|
|
{
|
|
"epoch": 2.56,
|
|
"grad_norm": 1.6830319166183472,
|
|
"learning_rate": 4.909264109724853e-05,
|
|
"loss": 3.0087,
|
|
"num_input_tokens_seen": 96704,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 2.7199999999999998,
|
|
"grad_norm": 1.941502332687378,
|
|
"learning_rate": 4.897648119753006e-05,
|
|
"loss": 3.0016,
|
|
"num_input_tokens_seen": 102352,
|
|
"step": 85
|
|
},
|
|
{
|
|
"epoch": 2.88,
|
|
"grad_norm": 1.8244129419326782,
|
|
"learning_rate": 4.885348141000122e-05,
|
|
"loss": 3.1813,
|
|
"num_input_tokens_seen": 108112,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 3.04,
|
|
"grad_norm": 1.6675082445144653,
|
|
"learning_rate": 4.872367682341173e-05,
|
|
"loss": 3.1158,
|
|
"num_input_tokens_seen": 114240,
|
|
"step": 95
|
|
},
|
|
{
|
|
"epoch": 3.2,
|
|
"grad_norm": 1.8476835489273071,
|
|
"learning_rate": 4.858710446774951e-05,
|
|
"loss": 2.9404,
|
|
"num_input_tokens_seen": 119936,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 3.36,
|
|
"grad_norm": 1.8608956336975098,
|
|
"learning_rate": 4.844380330367701e-05,
|
|
"loss": 2.9749,
|
|
"num_input_tokens_seen": 125984,
|
|
"step": 105
|
|
},
|
|
{
|
|
"epoch": 3.52,
|
|
"grad_norm": 1.9425894021987915,
|
|
"learning_rate": 4.829381421141671e-05,
|
|
"loss": 2.9456,
|
|
"num_input_tokens_seen": 131808,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 3.68,
|
|
"grad_norm": 2.114993095397949,
|
|
"learning_rate": 4.8137179979088995e-05,
|
|
"loss": 2.8505,
|
|
"num_input_tokens_seen": 137792,
|
|
"step": 115
|
|
},
|
|
{
|
|
"epoch": 3.84,
|
|
"grad_norm": 1.7963783740997314,
|
|
"learning_rate": 4.7973945290505766e-05,
|
|
"loss": 3.0044,
|
|
"num_input_tokens_seen": 144336,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 4.0,
|
|
"grad_norm": 1.7597969770431519,
|
|
"learning_rate": 4.780415671242334e-05,
|
|
"loss": 3.0709,
|
|
"num_input_tokens_seen": 150336,
|
|
"step": 125
|
|
},
|
|
{
|
|
"epoch": 4.16,
|
|
"grad_norm": 1.8621456623077393,
|
|
"learning_rate": 4.7627862681258037e-05,
|
|
"loss": 2.912,
|
|
"num_input_tokens_seen": 156768,
|
|
"step": 130
|
|
},
|
|
{
|
|
"epoch": 4.32,
|
|
"grad_norm": 2.021226167678833,
|
|
"learning_rate": 4.7445113489268544e-05,
|
|
"loss": 2.8063,
|
|
"num_input_tokens_seen": 163232,
|
|
"step": 135
|
|
},
|
|
{
|
|
"epoch": 4.48,
|
|
"grad_norm": 2.2448067665100098,
|
|
"learning_rate": 4.725596127020879e-05,
|
|
"loss": 2.7714,
|
|
"num_input_tokens_seen": 169616,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 4.64,
|
|
"grad_norm": 5.535823822021484,
|
|
"learning_rate": 4.706045998445548e-05,
|
|
"loss": 2.8277,
|
|
"num_input_tokens_seen": 175664,
|
|
"step": 145
|
|
},
|
|
{
|
|
"epoch": 4.8,
|
|
"grad_norm": 2.4096429347991943,
|
|
"learning_rate": 4.685866540361456e-05,
|
|
"loss": 2.7456,
|
|
"num_input_tokens_seen": 181232,
|
|
"step": 150
|
|
},
|
|
{
|
|
"epoch": 4.96,
|
|
"grad_norm": 2.699846029281616,
|
|
"learning_rate": 4.665063509461097e-05,
|
|
"loss": 2.7187,
|
|
"num_input_tokens_seen": 186960,
|
|
"step": 155
|
|
},
|
|
{
|
|
"epoch": 5.12,
|
|
"grad_norm": 2.661548376083374,
|
|
"learning_rate": 4.643642840326627e-05,
|
|
"loss": 2.7918,
|
|
"num_input_tokens_seen": 192640,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 5.28,
|
|
"grad_norm": 2.395092725753784,
|
|
"learning_rate": 4.621610643736878e-05,
|
|
"loss": 2.6223,
|
|
"num_input_tokens_seen": 198672,
|
|
"step": 165
|
|
},
|
|
{
|
|
"epoch": 5.44,
|
|
"grad_norm": 2.329503059387207,
|
|
"learning_rate": 4.598973204924097e-05,
|
|
"loss": 2.654,
|
|
"num_input_tokens_seen": 204976,
|
|
"step": 170
|
|
},
|
|
{
|
|
"epoch": 5.6,
|
|
"grad_norm": 2.3466107845306396,
|
|
"learning_rate": 4.5757369817809415e-05,
|
|
"loss": 2.7321,
|
|
"num_input_tokens_seen": 211168,
|
|
"step": 175
|
|
},
|
|
{
|
|
"epoch": 5.76,
|
|
"grad_norm": 2.9179327487945557,
|
|
"learning_rate": 4.551908603018191e-05,
|
|
"loss": 2.71,
|
|
"num_input_tokens_seen": 217072,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 5.92,
|
|
"grad_norm": 2.7011966705322266,
|
|
"learning_rate": 4.527494866273753e-05,
|
|
"loss": 2.7663,
|
|
"num_input_tokens_seen": 223136,
|
|
"step": 185
|
|
},
|
|
{
|
|
"epoch": 6.08,
|
|
"grad_norm": 2.610811948776245,
|
|
"learning_rate": 4.502502736173462e-05,
|
|
"loss": 2.4595,
|
|
"num_input_tokens_seen": 229088,
|
|
"step": 190
|
|
},
|
|
{
|
|
"epoch": 6.24,
|
|
"grad_norm": 2.6016838550567627,
|
|
"learning_rate": 4.476939342344246e-05,
|
|
"loss": 2.5734,
|
|
"num_input_tokens_seen": 235184,
|
|
"step": 195
|
|
},
|
|
{
|
|
"epoch": 6.4,
|
|
"grad_norm": 3.063948392868042,
|
|
"learning_rate": 4.45081197738023e-05,
|
|
"loss": 2.3804,
|
|
"num_input_tokens_seen": 241152,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 6.5600000000000005,
|
|
"grad_norm": 3.9621613025665283,
|
|
"learning_rate": 4.424128094762331e-05,
|
|
"loss": 2.4395,
|
|
"num_input_tokens_seen": 247136,
|
|
"step": 205
|
|
},
|
|
{
|
|
"epoch": 6.72,
|
|
"grad_norm": 2.887089967727661,
|
|
"learning_rate": 4.3968953067319777e-05,
|
|
"loss": 2.3477,
|
|
"num_input_tokens_seen": 253232,
|
|
"step": 210
|
|
},
|
|
{
|
|
"epoch": 6.88,
|
|
"grad_norm": 2.8491384983062744,
|
|
"learning_rate": 4.369121382119523e-05,
|
|
"loss": 2.5377,
|
|
"num_input_tokens_seen": 258992,
|
|
"step": 215
|
|
},
|
|
{
|
|
"epoch": 7.04,
|
|
"grad_norm": 3.3222053050994873,
|
|
"learning_rate": 4.340814244127993e-05,
|
|
"loss": 2.5576,
|
|
"num_input_tokens_seen": 265392,
|
|
"step": 220
|
|
},
|
|
{
|
|
"epoch": 7.2,
|
|
"grad_norm": 3.4029548168182373,
|
|
"learning_rate": 4.3119819680728e-05,
|
|
"loss": 2.2721,
|
|
"num_input_tokens_seen": 271344,
|
|
"step": 225
|
|
},
|
|
{
|
|
"epoch": 7.36,
|
|
"grad_norm": 3.4295144081115723,
|
|
"learning_rate": 4.282632779078051e-05,
|
|
"loss": 2.4025,
|
|
"num_input_tokens_seen": 277120,
|
|
"step": 230
|
|
},
|
|
{
|
|
"epoch": 7.52,
|
|
"grad_norm": 3.2438597679138184,
|
|
"learning_rate": 4.2527750497301323e-05,
|
|
"loss": 2.2175,
|
|
"num_input_tokens_seen": 283152,
|
|
"step": 235
|
|
},
|
|
{
|
|
"epoch": 7.68,
|
|
"grad_norm": 3.3982954025268555,
|
|
"learning_rate": 4.222417297689217e-05,
|
|
"loss": 2.1803,
|
|
"num_input_tokens_seen": 289136,
|
|
"step": 240
|
|
},
|
|
{
|
|
"epoch": 7.84,
|
|
"grad_norm": 2.70150089263916,
|
|
"learning_rate": 4.191568183259394e-05,
|
|
"loss": 2.4197,
|
|
"num_input_tokens_seen": 295968,
|
|
"step": 245
|
|
},
|
|
{
|
|
"epoch": 8.0,
|
|
"grad_norm": 3.426187515258789,
|
|
"learning_rate": 4.160236506918098e-05,
|
|
"loss": 2.2351,
|
|
"num_input_tokens_seen": 301472,
|
|
"step": 250
|
|
},
|
|
{
|
|
"epoch": 8.16,
|
|
"grad_norm": 4.023019313812256,
|
|
"learning_rate": 4.128431206805557e-05,
|
|
"loss": 2.1748,
|
|
"num_input_tokens_seen": 307216,
|
|
"step": 255
|
|
},
|
|
{
|
|
"epoch": 8.32,
|
|
"grad_norm": 3.1614391803741455,
|
|
"learning_rate": 4.096161356174959e-05,
|
|
"loss": 2.1173,
|
|
"num_input_tokens_seen": 313424,
|
|
"step": 260
|
|
},
|
|
{
|
|
"epoch": 8.48,
|
|
"grad_norm": 3.4006361961364746,
|
|
"learning_rate": 4.063436160804092e-05,
|
|
"loss": 2.0966,
|
|
"num_input_tokens_seen": 319472,
|
|
"step": 265
|
|
},
|
|
{
|
|
"epoch": 8.64,
|
|
"grad_norm": 4.167815208435059,
|
|
"learning_rate": 4.030264956369157e-05,
|
|
"loss": 2.0825,
|
|
"num_input_tokens_seen": 325504,
|
|
"step": 270
|
|
},
|
|
{
|
|
"epoch": 8.8,
|
|
"grad_norm": 4.040388584136963,
|
|
"learning_rate": 3.9966572057815373e-05,
|
|
"loss": 2.0189,
|
|
"num_input_tokens_seen": 331520,
|
|
"step": 275
|
|
},
|
|
{
|
|
"epoch": 8.96,
|
|
"grad_norm": 5.056305885314941,
|
|
"learning_rate": 3.962622496488269e-05,
|
|
"loss": 2.0671,
|
|
"num_input_tokens_seen": 337680,
|
|
"step": 280
|
|
},
|
|
{
|
|
"epoch": 9.12,
|
|
"grad_norm": 4.466280937194824,
|
|
"learning_rate": 3.928170537736981e-05,
|
|
"loss": 2.0337,
|
|
"num_input_tokens_seen": 343936,
|
|
"step": 285
|
|
},
|
|
{
|
|
"epoch": 9.28,
|
|
"grad_norm": 4.8046698570251465,
|
|
"learning_rate": 3.893311157806091e-05,
|
|
"loss": 1.7892,
|
|
"num_input_tokens_seen": 349632,
|
|
"step": 290
|
|
},
|
|
{
|
|
"epoch": 9.44,
|
|
"grad_norm": 5.066511631011963,
|
|
"learning_rate": 3.858054301201047e-05,
|
|
"loss": 1.8558,
|
|
"num_input_tokens_seen": 356080,
|
|
"step": 295
|
|
},
|
|
{
|
|
"epoch": 9.6,
|
|
"grad_norm": 4.678648948669434,
|
|
"learning_rate": 3.822410025817406e-05,
|
|
"loss": 1.9685,
|
|
"num_input_tokens_seen": 362048,
|
|
"step": 300
|
|
},
|
|
{
|
|
"epoch": 9.76,
|
|
"grad_norm": 4.883659839630127,
|
|
"learning_rate": 3.786388500071572e-05,
|
|
"loss": 1.6883,
|
|
"num_input_tokens_seen": 368128,
|
|
"step": 305
|
|
},
|
|
{
|
|
"epoch": 9.92,
|
|
"grad_norm": 4.799614429473877,
|
|
"learning_rate": 3.7500000000000003e-05,
|
|
"loss": 1.8049,
|
|
"num_input_tokens_seen": 373920,
|
|
"step": 310
|
|
},
|
|
{
|
|
"epoch": 10.08,
|
|
"grad_norm": 4.910759925842285,
|
|
"learning_rate": 3.713254906327703e-05,
|
|
"loss": 1.8785,
|
|
"num_input_tokens_seen": 379920,
|
|
"step": 315
|
|
},
|
|
{
|
|
"epoch": 10.24,
|
|
"grad_norm": 6.2078447341918945,
|
|
"learning_rate": 3.67616370150689e-05,
|
|
"loss": 1.8797,
|
|
"num_input_tokens_seen": 386464,
|
|
"step": 320
|
|
},
|
|
{
|
|
"epoch": 10.4,
|
|
"grad_norm": 6.023489952087402,
|
|
"learning_rate": 3.638736966726585e-05,
|
|
"loss": 1.6139,
|
|
"num_input_tokens_seen": 392608,
|
|
"step": 325
|
|
},
|
|
{
|
|
"epoch": 10.56,
|
|
"grad_norm": 5.2762017250061035,
|
|
"learning_rate": 3.600985378894086e-05,
|
|
"loss": 1.4428,
|
|
"num_input_tokens_seen": 398224,
|
|
"step": 330
|
|
},
|
|
{
|
|
"epoch": 10.72,
|
|
"grad_norm": 6.009916305541992,
|
|
"learning_rate": 3.562919707589102e-05,
|
|
"loss": 1.6858,
|
|
"num_input_tokens_seen": 404480,
|
|
"step": 335
|
|
},
|
|
{
|
|
"epoch": 10.88,
|
|
"grad_norm": 5.862870216369629,
|
|
"learning_rate": 3.5245508119914687e-05,
|
|
"loss": 1.5669,
|
|
"num_input_tokens_seen": 410448,
|
|
"step": 340
|
|
},
|
|
{
|
|
"epoch": 11.04,
|
|
"grad_norm": 5.693241119384766,
|
|
"learning_rate": 3.4858896377832966e-05,
|
|
"loss": 1.5857,
|
|
"num_input_tokens_seen": 416544,
|
|
"step": 345
|
|
},
|
|
{
|
|
"epoch": 11.2,
|
|
"grad_norm": 7.012350082397461,
|
|
"learning_rate": 3.44694721402644e-05,
|
|
"loss": 1.4511,
|
|
"num_input_tokens_seen": 422032,
|
|
"step": 350
|
|
},
|
|
{
|
|
"epoch": 11.36,
|
|
"grad_norm": 6.417391777038574,
|
|
"learning_rate": 3.407734650016187e-05,
|
|
"loss": 1.4448,
|
|
"num_input_tokens_seen": 428224,
|
|
"step": 355
|
|
},
|
|
{
|
|
"epoch": 11.52,
|
|
"grad_norm": 5.746353626251221,
|
|
"learning_rate": 3.3682631321120504e-05,
|
|
"loss": 1.4648,
|
|
"num_input_tokens_seen": 434368,
|
|
"step": 360
|
|
},
|
|
{
|
|
"epoch": 11.68,
|
|
"grad_norm": 6.427619457244873,
|
|
"learning_rate": 3.32854392054659e-05,
|
|
"loss": 1.4896,
|
|
"num_input_tokens_seen": 440816,
|
|
"step": 365
|
|
},
|
|
{
|
|
"epoch": 11.84,
|
|
"grad_norm": 6.434299468994141,
|
|
"learning_rate": 3.2885883462131394e-05,
|
|
"loss": 1.2713,
|
|
"num_input_tokens_seen": 446992,
|
|
"step": 370
|
|
},
|
|
{
|
|
"epoch": 12.0,
|
|
"grad_norm": 6.6553521156311035,
|
|
"learning_rate": 3.2484078074333954e-05,
|
|
"loss": 1.4016,
|
|
"num_input_tokens_seen": 452864,
|
|
"step": 375
|
|
},
|
|
{
|
|
"epoch": 12.16,
|
|
"grad_norm": 6.007937431335449,
|
|
"learning_rate": 3.2080137667057595e-05,
|
|
"loss": 1.1953,
|
|
"num_input_tokens_seen": 458800,
|
|
"step": 380
|
|
},
|
|
{
|
|
"epoch": 12.32,
|
|
"grad_norm": 7.414127826690674,
|
|
"learning_rate": 3.167417747435379e-05,
|
|
"loss": 1.235,
|
|
"num_input_tokens_seen": 465040,
|
|
"step": 385
|
|
},
|
|
{
|
|
"epoch": 12.48,
|
|
"grad_norm": 7.6087565422058105,
|
|
"learning_rate": 3.126631330646802e-05,
|
|
"loss": 1.1375,
|
|
"num_input_tokens_seen": 470544,
|
|
"step": 390
|
|
},
|
|
{
|
|
"epoch": 12.64,
|
|
"grad_norm": 7.889481067657471,
|
|
"learning_rate": 3.0856661516802054e-05,
|
|
"loss": 1.1427,
|
|
"num_input_tokens_seen": 476608,
|
|
"step": 395
|
|
},
|
|
{
|
|
"epoch": 12.8,
|
|
"grad_norm": 5.883874416351318,
|
|
"learning_rate": 3.0445338968721287e-05,
|
|
"loss": 1.3749,
|
|
"num_input_tokens_seen": 483136,
|
|
"step": 400
|
|
},
|
|
{
|
|
"epoch": 12.96,
|
|
"grad_norm": 5.944591045379639,
|
|
"learning_rate": 3.0032463002216505e-05,
|
|
"loss": 1.2318,
|
|
"num_input_tokens_seen": 488752,
|
|
"step": 405
|
|
},
|
|
{
|
|
"epoch": 13.12,
|
|
"grad_norm": 5.7361931800842285,
|
|
"learning_rate": 2.961815140042974e-05,
|
|
"loss": 1.0918,
|
|
"num_input_tokens_seen": 495040,
|
|
"step": 410
|
|
},
|
|
{
|
|
"epoch": 13.28,
|
|
"grad_norm": 5.651491165161133,
|
|
"learning_rate": 2.920252235605371e-05,
|
|
"loss": 1.0229,
|
|
"num_input_tokens_seen": 500816,
|
|
"step": 415
|
|
},
|
|
{
|
|
"epoch": 13.44,
|
|
"grad_norm": 7.200684547424316,
|
|
"learning_rate": 2.878569443761442e-05,
|
|
"loss": 0.9925,
|
|
"num_input_tokens_seen": 506336,
|
|
"step": 420
|
|
},
|
|
{
|
|
"epoch": 13.6,
|
|
"grad_norm": 5.146182060241699,
|
|
"learning_rate": 2.836778655564653e-05,
|
|
"loss": 1.05,
|
|
"num_input_tokens_seen": 512848,
|
|
"step": 425
|
|
},
|
|
{
|
|
"epoch": 13.76,
|
|
"grad_norm": 5.351944923400879,
|
|
"learning_rate": 2.7948917928771158e-05,
|
|
"loss": 0.9813,
|
|
"num_input_tokens_seen": 519024,
|
|
"step": 430
|
|
},
|
|
{
|
|
"epoch": 13.92,
|
|
"grad_norm": 6.804362773895264,
|
|
"learning_rate": 2.7529208049685807e-05,
|
|
"loss": 1.1116,
|
|
"num_input_tokens_seen": 525536,
|
|
"step": 435
|
|
},
|
|
{
|
|
"epoch": 14.08,
|
|
"grad_norm": 5.708404064178467,
|
|
"learning_rate": 2.7108776651076118e-05,
|
|
"loss": 0.9063,
|
|
"num_input_tokens_seen": 531200,
|
|
"step": 440
|
|
},
|
|
{
|
|
"epoch": 14.24,
|
|
"grad_norm": 5.502604007720947,
|
|
"learning_rate": 2.668774367145913e-05,
|
|
"loss": 0.8812,
|
|
"num_input_tokens_seen": 537232,
|
|
"step": 445
|
|
},
|
|
{
|
|
"epoch": 14.4,
|
|
"grad_norm": 5.171998500823975,
|
|
"learning_rate": 2.6266229220967818e-05,
|
|
"loss": 0.7509,
|
|
"num_input_tokens_seen": 543120,
|
|
"step": 450
|
|
},
|
|
{
|
|
"epoch": 14.56,
|
|
"grad_norm": 5.874445915222168,
|
|
"learning_rate": 2.584435354708671e-05,
|
|
"loss": 1.0827,
|
|
"num_input_tokens_seen": 549312,
|
|
"step": 455
|
|
},
|
|
{
|
|
"epoch": 14.72,
|
|
"grad_norm": 6.16728401184082,
|
|
"learning_rate": 2.5422237000348276e-05,
|
|
"loss": 0.8113,
|
|
"num_input_tokens_seen": 555408,
|
|
"step": 460
|
|
},
|
|
{
|
|
"epoch": 14.88,
|
|
"grad_norm": 7.150355339050293,
|
|
"learning_rate": 2.5e-05,
|
|
"loss": 0.8095,
|
|
"num_input_tokens_seen": 561648,
|
|
"step": 465
|
|
},
|
|
{
|
|
"epoch": 15.04,
|
|
"grad_norm": 4.866459369659424,
|
|
"learning_rate": 2.4577762999651726e-05,
|
|
"loss": 0.9768,
|
|
"num_input_tokens_seen": 568048,
|
|
"step": 470
|
|
},
|
|
{
|
|
"epoch": 15.2,
|
|
"grad_norm": 5.219202995300293,
|
|
"learning_rate": 2.4155646452913296e-05,
|
|
"loss": 0.8017,
|
|
"num_input_tokens_seen": 574352,
|
|
"step": 475
|
|
},
|
|
{
|
|
"epoch": 15.36,
|
|
"grad_norm": 5.405744552612305,
|
|
"learning_rate": 2.3733770779032184e-05,
|
|
"loss": 0.8299,
|
|
"num_input_tokens_seen": 580160,
|
|
"step": 480
|
|
},
|
|
{
|
|
"epoch": 15.52,
|
|
"grad_norm": 6.140029430389404,
|
|
"learning_rate": 2.331225632854087e-05,
|
|
"loss": 0.581,
|
|
"num_input_tokens_seen": 585952,
|
|
"step": 485
|
|
},
|
|
{
|
|
"epoch": 15.68,
|
|
"grad_norm": 5.105503559112549,
|
|
"learning_rate": 2.2891223348923884e-05,
|
|
"loss": 0.6687,
|
|
"num_input_tokens_seen": 591968,
|
|
"step": 490
|
|
},
|
|
{
|
|
"epoch": 15.84,
|
|
"grad_norm": 6.098103046417236,
|
|
"learning_rate": 2.24707919503142e-05,
|
|
"loss": 0.8088,
|
|
"num_input_tokens_seen": 598128,
|
|
"step": 495
|
|
},
|
|
{
|
|
"epoch": 16.0,
|
|
"grad_norm": 6.3401780128479,
|
|
"learning_rate": 2.2051082071228854e-05,
|
|
"loss": 0.8223,
|
|
"num_input_tokens_seen": 604192,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 16.16,
|
|
"grad_norm": 5.5078630447387695,
|
|
"learning_rate": 2.1632213444353482e-05,
|
|
"loss": 0.5949,
|
|
"num_input_tokens_seen": 609872,
|
|
"step": 505
|
|
},
|
|
{
|
|
"epoch": 16.32,
|
|
"grad_norm": 5.564432621002197,
|
|
"learning_rate": 2.1214305562385592e-05,
|
|
"loss": 0.705,
|
|
"num_input_tokens_seen": 615584,
|
|
"step": 510
|
|
},
|
|
{
|
|
"epoch": 16.48,
|
|
"grad_norm": 4.540548801422119,
|
|
"learning_rate": 2.07974776439463e-05,
|
|
"loss": 0.3912,
|
|
"num_input_tokens_seen": 621536,
|
|
"step": 515
|
|
},
|
|
{
|
|
"epoch": 16.64,
|
|
"grad_norm": 5.815915107727051,
|
|
"learning_rate": 2.0381848599570276e-05,
|
|
"loss": 0.6312,
|
|
"num_input_tokens_seen": 627616,
|
|
"step": 520
|
|
},
|
|
{
|
|
"epoch": 16.8,
|
|
"grad_norm": 5.4646501541137695,
|
|
"learning_rate": 1.9967536997783494e-05,
|
|
"loss": 0.8473,
|
|
"num_input_tokens_seen": 633952,
|
|
"step": 525
|
|
},
|
|
{
|
|
"epoch": 16.96,
|
|
"grad_norm": 6.268056869506836,
|
|
"learning_rate": 1.9554661031278712e-05,
|
|
"loss": 0.7956,
|
|
"num_input_tokens_seen": 640720,
|
|
"step": 530
|
|
},
|
|
{
|
|
"epoch": 17.12,
|
|
"grad_norm": 5.364756107330322,
|
|
"learning_rate": 1.914333848319795e-05,
|
|
"loss": 0.6384,
|
|
"num_input_tokens_seen": 646976,
|
|
"step": 535
|
|
},
|
|
{
|
|
"epoch": 17.28,
|
|
"grad_norm": 5.4850263595581055,
|
|
"learning_rate": 1.8733686693531985e-05,
|
|
"loss": 0.4768,
|
|
"num_input_tokens_seen": 653296,
|
|
"step": 540
|
|
},
|
|
{
|
|
"epoch": 17.44,
|
|
"grad_norm": 5.440878391265869,
|
|
"learning_rate": 1.8325822525646208e-05,
|
|
"loss": 0.6033,
|
|
"num_input_tokens_seen": 659232,
|
|
"step": 545
|
|
},
|
|
{
|
|
"epoch": 17.6,
|
|
"grad_norm": 7.1814374923706055,
|
|
"learning_rate": 1.79198623329424e-05,
|
|
"loss": 0.549,
|
|
"num_input_tokens_seen": 665248,
|
|
"step": 550
|
|
},
|
|
{
|
|
"epoch": 17.76,
|
|
"grad_norm": 4.951470851898193,
|
|
"learning_rate": 1.7515921925666052e-05,
|
|
"loss": 0.5032,
|
|
"num_input_tokens_seen": 671104,
|
|
"step": 555
|
|
},
|
|
{
|
|
"epoch": 17.92,
|
|
"grad_norm": 4.948657989501953,
|
|
"learning_rate": 1.711411653786861e-05,
|
|
"loss": 0.524,
|
|
"num_input_tokens_seen": 676720,
|
|
"step": 560
|
|
},
|
|
{
|
|
"epoch": 18.08,
|
|
"grad_norm": 4.3077898025512695,
|
|
"learning_rate": 1.6714560794534108e-05,
|
|
"loss": 0.5206,
|
|
"num_input_tokens_seen": 683104,
|
|
"step": 565
|
|
},
|
|
{
|
|
"epoch": 18.24,
|
|
"grad_norm": 4.8670220375061035,
|
|
"learning_rate": 1.6317368678879495e-05,
|
|
"loss": 0.4662,
|
|
"num_input_tokens_seen": 688928,
|
|
"step": 570
|
|
},
|
|
{
|
|
"epoch": 18.4,
|
|
"grad_norm": 5.523406028747559,
|
|
"learning_rate": 1.5922653499838137e-05,
|
|
"loss": 0.5093,
|
|
"num_input_tokens_seen": 695072,
|
|
"step": 575
|
|
},
|
|
{
|
|
"epoch": 18.56,
|
|
"grad_norm": 4.92819356918335,
|
|
"learning_rate": 1.55305278597356e-05,
|
|
"loss": 0.5162,
|
|
"num_input_tokens_seen": 701248,
|
|
"step": 580
|
|
},
|
|
{
|
|
"epoch": 18.72,
|
|
"grad_norm": 4.869935989379883,
|
|
"learning_rate": 1.5141103622167041e-05,
|
|
"loss": 0.5355,
|
|
"num_input_tokens_seen": 707296,
|
|
"step": 585
|
|
},
|
|
{
|
|
"epoch": 18.88,
|
|
"grad_norm": 5.4244818687438965,
|
|
"learning_rate": 1.475449188008532e-05,
|
|
"loss": 0.3836,
|
|
"num_input_tokens_seen": 713120,
|
|
"step": 590
|
|
},
|
|
{
|
|
"epoch": 19.04,
|
|
"grad_norm": 4.891148090362549,
|
|
"learning_rate": 1.437080292410899e-05,
|
|
"loss": 0.519,
|
|
"num_input_tokens_seen": 719664,
|
|
"step": 595
|
|
},
|
|
{
|
|
"epoch": 19.2,
|
|
"grad_norm": 4.736128330230713,
|
|
"learning_rate": 1.399014621105914e-05,
|
|
"loss": 0.4391,
|
|
"num_input_tokens_seen": 726224,
|
|
"step": 600
|
|
},
|
|
{
|
|
"epoch": 19.36,
|
|
"grad_norm": 4.512727737426758,
|
|
"learning_rate": 1.361263033273415e-05,
|
|
"loss": 0.4404,
|
|
"num_input_tokens_seen": 732144,
|
|
"step": 605
|
|
},
|
|
{
|
|
"epoch": 19.52,
|
|
"grad_norm": 4.712576389312744,
|
|
"learning_rate": 1.3238362984931113e-05,
|
|
"loss": 0.3758,
|
|
"num_input_tokens_seen": 737824,
|
|
"step": 610
|
|
},
|
|
{
|
|
"epoch": 19.68,
|
|
"grad_norm": 5.2509379386901855,
|
|
"learning_rate": 1.286745093672298e-05,
|
|
"loss": 0.483,
|
|
"num_input_tokens_seen": 743984,
|
|
"step": 615
|
|
},
|
|
{
|
|
"epoch": 19.84,
|
|
"grad_norm": 5.451502323150635,
|
|
"learning_rate": 1.2500000000000006e-05,
|
|
"loss": 0.4637,
|
|
"num_input_tokens_seen": 750016,
|
|
"step": 620
|
|
},
|
|
{
|
|
"epoch": 20.0,
|
|
"grad_norm": 5.625941753387451,
|
|
"learning_rate": 1.2136114999284288e-05,
|
|
"loss": 0.4252,
|
|
"num_input_tokens_seen": 756160,
|
|
"step": 625
|
|
},
|
|
{
|
|
"epoch": 20.16,
|
|
"grad_norm": 4.41165018081665,
|
|
"learning_rate": 1.1775899741825947e-05,
|
|
"loss": 0.3591,
|
|
"num_input_tokens_seen": 762576,
|
|
"step": 630
|
|
},
|
|
{
|
|
"epoch": 20.32,
|
|
"grad_norm": 4.646330833435059,
|
|
"learning_rate": 1.141945698798954e-05,
|
|
"loss": 0.5274,
|
|
"num_input_tokens_seen": 768768,
|
|
"step": 635
|
|
},
|
|
{
|
|
"epoch": 20.48,
|
|
"grad_norm": 3.5506694316864014,
|
|
"learning_rate": 1.1066888421939093e-05,
|
|
"loss": 0.4133,
|
|
"num_input_tokens_seen": 774480,
|
|
"step": 640
|
|
},
|
|
{
|
|
"epoch": 20.64,
|
|
"grad_norm": 4.34029483795166,
|
|
"learning_rate": 1.0718294622630188e-05,
|
|
"loss": 0.274,
|
|
"num_input_tokens_seen": 780368,
|
|
"step": 645
|
|
},
|
|
{
|
|
"epoch": 20.8,
|
|
"grad_norm": 4.8732476234436035,
|
|
"learning_rate": 1.0373775035117305e-05,
|
|
"loss": 0.3506,
|
|
"num_input_tokens_seen": 786368,
|
|
"step": 650
|
|
},
|
|
{
|
|
"epoch": 20.96,
|
|
"grad_norm": 3.6054153442382812,
|
|
"learning_rate": 1.0033427942184622e-05,
|
|
"loss": 0.3099,
|
|
"num_input_tokens_seen": 792304,
|
|
"step": 655
|
|
},
|
|
{
|
|
"epoch": 21.12,
|
|
"grad_norm": 4.9192914962768555,
|
|
"learning_rate": 9.697350436308427e-06,
|
|
"loss": 0.409,
|
|
"num_input_tokens_seen": 798464,
|
|
"step": 660
|
|
},
|
|
{
|
|
"epoch": 21.28,
|
|
"grad_norm": 3.9316163063049316,
|
|
"learning_rate": 9.36563839195908e-06,
|
|
"loss": 0.4555,
|
|
"num_input_tokens_seen": 805152,
|
|
"step": 665
|
|
},
|
|
{
|
|
"epoch": 21.44,
|
|
"grad_norm": 4.707286834716797,
|
|
"learning_rate": 9.038386438250415e-06,
|
|
"loss": 0.3132,
|
|
"num_input_tokens_seen": 810848,
|
|
"step": 670
|
|
},
|
|
{
|
|
"epoch": 21.6,
|
|
"grad_norm": 5.040837287902832,
|
|
"learning_rate": 8.715687931944449e-06,
|
|
"loss": 0.2986,
|
|
"num_input_tokens_seen": 816704,
|
|
"step": 675
|
|
},
|
|
{
|
|
"epoch": 21.76,
|
|
"grad_norm": 5.323684215545654,
|
|
"learning_rate": 8.397634930819021e-06,
|
|
"loss": 0.3246,
|
|
"num_input_tokens_seen": 822272,
|
|
"step": 680
|
|
},
|
|
{
|
|
"epoch": 21.92,
|
|
"grad_norm": 4.811425685882568,
|
|
"learning_rate": 8.084318167406066e-06,
|
|
"loss": 0.349,
|
|
"num_input_tokens_seen": 828256,
|
|
"step": 685
|
|
},
|
|
{
|
|
"epoch": 22.08,
|
|
"grad_norm": 3.185715675354004,
|
|
"learning_rate": 7.775827023107838e-06,
|
|
"loss": 0.2495,
|
|
"num_input_tokens_seen": 834672,
|
|
"step": 690
|
|
},
|
|
{
|
|
"epoch": 22.24,
|
|
"grad_norm": 5.034013748168945,
|
|
"learning_rate": 7.472249502698686e-06,
|
|
"loss": 0.4472,
|
|
"num_input_tokens_seen": 841184,
|
|
"step": 695
|
|
},
|
|
{
|
|
"epoch": 22.4,
|
|
"grad_norm": 3.9689395427703857,
|
|
"learning_rate": 7.173672209219495e-06,
|
|
"loss": 0.2427,
|
|
"num_input_tokens_seen": 847104,
|
|
"step": 700
|
|
},
|
|
{
|
|
"epoch": 22.56,
|
|
"grad_norm": 5.952221870422363,
|
|
"learning_rate": 6.880180319272006e-06,
|
|
"loss": 0.2333,
|
|
"num_input_tokens_seen": 853040,
|
|
"step": 705
|
|
},
|
|
{
|
|
"epoch": 22.72,
|
|
"grad_norm": 5.011849403381348,
|
|
"learning_rate": 6.591857558720071e-06,
|
|
"loss": 0.3539,
|
|
"num_input_tokens_seen": 859040,
|
|
"step": 710
|
|
},
|
|
{
|
|
"epoch": 22.88,
|
|
"grad_norm": 5.248806953430176,
|
|
"learning_rate": 6.308786178804782e-06,
|
|
"loss": 0.3516,
|
|
"num_input_tokens_seen": 865168,
|
|
"step": 715
|
|
},
|
|
{
|
|
"epoch": 23.04,
|
|
"grad_norm": 3.6265501976013184,
|
|
"learning_rate": 6.031046932680226e-06,
|
|
"loss": 0.2574,
|
|
"num_input_tokens_seen": 870624,
|
|
"step": 720
|
|
},
|
|
{
|
|
"epoch": 23.2,
|
|
"grad_norm": 4.028469562530518,
|
|
"learning_rate": 5.758719052376693e-06,
|
|
"loss": 0.2764,
|
|
"num_input_tokens_seen": 876288,
|
|
"step": 725
|
|
},
|
|
{
|
|
"epoch": 23.36,
|
|
"grad_norm": 4.448588848114014,
|
|
"learning_rate": 5.491880226197707e-06,
|
|
"loss": 0.2085,
|
|
"num_input_tokens_seen": 881968,
|
|
"step": 730
|
|
},
|
|
{
|
|
"epoch": 23.52,
|
|
"grad_norm": 4.58131742477417,
|
|
"learning_rate": 5.23060657655754e-06,
|
|
"loss": 0.2651,
|
|
"num_input_tokens_seen": 888496,
|
|
"step": 735
|
|
},
|
|
{
|
|
"epoch": 23.68,
|
|
"grad_norm": 3.284576416015625,
|
|
"learning_rate": 4.9749726382653905e-06,
|
|
"loss": 0.3042,
|
|
"num_input_tokens_seen": 894880,
|
|
"step": 740
|
|
},
|
|
{
|
|
"epoch": 23.84,
|
|
"grad_norm": 4.4178972244262695,
|
|
"learning_rate": 4.725051337262476e-06,
|
|
"loss": 0.334,
|
|
"num_input_tokens_seen": 900624,
|
|
"step": 745
|
|
},
|
|
{
|
|
"epoch": 24.0,
|
|
"grad_norm": 4.631526470184326,
|
|
"learning_rate": 4.480913969818098e-06,
|
|
"loss": 0.382,
|
|
"num_input_tokens_seen": 906880,
|
|
"step": 750
|
|
},
|
|
{
|
|
"epoch": 24.16,
|
|
"grad_norm": 3.961879014968872,
|
|
"learning_rate": 4.242630182190594e-06,
|
|
"loss": 0.2669,
|
|
"num_input_tokens_seen": 913472,
|
|
"step": 755
|
|
},
|
|
{
|
|
"epoch": 24.32,
|
|
"grad_norm": 4.3941779136657715,
|
|
"learning_rate": 4.010267950759025e-06,
|
|
"loss": 0.2149,
|
|
"num_input_tokens_seen": 919360,
|
|
"step": 760
|
|
},
|
|
{
|
|
"epoch": 24.48,
|
|
"grad_norm": 3.2143290042877197,
|
|
"learning_rate": 3.7838935626312242e-06,
|
|
"loss": 0.2215,
|
|
"num_input_tokens_seen": 924928,
|
|
"step": 765
|
|
},
|
|
{
|
|
"epoch": 24.64,
|
|
"grad_norm": 4.793003559112549,
|
|
"learning_rate": 3.5635715967337223e-06,
|
|
"loss": 0.2804,
|
|
"num_input_tokens_seen": 930816,
|
|
"step": 770
|
|
},
|
|
{
|
|
"epoch": 24.8,
|
|
"grad_norm": 3.6798858642578125,
|
|
"learning_rate": 3.3493649053890326e-06,
|
|
"loss": 0.348,
|
|
"num_input_tokens_seen": 937088,
|
|
"step": 775
|
|
},
|
|
{
|
|
"epoch": 24.96,
|
|
"grad_norm": 4.566779136657715,
|
|
"learning_rate": 3.141334596385448e-06,
|
|
"loss": 0.3526,
|
|
"num_input_tokens_seen": 943536,
|
|
"step": 780
|
|
},
|
|
{
|
|
"epoch": 25.12,
|
|
"grad_norm": 3.3694982528686523,
|
|
"learning_rate": 2.939540015544523e-06,
|
|
"loss": 0.1965,
|
|
"num_input_tokens_seen": 949360,
|
|
"step": 785
|
|
},
|
|
{
|
|
"epoch": 25.28,
|
|
"grad_norm": 3.4644641876220703,
|
|
"learning_rate": 2.7440387297912123e-06,
|
|
"loss": 0.2734,
|
|
"num_input_tokens_seen": 955952,
|
|
"step": 790
|
|
},
|
|
{
|
|
"epoch": 25.44,
|
|
"grad_norm": 3.234680652618408,
|
|
"learning_rate": 2.5548865107314607e-06,
|
|
"loss": 0.2059,
|
|
"num_input_tokens_seen": 961696,
|
|
"step": 795
|
|
},
|
|
{
|
|
"epoch": 25.6,
|
|
"grad_norm": 4.125905513763428,
|
|
"learning_rate": 2.372137318741968e-06,
|
|
"loss": 0.1972,
|
|
"num_input_tokens_seen": 967456,
|
|
"step": 800
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 930,
|
|
"num_input_tokens_seen": 967456,
|
|
"num_train_epochs": 30,
|
|
"save_steps": 100,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 7663517602578432.0,
|
|
"train_batch_size": 2,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|
|
|