f4e1c86447
Model: fpadovani/ita-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed455 Source: Original Platform
12101 lines
330 KiB
JSON
12101 lines
330 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 6.048411497730711,
|
|
"eval_steps": 1000,
|
|
"global_step": 6000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.432322406768799,
|
|
"epoch": 0.005042864346949067,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.8944,
|
|
"mean_token_accuracy": 0.08793810680508614,
|
|
"num_tokens": 10258.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.435951519012451,
|
|
"epoch": 0.010085728693898134,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.8302,
|
|
"mean_token_accuracy": 0.08808585703372955,
|
|
"num_tokens": 21166.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.430681848526001,
|
|
"epoch": 0.015128593040847202,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.9437,
|
|
"mean_token_accuracy": 0.08197088204324246,
|
|
"num_tokens": 31390.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.407356119155883,
|
|
"epoch": 0.020171457387796268,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.7616,
|
|
"mean_token_accuracy": 0.09650376290082932,
|
|
"num_tokens": 40983.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.418797492980957,
|
|
"epoch": 0.025214321734745335,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.9135,
|
|
"mean_token_accuracy": 0.09085965603590011,
|
|
"num_tokens": 52673.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.349091672897339,
|
|
"epoch": 0.030257186081694403,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.8145,
|
|
"mean_token_accuracy": 0.08841377198696136,
|
|
"num_tokens": 64956.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.354793548583984,
|
|
"epoch": 0.03530005042864347,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.8421,
|
|
"mean_token_accuracy": 0.08988854065537452,
|
|
"num_tokens": 76492.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.330742835998535,
|
|
"epoch": 0.040342914775592535,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.8362,
|
|
"mean_token_accuracy": 0.08618292436003686,
|
|
"num_tokens": 89191.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.338322544097901,
|
|
"epoch": 0.0453857791225416,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.8356,
|
|
"mean_token_accuracy": 0.09781758561730385,
|
|
"num_tokens": 100077.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.363365173339844,
|
|
"epoch": 0.05042864346949067,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.9068,
|
|
"mean_token_accuracy": 0.08456681296229362,
|
|
"num_tokens": 110530.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.334746885299682,
|
|
"epoch": 0.05547150781643974,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.8925,
|
|
"mean_token_accuracy": 0.08198079392313957,
|
|
"num_tokens": 121113.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.350234079360962,
|
|
"epoch": 0.060514372163388806,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.055,
|
|
"mean_token_accuracy": 0.08448217362165451,
|
|
"num_tokens": 133912.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.348407411575318,
|
|
"epoch": 0.06555723651033787,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.9755,
|
|
"mean_token_accuracy": 0.08814884722232819,
|
|
"num_tokens": 145322.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.38183012008667,
|
|
"epoch": 0.07060010085728693,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.8601,
|
|
"mean_token_accuracy": 0.08610649928450584,
|
|
"num_tokens": 156252.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.35923204421997,
|
|
"epoch": 0.07564296520423601,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.8284,
|
|
"mean_token_accuracy": 0.09050894230604171,
|
|
"num_tokens": 166622.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.273055934906006,
|
|
"epoch": 0.08068582955118507,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.6887,
|
|
"mean_token_accuracy": 0.0989742748439312,
|
|
"num_tokens": 177591.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.254341888427734,
|
|
"epoch": 0.08572869389813415,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.7421,
|
|
"mean_token_accuracy": 0.09370373338460922,
|
|
"num_tokens": 187764.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.251728105545044,
|
|
"epoch": 0.0907715582450832,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.7251,
|
|
"mean_token_accuracy": 0.09580637812614441,
|
|
"num_tokens": 197810.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.275364923477173,
|
|
"epoch": 0.09581442259203228,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.8163,
|
|
"mean_token_accuracy": 0.0883400447666645,
|
|
"num_tokens": 208552.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.275786542892456,
|
|
"epoch": 0.10085728693898134,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.826,
|
|
"mean_token_accuracy": 0.0916892133653164,
|
|
"num_tokens": 221628.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.29000997543335,
|
|
"epoch": 0.1059001512859304,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.7912,
|
|
"mean_token_accuracy": 0.09033632948994637,
|
|
"num_tokens": 231033.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.300166130065918,
|
|
"epoch": 0.11094301563287948,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.8315,
|
|
"mean_token_accuracy": 0.08547956682741642,
|
|
"num_tokens": 242259.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.323574876785278,
|
|
"epoch": 0.11598587997982854,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.8614,
|
|
"mean_token_accuracy": 0.08920874819159508,
|
|
"num_tokens": 254116.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.342547845840454,
|
|
"epoch": 0.12102874432677761,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.8024,
|
|
"mean_token_accuracy": 0.09266475886106491,
|
|
"num_tokens": 264518.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.254070568084717,
|
|
"epoch": 0.1260716086737267,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.6783,
|
|
"mean_token_accuracy": 0.09979371726512909,
|
|
"num_tokens": 273783.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.216696405410767,
|
|
"epoch": 0.13111447302067575,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.8518,
|
|
"mean_token_accuracy": 0.09130219370126724,
|
|
"num_tokens": 284568.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.356634569168091,
|
|
"epoch": 0.1361573373676248,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.8489,
|
|
"mean_token_accuracy": 0.08760668933391572,
|
|
"num_tokens": 296499.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.341347980499267,
|
|
"epoch": 0.14120020171457387,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.7447,
|
|
"mean_token_accuracy": 0.09345789924263954,
|
|
"num_tokens": 307013.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.212394428253174,
|
|
"epoch": 0.14624306606152296,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.8123,
|
|
"mean_token_accuracy": 0.08974087685346603,
|
|
"num_tokens": 318604.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.267185878753662,
|
|
"epoch": 0.15128593040847202,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.757,
|
|
"mean_token_accuracy": 0.08906264528632164,
|
|
"num_tokens": 328968.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.338633060455322,
|
|
"epoch": 0.15632879475542108,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.7681,
|
|
"mean_token_accuracy": 0.09419665113091469,
|
|
"num_tokens": 338950.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.276337623596191,
|
|
"epoch": 0.16137165910237014,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.8953,
|
|
"mean_token_accuracy": 0.09276964589953422,
|
|
"num_tokens": 350667.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.386541271209717,
|
|
"epoch": 0.1664145234493192,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.9239,
|
|
"mean_token_accuracy": 0.09228215366601944,
|
|
"num_tokens": 361781.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.312812662124633,
|
|
"epoch": 0.1714573877962683,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.8138,
|
|
"mean_token_accuracy": 0.09285491853952407,
|
|
"num_tokens": 373410.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.218026733398437,
|
|
"epoch": 0.17650025214321735,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.7653,
|
|
"mean_token_accuracy": 0.09918249770998955,
|
|
"num_tokens": 385302.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.297417688369751,
|
|
"epoch": 0.1815431164901664,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.7624,
|
|
"mean_token_accuracy": 0.09598183333873749,
|
|
"num_tokens": 395183.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.290459728240966,
|
|
"epoch": 0.18658598083711547,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.758,
|
|
"mean_token_accuracy": 0.09613369405269623,
|
|
"num_tokens": 405356.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.29151873588562,
|
|
"epoch": 0.19162884518406456,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.7434,
|
|
"mean_token_accuracy": 0.09359151124954224,
|
|
"num_tokens": 415864.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.1764427661895756,
|
|
"epoch": 0.19667170953101362,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.6799,
|
|
"mean_token_accuracy": 0.0958583302795887,
|
|
"num_tokens": 426225.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.251503944396973,
|
|
"epoch": 0.20171457387796268,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.8068,
|
|
"mean_token_accuracy": 0.09116024672985076,
|
|
"num_tokens": 436847.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.271901845932007,
|
|
"epoch": 0.20675743822491174,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.7809,
|
|
"mean_token_accuracy": 0.09316601976752281,
|
|
"num_tokens": 448807.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.312121725082397,
|
|
"epoch": 0.2118003025718608,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.7572,
|
|
"mean_token_accuracy": 0.09582966193556786,
|
|
"num_tokens": 459691.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.235625171661377,
|
|
"epoch": 0.2168431669188099,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.7575,
|
|
"mean_token_accuracy": 0.08772426173090934,
|
|
"num_tokens": 470097.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.220300769805908,
|
|
"epoch": 0.22188603126575895,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.6645,
|
|
"mean_token_accuracy": 0.0966582477092743,
|
|
"num_tokens": 482006.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.262398672103882,
|
|
"epoch": 0.22692889561270801,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.8118,
|
|
"mean_token_accuracy": 0.09521539956331253,
|
|
"num_tokens": 493134.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.295605850219727,
|
|
"epoch": 0.23197175995965708,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.743,
|
|
"mean_token_accuracy": 0.0946677066385746,
|
|
"num_tokens": 504801.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.281533098220825,
|
|
"epoch": 0.23701462430660616,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.8161,
|
|
"mean_token_accuracy": 0.09522581920027733,
|
|
"num_tokens": 517523.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.199113416671753,
|
|
"epoch": 0.24205748865355523,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.7182,
|
|
"mean_token_accuracy": 0.09818312823772431,
|
|
"num_tokens": 528267.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.226230573654175,
|
|
"epoch": 0.24710035300050429,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.7107,
|
|
"mean_token_accuracy": 0.09675659462809563,
|
|
"num_tokens": 538750.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.268562889099121,
|
|
"epoch": 0.2521432173474534,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.7346,
|
|
"mean_token_accuracy": 0.097308711707592,
|
|
"num_tokens": 549380.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.262276458740234,
|
|
"epoch": 0.25718608169440244,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.7255,
|
|
"mean_token_accuracy": 0.09805365800857543,
|
|
"num_tokens": 559643.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.235341882705688,
|
|
"epoch": 0.2622289460413515,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.7632,
|
|
"mean_token_accuracy": 0.09189907684922219,
|
|
"num_tokens": 571250.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.196493816375733,
|
|
"epoch": 0.26727181038830056,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.7442,
|
|
"mean_token_accuracy": 0.10515325963497162,
|
|
"num_tokens": 582406.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.352744102478027,
|
|
"epoch": 0.2723146747352496,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.8521,
|
|
"mean_token_accuracy": 0.09412558823823929,
|
|
"num_tokens": 593115.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.18369722366333,
|
|
"epoch": 0.2773575390821987,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.7276,
|
|
"mean_token_accuracy": 0.09833462312817573,
|
|
"num_tokens": 603938.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.247702264785767,
|
|
"epoch": 0.28240040342914774,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.771,
|
|
"mean_token_accuracy": 0.09771163985133172,
|
|
"num_tokens": 614067.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.243542671203613,
|
|
"epoch": 0.2874432677760968,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.7654,
|
|
"mean_token_accuracy": 0.09758238643407821,
|
|
"num_tokens": 626271.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.291503524780273,
|
|
"epoch": 0.2924861321230459,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.7423,
|
|
"mean_token_accuracy": 0.09706803634762765,
|
|
"num_tokens": 636580.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.140781545639038,
|
|
"epoch": 0.297528996469995,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.6248,
|
|
"mean_token_accuracy": 0.1026174359023571,
|
|
"num_tokens": 648202.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.259130382537842,
|
|
"epoch": 0.30257186081694404,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.7082,
|
|
"mean_token_accuracy": 0.09900272190570832,
|
|
"num_tokens": 659274.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.239129590988159,
|
|
"epoch": 0.3076147251638931,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.7624,
|
|
"mean_token_accuracy": 0.09335382506251336,
|
|
"num_tokens": 670782.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.163164758682251,
|
|
"epoch": 0.31265758951084216,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.6747,
|
|
"mean_token_accuracy": 0.09428300261497498,
|
|
"num_tokens": 681392.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.21016297340393,
|
|
"epoch": 0.3177004538577912,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.7037,
|
|
"mean_token_accuracy": 0.10150701180100441,
|
|
"num_tokens": 692779.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.229485654830933,
|
|
"epoch": 0.3227433182047403,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.7269,
|
|
"mean_token_accuracy": 0.10247877538204193,
|
|
"num_tokens": 702200.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.1149742126464846,
|
|
"epoch": 0.32778618255168934,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.7218,
|
|
"mean_token_accuracy": 0.09391158893704414,
|
|
"num_tokens": 714817.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.174094009399414,
|
|
"epoch": 0.3328290468986384,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.6052,
|
|
"mean_token_accuracy": 0.0979647196829319,
|
|
"num_tokens": 725555.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.13692626953125,
|
|
"epoch": 0.3378719112455875,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.6195,
|
|
"mean_token_accuracy": 0.09633892104029655,
|
|
"num_tokens": 737372.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.185083723068237,
|
|
"epoch": 0.3429147755925366,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.6887,
|
|
"mean_token_accuracy": 0.09809593409299851,
|
|
"num_tokens": 748515.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.191352319717407,
|
|
"epoch": 0.34795763993948564,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.7335,
|
|
"mean_token_accuracy": 0.09944713562726974,
|
|
"num_tokens": 760586.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.127374887466431,
|
|
"epoch": 0.3530005042864347,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.6665,
|
|
"mean_token_accuracy": 0.09854965955018997,
|
|
"num_tokens": 772228.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.2189923286437985,
|
|
"epoch": 0.35804336863338376,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.7453,
|
|
"mean_token_accuracy": 0.0974178895354271,
|
|
"num_tokens": 784035.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.065363168716431,
|
|
"epoch": 0.3630862329803328,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.6418,
|
|
"mean_token_accuracy": 0.10008223056793213,
|
|
"num_tokens": 794383.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.085238981246948,
|
|
"epoch": 0.3681290973272819,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.5745,
|
|
"mean_token_accuracy": 0.10618990287184715,
|
|
"num_tokens": 805192.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.1900914192199705,
|
|
"epoch": 0.37317196167423095,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7213,
|
|
"mean_token_accuracy": 0.10011058002710342,
|
|
"num_tokens": 816794.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.214580726623535,
|
|
"epoch": 0.37821482602118,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7313,
|
|
"mean_token_accuracy": 0.09720972329378127,
|
|
"num_tokens": 827505.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.1219089984893795,
|
|
"epoch": 0.3832576903681291,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.5946,
|
|
"mean_token_accuracy": 0.10458309799432755,
|
|
"num_tokens": 837462.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.100887250900269,
|
|
"epoch": 0.3883005547150782,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.552,
|
|
"mean_token_accuracy": 0.10412034988403321,
|
|
"num_tokens": 848309.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.13932957649231,
|
|
"epoch": 0.39334341906202724,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7236,
|
|
"mean_token_accuracy": 0.09661326780915261,
|
|
"num_tokens": 859474.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.078795003890991,
|
|
"epoch": 0.3983862834089763,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.5887,
|
|
"mean_token_accuracy": 0.10084822699427605,
|
|
"num_tokens": 871638.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.1444016456604,
|
|
"epoch": 0.40342914775592537,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.6305,
|
|
"mean_token_accuracy": 0.10135297477245331,
|
|
"num_tokens": 882480.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.110705089569092,
|
|
"epoch": 0.4084720121028744,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.6247,
|
|
"mean_token_accuracy": 0.10225840136408806,
|
|
"num_tokens": 893366.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.121734523773194,
|
|
"epoch": 0.4135148764498235,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.5849,
|
|
"mean_token_accuracy": 0.1083330750465393,
|
|
"num_tokens": 904678.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.0099626064300535,
|
|
"epoch": 0.41855774079677255,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.5967,
|
|
"mean_token_accuracy": 0.1049334615468979,
|
|
"num_tokens": 915756.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.100469732284546,
|
|
"epoch": 0.4236006051437216,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.581,
|
|
"mean_token_accuracy": 0.09910632744431495,
|
|
"num_tokens": 926204.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.060488557815551,
|
|
"epoch": 0.4286434694906707,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6065,
|
|
"mean_token_accuracy": 0.10566477999091148,
|
|
"num_tokens": 936894.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.137387037277222,
|
|
"epoch": 0.4336863338376198,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6479,
|
|
"mean_token_accuracy": 0.10312608256936073,
|
|
"num_tokens": 948987.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.084998321533203,
|
|
"epoch": 0.43872919818456885,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6537,
|
|
"mean_token_accuracy": 0.1018140584230423,
|
|
"num_tokens": 961036.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.125703001022339,
|
|
"epoch": 0.4437720625315179,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.6763,
|
|
"mean_token_accuracy": 0.09882956370711327,
|
|
"num_tokens": 972445.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.0276741027832035,
|
|
"epoch": 0.44881492687846697,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.6481,
|
|
"mean_token_accuracy": 0.09917017221450805,
|
|
"num_tokens": 982847.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.137781095504761,
|
|
"epoch": 0.45385779122541603,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6044,
|
|
"mean_token_accuracy": 0.10046214014291763,
|
|
"num_tokens": 994055.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.033148241043091,
|
|
"epoch": 0.4589006555723651,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.5887,
|
|
"mean_token_accuracy": 0.10494610518217087,
|
|
"num_tokens": 1003565.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.981668519973755,
|
|
"epoch": 0.46394351991931415,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.4992,
|
|
"mean_token_accuracy": 0.11249068826436996,
|
|
"num_tokens": 1013376.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.059791231155396,
|
|
"epoch": 0.4689863842662632,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.5858,
|
|
"mean_token_accuracy": 0.10482539832592011,
|
|
"num_tokens": 1024580.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.999967098236084,
|
|
"epoch": 0.47402924861321233,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5819,
|
|
"mean_token_accuracy": 0.10121265724301338,
|
|
"num_tokens": 1037172.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.029157018661499,
|
|
"epoch": 0.4790721129601614,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.5026,
|
|
"mean_token_accuracy": 0.10607488974928855,
|
|
"num_tokens": 1046690.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.958544731140137,
|
|
"epoch": 0.48411497730711045,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.5115,
|
|
"mean_token_accuracy": 0.10288126021623611,
|
|
"num_tokens": 1058341.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.991989469528198,
|
|
"epoch": 0.4891578416540595,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.4452,
|
|
"mean_token_accuracy": 0.10999866873025894,
|
|
"num_tokens": 1067544.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.035932159423828,
|
|
"epoch": 0.49420070600100857,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.5638,
|
|
"mean_token_accuracy": 0.10314799919724464,
|
|
"num_tokens": 1079028.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.959421110153198,
|
|
"epoch": 0.49924357034795763,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6327,
|
|
"mean_token_accuracy": 0.1083416223526001,
|
|
"num_tokens": 1090709.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.954031038284302,
|
|
"epoch": 0.5042864346949067,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.5021,
|
|
"mean_token_accuracy": 0.10255713239312172,
|
|
"num_tokens": 1102987.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.159888648986817,
|
|
"epoch": 0.5093292990418558,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.6089,
|
|
"mean_token_accuracy": 0.09792369678616523,
|
|
"num_tokens": 1113370.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.860920858383179,
|
|
"epoch": 0.5143721633888049,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.452,
|
|
"mean_token_accuracy": 0.11532516852021217,
|
|
"num_tokens": 1124054.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.0116840362548825,
|
|
"epoch": 0.5194150277357539,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.4394,
|
|
"mean_token_accuracy": 0.1111026681959629,
|
|
"num_tokens": 1134530.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.8797385692596436,
|
|
"epoch": 0.524457892082703,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.4534,
|
|
"mean_token_accuracy": 0.11024491488933563,
|
|
"num_tokens": 1146011.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.911460733413696,
|
|
"epoch": 0.529500756429652,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5475,
|
|
"mean_token_accuracy": 0.10476418659090996,
|
|
"num_tokens": 1155822.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.9672643661499025,
|
|
"epoch": 0.5345436207766011,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.4517,
|
|
"mean_token_accuracy": 0.10918923616409301,
|
|
"num_tokens": 1167398.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.951054525375366,
|
|
"epoch": 0.5395864851235502,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5179,
|
|
"mean_token_accuracy": 0.10848165675997734,
|
|
"num_tokens": 1179706.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.842685604095459,
|
|
"epoch": 0.5446293494704992,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.4623,
|
|
"mean_token_accuracy": 0.10590022280812264,
|
|
"num_tokens": 1190159.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.887945699691772,
|
|
"epoch": 0.5496722138174484,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.3957,
|
|
"mean_token_accuracy": 0.11316062286496162,
|
|
"num_tokens": 1200223.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.8363923072814945,
|
|
"epoch": 0.5547150781643974,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5028,
|
|
"mean_token_accuracy": 0.10700231790542603,
|
|
"num_tokens": 1212108.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.985593938827515,
|
|
"epoch": 0.5597579425113465,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4666,
|
|
"mean_token_accuracy": 0.1086841069161892,
|
|
"num_tokens": 1221940.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.873191499710083,
|
|
"epoch": 0.5648008068582955,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5171,
|
|
"mean_token_accuracy": 0.10610352382063866,
|
|
"num_tokens": 1232884.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.956163358688355,
|
|
"epoch": 0.5698436712052446,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4327,
|
|
"mean_token_accuracy": 0.10770275890827179,
|
|
"num_tokens": 1242758.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.814071226119995,
|
|
"epoch": 0.5748865355521936,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4253,
|
|
"mean_token_accuracy": 0.11039303541183472,
|
|
"num_tokens": 1254787.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.858562231063843,
|
|
"epoch": 0.5799293998991427,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.4618,
|
|
"mean_token_accuracy": 0.11011067107319832,
|
|
"num_tokens": 1266364.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.871497106552124,
|
|
"epoch": 0.5849722642460918,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.4473,
|
|
"mean_token_accuracy": 0.10705809071660041,
|
|
"num_tokens": 1277105.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.04167366027832,
|
|
"epoch": 0.5900151285930408,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6277,
|
|
"mean_token_accuracy": 0.10272014588117599,
|
|
"num_tokens": 1288470.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.948342657089233,
|
|
"epoch": 0.59505799293999,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5378,
|
|
"mean_token_accuracy": 0.10782962441444396,
|
|
"num_tokens": 1301596.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.915746879577637,
|
|
"epoch": 0.600100857286939,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5886,
|
|
"mean_token_accuracy": 0.10370224639773369,
|
|
"num_tokens": 1314102.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.862001037597656,
|
|
"epoch": 0.6051437216338881,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4664,
|
|
"mean_token_accuracy": 0.11232062503695488,
|
|
"num_tokens": 1326939.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.942681741714478,
|
|
"epoch": 0.6101865859808371,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.5184,
|
|
"mean_token_accuracy": 0.10658980160951614,
|
|
"num_tokens": 1337877.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.862700510025024,
|
|
"epoch": 0.6152294503277862,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.4309,
|
|
"mean_token_accuracy": 0.10537772327661514,
|
|
"num_tokens": 1348425.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.898571968078613,
|
|
"epoch": 0.6202723146747352,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4755,
|
|
"mean_token_accuracy": 0.11413818150758744,
|
|
"num_tokens": 1359216.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.801260757446289,
|
|
"epoch": 0.6253151790216843,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.4595,
|
|
"mean_token_accuracy": 0.11029788851737976,
|
|
"num_tokens": 1371181.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.974251842498779,
|
|
"epoch": 0.6303580433686334,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4479,
|
|
"mean_token_accuracy": 0.10635321885347367,
|
|
"num_tokens": 1382371.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.8364170551300045,
|
|
"epoch": 0.6354009077155824,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4388,
|
|
"mean_token_accuracy": 0.10637976825237275,
|
|
"num_tokens": 1392981.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.818234634399414,
|
|
"epoch": 0.6404437720625316,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.4792,
|
|
"mean_token_accuracy": 0.11108815595507622,
|
|
"num_tokens": 1405782.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.831936502456665,
|
|
"epoch": 0.6454866364094806,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4194,
|
|
"mean_token_accuracy": 0.11271615028381347,
|
|
"num_tokens": 1415809.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.828110933303833,
|
|
"epoch": 0.6505295007564297,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4539,
|
|
"mean_token_accuracy": 0.10585936233401298,
|
|
"num_tokens": 1427845.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.850209903717041,
|
|
"epoch": 0.6555723651033787,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.4514,
|
|
"mean_token_accuracy": 0.10457649230957031,
|
|
"num_tokens": 1439761.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.716944456100464,
|
|
"epoch": 0.6606152294503278,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4282,
|
|
"mean_token_accuracy": 0.11434582471847535,
|
|
"num_tokens": 1450373.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.7856956958770756,
|
|
"epoch": 0.6656580937972768,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.308,
|
|
"mean_token_accuracy": 0.11268317475914955,
|
|
"num_tokens": 1462134.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.804158306121826,
|
|
"epoch": 0.6707009581442259,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4147,
|
|
"mean_token_accuracy": 0.11142993420362472,
|
|
"num_tokens": 1473600.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.800200128555298,
|
|
"epoch": 0.675743822491175,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.361,
|
|
"mean_token_accuracy": 0.11059448346495629,
|
|
"num_tokens": 1484313.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.850586938858032,
|
|
"epoch": 0.680786686838124,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.445,
|
|
"mean_token_accuracy": 0.10799070671200753,
|
|
"num_tokens": 1495407.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.776702213287353,
|
|
"epoch": 0.6858295511850732,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.526,
|
|
"mean_token_accuracy": 0.10533422976732254,
|
|
"num_tokens": 1506707.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.7912431240081785,
|
|
"epoch": 0.6908724155320222,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.3557,
|
|
"mean_token_accuracy": 0.11059139668941498,
|
|
"num_tokens": 1517991.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.7320136547088625,
|
|
"epoch": 0.6959152798789713,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4017,
|
|
"mean_token_accuracy": 0.11187214478850364,
|
|
"num_tokens": 1529538.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.701151990890503,
|
|
"epoch": 0.7009581442259203,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.4026,
|
|
"mean_token_accuracy": 0.11320648416876793,
|
|
"num_tokens": 1540183.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.852318525314331,
|
|
"epoch": 0.7060010085728694,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4475,
|
|
"mean_token_accuracy": 0.10782841965556145,
|
|
"num_tokens": 1552899.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.773307371139526,
|
|
"epoch": 0.7110438729198184,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.424,
|
|
"mean_token_accuracy": 0.10968175902962685,
|
|
"num_tokens": 1563219.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.795606517791748,
|
|
"epoch": 0.7160867372667675,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3444,
|
|
"mean_token_accuracy": 0.1151584655046463,
|
|
"num_tokens": 1573508.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.708667278289795,
|
|
"epoch": 0.7211296016137166,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4283,
|
|
"mean_token_accuracy": 0.10857328176498413,
|
|
"num_tokens": 1584319.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.716313695907592,
|
|
"epoch": 0.7261724659606656,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.297,
|
|
"mean_token_accuracy": 0.12033721506595611,
|
|
"num_tokens": 1595986.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.664373064041138,
|
|
"epoch": 0.7312153303076148,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.37,
|
|
"mean_token_accuracy": 0.11175311356782913,
|
|
"num_tokens": 1606742.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.734103393554688,
|
|
"epoch": 0.7362581946545638,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3412,
|
|
"mean_token_accuracy": 0.11203160732984543,
|
|
"num_tokens": 1618325.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.739001607894897,
|
|
"epoch": 0.7413010590015129,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.3498,
|
|
"mean_token_accuracy": 0.1113662451505661,
|
|
"num_tokens": 1627693.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.693121767044067,
|
|
"epoch": 0.7463439233484619,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3145,
|
|
"mean_token_accuracy": 0.11420592591166497,
|
|
"num_tokens": 1637707.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.634211206436158,
|
|
"epoch": 0.751386787695411,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3234,
|
|
"mean_token_accuracy": 0.1172995112836361,
|
|
"num_tokens": 1647277.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.757878684997559,
|
|
"epoch": 0.75642965204236,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3699,
|
|
"mean_token_accuracy": 0.11142466813325883,
|
|
"num_tokens": 1657919.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.714281225204468,
|
|
"epoch": 0.7614725163893091,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4127,
|
|
"mean_token_accuracy": 0.11027689799666404,
|
|
"num_tokens": 1669107.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.73455228805542,
|
|
"epoch": 0.7665153807362582,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4201,
|
|
"mean_token_accuracy": 0.1129143200814724,
|
|
"num_tokens": 1681338.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.710708665847778,
|
|
"epoch": 0.7715582450832073,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.2621,
|
|
"mean_token_accuracy": 0.11415471211075782,
|
|
"num_tokens": 1691288.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.568854570388794,
|
|
"epoch": 0.7766011094301564,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3503,
|
|
"mean_token_accuracy": 0.11261749044060707,
|
|
"num_tokens": 1702864.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.679662752151489,
|
|
"epoch": 0.7816439737771054,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.2684,
|
|
"mean_token_accuracy": 0.12043757885694503,
|
|
"num_tokens": 1712454.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.683392810821533,
|
|
"epoch": 0.7866868381240545,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4333,
|
|
"mean_token_accuracy": 0.1112380564212799,
|
|
"num_tokens": 1723925.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.6579155921936035,
|
|
"epoch": 0.7917297024710035,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.36,
|
|
"mean_token_accuracy": 0.11308201998472214,
|
|
"num_tokens": 1735624.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.683883953094482,
|
|
"epoch": 0.7967725668179526,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3225,
|
|
"mean_token_accuracy": 0.11539738923311234,
|
|
"num_tokens": 1746288.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.768927764892578,
|
|
"epoch": 0.8018154311649016,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4913,
|
|
"mean_token_accuracy": 0.10619482472538948,
|
|
"num_tokens": 1758145.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.602115249633789,
|
|
"epoch": 0.8068582955118507,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3161,
|
|
"mean_token_accuracy": 0.1136295534670353,
|
|
"num_tokens": 1768586.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.709038305282593,
|
|
"epoch": 0.8119011598587998,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.321,
|
|
"mean_token_accuracy": 0.11802070513367653,
|
|
"num_tokens": 1780502.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.624473953247071,
|
|
"epoch": 0.8169440242057489,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3582,
|
|
"mean_token_accuracy": 0.11211898848414421,
|
|
"num_tokens": 1791450.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.6230384349823,
|
|
"epoch": 0.821986888552698,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3628,
|
|
"mean_token_accuracy": 0.1177310198545456,
|
|
"num_tokens": 1803690.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.609463548660278,
|
|
"epoch": 0.827029752899647,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2264,
|
|
"mean_token_accuracy": 0.11917722448706627,
|
|
"num_tokens": 1813630.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.637943077087402,
|
|
"epoch": 0.8320726172465961,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2778,
|
|
"mean_token_accuracy": 0.11222293600440025,
|
|
"num_tokens": 1824537.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.645944356918335,
|
|
"epoch": 0.8371154815935451,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3373,
|
|
"mean_token_accuracy": 0.1141052208840847,
|
|
"num_tokens": 1834404.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.700550842285156,
|
|
"epoch": 0.8421583459404942,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3104,
|
|
"mean_token_accuracy": 0.12324466183781624,
|
|
"num_tokens": 1845371.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.50260362625122,
|
|
"epoch": 0.8472012102874432,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.341,
|
|
"mean_token_accuracy": 0.11797986924648285,
|
|
"num_tokens": 1856749.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.63838267326355,
|
|
"epoch": 0.8522440746343923,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.3337,
|
|
"mean_token_accuracy": 0.109927898645401,
|
|
"num_tokens": 1866577.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.72889552116394,
|
|
"epoch": 0.8572869389813415,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.282,
|
|
"mean_token_accuracy": 0.1153933621942997,
|
|
"num_tokens": 1877123.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.602636194229126,
|
|
"epoch": 0.8623298033282905,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.3232,
|
|
"mean_token_accuracy": 0.11297061070799827,
|
|
"num_tokens": 1888171.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.546885442733765,
|
|
"epoch": 0.8673726676752396,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.2226,
|
|
"mean_token_accuracy": 0.11574787572026253,
|
|
"num_tokens": 1898150.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.649620103836059,
|
|
"epoch": 0.8724155320221886,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3305,
|
|
"mean_token_accuracy": 0.11207615062594414,
|
|
"num_tokens": 1907641.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.632721757888794,
|
|
"epoch": 0.8774583963691377,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2845,
|
|
"mean_token_accuracy": 0.11901096925139427,
|
|
"num_tokens": 1918297.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.578178358078003,
|
|
"epoch": 0.8825012607160867,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3186,
|
|
"mean_token_accuracy": 0.1133461907505989,
|
|
"num_tokens": 1930268.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.531199407577515,
|
|
"epoch": 0.8875441250630358,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2465,
|
|
"mean_token_accuracy": 0.12018090561032295,
|
|
"num_tokens": 1941762.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.59914002418518,
|
|
"epoch": 0.8925869894099848,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.3195,
|
|
"mean_token_accuracy": 0.11756950095295907,
|
|
"num_tokens": 1952416.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.5428722381591795,
|
|
"epoch": 0.8976298537569339,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2651,
|
|
"mean_token_accuracy": 0.11572367325425148,
|
|
"num_tokens": 1963451.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.58216700553894,
|
|
"epoch": 0.902672718103883,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.2794,
|
|
"mean_token_accuracy": 0.11693633943796158,
|
|
"num_tokens": 1974360.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.693656730651855,
|
|
"epoch": 0.9077155824508321,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3962,
|
|
"mean_token_accuracy": 0.10579293221235275,
|
|
"num_tokens": 1985823.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.767707967758179,
|
|
"epoch": 0.9127584467977812,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4287,
|
|
"mean_token_accuracy": 0.11259771957993507,
|
|
"num_tokens": 1997475.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.570524787902832,
|
|
"epoch": 0.9178013111447302,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2708,
|
|
"mean_token_accuracy": 0.11376104354858399,
|
|
"num_tokens": 2009029.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.4297045230865475,
|
|
"epoch": 0.9228441754916793,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.141,
|
|
"mean_token_accuracy": 0.11884743124246597,
|
|
"num_tokens": 2019564.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.618962907791138,
|
|
"epoch": 0.9278870398386283,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3949,
|
|
"mean_token_accuracy": 0.11527436822652817,
|
|
"num_tokens": 2031394.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.623352861404419,
|
|
"epoch": 0.9329299041855774,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2762,
|
|
"mean_token_accuracy": 0.1077704131603241,
|
|
"num_tokens": 2042132.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.538896608352661,
|
|
"epoch": 0.9379727685325264,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.2666,
|
|
"mean_token_accuracy": 0.1150762416422367,
|
|
"num_tokens": 2054469.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.48656005859375,
|
|
"epoch": 0.9430156328794755,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2445,
|
|
"mean_token_accuracy": 0.11287088468670844,
|
|
"num_tokens": 2065001.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.535896301269531,
|
|
"epoch": 0.9480584972264247,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2612,
|
|
"mean_token_accuracy": 0.11251269802451133,
|
|
"num_tokens": 2075786.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.53080735206604,
|
|
"epoch": 0.9531013615733737,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.2161,
|
|
"mean_token_accuracy": 0.1198991946876049,
|
|
"num_tokens": 2087818.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.592200040817261,
|
|
"epoch": 0.9581442259203228,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3004,
|
|
"mean_token_accuracy": 0.1103257454931736,
|
|
"num_tokens": 2098771.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.4617808818817135,
|
|
"epoch": 0.9631870902672718,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.2119,
|
|
"mean_token_accuracy": 0.12062049955129624,
|
|
"num_tokens": 2110036.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.536091947555542,
|
|
"epoch": 0.9682299546142209,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.312,
|
|
"mean_token_accuracy": 0.12011703029274941,
|
|
"num_tokens": 2120411.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.482806015014648,
|
|
"epoch": 0.9732728189611699,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2244,
|
|
"mean_token_accuracy": 0.11997658982872964,
|
|
"num_tokens": 2132831.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.493047666549683,
|
|
"epoch": 0.978315683308119,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1953,
|
|
"mean_token_accuracy": 0.11635629460215569,
|
|
"num_tokens": 2143166.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.481213617324829,
|
|
"epoch": 0.983358547655068,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.2695,
|
|
"mean_token_accuracy": 0.10903773382306099,
|
|
"num_tokens": 2154366.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.55472240447998,
|
|
"epoch": 0.9884014120020171,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2535,
|
|
"mean_token_accuracy": 0.11955308020114899,
|
|
"num_tokens": 2164396.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.478724098205566,
|
|
"epoch": 0.9934442763489663,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3211,
|
|
"mean_token_accuracy": 0.11328333243727684,
|
|
"num_tokens": 2177295.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.5510862350463865,
|
|
"epoch": 0.9984871406959153,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1761,
|
|
"mean_token_accuracy": 0.12193462178111077,
|
|
"num_tokens": 2188832.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.45083400938246,
|
|
"epoch": 1.0030257186081695,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1345,
|
|
"mean_token_accuracy": 0.11941885699828465,
|
|
"num_tokens": 2197926.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.500611925125122,
|
|
"epoch": 1.0080685829551186,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.173,
|
|
"mean_token_accuracy": 0.11789006367325783,
|
|
"num_tokens": 2209156.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0080685829551186,
|
|
"eval_entropy": 6.293738744337746,
|
|
"eval_loss": 6.301364898681641,
|
|
"eval_mean_token_accuracy": 0.11593409591655392,
|
|
"eval_num_tokens": 2209156.0,
|
|
"eval_runtime": 2.053,
|
|
"eval_samples_per_second": 1528.532,
|
|
"eval_steps_per_second": 191.432,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.380574512481689,
|
|
"epoch": 1.0131114473020675,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999997762223182,
|
|
"loss": 6.0618,
|
|
"mean_token_accuracy": 0.1228681594133377,
|
|
"num_tokens": 2220419.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.594297313690186,
|
|
"epoch": 1.0181543116490166,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999988671262489,
|
|
"loss": 6.2169,
|
|
"mean_token_accuracy": 0.1185660257935524,
|
|
"num_tokens": 2232671.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.441428279876709,
|
|
"epoch": 1.0231971759959657,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999972587285105,
|
|
"loss": 6.1056,
|
|
"mean_token_accuracy": 0.12909807413816451,
|
|
"num_tokens": 2243367.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.527055931091309,
|
|
"epoch": 1.0282400403429148,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999949510341015,
|
|
"loss": 6.0476,
|
|
"mean_token_accuracy": 0.1258266642689705,
|
|
"num_tokens": 2254120.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.445385408401489,
|
|
"epoch": 1.033282904689864,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999919440501947,
|
|
"loss": 6.0938,
|
|
"mean_token_accuracy": 0.12587995156645776,
|
|
"num_tokens": 2265430.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.474833059310913,
|
|
"epoch": 1.0383257690368128,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999882377861355,
|
|
"loss": 6.1379,
|
|
"mean_token_accuracy": 0.11899987384676933,
|
|
"num_tokens": 2277031.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.336647272109985,
|
|
"epoch": 1.043368633383762,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999838322534433,
|
|
"loss": 6.0122,
|
|
"mean_token_accuracy": 0.1258586823940277,
|
|
"num_tokens": 2287430.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.418465614318848,
|
|
"epoch": 1.048411497730711,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999787274658107,
|
|
"loss": 6.0937,
|
|
"mean_token_accuracy": 0.12477722540497779,
|
|
"num_tokens": 2298063.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.434740781784058,
|
|
"epoch": 1.0534543620776602,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999729234391032,
|
|
"loss": 6.14,
|
|
"mean_token_accuracy": 0.11822782978415489,
|
|
"num_tokens": 2308684.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.431240081787109,
|
|
"epoch": 1.058497226424609,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00049996642019136,
|
|
"loss": 6.069,
|
|
"mean_token_accuracy": 0.1249009370803833,
|
|
"num_tokens": 2319611.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.3972173690795895,
|
|
"epoch": 1.0635400907715582,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999592177427933,
|
|
"loss": 6.0429,
|
|
"mean_token_accuracy": 0.12159126326441765,
|
|
"num_tokens": 2330655.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.373932552337647,
|
|
"epoch": 1.0685829551185073,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999513161157885,
|
|
"loss": 6.0513,
|
|
"mean_token_accuracy": 0.1277623564004898,
|
|
"num_tokens": 2341717.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.368822145462036,
|
|
"epoch": 1.0736258194654564,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999427153349041,
|
|
"loss": 6.1222,
|
|
"mean_token_accuracy": 0.12355232909321785,
|
|
"num_tokens": 2352614.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.362414598464966,
|
|
"epoch": 1.0786686838124055,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999334154268714,
|
|
"loss": 6.0426,
|
|
"mean_token_accuracy": 0.12430263385176658,
|
|
"num_tokens": 2362651.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.453306579589844,
|
|
"epoch": 1.0837115481593544,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999234164205949,
|
|
"loss": 6.0332,
|
|
"mean_token_accuracy": 0.12451537624001503,
|
|
"num_tokens": 2375518.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.368288326263428,
|
|
"epoch": 1.0887544125063036,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999127183471517,
|
|
"loss": 6.1056,
|
|
"mean_token_accuracy": 0.11944749429821969,
|
|
"num_tokens": 2386908.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.233647203445434,
|
|
"epoch": 1.0937972768532527,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999013212397915,
|
|
"loss": 5.9485,
|
|
"mean_token_accuracy": 0.13038301914930345,
|
|
"num_tokens": 2398992.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.483757638931275,
|
|
"epoch": 1.0988401412002018,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998892251339368,
|
|
"loss": 6.1526,
|
|
"mean_token_accuracy": 0.11885706782341003,
|
|
"num_tokens": 2410741.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.373459053039551,
|
|
"epoch": 1.1038830055471507,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998764300671827,
|
|
"loss": 6.1014,
|
|
"mean_token_accuracy": 0.11981240957975388,
|
|
"num_tokens": 2421702.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.387602806091309,
|
|
"epoch": 1.1089258698940998,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998629360792965,
|
|
"loss": 6.1255,
|
|
"mean_token_accuracy": 0.12111065089702606,
|
|
"num_tokens": 2433676.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.424046039581299,
|
|
"epoch": 1.113968734241049,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998487432122179,
|
|
"loss": 6.0617,
|
|
"mean_token_accuracy": 0.13049159124493598,
|
|
"num_tokens": 2445415.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.266917037963867,
|
|
"epoch": 1.119011598587998,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004998338515100585,
|
|
"loss": 5.9754,
|
|
"mean_token_accuracy": 0.12859009951353073,
|
|
"num_tokens": 2456195.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.311762189865112,
|
|
"epoch": 1.1240544629349472,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998182610191022,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.12430239692330361,
|
|
"num_tokens": 2469448.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.343682909011841,
|
|
"epoch": 1.129097327281896,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004998019717878047,
|
|
"loss": 6.0318,
|
|
"mean_token_accuracy": 0.12581473588943481,
|
|
"num_tokens": 2480532.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.31264820098877,
|
|
"epoch": 1.1341401916288452,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499784983866793,
|
|
"loss": 5.9978,
|
|
"mean_token_accuracy": 0.12010813131928444,
|
|
"num_tokens": 2492203.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.282516193389893,
|
|
"epoch": 1.1391830559757943,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997672973088661,
|
|
"loss": 5.9818,
|
|
"mean_token_accuracy": 0.12565053179860114,
|
|
"num_tokens": 2502503.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.332369613647461,
|
|
"epoch": 1.1442259203227434,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997489121689944,
|
|
"loss": 6.0528,
|
|
"mean_token_accuracy": 0.12229174375534058,
|
|
"num_tokens": 2513981.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.361654996871948,
|
|
"epoch": 1.1492687846696923,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997298285043189,
|
|
"loss": 6.0537,
|
|
"mean_token_accuracy": 0.1268092177808285,
|
|
"num_tokens": 2524572.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.332787799835205,
|
|
"epoch": 1.1543116490166414,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997100463741527,
|
|
"loss": 6.0419,
|
|
"mean_token_accuracy": 0.1255467101931572,
|
|
"num_tokens": 2534159.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.339848136901855,
|
|
"epoch": 1.1593545133635905,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004996895658399785,
|
|
"loss": 6.0105,
|
|
"mean_token_accuracy": 0.12619809359312056,
|
|
"num_tokens": 2545486.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.28226490020752,
|
|
"epoch": 1.1643973777105396,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004996683869654507,
|
|
"loss": 6.0257,
|
|
"mean_token_accuracy": 0.12942853271961213,
|
|
"num_tokens": 2556288.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.308646821975708,
|
|
"epoch": 1.1694402420574885,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996465098163934,
|
|
"loss": 6.0005,
|
|
"mean_token_accuracy": 0.12609127834439277,
|
|
"num_tokens": 2567241.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.257927417755127,
|
|
"epoch": 1.1744831064044376,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996239344608015,
|
|
"loss": 6.0212,
|
|
"mean_token_accuracy": 0.12528160214424133,
|
|
"num_tokens": 2577697.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.353972053527832,
|
|
"epoch": 1.1795259707513868,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996006609688398,
|
|
"loss": 6.0098,
|
|
"mean_token_accuracy": 0.13097208365797997,
|
|
"num_tokens": 2588231.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.2682051181793215,
|
|
"epoch": 1.1845688350983359,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004995766894128426,
|
|
"loss": 5.9628,
|
|
"mean_token_accuracy": 0.13096450343728067,
|
|
"num_tokens": 2598750.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.331038904190064,
|
|
"epoch": 1.189611699445285,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995520198673143,
|
|
"loss": 6.0365,
|
|
"mean_token_accuracy": 0.12640778720378876,
|
|
"num_tokens": 2609481.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.301260614395142,
|
|
"epoch": 1.194654563792234,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995266524089283,
|
|
"loss": 6.0326,
|
|
"mean_token_accuracy": 0.12409081980586052,
|
|
"num_tokens": 2619746.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.2238668441772464,
|
|
"epoch": 1.199697428139183,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995005871165271,
|
|
"loss": 5.906,
|
|
"mean_token_accuracy": 0.13082837238907813,
|
|
"num_tokens": 2629578.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.251807498931885,
|
|
"epoch": 1.2047402924861321,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004994738240711225,
|
|
"loss": 5.9644,
|
|
"mean_token_accuracy": 0.13376058265566826,
|
|
"num_tokens": 2639063.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.277953433990478,
|
|
"epoch": 1.2097831568330812,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004994463633558945,
|
|
"loss": 6.0826,
|
|
"mean_token_accuracy": 0.1265018604695797,
|
|
"num_tokens": 2649304.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.371228885650635,
|
|
"epoch": 1.2148260211800301,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004994182050561917,
|
|
"loss": 6.0794,
|
|
"mean_token_accuracy": 0.12711584940552711,
|
|
"num_tokens": 2659917.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.348291826248169,
|
|
"epoch": 1.2198688855269793,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004993893492595306,
|
|
"loss": 6.1416,
|
|
"mean_token_accuracy": 0.12414629384875298,
|
|
"num_tokens": 2672136.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.322640752792358,
|
|
"epoch": 1.2249117498739284,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004993597960555959,
|
|
"loss": 6.0743,
|
|
"mean_token_accuracy": 0.12048955485224724,
|
|
"num_tokens": 2683656.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.207147073745728,
|
|
"epoch": 1.2299546142208775,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993295455362395,
|
|
"loss": 5.9307,
|
|
"mean_token_accuracy": 0.12895723581314086,
|
|
"num_tokens": 2693289.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.320633220672607,
|
|
"epoch": 1.2349974785678266,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004992985977954808,
|
|
"loss": 5.9874,
|
|
"mean_token_accuracy": 0.12993058934807777,
|
|
"num_tokens": 2703442.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.355954933166504,
|
|
"epoch": 1.2400403429147757,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499266952929506,
|
|
"loss": 6.0387,
|
|
"mean_token_accuracy": 0.12347332090139389,
|
|
"num_tokens": 2715229.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.2153572082519535,
|
|
"epoch": 1.2450832072617246,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004992346110366682,
|
|
"loss": 5.9723,
|
|
"mean_token_accuracy": 0.12580555379390718,
|
|
"num_tokens": 2725075.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.289635229110718,
|
|
"epoch": 1.2501260716086737,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004992015722174865,
|
|
"loss": 5.9753,
|
|
"mean_token_accuracy": 0.1257520206272602,
|
|
"num_tokens": 2735138.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.132227563858033,
|
|
"epoch": 1.2551689359556228,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991678365746466,
|
|
"loss": 5.9169,
|
|
"mean_token_accuracy": 0.13637566268444062,
|
|
"num_tokens": 2745824.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.316624927520752,
|
|
"epoch": 1.2602118003025717,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991334042129996,
|
|
"loss": 6.0001,
|
|
"mean_token_accuracy": 0.1267104282975197,
|
|
"num_tokens": 2756194.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.3138305187225345,
|
|
"epoch": 1.2652546646495209,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004990982752395618,
|
|
"loss": 6.1534,
|
|
"mean_token_accuracy": 0.12595434188842775,
|
|
"num_tokens": 2766811.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.329576921463013,
|
|
"epoch": 1.27029752899647,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004990624497635152,
|
|
"loss": 5.9701,
|
|
"mean_token_accuracy": 0.12633612081408502,
|
|
"num_tokens": 2777079.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.188473510742187,
|
|
"epoch": 1.275340393343419,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499025927896206,
|
|
"loss": 5.9974,
|
|
"mean_token_accuracy": 0.1308589056134224,
|
|
"num_tokens": 2788789.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.341365766525269,
|
|
"epoch": 1.2803832576903682,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004989887097511451,
|
|
"loss": 6.0693,
|
|
"mean_token_accuracy": 0.12354071140289306,
|
|
"num_tokens": 2801587.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.293736124038697,
|
|
"epoch": 1.2854261220373173,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004989507954440074,
|
|
"loss": 5.9865,
|
|
"mean_token_accuracy": 0.12447015196084976,
|
|
"num_tokens": 2812027.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.10911865234375,
|
|
"epoch": 1.2904689863842662,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004989121850926313,
|
|
"loss": 5.9818,
|
|
"mean_token_accuracy": 0.128651275485754,
|
|
"num_tokens": 2823791.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.327929878234864,
|
|
"epoch": 1.2955118507312153,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004988728788170188,
|
|
"loss": 6.0357,
|
|
"mean_token_accuracy": 0.12696870118379594,
|
|
"num_tokens": 2835015.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.332037115097046,
|
|
"epoch": 1.3005547150781644,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004988328767393346,
|
|
"loss": 5.9923,
|
|
"mean_token_accuracy": 0.12338553890585899,
|
|
"num_tokens": 2846078.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.206235599517822,
|
|
"epoch": 1.3055975794251133,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004987921789839061,
|
|
"loss": 5.9602,
|
|
"mean_token_accuracy": 0.13209799453616142,
|
|
"num_tokens": 2856168.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.26894588470459,
|
|
"epoch": 1.3106404437720625,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004987507856772229,
|
|
"loss": 6.0354,
|
|
"mean_token_accuracy": 0.1287067450582981,
|
|
"num_tokens": 2868258.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.223844909667969,
|
|
"epoch": 1.3156833081190116,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004987086969479364,
|
|
"loss": 5.9567,
|
|
"mean_token_accuracy": 0.12373382598161697,
|
|
"num_tokens": 2879202.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.128959083557129,
|
|
"epoch": 1.3207261724659607,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004986659129268592,
|
|
"loss": 5.9066,
|
|
"mean_token_accuracy": 0.13103111013770102,
|
|
"num_tokens": 2890669.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.189808559417725,
|
|
"epoch": 1.3257690368129098,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004986224337469652,
|
|
"loss": 5.9354,
|
|
"mean_token_accuracy": 0.12434953525662422,
|
|
"num_tokens": 2900799.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.236221122741699,
|
|
"epoch": 1.330811901159859,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004985782595433886,
|
|
"loss": 6.042,
|
|
"mean_token_accuracy": 0.12363123670220375,
|
|
"num_tokens": 2913166.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.242849397659302,
|
|
"epoch": 1.3358547655068078,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004985333904534237,
|
|
"loss": 6.0172,
|
|
"mean_token_accuracy": 0.1291557051241398,
|
|
"num_tokens": 2924750.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.271391916275024,
|
|
"epoch": 1.340897629853757,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000498487826616525,
|
|
"loss": 5.9567,
|
|
"mean_token_accuracy": 0.1324070945382118,
|
|
"num_tokens": 2935343.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.224100351333618,
|
|
"epoch": 1.345940494200706,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004984415681743056,
|
|
"loss": 5.9579,
|
|
"mean_token_accuracy": 0.13276781067252158,
|
|
"num_tokens": 2945420.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.310044288635254,
|
|
"epoch": 1.350983358547655,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004983946152705382,
|
|
"loss": 6.0671,
|
|
"mean_token_accuracy": 0.1256522938609123,
|
|
"num_tokens": 2955580.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.282643127441406,
|
|
"epoch": 1.356026222894604,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004983469680511531,
|
|
"loss": 5.9195,
|
|
"mean_token_accuracy": 0.13628479614853858,
|
|
"num_tokens": 2966192.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.144649648666382,
|
|
"epoch": 1.3610690872415532,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004982986266642393,
|
|
"loss": 5.9084,
|
|
"mean_token_accuracy": 0.12781317979097367,
|
|
"num_tokens": 2976715.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.261746740341186,
|
|
"epoch": 1.3661119515885023,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004982495912600426,
|
|
"loss": 6.0509,
|
|
"mean_token_accuracy": 0.12088210582733154,
|
|
"num_tokens": 2987763.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.208369159698487,
|
|
"epoch": 1.3711548159354514,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004981998619909664,
|
|
"loss": 5.8525,
|
|
"mean_token_accuracy": 0.13546379208564757,
|
|
"num_tokens": 2998518.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.262377691268921,
|
|
"epoch": 1.3761976802824005,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004981494390115704,
|
|
"loss": 6.0739,
|
|
"mean_token_accuracy": 0.13269724696874619,
|
|
"num_tokens": 3008953.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.125183010101319,
|
|
"epoch": 1.3812405446293494,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004980983224785704,
|
|
"loss": 5.92,
|
|
"mean_token_accuracy": 0.13169071897864343,
|
|
"num_tokens": 3018763.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.243589067459107,
|
|
"epoch": 1.3862834089762985,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004980465125508376,
|
|
"loss": 6.0157,
|
|
"mean_token_accuracy": 0.12686386331915855,
|
|
"num_tokens": 3030061.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.161712503433227,
|
|
"epoch": 1.3913262733232477,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004979940093893986,
|
|
"loss": 5.9565,
|
|
"mean_token_accuracy": 0.12839753404259682,
|
|
"num_tokens": 3040646.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.308729410171509,
|
|
"epoch": 1.3963691376701965,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004979408131574345,
|
|
"loss": 6.0451,
|
|
"mean_token_accuracy": 0.12857256084680557,
|
|
"num_tokens": 3051470.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.198573255538941,
|
|
"epoch": 1.4014120020171457,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004978869240202802,
|
|
"loss": 6.0424,
|
|
"mean_token_accuracy": 0.12916125878691673,
|
|
"num_tokens": 3063539.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.217213916778564,
|
|
"epoch": 1.4064548663640948,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004978323421454246,
|
|
"loss": 5.8652,
|
|
"mean_token_accuracy": 0.12893132865428925,
|
|
"num_tokens": 3073945.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.17272982597351,
|
|
"epoch": 1.411497730711044,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004977770677025092,
|
|
"loss": 5.9556,
|
|
"mean_token_accuracy": 0.12694748416543006,
|
|
"num_tokens": 3085276.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.168652772903442,
|
|
"epoch": 1.416540595057993,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004977211008633284,
|
|
"loss": 5.974,
|
|
"mean_token_accuracy": 0.1286464363336563,
|
|
"num_tokens": 3097599.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.152103853225708,
|
|
"epoch": 1.4215834594049421,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004976644418018285,
|
|
"loss": 5.9146,
|
|
"mean_token_accuracy": 0.1416459374129772,
|
|
"num_tokens": 3109536.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.224145460128784,
|
|
"epoch": 1.426626323751891,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000497607090694107,
|
|
"loss": 5.8913,
|
|
"mean_token_accuracy": 0.1341511972248554,
|
|
"num_tokens": 3119178.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.119580841064453,
|
|
"epoch": 1.4316691880988401,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004975490477184126,
|
|
"loss": 5.9717,
|
|
"mean_token_accuracy": 0.12969112098217012,
|
|
"num_tokens": 3131183.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.23134126663208,
|
|
"epoch": 1.4367120524457893,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000497490313055144,
|
|
"loss": 5.9428,
|
|
"mean_token_accuracy": 0.13121251612901688,
|
|
"num_tokens": 3142038.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.12155122756958,
|
|
"epoch": 1.4417549167927382,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004974308868868501,
|
|
"loss": 5.9582,
|
|
"mean_token_accuracy": 0.1304076172411442,
|
|
"num_tokens": 3152878.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.24698657989502,
|
|
"epoch": 1.4467977811396873,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004973707693982287,
|
|
"loss": 5.9284,
|
|
"mean_token_accuracy": 0.13284528702497483,
|
|
"num_tokens": 3165485.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.169107341766358,
|
|
"epoch": 1.4518406454866364,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004973099607761264,
|
|
"loss": 5.9444,
|
|
"mean_token_accuracy": 0.1265783950686455,
|
|
"num_tokens": 3175134.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.165101671218872,
|
|
"epoch": 1.4568835098335855,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004972484612095378,
|
|
"loss": 5.9255,
|
|
"mean_token_accuracy": 0.1292041137814522,
|
|
"num_tokens": 3185319.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.138285732269287,
|
|
"epoch": 1.4619263741805346,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000497186270889605,
|
|
"loss": 6.0183,
|
|
"mean_token_accuracy": 0.13033954948186874,
|
|
"num_tokens": 3196427.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.190555095672607,
|
|
"epoch": 1.4669692385274837,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004971233900096169,
|
|
"loss": 5.9092,
|
|
"mean_token_accuracy": 0.13712915629148484,
|
|
"num_tokens": 3207233.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.175220775604248,
|
|
"epoch": 1.4720121028744326,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004970598187650085,
|
|
"loss": 5.937,
|
|
"mean_token_accuracy": 0.12608329206705093,
|
|
"num_tokens": 3217999.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.170501661300659,
|
|
"epoch": 1.4770549672213817,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000496995557353361,
|
|
"loss": 5.9915,
|
|
"mean_token_accuracy": 0.12114647999405861,
|
|
"num_tokens": 3228965.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.154280710220337,
|
|
"epoch": 1.4820978315683309,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004969306059744004,
|
|
"loss": 5.9169,
|
|
"mean_token_accuracy": 0.13613021969795228,
|
|
"num_tokens": 3239960.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.200178766250611,
|
|
"epoch": 1.4871406959152798,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004968649648299968,
|
|
"loss": 5.9223,
|
|
"mean_token_accuracy": 0.12361980378627777,
|
|
"num_tokens": 3249250.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.096926069259643,
|
|
"epoch": 1.4921835602622289,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004967986341241646,
|
|
"loss": 5.966,
|
|
"mean_token_accuracy": 0.12727833837270736,
|
|
"num_tokens": 3260936.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.113045310974121,
|
|
"epoch": 1.497226424609178,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004967316140630611,
|
|
"loss": 5.783,
|
|
"mean_token_accuracy": 0.1374378450214863,
|
|
"num_tokens": 3271571.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.131659078598022,
|
|
"epoch": 1.502269288956127,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004966639048549862,
|
|
"loss": 5.9085,
|
|
"mean_token_accuracy": 0.13167561665177346,
|
|
"num_tokens": 3282216.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.030834054946899,
|
|
"epoch": 1.5073121533030762,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004965955067103816,
|
|
"loss": 5.8573,
|
|
"mean_token_accuracy": 0.13101960197091103,
|
|
"num_tokens": 3293791.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.221430778503418,
|
|
"epoch": 1.5123550176500253,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004965264198418305,
|
|
"loss": 5.8628,
|
|
"mean_token_accuracy": 0.13768607452511789,
|
|
"num_tokens": 3304260.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.028321027755737,
|
|
"epoch": 1.5173978819969742,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004964566444640564,
|
|
"loss": 5.8848,
|
|
"mean_token_accuracy": 0.13266941159963608,
|
|
"num_tokens": 3315882.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.226700925827027,
|
|
"epoch": 1.5224407463439233,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004963861807939227,
|
|
"loss": 5.8893,
|
|
"mean_token_accuracy": 0.12964478880167007,
|
|
"num_tokens": 3326649.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.0752287864685055,
|
|
"epoch": 1.5274836106908725,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004963150290504321,
|
|
"loss": 5.9676,
|
|
"mean_token_accuracy": 0.1294805459678173,
|
|
"num_tokens": 3338670.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.241375064849853,
|
|
"epoch": 1.5325264750378214,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004962431894547259,
|
|
"loss": 6.0023,
|
|
"mean_token_accuracy": 0.1268956780433655,
|
|
"num_tokens": 3350360.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.096974945068359,
|
|
"epoch": 1.5375693393847705,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004961706622300834,
|
|
"loss": 5.8394,
|
|
"mean_token_accuracy": 0.1453121781349182,
|
|
"num_tokens": 3360647.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.069055509567261,
|
|
"epoch": 1.5426122037317196,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004960974476019206,
|
|
"loss": 5.9402,
|
|
"mean_token_accuracy": 0.13708192557096482,
|
|
"num_tokens": 3371153.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.1496970653533936,
|
|
"epoch": 1.5476550680786687,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004960235457977905,
|
|
"loss": 5.9111,
|
|
"mean_token_accuracy": 0.13489063307642937,
|
|
"num_tokens": 3382027.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.172466611862182,
|
|
"epoch": 1.5526979324256178,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004959489570473814,
|
|
"loss": 5.8597,
|
|
"mean_token_accuracy": 0.13355688229203225,
|
|
"num_tokens": 3393122.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.046972990036011,
|
|
"epoch": 1.557740796772567,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000495873681582517,
|
|
"loss": 5.8587,
|
|
"mean_token_accuracy": 0.12852961122989653,
|
|
"num_tokens": 3403992.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.234967470169067,
|
|
"epoch": 1.5627836611195158,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004957977196371552,
|
|
"loss": 5.9556,
|
|
"mean_token_accuracy": 0.13196602687239647,
|
|
"num_tokens": 3415375.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.225420665740967,
|
|
"epoch": 1.567826525466465,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004957210714473873,
|
|
"loss": 5.9751,
|
|
"mean_token_accuracy": 0.12916150763630868,
|
|
"num_tokens": 3426670.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.183484077453613,
|
|
"epoch": 1.572869389813414,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004956437372514378,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.12805879786610602,
|
|
"num_tokens": 3438262.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.1148802757263185,
|
|
"epoch": 1.577912254160363,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004955657172896631,
|
|
"loss": 5.9104,
|
|
"mean_token_accuracy": 0.13370768427848817,
|
|
"num_tokens": 3451442.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.0889012813568115,
|
|
"epoch": 1.582955118507312,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004954870118045511,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.1385216996073723,
|
|
"num_tokens": 3462022.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.082703638076782,
|
|
"epoch": 1.5879979828542612,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004954076210407203,
|
|
"loss": 5.8538,
|
|
"mean_token_accuracy": 0.12839669212698937,
|
|
"num_tokens": 3472777.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.127910327911377,
|
|
"epoch": 1.5930408472012103,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004953275452449188,
|
|
"loss": 5.8214,
|
|
"mean_token_accuracy": 0.13631626442074776,
|
|
"num_tokens": 3482946.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.027885484695434,
|
|
"epoch": 1.5980837115481594,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004952467846660243,
|
|
"loss": 5.868,
|
|
"mean_token_accuracy": 0.1314691998064518,
|
|
"num_tokens": 3493930.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.154941272735596,
|
|
"epoch": 1.6031265758951085,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004951653395550423,
|
|
"loss": 5.9423,
|
|
"mean_token_accuracy": 0.13613177239894866,
|
|
"num_tokens": 3506139.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.081747388839721,
|
|
"epoch": 1.6081694402420574,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004950832101651062,
|
|
"loss": 5.8018,
|
|
"mean_token_accuracy": 0.13648922890424728,
|
|
"num_tokens": 3516844.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.068617916107177,
|
|
"epoch": 1.6132123045890066,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004950003967514762,
|
|
"loss": 5.8668,
|
|
"mean_token_accuracy": 0.13571436777710916,
|
|
"num_tokens": 3526815.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.036658430099488,
|
|
"epoch": 1.6182551689359557,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000494916899571538,
|
|
"loss": 5.8299,
|
|
"mean_token_accuracy": 0.13820403814315796,
|
|
"num_tokens": 3538041.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.102004766464233,
|
|
"epoch": 1.6232980332829046,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004948327188848029,
|
|
"loss": 5.8495,
|
|
"mean_token_accuracy": 0.1341189369559288,
|
|
"num_tokens": 3549359.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.0633875846862795,
|
|
"epoch": 1.6283408976298537,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004947478549529065,
|
|
"loss": 5.8386,
|
|
"mean_token_accuracy": 0.13996319174766542,
|
|
"num_tokens": 3561902.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.0684449672698975,
|
|
"epoch": 1.6333837619768028,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004946623080396077,
|
|
"loss": 5.8825,
|
|
"mean_token_accuracy": 0.13902646601200103,
|
|
"num_tokens": 3572584.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.17164568901062,
|
|
"epoch": 1.638426626323752,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004945760784107884,
|
|
"loss": 5.8355,
|
|
"mean_token_accuracy": 0.1415456347167492,
|
|
"num_tokens": 3582635.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.022398376464844,
|
|
"epoch": 1.643469490670701,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004944891663344522,
|
|
"loss": 5.8669,
|
|
"mean_token_accuracy": 0.1371048003435135,
|
|
"num_tokens": 3594140.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.008963584899902,
|
|
"epoch": 1.6485123550176501,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004944015720807239,
|
|
"loss": 5.7912,
|
|
"mean_token_accuracy": 0.1387906551361084,
|
|
"num_tokens": 3605748.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.025937128067016,
|
|
"epoch": 1.653555219364599,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004943132959218484,
|
|
"loss": 5.9483,
|
|
"mean_token_accuracy": 0.13064202591776847,
|
|
"num_tokens": 3618304.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.2111304759979244,
|
|
"epoch": 1.6585980837115482,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00049422433813219,
|
|
"loss": 5.8929,
|
|
"mean_token_accuracy": 0.13720419406890869,
|
|
"num_tokens": 3630708.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.045804977416992,
|
|
"epoch": 1.6636409480584973,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004941346989882318,
|
|
"loss": 5.8532,
|
|
"mean_token_accuracy": 0.12697262912988663,
|
|
"num_tokens": 3643188.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.9926165580749515,
|
|
"epoch": 1.6686838124054462,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004940443787685739,
|
|
"loss": 5.7436,
|
|
"mean_token_accuracy": 0.14092481806874274,
|
|
"num_tokens": 3653491.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.0171300888061525,
|
|
"epoch": 1.6737266767523953,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004939533777539341,
|
|
"loss": 5.7437,
|
|
"mean_token_accuracy": 0.1434481054544449,
|
|
"num_tokens": 3663695.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.129909181594849,
|
|
"epoch": 1.6787695410993444,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004938616962271453,
|
|
"loss": 5.9507,
|
|
"mean_token_accuracy": 0.13342082649469375,
|
|
"num_tokens": 3674093.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.03012490272522,
|
|
"epoch": 1.6838124054462935,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004937693344731561,
|
|
"loss": 5.8814,
|
|
"mean_token_accuracy": 0.13591884672641755,
|
|
"num_tokens": 3685362.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.124035739898682,
|
|
"epoch": 1.6888552697932426,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004936762927790286,
|
|
"loss": 5.8948,
|
|
"mean_token_accuracy": 0.13367810025811194,
|
|
"num_tokens": 3695963.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.061925315856934,
|
|
"epoch": 1.6938981341401917,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004935825714339389,
|
|
"loss": 5.8475,
|
|
"mean_token_accuracy": 0.1323956087231636,
|
|
"num_tokens": 3706805.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.1440023422241214,
|
|
"epoch": 1.6989409984871406,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004934881707291748,
|
|
"loss": 5.9564,
|
|
"mean_token_accuracy": 0.1275752753019333,
|
|
"num_tokens": 3718822.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.037684059143066,
|
|
"epoch": 1.7039838628340898,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004933930909581362,
|
|
"loss": 5.9123,
|
|
"mean_token_accuracy": 0.1372961588203907,
|
|
"num_tokens": 3730170.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.153582143783569,
|
|
"epoch": 1.7090267271810389,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004932973324163328,
|
|
"loss": 5.8998,
|
|
"mean_token_accuracy": 0.1329277604818344,
|
|
"num_tokens": 3741293.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.102115774154663,
|
|
"epoch": 1.7140695915279878,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004932008954013846,
|
|
"loss": 5.8938,
|
|
"mean_token_accuracy": 0.12907072007656098,
|
|
"num_tokens": 3753521.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.104592990875244,
|
|
"epoch": 1.7191124558749369,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004931037802130201,
|
|
"loss": 5.8856,
|
|
"mean_token_accuracy": 0.13684083744883538,
|
|
"num_tokens": 3765767.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.118759346008301,
|
|
"epoch": 1.724155320221886,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004930059871530755,
|
|
"loss": 5.8635,
|
|
"mean_token_accuracy": 0.1346114307641983,
|
|
"num_tokens": 3776330.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.085878133773804,
|
|
"epoch": 1.7291981845688351,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004929075165254938,
|
|
"loss": 5.8594,
|
|
"mean_token_accuracy": 0.1392315737903118,
|
|
"num_tokens": 3786937.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.036421394348144,
|
|
"epoch": 1.7342410489157842,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004928083686363241,
|
|
"loss": 5.8928,
|
|
"mean_token_accuracy": 0.1338902086019516,
|
|
"num_tokens": 3798546.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.0134021759033205,
|
|
"epoch": 1.7392839132627333,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004927085437937204,
|
|
"loss": 5.827,
|
|
"mean_token_accuracy": 0.13127774894237518,
|
|
"num_tokens": 3809893.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.0413683414459225,
|
|
"epoch": 1.7443267776096822,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004926080423079404,
|
|
"loss": 5.8463,
|
|
"mean_token_accuracy": 0.13201947510242462,
|
|
"num_tokens": 3820570.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.060141515731812,
|
|
"epoch": 1.7493696419566314,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000492506864491345,
|
|
"loss": 5.8983,
|
|
"mean_token_accuracy": 0.134102962911129,
|
|
"num_tokens": 3833048.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.1224559307098385,
|
|
"epoch": 1.7544125063035805,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004924050106583976,
|
|
"loss": 5.848,
|
|
"mean_token_accuracy": 0.13240425437688827,
|
|
"num_tokens": 3844375.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.000351428985596,
|
|
"epoch": 1.7594553706505294,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004923024811256618,
|
|
"loss": 5.7444,
|
|
"mean_token_accuracy": 0.14447227269411086,
|
|
"num_tokens": 3855103.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.04535436630249,
|
|
"epoch": 1.7644982349974785,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000492199276211802,
|
|
"loss": 5.7888,
|
|
"mean_token_accuracy": 0.13837607055902482,
|
|
"num_tokens": 3865303.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.979665994644165,
|
|
"epoch": 1.7695410993444276,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004920953962375813,
|
|
"loss": 5.865,
|
|
"mean_token_accuracy": 0.13156018555164337,
|
|
"num_tokens": 3877344.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.087999963760376,
|
|
"epoch": 1.7745839636913767,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004919908415258612,
|
|
"loss": 5.8169,
|
|
"mean_token_accuracy": 0.1383423887193203,
|
|
"num_tokens": 3887471.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.955911159515381,
|
|
"epoch": 1.7796268280383258,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004918856124016002,
|
|
"loss": 5.7542,
|
|
"mean_token_accuracy": 0.14216471314430237,
|
|
"num_tokens": 3898486.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.015568399429322,
|
|
"epoch": 1.784669692385275,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004917797091918526,
|
|
"loss": 5.8663,
|
|
"mean_token_accuracy": 0.1332813784480095,
|
|
"num_tokens": 3909864.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.135464429855347,
|
|
"epoch": 1.789712556732224,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004916731322257683,
|
|
"loss": 5.9511,
|
|
"mean_token_accuracy": 0.1283704347908497,
|
|
"num_tokens": 3921308.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.080710220336914,
|
|
"epoch": 1.794755421079173,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004915658818345908,
|
|
"loss": 5.7877,
|
|
"mean_token_accuracy": 0.14001286923885345,
|
|
"num_tokens": 3931469.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.9761453628540036,
|
|
"epoch": 1.799798285426122,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004914579583516568,
|
|
"loss": 5.8894,
|
|
"mean_token_accuracy": 0.12953986525535582,
|
|
"num_tokens": 3943348.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.1129004001617435,
|
|
"epoch": 1.804841149773071,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000491349362112395,
|
|
"loss": 5.7944,
|
|
"mean_token_accuracy": 0.13509925082325935,
|
|
"num_tokens": 3954491.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.962020349502564,
|
|
"epoch": 1.80988401412002,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000491240093454325,
|
|
"loss": 5.7359,
|
|
"mean_token_accuracy": 0.13744344264268876,
|
|
"num_tokens": 3965151.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.044571113586426,
|
|
"epoch": 1.8149268784669692,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004911301527170563,
|
|
"loss": 5.836,
|
|
"mean_token_accuracy": 0.13276761174201965,
|
|
"num_tokens": 3975482.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.018252897262573,
|
|
"epoch": 1.8199697428139183,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004910195402422873,
|
|
"loss": 5.8106,
|
|
"mean_token_accuracy": 0.1397517703473568,
|
|
"num_tokens": 3985295.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.990886116027832,
|
|
"epoch": 1.8250126071608674,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004909082563738042,
|
|
"loss": 5.8606,
|
|
"mean_token_accuracy": 0.13000920787453651,
|
|
"num_tokens": 3997058.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.049672079086304,
|
|
"epoch": 1.8300554715078166,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004907963014574797,
|
|
"loss": 5.8038,
|
|
"mean_token_accuracy": 0.13595991358160972,
|
|
"num_tokens": 4008224.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.9268725395202635,
|
|
"epoch": 1.8350983358547657,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004906836758412722,
|
|
"loss": 5.7439,
|
|
"mean_token_accuracy": 0.13537646234035491,
|
|
"num_tokens": 4019264.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.027598571777344,
|
|
"epoch": 1.8401412002017146,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004905703798752248,
|
|
"loss": 5.8024,
|
|
"mean_token_accuracy": 0.1361069694161415,
|
|
"num_tokens": 4031974.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.995775747299194,
|
|
"epoch": 1.8451840645486637,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000490456413911464,
|
|
"loss": 5.8282,
|
|
"mean_token_accuracy": 0.1380675122141838,
|
|
"num_tokens": 4042023.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.093371391296387,
|
|
"epoch": 1.8502269288956126,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004903417783041987,
|
|
"loss": 5.8488,
|
|
"mean_token_accuracy": 0.13388717770576478,
|
|
"num_tokens": 4053488.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.997908878326416,
|
|
"epoch": 1.8552697932425617,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004902264734097188,
|
|
"loss": 5.7,
|
|
"mean_token_accuracy": 0.1459822252392769,
|
|
"num_tokens": 4063576.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.934653806686401,
|
|
"epoch": 1.8603126575895108,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004901104995863948,
|
|
"loss": 5.7515,
|
|
"mean_token_accuracy": 0.13665405064821243,
|
|
"num_tokens": 4073767.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.051873254776001,
|
|
"epoch": 1.86535552193646,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004899938571946759,
|
|
"loss": 5.8895,
|
|
"mean_token_accuracy": 0.13628361895680427,
|
|
"num_tokens": 4086112.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.005042839050293,
|
|
"epoch": 1.870398386283409,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004898765465970894,
|
|
"loss": 5.7576,
|
|
"mean_token_accuracy": 0.13836311623454095,
|
|
"num_tokens": 4097172.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.969844341278076,
|
|
"epoch": 1.8754412506303582,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004897585681582393,
|
|
"loss": 5.7747,
|
|
"mean_token_accuracy": 0.13562445044517518,
|
|
"num_tokens": 4108034.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.010134601593018,
|
|
"epoch": 1.8804841149773073,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004896399222448053,
|
|
"loss": 5.9158,
|
|
"mean_token_accuracy": 0.13355276435613633,
|
|
"num_tokens": 4118787.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.114977502822876,
|
|
"epoch": 1.8855269793242562,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004895206092255417,
|
|
"loss": 5.855,
|
|
"mean_token_accuracy": 0.13844020664691925,
|
|
"num_tokens": 4129427.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.920868730545044,
|
|
"epoch": 1.8905698436712053,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004894006294712761,
|
|
"loss": 5.746,
|
|
"mean_token_accuracy": 0.140569781512022,
|
|
"num_tokens": 4139750.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.022733497619629,
|
|
"epoch": 1.8956127080181542,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004892799833549083,
|
|
"loss": 5.8036,
|
|
"mean_token_accuracy": 0.13444036543369292,
|
|
"num_tokens": 4150966.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.029891872406006,
|
|
"epoch": 1.9006555723651033,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004891586712514091,
|
|
"loss": 5.8361,
|
|
"mean_token_accuracy": 0.13739247769117355,
|
|
"num_tokens": 4162002.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.039386367797851,
|
|
"epoch": 1.9056984367120524,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004890366935378196,
|
|
"loss": 5.8781,
|
|
"mean_token_accuracy": 0.1344475492835045,
|
|
"num_tokens": 4173529.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.00244288444519,
|
|
"epoch": 1.9107413010590015,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004889140505932492,
|
|
"loss": 5.888,
|
|
"mean_token_accuracy": 0.13375614508986472,
|
|
"num_tokens": 4184743.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.995673084259034,
|
|
"epoch": 1.9157841654059506,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004887907427988749,
|
|
"loss": 5.7513,
|
|
"mean_token_accuracy": 0.1396537348628044,
|
|
"num_tokens": 4195484.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.967873668670654,
|
|
"epoch": 1.9208270297528998,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004886667705379403,
|
|
"loss": 5.8457,
|
|
"mean_token_accuracy": 0.1371358223259449,
|
|
"num_tokens": 4207564.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.952332258224487,
|
|
"epoch": 1.9258698940998489,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004885421341957541,
|
|
"loss": 5.7435,
|
|
"mean_token_accuracy": 0.13553724884986879,
|
|
"num_tokens": 4219165.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.005974435806275,
|
|
"epoch": 1.9309127584467978,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004884168341596888,
|
|
"loss": 5.8068,
|
|
"mean_token_accuracy": 0.13346681967377663,
|
|
"num_tokens": 4229444.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.964589071273804,
|
|
"epoch": 1.9359556227937469,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004882908708191797,
|
|
"loss": 5.7582,
|
|
"mean_token_accuracy": 0.13898043781518937,
|
|
"num_tokens": 4240682.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.1030495166778564,
|
|
"epoch": 1.9409984871406958,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048816424456572403,
|
|
"loss": 6.0139,
|
|
"mean_token_accuracy": 0.1314317248761654,
|
|
"num_tokens": 4253469.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.1366808891296385,
|
|
"epoch": 1.946041351487645,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048803695579287905,
|
|
"loss": 5.8504,
|
|
"mean_token_accuracy": 0.13721998855471612,
|
|
"num_tokens": 4264173.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.989302253723144,
|
|
"epoch": 1.951084215834594,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00048790900489626104,
|
|
"loss": 5.8217,
|
|
"mean_token_accuracy": 0.1348733127117157,
|
|
"num_tokens": 4275748.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.019631099700928,
|
|
"epoch": 1.9561270801815431,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004877803922735445,
|
|
"loss": 5.7563,
|
|
"mean_token_accuracy": 0.13681667894124985,
|
|
"num_tokens": 4286624.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.970220470428467,
|
|
"epoch": 1.9611699445284922,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004876511183244603,
|
|
"loss": 5.7783,
|
|
"mean_token_accuracy": 0.13412140533328057,
|
|
"num_tokens": 4297722.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.989438676834107,
|
|
"epoch": 1.9662128088754414,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048752118345079496,
|
|
"loss": 5.819,
|
|
"mean_token_accuracy": 0.1387975536286831,
|
|
"num_tokens": 4307718.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.997930192947388,
|
|
"epoch": 1.9712556732223905,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048739058805638905,
|
|
"loss": 5.8094,
|
|
"mean_token_accuracy": 0.13116757497191428,
|
|
"num_tokens": 4320384.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.005349445343017,
|
|
"epoch": 1.9762985375693394,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000487259332547136,
|
|
"loss": 5.7605,
|
|
"mean_token_accuracy": 0.1417269617319107,
|
|
"num_tokens": 4331300.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.93936595916748,
|
|
"epoch": 1.9813414019162885,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000487127417330981,
|
|
"loss": 5.7655,
|
|
"mean_token_accuracy": 0.13874521404504775,
|
|
"num_tokens": 4341678.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.00667085647583,
|
|
"epoch": 1.9863842662632374,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004869948428179196,
|
|
"loss": 5.8158,
|
|
"mean_token_accuracy": 0.1337434582412243,
|
|
"num_tokens": 4352924.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.990743732452392,
|
|
"epoch": 1.9914271306101865,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004868616094199964,
|
|
"loss": 5.7495,
|
|
"mean_token_accuracy": 0.1391561046242714,
|
|
"num_tokens": 4363133.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.010904788970947,
|
|
"epoch": 1.9964699949571356,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000486727717551304,
|
|
"loss": 5.8117,
|
|
"mean_token_accuracy": 0.13316915854811667,
|
|
"num_tokens": 4376024.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.940850469801161,
|
|
"epoch": 2.00100857286939,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004865931676279813,
|
|
"loss": 5.6394,
|
|
"mean_token_accuracy": 0.14257431609763038,
|
|
"num_tokens": 4385440.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.965998840332031,
|
|
"epoch": 2.006051437216339,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004864579600682125,
|
|
"loss": 5.484,
|
|
"mean_token_accuracy": 0.14857447966933252,
|
|
"num_tokens": 4397208.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.796861696243286,
|
|
"epoch": 2.011094301563288,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048632209529222607,
|
|
"loss": 5.4264,
|
|
"mean_token_accuracy": 0.15789164751768112,
|
|
"num_tokens": 4406863.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.018551921844482,
|
|
"epoch": 2.016137165910237,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048618557372229253,
|
|
"loss": 5.5849,
|
|
"mean_token_accuracy": 0.14688986241817475,
|
|
"num_tokens": 4419069.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.016137165910237,
|
|
"eval_entropy": 5.81801115222863,
|
|
"eval_loss": 5.917318820953369,
|
|
"eval_mean_token_accuracy": 0.13670703228420883,
|
|
"eval_num_tokens": 4419069.0,
|
|
"eval_runtime": 2.0283,
|
|
"eval_samples_per_second": 1547.125,
|
|
"eval_steps_per_second": 193.76,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.943663930892944,
|
|
"epoch": 2.0211800302571863,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004860483957827243,
|
|
"loss": 5.5384,
|
|
"mean_token_accuracy": 0.15685443133115767,
|
|
"num_tokens": 4432237.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.806105899810791,
|
|
"epoch": 2.026222894604135,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004859105618998735,
|
|
"loss": 5.4086,
|
|
"mean_token_accuracy": 0.1614102154970169,
|
|
"num_tokens": 4441256.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.956412267684937,
|
|
"epoch": 2.031265758951084,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004857720725021311,
|
|
"loss": 5.5642,
|
|
"mean_token_accuracy": 0.15119071453809738,
|
|
"num_tokens": 4452509.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.99310998916626,
|
|
"epoch": 2.036308623298033,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048563292801992534,
|
|
"loss": 5.5261,
|
|
"mean_token_accuracy": 0.1501511499285698,
|
|
"num_tokens": 4464819.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.83888840675354,
|
|
"epoch": 2.0413514876449823,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004854931288857203,
|
|
"loss": 5.4698,
|
|
"mean_token_accuracy": 0.15105220377445222,
|
|
"num_tokens": 4474867.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.885752773284912,
|
|
"epoch": 2.0463943519919314,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048535267553401505,
|
|
"loss": 5.4419,
|
|
"mean_token_accuracy": 0.15685297697782516,
|
|
"num_tokens": 4485229.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.856101560592651,
|
|
"epoch": 2.0514372163388805,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048521156840134193,
|
|
"loss": 5.4469,
|
|
"mean_token_accuracy": 0.15147644877433777,
|
|
"num_tokens": 4496589.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.864161205291748,
|
|
"epoch": 2.0564800806858297,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000485069807926265,
|
|
"loss": 5.5305,
|
|
"mean_token_accuracy": 0.15417610704898835,
|
|
"num_tokens": 4508179.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.930064678192139,
|
|
"epoch": 2.0615229450327788,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004849273945493792,
|
|
"loss": 5.5212,
|
|
"mean_token_accuracy": 0.1541721999645233,
|
|
"num_tokens": 4518796.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.883697271347046,
|
|
"epoch": 2.066565809379728,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004847843287133084,
|
|
"loss": 5.5185,
|
|
"mean_token_accuracy": 0.1513978511095047,
|
|
"num_tokens": 4530544.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.884379053115845,
|
|
"epoch": 2.0716086737266766,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004846406108627048,
|
|
"loss": 5.4441,
|
|
"mean_token_accuracy": 0.15471192747354506,
|
|
"num_tokens": 4541382.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.9226783275604244,
|
|
"epoch": 2.0766515380736257,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048449624144424665,
|
|
"loss": 5.623,
|
|
"mean_token_accuracy": 0.1488088011741638,
|
|
"num_tokens": 4552318.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.879350566864014,
|
|
"epoch": 2.081694402420575,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048435122090663744,
|
|
"loss": 5.4473,
|
|
"mean_token_accuracy": 0.15576326847076416,
|
|
"num_tokens": 4562746.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.873903560638428,
|
|
"epoch": 2.086737266767524,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048420554970060424,
|
|
"loss": 5.5283,
|
|
"mean_token_accuracy": 0.14986108988523483,
|
|
"num_tokens": 4574635.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.898341703414917,
|
|
"epoch": 2.091780131114473,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048405922827889677,
|
|
"loss": 5.5719,
|
|
"mean_token_accuracy": 0.1513387531042099,
|
|
"num_tokens": 4584735.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.915944051742554,
|
|
"epoch": 2.096822995461422,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048391225709628515,
|
|
"loss": 5.4703,
|
|
"mean_token_accuracy": 0.14948942512273788,
|
|
"num_tokens": 4596090.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.827583885192871,
|
|
"epoch": 2.1018658598083713,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004837646366095593,
|
|
"loss": 5.56,
|
|
"mean_token_accuracy": 0.15303790420293809,
|
|
"num_tokens": 4607692.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.810319614410401,
|
|
"epoch": 2.1069087241553204,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048361636727752716,
|
|
"loss": 5.4394,
|
|
"mean_token_accuracy": 0.15243404358625412,
|
|
"num_tokens": 4618319.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.902970266342163,
|
|
"epoch": 2.1119515885022695,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004834674495610131,
|
|
"loss": 5.5708,
|
|
"mean_token_accuracy": 0.15355473160743713,
|
|
"num_tokens": 4628650.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.942028570175171,
|
|
"epoch": 2.116994452849218,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048331788392285693,
|
|
"loss": 5.5637,
|
|
"mean_token_accuracy": 0.15165936648845674,
|
|
"num_tokens": 4639883.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.823600196838379,
|
|
"epoch": 2.1220373171961673,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048316767082791205,
|
|
"loss": 5.4434,
|
|
"mean_token_accuracy": 0.15266976952552797,
|
|
"num_tokens": 4650041.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.850564002990723,
|
|
"epoch": 2.1270801815431164,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048301681074304427,
|
|
"loss": 5.5098,
|
|
"mean_token_accuracy": 0.14926768094301224,
|
|
"num_tokens": 4661226.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.882189083099365,
|
|
"epoch": 2.1321230458900655,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048286530413713016,
|
|
"loss": 5.5996,
|
|
"mean_token_accuracy": 0.14671720415353776,
|
|
"num_tokens": 4672658.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.835665464401245,
|
|
"epoch": 2.1371659102370146,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048271315148105574,
|
|
"loss": 5.4562,
|
|
"mean_token_accuracy": 0.15615704357624055,
|
|
"num_tokens": 4683214.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.7868915557861325,
|
|
"epoch": 2.1422087745839637,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000482560353247715,
|
|
"loss": 5.4983,
|
|
"mean_token_accuracy": 0.1570190817117691,
|
|
"num_tokens": 4694787.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.894145154953003,
|
|
"epoch": 2.147251638930913,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004824069099120084,
|
|
"loss": 5.4587,
|
|
"mean_token_accuracy": 0.15608564168214797,
|
|
"num_tokens": 4705827.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.781821918487549,
|
|
"epoch": 2.152294503277862,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004822528219508413,
|
|
"loss": 5.5101,
|
|
"mean_token_accuracy": 0.1537075772881508,
|
|
"num_tokens": 4717197.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.969233179092408,
|
|
"epoch": 2.157337367624811,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004820980898431227,
|
|
"loss": 5.6358,
|
|
"mean_token_accuracy": 0.15112573504447938,
|
|
"num_tokens": 4729000.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.8603473663330075,
|
|
"epoch": 2.1623802319717598,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004819427140697635,
|
|
"loss": 5.523,
|
|
"mean_token_accuracy": 0.14781847298145295,
|
|
"num_tokens": 4739999.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.883710241317749,
|
|
"epoch": 2.167423096318709,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004817866951136753,
|
|
"loss": 5.5138,
|
|
"mean_token_accuracy": 0.1504565268754959,
|
|
"num_tokens": 4752461.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.909519195556641,
|
|
"epoch": 2.172465960665658,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048163003345976847,
|
|
"loss": 5.5642,
|
|
"mean_token_accuracy": 0.1481477752327919,
|
|
"num_tokens": 4763114.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.875637865066528,
|
|
"epoch": 2.177508825012607,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004814727295949512,
|
|
"loss": 5.5716,
|
|
"mean_token_accuracy": 0.14529902637004852,
|
|
"num_tokens": 4773926.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.878474426269531,
|
|
"epoch": 2.1825516893595562,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048131478400812735,
|
|
"loss": 5.5022,
|
|
"mean_token_accuracy": 0.15807463079690934,
|
|
"num_tokens": 4784062.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.951031494140625,
|
|
"epoch": 2.1875945537065054,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004811561971901956,
|
|
"loss": 5.4972,
|
|
"mean_token_accuracy": 0.1530397966504097,
|
|
"num_tokens": 4794763.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.868496370315552,
|
|
"epoch": 2.1926374180534545,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004809969696340474,
|
|
"loss": 5.5282,
|
|
"mean_token_accuracy": 0.15282675623893738,
|
|
"num_tokens": 4804317.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.7768796443939205,
|
|
"epoch": 2.1976802824004036,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048083710183456554,
|
|
"loss": 5.5048,
|
|
"mean_token_accuracy": 0.15021117329597472,
|
|
"num_tokens": 4816229.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.844064044952392,
|
|
"epoch": 2.2027231467473527,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048067659428862306,
|
|
"loss": 5.5178,
|
|
"mean_token_accuracy": 0.16275418549776077,
|
|
"num_tokens": 4827317.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.868047475814819,
|
|
"epoch": 2.2077660110943014,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048051544749508096,
|
|
"loss": 5.5199,
|
|
"mean_token_accuracy": 0.14801972061395646,
|
|
"num_tokens": 4838519.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.797014331817627,
|
|
"epoch": 2.2128088754412505,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004803536619547872,
|
|
"loss": 5.5181,
|
|
"mean_token_accuracy": 0.15277098268270492,
|
|
"num_tokens": 4848916.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.953539609909058,
|
|
"epoch": 2.2178517397881996,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000480191238170575,
|
|
"loss": 5.6006,
|
|
"mean_token_accuracy": 0.14733142107725145,
|
|
"num_tokens": 4860977.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.769330167770386,
|
|
"epoch": 2.2228946041351487,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004800281766472612,
|
|
"loss": 5.3685,
|
|
"mean_token_accuracy": 0.16047884821891784,
|
|
"num_tokens": 4871332.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.88454384803772,
|
|
"epoch": 2.227937468482098,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047986447789164486,
|
|
"loss": 5.581,
|
|
"mean_token_accuracy": 0.14643221944570542,
|
|
"num_tokens": 4883139.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.944646644592285,
|
|
"epoch": 2.232980332829047,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004797001424125054,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.1498621180653572,
|
|
"num_tokens": 4894892.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.8559250831604,
|
|
"epoch": 2.238023197175996,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047953517072060124,
|
|
"loss": 5.5692,
|
|
"mean_token_accuracy": 0.1519879698753357,
|
|
"num_tokens": 4906059.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.891297197341919,
|
|
"epoch": 2.243066061522945,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004793695633286684,
|
|
"loss": 5.5214,
|
|
"mean_token_accuracy": 0.15420397147536277,
|
|
"num_tokens": 4917527.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.8056127548217775,
|
|
"epoch": 2.2481089258698943,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047920332075141825,
|
|
"loss": 5.4864,
|
|
"mean_token_accuracy": 0.1573357716202736,
|
|
"num_tokens": 4928502.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.855831098556519,
|
|
"epoch": 2.253151790216843,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004790364435055367,
|
|
"loss": 5.4629,
|
|
"mean_token_accuracy": 0.15087850987911225,
|
|
"num_tokens": 4938179.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.902001523971558,
|
|
"epoch": 2.258194654563792,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000478868932109682,
|
|
"loss": 5.5543,
|
|
"mean_token_accuracy": 0.14623372554779052,
|
|
"num_tokens": 4948979.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.75980167388916,
|
|
"epoch": 2.263237518910741,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004787007870844835,
|
|
"loss": 5.4403,
|
|
"mean_token_accuracy": 0.15094903409481047,
|
|
"num_tokens": 4958459.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.93130464553833,
|
|
"epoch": 2.2682803832576903,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047853200895253976,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.14569308310747148,
|
|
"num_tokens": 4970680.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.8348475933074955,
|
|
"epoch": 2.2733232476046394,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047836259823841716,
|
|
"loss": 5.5944,
|
|
"mean_token_accuracy": 0.15193736106157302,
|
|
"num_tokens": 4981635.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.810261869430542,
|
|
"epoch": 2.2783661119515886,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047819255546864797,
|
|
"loss": 5.5199,
|
|
"mean_token_accuracy": 0.1543443039059639,
|
|
"num_tokens": 4992661.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.896259546279907,
|
|
"epoch": 2.2834089762985377,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004780218811717293,
|
|
"loss": 5.5618,
|
|
"mean_token_accuracy": 0.15468370020389557,
|
|
"num_tokens": 5004364.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.841471099853516,
|
|
"epoch": 2.288451840645487,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047785057587812057,
|
|
"loss": 5.5862,
|
|
"mean_token_accuracy": 0.15330396443605424,
|
|
"num_tokens": 5015575.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.87632269859314,
|
|
"epoch": 2.293494704992436,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047767864012024266,
|
|
"loss": 5.5053,
|
|
"mean_token_accuracy": 0.15744749456644058,
|
|
"num_tokens": 5026984.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.802755498886109,
|
|
"epoch": 2.2985375693393846,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047750607443247585,
|
|
"loss": 5.5077,
|
|
"mean_token_accuracy": 0.146403731405735,
|
|
"num_tokens": 5039835.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.824838447570801,
|
|
"epoch": 2.3035804336863337,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047733287935115837,
|
|
"loss": 5.5635,
|
|
"mean_token_accuracy": 0.14812358915805818,
|
|
"num_tokens": 5052284.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.830850887298584,
|
|
"epoch": 2.308623298033283,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004771590554145844,
|
|
"loss": 5.4697,
|
|
"mean_token_accuracy": 0.15186792463064194,
|
|
"num_tokens": 5062535.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.804275608062744,
|
|
"epoch": 2.313666162380232,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047698460316300285,
|
|
"loss": 5.5792,
|
|
"mean_token_accuracy": 0.15214556753635405,
|
|
"num_tokens": 5074065.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.76812596321106,
|
|
"epoch": 2.318709026727181,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004768095231386153,
|
|
"loss": 5.4481,
|
|
"mean_token_accuracy": 0.1508024200797081,
|
|
"num_tokens": 5085269.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.8601611137390135,
|
|
"epoch": 2.32375189107413,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004766338158855744,
|
|
"loss": 5.5466,
|
|
"mean_token_accuracy": 0.1513417288661003,
|
|
"num_tokens": 5096537.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.889308166503906,
|
|
"epoch": 2.3287947554210793,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004764574819499826,
|
|
"loss": 5.5617,
|
|
"mean_token_accuracy": 0.14688292294740676,
|
|
"num_tokens": 5109554.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.834060573577881,
|
|
"epoch": 2.3338376197680284,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047628052187988955,
|
|
"loss": 5.5214,
|
|
"mean_token_accuracy": 0.15323757827281953,
|
|
"num_tokens": 5119740.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.76870756149292,
|
|
"epoch": 2.338880484114977,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004761029362252914,
|
|
"loss": 5.5789,
|
|
"mean_token_accuracy": 0.14972641170024872,
|
|
"num_tokens": 5130682.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.866538429260254,
|
|
"epoch": 2.343923348461926,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047592472553812856,
|
|
"loss": 5.4855,
|
|
"mean_token_accuracy": 0.1567153364419937,
|
|
"num_tokens": 5141290.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.753556871414185,
|
|
"epoch": 2.3489662128088753,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004757458903722837,
|
|
"loss": 5.4612,
|
|
"mean_token_accuracy": 0.15005119442939757,
|
|
"num_tokens": 5152930.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.791133117675781,
|
|
"epoch": 2.3540090771558244,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047556643128358075,
|
|
"loss": 5.508,
|
|
"mean_token_accuracy": 0.153596168756485,
|
|
"num_tokens": 5164031.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.792740106582642,
|
|
"epoch": 2.3590519415027735,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047538634882978277,
|
|
"loss": 5.499,
|
|
"mean_token_accuracy": 0.15123484134674073,
|
|
"num_tokens": 5174833.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.781650400161743,
|
|
"epoch": 2.3640948058497226,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047520564357059013,
|
|
"loss": 5.4348,
|
|
"mean_token_accuracy": 0.15593076795339583,
|
|
"num_tokens": 5184020.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.7946364402771,
|
|
"epoch": 2.3691376701966718,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004750243160676389,
|
|
"loss": 5.4305,
|
|
"mean_token_accuracy": 0.15806339532136918,
|
|
"num_tokens": 5193812.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.909400987625122,
|
|
"epoch": 2.374180534543621,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004748423668844992,
|
|
"loss": 5.6468,
|
|
"mean_token_accuracy": 0.1561293825507164,
|
|
"num_tokens": 5204663.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.850083351135254,
|
|
"epoch": 2.37922339889057,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004746597965866733,
|
|
"loss": 5.4514,
|
|
"mean_token_accuracy": 0.15195969939231874,
|
|
"num_tokens": 5215582.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.868561172485352,
|
|
"epoch": 2.384266263237519,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004744766057415938,
|
|
"loss": 5.5234,
|
|
"mean_token_accuracy": 0.15315358638763427,
|
|
"num_tokens": 5227185.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.7468945503234865,
|
|
"epoch": 2.389309127584468,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004742927949186222,
|
|
"loss": 5.4745,
|
|
"mean_token_accuracy": 0.15835718661546708,
|
|
"num_tokens": 5238800.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.728143310546875,
|
|
"epoch": 2.394351991931417,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047410836468904664,
|
|
"loss": 5.4446,
|
|
"mean_token_accuracy": 0.15594211220741272,
|
|
"num_tokens": 5249149.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.749974918365479,
|
|
"epoch": 2.399394856278366,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047392331562608073,
|
|
"loss": 5.4105,
|
|
"mean_token_accuracy": 0.15610798448324203,
|
|
"num_tokens": 5259753.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.747346830368042,
|
|
"epoch": 2.404437720625315,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004737376483048609,
|
|
"loss": 5.464,
|
|
"mean_token_accuracy": 0.15509508103132247,
|
|
"num_tokens": 5269954.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.855034351348877,
|
|
"epoch": 2.4094805849722642,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004735513633024458,
|
|
"loss": 5.4991,
|
|
"mean_token_accuracy": 0.15720080435276032,
|
|
"num_tokens": 5280519.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.713812160491943,
|
|
"epoch": 2.4145234493192134,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004733644611978134,
|
|
"loss": 5.5227,
|
|
"mean_token_accuracy": 0.15511042177677153,
|
|
"num_tokens": 5291729.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.7760590553283695,
|
|
"epoch": 2.4195663136661625,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004731769425718597,
|
|
"loss": 5.4847,
|
|
"mean_token_accuracy": 0.15621420443058015,
|
|
"num_tokens": 5302277.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.745872354507446,
|
|
"epoch": 2.4246091780131116,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000472988808007397,
|
|
"loss": 5.4534,
|
|
"mean_token_accuracy": 0.15893562734127045,
|
|
"num_tokens": 5314432.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.661859083175659,
|
|
"epoch": 2.4296520423600603,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047280005808915176,
|
|
"loss": 5.4465,
|
|
"mean_token_accuracy": 0.15033914148807526,
|
|
"num_tokens": 5324707.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.794715404510498,
|
|
"epoch": 2.4346949067070094,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004726106934037631,
|
|
"loss": 5.465,
|
|
"mean_token_accuracy": 0.15631193071603774,
|
|
"num_tokens": 5334414.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.789452266693115,
|
|
"epoch": 2.4397377710539585,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004724207145397809,
|
|
"loss": 5.4673,
|
|
"mean_token_accuracy": 0.1548281952738762,
|
|
"num_tokens": 5346372.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.732959461212158,
|
|
"epoch": 2.4447806354009076,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047223012208766384,
|
|
"loss": 5.5026,
|
|
"mean_token_accuracy": 0.15159205645322799,
|
|
"num_tokens": 5356720.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.893084955215454,
|
|
"epoch": 2.4498234997478567,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004720389166397777,
|
|
"loss": 5.5696,
|
|
"mean_token_accuracy": 0.15283028185367584,
|
|
"num_tokens": 5367959.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.750197505950927,
|
|
"epoch": 2.454866364094806,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047184709879039325,
|
|
"loss": 5.4695,
|
|
"mean_token_accuracy": 0.15502368956804274,
|
|
"num_tokens": 5378490.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.638679313659668,
|
|
"epoch": 2.459909228441755,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047165466913568497,
|
|
"loss": 5.423,
|
|
"mean_token_accuracy": 0.15600484013557434,
|
|
"num_tokens": 5389326.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.778469181060791,
|
|
"epoch": 2.464952092788704,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004714616282737285,
|
|
"loss": 5.4753,
|
|
"mean_token_accuracy": 0.152701935172081,
|
|
"num_tokens": 5402289.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.789434337615967,
|
|
"epoch": 2.469994957135653,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004712679768044995,
|
|
"loss": 5.4628,
|
|
"mean_token_accuracy": 0.15229596346616744,
|
|
"num_tokens": 5412487.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.767400121688842,
|
|
"epoch": 2.4750378214826023,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000471073715329871,
|
|
"loss": 5.4818,
|
|
"mean_token_accuracy": 0.1478862538933754,
|
|
"num_tokens": 5423939.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.78984751701355,
|
|
"epoch": 2.4800806858295514,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004708788444536123,
|
|
"loss": 5.54,
|
|
"mean_token_accuracy": 0.1537518873810768,
|
|
"num_tokens": 5434623.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.758418273925781,
|
|
"epoch": 2.4851235501765,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004706833647813865,
|
|
"loss": 5.5085,
|
|
"mean_token_accuracy": 0.15297200083732604,
|
|
"num_tokens": 5445368.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.83281774520874,
|
|
"epoch": 2.4901664145234492,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047048727692074893,
|
|
"loss": 5.4902,
|
|
"mean_token_accuracy": 0.15309383273124694,
|
|
"num_tokens": 5456678.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.736118793487549,
|
|
"epoch": 2.4952092788703983,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004702905814811452,
|
|
"loss": 5.5078,
|
|
"mean_token_accuracy": 0.1514161393046379,
|
|
"num_tokens": 5468181.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.833150863647461,
|
|
"epoch": 2.5002521432173475,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047009327907390926,
|
|
"loss": 5.6341,
|
|
"mean_token_accuracy": 0.14663451313972473,
|
|
"num_tokens": 5479858.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.7969193935394285,
|
|
"epoch": 2.5052950075642966,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004698953703122616,
|
|
"loss": 5.4655,
|
|
"mean_token_accuracy": 0.15067108720541,
|
|
"num_tokens": 5489885.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.73531322479248,
|
|
"epoch": 2.5103378719112457,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046969685581130715,
|
|
"loss": 5.4311,
|
|
"mean_token_accuracy": 0.14690714105963706,
|
|
"num_tokens": 5500597.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.7298362255096436,
|
|
"epoch": 2.515380736258195,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046949773618803353,
|
|
"loss": 5.4247,
|
|
"mean_token_accuracy": 0.15728699415922165,
|
|
"num_tokens": 5513830.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.605797576904297,
|
|
"epoch": 2.5204236006051435,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046929801206130925,
|
|
"loss": 5.5076,
|
|
"mean_token_accuracy": 0.1480040967464447,
|
|
"num_tokens": 5525318.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.753169536590576,
|
|
"epoch": 2.5254664649520926,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046909768405188135,
|
|
"loss": 5.4669,
|
|
"mean_token_accuracy": 0.160452176630497,
|
|
"num_tokens": 5536338.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.790803241729736,
|
|
"epoch": 2.5305093292990417,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00046889675278237404,
|
|
"loss": 5.4771,
|
|
"mean_token_accuracy": 0.15563293546438217,
|
|
"num_tokens": 5550114.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.788385915756225,
|
|
"epoch": 2.535552193645991,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004686952188772863,
|
|
"loss": 5.5352,
|
|
"mean_token_accuracy": 0.14535831958055495,
|
|
"num_tokens": 5562066.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.717385292053223,
|
|
"epoch": 2.54059505799294,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046849308296299005,
|
|
"loss": 5.441,
|
|
"mean_token_accuracy": 0.15885742306709288,
|
|
"num_tokens": 5573177.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.68976149559021,
|
|
"epoch": 2.545637922339889,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004682903456677285,
|
|
"loss": 5.4345,
|
|
"mean_token_accuracy": 0.15550419092178344,
|
|
"num_tokens": 5583612.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.744239616394043,
|
|
"epoch": 2.550680786686838,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046808700762161385,
|
|
"loss": 5.4654,
|
|
"mean_token_accuracy": 0.15590668618679046,
|
|
"num_tokens": 5594188.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.73140230178833,
|
|
"epoch": 2.5557236510337873,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004678830694566253,
|
|
"loss": 5.4989,
|
|
"mean_token_accuracy": 0.15542927384376526,
|
|
"num_tokens": 5604779.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.738622331619263,
|
|
"epoch": 2.5607665153807364,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004676785318066075,
|
|
"loss": 5.5696,
|
|
"mean_token_accuracy": 0.15190335214138032,
|
|
"num_tokens": 5615769.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.814460611343383,
|
|
"epoch": 2.5658093797276855,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004674733953072682,
|
|
"loss": 5.4405,
|
|
"mean_token_accuracy": 0.1664976730942726,
|
|
"num_tokens": 5626402.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.641471862792969,
|
|
"epoch": 2.5708522440746346,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004672676605961763,
|
|
"loss": 5.4143,
|
|
"mean_token_accuracy": 0.15526661574840545,
|
|
"num_tokens": 5637001.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.8195109367370605,
|
|
"epoch": 2.5758951084215833,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046706132831276007,
|
|
"loss": 5.5429,
|
|
"mean_token_accuracy": 0.15143456161022187,
|
|
"num_tokens": 5649265.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.7367987632751465,
|
|
"epoch": 2.5809379727685324,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046685439909830486,
|
|
"loss": 5.47,
|
|
"mean_token_accuracy": 0.156496499478817,
|
|
"num_tokens": 5660755.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.706658458709716,
|
|
"epoch": 2.5859808371154815,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004666468735959516,
|
|
"loss": 5.4368,
|
|
"mean_token_accuracy": 0.1559344068169594,
|
|
"num_tokens": 5673084.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.739587211608887,
|
|
"epoch": 2.5910237014624307,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004664387524506943,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.15186764299869537,
|
|
"num_tokens": 5683524.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.703453207015992,
|
|
"epoch": 2.59606656580938,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004662300363093784,
|
|
"loss": 5.4108,
|
|
"mean_token_accuracy": 0.1567538633942604,
|
|
"num_tokens": 5693847.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.737986516952515,
|
|
"epoch": 2.601109430156329,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004660207258206983,
|
|
"loss": 5.505,
|
|
"mean_token_accuracy": 0.15696525871753692,
|
|
"num_tokens": 5704915.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.7901694774627686,
|
|
"epoch": 2.606152294503278,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00046581082163519585,
|
|
"loss": 5.5535,
|
|
"mean_token_accuracy": 0.1479012444615364,
|
|
"num_tokens": 5716783.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.745092058181763,
|
|
"epoch": 2.6111951588502267,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004656003244052582,
|
|
"loss": 5.5246,
|
|
"mean_token_accuracy": 0.15058010220527648,
|
|
"num_tokens": 5728993.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.761765193939209,
|
|
"epoch": 2.616238023197176,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004653892347851155,
|
|
"loss": 5.5068,
|
|
"mean_token_accuracy": 0.1499432608485222,
|
|
"num_tokens": 5740340.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.774857854843139,
|
|
"epoch": 2.621280887544125,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004651775534308392,
|
|
"loss": 5.4984,
|
|
"mean_token_accuracy": 0.15435360968112946,
|
|
"num_tokens": 5751545.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.687238502502441,
|
|
"epoch": 2.626323751891074,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004649652810003398,
|
|
"loss": 5.4287,
|
|
"mean_token_accuracy": 0.15567395687103272,
|
|
"num_tokens": 5761604.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.688789033889771,
|
|
"epoch": 2.631366616238023,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046475241815336504,
|
|
"loss": 5.4446,
|
|
"mean_token_accuracy": 0.15180650651454924,
|
|
"num_tokens": 5772350.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.771944618225097,
|
|
"epoch": 2.6364094805849723,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004645389655514974,
|
|
"loss": 5.5181,
|
|
"mean_token_accuracy": 0.14686440974473952,
|
|
"num_tokens": 5783675.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.831314373016357,
|
|
"epoch": 2.6414523449319214,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004643249238581525,
|
|
"loss": 5.5434,
|
|
"mean_token_accuracy": 0.15259119123220444,
|
|
"num_tokens": 5794430.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.794641542434692,
|
|
"epoch": 2.6464952092788705,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004641102937385769,
|
|
"loss": 5.5542,
|
|
"mean_token_accuracy": 0.15364048182964324,
|
|
"num_tokens": 5806175.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.70883526802063,
|
|
"epoch": 2.6515380736258196,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004638950758598461,
|
|
"loss": 5.5536,
|
|
"mean_token_accuracy": 0.14425939917564393,
|
|
"num_tokens": 5817410.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.80547342300415,
|
|
"epoch": 2.6565809379727687,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046367927089086203,
|
|
"loss": 5.448,
|
|
"mean_token_accuracy": 0.15607620179653167,
|
|
"num_tokens": 5828912.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.748285531997681,
|
|
"epoch": 2.661623802319718,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046346287950235163,
|
|
"loss": 5.5401,
|
|
"mean_token_accuracy": 0.15586501061916352,
|
|
"num_tokens": 5839258.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.737153673171997,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004632459023668642,
|
|
"loss": 5.5079,
|
|
"mean_token_accuracy": 0.15383340567350387,
|
|
"num_tokens": 5849959.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.763741827011108,
|
|
"epoch": 2.6717095310136156,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004630283401587699,
|
|
"loss": 5.4789,
|
|
"mean_token_accuracy": 0.15776506662368775,
|
|
"num_tokens": 5860426.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.854124212265015,
|
|
"epoch": 2.6767523953605648,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004628101935542568,
|
|
"loss": 5.542,
|
|
"mean_token_accuracy": 0.148340904712677,
|
|
"num_tokens": 5871388.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.688088989257812,
|
|
"epoch": 2.681795259707514,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046259146323132986,
|
|
"loss": 5.4544,
|
|
"mean_token_accuracy": 0.15273037701845169,
|
|
"num_tokens": 5881949.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.818883419036865,
|
|
"epoch": 2.686838124054463,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046237214986980776,
|
|
"loss": 5.558,
|
|
"mean_token_accuracy": 0.14223721921443938,
|
|
"num_tokens": 5893144.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.765694522857666,
|
|
"epoch": 2.691880988401412,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004621522541513216,
|
|
"loss": 5.4284,
|
|
"mean_token_accuracy": 0.15049116015434266,
|
|
"num_tokens": 5903305.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.6631287097930905,
|
|
"epoch": 2.696923852748361,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004619317767593123,
|
|
"loss": 5.5171,
|
|
"mean_token_accuracy": 0.1517879694700241,
|
|
"num_tokens": 5914713.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.818699312210083,
|
|
"epoch": 2.70196671709531,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046171071837902873,
|
|
"loss": 5.4819,
|
|
"mean_token_accuracy": 0.15786724239587785,
|
|
"num_tokens": 5925576.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.696503210067749,
|
|
"epoch": 2.707009581442259,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004614890796975254,
|
|
"loss": 5.4826,
|
|
"mean_token_accuracy": 0.15257023125886918,
|
|
"num_tokens": 5935686.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.718165111541748,
|
|
"epoch": 2.712052445789208,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004612668614036605,
|
|
"loss": 5.4736,
|
|
"mean_token_accuracy": 0.14945560395717622,
|
|
"num_tokens": 5947741.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.7087709426879885,
|
|
"epoch": 2.7170953101361572,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004610440641880934,
|
|
"loss": 5.4841,
|
|
"mean_token_accuracy": 0.15218786150217056,
|
|
"num_tokens": 5958288.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.716747570037842,
|
|
"epoch": 2.7221381744831064,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046082068874328324,
|
|
"loss": 5.4522,
|
|
"mean_token_accuracy": 0.15020226687192917,
|
|
"num_tokens": 5969815.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.726864147186279,
|
|
"epoch": 2.7271810388300555,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046059673576348596,
|
|
"loss": 5.4902,
|
|
"mean_token_accuracy": 0.15824628323316575,
|
|
"num_tokens": 5980795.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.73177137374878,
|
|
"epoch": 2.7322239031770046,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046037220594475264,
|
|
"loss": 5.444,
|
|
"mean_token_accuracy": 0.1508912980556488,
|
|
"num_tokens": 5992284.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.7489784240722654,
|
|
"epoch": 2.7372667675239537,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000460147099984927,
|
|
"loss": 5.4116,
|
|
"mean_token_accuracy": 0.1543903425335884,
|
|
"num_tokens": 6002161.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.680394983291626,
|
|
"epoch": 2.742309631870903,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004599214185836438,
|
|
"loss": 5.3987,
|
|
"mean_token_accuracy": 0.15582872778177262,
|
|
"num_tokens": 6013154.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.742609643936158,
|
|
"epoch": 2.747352496217852,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045969516244232574,
|
|
"loss": 5.5027,
|
|
"mean_token_accuracy": 0.14868493303656577,
|
|
"num_tokens": 6023567.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.779243469238281,
|
|
"epoch": 2.752395360564801,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045946833226418235,
|
|
"loss": 5.5049,
|
|
"mean_token_accuracy": 0.1554177589714527,
|
|
"num_tokens": 6034381.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.7479206085205075,
|
|
"epoch": 2.7574382249117497,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004592409287542069,
|
|
"loss": 5.533,
|
|
"mean_token_accuracy": 0.15534146279096603,
|
|
"num_tokens": 6045899.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.6765319347381595,
|
|
"epoch": 2.762481089258699,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004590129526191748,
|
|
"loss": 5.4588,
|
|
"mean_token_accuracy": 0.16064525693655013,
|
|
"num_tokens": 6056852.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.6774509906768795,
|
|
"epoch": 2.767523953605648,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004587844045676413,
|
|
"loss": 5.4993,
|
|
"mean_token_accuracy": 0.15331972688436507,
|
|
"num_tokens": 6067345.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.847537851333618,
|
|
"epoch": 2.772566817952597,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045855528530993874,
|
|
"loss": 5.5506,
|
|
"mean_token_accuracy": 0.15039721131324768,
|
|
"num_tokens": 6078749.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.746519994735718,
|
|
"epoch": 2.777609682299546,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045832559555817525,
|
|
"loss": 5.42,
|
|
"mean_token_accuracy": 0.15581451654434203,
|
|
"num_tokens": 6089577.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.669173908233643,
|
|
"epoch": 2.7826525466464953,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045809533602623185,
|
|
"loss": 5.4293,
|
|
"mean_token_accuracy": 0.15584837794303893,
|
|
"num_tokens": 6101253.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.8013957977294925,
|
|
"epoch": 2.7876954109934444,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004578645074297603,
|
|
"loss": 5.495,
|
|
"mean_token_accuracy": 0.14859967827796935,
|
|
"num_tokens": 6113342.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.718002796173096,
|
|
"epoch": 2.792738275340393,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045763311048618144,
|
|
"loss": 5.411,
|
|
"mean_token_accuracy": 0.152135494351387,
|
|
"num_tokens": 6124785.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.641060495376587,
|
|
"epoch": 2.797781139687342,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004574011459146822,
|
|
"loss": 5.4503,
|
|
"mean_token_accuracy": 0.15371314883232118,
|
|
"num_tokens": 6137592.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.749881744384766,
|
|
"epoch": 2.8028240040342913,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045716861443621386,
|
|
"loss": 5.4697,
|
|
"mean_token_accuracy": 0.15637175291776656,
|
|
"num_tokens": 6148667.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.770527267456055,
|
|
"epoch": 2.8078668683812404,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045693551677348967,
|
|
"loss": 5.451,
|
|
"mean_token_accuracy": 0.15207348614931107,
|
|
"num_tokens": 6160713.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.779430723190307,
|
|
"epoch": 2.8129097327281896,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004567018536509823,
|
|
"loss": 5.5755,
|
|
"mean_token_accuracy": 0.15166713744401933,
|
|
"num_tokens": 6172272.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.721499395370484,
|
|
"epoch": 2.8179525970751387,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045646762579492245,
|
|
"loss": 5.4229,
|
|
"mean_token_accuracy": 0.14775012731552123,
|
|
"num_tokens": 6183785.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.688457107543945,
|
|
"epoch": 2.822995461422088,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045623283393329546,
|
|
"loss": 5.3775,
|
|
"mean_token_accuracy": 0.15840678364038469,
|
|
"num_tokens": 6194464.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.775582647323608,
|
|
"epoch": 2.828038325769037,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045599747879584,
|
|
"loss": 5.426,
|
|
"mean_token_accuracy": 0.1546274319291115,
|
|
"num_tokens": 6204671.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.640670299530029,
|
|
"epoch": 2.833081190115986,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004557615611140451,
|
|
"loss": 5.4146,
|
|
"mean_token_accuracy": 0.1612096130847931,
|
|
"num_tokens": 6215385.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.630797576904297,
|
|
"epoch": 2.838124054462935,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004555250816211485,
|
|
"loss": 5.357,
|
|
"mean_token_accuracy": 0.15553602576255798,
|
|
"num_tokens": 6225912.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.752071332931519,
|
|
"epoch": 2.8431669188098843,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045528804105213395,
|
|
"loss": 5.4612,
|
|
"mean_token_accuracy": 0.1602766200900078,
|
|
"num_tokens": 6236559.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.713701391220093,
|
|
"epoch": 2.848209783156833,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004550504401437289,
|
|
"loss": 5.4315,
|
|
"mean_token_accuracy": 0.1578310638666153,
|
|
"num_tokens": 6247711.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.672023010253906,
|
|
"epoch": 2.853252647503782,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045481227963440235,
|
|
"loss": 5.5328,
|
|
"mean_token_accuracy": 0.14823448956012725,
|
|
"num_tokens": 6258221.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.652795839309692,
|
|
"epoch": 2.858295511850731,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00045457356026436295,
|
|
"loss": 5.4001,
|
|
"mean_token_accuracy": 0.1583421677350998,
|
|
"num_tokens": 6270839.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.733667516708374,
|
|
"epoch": 2.8633383761976803,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045433428277555577,
|
|
"loss": 5.4762,
|
|
"mean_token_accuracy": 0.15817931145429612,
|
|
"num_tokens": 6281014.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.711543846130371,
|
|
"epoch": 2.8683812405446294,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004540944479116609,
|
|
"loss": 5.3434,
|
|
"mean_token_accuracy": 0.16303679645061492,
|
|
"num_tokens": 6291585.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.662552165985107,
|
|
"epoch": 2.8734241048915785,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004538540564180906,
|
|
"loss": 5.4371,
|
|
"mean_token_accuracy": 0.1538010999560356,
|
|
"num_tokens": 6302474.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.651121282577515,
|
|
"epoch": 2.8784669692385276,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004536131090419871,
|
|
"loss": 5.4358,
|
|
"mean_token_accuracy": 0.1578834153711796,
|
|
"num_tokens": 6314294.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.707853221893311,
|
|
"epoch": 2.8835098335854763,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004533716065322205,
|
|
"loss": 5.4801,
|
|
"mean_token_accuracy": 0.1580565854907036,
|
|
"num_tokens": 6325465.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.680128335952759,
|
|
"epoch": 2.8885526979324254,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004531295496393861,
|
|
"loss": 5.4001,
|
|
"mean_token_accuracy": 0.15744387805461885,
|
|
"num_tokens": 6335010.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.701512670516967,
|
|
"epoch": 2.8935955622793745,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004528869391158023,
|
|
"loss": 5.4527,
|
|
"mean_token_accuracy": 0.1601644814014435,
|
|
"num_tokens": 6345640.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.62476921081543,
|
|
"epoch": 2.8986384266263236,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045264377571550834,
|
|
"loss": 5.4294,
|
|
"mean_token_accuracy": 0.16621643155813218,
|
|
"num_tokens": 6355539.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.619204473495484,
|
|
"epoch": 2.9036812909732728,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004524000601942615,
|
|
"loss": 5.3329,
|
|
"mean_token_accuracy": 0.15699979960918425,
|
|
"num_tokens": 6366542.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.751354694366455,
|
|
"epoch": 2.908724155320222,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004521557933095354,
|
|
"loss": 5.5287,
|
|
"mean_token_accuracy": 0.15396645069122314,
|
|
"num_tokens": 6376735.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.64409499168396,
|
|
"epoch": 2.913767019667171,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000451910975820517,
|
|
"loss": 5.3584,
|
|
"mean_token_accuracy": 0.15932220369577407,
|
|
"num_tokens": 6386663.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.62259087562561,
|
|
"epoch": 2.91880988401412,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00045166560848810476,
|
|
"loss": 5.4581,
|
|
"mean_token_accuracy": 0.1552051857113838,
|
|
"num_tokens": 6397886.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.692402982711792,
|
|
"epoch": 2.9238527483610692,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045141969207490606,
|
|
"loss": 5.4081,
|
|
"mean_token_accuracy": 0.16515503227710723,
|
|
"num_tokens": 6408768.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.713067817687988,
|
|
"epoch": 2.9288956127080183,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004511732273452347,
|
|
"loss": 5.4582,
|
|
"mean_token_accuracy": 0.15396346151828766,
|
|
"num_tokens": 6419427.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.73297438621521,
|
|
"epoch": 2.9339384770549675,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004509262150651089,
|
|
"loss": 5.4812,
|
|
"mean_token_accuracy": 0.15133209377527237,
|
|
"num_tokens": 6430135.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.707785749435425,
|
|
"epoch": 2.938981341401916,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045067865600224833,
|
|
"loss": 5.425,
|
|
"mean_token_accuracy": 0.15304721444845198,
|
|
"num_tokens": 6440977.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.713269805908203,
|
|
"epoch": 2.9440242057488653,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004504305509260724,
|
|
"loss": 5.478,
|
|
"mean_token_accuracy": 0.1551429495215416,
|
|
"num_tokens": 6452319.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.829064273834229,
|
|
"epoch": 2.9490670700958144,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004501819006076974,
|
|
"loss": 5.482,
|
|
"mean_token_accuracy": 0.15775506496429442,
|
|
"num_tokens": 6463970.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.789419078826905,
|
|
"epoch": 2.9541099344427635,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044993270581993413,
|
|
"loss": 5.495,
|
|
"mean_token_accuracy": 0.14951940700411798,
|
|
"num_tokens": 6474113.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.697580623626709,
|
|
"epoch": 2.9591527987897126,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00044968296733728587,
|
|
"loss": 5.4777,
|
|
"mean_token_accuracy": 0.1503283739089966,
|
|
"num_tokens": 6485596.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.704351806640625,
|
|
"epoch": 2.9641956631366617,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004494326859359456,
|
|
"loss": 5.4802,
|
|
"mean_token_accuracy": 0.15750115364789963,
|
|
"num_tokens": 6497521.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.6369767665863035,
|
|
"epoch": 2.969238527483611,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044918186239379363,
|
|
"loss": 5.2882,
|
|
"mean_token_accuracy": 0.1666715368628502,
|
|
"num_tokens": 6508034.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.607600402832031,
|
|
"epoch": 2.9742813918305595,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004489304974903953,
|
|
"loss": 5.4667,
|
|
"mean_token_accuracy": 0.15410790145397185,
|
|
"num_tokens": 6518640.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.7324717998504635,
|
|
"epoch": 2.9793242561775086,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004486785920069985,
|
|
"loss": 5.4666,
|
|
"mean_token_accuracy": 0.1566634237766266,
|
|
"num_tokens": 6528856.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.747164154052735,
|
|
"epoch": 2.9843671205244577,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004484261467265313,
|
|
"loss": 5.5008,
|
|
"mean_token_accuracy": 0.15833084285259247,
|
|
"num_tokens": 6539385.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.621426057815552,
|
|
"epoch": 2.989409984871407,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044817316243359956,
|
|
"loss": 5.4573,
|
|
"mean_token_accuracy": 0.15738039314746857,
|
|
"num_tokens": 6550327.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.802563714981079,
|
|
"epoch": 2.994452849218356,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044791963991448427,
|
|
"loss": 5.5538,
|
|
"mean_token_accuracy": 0.14475565552711486,
|
|
"num_tokens": 6563230.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.689474630355835,
|
|
"epoch": 2.999495713565305,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00044766557995713915,
|
|
"loss": 5.4316,
|
|
"mean_token_accuracy": 0.15604960024356843,
|
|
"num_tokens": 6573947.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.714105500115289,
|
|
"epoch": 3.0040342914775593,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004474109833511885,
|
|
"loss": 5.3151,
|
|
"mean_token_accuracy": 0.16546918120649126,
|
|
"num_tokens": 6583671.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.67859673500061,
|
|
"epoch": 3.0090771558245084,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004471558508879244,
|
|
"loss": 5.1105,
|
|
"mean_token_accuracy": 0.1719118610024452,
|
|
"num_tokens": 6594270.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.638429021835327,
|
|
"epoch": 3.0141200201714575,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004469001833603045,
|
|
"loss": 5.2033,
|
|
"mean_token_accuracy": 0.17100471556186675,
|
|
"num_tokens": 6605849.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.6778584003448485,
|
|
"epoch": 3.0191628845184066,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004466439815629493,
|
|
"loss": 5.14,
|
|
"mean_token_accuracy": 0.1727638065814972,
|
|
"num_tokens": 6615630.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.664512825012207,
|
|
"epoch": 3.0242057488653553,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044638724629214,
|
|
"loss": 5.153,
|
|
"mean_token_accuracy": 0.17399086505174638,
|
|
"num_tokens": 6626467.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.0242057488653553,
|
|
"eval_entropy": 5.454597516824271,
|
|
"eval_loss": 5.765429496765137,
|
|
"eval_mean_token_accuracy": 0.14664496775271024,
|
|
"eval_num_tokens": 6626467.0,
|
|
"eval_runtime": 2.0324,
|
|
"eval_samples_per_second": 1543.982,
|
|
"eval_steps_per_second": 193.367,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.624707126617432,
|
|
"epoch": 3.0292486132123044,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044612997834581557,
|
|
"loss": 5.1775,
|
|
"mean_token_accuracy": 0.16961953043937683,
|
|
"num_tokens": 6638599.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.630809354782104,
|
|
"epoch": 3.0342914775592535,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044587217852357086,
|
|
"loss": 5.123,
|
|
"mean_token_accuracy": 0.1691894441843033,
|
|
"num_tokens": 6648375.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.585572862625122,
|
|
"epoch": 3.0393343419062027,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044561384762665363,
|
|
"loss": 5.2175,
|
|
"mean_token_accuracy": 0.17457871437072753,
|
|
"num_tokens": 6659910.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.675475311279297,
|
|
"epoch": 3.044377206253152,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004453549864579623,
|
|
"loss": 5.2177,
|
|
"mean_token_accuracy": 0.17402236610651017,
|
|
"num_tokens": 6670719.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.612090826034546,
|
|
"epoch": 3.049420070600101,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004450955958220433,
|
|
"loss": 5.0868,
|
|
"mean_token_accuracy": 0.17708014696836472,
|
|
"num_tokens": 6681168.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.632619190216064,
|
|
"epoch": 3.05446293494705,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044483567652508864,
|
|
"loss": 5.2205,
|
|
"mean_token_accuracy": 0.16884091645479202,
|
|
"num_tokens": 6694026.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.628296947479248,
|
|
"epoch": 3.059505799293999,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004445752293749337,
|
|
"loss": 5.0936,
|
|
"mean_token_accuracy": 0.1738958701491356,
|
|
"num_tokens": 6703666.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.514598226547241,
|
|
"epoch": 3.0645486636409482,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004443142551810541,
|
|
"loss": 5.1048,
|
|
"mean_token_accuracy": 0.17896140217781067,
|
|
"num_tokens": 6713506.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.579556608200074,
|
|
"epoch": 3.069591527987897,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044405275475456365,
|
|
"loss": 5.0987,
|
|
"mean_token_accuracy": 0.17890556901693344,
|
|
"num_tokens": 6724184.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.552690172195435,
|
|
"epoch": 3.074634392334846,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004437907289082116,
|
|
"loss": 5.2027,
|
|
"mean_token_accuracy": 0.17316050678491593,
|
|
"num_tokens": 6734843.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.7205370426177975,
|
|
"epoch": 3.079677256681795,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044352817845638053,
|
|
"loss": 5.19,
|
|
"mean_token_accuracy": 0.1774067610502243,
|
|
"num_tokens": 6746194.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.671412706375122,
|
|
"epoch": 3.0847201210287443,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004432651042150833,
|
|
"loss": 5.1768,
|
|
"mean_token_accuracy": 0.17474347949028016,
|
|
"num_tokens": 6757617.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.6246847152709964,
|
|
"epoch": 3.0897629853756934,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00044300150700196054,
|
|
"loss": 5.2724,
|
|
"mean_token_accuracy": 0.17038802802562714,
|
|
"num_tokens": 6770026.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.656072330474854,
|
|
"epoch": 3.0948058497226425,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004427373876362786,
|
|
"loss": 5.1523,
|
|
"mean_token_accuracy": 0.17798729091882706,
|
|
"num_tokens": 6781072.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.605785083770752,
|
|
"epoch": 3.0998487140695916,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004424727469389265,
|
|
"loss": 5.1965,
|
|
"mean_token_accuracy": 0.1710650458931923,
|
|
"num_tokens": 6790758.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.545757865905761,
|
|
"epoch": 3.1048915784165407,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004422075857324137,
|
|
"loss": 5.1302,
|
|
"mean_token_accuracy": 0.17503484785556794,
|
|
"num_tokens": 6800929.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.638562488555908,
|
|
"epoch": 3.10993444276349,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004419419048408673,
|
|
"loss": 5.2069,
|
|
"mean_token_accuracy": 0.17085701674222947,
|
|
"num_tokens": 6811565.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.619118499755859,
|
|
"epoch": 3.1149773071104385,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004416757050900297,
|
|
"loss": 5.1458,
|
|
"mean_token_accuracy": 0.17683406919240952,
|
|
"num_tokens": 6822088.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.572403144836426,
|
|
"epoch": 3.1200201714573876,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004414089873072557,
|
|
"loss": 5.2257,
|
|
"mean_token_accuracy": 0.16655576080083848,
|
|
"num_tokens": 6833032.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.581235647201538,
|
|
"epoch": 3.1250630358043368,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004411417523215105,
|
|
"loss": 5.1334,
|
|
"mean_token_accuracy": 0.17942795902490616,
|
|
"num_tokens": 6844224.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.653260564804077,
|
|
"epoch": 3.130105900151286,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004408740009633665,
|
|
"loss": 5.1862,
|
|
"mean_token_accuracy": 0.17529474645853044,
|
|
"num_tokens": 6855620.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.71563310623169,
|
|
"epoch": 3.135148764498235,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044060573406500117,
|
|
"loss": 5.2117,
|
|
"mean_token_accuracy": 0.17314352393150328,
|
|
"num_tokens": 6869080.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.541918420791626,
|
|
"epoch": 3.140191628845184,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004403369524601941,
|
|
"loss": 5.0713,
|
|
"mean_token_accuracy": 0.18644551634788514,
|
|
"num_tokens": 6878893.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.568064546585083,
|
|
"epoch": 3.145234493192133,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004400676569843249,
|
|
"loss": 5.1767,
|
|
"mean_token_accuracy": 0.17243396490812302,
|
|
"num_tokens": 6889641.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.574889373779297,
|
|
"epoch": 3.1502773575390823,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043979784847436994,
|
|
"loss": 5.1825,
|
|
"mean_token_accuracy": 0.17157595604658127,
|
|
"num_tokens": 6900930.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.7191845893859865,
|
|
"epoch": 3.1553202218860315,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004395275277689004,
|
|
"loss": 5.2806,
|
|
"mean_token_accuracy": 0.17274542152881622,
|
|
"num_tokens": 6912083.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.633571147918701,
|
|
"epoch": 3.16036308623298,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004392566957080793,
|
|
"loss": 5.2492,
|
|
"mean_token_accuracy": 0.1637856423854828,
|
|
"num_tokens": 6924677.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.556624555587769,
|
|
"epoch": 3.1654059505799292,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043898535313365884,
|
|
"loss": 5.1674,
|
|
"mean_token_accuracy": 0.17192815840244294,
|
|
"num_tokens": 6936148.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.636176872253418,
|
|
"epoch": 3.1704488149268784,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000438713500888978,
|
|
"loss": 5.1623,
|
|
"mean_token_accuracy": 0.17791370004415513,
|
|
"num_tokens": 6945956.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.544927453994751,
|
|
"epoch": 3.1754916792738275,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004384411398189599,
|
|
"loss": 5.1143,
|
|
"mean_token_accuracy": 0.1729247212409973,
|
|
"num_tokens": 6956785.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.588227605819702,
|
|
"epoch": 3.1805345436207766,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000438168270770109,
|
|
"loss": 5.2307,
|
|
"mean_token_accuracy": 0.17140598893165587,
|
|
"num_tokens": 6967652.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.6545312881469725,
|
|
"epoch": 3.1855774079677257,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004378948945905085,
|
|
"loss": 5.205,
|
|
"mean_token_accuracy": 0.16459373533725738,
|
|
"num_tokens": 6979682.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.530513000488281,
|
|
"epoch": 3.190620272314675,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043762101212981805,
|
|
"loss": 5.1728,
|
|
"mean_token_accuracy": 0.1710875228047371,
|
|
"num_tokens": 6989733.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.536697578430176,
|
|
"epoch": 3.195663136661624,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043734662423927054,
|
|
"loss": 5.1218,
|
|
"mean_token_accuracy": 0.16705944687128066,
|
|
"num_tokens": 6999713.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.6293610572814945,
|
|
"epoch": 3.200706001008573,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004370717317716699,
|
|
"loss": 5.1607,
|
|
"mean_token_accuracy": 0.1728067383170128,
|
|
"num_tokens": 7010066.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.566136169433594,
|
|
"epoch": 3.205748865355522,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004367963355813882,
|
|
"loss": 5.191,
|
|
"mean_token_accuracy": 0.16791414767503737,
|
|
"num_tokens": 7020149.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.564663887023926,
|
|
"epoch": 3.210791729702471,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043652043652436323,
|
|
"loss": 5.2088,
|
|
"mean_token_accuracy": 0.16811802834272385,
|
|
"num_tokens": 7031417.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.649209117889404,
|
|
"epoch": 3.21583459404942,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004362440354580956,
|
|
"loss": 5.2171,
|
|
"mean_token_accuracy": 0.1664293885231018,
|
|
"num_tokens": 7042619.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.573532295227051,
|
|
"epoch": 3.220877458396369,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004359671332416462,
|
|
"loss": 5.1925,
|
|
"mean_token_accuracy": 0.16897227466106415,
|
|
"num_tokens": 7054003.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.554476642608643,
|
|
"epoch": 3.225920322743318,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004356897307356335,
|
|
"loss": 5.2428,
|
|
"mean_token_accuracy": 0.17232795655727387,
|
|
"num_tokens": 7065837.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.677301025390625,
|
|
"epoch": 3.2309631870902673,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004354118288022309,
|
|
"loss": 5.2262,
|
|
"mean_token_accuracy": 0.1688625156879425,
|
|
"num_tokens": 7076059.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.554659843444824,
|
|
"epoch": 3.2360060514372164,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043513342830516405,
|
|
"loss": 5.199,
|
|
"mean_token_accuracy": 0.17233071476221085,
|
|
"num_tokens": 7086163.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.5844423294067385,
|
|
"epoch": 3.2410489157841655,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000434854530109708,
|
|
"loss": 5.1042,
|
|
"mean_token_accuracy": 0.1778377577662468,
|
|
"num_tokens": 7096044.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.605183219909668,
|
|
"epoch": 3.2460917801311147,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043457513508268487,
|
|
"loss": 5.2085,
|
|
"mean_token_accuracy": 0.16744081825017929,
|
|
"num_tokens": 7109189.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.528883600234986,
|
|
"epoch": 3.2511346444780633,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043429524409246083,
|
|
"loss": 5.1615,
|
|
"mean_token_accuracy": 0.1684742361307144,
|
|
"num_tokens": 7120418.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.597215890884399,
|
|
"epoch": 3.2561775088250124,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004340148580089436,
|
|
"loss": 5.2217,
|
|
"mean_token_accuracy": 0.17177551239728928,
|
|
"num_tokens": 7131895.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.531784629821777,
|
|
"epoch": 3.2612203731719616,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004337339777035794,
|
|
"loss": 5.0864,
|
|
"mean_token_accuracy": 0.17491756081581117,
|
|
"num_tokens": 7141608.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.65837287902832,
|
|
"epoch": 3.2662632375189107,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000433452604049351,
|
|
"loss": 5.2354,
|
|
"mean_token_accuracy": 0.1671900361776352,
|
|
"num_tokens": 7153550.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.607674837112427,
|
|
"epoch": 3.27130610186586,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043317073792077394,
|
|
"loss": 5.136,
|
|
"mean_token_accuracy": 0.17821951061487198,
|
|
"num_tokens": 7162945.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.5570567607879635,
|
|
"epoch": 3.276348966212809,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004328883801938947,
|
|
"loss": 5.1426,
|
|
"mean_token_accuracy": 0.17293958365917206,
|
|
"num_tokens": 7173426.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.569593620300293,
|
|
"epoch": 3.281391830559758,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00043260553174628776,
|
|
"loss": 5.1181,
|
|
"mean_token_accuracy": 0.16555063426494598,
|
|
"num_tokens": 7184335.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.510777854919434,
|
|
"epoch": 3.286434694906707,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004323221934570524,
|
|
"loss": 5.1753,
|
|
"mean_token_accuracy": 0.17292621433734895,
|
|
"num_tokens": 7195587.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.657611513137818,
|
|
"epoch": 3.2914775592536563,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043203836620681074,
|
|
"loss": 5.213,
|
|
"mean_token_accuracy": 0.1684402734041214,
|
|
"num_tokens": 7205616.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.532323741912842,
|
|
"epoch": 3.2965204236006054,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043175405087770435,
|
|
"loss": 5.0848,
|
|
"mean_token_accuracy": 0.17454392611980438,
|
|
"num_tokens": 7215642.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.537416124343872,
|
|
"epoch": 3.301563287947554,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00043146924835339186,
|
|
"loss": 5.1045,
|
|
"mean_token_accuracy": 0.1720063418149948,
|
|
"num_tokens": 7226607.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.5237953662872314,
|
|
"epoch": 3.306606152294503,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043118395951904605,
|
|
"loss": 5.1983,
|
|
"mean_token_accuracy": 0.16655222028493882,
|
|
"num_tokens": 7238035.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.620041179656982,
|
|
"epoch": 3.3116490166414523,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004308981852613512,
|
|
"loss": 5.1871,
|
|
"mean_token_accuracy": 0.16416720151901246,
|
|
"num_tokens": 7248183.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.61048731803894,
|
|
"epoch": 3.3166918809884014,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004306119264685003,
|
|
"loss": 5.2073,
|
|
"mean_token_accuracy": 0.1729658141732216,
|
|
"num_tokens": 7260092.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.5868383884429935,
|
|
"epoch": 3.3217347453353505,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004303251840301923,
|
|
"loss": 5.1853,
|
|
"mean_token_accuracy": 0.16900800466537474,
|
|
"num_tokens": 7272262.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.602219533920288,
|
|
"epoch": 3.3267776096822996,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043003795883762933,
|
|
"loss": 5.2445,
|
|
"mean_token_accuracy": 0.1740011304616928,
|
|
"num_tokens": 7283061.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.639716911315918,
|
|
"epoch": 3.3318204740292487,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042975025178351386,
|
|
"loss": 5.2361,
|
|
"mean_token_accuracy": 0.17260727137327195,
|
|
"num_tokens": 7292904.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.582015132904052,
|
|
"epoch": 3.336863338376198,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000429462063762046,
|
|
"loss": 5.1816,
|
|
"mean_token_accuracy": 0.1696217879652977,
|
|
"num_tokens": 7304368.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.599396991729736,
|
|
"epoch": 3.3419062027231465,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042917339566892095,
|
|
"loss": 5.2711,
|
|
"mean_token_accuracy": 0.16964879333972932,
|
|
"num_tokens": 7315500.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.59840669631958,
|
|
"epoch": 3.3469490670700957,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004288842484013257,
|
|
"loss": 5.141,
|
|
"mean_token_accuracy": 0.17560205608606339,
|
|
"num_tokens": 7326479.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.613791799545288,
|
|
"epoch": 3.3519919314170448,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004285946228579365,
|
|
"loss": 5.2597,
|
|
"mean_token_accuracy": 0.16658831536769866,
|
|
"num_tokens": 7338284.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.645831203460693,
|
|
"epoch": 3.357034795763994,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004283045199389164,
|
|
"loss": 5.1938,
|
|
"mean_token_accuracy": 0.16851962357759476,
|
|
"num_tokens": 7348601.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.586950302124023,
|
|
"epoch": 3.362077660110943,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00042801394054591197,
|
|
"loss": 5.2191,
|
|
"mean_token_accuracy": 0.1698923408985138,
|
|
"num_tokens": 7360282.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.636523103713989,
|
|
"epoch": 3.367120524457892,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042772288558205057,
|
|
"loss": 5.204,
|
|
"mean_token_accuracy": 0.17189548760652543,
|
|
"num_tokens": 7371377.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.523115110397339,
|
|
"epoch": 3.3721633888048412,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004274313559519377,
|
|
"loss": 5.1369,
|
|
"mean_token_accuracy": 0.17882900387048722,
|
|
"num_tokens": 7381895.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.621546792984009,
|
|
"epoch": 3.3772062531517903,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004271393525616544,
|
|
"loss": 5.2463,
|
|
"mean_token_accuracy": 0.1715021014213562,
|
|
"num_tokens": 7393211.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.56148624420166,
|
|
"epoch": 3.3822491174987395,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042684687631875377,
|
|
"loss": 5.178,
|
|
"mean_token_accuracy": 0.1684443712234497,
|
|
"num_tokens": 7404618.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.640697002410889,
|
|
"epoch": 3.3872919818456886,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042655392813225864,
|
|
"loss": 5.1458,
|
|
"mean_token_accuracy": 0.17137933224439622,
|
|
"num_tokens": 7415467.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.532991790771485,
|
|
"epoch": 3.3923348461926373,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042626050891265884,
|
|
"loss": 5.1787,
|
|
"mean_token_accuracy": 0.17038492113351822,
|
|
"num_tokens": 7426594.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.627938890457154,
|
|
"epoch": 3.3973777105395864,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004259666195719081,
|
|
"loss": 5.2112,
|
|
"mean_token_accuracy": 0.17373456209897994,
|
|
"num_tokens": 7436920.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.611821985244751,
|
|
"epoch": 3.4024205748865355,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042567226102342134,
|
|
"loss": 5.2864,
|
|
"mean_token_accuracy": 0.1682152509689331,
|
|
"num_tokens": 7448499.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.56901707649231,
|
|
"epoch": 3.4074634392334846,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004253774341820715,
|
|
"loss": 5.1987,
|
|
"mean_token_accuracy": 0.17500097304582596,
|
|
"num_tokens": 7459788.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.630524682998657,
|
|
"epoch": 3.4125063035804337,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004250821399641874,
|
|
"loss": 5.1772,
|
|
"mean_token_accuracy": 0.1729726031422615,
|
|
"num_tokens": 7472387.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.558799314498901,
|
|
"epoch": 3.417549167927383,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004247863792875502,
|
|
"loss": 5.1668,
|
|
"mean_token_accuracy": 0.17443351447582245,
|
|
"num_tokens": 7483242.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.50960865020752,
|
|
"epoch": 3.422592032274332,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042449015307139076,
|
|
"loss": 5.2067,
|
|
"mean_token_accuracy": 0.17049969583749772,
|
|
"num_tokens": 7494908.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.585347986221313,
|
|
"epoch": 3.427634896621281,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042419346223638725,
|
|
"loss": 5.2319,
|
|
"mean_token_accuracy": 0.1668471336364746,
|
|
"num_tokens": 7505468.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.6179139614105225,
|
|
"epoch": 3.4326777609682297,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004238963077046616,
|
|
"loss": 5.2232,
|
|
"mean_token_accuracy": 0.165214341878891,
|
|
"num_tokens": 7517281.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.544996356964111,
|
|
"epoch": 3.437720625315179,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004235986903997767,
|
|
"loss": 5.1942,
|
|
"mean_token_accuracy": 0.16933305859565734,
|
|
"num_tokens": 7528680.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.556524276733398,
|
|
"epoch": 3.442763489662128,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042330061124673437,
|
|
"loss": 5.1999,
|
|
"mean_token_accuracy": 0.16344990730285644,
|
|
"num_tokens": 7540927.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.627091646194458,
|
|
"epoch": 3.447806354009077,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042300207117197116,
|
|
"loss": 5.2198,
|
|
"mean_token_accuracy": 0.17462050765752793,
|
|
"num_tokens": 7551827.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.555369091033936,
|
|
"epoch": 3.452849218356026,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042270307110335657,
|
|
"loss": 5.1695,
|
|
"mean_token_accuracy": 0.17428209483623505,
|
|
"num_tokens": 7561721.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.566034984588623,
|
|
"epoch": 3.4578920827029753,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004224036119701899,
|
|
"loss": 5.2021,
|
|
"mean_token_accuracy": 0.1652482956647873,
|
|
"num_tokens": 7571932.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.629524374008179,
|
|
"epoch": 3.4629349470499244,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042210369470319694,
|
|
"loss": 5.1684,
|
|
"mean_token_accuracy": 0.17760788053274154,
|
|
"num_tokens": 7581859.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.498959589004516,
|
|
"epoch": 3.4679778113968736,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004218033202345274,
|
|
"loss": 5.1651,
|
|
"mean_token_accuracy": 0.16923027336597443,
|
|
"num_tokens": 7592811.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.609005641937256,
|
|
"epoch": 3.4730206757438227,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004215024894977522,
|
|
"loss": 5.2602,
|
|
"mean_token_accuracy": 0.16206480115652083,
|
|
"num_tokens": 7604332.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.537288188934326,
|
|
"epoch": 3.478063540090772,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004212012034278601,
|
|
"loss": 5.201,
|
|
"mean_token_accuracy": 0.16690837293863298,
|
|
"num_tokens": 7615387.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.572259140014649,
|
|
"epoch": 3.4831064044377205,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042089946296125513,
|
|
"loss": 5.249,
|
|
"mean_token_accuracy": 0.1678703710436821,
|
|
"num_tokens": 7627381.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.603390121459961,
|
|
"epoch": 3.4881492687846696,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042059726903575373,
|
|
"loss": 5.1732,
|
|
"mean_token_accuracy": 0.16659764498472213,
|
|
"num_tokens": 7637327.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.4189528465271,
|
|
"epoch": 3.4931921331316187,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004202946225905815,
|
|
"loss": 5.1187,
|
|
"mean_token_accuracy": 0.17738016843795776,
|
|
"num_tokens": 7647709.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.524550342559815,
|
|
"epoch": 3.498234997478568,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004199915245663706,
|
|
"loss": 5.2145,
|
|
"mean_token_accuracy": 0.16580404192209244,
|
|
"num_tokens": 7660003.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.57571120262146,
|
|
"epoch": 3.503277861825517,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041968797590515673,
|
|
"loss": 5.1456,
|
|
"mean_token_accuracy": 0.17444880902767182,
|
|
"num_tokens": 7671667.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.544626379013062,
|
|
"epoch": 3.508320726172466,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004193839775503761,
|
|
"loss": 5.1862,
|
|
"mean_token_accuracy": 0.1735386520624161,
|
|
"num_tokens": 7682653.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.5591529369354244,
|
|
"epoch": 3.513363590519415,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041907953044686253,
|
|
"loss": 5.2008,
|
|
"mean_token_accuracy": 0.17169943004846572,
|
|
"num_tokens": 7693834.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.597692251205444,
|
|
"epoch": 3.518406454866364,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004187746355408448,
|
|
"loss": 5.2987,
|
|
"mean_token_accuracy": 0.16503139734268188,
|
|
"num_tokens": 7706323.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.621932125091552,
|
|
"epoch": 3.523449319213313,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004184692937799431,
|
|
"loss": 5.2134,
|
|
"mean_token_accuracy": 0.17117932587862014,
|
|
"num_tokens": 7719019.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.534463167190552,
|
|
"epoch": 3.528492183560262,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041816350611316693,
|
|
"loss": 5.1635,
|
|
"mean_token_accuracy": 0.16748429834842682,
|
|
"num_tokens": 7729312.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.477231693267822,
|
|
"epoch": 3.533535047907211,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041785727349091117,
|
|
"loss": 5.1077,
|
|
"mean_token_accuracy": 0.18037325739860535,
|
|
"num_tokens": 7739049.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.546959733963012,
|
|
"epoch": 3.5385779122541603,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000417550596864954,
|
|
"loss": 5.2936,
|
|
"mean_token_accuracy": 0.1701042428612709,
|
|
"num_tokens": 7750076.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.585612487792969,
|
|
"epoch": 3.5436207766011094,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041724347718845344,
|
|
"loss": 5.1787,
|
|
"mean_token_accuracy": 0.17000592797994613,
|
|
"num_tokens": 7761442.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.61525149345398,
|
|
"epoch": 3.5486636409480585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041693591541594455,
|
|
"loss": 5.1629,
|
|
"mean_token_accuracy": 0.17229202687740325,
|
|
"num_tokens": 7772061.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.5182301044464115,
|
|
"epoch": 3.5537065052950076,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004166279125033363,
|
|
"loss": 5.1521,
|
|
"mean_token_accuracy": 0.1691966325044632,
|
|
"num_tokens": 7783200.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.538359928131103,
|
|
"epoch": 3.5587493696419568,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041631946940790893,
|
|
"loss": 5.2372,
|
|
"mean_token_accuracy": 0.169610658288002,
|
|
"num_tokens": 7793737.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.606062221527099,
|
|
"epoch": 3.563792233988906,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041601058708831077,
|
|
"loss": 5.1918,
|
|
"mean_token_accuracy": 0.16978860646486282,
|
|
"num_tokens": 7804894.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.522193765640258,
|
|
"epoch": 3.568835098335855,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000415701266504555,
|
|
"loss": 5.1683,
|
|
"mean_token_accuracy": 0.16874499917030333,
|
|
"num_tokens": 7815773.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.513698244094849,
|
|
"epoch": 3.5738779626828037,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004153915086180173,
|
|
"loss": 5.2102,
|
|
"mean_token_accuracy": 0.1681763008236885,
|
|
"num_tokens": 7825857.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.572156953811645,
|
|
"epoch": 3.578920827029753,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004150813143914323,
|
|
"loss": 5.2316,
|
|
"mean_token_accuracy": 0.17015199810266496,
|
|
"num_tokens": 7836878.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.5560825824737545,
|
|
"epoch": 3.583963691376702,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004147706847888908,
|
|
"loss": 5.1864,
|
|
"mean_token_accuracy": 0.17338349372148515,
|
|
"num_tokens": 7848382.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.492858743667602,
|
|
"epoch": 3.589006555723651,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004144596207758367,
|
|
"loss": 5.1072,
|
|
"mean_token_accuracy": 0.17649791538715362,
|
|
"num_tokens": 7858736.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.462978792190552,
|
|
"epoch": 3.5940494200706,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004141481233190642,
|
|
"loss": 5.1454,
|
|
"mean_token_accuracy": 0.16892780661582946,
|
|
"num_tokens": 7869396.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.531047677993774,
|
|
"epoch": 3.5990922844175492,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004138361933867145,
|
|
"loss": 5.1796,
|
|
"mean_token_accuracy": 0.17372691184282302,
|
|
"num_tokens": 7880072.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.568310928344727,
|
|
"epoch": 3.6041351487644984,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004135238319482731,
|
|
"loss": 5.2331,
|
|
"mean_token_accuracy": 0.16947868466377258,
|
|
"num_tokens": 7892011.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.589046621322632,
|
|
"epoch": 3.609178013111447,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004132110399745666,
|
|
"loss": 5.2212,
|
|
"mean_token_accuracy": 0.16981034129858016,
|
|
"num_tokens": 7903586.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.527681827545166,
|
|
"epoch": 3.614220877458396,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004128978184377596,
|
|
"loss": 5.1576,
|
|
"mean_token_accuracy": 0.16689536422491075,
|
|
"num_tokens": 7912999.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.544303846359253,
|
|
"epoch": 3.6192637418053453,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004125841683113518,
|
|
"loss": 5.1951,
|
|
"mean_token_accuracy": 0.17250838726758957,
|
|
"num_tokens": 7924969.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.594044017791748,
|
|
"epoch": 3.6243066061522944,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041227009057017523,
|
|
"loss": 5.2287,
|
|
"mean_token_accuracy": 0.17300127297639847,
|
|
"num_tokens": 7936811.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.580919218063355,
|
|
"epoch": 3.6293494704992435,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004119555861903906,
|
|
"loss": 5.1112,
|
|
"mean_token_accuracy": 0.17159901559352875,
|
|
"num_tokens": 7946886.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.598188829421997,
|
|
"epoch": 3.6343923348461926,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004116406561494847,
|
|
"loss": 5.2436,
|
|
"mean_token_accuracy": 0.16683561801910402,
|
|
"num_tokens": 7960485.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.525617361068726,
|
|
"epoch": 3.6394351991931417,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041132530142626756,
|
|
"loss": 5.1817,
|
|
"mean_token_accuracy": 0.1733034983277321,
|
|
"num_tokens": 7971460.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.5784553527832035,
|
|
"epoch": 3.644478063540091,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041100952300086884,
|
|
"loss": 5.2436,
|
|
"mean_token_accuracy": 0.16418713033199311,
|
|
"num_tokens": 7984060.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.591583967208862,
|
|
"epoch": 3.64952092788704,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041069332185473514,
|
|
"loss": 5.2653,
|
|
"mean_token_accuracy": 0.1742211565375328,
|
|
"num_tokens": 7996547.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.575674962997437,
|
|
"epoch": 3.654563792233989,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000410376698970627,
|
|
"loss": 5.1662,
|
|
"mean_token_accuracy": 0.17636522203683852,
|
|
"num_tokens": 8006952.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.547354316711425,
|
|
"epoch": 3.659606656580938,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041005965533261554,
|
|
"loss": 5.3228,
|
|
"mean_token_accuracy": 0.1653442472219467,
|
|
"num_tokens": 8020293.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.541851329803467,
|
|
"epoch": 3.6646495209278873,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040974219192607973,
|
|
"loss": 5.1685,
|
|
"mean_token_accuracy": 0.1649821951985359,
|
|
"num_tokens": 8030736.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.5063081741333,
|
|
"epoch": 3.669692385274836,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040942430973770323,
|
|
"loss": 5.1576,
|
|
"mean_token_accuracy": 0.1750467225909233,
|
|
"num_tokens": 8041206.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.603054332733154,
|
|
"epoch": 3.674735249621785,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004091060097554712,
|
|
"loss": 5.1878,
|
|
"mean_token_accuracy": 0.16984860897064208,
|
|
"num_tokens": 8052549.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.589501905441284,
|
|
"epoch": 3.679778113968734,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004087872929686673,
|
|
"loss": 5.2318,
|
|
"mean_token_accuracy": 0.16828680634498597,
|
|
"num_tokens": 8064011.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.503438901901245,
|
|
"epoch": 3.6848209783156833,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040846816036787056,
|
|
"loss": 5.0901,
|
|
"mean_token_accuracy": 0.17338352501392365,
|
|
"num_tokens": 8074824.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.588330602645874,
|
|
"epoch": 3.6898638426626325,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004081486129449527,
|
|
"loss": 5.2096,
|
|
"mean_token_accuracy": 0.16712614744901658,
|
|
"num_tokens": 8086207.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.569873952865601,
|
|
"epoch": 3.6949067070095816,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00040782865169307423,
|
|
"loss": 5.0798,
|
|
"mean_token_accuracy": 0.18110066801309585,
|
|
"num_tokens": 8097384.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.458026647567749,
|
|
"epoch": 3.6999495713565302,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004075082776066822,
|
|
"loss": 5.094,
|
|
"mean_token_accuracy": 0.1761010468006134,
|
|
"num_tokens": 8110306.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.500713920593261,
|
|
"epoch": 3.7049924357034794,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004071874916815068,
|
|
"loss": 5.179,
|
|
"mean_token_accuracy": 0.16946810930967332,
|
|
"num_tokens": 8121120.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.614670181274414,
|
|
"epoch": 3.7100353000504285,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00040686629491455787,
|
|
"loss": 5.1903,
|
|
"mean_token_accuracy": 0.16760142594575883,
|
|
"num_tokens": 8131657.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.521497011184692,
|
|
"epoch": 3.7150781643973776,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00040654468830412247,
|
|
"loss": 5.0926,
|
|
"mean_token_accuracy": 0.18021889328956603,
|
|
"num_tokens": 8141846.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.584207773208618,
|
|
"epoch": 3.7201210287443267,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040622267284976134,
|
|
"loss": 5.2331,
|
|
"mean_token_accuracy": 0.1693637952208519,
|
|
"num_tokens": 8152713.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.461621713638306,
|
|
"epoch": 3.725163893091276,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040590024955230596,
|
|
"loss": 5.0776,
|
|
"mean_token_accuracy": 0.17957098335027694,
|
|
"num_tokens": 8164805.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.630222606658935,
|
|
"epoch": 3.730206757438225,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004055774194138554,
|
|
"loss": 5.3166,
|
|
"mean_token_accuracy": 0.16696483343839646,
|
|
"num_tokens": 8176932.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.579493188858033,
|
|
"epoch": 3.735249621785174,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004052541834377731,
|
|
"loss": 5.1268,
|
|
"mean_token_accuracy": 0.17967449724674225,
|
|
"num_tokens": 8189199.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.4520971298217775,
|
|
"epoch": 3.740292486132123,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040493054262868397,
|
|
"loss": 5.127,
|
|
"mean_token_accuracy": 0.17680779993534088,
|
|
"num_tokens": 8200132.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.561750507354736,
|
|
"epoch": 3.7453353504790723,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040460649799247116,
|
|
"loss": 5.1938,
|
|
"mean_token_accuracy": 0.16911092400550842,
|
|
"num_tokens": 8211392.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.5382912158966064,
|
|
"epoch": 3.7503782148260214,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004042820505362728,
|
|
"loss": 5.1677,
|
|
"mean_token_accuracy": 0.17133865207433702,
|
|
"num_tokens": 8222831.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.500017786026001,
|
|
"epoch": 3.7554210791729705,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004039572012684792,
|
|
"loss": 5.2038,
|
|
"mean_token_accuracy": 0.17324525117874146,
|
|
"num_tokens": 8234077.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.593252038955688,
|
|
"epoch": 3.760463943519919,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040363195119872916,
|
|
"loss": 5.259,
|
|
"mean_token_accuracy": 0.1709211975336075,
|
|
"num_tokens": 8245419.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.679217767715454,
|
|
"epoch": 3.7655068078668683,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040330630133790757,
|
|
"loss": 5.3014,
|
|
"mean_token_accuracy": 0.16429313123226166,
|
|
"num_tokens": 8256185.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.458292388916016,
|
|
"epoch": 3.7705496722138174,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040298025269814165,
|
|
"loss": 5.1148,
|
|
"mean_token_accuracy": 0.17420152872800826,
|
|
"num_tokens": 8265812.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.533199834823608,
|
|
"epoch": 3.7755925365607665,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004026538062927981,
|
|
"loss": 5.1061,
|
|
"mean_token_accuracy": 0.17417732179164885,
|
|
"num_tokens": 8275544.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.48965048789978,
|
|
"epoch": 3.7806354009077157,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040232696313647985,
|
|
"loss": 5.1825,
|
|
"mean_token_accuracy": 0.17443122416734697,
|
|
"num_tokens": 8286774.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.505076122283936,
|
|
"epoch": 3.7856782652546648,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004019997242450231,
|
|
"loss": 5.2117,
|
|
"mean_token_accuracy": 0.16918815523386002,
|
|
"num_tokens": 8298027.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.622967720031738,
|
|
"epoch": 3.7907211296016134,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004016720906354937,
|
|
"loss": 5.2663,
|
|
"mean_token_accuracy": 0.1656852439045906,
|
|
"num_tokens": 8308574.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.532533407211304,
|
|
"epoch": 3.7957639939485626,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004013440633261845,
|
|
"loss": 5.1606,
|
|
"mean_token_accuracy": 0.179658842086792,
|
|
"num_tokens": 8318085.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.482535552978516,
|
|
"epoch": 3.8008068582955117,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00040101564333661213,
|
|
"loss": 5.1024,
|
|
"mean_token_accuracy": 0.1797365203499794,
|
|
"num_tokens": 8328470.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.5581197261810305,
|
|
"epoch": 3.805849722642461,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040068683168751316,
|
|
"loss": 5.238,
|
|
"mean_token_accuracy": 0.16978387236595155,
|
|
"num_tokens": 8338680.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.529878234863281,
|
|
"epoch": 3.81089258698941,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040035762940084205,
|
|
"loss": 5.1641,
|
|
"mean_token_accuracy": 0.1718604162335396,
|
|
"num_tokens": 8349926.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.543536138534546,
|
|
"epoch": 3.815935451336359,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000400028037499767,
|
|
"loss": 5.2463,
|
|
"mean_token_accuracy": 0.166019806265831,
|
|
"num_tokens": 8361035.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.6012788772583,
|
|
"epoch": 3.820978315683308,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003996980570086673,
|
|
"loss": 5.1466,
|
|
"mean_token_accuracy": 0.1742403820157051,
|
|
"num_tokens": 8371201.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.442079782485962,
|
|
"epoch": 3.8260211800302573,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039936768895312987,
|
|
"loss": 5.0571,
|
|
"mean_token_accuracy": 0.1846223682165146,
|
|
"num_tokens": 8382078.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.448010444641113,
|
|
"epoch": 3.8310640443772064,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00039903693435994626,
|
|
"loss": 5.1149,
|
|
"mean_token_accuracy": 0.16895634680986404,
|
|
"num_tokens": 8394072.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.537779521942139,
|
|
"epoch": 3.8361069087241555,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039870579425710937,
|
|
"loss": 5.2554,
|
|
"mean_token_accuracy": 0.169295272231102,
|
|
"num_tokens": 8405077.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.54584436416626,
|
|
"epoch": 3.8411497730711046,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003983742696738103,
|
|
"loss": 5.2189,
|
|
"mean_token_accuracy": 0.17123345285654068,
|
|
"num_tokens": 8416009.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.544118452072143,
|
|
"epoch": 3.8461926374180537,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003980423616404351,
|
|
"loss": 5.2399,
|
|
"mean_token_accuracy": 0.16529951393604278,
|
|
"num_tokens": 8426756.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.606235837936401,
|
|
"epoch": 3.8512355017650024,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003977100711885616,
|
|
"loss": 5.1831,
|
|
"mean_token_accuracy": 0.17143406420946122,
|
|
"num_tokens": 8438052.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.504273700714111,
|
|
"epoch": 3.8562783661119515,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039737739935095614,
|
|
"loss": 5.1765,
|
|
"mean_token_accuracy": 0.1669452667236328,
|
|
"num_tokens": 8448481.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.53804202079773,
|
|
"epoch": 3.8613212304589006,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003970443471615705,
|
|
"loss": 5.2245,
|
|
"mean_token_accuracy": 0.165435391664505,
|
|
"num_tokens": 8460199.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.603448534011841,
|
|
"epoch": 3.8663640948058497,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003967109156555386,
|
|
"loss": 5.1554,
|
|
"mean_token_accuracy": 0.1741995930671692,
|
|
"num_tokens": 8470842.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.599123954772949,
|
|
"epoch": 3.871406959152799,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039637710586917324,
|
|
"loss": 5.2734,
|
|
"mean_token_accuracy": 0.1638499453663826,
|
|
"num_tokens": 8482631.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.530108976364136,
|
|
"epoch": 3.876449823499748,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039604291883996287,
|
|
"loss": 5.1292,
|
|
"mean_token_accuracy": 0.16958068907260895,
|
|
"num_tokens": 8495178.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.548141574859619,
|
|
"epoch": 3.8814926878466967,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039570835560656854,
|
|
"loss": 5.1354,
|
|
"mean_token_accuracy": 0.18214099258184432,
|
|
"num_tokens": 8505257.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.552135944366455,
|
|
"epoch": 3.8865355521936458,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003953734172088205,
|
|
"loss": 5.2628,
|
|
"mean_token_accuracy": 0.16553220599889756,
|
|
"num_tokens": 8517124.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.474930334091186,
|
|
"epoch": 3.891578416540595,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039503810468771494,
|
|
"loss": 5.1272,
|
|
"mean_token_accuracy": 0.17467139214277266,
|
|
"num_tokens": 8528201.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.552065467834472,
|
|
"epoch": 3.896621280887544,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003947024190854111,
|
|
"loss": 5.276,
|
|
"mean_token_accuracy": 0.16642479002475738,
|
|
"num_tokens": 8539900.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.5774843215942385,
|
|
"epoch": 3.901664145234493,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039436636144522737,
|
|
"loss": 5.2117,
|
|
"mean_token_accuracy": 0.17163171470165253,
|
|
"num_tokens": 8549199.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.547877407073974,
|
|
"epoch": 3.9067070095814422,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039402993281163875,
|
|
"loss": 5.1093,
|
|
"mean_token_accuracy": 0.16864251494407653,
|
|
"num_tokens": 8560626.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.571566009521485,
|
|
"epoch": 3.9117498739283914,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003936931342302732,
|
|
"loss": 5.2851,
|
|
"mean_token_accuracy": 0.16724284142255783,
|
|
"num_tokens": 8571706.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.540174150466919,
|
|
"epoch": 3.9167927382753405,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003933559667479084,
|
|
"loss": 5.2048,
|
|
"mean_token_accuracy": 0.1692056193947792,
|
|
"num_tokens": 8583101.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.600929307937622,
|
|
"epoch": 3.9218356026222896,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039301843141246876,
|
|
"loss": 5.2951,
|
|
"mean_token_accuracy": 0.16274531185626984,
|
|
"num_tokens": 8594742.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.540600728988648,
|
|
"epoch": 3.9268784669692387,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039268052927302185,
|
|
"loss": 5.1725,
|
|
"mean_token_accuracy": 0.17035382688045503,
|
|
"num_tokens": 8606609.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.531373834609985,
|
|
"epoch": 3.931921331316188,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003923422613797753,
|
|
"loss": 5.1982,
|
|
"mean_token_accuracy": 0.17010482251644135,
|
|
"num_tokens": 8618291.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.605801773071289,
|
|
"epoch": 3.936964195663137,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003920036287840734,
|
|
"loss": 5.2477,
|
|
"mean_token_accuracy": 0.1721893161535263,
|
|
"num_tokens": 8629850.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.514268398284912,
|
|
"epoch": 3.9420070600100856,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039166463253839434,
|
|
"loss": 5.2419,
|
|
"mean_token_accuracy": 0.1722009301185608,
|
|
"num_tokens": 8640363.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.500269651412964,
|
|
"epoch": 3.9470499243570347,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003913252736963462,
|
|
"loss": 5.2264,
|
|
"mean_token_accuracy": 0.17108370512723922,
|
|
"num_tokens": 8651869.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.567825078964233,
|
|
"epoch": 3.952092788703984,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000390985553312664,
|
|
"loss": 5.0944,
|
|
"mean_token_accuracy": 0.17673670500516891,
|
|
"num_tokens": 8662032.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.550057506561279,
|
|
"epoch": 3.957135653050933,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003906454724432065,
|
|
"loss": 5.1629,
|
|
"mean_token_accuracy": 0.17670822143554688,
|
|
"num_tokens": 8673186.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.504792022705078,
|
|
"epoch": 3.962178517397882,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039030503214495296,
|
|
"loss": 5.0819,
|
|
"mean_token_accuracy": 0.18184632956981658,
|
|
"num_tokens": 8683176.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.47991909980774,
|
|
"epoch": 3.967221381744831,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003899642334759996,
|
|
"loss": 5.1435,
|
|
"mean_token_accuracy": 0.1776417836546898,
|
|
"num_tokens": 8694799.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.562114810943603,
|
|
"epoch": 3.97226424609178,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003896230774955567,
|
|
"loss": 5.1989,
|
|
"mean_token_accuracy": 0.17264344543218613,
|
|
"num_tokens": 8706356.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.471002006530762,
|
|
"epoch": 3.977307110438729,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038928156526394476,
|
|
"loss": 5.1266,
|
|
"mean_token_accuracy": 0.17723785936832429,
|
|
"num_tokens": 8717340.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.4876954555511475,
|
|
"epoch": 3.982349974785678,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003889396978425918,
|
|
"loss": 5.0846,
|
|
"mean_token_accuracy": 0.1728333756327629,
|
|
"num_tokens": 8728902.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.48401951789856,
|
|
"epoch": 3.987392839132627,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003885974762940295,
|
|
"loss": 5.1476,
|
|
"mean_token_accuracy": 0.17540051639080048,
|
|
"num_tokens": 8739033.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.533904981613159,
|
|
"epoch": 3.9924357034795763,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003882549016818905,
|
|
"loss": 5.151,
|
|
"mean_token_accuracy": 0.1737613081932068,
|
|
"num_tokens": 8750107.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.5011085033416744,
|
|
"epoch": 3.9974785678265254,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00038791197507090456,
|
|
"loss": 5.1714,
|
|
"mean_token_accuracy": 0.17420586049556733,
|
|
"num_tokens": 8762486.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.4726547135247126,
|
|
"epoch": 4.00201714573878,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038756869752689527,
|
|
"loss": 5.002,
|
|
"mean_token_accuracy": 0.17883256243334877,
|
|
"num_tokens": 8770890.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.476257181167602,
|
|
"epoch": 4.007060010085729,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003872250701167775,
|
|
"loss": 4.9285,
|
|
"mean_token_accuracy": 0.1875666156411171,
|
|
"num_tokens": 8780880.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.516247749328613,
|
|
"epoch": 4.012102874432678,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038688109390855297,
|
|
"loss": 4.872,
|
|
"mean_token_accuracy": 0.19570760428905487,
|
|
"num_tokens": 8792616.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.497460126876831,
|
|
"epoch": 4.017145738779627,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003865367699713077,
|
|
"loss": 4.9809,
|
|
"mean_token_accuracy": 0.1902925744652748,
|
|
"num_tokens": 8803630.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.45757417678833,
|
|
"epoch": 4.022188603126576,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038619209937520836,
|
|
"loss": 4.9198,
|
|
"mean_token_accuracy": 0.19390207678079605,
|
|
"num_tokens": 8814824.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.566776752471924,
|
|
"epoch": 4.027231467473525,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038584708319149924,
|
|
"loss": 4.9239,
|
|
"mean_token_accuracy": 0.19280405938625336,
|
|
"num_tokens": 8824240.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.4058189868927,
|
|
"epoch": 4.032274331820474,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038550172249249856,
|
|
"loss": 4.873,
|
|
"mean_token_accuracy": 0.19107017517089844,
|
|
"num_tokens": 8835129.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.032274331820474,
|
|
"eval_entropy": 5.298038079235087,
|
|
"eval_loss": 5.699831962585449,
|
|
"eval_mean_token_accuracy": 0.1527644212312674,
|
|
"eval_num_tokens": 8835129.0,
|
|
"eval_runtime": 2.0265,
|
|
"eval_samples_per_second": 1548.455,
|
|
"eval_steps_per_second": 193.927,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.507882356643677,
|
|
"epoch": 4.0373171961674235,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003851560183515953,
|
|
"loss": 4.9006,
|
|
"mean_token_accuracy": 0.19513047337532044,
|
|
"num_tokens": 8846883.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.636105680465699,
|
|
"epoch": 4.042360060514373,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000384809971843246,
|
|
"loss": 5.0285,
|
|
"mean_token_accuracy": 0.18483300358057023,
|
|
"num_tokens": 8857592.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.475295829772949,
|
|
"epoch": 4.047402924861321,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000384463584042971,
|
|
"loss": 4.8952,
|
|
"mean_token_accuracy": 0.19550566673278807,
|
|
"num_tokens": 8868527.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.454197931289673,
|
|
"epoch": 4.05244578920827,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038411685602735175,
|
|
"loss": 4.9249,
|
|
"mean_token_accuracy": 0.1891388326883316,
|
|
"num_tokens": 8879416.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.4192150115966795,
|
|
"epoch": 4.057488653555219,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038376978887402687,
|
|
"loss": 4.9098,
|
|
"mean_token_accuracy": 0.19191017746925354,
|
|
"num_tokens": 8889452.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.370370864868164,
|
|
"epoch": 4.062531517902168,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003834223836616891,
|
|
"loss": 4.7748,
|
|
"mean_token_accuracy": 0.19761549085378646,
|
|
"num_tokens": 8900447.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.534723281860352,
|
|
"epoch": 4.067574382249117,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038307464147008176,
|
|
"loss": 4.9709,
|
|
"mean_token_accuracy": 0.18329481780529022,
|
|
"num_tokens": 8911298.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.439350223541259,
|
|
"epoch": 4.072617246596066,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003827265633799957,
|
|
"loss": 4.9157,
|
|
"mean_token_accuracy": 0.1890334367752075,
|
|
"num_tokens": 8922279.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.486150932312012,
|
|
"epoch": 4.0776601109430155,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003823781504732657,
|
|
"loss": 4.9313,
|
|
"mean_token_accuracy": 0.18667482137680053,
|
|
"num_tokens": 8934780.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.514814758300782,
|
|
"epoch": 4.082702975289965,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003820294038327669,
|
|
"loss": 4.8954,
|
|
"mean_token_accuracy": 0.1852332517504692,
|
|
"num_tokens": 8946114.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.417545366287231,
|
|
"epoch": 4.087745839636914,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003816803245424121,
|
|
"loss": 4.8903,
|
|
"mean_token_accuracy": 0.19420945197343825,
|
|
"num_tokens": 8957830.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.4529143333435055,
|
|
"epoch": 4.092788703983863,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003813309136871478,
|
|
"loss": 4.9542,
|
|
"mean_token_accuracy": 0.19547271430492402,
|
|
"num_tokens": 8968312.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.488467979431152,
|
|
"epoch": 4.097831568330812,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038098117235295083,
|
|
"loss": 4.8996,
|
|
"mean_token_accuracy": 0.1928773194551468,
|
|
"num_tokens": 8979502.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.448248672485351,
|
|
"epoch": 4.102874432677761,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003806311016268254,
|
|
"loss": 4.9214,
|
|
"mean_token_accuracy": 0.18985837399959565,
|
|
"num_tokens": 8991934.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.45822057723999,
|
|
"epoch": 4.10791729702471,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003802807025967995,
|
|
"loss": 4.8492,
|
|
"mean_token_accuracy": 0.1978401616215706,
|
|
"num_tokens": 9001669.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.462690544128418,
|
|
"epoch": 4.112960161371659,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003799299763519213,
|
|
"loss": 4.9006,
|
|
"mean_token_accuracy": 0.19447372406721114,
|
|
"num_tokens": 9011904.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.428786897659302,
|
|
"epoch": 4.118003025718608,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003795789239822561,
|
|
"loss": 4.91,
|
|
"mean_token_accuracy": 0.18757592886686325,
|
|
"num_tokens": 9023024.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.474789571762085,
|
|
"epoch": 4.1230458900655576,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003792275465788826,
|
|
"loss": 5.0288,
|
|
"mean_token_accuracy": 0.18281726986169816,
|
|
"num_tokens": 9035721.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.549254083633423,
|
|
"epoch": 4.128088754412507,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037887584523389013,
|
|
"loss": 4.9429,
|
|
"mean_token_accuracy": 0.1963011309504509,
|
|
"num_tokens": 9046541.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.51261191368103,
|
|
"epoch": 4.133131618759456,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00037852382104037454,
|
|
"loss": 4.9209,
|
|
"mean_token_accuracy": 0.1832383990287781,
|
|
"num_tokens": 9057696.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.49119758605957,
|
|
"epoch": 4.138174483106404,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00037817147509243497,
|
|
"loss": 5.0707,
|
|
"mean_token_accuracy": 0.18049182891845703,
|
|
"num_tokens": 9070737.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.50582594871521,
|
|
"epoch": 4.143217347453353,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003778188084851709,
|
|
"loss": 4.935,
|
|
"mean_token_accuracy": 0.19351411312818528,
|
|
"num_tokens": 9081582.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.539880180358887,
|
|
"epoch": 4.148260211800302,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00037746582231467816,
|
|
"loss": 5.0069,
|
|
"mean_token_accuracy": 0.18012915998697282,
|
|
"num_tokens": 9092571.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.47385425567627,
|
|
"epoch": 4.153303076147251,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003771125176780461,
|
|
"loss": 4.9757,
|
|
"mean_token_accuracy": 0.18279217034578324,
|
|
"num_tokens": 9103172.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.495398378372192,
|
|
"epoch": 4.1583459404942005,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037675889567335345,
|
|
"loss": 4.9165,
|
|
"mean_token_accuracy": 0.18877640664577483,
|
|
"num_tokens": 9114407.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.4332798480987545,
|
|
"epoch": 4.16338880484115,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00037640495739966573,
|
|
"loss": 4.871,
|
|
"mean_token_accuracy": 0.19287270307540894,
|
|
"num_tokens": 9126175.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.474513959884644,
|
|
"epoch": 4.168431669188099,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00037605070395703124,
|
|
"loss": 5.0226,
|
|
"mean_token_accuracy": 0.18439341336488724,
|
|
"num_tokens": 9139046.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.486795473098755,
|
|
"epoch": 4.173474533535048,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00037569613644647783,
|
|
"loss": 4.9218,
|
|
"mean_token_accuracy": 0.1871136710047722,
|
|
"num_tokens": 9150413.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.5463550090789795,
|
|
"epoch": 4.178517397881997,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003753412559700094,
|
|
"loss": 4.9098,
|
|
"mean_token_accuracy": 0.1975122556090355,
|
|
"num_tokens": 9161831.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.437201404571534,
|
|
"epoch": 4.183560262228946,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037498606363060285,
|
|
"loss": 4.9258,
|
|
"mean_token_accuracy": 0.1963041603565216,
|
|
"num_tokens": 9172381.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.364158773422242,
|
|
"epoch": 4.188603126575895,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003746305605322041,
|
|
"loss": 4.8546,
|
|
"mean_token_accuracy": 0.19601071029901504,
|
|
"num_tokens": 9182212.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.442264175415039,
|
|
"epoch": 4.193645990922844,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037427474777972477,
|
|
"loss": 4.9478,
|
|
"mean_token_accuracy": 0.19168211966753007,
|
|
"num_tokens": 9192207.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.51265287399292,
|
|
"epoch": 4.198688855269793,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037391862647903943,
|
|
"loss": 4.9895,
|
|
"mean_token_accuracy": 0.18737609982490538,
|
|
"num_tokens": 9203693.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.489502048492431,
|
|
"epoch": 4.2037317196167425,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037356219773698106,
|
|
"loss": 4.93,
|
|
"mean_token_accuracy": 0.19232701063156127,
|
|
"num_tokens": 9215201.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.35007643699646,
|
|
"epoch": 4.208774583963692,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00037320546266133854,
|
|
"loss": 4.8919,
|
|
"mean_token_accuracy": 0.18937314301729202,
|
|
"num_tokens": 9227705.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.513562440872192,
|
|
"epoch": 4.213817448310641,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003728484223608526,
|
|
"loss": 5.0298,
|
|
"mean_token_accuracy": 0.19036928117275237,
|
|
"num_tokens": 9238372.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.479854774475098,
|
|
"epoch": 4.21886031265759,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037249107794521286,
|
|
"loss": 4.9577,
|
|
"mean_token_accuracy": 0.18163685202598573,
|
|
"num_tokens": 9250303.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.421753692626953,
|
|
"epoch": 4.223903177004539,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037213343052505394,
|
|
"loss": 5.0321,
|
|
"mean_token_accuracy": 0.18563628941774368,
|
|
"num_tokens": 9261750.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.509016704559326,
|
|
"epoch": 4.228946041351487,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003717754812119523,
|
|
"loss": 4.9431,
|
|
"mean_token_accuracy": 0.19167808294296265,
|
|
"num_tokens": 9273808.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.454015445709229,
|
|
"epoch": 4.233988905698436,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003714172311184226,
|
|
"loss": 4.9408,
|
|
"mean_token_accuracy": 0.18980903923511505,
|
|
"num_tokens": 9285099.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.498741102218628,
|
|
"epoch": 4.2390317700453854,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003710586813579144,
|
|
"loss": 5.0256,
|
|
"mean_token_accuracy": 0.18340314030647278,
|
|
"num_tokens": 9295513.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.5043713569641115,
|
|
"epoch": 4.244074634392335,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037069983304480875,
|
|
"loss": 5.0181,
|
|
"mean_token_accuracy": 0.17525929361581802,
|
|
"num_tokens": 9307243.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.41992335319519,
|
|
"epoch": 4.249117498739284,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003703406872944144,
|
|
"loss": 4.9652,
|
|
"mean_token_accuracy": 0.1928505852818489,
|
|
"num_tokens": 9319289.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.529471921920776,
|
|
"epoch": 4.254160363086233,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003699812452229645,
|
|
"loss": 5.0786,
|
|
"mean_token_accuracy": 0.1831396237015724,
|
|
"num_tokens": 9332273.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.502027702331543,
|
|
"epoch": 4.259203227433182,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00036962150794761354,
|
|
"loss": 4.9647,
|
|
"mean_token_accuracy": 0.18839758336544038,
|
|
"num_tokens": 9343703.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.409524011611938,
|
|
"epoch": 4.264246091780131,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003692614765864331,
|
|
"loss": 4.9535,
|
|
"mean_token_accuracy": 0.18380556404590606,
|
|
"num_tokens": 9355371.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.415918779373169,
|
|
"epoch": 4.26928895612708,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00036890115225840904,
|
|
"loss": 4.9783,
|
|
"mean_token_accuracy": 0.18282967656850815,
|
|
"num_tokens": 9367337.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.495303773880005,
|
|
"epoch": 4.274331820474029,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003685405360834376,
|
|
"loss": 4.9964,
|
|
"mean_token_accuracy": 0.1882554829120636,
|
|
"num_tokens": 9378613.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.522206258773804,
|
|
"epoch": 4.279374684820978,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00036817962918232234,
|
|
"loss": 4.9605,
|
|
"mean_token_accuracy": 0.18544823825359344,
|
|
"num_tokens": 9390576.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.513616561889648,
|
|
"epoch": 4.2844175491679275,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00036781843267677,
|
|
"loss": 4.9929,
|
|
"mean_token_accuracy": 0.1839417487382889,
|
|
"num_tokens": 9401828.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.4544768810272215,
|
|
"epoch": 4.289460413514877,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003674569476893878,
|
|
"loss": 4.9674,
|
|
"mean_token_accuracy": 0.18869195580482484,
|
|
"num_tokens": 9414065.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.501214170455933,
|
|
"epoch": 4.294503277861826,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003670951753436794,
|
|
"loss": 4.9991,
|
|
"mean_token_accuracy": 0.193516606092453,
|
|
"num_tokens": 9425025.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.451015043258667,
|
|
"epoch": 4.299546142208775,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036673311676404163,
|
|
"loss": 4.9283,
|
|
"mean_token_accuracy": 0.1888926148414612,
|
|
"num_tokens": 9435562.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.448048305511475,
|
|
"epoch": 4.304589006555724,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003663707730757608,
|
|
"loss": 4.9323,
|
|
"mean_token_accuracy": 0.1817737638950348,
|
|
"num_tokens": 9445963.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.42729001045227,
|
|
"epoch": 4.309631870902673,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036600814540500956,
|
|
"loss": 4.9144,
|
|
"mean_token_accuracy": 0.18806320428848267,
|
|
"num_tokens": 9456810.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.404821681976318,
|
|
"epoch": 4.314674735249622,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000365645234878843,
|
|
"loss": 4.944,
|
|
"mean_token_accuracy": 0.18662286251783372,
|
|
"num_tokens": 9466527.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.460329341888428,
|
|
"epoch": 4.31971759959657,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00036528204262519545,
|
|
"loss": 4.9273,
|
|
"mean_token_accuracy": 0.19052704572677612,
|
|
"num_tokens": 9477770.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.477109098434449,
|
|
"epoch": 4.3247604639435195,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003649185697728768,
|
|
"loss": 4.9679,
|
|
"mean_token_accuracy": 0.1862930953502655,
|
|
"num_tokens": 9487723.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.513242292404175,
|
|
"epoch": 4.329803328290469,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00036455481745156906,
|
|
"loss": 4.9169,
|
|
"mean_token_accuracy": 0.19376833438873292,
|
|
"num_tokens": 9497286.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.3801576614379885,
|
|
"epoch": 4.334846192637418,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003641907867918228,
|
|
"loss": 4.8533,
|
|
"mean_token_accuracy": 0.1950397878885269,
|
|
"num_tokens": 9508535.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.381755065917969,
|
|
"epoch": 4.339889056984367,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003638264789250538,
|
|
"loss": 4.9355,
|
|
"mean_token_accuracy": 0.18968858420848847,
|
|
"num_tokens": 9518165.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.4629909038543705,
|
|
"epoch": 4.344931921331316,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036346189498353914,
|
|
"loss": 4.9332,
|
|
"mean_token_accuracy": 0.18796923905611038,
|
|
"num_tokens": 9528197.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.469100570678711,
|
|
"epoch": 4.349974785678265,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003630970361004143,
|
|
"loss": 4.9514,
|
|
"mean_token_accuracy": 0.1987667128443718,
|
|
"num_tokens": 9539024.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.498983573913574,
|
|
"epoch": 4.355017650025214,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003627319034096691,
|
|
"loss": 4.9047,
|
|
"mean_token_accuracy": 0.1949477717280388,
|
|
"num_tokens": 9548701.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.3508720874786375,
|
|
"epoch": 4.360060514372163,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003623664980461443,
|
|
"loss": 4.8769,
|
|
"mean_token_accuracy": 0.19532206803560256,
|
|
"num_tokens": 9558576.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.39421124458313,
|
|
"epoch": 4.3651033787191125,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00036200082114552806,
|
|
"loss": 4.9565,
|
|
"mean_token_accuracy": 0.18533562421798705,
|
|
"num_tokens": 9567464.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.425723648071289,
|
|
"epoch": 4.370146243066062,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036163487384435294,
|
|
"loss": 4.8855,
|
|
"mean_token_accuracy": 0.19447901099920273,
|
|
"num_tokens": 9578011.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.549548530578614,
|
|
"epoch": 4.375189107413011,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003612686572799914,
|
|
"loss": 5.0191,
|
|
"mean_token_accuracy": 0.1840487852692604,
|
|
"num_tokens": 9589279.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.472797155380249,
|
|
"epoch": 4.38023197175996,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003609021725906529,
|
|
"loss": 4.9258,
|
|
"mean_token_accuracy": 0.18987287431955338,
|
|
"num_tokens": 9600170.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.4490509033203125,
|
|
"epoch": 4.385274836106909,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036053542091538045,
|
|
"loss": 4.925,
|
|
"mean_token_accuracy": 0.19032368808984756,
|
|
"num_tokens": 9611006.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.473626661300659,
|
|
"epoch": 4.390317700453858,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00036016840339404664,
|
|
"loss": 5.0008,
|
|
"mean_token_accuracy": 0.18587622791528702,
|
|
"num_tokens": 9621645.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.423802852630615,
|
|
"epoch": 4.395360564800807,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003598011211673505,
|
|
"loss": 5.016,
|
|
"mean_token_accuracy": 0.18200191259384155,
|
|
"num_tokens": 9633221.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.462389421463013,
|
|
"epoch": 4.400403429147756,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003594335753768135,
|
|
"loss": 5.0055,
|
|
"mean_token_accuracy": 0.1831961914896965,
|
|
"num_tokens": 9644616.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.469020700454712,
|
|
"epoch": 4.405446293494705,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003590657671647766,
|
|
"loss": 4.939,
|
|
"mean_token_accuracy": 0.19091181010007857,
|
|
"num_tokens": 9656072.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.4417582035064695,
|
|
"epoch": 4.4104891578416545,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00035869769767439606,
|
|
"loss": 4.9764,
|
|
"mean_token_accuracy": 0.18057194650173186,
|
|
"num_tokens": 9667252.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.527696800231934,
|
|
"epoch": 4.415532022188603,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003583293680496404,
|
|
"loss": 5.0256,
|
|
"mean_token_accuracy": 0.18298113197088242,
|
|
"num_tokens": 9677810.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.482815313339233,
|
|
"epoch": 4.420574886535552,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003579607794352867,
|
|
"loss": 4.9634,
|
|
"mean_token_accuracy": 0.19282228946685792,
|
|
"num_tokens": 9688950.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.51417818069458,
|
|
"epoch": 4.425617750882501,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035759193297691674,
|
|
"loss": 5.0062,
|
|
"mean_token_accuracy": 0.1773902341723442,
|
|
"num_tokens": 9699961.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.441384744644165,
|
|
"epoch": 4.43066061522945,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00035722282982091373,
|
|
"loss": 4.968,
|
|
"mean_token_accuracy": 0.18881333619356155,
|
|
"num_tokens": 9711548.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.4814496517181395,
|
|
"epoch": 4.435703479576399,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003568534711144591,
|
|
"loss": 5.0064,
|
|
"mean_token_accuracy": 0.19238194972276687,
|
|
"num_tokens": 9722237.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.461666250228882,
|
|
"epoch": 4.440746343923348,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000356483858005528,
|
|
"loss": 5.02,
|
|
"mean_token_accuracy": 0.18607597798109055,
|
|
"num_tokens": 9733654.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.396451997756958,
|
|
"epoch": 4.445789208270297,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00035611399164288654,
|
|
"loss": 4.8687,
|
|
"mean_token_accuracy": 0.19753507524728775,
|
|
"num_tokens": 9745012.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.431181240081787,
|
|
"epoch": 4.450832072617247,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003557438731760879,
|
|
"loss": 4.9509,
|
|
"mean_token_accuracy": 0.18538981080055236,
|
|
"num_tokens": 9755851.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.40848217010498,
|
|
"epoch": 4.455874936964196,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003553735037554689,
|
|
"loss": 4.8793,
|
|
"mean_token_accuracy": 0.19078117311000825,
|
|
"num_tokens": 9765652.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.463437080383301,
|
|
"epoch": 4.460917801311145,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003550028845321461,
|
|
"loss": 4.9528,
|
|
"mean_token_accuracy": 0.18317756056785583,
|
|
"num_tokens": 9776618.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.437930488586426,
|
|
"epoch": 4.465960665658094,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003546320166580127,
|
|
"loss": 4.912,
|
|
"mean_token_accuracy": 0.18995112627744676,
|
|
"num_tokens": 9786523.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.503436660766601,
|
|
"epoch": 4.471003530005043,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035426090128573464,
|
|
"loss": 4.9902,
|
|
"mean_token_accuracy": 0.18866927772760392,
|
|
"num_tokens": 9797013.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.505742359161377,
|
|
"epoch": 4.476046394351992,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035388953956874686,
|
|
"loss": 4.978,
|
|
"mean_token_accuracy": 0.1855326250195503,
|
|
"num_tokens": 9807431.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.459383773803711,
|
|
"epoch": 4.481089258698941,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003535179326612504,
|
|
"loss": 5.0067,
|
|
"mean_token_accuracy": 0.18672738820314408,
|
|
"num_tokens": 9817560.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.431673955917359,
|
|
"epoch": 4.48613212304589,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00035314608171820787,
|
|
"loss": 4.9291,
|
|
"mean_token_accuracy": 0.18797255009412767,
|
|
"num_tokens": 9828233.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.450408601760865,
|
|
"epoch": 4.4911749873928395,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00035277398789534075,
|
|
"loss": 4.9298,
|
|
"mean_token_accuracy": 0.19417595118284225,
|
|
"num_tokens": 9838699.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.42654972076416,
|
|
"epoch": 4.496217851739789,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003524016523491252,
|
|
"loss": 4.9502,
|
|
"mean_token_accuracy": 0.1807286635041237,
|
|
"num_tokens": 9849635.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.424818992614746,
|
|
"epoch": 4.501260716086737,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035202907623678865,
|
|
"loss": 4.9072,
|
|
"mean_token_accuracy": 0.18710703998804093,
|
|
"num_tokens": 9861822.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.43951678276062,
|
|
"epoch": 4.506303580433686,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003516562607163062,
|
|
"loss": 4.9867,
|
|
"mean_token_accuracy": 0.18685145676136017,
|
|
"num_tokens": 9872341.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.522892475128174,
|
|
"epoch": 4.511346444780635,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003512832069463971,
|
|
"loss": 5.009,
|
|
"mean_token_accuracy": 0.18200730085372924,
|
|
"num_tokens": 9883078.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.457088470458984,
|
|
"epoch": 4.516389309127584,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035090991608652125,
|
|
"loss": 5.0672,
|
|
"mean_token_accuracy": 0.18591289818286896,
|
|
"num_tokens": 9894688.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.392157983779907,
|
|
"epoch": 4.521432173474533,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00035053638929687504,
|
|
"loss": 4.8778,
|
|
"mean_token_accuracy": 0.18869801312685014,
|
|
"num_tokens": 9904409.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.485831022262573,
|
|
"epoch": 4.526475037821482,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003501626277383884,
|
|
"loss": 5.0515,
|
|
"mean_token_accuracy": 0.1752968832850456,
|
|
"num_tokens": 9915204.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.5224214553833,
|
|
"epoch": 4.5315179021684315,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000349788632572721,
|
|
"loss": 4.9806,
|
|
"mean_token_accuracy": 0.18406761437654495,
|
|
"num_tokens": 9927354.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.412999963760376,
|
|
"epoch": 4.536560766515381,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003494144049622582,
|
|
"loss": 4.902,
|
|
"mean_token_accuracy": 0.1886482670903206,
|
|
"num_tokens": 9939009.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.391222190856934,
|
|
"epoch": 4.54160363086233,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003490399460701083,
|
|
"loss": 5.0147,
|
|
"mean_token_accuracy": 0.18851865231990814,
|
|
"num_tokens": 9950332.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.448312902450562,
|
|
"epoch": 4.546646495209279,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000348665257060098,
|
|
"loss": 4.922,
|
|
"mean_token_accuracy": 0.1947105810046196,
|
|
"num_tokens": 9961206.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.484153413772583,
|
|
"epoch": 4.551689359556228,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003482903390967695,
|
|
"loss": 4.917,
|
|
"mean_token_accuracy": 0.19644837379455565,
|
|
"num_tokens": 9972272.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.377922439575196,
|
|
"epoch": 4.556732223903177,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00034791519334537635,
|
|
"loss": 4.9262,
|
|
"mean_token_accuracy": 0.19269737154245375,
|
|
"num_tokens": 9982189.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.377587604522705,
|
|
"epoch": 4.561775088250126,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003475398209718804,
|
|
"loss": 4.8724,
|
|
"mean_token_accuracy": 0.1955879732966423,
|
|
"num_tokens": 9993122.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.44607400894165,
|
|
"epoch": 4.566817952597075,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003471642231429475,
|
|
"loss": 4.8815,
|
|
"mean_token_accuracy": 0.19085957556962968,
|
|
"num_tokens": 10004043.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.408733749389649,
|
|
"epoch": 4.5718608169440245,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003467884010259443,
|
|
"loss": 4.8907,
|
|
"mean_token_accuracy": 0.19512842446565629,
|
|
"num_tokens": 10014042.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.363145399093628,
|
|
"epoch": 4.576903681290974,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00034641235578893476,
|
|
"loss": 4.8653,
|
|
"mean_token_accuracy": 0.1962948814034462,
|
|
"num_tokens": 10025742.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.402016878128052,
|
|
"epoch": 4.581946545637923,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003460360886006761,
|
|
"loss": 4.9674,
|
|
"mean_token_accuracy": 0.18587110042572022,
|
|
"num_tokens": 10036727.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.4374651432037355,
|
|
"epoch": 4.586989409984872,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00034565960063061534,
|
|
"loss": 4.9226,
|
|
"mean_token_accuracy": 0.18302160054445266,
|
|
"num_tokens": 10047246.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.398792886734009,
|
|
"epoch": 4.592032274331821,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003452828930488859,
|
|
"loss": 4.9253,
|
|
"mean_token_accuracy": 0.1938019722700119,
|
|
"num_tokens": 10059750.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.455434322357178,
|
|
"epoch": 4.597075138678769,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00034490596702630355,
|
|
"loss": 4.9739,
|
|
"mean_token_accuracy": 0.1875145599246025,
|
|
"num_tokens": 10071431.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.463915681838989,
|
|
"epoch": 4.602118003025718,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003445288237343632,
|
|
"loss": 4.9644,
|
|
"mean_token_accuracy": 0.18932064026594161,
|
|
"num_tokens": 10081567.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.497029161453247,
|
|
"epoch": 4.607160867372667,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00034415146434523463,
|
|
"loss": 5.0002,
|
|
"mean_token_accuracy": 0.18606244027614594,
|
|
"num_tokens": 10092506.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.480929279327393,
|
|
"epoch": 4.6122037317196165,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00034377389003175984,
|
|
"loss": 4.9801,
|
|
"mean_token_accuracy": 0.1865699604153633,
|
|
"num_tokens": 10103896.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.372171783447266,
|
|
"epoch": 4.617246596066566,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003433961019674483,
|
|
"loss": 4.9572,
|
|
"mean_token_accuracy": 0.1918066293001175,
|
|
"num_tokens": 10113646.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.419704341888428,
|
|
"epoch": 4.622289460413515,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003430181013264741,
|
|
"loss": 5.0208,
|
|
"mean_token_accuracy": 0.1797514334321022,
|
|
"num_tokens": 10125097.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.5000776767730715,
|
|
"epoch": 4.627332324760464,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003426398892836719,
|
|
"loss": 4.9718,
|
|
"mean_token_accuracy": 0.18807972222566605,
|
|
"num_tokens": 10137396.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.483876419067383,
|
|
"epoch": 4.632375189107413,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003422614670145336,
|
|
"loss": 5.0777,
|
|
"mean_token_accuracy": 0.18274005502462387,
|
|
"num_tokens": 10147483.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.432821559906006,
|
|
"epoch": 4.637418053454362,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003418828356952041,
|
|
"loss": 4.951,
|
|
"mean_token_accuracy": 0.18625150471925736,
|
|
"num_tokens": 10159693.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.431114387512207,
|
|
"epoch": 4.642460917801311,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034150399650247846,
|
|
"loss": 4.84,
|
|
"mean_token_accuracy": 0.19102736562490463,
|
|
"num_tokens": 10170371.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.411733675003052,
|
|
"epoch": 4.64750378214826,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034112495061379764,
|
|
"loss": 4.9494,
|
|
"mean_token_accuracy": 0.19134079068899154,
|
|
"num_tokens": 10181487.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.41342191696167,
|
|
"epoch": 4.652546646495209,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034074569920724483,
|
|
"loss": 4.9766,
|
|
"mean_token_accuracy": 0.18301253765821457,
|
|
"num_tokens": 10193386.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.498093175888061,
|
|
"epoch": 4.6575895108421586,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003403662434615423,
|
|
"loss": 5.0328,
|
|
"mean_token_accuracy": 0.18311253786087037,
|
|
"num_tokens": 10205531.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.456884098052979,
|
|
"epoch": 4.662632375189108,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033998658455604736,
|
|
"loss": 4.9797,
|
|
"mean_token_accuracy": 0.18898638486862182,
|
|
"num_tokens": 10216586.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.431148862838745,
|
|
"epoch": 4.667675239536057,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033960672367074827,
|
|
"loss": 4.8987,
|
|
"mean_token_accuracy": 0.19037448465824128,
|
|
"num_tokens": 10227330.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.440126276016235,
|
|
"epoch": 4.672718103883006,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003392266619862619,
|
|
"loss": 5.0402,
|
|
"mean_token_accuracy": 0.17827739417552949,
|
|
"num_tokens": 10238295.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.449625444412232,
|
|
"epoch": 4.677760968229954,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003388464006838284,
|
|
"loss": 4.9925,
|
|
"mean_token_accuracy": 0.19437420070171357,
|
|
"num_tokens": 10249011.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.503025913238526,
|
|
"epoch": 4.682803832576903,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033846594094530874,
|
|
"loss": 4.9525,
|
|
"mean_token_accuracy": 0.18661330491304398,
|
|
"num_tokens": 10260986.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.403000020980835,
|
|
"epoch": 4.687846696923852,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003380852839531806,
|
|
"loss": 4.9642,
|
|
"mean_token_accuracy": 0.18330392688512803,
|
|
"num_tokens": 10271557.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.465458154678345,
|
|
"epoch": 4.6928895612708015,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033770443089053476,
|
|
"loss": 4.9798,
|
|
"mean_token_accuracy": 0.18609561026096344,
|
|
"num_tokens": 10281458.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.459237432479858,
|
|
"epoch": 4.697932425617751,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033732338294107116,
|
|
"loss": 4.9671,
|
|
"mean_token_accuracy": 0.1871896654367447,
|
|
"num_tokens": 10293336.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.432980871200561,
|
|
"epoch": 4.7029752899647,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033694214128909567,
|
|
"loss": 4.9673,
|
|
"mean_token_accuracy": 0.18993277698755265,
|
|
"num_tokens": 10303633.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.44464521408081,
|
|
"epoch": 4.708018154311649,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033656070711951633,
|
|
"loss": 4.9845,
|
|
"mean_token_accuracy": 0.18049918115139008,
|
|
"num_tokens": 10314776.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.4221460819244385,
|
|
"epoch": 4.713061018658598,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003361790816178391,
|
|
"loss": 4.9222,
|
|
"mean_token_accuracy": 0.19435497671365737,
|
|
"num_tokens": 10325792.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.3949438571929935,
|
|
"epoch": 4.718103883005547,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003357972659701649,
|
|
"loss": 5.0,
|
|
"mean_token_accuracy": 0.18047935217618943,
|
|
"num_tokens": 10339024.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.466359615325928,
|
|
"epoch": 4.723146747352496,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033541526136318564,
|
|
"loss": 4.9542,
|
|
"mean_token_accuracy": 0.18820637315511704,
|
|
"num_tokens": 10349198.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.476608657836914,
|
|
"epoch": 4.728189611699445,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003350330689841804,
|
|
"loss": 5.047,
|
|
"mean_token_accuracy": 0.17754877507686614,
|
|
"num_tokens": 10360925.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.471492719650269,
|
|
"epoch": 4.733232476046394,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003346506900210118,
|
|
"loss": 4.9786,
|
|
"mean_token_accuracy": 0.1891571193933487,
|
|
"num_tokens": 10372074.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.40898289680481,
|
|
"epoch": 4.7382753403933435,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003342681256621227,
|
|
"loss": 4.8793,
|
|
"mean_token_accuracy": 0.19103499352931977,
|
|
"num_tokens": 10384050.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.43700852394104,
|
|
"epoch": 4.743318204740293,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003338853770965319,
|
|
"loss": 4.9256,
|
|
"mean_token_accuracy": 0.18500882834196092,
|
|
"num_tokens": 10394466.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.408495807647705,
|
|
"epoch": 4.748361069087242,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033350244551383064,
|
|
"loss": 4.9981,
|
|
"mean_token_accuracy": 0.18811080902814864,
|
|
"num_tokens": 10405279.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.397424650192261,
|
|
"epoch": 4.753403933434191,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003331193321041793,
|
|
"loss": 4.9563,
|
|
"mean_token_accuracy": 0.1881820723414421,
|
|
"num_tokens": 10417504.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.425117683410645,
|
|
"epoch": 4.75844679778114,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033273603805830296,
|
|
"loss": 4.9777,
|
|
"mean_token_accuracy": 0.18997900187969208,
|
|
"num_tokens": 10428824.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.47361912727356,
|
|
"epoch": 4.763489662128089,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00033235256456748847,
|
|
"loss": 5.0229,
|
|
"mean_token_accuracy": 0.1850911021232605,
|
|
"num_tokens": 10440164.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.507453775405883,
|
|
"epoch": 4.768532526475038,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003319689128235804,
|
|
"loss": 4.9309,
|
|
"mean_token_accuracy": 0.19020796716213226,
|
|
"num_tokens": 10450216.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.436894702911377,
|
|
"epoch": 4.773575390821987,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00033158508401897707,
|
|
"loss": 4.9579,
|
|
"mean_token_accuracy": 0.19510447680950166,
|
|
"num_tokens": 10461521.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.441084003448486,
|
|
"epoch": 4.778618255168936,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003312010793466273,
|
|
"loss": 5.0663,
|
|
"mean_token_accuracy": 0.17662031203508377,
|
|
"num_tokens": 10473052.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.386702537536621,
|
|
"epoch": 4.783661119515885,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003308169000000265,
|
|
"loss": 4.9103,
|
|
"mean_token_accuracy": 0.19201598614454268,
|
|
"num_tokens": 10483197.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.4234694480896,
|
|
"epoch": 4.788703983862834,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003304325471732129,
|
|
"loss": 4.9217,
|
|
"mean_token_accuracy": 0.18737069070339202,
|
|
"num_tokens": 10493877.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.445645141601562,
|
|
"epoch": 4.793746848209783,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033004802206076396,
|
|
"loss": 4.9896,
|
|
"mean_token_accuracy": 0.1867102414369583,
|
|
"num_tokens": 10505457.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.451334476470947,
|
|
"epoch": 4.798789712556732,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032966332585779263,
|
|
"loss": 5.0065,
|
|
"mean_token_accuracy": 0.18804700821638107,
|
|
"num_tokens": 10516746.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.501697826385498,
|
|
"epoch": 4.803832576903681,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032927845975994356,
|
|
"loss": 4.987,
|
|
"mean_token_accuracy": 0.18748954087495803,
|
|
"num_tokens": 10527879.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.478173398971558,
|
|
"epoch": 4.80887544125063,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032889342496338953,
|
|
"loss": 5.043,
|
|
"mean_token_accuracy": 0.18037958443164825,
|
|
"num_tokens": 10539729.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.456065368652344,
|
|
"epoch": 4.813918305597579,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003285082226648273,
|
|
"loss": 5.061,
|
|
"mean_token_accuracy": 0.18717390894889832,
|
|
"num_tokens": 10551343.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.458584022521973,
|
|
"epoch": 4.8189611699445285,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003281228540614749,
|
|
"loss": 4.9506,
|
|
"mean_token_accuracy": 0.19011957943439484,
|
|
"num_tokens": 10562362.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.431289768218994,
|
|
"epoch": 4.824004034291478,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032773732035106657,
|
|
"loss": 4.9544,
|
|
"mean_token_accuracy": 0.18818462193012236,
|
|
"num_tokens": 10574203.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.452210330963135,
|
|
"epoch": 4.829046898638427,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032735162273185006,
|
|
"loss": 4.9991,
|
|
"mean_token_accuracy": 0.17896709591150284,
|
|
"num_tokens": 10585732.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.494658279418945,
|
|
"epoch": 4.834089762985376,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003269657624025824,
|
|
"loss": 5.0101,
|
|
"mean_token_accuracy": 0.19213832318782806,
|
|
"num_tokens": 10596414.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.473741817474365,
|
|
"epoch": 4.839132627332325,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003265797405625265,
|
|
"loss": 4.947,
|
|
"mean_token_accuracy": 0.189757277071476,
|
|
"num_tokens": 10608893.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.353061628341675,
|
|
"epoch": 4.844175491679274,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003261935584114471,
|
|
"loss": 4.869,
|
|
"mean_token_accuracy": 0.19499963223934175,
|
|
"num_tokens": 10619229.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.358573865890503,
|
|
"epoch": 4.849218356026223,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003258072171496072,
|
|
"loss": 4.9867,
|
|
"mean_token_accuracy": 0.18436321318149568,
|
|
"num_tokens": 10630811.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.45855302810669,
|
|
"epoch": 4.854261220373172,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003254207179777644,
|
|
"loss": 4.9808,
|
|
"mean_token_accuracy": 0.19197733849287033,
|
|
"num_tokens": 10641191.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.453788089752197,
|
|
"epoch": 4.8593040847201205,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000325034062097167,
|
|
"loss": 5.0329,
|
|
"mean_token_accuracy": 0.18242633044719697,
|
|
"num_tokens": 10652019.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.456726026535034,
|
|
"epoch": 4.86434694906707,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032464725070955046,
|
|
"loss": 5.0328,
|
|
"mean_token_accuracy": 0.18062737584114075,
|
|
"num_tokens": 10664820.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.48663558959961,
|
|
"epoch": 4.869389813414019,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032426028501713347,
|
|
"loss": 4.9731,
|
|
"mean_token_accuracy": 0.18408883810043336,
|
|
"num_tokens": 10676343.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.445742988586426,
|
|
"epoch": 4.874432677760968,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032387316622261425,
|
|
"loss": 5.0056,
|
|
"mean_token_accuracy": 0.18338582515716553,
|
|
"num_tokens": 10687464.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.407337713241577,
|
|
"epoch": 4.879475542107917,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032348589552916703,
|
|
"loss": 4.9456,
|
|
"mean_token_accuracy": 0.18678104132413864,
|
|
"num_tokens": 10698796.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.440946578979492,
|
|
"epoch": 4.884518406454866,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003230984741404381,
|
|
"loss": 4.9428,
|
|
"mean_token_accuracy": 0.18648777455091475,
|
|
"num_tokens": 10708759.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.520267152786255,
|
|
"epoch": 4.889561270801815,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003227109032605419,
|
|
"loss": 4.9954,
|
|
"mean_token_accuracy": 0.18635909855365754,
|
|
"num_tokens": 10718518.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.474165201187134,
|
|
"epoch": 4.894604135148764,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003223231840940578,
|
|
"loss": 5.0391,
|
|
"mean_token_accuracy": 0.18102305382490158,
|
|
"num_tokens": 10730091.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.400430488586426,
|
|
"epoch": 4.8996469994957135,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032193531784602595,
|
|
"loss": 4.9725,
|
|
"mean_token_accuracy": 0.18741223067045212,
|
|
"num_tokens": 10739664.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.423764419555664,
|
|
"epoch": 4.904689863842663,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00032154730572194346,
|
|
"loss": 4.9291,
|
|
"mean_token_accuracy": 0.19145375043153762,
|
|
"num_tokens": 10749573.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.503514146804809,
|
|
"epoch": 4.909732728189612,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000321159148927761,
|
|
"loss": 5.0556,
|
|
"mean_token_accuracy": 0.18899090141057967,
|
|
"num_tokens": 10760455.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.5097449779510494,
|
|
"epoch": 4.914775592536561,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000320770848669879,
|
|
"loss": 5.0187,
|
|
"mean_token_accuracy": 0.18463164716959,
|
|
"num_tokens": 10771845.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.438835954666137,
|
|
"epoch": 4.91981845688351,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032038240615514345,
|
|
"loss": 4.9946,
|
|
"mean_token_accuracy": 0.18855448365211486,
|
|
"num_tokens": 10783812.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.403696203231812,
|
|
"epoch": 4.924861321230459,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003199938225908426,
|
|
"loss": 4.9135,
|
|
"mean_token_accuracy": 0.18758562356233596,
|
|
"num_tokens": 10794390.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.430738544464111,
|
|
"epoch": 4.929904185577408,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003196050991847033,
|
|
"loss": 4.9578,
|
|
"mean_token_accuracy": 0.18548450171947478,
|
|
"num_tokens": 10805521.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.4192510604858395,
|
|
"epoch": 4.934947049924357,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003192162371448868,
|
|
"loss": 4.9664,
|
|
"mean_token_accuracy": 0.1857212409377098,
|
|
"num_tokens": 10816265.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.4131299495697025,
|
|
"epoch": 4.939989914271306,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00031882723767998517,
|
|
"loss": 4.9921,
|
|
"mean_token_accuracy": 0.18737585991621017,
|
|
"num_tokens": 10826285.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.41847596168518,
|
|
"epoch": 4.9450327786182555,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031843810199901795,
|
|
"loss": 4.9347,
|
|
"mean_token_accuracy": 0.19147606939077377,
|
|
"num_tokens": 10836817.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.45005612373352,
|
|
"epoch": 4.950075642965205,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003180488313114277,
|
|
"loss": 4.9638,
|
|
"mean_token_accuracy": 0.18842111229896547,
|
|
"num_tokens": 10848186.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.374664354324341,
|
|
"epoch": 4.955118507312154,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003176594268270766,
|
|
"loss": 4.9524,
|
|
"mean_token_accuracy": 0.19444430619478226,
|
|
"num_tokens": 10858851.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.424822568893433,
|
|
"epoch": 4.960161371659103,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00031726988975624293,
|
|
"loss": 4.9957,
|
|
"mean_token_accuracy": 0.18679312616586685,
|
|
"num_tokens": 10870242.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.4419248580932615,
|
|
"epoch": 4.965204236006051,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00031688022130961675,
|
|
"loss": 4.9299,
|
|
"mean_token_accuracy": 0.18993133008480073,
|
|
"num_tokens": 10881580.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.439469051361084,
|
|
"epoch": 4.970247100353,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003164904226982967,
|
|
"loss": 4.9934,
|
|
"mean_token_accuracy": 0.19487899541854858,
|
|
"num_tokens": 10893140.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.448909187316895,
|
|
"epoch": 4.975289964699949,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00031610049513378574,
|
|
"loss": 4.9945,
|
|
"mean_token_accuracy": 0.1908086955547333,
|
|
"num_tokens": 10904373.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.507550525665283,
|
|
"epoch": 4.9803328290468984,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00031571043982798786,
|
|
"loss": 5.0025,
|
|
"mean_token_accuracy": 0.18820565193891525,
|
|
"num_tokens": 10914155.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.516139650344849,
|
|
"epoch": 4.985375693393848,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003153202579932039,
|
|
"loss": 5.009,
|
|
"mean_token_accuracy": 0.1866231933236122,
|
|
"num_tokens": 10926325.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.493792581558227,
|
|
"epoch": 4.990418557740797,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00031492995084212794,
|
|
"loss": 5.0369,
|
|
"mean_token_accuracy": 0.18808876425027848,
|
|
"num_tokens": 10938799.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.443079853057862,
|
|
"epoch": 4.995461422087746,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003145395195878437,
|
|
"loss": 4.8587,
|
|
"mean_token_accuracy": 0.19427645206451416,
|
|
"num_tokens": 10949918.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.373980893029107,
|
|
"epoch": 5.0,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 0.00031414896544382047,
|
|
"loss": 4.8479,
|
|
"mean_token_accuracy": 0.1932825760708915,
|
|
"num_tokens": 10958200.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.377087545394898,
|
|
"epoch": 5.005042864346949,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003137582896239095,
|
|
"loss": 4.7798,
|
|
"mean_token_accuracy": 0.19978601783514022,
|
|
"num_tokens": 10969020.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.3690825462341305,
|
|
"epoch": 5.010085728693898,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003133674933423404,
|
|
"loss": 4.6958,
|
|
"mean_token_accuracy": 0.21049367189407348,
|
|
"num_tokens": 10978191.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.39783124923706,
|
|
"epoch": 5.015128593040847,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00031297657781371687,
|
|
"loss": 4.769,
|
|
"mean_token_accuracy": 0.20987659692764282,
|
|
"num_tokens": 10989022.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.46621470451355,
|
|
"epoch": 5.0201714573877965,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003125855442530135,
|
|
"loss": 4.8209,
|
|
"mean_token_accuracy": 0.20026374459266663,
|
|
"num_tokens": 11002727.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.405496120452881,
|
|
"epoch": 5.025214321734746,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00031219439387557155,
|
|
"loss": 4.762,
|
|
"mean_token_accuracy": 0.2060738682746887,
|
|
"num_tokens": 11013723.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.470185852050781,
|
|
"epoch": 5.030257186081695,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031180312789709535,
|
|
"loss": 4.7332,
|
|
"mean_token_accuracy": 0.20676991194486619,
|
|
"num_tokens": 11024676.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.4253129959106445,
|
|
"epoch": 5.035300050428644,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003114117475336487,
|
|
"loss": 4.7794,
|
|
"mean_token_accuracy": 0.2019949272274971,
|
|
"num_tokens": 11036314.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.399463033676147,
|
|
"epoch": 5.040342914775593,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003110202540016509,
|
|
"loss": 4.7499,
|
|
"mean_token_accuracy": 0.2073090746998787,
|
|
"num_tokens": 11046321.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.040342914775593,
|
|
"eval_entropy": 5.259037116708343,
|
|
"eval_loss": 5.665579795837402,
|
|
"eval_mean_token_accuracy": 0.1547874487538374,
|
|
"eval_num_tokens": 11046321.0,
|
|
"eval_runtime": 2.0489,
|
|
"eval_samples_per_second": 1531.555,
|
|
"eval_steps_per_second": 191.81,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.422780227661133,
|
|
"epoch": 5.045385779122542,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00031062864851787267,
|
|
"loss": 4.7483,
|
|
"mean_token_accuracy": 0.20660395473241805,
|
|
"num_tokens": 11059021.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.413686323165893,
|
|
"epoch": 5.05042864346949,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00031023693229943294,
|
|
"loss": 4.7311,
|
|
"mean_token_accuracy": 0.20787329226732254,
|
|
"num_tokens": 11069785.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.442879867553711,
|
|
"epoch": 5.055471507816439,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003098451065637948,
|
|
"loss": 4.789,
|
|
"mean_token_accuracy": 0.20268239378929137,
|
|
"num_tokens": 11079439.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.377065849304199,
|
|
"epoch": 5.0605143721633885,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030945317252876163,
|
|
"loss": 4.7887,
|
|
"mean_token_accuracy": 0.20264918059110643,
|
|
"num_tokens": 11089933.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.395506191253662,
|
|
"epoch": 5.065557236510338,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00030906113141247353,
|
|
"loss": 4.7127,
|
|
"mean_token_accuracy": 0.2048274829983711,
|
|
"num_tokens": 11099743.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.2882936000823975,
|
|
"epoch": 5.070600100857287,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00030866898443340324,
|
|
"loss": 4.66,
|
|
"mean_token_accuracy": 0.21423418521881105,
|
|
"num_tokens": 11109993.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.454429244995117,
|
|
"epoch": 5.075642965204236,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003082767328103527,
|
|
"loss": 4.7316,
|
|
"mean_token_accuracy": 0.19962920844554902,
|
|
"num_tokens": 11121071.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.468168592453003,
|
|
"epoch": 5.080685829551185,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00030788437776244895,
|
|
"loss": 4.707,
|
|
"mean_token_accuracy": 0.21180965304374694,
|
|
"num_tokens": 11130402.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.363940095901489,
|
|
"epoch": 5.085728693898134,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003074919205091405,
|
|
"loss": 4.815,
|
|
"mean_token_accuracy": 0.200840762257576,
|
|
"num_tokens": 11141272.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.448993110656739,
|
|
"epoch": 5.090771558245083,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003070993622701936,
|
|
"loss": 4.7531,
|
|
"mean_token_accuracy": 0.2033784419298172,
|
|
"num_tokens": 11152254.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.383047008514405,
|
|
"epoch": 5.095814422592032,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00030670670426568837,
|
|
"loss": 4.6806,
|
|
"mean_token_accuracy": 0.21028010547161102,
|
|
"num_tokens": 11162848.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.356310844421387,
|
|
"epoch": 5.100857286938981,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003063139477160147,
|
|
"loss": 4.6863,
|
|
"mean_token_accuracy": 0.20928248167037963,
|
|
"num_tokens": 11173896.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.398014307022095,
|
|
"epoch": 5.1059001512859306,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030592109384186935,
|
|
"loss": 4.7307,
|
|
"mean_token_accuracy": 0.21019263416528702,
|
|
"num_tokens": 11184423.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.388689661026001,
|
|
"epoch": 5.11094301563288,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000305528143864251,
|
|
"loss": 4.7624,
|
|
"mean_token_accuracy": 0.2092597559094429,
|
|
"num_tokens": 11195128.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.373467206954956,
|
|
"epoch": 5.115985879979829,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00030513509900445725,
|
|
"loss": 4.7285,
|
|
"mean_token_accuracy": 0.21020106822252274,
|
|
"num_tokens": 11205628.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.383992910385132,
|
|
"epoch": 5.121028744326778,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00030474196048408065,
|
|
"loss": 4.7651,
|
|
"mean_token_accuracy": 0.20403726100921632,
|
|
"num_tokens": 11216835.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.391399335861206,
|
|
"epoch": 5.126071608673727,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003043487295250049,
|
|
"loss": 4.7863,
|
|
"mean_token_accuracy": 0.19799292981624603,
|
|
"num_tokens": 11228763.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.366147708892822,
|
|
"epoch": 5.131114473020676,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003039554073494006,
|
|
"loss": 4.7319,
|
|
"mean_token_accuracy": 0.21267775297164918,
|
|
"num_tokens": 11239432.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.470732545852661,
|
|
"epoch": 5.136157337367624,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003035619951797224,
|
|
"loss": 4.8514,
|
|
"mean_token_accuracy": 0.1952908530831337,
|
|
"num_tokens": 11249618.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.432192802429199,
|
|
"epoch": 5.1412002017145735,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003031684942387043,
|
|
"loss": 4.7746,
|
|
"mean_token_accuracy": 0.2070817768573761,
|
|
"num_tokens": 11260359.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.400609254837036,
|
|
"epoch": 5.146243066061523,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00030277490574935633,
|
|
"loss": 4.8143,
|
|
"mean_token_accuracy": 0.1999165177345276,
|
|
"num_tokens": 11272003.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.39234561920166,
|
|
"epoch": 5.151285930408472,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003023812309349605,
|
|
"loss": 4.7549,
|
|
"mean_token_accuracy": 0.20627606362104417,
|
|
"num_tokens": 11283649.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.5274261951446535,
|
|
"epoch": 5.156328794755421,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003019874710190674,
|
|
"loss": 4.8718,
|
|
"mean_token_accuracy": 0.1988748401403427,
|
|
"num_tokens": 11295844.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.449365043640137,
|
|
"epoch": 5.16137165910237,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003015936272254918,
|
|
"loss": 4.7737,
|
|
"mean_token_accuracy": 0.2059719055891037,
|
|
"num_tokens": 11306676.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.384851932525635,
|
|
"epoch": 5.166414523449319,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003011997007783094,
|
|
"loss": 4.7456,
|
|
"mean_token_accuracy": 0.20340470224618912,
|
|
"num_tokens": 11317801.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.382043647766113,
|
|
"epoch": 5.171457387796268,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003008056929018525,
|
|
"loss": 4.7306,
|
|
"mean_token_accuracy": 0.20402662605047225,
|
|
"num_tokens": 11327530.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.5060724258422855,
|
|
"epoch": 5.176500252143217,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00030041160482070686,
|
|
"loss": 4.7442,
|
|
"mean_token_accuracy": 0.20388694554567338,
|
|
"num_tokens": 11338784.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.406298446655273,
|
|
"epoch": 5.181543116490166,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00030001743775970714,
|
|
"loss": 4.7815,
|
|
"mean_token_accuracy": 0.20606305152177812,
|
|
"num_tokens": 11348666.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.387162542343139,
|
|
"epoch": 5.1865859808371155,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00029962319294393383,
|
|
"loss": 4.73,
|
|
"mean_token_accuracy": 0.2070406809449196,
|
|
"num_tokens": 11358247.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.343744802474975,
|
|
"epoch": 5.191628845184065,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002992288715987088,
|
|
"loss": 4.754,
|
|
"mean_token_accuracy": 0.20762002915143968,
|
|
"num_tokens": 11369897.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.429714822769165,
|
|
"epoch": 5.196671709531014,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002988344749495919,
|
|
"loss": 4.8356,
|
|
"mean_token_accuracy": 0.20064053535461426,
|
|
"num_tokens": 11381747.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.450630521774292,
|
|
"epoch": 5.201714573877963,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00029844000422237714,
|
|
"loss": 4.8277,
|
|
"mean_token_accuracy": 0.19658990800380707,
|
|
"num_tokens": 11393725.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.3230180740356445,
|
|
"epoch": 5.206757438224912,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00029804546064308854,
|
|
"loss": 4.7771,
|
|
"mean_token_accuracy": 0.2036718338727951,
|
|
"num_tokens": 11405223.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.4090415954589846,
|
|
"epoch": 5.211800302571861,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00029765084543797657,
|
|
"loss": 4.7617,
|
|
"mean_token_accuracy": 0.20066658705472945,
|
|
"num_tokens": 11416334.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.387953090667724,
|
|
"epoch": 5.21684316691881,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002972561598335145,
|
|
"loss": 4.7496,
|
|
"mean_token_accuracy": 0.20192946195602418,
|
|
"num_tokens": 11427202.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.42350058555603,
|
|
"epoch": 5.221886031265759,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002968614050563943,
|
|
"loss": 4.8457,
|
|
"mean_token_accuracy": 0.19637791067361832,
|
|
"num_tokens": 11438302.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.411188840866089,
|
|
"epoch": 5.2269288956127085,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002964665823335229,
|
|
"loss": 4.7577,
|
|
"mean_token_accuracy": 0.20150525718927384,
|
|
"num_tokens": 11450551.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.3604906558990475,
|
|
"epoch": 5.231971759959657,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00029607169289201836,
|
|
"loss": 4.741,
|
|
"mean_token_accuracy": 0.21597910523414612,
|
|
"num_tokens": 11460907.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.375416374206543,
|
|
"epoch": 5.237014624306606,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00029567673795920625,
|
|
"loss": 4.7224,
|
|
"mean_token_accuracy": 0.207424658536911,
|
|
"num_tokens": 11472290.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.276330804824829,
|
|
"epoch": 5.242057488653555,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00029528171876261555,
|
|
"loss": 4.7787,
|
|
"mean_token_accuracy": 0.20905390083789827,
|
|
"num_tokens": 11484946.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.454510831832886,
|
|
"epoch": 5.247100353000504,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002948866365299751,
|
|
"loss": 4.8389,
|
|
"mean_token_accuracy": 0.1950281471014023,
|
|
"num_tokens": 11494845.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.511236238479614,
|
|
"epoch": 5.252143217347453,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002944914924892096,
|
|
"loss": 4.852,
|
|
"mean_token_accuracy": 0.19927518665790558,
|
|
"num_tokens": 11506533.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.375716781616211,
|
|
"epoch": 5.257186081694402,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002940962878684358,
|
|
"loss": 4.7628,
|
|
"mean_token_accuracy": 0.20310329794883727,
|
|
"num_tokens": 11516985.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.324270009994507,
|
|
"epoch": 5.262228946041351,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002937010238959588,
|
|
"loss": 4.7455,
|
|
"mean_token_accuracy": 0.20899288654327391,
|
|
"num_tokens": 11526203.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.351016664505005,
|
|
"epoch": 5.2672718103883005,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002933057018002681,
|
|
"loss": 4.8466,
|
|
"mean_token_accuracy": 0.2013192817568779,
|
|
"num_tokens": 11536676.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.415819597244263,
|
|
"epoch": 5.27231467473525,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000292910322810034,
|
|
"loss": 4.7999,
|
|
"mean_token_accuracy": 0.20850418359041215,
|
|
"num_tokens": 11548466.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.401542377471924,
|
|
"epoch": 5.277357539082199,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00029251488815410344,
|
|
"loss": 4.8248,
|
|
"mean_token_accuracy": 0.19646383076906204,
|
|
"num_tokens": 11560065.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.407020950317383,
|
|
"epoch": 5.282400403429148,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00029211939906149644,
|
|
"loss": 4.8254,
|
|
"mean_token_accuracy": 0.1985974207520485,
|
|
"num_tokens": 11571718.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.37835693359375,
|
|
"epoch": 5.287443267776097,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00029172385676140225,
|
|
"loss": 4.8058,
|
|
"mean_token_accuracy": 0.20265444219112397,
|
|
"num_tokens": 11583046.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.317554950714111,
|
|
"epoch": 5.292486132123046,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00029132826248317537,
|
|
"loss": 4.6898,
|
|
"mean_token_accuracy": 0.20848133116960527,
|
|
"num_tokens": 11593744.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.385888576507568,
|
|
"epoch": 5.297528996469995,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002909326174563322,
|
|
"loss": 4.7439,
|
|
"mean_token_accuracy": 0.20916520357131957,
|
|
"num_tokens": 11604337.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.394536018371582,
|
|
"epoch": 5.302571860816944,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00029053692291054636,
|
|
"loss": 4.8092,
|
|
"mean_token_accuracy": 0.2000624418258667,
|
|
"num_tokens": 11615712.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.324703073501587,
|
|
"epoch": 5.307614725163893,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00029014118007564564,
|
|
"loss": 4.7372,
|
|
"mean_token_accuracy": 0.20561733543872834,
|
|
"num_tokens": 11625846.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.344365358352661,
|
|
"epoch": 5.3126575895108425,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002897453901816081,
|
|
"loss": 4.7914,
|
|
"mean_token_accuracy": 0.20059997588396072,
|
|
"num_tokens": 11635968.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.4661658763885494,
|
|
"epoch": 5.317700453857791,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00028934955445855773,
|
|
"loss": 4.8665,
|
|
"mean_token_accuracy": 0.20574951171875,
|
|
"num_tokens": 11646858.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.3323393821716305,
|
|
"epoch": 5.32274331820474,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00028895367413676095,
|
|
"loss": 4.7689,
|
|
"mean_token_accuracy": 0.20801410377025603,
|
|
"num_tokens": 11657728.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.462389087677002,
|
|
"epoch": 5.327786182551689,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002885577504466232,
|
|
"loss": 4.8162,
|
|
"mean_token_accuracy": 0.20330755561590194,
|
|
"num_tokens": 11668288.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.37921552658081,
|
|
"epoch": 5.332829046898638,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028816178461868425,
|
|
"loss": 4.7375,
|
|
"mean_token_accuracy": 0.2022107347846031,
|
|
"num_tokens": 11679883.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.415158128738403,
|
|
"epoch": 5.337871911245587,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002877657778836151,
|
|
"loss": 4.8363,
|
|
"mean_token_accuracy": 0.19621689319610597,
|
|
"num_tokens": 11690643.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.428422117233277,
|
|
"epoch": 5.342914775592536,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002873697314722138,
|
|
"loss": 4.8324,
|
|
"mean_token_accuracy": 0.20309034287929534,
|
|
"num_tokens": 11701687.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.359555959701538,
|
|
"epoch": 5.3479576399394855,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002869736466154018,
|
|
"loss": 4.7564,
|
|
"mean_token_accuracy": 0.20772893279790877,
|
|
"num_tokens": 11713432.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.408741807937622,
|
|
"epoch": 5.353000504286435,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00028657752454421974,
|
|
"loss": 4.8059,
|
|
"mean_token_accuracy": 0.20484850853681563,
|
|
"num_tokens": 11725009.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.435963106155396,
|
|
"epoch": 5.358043368633384,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002861813664898245,
|
|
"loss": 4.7478,
|
|
"mean_token_accuracy": 0.21129528284072877,
|
|
"num_tokens": 11735576.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.396055507659912,
|
|
"epoch": 5.363086232980333,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00028578517368348425,
|
|
"loss": 4.7527,
|
|
"mean_token_accuracy": 0.2051149368286133,
|
|
"num_tokens": 11746334.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.342900180816651,
|
|
"epoch": 5.368129097327282,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002853889473565753,
|
|
"loss": 4.7693,
|
|
"mean_token_accuracy": 0.2017793208360672,
|
|
"num_tokens": 11757093.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.349963808059693,
|
|
"epoch": 5.373171961674231,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002849926887405784,
|
|
"loss": 4.7967,
|
|
"mean_token_accuracy": 0.19983594566583635,
|
|
"num_tokens": 11767141.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.512475156784058,
|
|
"epoch": 5.37821482602118,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002845963990670744,
|
|
"loss": 4.9514,
|
|
"mean_token_accuracy": 0.19021792709827423,
|
|
"num_tokens": 11778944.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.40713062286377,
|
|
"epoch": 5.383257690368129,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00028420007956774077,
|
|
"loss": 4.6902,
|
|
"mean_token_accuracy": 0.20905327051877975,
|
|
"num_tokens": 11789081.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.3399951457977295,
|
|
"epoch": 5.388300554715078,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00028380373147434746,
|
|
"loss": 4.7737,
|
|
"mean_token_accuracy": 0.2040804609656334,
|
|
"num_tokens": 11800626.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.378877353668213,
|
|
"epoch": 5.3933434190620275,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00028340735601875376,
|
|
"loss": 4.7866,
|
|
"mean_token_accuracy": 0.19718751609325408,
|
|
"num_tokens": 11810899.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.417437934875489,
|
|
"epoch": 5.398386283408977,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002830109544329036,
|
|
"loss": 4.8029,
|
|
"mean_token_accuracy": 0.20603054910898208,
|
|
"num_tokens": 11822599.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.413007068634033,
|
|
"epoch": 5.403429147755926,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00028261452794882223,
|
|
"loss": 4.8487,
|
|
"mean_token_accuracy": 0.20448752641677856,
|
|
"num_tokens": 11836199.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.285883283615112,
|
|
"epoch": 5.408472012102875,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00028221807779861243,
|
|
"loss": 4.6572,
|
|
"mean_token_accuracy": 0.21588353663682938,
|
|
"num_tokens": 11846220.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.366286277770996,
|
|
"epoch": 5.413514876449823,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002818216052144502,
|
|
"loss": 4.7719,
|
|
"mean_token_accuracy": 0.20182025879621507,
|
|
"num_tokens": 11857356.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.40411696434021,
|
|
"epoch": 5.418557740796772,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002814251114285816,
|
|
"loss": 4.759,
|
|
"mean_token_accuracy": 0.20002357810735702,
|
|
"num_tokens": 11869088.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.362497329711914,
|
|
"epoch": 5.423600605143721,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002810285976733186,
|
|
"loss": 4.7575,
|
|
"mean_token_accuracy": 0.2007217139005661,
|
|
"num_tokens": 11878805.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.344033241271973,
|
|
"epoch": 5.4286434694906704,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002806320651810349,
|
|
"loss": 4.7229,
|
|
"mean_token_accuracy": 0.20748991668224334,
|
|
"num_tokens": 11889656.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.390753555297851,
|
|
"epoch": 5.43368633383762,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002802355151841627,
|
|
"loss": 4.7644,
|
|
"mean_token_accuracy": 0.20545733124017715,
|
|
"num_tokens": 11901816.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.306405162811279,
|
|
"epoch": 5.438729198184569,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002798389489151886,
|
|
"loss": 4.7576,
|
|
"mean_token_accuracy": 0.2034154698252678,
|
|
"num_tokens": 11915049.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.366006183624267,
|
|
"epoch": 5.443772062531518,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00027944236760664947,
|
|
"loss": 4.7347,
|
|
"mean_token_accuracy": 0.2040851205587387,
|
|
"num_tokens": 11925574.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.373019075393676,
|
|
"epoch": 5.448814926878467,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00027904577249112945,
|
|
"loss": 4.7729,
|
|
"mean_token_accuracy": 0.20976183712482452,
|
|
"num_tokens": 11936790.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.37097053527832,
|
|
"epoch": 5.453857791225416,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002786491648012552,
|
|
"loss": 4.7702,
|
|
"mean_token_accuracy": 0.19968913346529008,
|
|
"num_tokens": 11947430.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.392760419845581,
|
|
"epoch": 5.458900655572365,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002782525457696926,
|
|
"loss": 4.8093,
|
|
"mean_token_accuracy": 0.2016148239374161,
|
|
"num_tokens": 11959385.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.367619800567627,
|
|
"epoch": 5.463943519919314,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002778559166291427,
|
|
"loss": 4.7897,
|
|
"mean_token_accuracy": 0.20243736058473588,
|
|
"num_tokens": 11972117.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.404072093963623,
|
|
"epoch": 5.468986384266263,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002774592786123379,
|
|
"loss": 4.8294,
|
|
"mean_token_accuracy": 0.19929574579000472,
|
|
"num_tokens": 11984195.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.384224510192871,
|
|
"epoch": 5.4740292486132125,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002770626329520386,
|
|
"loss": 4.7591,
|
|
"mean_token_accuracy": 0.20624046474695207,
|
|
"num_tokens": 11994875.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.42634220123291,
|
|
"epoch": 5.479072112960162,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002766659808810285,
|
|
"loss": 4.8343,
|
|
"mean_token_accuracy": 0.20160355269908906,
|
|
"num_tokens": 12006408.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.4102010250091555,
|
|
"epoch": 5.484114977307111,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027626932363211134,
|
|
"loss": 4.7882,
|
|
"mean_token_accuracy": 0.2065979465842247,
|
|
"num_tokens": 12016764.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.375625371932983,
|
|
"epoch": 5.48915784165406,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002758726624381071,
|
|
"loss": 4.7426,
|
|
"mean_token_accuracy": 0.21021603494882585,
|
|
"num_tokens": 12027900.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.3879789352417,
|
|
"epoch": 5.494200706001009,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002754759985318478,
|
|
"loss": 4.8359,
|
|
"mean_token_accuracy": 0.19971801191568375,
|
|
"num_tokens": 12038641.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.400270891189575,
|
|
"epoch": 5.499243570347957,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002750793331461741,
|
|
"loss": 4.8084,
|
|
"mean_token_accuracy": 0.19689522981643676,
|
|
"num_tokens": 12050938.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.380944490432739,
|
|
"epoch": 5.504286434694906,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002746826675139313,
|
|
"loss": 4.7536,
|
|
"mean_token_accuracy": 0.2038170129060745,
|
|
"num_tokens": 12061483.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.428875350952149,
|
|
"epoch": 5.509329299041855,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00027428600286796503,
|
|
"loss": 4.8109,
|
|
"mean_token_accuracy": 0.19564636945724487,
|
|
"num_tokens": 12073214.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.413195037841797,
|
|
"epoch": 5.5143721633888045,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00027388934044111836,
|
|
"loss": 4.7896,
|
|
"mean_token_accuracy": 0.20607620030641555,
|
|
"num_tokens": 12082536.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.447469186782837,
|
|
"epoch": 5.519415027735754,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002734926814662271,
|
|
"loss": 4.8629,
|
|
"mean_token_accuracy": 0.1926976814866066,
|
|
"num_tokens": 12094867.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.416417980194092,
|
|
"epoch": 5.524457892082703,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00027309602717611673,
|
|
"loss": 4.8627,
|
|
"mean_token_accuracy": 0.20292915999889374,
|
|
"num_tokens": 12105385.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.364644479751587,
|
|
"epoch": 5.529500756429652,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00027269937880359786,
|
|
"loss": 4.7571,
|
|
"mean_token_accuracy": 0.2100677892565727,
|
|
"num_tokens": 12117652.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.402087545394897,
|
|
"epoch": 5.534543620776601,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00027230273758146263,
|
|
"loss": 4.8328,
|
|
"mean_token_accuracy": 0.20247368216514589,
|
|
"num_tokens": 12128372.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.426740074157715,
|
|
"epoch": 5.53958648512355,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002719061047424813,
|
|
"loss": 4.818,
|
|
"mean_token_accuracy": 0.2003900319337845,
|
|
"num_tokens": 12138943.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.3167243003845215,
|
|
"epoch": 5.544629349470499,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002715094815193978,
|
|
"loss": 4.7368,
|
|
"mean_token_accuracy": 0.20867759585380555,
|
|
"num_tokens": 12149329.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.331606101989746,
|
|
"epoch": 5.549672213817448,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002711128691449262,
|
|
"loss": 4.7266,
|
|
"mean_token_accuracy": 0.20735112428665162,
|
|
"num_tokens": 12160498.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.4235076904296875,
|
|
"epoch": 5.5547150781643975,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000270716268851747,
|
|
"loss": 4.8422,
|
|
"mean_token_accuracy": 0.19248930662870406,
|
|
"num_tokens": 12171110.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.297978353500366,
|
|
"epoch": 5.559757942511347,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00027031968187250305,
|
|
"loss": 4.7609,
|
|
"mean_token_accuracy": 0.2007995918393135,
|
|
"num_tokens": 12182706.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.36890983581543,
|
|
"epoch": 5.564800806858296,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002699231094397957,
|
|
"loss": 4.7555,
|
|
"mean_token_accuracy": 0.20540590286254884,
|
|
"num_tokens": 12194023.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.39328031539917,
|
|
"epoch": 5.569843671205245,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002695265527861815,
|
|
"loss": 4.7822,
|
|
"mean_token_accuracy": 0.19761642217636108,
|
|
"num_tokens": 12204165.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.416159820556641,
|
|
"epoch": 5.574886535552194,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002691300131441674,
|
|
"loss": 4.8489,
|
|
"mean_token_accuracy": 0.1993971899151802,
|
|
"num_tokens": 12214840.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.398578071594239,
|
|
"epoch": 5.579929399899143,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002687334917462079,
|
|
"loss": 4.8122,
|
|
"mean_token_accuracy": 0.20347722321748735,
|
|
"num_tokens": 12226454.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.382529306411743,
|
|
"epoch": 5.584972264246092,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002683369898247005,
|
|
"loss": 4.8185,
|
|
"mean_token_accuracy": 0.19594880789518357,
|
|
"num_tokens": 12236958.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.407418203353882,
|
|
"epoch": 5.590015128593041,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002679405086119825,
|
|
"loss": 4.7829,
|
|
"mean_token_accuracy": 0.20016122311353685,
|
|
"num_tokens": 12248050.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.418981838226318,
|
|
"epoch": 5.59505799293999,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026754404934032657,
|
|
"loss": 4.7709,
|
|
"mean_token_accuracy": 0.19695729166269302,
|
|
"num_tokens": 12258230.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.2918905258178714,
|
|
"epoch": 5.600100857286939,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002671476132419374,
|
|
"loss": 4.7131,
|
|
"mean_token_accuracy": 0.2035423129796982,
|
|
"num_tokens": 12268708.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.359189367294311,
|
|
"epoch": 5.605143721633888,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002667512015489472,
|
|
"loss": 4.8098,
|
|
"mean_token_accuracy": 0.198850516974926,
|
|
"num_tokens": 12279363.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.361038446426392,
|
|
"epoch": 5.610186585980837,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000266354815493413,
|
|
"loss": 4.6775,
|
|
"mean_token_accuracy": 0.2127465561032295,
|
|
"num_tokens": 12290525.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.3382140636444095,
|
|
"epoch": 5.615229450327786,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00026595845630731166,
|
|
"loss": 4.7394,
|
|
"mean_token_accuracy": 0.21278925836086274,
|
|
"num_tokens": 12301404.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.28578896522522,
|
|
"epoch": 5.620272314674735,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026556212522253666,
|
|
"loss": 4.7747,
|
|
"mean_token_accuracy": 0.20052016228437425,
|
|
"num_tokens": 12312601.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.371021366119384,
|
|
"epoch": 5.625315179021684,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002651658234708941,
|
|
"loss": 4.7389,
|
|
"mean_token_accuracy": 0.20447789281606674,
|
|
"num_tokens": 12324192.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.4337811946868895,
|
|
"epoch": 5.630358043368633,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026476955228409924,
|
|
"loss": 4.7604,
|
|
"mean_token_accuracy": 0.21026771813631057,
|
|
"num_tokens": 12335358.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.328546142578125,
|
|
"epoch": 5.635400907715582,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00026437331289377175,
|
|
"loss": 4.7659,
|
|
"mean_token_accuracy": 0.2018200382590294,
|
|
"num_tokens": 12347000.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.434003067016602,
|
|
"epoch": 5.640443772062532,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002639771065314329,
|
|
"loss": 4.8828,
|
|
"mean_token_accuracy": 0.19211796522140503,
|
|
"num_tokens": 12358067.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.412813138961792,
|
|
"epoch": 5.645486636409481,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026358093442850134,
|
|
"loss": 4.7987,
|
|
"mean_token_accuracy": 0.19619182050228118,
|
|
"num_tokens": 12369359.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.379443597793579,
|
|
"epoch": 5.65052950075643,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002631847978162889,
|
|
"loss": 4.8384,
|
|
"mean_token_accuracy": 0.20223764479160308,
|
|
"num_tokens": 12380789.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.349356412887573,
|
|
"epoch": 5.655572365103379,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00026278869792599745,
|
|
"loss": 4.7751,
|
|
"mean_token_accuracy": 0.20764461308717727,
|
|
"num_tokens": 12392591.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.34460825920105,
|
|
"epoch": 5.660615229450328,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002623926359887145,
|
|
"loss": 4.8184,
|
|
"mean_token_accuracy": 0.19706828147172928,
|
|
"num_tokens": 12404269.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.430381488800049,
|
|
"epoch": 5.665658093797277,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00026199661323540973,
|
|
"loss": 4.8309,
|
|
"mean_token_accuracy": 0.20572761446237564,
|
|
"num_tokens": 12414356.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.4070611000061035,
|
|
"epoch": 5.670700958144226,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026160063089693096,
|
|
"loss": 4.7887,
|
|
"mean_token_accuracy": 0.20611845701932907,
|
|
"num_tokens": 12424480.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.383074998855591,
|
|
"epoch": 5.675743822491175,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002612046902040006,
|
|
"loss": 4.8203,
|
|
"mean_token_accuracy": 0.19679659008979797,
|
|
"num_tokens": 12437253.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.478241586685181,
|
|
"epoch": 5.680786686838124,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002608087923872111,
|
|
"loss": 4.7828,
|
|
"mean_token_accuracy": 0.2005940333008766,
|
|
"num_tokens": 12448297.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.431466722488404,
|
|
"epoch": 5.685829551185073,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026041293867702235,
|
|
"loss": 4.835,
|
|
"mean_token_accuracy": 0.20168797373771669,
|
|
"num_tokens": 12460066.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.405737352371216,
|
|
"epoch": 5.690872415532022,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026001713030375664,
|
|
"loss": 4.8312,
|
|
"mean_token_accuracy": 0.19425902664661407,
|
|
"num_tokens": 12469806.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.340780353546142,
|
|
"epoch": 5.695915279878971,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00025962136849759564,
|
|
"loss": 4.8248,
|
|
"mean_token_accuracy": 0.1924226090312004,
|
|
"num_tokens": 12480713.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.335066604614258,
|
|
"epoch": 5.70095814422592,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00025922565448857605,
|
|
"loss": 4.7662,
|
|
"mean_token_accuracy": 0.19760394990444183,
|
|
"num_tokens": 12491635.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.405575895309449,
|
|
"epoch": 5.706001008572869,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00025882998950658623,
|
|
"loss": 4.7873,
|
|
"mean_token_accuracy": 0.19493716955184937,
|
|
"num_tokens": 12502064.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.398749971389771,
|
|
"epoch": 5.711043872919818,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025843437478136196,
|
|
"loss": 4.8079,
|
|
"mean_token_accuracy": 0.20524186491966248,
|
|
"num_tokens": 12512525.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.335463333129883,
|
|
"epoch": 5.716086737266767,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00025803881154248305,
|
|
"loss": 4.7732,
|
|
"mean_token_accuracy": 0.1931627333164215,
|
|
"num_tokens": 12523512.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.351618337631225,
|
|
"epoch": 5.7211296016137165,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002576433010193691,
|
|
"loss": 4.7278,
|
|
"mean_token_accuracy": 0.19881056994199753,
|
|
"num_tokens": 12533678.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.355673885345459,
|
|
"epoch": 5.726172465960666,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000257247844441276,
|
|
"loss": 4.8625,
|
|
"mean_token_accuracy": 0.20028068572282792,
|
|
"num_tokens": 12545519.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.392779731750489,
|
|
"epoch": 5.731215330307615,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00025685244303729187,
|
|
"loss": 4.7666,
|
|
"mean_token_accuracy": 0.20816505402326585,
|
|
"num_tokens": 12556960.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.359384536743164,
|
|
"epoch": 5.736258194654564,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002564570980363334,
|
|
"loss": 4.7361,
|
|
"mean_token_accuracy": 0.20638254135847092,
|
|
"num_tokens": 12567085.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.3717506408691404,
|
|
"epoch": 5.741301059001513,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025606181066714195,
|
|
"loss": 4.7859,
|
|
"mean_token_accuracy": 0.19550453275442123,
|
|
"num_tokens": 12576382.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.380795001983643,
|
|
"epoch": 5.746343923348462,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002556665821582798,
|
|
"loss": 4.804,
|
|
"mean_token_accuracy": 0.2034294456243515,
|
|
"num_tokens": 12587332.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.329462766647339,
|
|
"epoch": 5.751386787695411,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002552714137381264,
|
|
"loss": 4.775,
|
|
"mean_token_accuracy": 0.20877379626035691,
|
|
"num_tokens": 12597816.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.419559144973755,
|
|
"epoch": 5.75642965204236,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002548763066348741,
|
|
"loss": 4.8104,
|
|
"mean_token_accuracy": 0.2014116421341896,
|
|
"num_tokens": 12609463.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.347645282745361,
|
|
"epoch": 5.7614725163893095,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002544812620765251,
|
|
"loss": 4.8199,
|
|
"mean_token_accuracy": 0.20352404415607453,
|
|
"num_tokens": 12621551.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.40521354675293,
|
|
"epoch": 5.766515380736259,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002540862812908868,
|
|
"loss": 4.7497,
|
|
"mean_token_accuracy": 0.2069175735116005,
|
|
"num_tokens": 12632246.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.376331758499146,
|
|
"epoch": 5.771558245083208,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002536913655055687,
|
|
"loss": 4.7504,
|
|
"mean_token_accuracy": 0.20739430338144302,
|
|
"num_tokens": 12642398.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.343450546264648,
|
|
"epoch": 5.776601109430157,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002532965159479781,
|
|
"loss": 4.7801,
|
|
"mean_token_accuracy": 0.20516646802425384,
|
|
"num_tokens": 12652454.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.403167772293091,
|
|
"epoch": 5.781643973777105,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002529017338453167,
|
|
"loss": 4.8266,
|
|
"mean_token_accuracy": 0.20408095866441728,
|
|
"num_tokens": 12663757.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.44119873046875,
|
|
"epoch": 5.786686838124054,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002525070204245761,
|
|
"loss": 4.8073,
|
|
"mean_token_accuracy": 0.20163584649562835,
|
|
"num_tokens": 12674795.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.388897228240967,
|
|
"epoch": 5.791729702471003,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025211237691253507,
|
|
"loss": 4.8593,
|
|
"mean_token_accuracy": 0.19761030226945878,
|
|
"num_tokens": 12686162.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.411794424057007,
|
|
"epoch": 5.796772566817952,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025171780453575447,
|
|
"loss": 4.7833,
|
|
"mean_token_accuracy": 0.19894900172948837,
|
|
"num_tokens": 12696930.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.430048322677612,
|
|
"epoch": 5.8018154311649015,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002513233045205746,
|
|
"loss": 4.8304,
|
|
"mean_token_accuracy": 0.20307109504938126,
|
|
"num_tokens": 12708083.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.339283895492554,
|
|
"epoch": 5.806858295511851,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002509288780931102,
|
|
"loss": 4.7246,
|
|
"mean_token_accuracy": 0.20105741918087006,
|
|
"num_tokens": 12718852.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.389208507537842,
|
|
"epoch": 5.8119011598588,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002505345264792482,
|
|
"loss": 4.8418,
|
|
"mean_token_accuracy": 0.2007734403014183,
|
|
"num_tokens": 12729276.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.384714031219483,
|
|
"epoch": 5.816944024205749,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00025014025090464205,
|
|
"loss": 4.7587,
|
|
"mean_token_accuracy": 0.20847140103578568,
|
|
"num_tokens": 12739442.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.395238590240479,
|
|
"epoch": 5.821986888552698,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00024974605259470975,
|
|
"loss": 4.8388,
|
|
"mean_token_accuracy": 0.20188961178064346,
|
|
"num_tokens": 12750315.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.367918825149536,
|
|
"epoch": 5.827029752899647,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00024935193277462835,
|
|
"loss": 4.7595,
|
|
"mean_token_accuracy": 0.20250935703516007,
|
|
"num_tokens": 12761306.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.399157953262329,
|
|
"epoch": 5.832072617246596,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002489578926693317,
|
|
"loss": 4.767,
|
|
"mean_token_accuracy": 0.20976047068834305,
|
|
"num_tokens": 12772182.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.362395763397217,
|
|
"epoch": 5.837115481593545,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002485639335035053,
|
|
"loss": 4.7758,
|
|
"mean_token_accuracy": 0.2043982192873955,
|
|
"num_tokens": 12783455.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.37901291847229,
|
|
"epoch": 5.842158345940494,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00024817005650158345,
|
|
"loss": 4.7832,
|
|
"mean_token_accuracy": 0.20846178829669954,
|
|
"num_tokens": 12794756.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.333470296859741,
|
|
"epoch": 5.8472012102874436,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00024777626288774473,
|
|
"loss": 4.7694,
|
|
"mean_token_accuracy": 0.20354427844285966,
|
|
"num_tokens": 12807748.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.330673408508301,
|
|
"epoch": 5.852244074634393,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002473825538859091,
|
|
"loss": 4.7809,
|
|
"mean_token_accuracy": 0.20519086122512817,
|
|
"num_tokens": 12818642.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.362842655181884,
|
|
"epoch": 5.857286938981342,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002469889307197331,
|
|
"loss": 4.7522,
|
|
"mean_token_accuracy": 0.2019769474864006,
|
|
"num_tokens": 12830873.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.417681550979614,
|
|
"epoch": 5.86232980332829,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002465953946126065,
|
|
"loss": 4.8827,
|
|
"mean_token_accuracy": 0.19947465062141417,
|
|
"num_tokens": 12842545.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.452762079238892,
|
|
"epoch": 5.867372667675239,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00024620194678764875,
|
|
"loss": 4.8285,
|
|
"mean_token_accuracy": 0.20198632925748825,
|
|
"num_tokens": 12853398.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.382379150390625,
|
|
"epoch": 5.872415532022188,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002458085884677046,
|
|
"loss": 4.7487,
|
|
"mean_token_accuracy": 0.20454103201627732,
|
|
"num_tokens": 12865637.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.35223274230957,
|
|
"epoch": 5.877458396369137,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00024541532087534097,
|
|
"loss": 4.7892,
|
|
"mean_token_accuracy": 0.20246104747056962,
|
|
"num_tokens": 12877055.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.314401578903198,
|
|
"epoch": 5.8825012607160865,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002450221452328424,
|
|
"loss": 4.6618,
|
|
"mean_token_accuracy": 0.21492190062999725,
|
|
"num_tokens": 12887609.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.378951358795166,
|
|
"epoch": 5.887544125063036,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024462906276220814,
|
|
"loss": 4.7694,
|
|
"mean_token_accuracy": 0.20678590536117553,
|
|
"num_tokens": 12898327.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.352730464935303,
|
|
"epoch": 5.892586989409985,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002442360746851473,
|
|
"loss": 4.8597,
|
|
"mean_token_accuracy": 0.19703632742166519,
|
|
"num_tokens": 12909837.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.38573989868164,
|
|
"epoch": 5.897629853756934,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00024384318222307605,
|
|
"loss": 4.8392,
|
|
"mean_token_accuracy": 0.2016778990626335,
|
|
"num_tokens": 12921386.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.345422220230103,
|
|
"epoch": 5.902672718103883,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002434503865971131,
|
|
"loss": 4.7784,
|
|
"mean_token_accuracy": 0.21095688492059708,
|
|
"num_tokens": 12932099.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.382601737976074,
|
|
"epoch": 5.907715582450832,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002430576890280765,
|
|
"loss": 4.8036,
|
|
"mean_token_accuracy": 0.20520464032888414,
|
|
"num_tokens": 12944151.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.406586980819702,
|
|
"epoch": 5.912758446797781,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00024266509073647913,
|
|
"loss": 4.8275,
|
|
"mean_token_accuracy": 0.2021748811006546,
|
|
"num_tokens": 12956399.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.408793163299561,
|
|
"epoch": 5.91780131114473,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00024227259294252578,
|
|
"loss": 4.8454,
|
|
"mean_token_accuracy": 0.2002648815512657,
|
|
"num_tokens": 12968332.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.366849565505982,
|
|
"epoch": 5.922844175491679,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002418801968661084,
|
|
"loss": 4.7667,
|
|
"mean_token_accuracy": 0.20432819426059723,
|
|
"num_tokens": 12979647.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.388034963607788,
|
|
"epoch": 5.9278870398386285,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00024148790372680324,
|
|
"loss": 4.8368,
|
|
"mean_token_accuracy": 0.1974472373723984,
|
|
"num_tokens": 12992115.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.358849620819091,
|
|
"epoch": 5.932929904185578,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024109571474386631,
|
|
"loss": 4.6664,
|
|
"mean_token_accuracy": 0.21497285068035127,
|
|
"num_tokens": 13002611.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.371519708633423,
|
|
"epoch": 5.937972768532527,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024070363113623018,
|
|
"loss": 4.7082,
|
|
"mean_token_accuracy": 0.21265474408864976,
|
|
"num_tokens": 13013510.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.40354437828064,
|
|
"epoch": 5.943015632879476,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00024031165412249938,
|
|
"loss": 4.8571,
|
|
"mean_token_accuracy": 0.19604741930961608,
|
|
"num_tokens": 13025962.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.315206623077392,
|
|
"epoch": 5.948058497226425,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000239919784920948,
|
|
"loss": 4.7802,
|
|
"mean_token_accuracy": 0.2017505183815956,
|
|
"num_tokens": 13036751.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.318692922592163,
|
|
"epoch": 5.953101361573374,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023952802474951418,
|
|
"loss": 4.7549,
|
|
"mean_token_accuracy": 0.20791540592908858,
|
|
"num_tokens": 13048302.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.40013747215271,
|
|
"epoch": 5.958144225920323,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00023913637482579774,
|
|
"loss": 4.8031,
|
|
"mean_token_accuracy": 0.2007676437497139,
|
|
"num_tokens": 13059879.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.39221043586731,
|
|
"epoch": 5.9631870902672715,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023874483636705563,
|
|
"loss": 4.7488,
|
|
"mean_token_accuracy": 0.20411685705184937,
|
|
"num_tokens": 13071540.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.358206176757813,
|
|
"epoch": 5.968229954614221,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023835341059019838,
|
|
"loss": 4.7673,
|
|
"mean_token_accuracy": 0.2024170696735382,
|
|
"num_tokens": 13082981.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.371860218048096,
|
|
"epoch": 5.97327281896117,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023796209871178642,
|
|
"loss": 4.7424,
|
|
"mean_token_accuracy": 0.20808600932359694,
|
|
"num_tokens": 13092669.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.33985652923584,
|
|
"epoch": 5.978315683308119,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023757090194802605,
|
|
"loss": 4.807,
|
|
"mean_token_accuracy": 0.20043543875217437,
|
|
"num_tokens": 13104081.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.336535930633545,
|
|
"epoch": 5.983358547655068,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023717982151476592,
|
|
"loss": 4.8211,
|
|
"mean_token_accuracy": 0.20272261053323745,
|
|
"num_tokens": 13115068.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.359402418136597,
|
|
"epoch": 5.988401412002017,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023678885862749282,
|
|
"loss": 4.7476,
|
|
"mean_token_accuracy": 0.2031234934926033,
|
|
"num_tokens": 13125930.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.453249120712281,
|
|
"epoch": 5.993444276348966,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023639801450132863,
|
|
"loss": 4.8517,
|
|
"mean_token_accuracy": 0.2007930040359497,
|
|
"num_tokens": 13137487.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.37138524055481,
|
|
"epoch": 5.998487140695915,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00023600729035102585,
|
|
"loss": 4.6566,
|
|
"mean_token_accuracy": 0.21346824765205383,
|
|
"num_tokens": 13147542.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.359211126963298,
|
|
"epoch": 6.0030257186081695,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002356166873909642,
|
|
"loss": 4.6838,
|
|
"mean_token_accuracy": 0.2145542949438095,
|
|
"num_tokens": 13156473.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.336161470413208,
|
|
"epoch": 6.008068582955119,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023522620683514645,
|
|
"loss": 4.6062,
|
|
"mean_token_accuracy": 0.2217238202691078,
|
|
"num_tokens": 13167225.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.365195655822754,
|
|
"epoch": 6.013111447302068,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00023483584989719553,
|
|
"loss": 4.5606,
|
|
"mean_token_accuracy": 0.220480115711689,
|
|
"num_tokens": 13178034.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.463253307342529,
|
|
"epoch": 6.018154311649017,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023444561779034957,
|
|
"loss": 4.7097,
|
|
"mean_token_accuracy": 0.21369642317295073,
|
|
"num_tokens": 13189610.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.399715805053711,
|
|
"epoch": 6.023197175995966,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023405551172745904,
|
|
"loss": 4.6941,
|
|
"mean_token_accuracy": 0.21796804219484328,
|
|
"num_tokens": 13201392.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.376670169830322,
|
|
"epoch": 6.028240040342915,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023366553292098247,
|
|
"loss": 4.6309,
|
|
"mean_token_accuracy": 0.22139676958322524,
|
|
"num_tokens": 13213125.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.390435981750488,
|
|
"epoch": 6.033282904689864,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002332756825829832,
|
|
"loss": 4.6585,
|
|
"mean_token_accuracy": 0.21771981567144394,
|
|
"num_tokens": 13224287.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.368021631240845,
|
|
"epoch": 6.038325769036813,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002328859619251249,
|
|
"loss": 4.6452,
|
|
"mean_token_accuracy": 0.2108944609761238,
|
|
"num_tokens": 13234035.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.319368362426758,
|
|
"epoch": 6.043368633383762,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023249637215866847,
|
|
"loss": 4.5883,
|
|
"mean_token_accuracy": 0.22029249966144562,
|
|
"num_tokens": 13245029.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.3847626686096195,
|
|
"epoch": 6.048411497730711,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002321069144944678,
|
|
"loss": 4.7143,
|
|
"mean_token_accuracy": 0.2131657525897026,
|
|
"num_tokens": 13256844.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 6.048411497730711,
|
|
"eval_entropy": 5.204124319644374,
|
|
"eval_loss": 5.655833721160889,
|
|
"eval_mean_token_accuracy": 0.15606892571497813,
|
|
"eval_num_tokens": 13256844.0,
|
|
"eval_runtime": 2.0456,
|
|
"eval_samples_per_second": 1534.049,
|
|
"eval_steps_per_second": 192.123,
|
|
"step": 6000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9910,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 3279894447237120.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|