f4e1c86447
Model: fpadovani/ita-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed455 Source: Original Platform
6068 lines
164 KiB
JSON
6068 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.0242057488653553,
|
|
"eval_steps": 1000,
|
|
"global_step": 3000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 1639493659607040.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|