3035 lines
82 KiB
JSON
3035 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.09896417496866135,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.74233112335205,
|
|
"epoch": 0.00032988058322887113,
|
|
"grad_norm": 4.78125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8047,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 11578.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742351722717284,
|
|
"epoch": 0.0006597611664577423,
|
|
"grad_norm": 4.53125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7851,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 22232.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742332267761231,
|
|
"epoch": 0.0009896417496866134,
|
|
"grad_norm": 4.65625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.8042,
|
|
"mean_token_accuracy": 7.412898703478277e-05,
|
|
"num_tokens": 35495.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.74229564666748,
|
|
"epoch": 0.0013195223329154845,
|
|
"grad_norm": 5.03125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.7299,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 47005.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742135810852051,
|
|
"epoch": 0.0016494029161443558,
|
|
"grad_norm": 4.21875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.6387,
|
|
"mean_token_accuracy": 0.00025946117821149526,
|
|
"num_tokens": 57675.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.741827201843261,
|
|
"epoch": 0.001979283499373227,
|
|
"grad_norm": 3.9375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.5537,
|
|
"mean_token_accuracy": 0.008250601799227297,
|
|
"num_tokens": 70455.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.7412504196167,
|
|
"epoch": 0.002309164082602098,
|
|
"grad_norm": 3.046875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.4147,
|
|
"mean_token_accuracy": 0.03285279367119074,
|
|
"num_tokens": 82422.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.7402099609375,
|
|
"epoch": 0.002639044665830969,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2672,
|
|
"mean_token_accuracy": 0.036276179179549214,
|
|
"num_tokens": 92109.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.738385868072509,
|
|
"epoch": 0.0029689252490598403,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.18,
|
|
"mean_token_accuracy": 0.04042476825416088,
|
|
"num_tokens": 103440.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.736727714538574,
|
|
"epoch": 0.0032988058322887116,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.1099,
|
|
"mean_token_accuracy": 0.04008350595831871,
|
|
"num_tokens": 114742.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.73554048538208,
|
|
"epoch": 0.0036286864155175825,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.0092,
|
|
"mean_token_accuracy": 0.043487312830984594,
|
|
"num_tokens": 125626.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.733216762542725,
|
|
"epoch": 0.003958566998746454,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.976,
|
|
"mean_token_accuracy": 0.043048874288797376,
|
|
"num_tokens": 137957.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.730610179901124,
|
|
"epoch": 0.004288447581975325,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.9097,
|
|
"mean_token_accuracy": 0.036802740022540095,
|
|
"num_tokens": 148277.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.729512119293213,
|
|
"epoch": 0.004618328165204196,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8853,
|
|
"mean_token_accuracy": 0.04078914504498243,
|
|
"num_tokens": 159878.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.727227592468262,
|
|
"epoch": 0.004948208748433067,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7999,
|
|
"mean_token_accuracy": 0.04081474579870701,
|
|
"num_tokens": 172156.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.724674892425536,
|
|
"epoch": 0.005278089331661938,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.7314,
|
|
"mean_token_accuracy": 0.0359017526730895,
|
|
"num_tokens": 183913.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.720765209197998,
|
|
"epoch": 0.00560796991489081,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.6786,
|
|
"mean_token_accuracy": 0.041428479738533495,
|
|
"num_tokens": 195817.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.713240337371825,
|
|
"epoch": 0.005937850498119681,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.6194,
|
|
"mean_token_accuracy": 0.0394784739241004,
|
|
"num_tokens": 207962.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.704511070251465,
|
|
"epoch": 0.0062677310813485515,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.5465,
|
|
"mean_token_accuracy": 0.03562910072505474,
|
|
"num_tokens": 219781.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.693966960906982,
|
|
"epoch": 0.006597611664577423,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.4197,
|
|
"mean_token_accuracy": 0.037568676844239236,
|
|
"num_tokens": 231163.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.679013442993163,
|
|
"epoch": 0.006927492247806294,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.326,
|
|
"mean_token_accuracy": 0.03616092000156641,
|
|
"num_tokens": 242141.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.657819652557373,
|
|
"epoch": 0.007257372831035165,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.2134,
|
|
"mean_token_accuracy": 0.03960070498287678,
|
|
"num_tokens": 253879.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.626339530944824,
|
|
"epoch": 0.007587253414264037,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0565,
|
|
"mean_token_accuracy": 0.044448528438806534,
|
|
"num_tokens": 264486.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.585787487030029,
|
|
"epoch": 0.007917133997492908,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9692,
|
|
"mean_token_accuracy": 0.037905681505799295,
|
|
"num_tokens": 275750.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.545704460144043,
|
|
"epoch": 0.008247014580721778,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.894,
|
|
"mean_token_accuracy": 0.042156589217484,
|
|
"num_tokens": 287746.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.492252826690674,
|
|
"epoch": 0.00857689516395065,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.7427,
|
|
"mean_token_accuracy": 0.044464115798473355,
|
|
"num_tokens": 300383.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.41234998703003,
|
|
"epoch": 0.008906775747179522,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.6308,
|
|
"mean_token_accuracy": 0.05071255713701248,
|
|
"num_tokens": 313091.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.307254981994628,
|
|
"epoch": 0.009236656330408393,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.4857,
|
|
"mean_token_accuracy": 0.04862055517733097,
|
|
"num_tokens": 323251.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.19880542755127,
|
|
"epoch": 0.009566536913637264,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.3857,
|
|
"mean_token_accuracy": 0.04940937012434006,
|
|
"num_tokens": 333707.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.061772537231445,
|
|
"epoch": 0.009896417496866134,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.3134,
|
|
"mean_token_accuracy": 0.04397304020822048,
|
|
"num_tokens": 345426.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.90941858291626,
|
|
"epoch": 0.010226298080095005,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.2127,
|
|
"mean_token_accuracy": 0.04301855657249689,
|
|
"num_tokens": 358876.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.72692346572876,
|
|
"epoch": 0.010556178663323876,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.0847,
|
|
"mean_token_accuracy": 0.04769652187824249,
|
|
"num_tokens": 370955.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.496499633789062,
|
|
"epoch": 0.010886059246552749,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.9552,
|
|
"mean_token_accuracy": 0.04982128106057644,
|
|
"num_tokens": 383667.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.298235988616943,
|
|
"epoch": 0.01121593982978162,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.8896,
|
|
"mean_token_accuracy": 0.047108123824000356,
|
|
"num_tokens": 394347.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.061514949798584,
|
|
"epoch": 0.01154582041301049,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.7922,
|
|
"mean_token_accuracy": 0.05443224273622036,
|
|
"num_tokens": 406589.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.849029541015625,
|
|
"epoch": 0.011875700996239361,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7205,
|
|
"mean_token_accuracy": 0.06005025506019592,
|
|
"num_tokens": 417462.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.702647686004639,
|
|
"epoch": 0.012205581579468232,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7819,
|
|
"mean_token_accuracy": 0.05638592652976513,
|
|
"num_tokens": 428076.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.6129864692688,
|
|
"epoch": 0.012535462162697103,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.6524,
|
|
"mean_token_accuracy": 0.059817750751972196,
|
|
"num_tokens": 438714.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.444343852996827,
|
|
"epoch": 0.012865342745925976,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.6408,
|
|
"mean_token_accuracy": 0.06158410720527172,
|
|
"num_tokens": 450373.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.418584156036378,
|
|
"epoch": 0.013195223329154846,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.5929,
|
|
"mean_token_accuracy": 0.057680726051330566,
|
|
"num_tokens": 461201.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.309362125396728,
|
|
"epoch": 0.013525103912383717,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5542,
|
|
"mean_token_accuracy": 0.06454258635640145,
|
|
"num_tokens": 471519.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.26671142578125,
|
|
"epoch": 0.013854984495612588,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.5927,
|
|
"mean_token_accuracy": 0.05961795505136251,
|
|
"num_tokens": 483245.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.234414291381835,
|
|
"epoch": 0.014184865078841459,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.4793,
|
|
"mean_token_accuracy": 0.06982782371342182,
|
|
"num_tokens": 494398.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.201716423034668,
|
|
"epoch": 0.01451474566207033,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.4962,
|
|
"mean_token_accuracy": 0.06742062233388424,
|
|
"num_tokens": 505550.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.13823642730713,
|
|
"epoch": 0.014844626245299202,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.4418,
|
|
"mean_token_accuracy": 0.06280249543488026,
|
|
"num_tokens": 515824.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.11952476501465,
|
|
"epoch": 0.015174506828528073,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.4716,
|
|
"mean_token_accuracy": 0.06296740882098675,
|
|
"num_tokens": 527272.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.108990573883057,
|
|
"epoch": 0.015504387411756944,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.4469,
|
|
"mean_token_accuracy": 0.06231762431561947,
|
|
"num_tokens": 537970.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.067597770690918,
|
|
"epoch": 0.015834267994985815,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.3392,
|
|
"mean_token_accuracy": 0.06825867146253586,
|
|
"num_tokens": 549026.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.033491706848144,
|
|
"epoch": 0.016164148578214688,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.4567,
|
|
"mean_token_accuracy": 0.06924333348870278,
|
|
"num_tokens": 561001.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.013448429107665,
|
|
"epoch": 0.016494029161443557,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.3986,
|
|
"mean_token_accuracy": 0.07052349597215653,
|
|
"num_tokens": 571910.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.004467582702636,
|
|
"epoch": 0.01682390974467243,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.3889,
|
|
"mean_token_accuracy": 0.0650747399777174,
|
|
"num_tokens": 583073.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.977169990539551,
|
|
"epoch": 0.0171537903279013,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3194,
|
|
"mean_token_accuracy": 0.07421667948365211,
|
|
"num_tokens": 593909.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.92542724609375,
|
|
"epoch": 0.01748367091113017,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.347,
|
|
"mean_token_accuracy": 0.07075322493910789,
|
|
"num_tokens": 605235.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.959033679962158,
|
|
"epoch": 0.017813551494359044,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.2928,
|
|
"mean_token_accuracy": 0.07352816425263882,
|
|
"num_tokens": 616491.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.854368066787719,
|
|
"epoch": 0.018143432077587913,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.301,
|
|
"mean_token_accuracy": 0.07566386498510838,
|
|
"num_tokens": 627576.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.89704909324646,
|
|
"epoch": 0.018473312660816785,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2279,
|
|
"mean_token_accuracy": 0.07298410087823867,
|
|
"num_tokens": 638833.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.945636034011841,
|
|
"epoch": 0.018803193244045655,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.3253,
|
|
"mean_token_accuracy": 0.06714979782700539,
|
|
"num_tokens": 648954.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.888553476333618,
|
|
"epoch": 0.019133073827274527,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.2291,
|
|
"mean_token_accuracy": 0.07276964522898197,
|
|
"num_tokens": 659653.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.827844858169556,
|
|
"epoch": 0.019462954410503396,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.3204,
|
|
"mean_token_accuracy": 0.0698035504668951,
|
|
"num_tokens": 671282.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.846695137023926,
|
|
"epoch": 0.01979283499373227,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2734,
|
|
"mean_token_accuracy": 0.06930247582495212,
|
|
"num_tokens": 683096.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.817320489883423,
|
|
"epoch": 0.02012271557696114,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2386,
|
|
"mean_token_accuracy": 0.07485006004571915,
|
|
"num_tokens": 694278.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.90717658996582,
|
|
"epoch": 0.02045259616019001,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2358,
|
|
"mean_token_accuracy": 0.0718875303864479,
|
|
"num_tokens": 705878.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.742885494232178,
|
|
"epoch": 0.020782476743418883,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.1725,
|
|
"mean_token_accuracy": 0.0757001593708992,
|
|
"num_tokens": 717459.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.795225667953491,
|
|
"epoch": 0.021112357326647752,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.1761,
|
|
"mean_token_accuracy": 0.07542249858379364,
|
|
"num_tokens": 726708.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.819357681274414,
|
|
"epoch": 0.021442237909876625,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1691,
|
|
"mean_token_accuracy": 0.07912719771265983,
|
|
"num_tokens": 736799.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.716309642791748,
|
|
"epoch": 0.021772118493105497,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1528,
|
|
"mean_token_accuracy": 0.07632440999150276,
|
|
"num_tokens": 747820.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.650016450881958,
|
|
"epoch": 0.022101999076334367,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1569,
|
|
"mean_token_accuracy": 0.07818866521120071,
|
|
"num_tokens": 757938.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.751991653442383,
|
|
"epoch": 0.02243187965956324,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.1513,
|
|
"mean_token_accuracy": 0.07743561789393424,
|
|
"num_tokens": 768900.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.693883180618286,
|
|
"epoch": 0.02276176024279211,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.218,
|
|
"mean_token_accuracy": 0.07597828209400177,
|
|
"num_tokens": 780839.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.725241041183471,
|
|
"epoch": 0.02309164082602098,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1578,
|
|
"mean_token_accuracy": 0.07896500080823898,
|
|
"num_tokens": 793144.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.7039953708648685,
|
|
"epoch": 0.02342152140924985,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.1385,
|
|
"mean_token_accuracy": 0.0761062603443861,
|
|
"num_tokens": 805293.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.650927495956421,
|
|
"epoch": 0.023751401992478723,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.0847,
|
|
"mean_token_accuracy": 0.07881326377391815,
|
|
"num_tokens": 817313.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.617914915084839,
|
|
"epoch": 0.024081282575707595,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.039,
|
|
"mean_token_accuracy": 0.07709232643246651,
|
|
"num_tokens": 827823.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.612017059326172,
|
|
"epoch": 0.024411163158936464,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.0581,
|
|
"mean_token_accuracy": 0.07990295961499214,
|
|
"num_tokens": 839221.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.634074735641479,
|
|
"epoch": 0.024741043742165337,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0299,
|
|
"mean_token_accuracy": 0.0763829343020916,
|
|
"num_tokens": 850616.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.552822065353394,
|
|
"epoch": 0.025070924325394206,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0531,
|
|
"mean_token_accuracy": 0.07905212491750717,
|
|
"num_tokens": 862467.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.6426232814788815,
|
|
"epoch": 0.02540080490862308,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.0412,
|
|
"mean_token_accuracy": 0.07432318702340127,
|
|
"num_tokens": 872130.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.561140012741089,
|
|
"epoch": 0.02573068549185195,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.0034,
|
|
"mean_token_accuracy": 0.08546995297074318,
|
|
"num_tokens": 883598.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.582135248184204,
|
|
"epoch": 0.02606056607508082,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0003,
|
|
"mean_token_accuracy": 0.07829435840249062,
|
|
"num_tokens": 896478.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.555583810806274,
|
|
"epoch": 0.026390446658309693,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.0712,
|
|
"mean_token_accuracy": 0.07842531427741051,
|
|
"num_tokens": 909351.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.550145435333252,
|
|
"epoch": 0.026720327241538562,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0363,
|
|
"mean_token_accuracy": 0.08111794963479042,
|
|
"num_tokens": 920170.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.451640224456787,
|
|
"epoch": 0.027050207824767435,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.978,
|
|
"mean_token_accuracy": 0.0850048691034317,
|
|
"num_tokens": 929498.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.552112293243408,
|
|
"epoch": 0.027380088407996304,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.9678,
|
|
"mean_token_accuracy": 0.07753312811255456,
|
|
"num_tokens": 939975.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.444381189346314,
|
|
"epoch": 0.027709968991225176,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9504,
|
|
"mean_token_accuracy": 0.0817045159637928,
|
|
"num_tokens": 950593.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.561855220794678,
|
|
"epoch": 0.02803984957445405,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.9561,
|
|
"mean_token_accuracy": 0.08070995509624482,
|
|
"num_tokens": 962281.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.47752628326416,
|
|
"epoch": 0.028369730157682918,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8784,
|
|
"mean_token_accuracy": 0.08914879560470582,
|
|
"num_tokens": 972717.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.474664211273193,
|
|
"epoch": 0.02869961074091179,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8833,
|
|
"mean_token_accuracy": 0.08555072918534279,
|
|
"num_tokens": 983669.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.426136112213134,
|
|
"epoch": 0.02902949132414066,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.8871,
|
|
"mean_token_accuracy": 0.08220960162580013,
|
|
"num_tokens": 994528.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.392468643188477,
|
|
"epoch": 0.029359371907369532,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8482,
|
|
"mean_token_accuracy": 0.0958639606833458,
|
|
"num_tokens": 1005445.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.446725940704345,
|
|
"epoch": 0.029689252490598405,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.0157,
|
|
"mean_token_accuracy": 0.0827529065310955,
|
|
"num_tokens": 1018747.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.441160249710083,
|
|
"epoch": 0.030019133073827274,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.8981,
|
|
"mean_token_accuracy": 0.08528381213545799,
|
|
"num_tokens": 1028956.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.43569803237915,
|
|
"epoch": 0.030349013657056147,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.0367,
|
|
"mean_token_accuracy": 0.07986071296036243,
|
|
"num_tokens": 1041227.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.422861957550049,
|
|
"epoch": 0.030678894240285016,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.9143,
|
|
"mean_token_accuracy": 0.07918849810957909,
|
|
"num_tokens": 1053002.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.4423298835754395,
|
|
"epoch": 0.03100877482351389,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.9095,
|
|
"mean_token_accuracy": 0.08558695167303085,
|
|
"num_tokens": 1063693.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.416697883605957,
|
|
"epoch": 0.03133865540674276,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.88,
|
|
"mean_token_accuracy": 0.09078652560710906,
|
|
"num_tokens": 1074245.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.32618203163147,
|
|
"epoch": 0.03166853598997163,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8789,
|
|
"mean_token_accuracy": 0.08656272403895855,
|
|
"num_tokens": 1084663.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.368553400039673,
|
|
"epoch": 0.0319984165732005,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8227,
|
|
"mean_token_accuracy": 0.0852760024368763,
|
|
"num_tokens": 1094777.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.348766326904297,
|
|
"epoch": 0.032328297156429375,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.8406,
|
|
"mean_token_accuracy": 0.08816310912370681,
|
|
"num_tokens": 1106712.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.286499500274658,
|
|
"epoch": 0.03265817773965824,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8402,
|
|
"mean_token_accuracy": 0.09093867242336273,
|
|
"num_tokens": 1118488.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.348550319671631,
|
|
"epoch": 0.032988058322887114,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8115,
|
|
"mean_token_accuracy": 0.08822314143180847,
|
|
"num_tokens": 1128803.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.325877618789673,
|
|
"epoch": 0.033317938906115986,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7801,
|
|
"mean_token_accuracy": 0.09269751161336899,
|
|
"num_tokens": 1140081.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.249666023254394,
|
|
"epoch": 0.03364781948934486,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.8423,
|
|
"mean_token_accuracy": 0.08509944006800652,
|
|
"num_tokens": 1151740.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.387292528152466,
|
|
"epoch": 0.03397770007257373,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.9424,
|
|
"mean_token_accuracy": 0.08753601871430874,
|
|
"num_tokens": 1163554.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.363021564483643,
|
|
"epoch": 0.0343075806558026,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.8149,
|
|
"mean_token_accuracy": 0.09259535372257233,
|
|
"num_tokens": 1174634.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.142534828186035,
|
|
"epoch": 0.03463746123903147,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.7739,
|
|
"mean_token_accuracy": 0.08925415575504303,
|
|
"num_tokens": 1186640.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.303580617904663,
|
|
"epoch": 0.03496734182226034,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7195,
|
|
"mean_token_accuracy": 0.08907380253076554,
|
|
"num_tokens": 1197168.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.284749698638916,
|
|
"epoch": 0.035297222405489215,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.7724,
|
|
"mean_token_accuracy": 0.09144287258386612,
|
|
"num_tokens": 1208560.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.210066461563111,
|
|
"epoch": 0.03562710298871809,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.753,
|
|
"mean_token_accuracy": 0.0977034792304039,
|
|
"num_tokens": 1220492.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.2108320713043215,
|
|
"epoch": 0.03595698357194695,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.7249,
|
|
"mean_token_accuracy": 0.09175074473023415,
|
|
"num_tokens": 1232753.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.252335834503174,
|
|
"epoch": 0.036286864155175826,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.845,
|
|
"mean_token_accuracy": 0.08571628406643868,
|
|
"num_tokens": 1244972.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.234958124160767,
|
|
"epoch": 0.0366167447384047,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.7357,
|
|
"mean_token_accuracy": 0.09484243839979171,
|
|
"num_tokens": 1256101.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.181823444366455,
|
|
"epoch": 0.03694662532163357,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.7857,
|
|
"mean_token_accuracy": 0.0866756096482277,
|
|
"num_tokens": 1268932.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.200888061523438,
|
|
"epoch": 0.037276505904862436,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.695,
|
|
"mean_token_accuracy": 0.08639899343252182,
|
|
"num_tokens": 1278799.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.22176570892334,
|
|
"epoch": 0.03760638648809131,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7409,
|
|
"mean_token_accuracy": 0.09684469699859619,
|
|
"num_tokens": 1288703.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.088978052139282,
|
|
"epoch": 0.03793626707132018,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6848,
|
|
"mean_token_accuracy": 0.09133393913507462,
|
|
"num_tokens": 1300718.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.159674024581909,
|
|
"epoch": 0.038266147654549054,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6556,
|
|
"mean_token_accuracy": 0.09586420208215714,
|
|
"num_tokens": 1311735.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.153814125061035,
|
|
"epoch": 0.03859602823777793,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6843,
|
|
"mean_token_accuracy": 0.09448647424578667,
|
|
"num_tokens": 1322027.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.167168998718262,
|
|
"epoch": 0.03892590882100679,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6957,
|
|
"mean_token_accuracy": 0.09144030138850212,
|
|
"num_tokens": 1331326.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.153041696548462,
|
|
"epoch": 0.039255789404235665,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7088,
|
|
"mean_token_accuracy": 0.09533656612038613,
|
|
"num_tokens": 1342064.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.149514245986938,
|
|
"epoch": 0.03958566998746454,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.7807,
|
|
"mean_token_accuracy": 0.09185100421309471,
|
|
"num_tokens": 1354257.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.18934383392334,
|
|
"epoch": 0.03991555057069341,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6781,
|
|
"mean_token_accuracy": 0.09455206841230393,
|
|
"num_tokens": 1365757.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.1387965202331545,
|
|
"epoch": 0.04024543115392228,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.7468,
|
|
"mean_token_accuracy": 0.09017066359519958,
|
|
"num_tokens": 1376655.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.205080175399781,
|
|
"epoch": 0.04057531173715115,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7564,
|
|
"mean_token_accuracy": 0.09203101694583893,
|
|
"num_tokens": 1390293.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.02491946220398,
|
|
"epoch": 0.04090519232038002,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6195,
|
|
"mean_token_accuracy": 0.0970487006008625,
|
|
"num_tokens": 1402595.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.1582615852355955,
|
|
"epoch": 0.041235072903608894,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6489,
|
|
"mean_token_accuracy": 0.09517830908298493,
|
|
"num_tokens": 1413574.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.061727285385132,
|
|
"epoch": 0.041564953486837766,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.6651,
|
|
"mean_token_accuracy": 0.08918783366680146,
|
|
"num_tokens": 1425520.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.056548929214477,
|
|
"epoch": 0.04189483407006664,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6162,
|
|
"mean_token_accuracy": 0.09108572974801063,
|
|
"num_tokens": 1436285.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.118799352645874,
|
|
"epoch": 0.042224714653295504,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.6136,
|
|
"mean_token_accuracy": 0.10142198577523232,
|
|
"num_tokens": 1447194.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.015072345733643,
|
|
"epoch": 0.04255459523652438,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6271,
|
|
"mean_token_accuracy": 0.09909635856747627,
|
|
"num_tokens": 1459587.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.023969030380249,
|
|
"epoch": 0.04288447581975325,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7239,
|
|
"mean_token_accuracy": 0.08922117725014686,
|
|
"num_tokens": 1472677.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.125645780563355,
|
|
"epoch": 0.04321435640298212,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6676,
|
|
"mean_token_accuracy": 0.09377843365073205,
|
|
"num_tokens": 1484307.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.98760347366333,
|
|
"epoch": 0.043544236986210995,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5969,
|
|
"mean_token_accuracy": 0.09738756716251373,
|
|
"num_tokens": 1496198.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.049053001403808,
|
|
"epoch": 0.04387411756943986,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5415,
|
|
"mean_token_accuracy": 0.09548762142658233,
|
|
"num_tokens": 1507439.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.902269315719605,
|
|
"epoch": 0.04420399815266873,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.5666,
|
|
"mean_token_accuracy": 0.09989230334758759,
|
|
"num_tokens": 1518176.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.977385139465332,
|
|
"epoch": 0.044533878735897606,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5591,
|
|
"mean_token_accuracy": 0.09720836281776428,
|
|
"num_tokens": 1529095.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.00214319229126,
|
|
"epoch": 0.04486375931912648,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5559,
|
|
"mean_token_accuracy": 0.09865213707089424,
|
|
"num_tokens": 1539431.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.024150228500366,
|
|
"epoch": 0.045193639902355344,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6248,
|
|
"mean_token_accuracy": 0.09451132789254188,
|
|
"num_tokens": 1550878.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.005107450485229,
|
|
"epoch": 0.04552352048558422,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.5717,
|
|
"mean_token_accuracy": 0.10220879390835762,
|
|
"num_tokens": 1561416.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.985655498504639,
|
|
"epoch": 0.04585340106881309,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5954,
|
|
"mean_token_accuracy": 0.10094324424862862,
|
|
"num_tokens": 1574470.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.93547043800354,
|
|
"epoch": 0.04618328165204196,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5503,
|
|
"mean_token_accuracy": 0.09845736622810364,
|
|
"num_tokens": 1586021.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.906772899627685,
|
|
"epoch": 0.046513162235270834,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.563,
|
|
"mean_token_accuracy": 0.09999166876077652,
|
|
"num_tokens": 1597500.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.013272094726562,
|
|
"epoch": 0.0468430428184997,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5971,
|
|
"mean_token_accuracy": 0.0965078391134739,
|
|
"num_tokens": 1609797.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.930358076095581,
|
|
"epoch": 0.04717292340172857,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.551,
|
|
"mean_token_accuracy": 0.1009341761469841,
|
|
"num_tokens": 1620613.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.000537967681884,
|
|
"epoch": 0.047502803984957445,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.553,
|
|
"mean_token_accuracy": 0.10127959176898002,
|
|
"num_tokens": 1631542.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.929404783248901,
|
|
"epoch": 0.04783268456818632,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4644,
|
|
"mean_token_accuracy": 0.10010106787085533,
|
|
"num_tokens": 1641783.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.827300930023194,
|
|
"epoch": 0.04816256515141519,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6008,
|
|
"mean_token_accuracy": 0.09775333851575851,
|
|
"num_tokens": 1652938.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.95234055519104,
|
|
"epoch": 0.048492445734644056,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.4386,
|
|
"mean_token_accuracy": 0.10487373694777488,
|
|
"num_tokens": 1663214.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.972109842300415,
|
|
"epoch": 0.04882232631787293,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.6092,
|
|
"mean_token_accuracy": 0.09515793696045875,
|
|
"num_tokens": 1674845.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.853538131713867,
|
|
"epoch": 0.0491522069011018,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4916,
|
|
"mean_token_accuracy": 0.10413750782608985,
|
|
"num_tokens": 1685222.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.949459981918335,
|
|
"epoch": 0.049482087484330674,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5191,
|
|
"mean_token_accuracy": 0.09741224348545074,
|
|
"num_tokens": 1696520.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.861531782150268,
|
|
"epoch": 0.049811968067559546,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5043,
|
|
"mean_token_accuracy": 0.10009976252913474,
|
|
"num_tokens": 1706888.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.85313572883606,
|
|
"epoch": 0.05014184865078841,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.5053,
|
|
"mean_token_accuracy": 0.10150282755494118,
|
|
"num_tokens": 1718023.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.816460084915161,
|
|
"epoch": 0.050471729234017285,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.447,
|
|
"mean_token_accuracy": 0.10082821324467658,
|
|
"num_tokens": 1729291.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.867931795120239,
|
|
"epoch": 0.05080160981724616,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5009,
|
|
"mean_token_accuracy": 0.10062082186341285,
|
|
"num_tokens": 1740801.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.793792390823365,
|
|
"epoch": 0.05113149040047503,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4479,
|
|
"mean_token_accuracy": 0.10480978637933731,
|
|
"num_tokens": 1750933.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.777090930938721,
|
|
"epoch": 0.0514613709837039,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5029,
|
|
"mean_token_accuracy": 0.09836127161979676,
|
|
"num_tokens": 1761820.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.843807744979858,
|
|
"epoch": 0.05179125156693277,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3851,
|
|
"mean_token_accuracy": 0.10793521851301194,
|
|
"num_tokens": 1771628.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.779563665390015,
|
|
"epoch": 0.05212113215016164,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.4118,
|
|
"mean_token_accuracy": 0.10292888209223747,
|
|
"num_tokens": 1783360.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.869561862945557,
|
|
"epoch": 0.05245101273339051,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4721,
|
|
"mean_token_accuracy": 0.10269110351800918,
|
|
"num_tokens": 1794831.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.757121276855469,
|
|
"epoch": 0.052780893316619386,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3996,
|
|
"mean_token_accuracy": 0.10898642167448998,
|
|
"num_tokens": 1805056.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.742442893981933,
|
|
"epoch": 0.05311077389984825,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3521,
|
|
"mean_token_accuracy": 0.10676872208714486,
|
|
"num_tokens": 1815233.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.787021493911743,
|
|
"epoch": 0.053440654483077124,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.611,
|
|
"mean_token_accuracy": 0.09833935126662255,
|
|
"num_tokens": 1826376.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.929257822036743,
|
|
"epoch": 0.053770535066306,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5177,
|
|
"mean_token_accuracy": 0.1020083598792553,
|
|
"num_tokens": 1836326.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.732786750793457,
|
|
"epoch": 0.05410041564953487,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4724,
|
|
"mean_token_accuracy": 0.10276546701788902,
|
|
"num_tokens": 1847796.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.75844931602478,
|
|
"epoch": 0.05443029623276374,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3888,
|
|
"mean_token_accuracy": 0.10622160062193871,
|
|
"num_tokens": 1858470.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.7840691089630125,
|
|
"epoch": 0.05476017681599261,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4729,
|
|
"mean_token_accuracy": 0.10429331809282302,
|
|
"num_tokens": 1870368.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.742818212509155,
|
|
"epoch": 0.05509005739922148,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4328,
|
|
"mean_token_accuracy": 0.10953493639826775,
|
|
"num_tokens": 1881841.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.755731439590454,
|
|
"epoch": 0.05541993798245035,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3925,
|
|
"mean_token_accuracy": 0.1099935881793499,
|
|
"num_tokens": 1892589.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.69039740562439,
|
|
"epoch": 0.055749818565679225,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.3673,
|
|
"mean_token_accuracy": 0.11012204438447952,
|
|
"num_tokens": 1903730.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.7573614597320555,
|
|
"epoch": 0.0560796991489081,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4302,
|
|
"mean_token_accuracy": 0.1086188443005085,
|
|
"num_tokens": 1914872.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.760918712615966,
|
|
"epoch": 0.056409579732136964,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.458,
|
|
"mean_token_accuracy": 0.10401058197021484,
|
|
"num_tokens": 1926127.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.707938003540039,
|
|
"epoch": 0.056739460315365836,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4261,
|
|
"mean_token_accuracy": 0.10297657176852226,
|
|
"num_tokens": 1938867.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.701300954818725,
|
|
"epoch": 0.05706934089859471,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3721,
|
|
"mean_token_accuracy": 0.11107245236635208,
|
|
"num_tokens": 1950379.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.717032527923584,
|
|
"epoch": 0.05739922148182358,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3958,
|
|
"mean_token_accuracy": 0.10820549428462982,
|
|
"num_tokens": 1961628.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.7236415386199955,
|
|
"epoch": 0.057729102065052454,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4058,
|
|
"mean_token_accuracy": 0.10361326783895493,
|
|
"num_tokens": 1971580.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.726569175720215,
|
|
"epoch": 0.05805898264828132,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4085,
|
|
"mean_token_accuracy": 0.10312368124723434,
|
|
"num_tokens": 1981789.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.681711912155151,
|
|
"epoch": 0.05838886323151019,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.3067,
|
|
"mean_token_accuracy": 0.1060155376791954,
|
|
"num_tokens": 1992667.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.598890781402588,
|
|
"epoch": 0.058718743814739065,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3319,
|
|
"mean_token_accuracy": 0.11056532859802246,
|
|
"num_tokens": 2003864.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.65644645690918,
|
|
"epoch": 0.05904862439796794,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4309,
|
|
"mean_token_accuracy": 0.10145750418305396,
|
|
"num_tokens": 2017399.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.609835195541382,
|
|
"epoch": 0.05937850498119681,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3656,
|
|
"mean_token_accuracy": 0.10574300587177277,
|
|
"num_tokens": 2027924.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.7295098304748535,
|
|
"epoch": 0.059708385564425676,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2793,
|
|
"mean_token_accuracy": 0.11127018183469772,
|
|
"num_tokens": 2040174.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.5820475101470945,
|
|
"epoch": 0.06003826614765455,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3755,
|
|
"mean_token_accuracy": 0.10315576866269112,
|
|
"num_tokens": 2052564.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.776964807510376,
|
|
"epoch": 0.06036814673088342,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4174,
|
|
"mean_token_accuracy": 0.10292134135961532,
|
|
"num_tokens": 2063211.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.646928071975708,
|
|
"epoch": 0.06069802731411229,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.327,
|
|
"mean_token_accuracy": 0.10777697786688804,
|
|
"num_tokens": 2075434.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.594566106796265,
|
|
"epoch": 0.061027907897341166,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3556,
|
|
"mean_token_accuracy": 0.1038420632481575,
|
|
"num_tokens": 2087969.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.745871925354004,
|
|
"epoch": 0.06135778848057003,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.393,
|
|
"mean_token_accuracy": 0.10197513997554779,
|
|
"num_tokens": 2098469.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.559610033035279,
|
|
"epoch": 0.061687669063798904,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2796,
|
|
"mean_token_accuracy": 0.10989089384675026,
|
|
"num_tokens": 2108739.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.695238208770752,
|
|
"epoch": 0.06201754964702778,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3878,
|
|
"mean_token_accuracy": 0.10769468769431115,
|
|
"num_tokens": 2120234.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.584358024597168,
|
|
"epoch": 0.06234743023025665,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.2652,
|
|
"mean_token_accuracy": 0.10648446381092072,
|
|
"num_tokens": 2131585.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.645196437835693,
|
|
"epoch": 0.06267731081348552,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4332,
|
|
"mean_token_accuracy": 0.10639934986829758,
|
|
"num_tokens": 2144537.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.700238752365112,
|
|
"epoch": 0.06300719139671439,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3867,
|
|
"mean_token_accuracy": 0.10446119010448456,
|
|
"num_tokens": 2157468.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.525477313995362,
|
|
"epoch": 0.06333707197994326,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.3074,
|
|
"mean_token_accuracy": 0.10612000674009323,
|
|
"num_tokens": 2169099.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.564707183837891,
|
|
"epoch": 0.06366695256317213,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2871,
|
|
"mean_token_accuracy": 0.11626865863800048,
|
|
"num_tokens": 2179970.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.564694499969482,
|
|
"epoch": 0.063996833146401,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3363,
|
|
"mean_token_accuracy": 0.10973790585994721,
|
|
"num_tokens": 2191455.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.749835729598999,
|
|
"epoch": 0.06432671372962988,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4757,
|
|
"mean_token_accuracy": 0.09981689378619193,
|
|
"num_tokens": 2202711.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.565635728836059,
|
|
"epoch": 0.06465659431285875,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2468,
|
|
"mean_token_accuracy": 0.11120761334896087,
|
|
"num_tokens": 2214035.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.568064403533936,
|
|
"epoch": 0.06498647489608762,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2674,
|
|
"mean_token_accuracy": 0.1099687896668911,
|
|
"num_tokens": 2225437.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.664786672592163,
|
|
"epoch": 0.06531635547931648,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3445,
|
|
"mean_token_accuracy": 0.10701494589447975,
|
|
"num_tokens": 2235693.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.5984398365020756,
|
|
"epoch": 0.06564623606254535,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2887,
|
|
"mean_token_accuracy": 0.11571999564766884,
|
|
"num_tokens": 2245850.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.527003002166748,
|
|
"epoch": 0.06597611664577423,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.346,
|
|
"mean_token_accuracy": 0.10999381244182586,
|
|
"num_tokens": 2257056.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.618610429763794,
|
|
"epoch": 0.0663059972290031,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.4444,
|
|
"mean_token_accuracy": 0.10575283989310265,
|
|
"num_tokens": 2270421.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.540064334869385,
|
|
"epoch": 0.06663587781223197,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.2778,
|
|
"mean_token_accuracy": 0.11337646692991257,
|
|
"num_tokens": 2282404.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.535987138748169,
|
|
"epoch": 0.06696575839546084,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.2074,
|
|
"mean_token_accuracy": 0.11448908373713493,
|
|
"num_tokens": 2292610.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.527937507629394,
|
|
"epoch": 0.06729563897868972,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.2561,
|
|
"mean_token_accuracy": 0.1101236142218113,
|
|
"num_tokens": 2303230.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.496319580078125,
|
|
"epoch": 0.06762551956191859,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.3138,
|
|
"mean_token_accuracy": 0.10930130183696747,
|
|
"num_tokens": 2316522.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.530350017547607,
|
|
"epoch": 0.06795540014514746,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.2444,
|
|
"mean_token_accuracy": 0.11181065291166306,
|
|
"num_tokens": 2327136.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.482801723480224,
|
|
"epoch": 0.06828528072837632,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.2218,
|
|
"mean_token_accuracy": 0.11561196967959404,
|
|
"num_tokens": 2337787.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.492512369155884,
|
|
"epoch": 0.0686151613116052,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.2251,
|
|
"mean_token_accuracy": 0.11074953898787498,
|
|
"num_tokens": 2349983.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.534840440750122,
|
|
"epoch": 0.06894504189483407,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.3348,
|
|
"mean_token_accuracy": 0.10682928562164307,
|
|
"num_tokens": 2361673.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.641943311691284,
|
|
"epoch": 0.06927492247806294,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.3282,
|
|
"mean_token_accuracy": 0.11464205458760261,
|
|
"num_tokens": 2373877.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.475301933288574,
|
|
"epoch": 0.06960480306129181,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.1952,
|
|
"mean_token_accuracy": 0.1164882704615593,
|
|
"num_tokens": 2384558.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.53571629524231,
|
|
"epoch": 0.06993468364452068,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.2885,
|
|
"mean_token_accuracy": 0.10790699049830436,
|
|
"num_tokens": 2395333.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.446133232116699,
|
|
"epoch": 0.07026456422774956,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.214,
|
|
"mean_token_accuracy": 0.11159637048840523,
|
|
"num_tokens": 2407100.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.443557214736939,
|
|
"epoch": 0.07059444481097843,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.1596,
|
|
"mean_token_accuracy": 0.11595186367630958,
|
|
"num_tokens": 2418137.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.414693832397461,
|
|
"epoch": 0.0709243253942073,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.2505,
|
|
"mean_token_accuracy": 0.11329601109027862,
|
|
"num_tokens": 2429058.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.513547801971436,
|
|
"epoch": 0.07125420597743617,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.24,
|
|
"mean_token_accuracy": 0.11459704115986824,
|
|
"num_tokens": 2440716.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.54072585105896,
|
|
"epoch": 0.07158408656066503,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.2677,
|
|
"mean_token_accuracy": 0.11517914533615112,
|
|
"num_tokens": 2451189.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.368064117431641,
|
|
"epoch": 0.0719139671438939,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.2162,
|
|
"mean_token_accuracy": 0.11259195283055305,
|
|
"num_tokens": 2462860.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.493119239807129,
|
|
"epoch": 0.07224384772712278,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.2409,
|
|
"mean_token_accuracy": 0.10539431199431419,
|
|
"num_tokens": 2475555.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.490513038635254,
|
|
"epoch": 0.07257372831035165,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.2706,
|
|
"mean_token_accuracy": 0.10943889692425728,
|
|
"num_tokens": 2485599.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.487355518341064,
|
|
"epoch": 0.07290360889358052,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.1315,
|
|
"mean_token_accuracy": 0.11551824659109115,
|
|
"num_tokens": 2495857.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.396845865249634,
|
|
"epoch": 0.0732334894768094,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.2302,
|
|
"mean_token_accuracy": 0.11530223116278648,
|
|
"num_tokens": 2506330.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.506023502349853,
|
|
"epoch": 0.07356337006003827,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.2103,
|
|
"mean_token_accuracy": 0.10762735083699226,
|
|
"num_tokens": 2518181.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.457603454589844,
|
|
"epoch": 0.07389325064326714,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.1525,
|
|
"mean_token_accuracy": 0.11814433187246323,
|
|
"num_tokens": 2528346.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.366501569747925,
|
|
"epoch": 0.07422313122649601,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.2237,
|
|
"mean_token_accuracy": 0.10850833207368851,
|
|
"num_tokens": 2539922.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.438397979736328,
|
|
"epoch": 0.07455301180972487,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.1788,
|
|
"mean_token_accuracy": 0.11676511541008949,
|
|
"num_tokens": 2551145.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.468164253234863,
|
|
"epoch": 0.07488289239295375,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.2688,
|
|
"mean_token_accuracy": 0.11262761503458023,
|
|
"num_tokens": 2562178.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.552046442031861,
|
|
"epoch": 0.07521277297618262,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.3198,
|
|
"mean_token_accuracy": 0.11036192998290062,
|
|
"num_tokens": 2574002.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.409352016448975,
|
|
"epoch": 0.07554265355941149,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.2251,
|
|
"mean_token_accuracy": 0.11538875252008438,
|
|
"num_tokens": 2584049.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.411483907699585,
|
|
"epoch": 0.07587253414264036,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.2128,
|
|
"mean_token_accuracy": 0.11105952560901641,
|
|
"num_tokens": 2597415.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.43424391746521,
|
|
"epoch": 0.07620241472586924,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.1614,
|
|
"mean_token_accuracy": 0.11391298174858093,
|
|
"num_tokens": 2609381.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.408471488952637,
|
|
"epoch": 0.07653229530909811,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.2488,
|
|
"mean_token_accuracy": 0.11827918812632561,
|
|
"num_tokens": 2620525.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.377165603637695,
|
|
"epoch": 0.07686217589232698,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.1652,
|
|
"mean_token_accuracy": 0.11539266556501389,
|
|
"num_tokens": 2631978.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.42329421043396,
|
|
"epoch": 0.07719205647555585,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.1672,
|
|
"mean_token_accuracy": 0.11286110579967498,
|
|
"num_tokens": 2643083.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.415650844573975,
|
|
"epoch": 0.07752193705878473,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.1571,
|
|
"mean_token_accuracy": 0.11403257474303245,
|
|
"num_tokens": 2654175.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.43393087387085,
|
|
"epoch": 0.07785181764201358,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.1244,
|
|
"mean_token_accuracy": 0.11616934239864349,
|
|
"num_tokens": 2665513.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.3165178298950195,
|
|
"epoch": 0.07818169822524246,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.1381,
|
|
"mean_token_accuracy": 0.1166747473180294,
|
|
"num_tokens": 2675842.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.44808874130249,
|
|
"epoch": 0.07851157880847133,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.289,
|
|
"mean_token_accuracy": 0.11076370999217033,
|
|
"num_tokens": 2687289.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.45378532409668,
|
|
"epoch": 0.0788414593917002,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.1818,
|
|
"mean_token_accuracy": 0.11144861951470375,
|
|
"num_tokens": 2698409.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.291270208358765,
|
|
"epoch": 0.07917133997492908,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.1305,
|
|
"mean_token_accuracy": 0.11825140044093133,
|
|
"num_tokens": 2708695.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.438046884536743,
|
|
"epoch": 0.07950122055815795,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.1106,
|
|
"mean_token_accuracy": 0.1234841413795948,
|
|
"num_tokens": 2719631.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.41588134765625,
|
|
"epoch": 0.07983110114138682,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.2096,
|
|
"mean_token_accuracy": 0.11646713688969612,
|
|
"num_tokens": 2730070.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.370893335342407,
|
|
"epoch": 0.08016098172461569,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.2039,
|
|
"mean_token_accuracy": 0.11152140498161316,
|
|
"num_tokens": 2741450.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.430210828781128,
|
|
"epoch": 0.08049086230784457,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.1301,
|
|
"mean_token_accuracy": 0.12166139110922813,
|
|
"num_tokens": 2752108.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.321960210800171,
|
|
"epoch": 0.08082074289107344,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.1902,
|
|
"mean_token_accuracy": 0.11438200101256371,
|
|
"num_tokens": 2763041.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.4781797409057615,
|
|
"epoch": 0.0811506234743023,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.142,
|
|
"mean_token_accuracy": 0.11679482907056808,
|
|
"num_tokens": 2773140.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.29995551109314,
|
|
"epoch": 0.08148050405753117,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.1152,
|
|
"mean_token_accuracy": 0.1186187893152237,
|
|
"num_tokens": 2783811.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.400778293609619,
|
|
"epoch": 0.08181038464076004,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.0916,
|
|
"mean_token_accuracy": 0.11716892346739768,
|
|
"num_tokens": 2793957.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.376893854141235,
|
|
"epoch": 0.08214026522398891,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.1916,
|
|
"mean_token_accuracy": 0.11219452396035194,
|
|
"num_tokens": 2804750.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.390607213973999,
|
|
"epoch": 0.08247014580721779,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.2091,
|
|
"mean_token_accuracy": 0.11758575886487961,
|
|
"num_tokens": 2815559.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.337049102783203,
|
|
"epoch": 0.08280002639044666,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.0389,
|
|
"mean_token_accuracy": 0.1198160044848919,
|
|
"num_tokens": 2826450.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.270413160324097,
|
|
"epoch": 0.08312990697367553,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.2178,
|
|
"mean_token_accuracy": 0.11123397275805473,
|
|
"num_tokens": 2837471.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.503760528564453,
|
|
"epoch": 0.0834597875569044,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.2566,
|
|
"mean_token_accuracy": 0.11070210263133048,
|
|
"num_tokens": 2849110.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.296745872497558,
|
|
"epoch": 0.08378966814013328,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.0871,
|
|
"mean_token_accuracy": 0.12081556469202041,
|
|
"num_tokens": 2860675.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.326677370071411,
|
|
"epoch": 0.08411954872336214,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.1927,
|
|
"mean_token_accuracy": 0.11729966774582863,
|
|
"num_tokens": 2872482.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.381938552856445,
|
|
"epoch": 0.08444942930659101,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.0714,
|
|
"mean_token_accuracy": 0.1179822638630867,
|
|
"num_tokens": 2882199.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.34971432685852,
|
|
"epoch": 0.08477930988981988,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.1134,
|
|
"mean_token_accuracy": 0.11833830401301385,
|
|
"num_tokens": 2892581.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.351896286010742,
|
|
"epoch": 0.08510919047304875,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.1352,
|
|
"mean_token_accuracy": 0.1160293884575367,
|
|
"num_tokens": 2904105.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.32787446975708,
|
|
"epoch": 0.08543907105627763,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.0727,
|
|
"mean_token_accuracy": 0.11833698526024819,
|
|
"num_tokens": 2914818.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.283242845535279,
|
|
"epoch": 0.0857689516395065,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.0902,
|
|
"mean_token_accuracy": 0.11517246216535568,
|
|
"num_tokens": 2925904.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.411873006820679,
|
|
"epoch": 0.08609883222273537,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.1266,
|
|
"mean_token_accuracy": 0.12158060446381569,
|
|
"num_tokens": 2937177.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.264025449752808,
|
|
"epoch": 0.08642871280596424,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.1313,
|
|
"mean_token_accuracy": 0.11671579852700234,
|
|
"num_tokens": 2948641.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.323843240737915,
|
|
"epoch": 0.08675859338919312,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.1582,
|
|
"mean_token_accuracy": 0.11641139760613442,
|
|
"num_tokens": 2960308.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.296243333816529,
|
|
"epoch": 0.08708847397242199,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.1697,
|
|
"mean_token_accuracy": 0.12116853073239327,
|
|
"num_tokens": 2972666.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.374683809280396,
|
|
"epoch": 0.08741835455565085,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.163,
|
|
"mean_token_accuracy": 0.1157516896724701,
|
|
"num_tokens": 2982894.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.366577816009522,
|
|
"epoch": 0.08774823513887972,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.0644,
|
|
"mean_token_accuracy": 0.12436588779091835,
|
|
"num_tokens": 2992617.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.234115743637085,
|
|
"epoch": 0.0880781157221086,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.0646,
|
|
"mean_token_accuracy": 0.11831269934773445,
|
|
"num_tokens": 3003439.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.431392860412598,
|
|
"epoch": 0.08840799630533747,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.2486,
|
|
"mean_token_accuracy": 0.11635357066988945,
|
|
"num_tokens": 3015204.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.386788988113404,
|
|
"epoch": 0.08873787688856634,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.1794,
|
|
"mean_token_accuracy": 0.12042261585593224,
|
|
"num_tokens": 3026624.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.286476135253906,
|
|
"epoch": 0.08906775747179521,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.0423,
|
|
"mean_token_accuracy": 0.12152522057294846,
|
|
"num_tokens": 3037828.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.296923875808716,
|
|
"epoch": 0.08939763805502408,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.0494,
|
|
"mean_token_accuracy": 0.12242811545729637,
|
|
"num_tokens": 3049666.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.2433209896087645,
|
|
"epoch": 0.08972751863825296,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.0924,
|
|
"mean_token_accuracy": 0.1256846785545349,
|
|
"num_tokens": 3061090.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.278544187545776,
|
|
"epoch": 0.09005739922148183,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.1306,
|
|
"mean_token_accuracy": 0.11453063562512397,
|
|
"num_tokens": 3073031.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.26322603225708,
|
|
"epoch": 0.09038727980471069,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.0688,
|
|
"mean_token_accuracy": 0.12036575302481652,
|
|
"num_tokens": 3084858.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.407075834274292,
|
|
"epoch": 0.09071716038793956,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.1054,
|
|
"mean_token_accuracy": 0.11281666085124016,
|
|
"num_tokens": 3095554.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.248094177246093,
|
|
"epoch": 0.09104704097116843,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.0643,
|
|
"mean_token_accuracy": 0.1204713374376297,
|
|
"num_tokens": 3107164.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.2559247493743895,
|
|
"epoch": 0.0913769215543973,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.1536,
|
|
"mean_token_accuracy": 0.1127657175064087,
|
|
"num_tokens": 3118681.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.407009935379028,
|
|
"epoch": 0.09170680213762618,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.1101,
|
|
"mean_token_accuracy": 0.11909232214093209,
|
|
"num_tokens": 3130604.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.287244367599487,
|
|
"epoch": 0.09203668272085505,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.1052,
|
|
"mean_token_accuracy": 0.116939115524292,
|
|
"num_tokens": 3141774.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.28112907409668,
|
|
"epoch": 0.09236656330408392,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.0033,
|
|
"mean_token_accuracy": 0.12666950821876527,
|
|
"num_tokens": 3151693.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.246624565124511,
|
|
"epoch": 0.0926964438873128,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 5.9844,
|
|
"mean_token_accuracy": 0.1263419881463051,
|
|
"num_tokens": 3162832.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.230603456497192,
|
|
"epoch": 0.09302632447054167,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.0362,
|
|
"mean_token_accuracy": 0.13186612203717232,
|
|
"num_tokens": 3174048.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.197139406204224,
|
|
"epoch": 0.09335620505377054,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.0143,
|
|
"mean_token_accuracy": 0.12525417655706406,
|
|
"num_tokens": 3185451.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.291043329238891,
|
|
"epoch": 0.0936860856369994,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.0907,
|
|
"mean_token_accuracy": 0.12228038236498832,
|
|
"num_tokens": 3197405.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.288193607330323,
|
|
"epoch": 0.09401596622022827,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.0708,
|
|
"mean_token_accuracy": 0.12197377160191536,
|
|
"num_tokens": 3208421.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.2467175960540775,
|
|
"epoch": 0.09434584680345715,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 5.9793,
|
|
"mean_token_accuracy": 0.12171160653233529,
|
|
"num_tokens": 3219745.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.232208776473999,
|
|
"epoch": 0.09467572738668602,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.0716,
|
|
"mean_token_accuracy": 0.12090381532907486,
|
|
"num_tokens": 3231347.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.211804819107056,
|
|
"epoch": 0.09500560796991489,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.0455,
|
|
"mean_token_accuracy": 0.12001732885837554,
|
|
"num_tokens": 3242616.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.160833644866943,
|
|
"epoch": 0.09533548855314376,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.9419,
|
|
"mean_token_accuracy": 0.1280081771314144,
|
|
"num_tokens": 3253697.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.303204488754273,
|
|
"epoch": 0.09566536913637264,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.0534,
|
|
"mean_token_accuracy": 0.1217152826488018,
|
|
"num_tokens": 3263800.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.257365798950195,
|
|
"epoch": 0.09599524971960151,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.0204,
|
|
"mean_token_accuracy": 0.12240952104330063,
|
|
"num_tokens": 3274538.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.264223718643189,
|
|
"epoch": 0.09632513030283038,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.0163,
|
|
"mean_token_accuracy": 0.1247081845998764,
|
|
"num_tokens": 3285719.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.265989780426025,
|
|
"epoch": 0.09665501088605925,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.1308,
|
|
"mean_token_accuracy": 0.11454476118087768,
|
|
"num_tokens": 3296715.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.278552579879761,
|
|
"epoch": 0.09698489146928811,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.0749,
|
|
"mean_token_accuracy": 0.12230216190218926,
|
|
"num_tokens": 3308800.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.281282806396485,
|
|
"epoch": 0.09731477205251698,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.1972,
|
|
"mean_token_accuracy": 0.12018877789378166,
|
|
"num_tokens": 3322180.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.260345363616944,
|
|
"epoch": 0.09764465263574586,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 5.9877,
|
|
"mean_token_accuracy": 0.11602422222495079,
|
|
"num_tokens": 3333363.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.2602760791778564,
|
|
"epoch": 0.09797453321897473,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.0591,
|
|
"mean_token_accuracy": 0.12210157290101051,
|
|
"num_tokens": 3345040.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.321555471420288,
|
|
"epoch": 0.0983044138022036,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.1333,
|
|
"mean_token_accuracy": 0.1185726396739483,
|
|
"num_tokens": 3356445.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.195455074310303,
|
|
"epoch": 0.09863429438543247,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 5.9938,
|
|
"mean_token_accuracy": 0.1257028914988041,
|
|
"num_tokens": 3366948.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.255468225479126,
|
|
"epoch": 0.09896417496866135,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.0501,
|
|
"mean_token_accuracy": 0.11747408881783486,
|
|
"num_tokens": 3378616.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 835065633669120.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|