c375242180
Model: fpadovani/isl-latn-100mb-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": 3.0550178298522668,
|
|
"eval_steps": 1000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.044518566131591,
|
|
"epoch": 0.005094243504839531,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.993,
|
|
"mean_token_accuracy": 0.09492013677954673,
|
|
"num_tokens": 12235.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.07644681930542,
|
|
"epoch": 0.010188487009679063,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.1362,
|
|
"mean_token_accuracy": 0.09126132428646087,
|
|
"num_tokens": 24341.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.072799730300903,
|
|
"epoch": 0.015282730514518594,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.9928,
|
|
"mean_token_accuracy": 0.10187065303325653,
|
|
"num_tokens": 34505.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.125888061523438,
|
|
"epoch": 0.020376974019358125,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.0829,
|
|
"mean_token_accuracy": 0.10146418437361718,
|
|
"num_tokens": 45816.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.176222467422486,
|
|
"epoch": 0.02547121752419766,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.1033,
|
|
"mean_token_accuracy": 0.09227820709347725,
|
|
"num_tokens": 57292.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.191669178009033,
|
|
"epoch": 0.030565461029037188,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.9706,
|
|
"mean_token_accuracy": 0.10464787930250168,
|
|
"num_tokens": 70047.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.173120975494385,
|
|
"epoch": 0.03565970453387672,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.8924,
|
|
"mean_token_accuracy": 0.10731507688760758,
|
|
"num_tokens": 82218.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.1837273120880125,
|
|
"epoch": 0.04075394803871625,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.9774,
|
|
"mean_token_accuracy": 0.10706076249480248,
|
|
"num_tokens": 94126.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.132718229293824,
|
|
"epoch": 0.04584819154355578,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.0128,
|
|
"mean_token_accuracy": 0.09793755188584327,
|
|
"num_tokens": 106527.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.159356594085693,
|
|
"epoch": 0.05094243504839532,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.9901,
|
|
"mean_token_accuracy": 0.10278031975030899,
|
|
"num_tokens": 118284.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.237170743942261,
|
|
"epoch": 0.056036678553234846,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.0489,
|
|
"mean_token_accuracy": 0.09143339470028877,
|
|
"num_tokens": 129462.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.230800580978394,
|
|
"epoch": 0.061130922058074376,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.0246,
|
|
"mean_token_accuracy": 0.094233488291502,
|
|
"num_tokens": 140488.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.194861650466919,
|
|
"epoch": 0.06622516556291391,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.0597,
|
|
"mean_token_accuracy": 0.0934529423713684,
|
|
"num_tokens": 151931.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.164480590820313,
|
|
"epoch": 0.07131940906775344,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.9843,
|
|
"mean_token_accuracy": 0.10014167577028274,
|
|
"num_tokens": 162824.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.177510499954224,
|
|
"epoch": 0.07641365257259297,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.9969,
|
|
"mean_token_accuracy": 0.09918644800782203,
|
|
"num_tokens": 173502.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.232845878601074,
|
|
"epoch": 0.0815078960774325,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.9813,
|
|
"mean_token_accuracy": 0.10139667466282845,
|
|
"num_tokens": 183899.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.129962968826294,
|
|
"epoch": 0.08660213958227203,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.9268,
|
|
"mean_token_accuracy": 0.10928690582513809,
|
|
"num_tokens": 195359.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.133749294281006,
|
|
"epoch": 0.09169638308711156,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.9783,
|
|
"mean_token_accuracy": 0.10235027223825455,
|
|
"num_tokens": 206472.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.171773624420166,
|
|
"epoch": 0.09679062659195109,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.9618,
|
|
"mean_token_accuracy": 0.1039932630956173,
|
|
"num_tokens": 217846.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.125865650177002,
|
|
"epoch": 0.10188487009679063,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.8912,
|
|
"mean_token_accuracy": 0.10703418701887131,
|
|
"num_tokens": 229036.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.243703031539917,
|
|
"epoch": 0.10697911360163016,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.0957,
|
|
"mean_token_accuracy": 0.09529214948415757,
|
|
"num_tokens": 239838.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.230321931838989,
|
|
"epoch": 0.11207335710646969,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.0036,
|
|
"mean_token_accuracy": 0.10142878666520119,
|
|
"num_tokens": 252006.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.1257548332214355,
|
|
"epoch": 0.11716760061130922,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.9558,
|
|
"mean_token_accuracy": 0.10273786783218383,
|
|
"num_tokens": 264027.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.194825267791748,
|
|
"epoch": 0.12226184411614875,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.938,
|
|
"mean_token_accuracy": 0.10348320975899697,
|
|
"num_tokens": 275156.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.3084108352661135,
|
|
"epoch": 0.1273560876209883,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.9773,
|
|
"mean_token_accuracy": 0.10230008214712143,
|
|
"num_tokens": 286240.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.107938385009765,
|
|
"epoch": 0.13245033112582782,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.0216,
|
|
"mean_token_accuracy": 0.09936515465378762,
|
|
"num_tokens": 298136.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.191159009933472,
|
|
"epoch": 0.13754457463066735,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.9527,
|
|
"mean_token_accuracy": 0.1035711295902729,
|
|
"num_tokens": 309682.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.143428659439087,
|
|
"epoch": 0.14263881813550688,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.9154,
|
|
"mean_token_accuracy": 0.11162546351552009,
|
|
"num_tokens": 321330.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.213088083267212,
|
|
"epoch": 0.1477330616403464,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.9336,
|
|
"mean_token_accuracy": 0.10514834746718407,
|
|
"num_tokens": 333364.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.1018579483032225,
|
|
"epoch": 0.15282730514518594,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.8841,
|
|
"mean_token_accuracy": 0.10978019163012505,
|
|
"num_tokens": 345994.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.230183935165405,
|
|
"epoch": 0.15792154865002547,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.9735,
|
|
"mean_token_accuracy": 0.10223098248243331,
|
|
"num_tokens": 357966.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.25154333114624,
|
|
"epoch": 0.163015792154865,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.9546,
|
|
"mean_token_accuracy": 0.10012312904000283,
|
|
"num_tokens": 368890.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.190155982971191,
|
|
"epoch": 0.16811003565970453,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.8739,
|
|
"mean_token_accuracy": 0.10893138274550437,
|
|
"num_tokens": 379995.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.124640941619873,
|
|
"epoch": 0.17320427916454406,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.9154,
|
|
"mean_token_accuracy": 0.10028746277093888,
|
|
"num_tokens": 392482.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.157647562026978,
|
|
"epoch": 0.1782985226693836,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.9532,
|
|
"mean_token_accuracy": 0.10032520517706871,
|
|
"num_tokens": 404318.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.221301460266114,
|
|
"epoch": 0.18339276617422312,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.9243,
|
|
"mean_token_accuracy": 0.10627947300672531,
|
|
"num_tokens": 416223.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.171807718276978,
|
|
"epoch": 0.18848700967906265,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.8954,
|
|
"mean_token_accuracy": 0.10836421623826027,
|
|
"num_tokens": 428375.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.175302124023437,
|
|
"epoch": 0.19358125318390218,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.9459,
|
|
"mean_token_accuracy": 0.10309424996376038,
|
|
"num_tokens": 442032.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.174814701080322,
|
|
"epoch": 0.1986754966887417,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.8798,
|
|
"mean_token_accuracy": 0.10339492037892342,
|
|
"num_tokens": 452898.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.154580116271973,
|
|
"epoch": 0.20376974019358127,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.9385,
|
|
"mean_token_accuracy": 0.10731036365032195,
|
|
"num_tokens": 465037.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.2330575466156,
|
|
"epoch": 0.2088639836984208,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.9432,
|
|
"mean_token_accuracy": 0.09682015776634216,
|
|
"num_tokens": 477816.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.174437618255615,
|
|
"epoch": 0.21395822720326033,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.0381,
|
|
"mean_token_accuracy": 0.1009465865790844,
|
|
"num_tokens": 489226.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.253626441955566,
|
|
"epoch": 0.21905247070809986,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.9718,
|
|
"mean_token_accuracy": 0.10117456316947937,
|
|
"num_tokens": 501391.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.217381477355957,
|
|
"epoch": 0.22414671421293939,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.0043,
|
|
"mean_token_accuracy": 0.0970021516084671,
|
|
"num_tokens": 513755.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.134067535400391,
|
|
"epoch": 0.22924095771777891,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.8867,
|
|
"mean_token_accuracy": 0.10816463977098464,
|
|
"num_tokens": 525814.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.201694059371948,
|
|
"epoch": 0.23433520122261844,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.9078,
|
|
"mean_token_accuracy": 0.11500680074095726,
|
|
"num_tokens": 537982.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.095086097717285,
|
|
"epoch": 0.23942944472745797,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.8092,
|
|
"mean_token_accuracy": 0.11077843755483627,
|
|
"num_tokens": 549795.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.057131576538086,
|
|
"epoch": 0.2445236882322975,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.8448,
|
|
"mean_token_accuracy": 0.10624202638864517,
|
|
"num_tokens": 561704.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.160128021240235,
|
|
"epoch": 0.24961793173713703,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.8942,
|
|
"mean_token_accuracy": 0.10151438117027282,
|
|
"num_tokens": 574736.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.185756397247315,
|
|
"epoch": 0.2547121752419766,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.9737,
|
|
"mean_token_accuracy": 0.10246768370270729,
|
|
"num_tokens": 586785.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.114655065536499,
|
|
"epoch": 0.2598064187468161,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.8574,
|
|
"mean_token_accuracy": 0.10742970556020737,
|
|
"num_tokens": 598738.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.136599349975586,
|
|
"epoch": 0.26490066225165565,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.8983,
|
|
"mean_token_accuracy": 0.10753594040870666,
|
|
"num_tokens": 611431.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.19928994178772,
|
|
"epoch": 0.26999490575649515,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9071,
|
|
"mean_token_accuracy": 0.10483545064926147,
|
|
"num_tokens": 623253.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.116682386398315,
|
|
"epoch": 0.2750891492613347,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.9065,
|
|
"mean_token_accuracy": 0.10133408531546592,
|
|
"num_tokens": 634663.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.220209312438965,
|
|
"epoch": 0.2801833927661742,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.8353,
|
|
"mean_token_accuracy": 0.11484303623437882,
|
|
"num_tokens": 644578.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.124927711486817,
|
|
"epoch": 0.28527763627101377,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.9214,
|
|
"mean_token_accuracy": 0.10066071227192878,
|
|
"num_tokens": 656226.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.117072820663452,
|
|
"epoch": 0.29037187977585327,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.8502,
|
|
"mean_token_accuracy": 0.10427462831139564,
|
|
"num_tokens": 668863.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.118947982788086,
|
|
"epoch": 0.2954661232806928,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8088,
|
|
"mean_token_accuracy": 0.10940986797213555,
|
|
"num_tokens": 679993.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.157330179214478,
|
|
"epoch": 0.30056036678553233,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.8971,
|
|
"mean_token_accuracy": 0.10062931329011918,
|
|
"num_tokens": 692197.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.231019306182861,
|
|
"epoch": 0.3056546102903719,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9478,
|
|
"mean_token_accuracy": 0.09776001274585724,
|
|
"num_tokens": 703363.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.169440603256225,
|
|
"epoch": 0.3107488537952114,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9333,
|
|
"mean_token_accuracy": 0.10456420406699181,
|
|
"num_tokens": 714335.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.152110528945923,
|
|
"epoch": 0.31584309730005095,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.9296,
|
|
"mean_token_accuracy": 0.10086682066321373,
|
|
"num_tokens": 726638.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.169084215164185,
|
|
"epoch": 0.32093734080489045,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.8097,
|
|
"mean_token_accuracy": 0.1081866279244423,
|
|
"num_tokens": 737115.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.0811069965362545,
|
|
"epoch": 0.32603158430973,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.8297,
|
|
"mean_token_accuracy": 0.1085408754646778,
|
|
"num_tokens": 748999.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.146255779266357,
|
|
"epoch": 0.33112582781456956,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.8223,
|
|
"mean_token_accuracy": 0.10416458621621132,
|
|
"num_tokens": 759858.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.081164216995239,
|
|
"epoch": 0.33622007131940906,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.7889,
|
|
"mean_token_accuracy": 0.10959977135062218,
|
|
"num_tokens": 770636.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.057614755630493,
|
|
"epoch": 0.3413143148242486,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.865,
|
|
"mean_token_accuracy": 0.10254048258066177,
|
|
"num_tokens": 783221.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.082399082183838,
|
|
"epoch": 0.3464085583290881,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.7952,
|
|
"mean_token_accuracy": 0.10752720311284066,
|
|
"num_tokens": 793548.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.148776912689209,
|
|
"epoch": 0.3515028018339277,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8156,
|
|
"mean_token_accuracy": 0.10141635611653328,
|
|
"num_tokens": 805553.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.078239059448242,
|
|
"epoch": 0.3565970453387672,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.8902,
|
|
"mean_token_accuracy": 0.09888725131750106,
|
|
"num_tokens": 818475.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.004705142974854,
|
|
"epoch": 0.36169128884360674,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.7757,
|
|
"mean_token_accuracy": 0.11100321486592293,
|
|
"num_tokens": 830826.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.182587623596191,
|
|
"epoch": 0.36678553234844624,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8619,
|
|
"mean_token_accuracy": 0.10591038465499877,
|
|
"num_tokens": 842850.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.009341144561768,
|
|
"epoch": 0.3718797758532858,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.8421,
|
|
"mean_token_accuracy": 0.09876091927289962,
|
|
"num_tokens": 855954.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.163160562515259,
|
|
"epoch": 0.3769740193581253,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9058,
|
|
"mean_token_accuracy": 0.10104744359850884,
|
|
"num_tokens": 867986.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.08941912651062,
|
|
"epoch": 0.38206826286296486,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.829,
|
|
"mean_token_accuracy": 0.10875942558050156,
|
|
"num_tokens": 880063.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.049508047103882,
|
|
"epoch": 0.38716250636780436,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8143,
|
|
"mean_token_accuracy": 0.10336368456482888,
|
|
"num_tokens": 893320.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.012372732162476,
|
|
"epoch": 0.3922567498726439,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.7175,
|
|
"mean_token_accuracy": 0.11578772887587548,
|
|
"num_tokens": 905118.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.087364292144775,
|
|
"epoch": 0.3973509933774834,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8228,
|
|
"mean_token_accuracy": 0.10489340797066689,
|
|
"num_tokens": 916403.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.149967098236084,
|
|
"epoch": 0.402445236882323,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8766,
|
|
"mean_token_accuracy": 0.10522974282503128,
|
|
"num_tokens": 927811.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.91042160987854,
|
|
"epoch": 0.40753948038716253,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.7217,
|
|
"mean_token_accuracy": 0.10639820843935013,
|
|
"num_tokens": 940095.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.022790098190308,
|
|
"epoch": 0.41263372389200204,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7787,
|
|
"mean_token_accuracy": 0.11006416901946067,
|
|
"num_tokens": 951985.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.055878973007202,
|
|
"epoch": 0.4177279673968416,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7782,
|
|
"mean_token_accuracy": 0.11026687249541282,
|
|
"num_tokens": 963981.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.1019713401794435,
|
|
"epoch": 0.4228222109016811,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8659,
|
|
"mean_token_accuracy": 0.10093940123915672,
|
|
"num_tokens": 976389.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.013764762878418,
|
|
"epoch": 0.42791645440652065,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7817,
|
|
"mean_token_accuracy": 0.10758809745311737,
|
|
"num_tokens": 987712.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.016055631637573,
|
|
"epoch": 0.43301069791136015,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7678,
|
|
"mean_token_accuracy": 0.10572115927934647,
|
|
"num_tokens": 998407.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.109854650497437,
|
|
"epoch": 0.4381049414161997,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8343,
|
|
"mean_token_accuracy": 0.1021016150712967,
|
|
"num_tokens": 1010275.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.032284593582153,
|
|
"epoch": 0.4431991849210392,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7488,
|
|
"mean_token_accuracy": 0.1083029642701149,
|
|
"num_tokens": 1022357.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.910487174987793,
|
|
"epoch": 0.44829342842587877,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.6801,
|
|
"mean_token_accuracy": 0.1184845358133316,
|
|
"num_tokens": 1033751.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.059388017654419,
|
|
"epoch": 0.4533876719307183,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7495,
|
|
"mean_token_accuracy": 0.11273220926523209,
|
|
"num_tokens": 1045464.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.905271768569946,
|
|
"epoch": 0.45848191543555783,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7088,
|
|
"mean_token_accuracy": 0.11199920400977134,
|
|
"num_tokens": 1057281.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.027792119979859,
|
|
"epoch": 0.46357615894039733,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.642,
|
|
"mean_token_accuracy": 0.11511161923408508,
|
|
"num_tokens": 1070169.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.8752524852752686,
|
|
"epoch": 0.4686704024452369,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6888,
|
|
"mean_token_accuracy": 0.12004167884588242,
|
|
"num_tokens": 1081207.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.927147293090821,
|
|
"epoch": 0.4737646459500764,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7989,
|
|
"mean_token_accuracy": 0.10520728453993797,
|
|
"num_tokens": 1093846.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.048350286483765,
|
|
"epoch": 0.47885888945491595,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7787,
|
|
"mean_token_accuracy": 0.10956415981054306,
|
|
"num_tokens": 1106360.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.9273552894592285,
|
|
"epoch": 0.48395313295975545,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6489,
|
|
"mean_token_accuracy": 0.1187779150903225,
|
|
"num_tokens": 1118042.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.917570686340332,
|
|
"epoch": 0.489047376464595,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6714,
|
|
"mean_token_accuracy": 0.10860469713807105,
|
|
"num_tokens": 1129625.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.9747683048248295,
|
|
"epoch": 0.49414161996943456,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6961,
|
|
"mean_token_accuracy": 0.11883748769760132,
|
|
"num_tokens": 1140416.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.994707870483398,
|
|
"epoch": 0.49923586347427407,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6787,
|
|
"mean_token_accuracy": 0.11599407494068145,
|
|
"num_tokens": 1152273.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.902123832702637,
|
|
"epoch": 0.5043301069791136,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6208,
|
|
"mean_token_accuracy": 0.11468892842531205,
|
|
"num_tokens": 1163621.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.947891044616699,
|
|
"epoch": 0.5094243504839532,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.7869,
|
|
"mean_token_accuracy": 0.1068939983844757,
|
|
"num_tokens": 1174445.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.92851185798645,
|
|
"epoch": 0.5145185939887926,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7171,
|
|
"mean_token_accuracy": 0.11412775367498398,
|
|
"num_tokens": 1187816.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.922699451446533,
|
|
"epoch": 0.5196128374936322,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.5966,
|
|
"mean_token_accuracy": 0.1230049654841423,
|
|
"num_tokens": 1200523.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.88109769821167,
|
|
"epoch": 0.5247070809984717,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6693,
|
|
"mean_token_accuracy": 0.11688895672559738,
|
|
"num_tokens": 1211841.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.85207200050354,
|
|
"epoch": 0.5298013245033113,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6275,
|
|
"mean_token_accuracy": 0.1219951406121254,
|
|
"num_tokens": 1224076.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.885891246795654,
|
|
"epoch": 0.5348955680081507,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6764,
|
|
"mean_token_accuracy": 0.1159809947013855,
|
|
"num_tokens": 1236079.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.935859680175781,
|
|
"epoch": 0.5399898115129903,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6735,
|
|
"mean_token_accuracy": 0.11025876700878143,
|
|
"num_tokens": 1248922.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.934536552429199,
|
|
"epoch": 0.5450840550178299,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.685,
|
|
"mean_token_accuracy": 0.10967916175723076,
|
|
"num_tokens": 1260471.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.866807699203491,
|
|
"epoch": 0.5501782985226694,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7325,
|
|
"mean_token_accuracy": 0.11336875557899476,
|
|
"num_tokens": 1272295.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.961508083343506,
|
|
"epoch": 0.5552725420275089,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.7008,
|
|
"mean_token_accuracy": 0.10831127315759659,
|
|
"num_tokens": 1285056.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.934025669097901,
|
|
"epoch": 0.5603667855323484,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6645,
|
|
"mean_token_accuracy": 0.1119805321097374,
|
|
"num_tokens": 1296371.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.887786865234375,
|
|
"epoch": 0.565461029037188,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6499,
|
|
"mean_token_accuracy": 0.12152763158082962,
|
|
"num_tokens": 1308428.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.899425601959228,
|
|
"epoch": 0.5705552725420275,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.7455,
|
|
"mean_token_accuracy": 0.1115893378853798,
|
|
"num_tokens": 1320586.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.900220680236816,
|
|
"epoch": 0.5756495160468671,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7336,
|
|
"mean_token_accuracy": 0.10702209770679474,
|
|
"num_tokens": 1332162.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.933666753768921,
|
|
"epoch": 0.5807437595517065,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7132,
|
|
"mean_token_accuracy": 0.11646312922239303,
|
|
"num_tokens": 1343574.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.876138305664062,
|
|
"epoch": 0.5858380030565461,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5788,
|
|
"mean_token_accuracy": 0.113432015478611,
|
|
"num_tokens": 1353970.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.9562897205352785,
|
|
"epoch": 0.5909322465613857,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7187,
|
|
"mean_token_accuracy": 0.11172034293413162,
|
|
"num_tokens": 1366324.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.93915491104126,
|
|
"epoch": 0.5960264900662252,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.7446,
|
|
"mean_token_accuracy": 0.11083488464355469,
|
|
"num_tokens": 1378778.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.843048238754273,
|
|
"epoch": 0.6011207335710647,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6485,
|
|
"mean_token_accuracy": 0.11136462986469269,
|
|
"num_tokens": 1390305.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.851701068878174,
|
|
"epoch": 0.6062149770759042,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6333,
|
|
"mean_token_accuracy": 0.12062104865908622,
|
|
"num_tokens": 1401122.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.851728773117065,
|
|
"epoch": 0.6113092205807438,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.6369,
|
|
"mean_token_accuracy": 0.1119437888264656,
|
|
"num_tokens": 1413598.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.8240314483642575,
|
|
"epoch": 0.6164034640855833,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6199,
|
|
"mean_token_accuracy": 0.11327560693025589,
|
|
"num_tokens": 1425344.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.848186206817627,
|
|
"epoch": 0.6214977075904228,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6384,
|
|
"mean_token_accuracy": 0.11503277495503425,
|
|
"num_tokens": 1439096.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.904638338088989,
|
|
"epoch": 0.6265919510952623,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7713,
|
|
"mean_token_accuracy": 0.10886242017149925,
|
|
"num_tokens": 1451207.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.877414178848267,
|
|
"epoch": 0.6316861946001019,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6737,
|
|
"mean_token_accuracy": 0.1112053357064724,
|
|
"num_tokens": 1463404.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.744198608398437,
|
|
"epoch": 0.6367804381049414,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5379,
|
|
"mean_token_accuracy": 0.12282706797122955,
|
|
"num_tokens": 1474609.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.912109327316284,
|
|
"epoch": 0.6418746816097809,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7093,
|
|
"mean_token_accuracy": 0.11195330247282982,
|
|
"num_tokens": 1486323.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.812345170974732,
|
|
"epoch": 0.6469689251146205,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5772,
|
|
"mean_token_accuracy": 0.11727494150400161,
|
|
"num_tokens": 1497325.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.790488815307617,
|
|
"epoch": 0.65206316861946,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5539,
|
|
"mean_token_accuracy": 0.1198957234621048,
|
|
"num_tokens": 1507329.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.799358463287353,
|
|
"epoch": 0.6571574121242996,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5975,
|
|
"mean_token_accuracy": 0.1194250911474228,
|
|
"num_tokens": 1518927.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.782226657867431,
|
|
"epoch": 0.6622516556291391,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.6371,
|
|
"mean_token_accuracy": 0.11219422966241836,
|
|
"num_tokens": 1529171.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.951243925094604,
|
|
"epoch": 0.6673458991339786,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6753,
|
|
"mean_token_accuracy": 0.11055781468749046,
|
|
"num_tokens": 1541295.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.83675708770752,
|
|
"epoch": 0.6724401426388181,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.6878,
|
|
"mean_token_accuracy": 0.10897862017154694,
|
|
"num_tokens": 1553865.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.816014432907105,
|
|
"epoch": 0.6775343861436577,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5663,
|
|
"mean_token_accuracy": 0.11735440343618393,
|
|
"num_tokens": 1565092.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.822263717651367,
|
|
"epoch": 0.6826286296484972,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6335,
|
|
"mean_token_accuracy": 0.11954260692000389,
|
|
"num_tokens": 1576170.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.819930934906006,
|
|
"epoch": 0.6877228731533367,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5868,
|
|
"mean_token_accuracy": 0.11839340776205062,
|
|
"num_tokens": 1588121.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.731348991394043,
|
|
"epoch": 0.6928171166581762,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.6231,
|
|
"mean_token_accuracy": 0.1168064221739769,
|
|
"num_tokens": 1601141.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.750489807128906,
|
|
"epoch": 0.6979113601630158,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.5861,
|
|
"mean_token_accuracy": 0.11778726801276207,
|
|
"num_tokens": 1613188.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.843847274780273,
|
|
"epoch": 0.7030056036678554,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6027,
|
|
"mean_token_accuracy": 0.12430845946073532,
|
|
"num_tokens": 1625889.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.780045652389527,
|
|
"epoch": 0.7080998471726948,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5655,
|
|
"mean_token_accuracy": 0.1163877122104168,
|
|
"num_tokens": 1636350.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.62624773979187,
|
|
"epoch": 0.7131940906775344,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5472,
|
|
"mean_token_accuracy": 0.1218593180179596,
|
|
"num_tokens": 1648290.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.788105869293213,
|
|
"epoch": 0.7182883341823739,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4555,
|
|
"mean_token_accuracy": 0.12383097037672997,
|
|
"num_tokens": 1658781.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.737675666809082,
|
|
"epoch": 0.7233825776872135,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5874,
|
|
"mean_token_accuracy": 0.11333777979016305,
|
|
"num_tokens": 1671779.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.791052627563476,
|
|
"epoch": 0.7284768211920529,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5221,
|
|
"mean_token_accuracy": 0.12146528512239456,
|
|
"num_tokens": 1683821.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.62400598526001,
|
|
"epoch": 0.7335710646968925,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4923,
|
|
"mean_token_accuracy": 0.12247836142778397,
|
|
"num_tokens": 1695754.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.735420274734497,
|
|
"epoch": 0.738665308201732,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6048,
|
|
"mean_token_accuracy": 0.11522219553589821,
|
|
"num_tokens": 1707398.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.739327764511108,
|
|
"epoch": 0.7437595517065716,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.5239,
|
|
"mean_token_accuracy": 0.11949873492121696,
|
|
"num_tokens": 1719315.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.7130818367004395,
|
|
"epoch": 0.7488537952114112,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5089,
|
|
"mean_token_accuracy": 0.11913193985819817,
|
|
"num_tokens": 1731406.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.729084157943726,
|
|
"epoch": 0.7539480387162506,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5829,
|
|
"mean_token_accuracy": 0.11685697734355927,
|
|
"num_tokens": 1744941.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.678062200546265,
|
|
"epoch": 0.7590422822210902,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.5819,
|
|
"mean_token_accuracy": 0.11202071979641914,
|
|
"num_tokens": 1757824.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.774014949798584,
|
|
"epoch": 0.7641365257259297,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5827,
|
|
"mean_token_accuracy": 0.11595133021473884,
|
|
"num_tokens": 1770050.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.656177711486817,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5539,
|
|
"mean_token_accuracy": 0.11940988376736641,
|
|
"num_tokens": 1782083.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.733289003372192,
|
|
"epoch": 0.7743250127356087,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4976,
|
|
"mean_token_accuracy": 0.11968936026096344,
|
|
"num_tokens": 1792812.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.62746295928955,
|
|
"epoch": 0.7794192562404483,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4971,
|
|
"mean_token_accuracy": 0.11665747761726379,
|
|
"num_tokens": 1804724.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.635995149612427,
|
|
"epoch": 0.7845134997452878,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5212,
|
|
"mean_token_accuracy": 0.11737691909074784,
|
|
"num_tokens": 1817051.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.7186009883880615,
|
|
"epoch": 0.7896077432501274,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5,
|
|
"mean_token_accuracy": 0.11847119778394699,
|
|
"num_tokens": 1828328.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.57848711013794,
|
|
"epoch": 0.7947019867549668,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4701,
|
|
"mean_token_accuracy": 0.1233375109732151,
|
|
"num_tokens": 1840117.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.669107103347779,
|
|
"epoch": 0.7997962302598064,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4462,
|
|
"mean_token_accuracy": 0.12489347085356713,
|
|
"num_tokens": 1851097.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.625661754608155,
|
|
"epoch": 0.804890473764646,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5077,
|
|
"mean_token_accuracy": 0.1241062693297863,
|
|
"num_tokens": 1862282.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.711694717407227,
|
|
"epoch": 0.8099847172694855,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4885,
|
|
"mean_token_accuracy": 0.12008807882666588,
|
|
"num_tokens": 1874936.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.657589626312256,
|
|
"epoch": 0.8150789607743251,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5164,
|
|
"mean_token_accuracy": 0.12074018567800522,
|
|
"num_tokens": 1887429.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.609248161315918,
|
|
"epoch": 0.8201732042791645,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.495,
|
|
"mean_token_accuracy": 0.11515450328588486,
|
|
"num_tokens": 1899378.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.573548364639282,
|
|
"epoch": 0.8252674477840041,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4321,
|
|
"mean_token_accuracy": 0.12534051835536958,
|
|
"num_tokens": 1910723.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.696687459945679,
|
|
"epoch": 0.8303616912888436,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.483,
|
|
"mean_token_accuracy": 0.12311445102095604,
|
|
"num_tokens": 1921913.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.580182313919067,
|
|
"epoch": 0.8354559347936832,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4802,
|
|
"mean_token_accuracy": 0.12698968648910522,
|
|
"num_tokens": 1934052.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.595705938339234,
|
|
"epoch": 0.8405501782985226,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.5471,
|
|
"mean_token_accuracy": 0.12124906629323959,
|
|
"num_tokens": 1947512.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.63047308921814,
|
|
"epoch": 0.8456444218033622,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4808,
|
|
"mean_token_accuracy": 0.12023026198148727,
|
|
"num_tokens": 1959569.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.630491256713867,
|
|
"epoch": 0.8507386653082017,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4583,
|
|
"mean_token_accuracy": 0.12822288647294044,
|
|
"num_tokens": 1970159.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.557434034347534,
|
|
"epoch": 0.8558329088130413,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5458,
|
|
"mean_token_accuracy": 0.12084418758749962,
|
|
"num_tokens": 1982688.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.6066173076629635,
|
|
"epoch": 0.8609271523178808,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.437,
|
|
"mean_token_accuracy": 0.12091387584805488,
|
|
"num_tokens": 1993733.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.671035385131836,
|
|
"epoch": 0.8660213958227203,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.497,
|
|
"mean_token_accuracy": 0.12560601383447648,
|
|
"num_tokens": 2004880.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.506956005096436,
|
|
"epoch": 0.8711156393275599,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.4735,
|
|
"mean_token_accuracy": 0.12167876958847046,
|
|
"num_tokens": 2017018.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.637276363372803,
|
|
"epoch": 0.8762098828323994,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5467,
|
|
"mean_token_accuracy": 0.11587712764739991,
|
|
"num_tokens": 2028930.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.596051549911499,
|
|
"epoch": 0.8813041263372389,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4726,
|
|
"mean_token_accuracy": 0.11901551336050034,
|
|
"num_tokens": 2039846.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.633983707427978,
|
|
"epoch": 0.8863983698420784,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4729,
|
|
"mean_token_accuracy": 0.12403136044740677,
|
|
"num_tokens": 2051021.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.65854320526123,
|
|
"epoch": 0.891492613346918,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.5927,
|
|
"mean_token_accuracy": 0.11336434632539749,
|
|
"num_tokens": 2064862.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.497711229324341,
|
|
"epoch": 0.8965868568517575,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.435,
|
|
"mean_token_accuracy": 0.12588154077529906,
|
|
"num_tokens": 2076316.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.637341451644898,
|
|
"epoch": 0.9016811003565971,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4642,
|
|
"mean_token_accuracy": 0.12097790688276291,
|
|
"num_tokens": 2088115.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.72122859954834,
|
|
"epoch": 0.9067753438614365,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.6047,
|
|
"mean_token_accuracy": 0.1152612641453743,
|
|
"num_tokens": 2100665.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.499973011016846,
|
|
"epoch": 0.9118695873662761,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4103,
|
|
"mean_token_accuracy": 0.12955493479967117,
|
|
"num_tokens": 2112231.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.561792182922363,
|
|
"epoch": 0.9169638308711157,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3457,
|
|
"mean_token_accuracy": 0.13369542136788368,
|
|
"num_tokens": 2123920.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.54530553817749,
|
|
"epoch": 0.9220580743759552,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4631,
|
|
"mean_token_accuracy": 0.11725258454680443,
|
|
"num_tokens": 2135167.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.622091150283813,
|
|
"epoch": 0.9271523178807947,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.4621,
|
|
"mean_token_accuracy": 0.12155643627047538,
|
|
"num_tokens": 2146809.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.567164516448974,
|
|
"epoch": 0.9322465613856342,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3959,
|
|
"mean_token_accuracy": 0.12437484413385391,
|
|
"num_tokens": 2157784.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.522310543060303,
|
|
"epoch": 0.9373408048904738,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4747,
|
|
"mean_token_accuracy": 0.12132963612675667,
|
|
"num_tokens": 2169748.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.603588724136353,
|
|
"epoch": 0.9424350483953133,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.487,
|
|
"mean_token_accuracy": 0.12010061740875244,
|
|
"num_tokens": 2182370.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.457374954223633,
|
|
"epoch": 0.9475292919001528,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4562,
|
|
"mean_token_accuracy": 0.11921827495098114,
|
|
"num_tokens": 2196043.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.589722347259522,
|
|
"epoch": 0.9526235354049923,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4824,
|
|
"mean_token_accuracy": 0.12165994271636009,
|
|
"num_tokens": 2207240.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.645488262176514,
|
|
"epoch": 0.9577177789098319,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.4934,
|
|
"mean_token_accuracy": 0.1202242136001587,
|
|
"num_tokens": 2219078.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.446953392028808,
|
|
"epoch": 0.9628120224146715,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4912,
|
|
"mean_token_accuracy": 0.12080769687891006,
|
|
"num_tokens": 2231434.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.617646169662476,
|
|
"epoch": 0.9679062659195109,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4011,
|
|
"mean_token_accuracy": 0.13057347685098647,
|
|
"num_tokens": 2241375.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.455522108078003,
|
|
"epoch": 0.9730005094243505,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4403,
|
|
"mean_token_accuracy": 0.12370983511209488,
|
|
"num_tokens": 2253520.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.472168827056885,
|
|
"epoch": 0.97809475292919,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.3584,
|
|
"mean_token_accuracy": 0.12518140971660613,
|
|
"num_tokens": 2264938.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.437531042098999,
|
|
"epoch": 0.9831889964340296,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4799,
|
|
"mean_token_accuracy": 0.11748291850090027,
|
|
"num_tokens": 2277306.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.539975261688232,
|
|
"epoch": 0.9882832399388691,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3544,
|
|
"mean_token_accuracy": 0.11762910708785057,
|
|
"num_tokens": 2288766.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.488857221603394,
|
|
"epoch": 0.9933774834437086,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3959,
|
|
"mean_token_accuracy": 0.12635208591818808,
|
|
"num_tokens": 2299487.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.41902174949646,
|
|
"epoch": 0.9984717269485481,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.3324,
|
|
"mean_token_accuracy": 0.13361692279577256,
|
|
"num_tokens": 2311302.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.477914651234944,
|
|
"epoch": 1.0030565461029037,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3057,
|
|
"mean_token_accuracy": 0.12617213858498466,
|
|
"num_tokens": 2321410.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.458166980743409,
|
|
"epoch": 1.0081507896077433,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.2389,
|
|
"mean_token_accuracy": 0.1335169292986393,
|
|
"num_tokens": 2332931.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.449419021606445,
|
|
"epoch": 1.0132450331125828,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2496,
|
|
"mean_token_accuracy": 0.13030336275696755,
|
|
"num_tokens": 2344207.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.343480253219605,
|
|
"epoch": 1.0183392766174224,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2018,
|
|
"mean_token_accuracy": 0.13757490888237953,
|
|
"num_tokens": 2356470.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0183392766174224,
|
|
"eval_entropy": 6.176096528842125,
|
|
"eval_loss": 6.380763053894043,
|
|
"eval_mean_token_accuracy": 0.13155310050940808,
|
|
"eval_num_tokens": 2356470.0,
|
|
"eval_runtime": 2.7368,
|
|
"eval_samples_per_second": 1183.865,
|
|
"eval_steps_per_second": 147.983,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.544700241088867,
|
|
"epoch": 1.0234335201222617,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999997711134041,
|
|
"loss": 6.2456,
|
|
"mean_token_accuracy": 0.129035347700119,
|
|
"num_tokens": 2368744.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.383368349075317,
|
|
"epoch": 1.0285277636271013,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999988412624068,
|
|
"loss": 6.2642,
|
|
"mean_token_accuracy": 0.12989430651068687,
|
|
"num_tokens": 2380906.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.347857427597046,
|
|
"epoch": 1.0336220071319409,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999971961445492,
|
|
"loss": 6.2064,
|
|
"mean_token_accuracy": 0.13972920179367065,
|
|
"num_tokens": 2392358.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.39717173576355,
|
|
"epoch": 1.0387162506367804,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999948357650613,
|
|
"loss": 6.1755,
|
|
"mean_token_accuracy": 0.13360414057970046,
|
|
"num_tokens": 2405509.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.437627983093262,
|
|
"epoch": 1.04381049414162,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999917601314468,
|
|
"loss": 6.2318,
|
|
"mean_token_accuracy": 0.12813753336668016,
|
|
"num_tokens": 2417890.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.364025545120239,
|
|
"epoch": 1.0489047376464595,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999879692534828,
|
|
"loss": 6.2279,
|
|
"mean_token_accuracy": 0.12833043411374093,
|
|
"num_tokens": 2429739.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.4404607772827145,
|
|
"epoch": 1.053998981151299,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999834631432208,
|
|
"loss": 6.1682,
|
|
"mean_token_accuracy": 0.13368272110819818,
|
|
"num_tokens": 2441374.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.3169389247894285,
|
|
"epoch": 1.0590932246561386,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999782418149852,
|
|
"loss": 6.1956,
|
|
"mean_token_accuracy": 0.12824264094233512,
|
|
"num_tokens": 2452914.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.3888952255249025,
|
|
"epoch": 1.064187468160978,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999723052853747,
|
|
"loss": 6.2455,
|
|
"mean_token_accuracy": 0.13443526551127433,
|
|
"num_tokens": 2464970.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.4613734722137455,
|
|
"epoch": 1.0692817116658175,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999656535732615,
|
|
"loss": 6.2936,
|
|
"mean_token_accuracy": 0.12898464128375053,
|
|
"num_tokens": 2477563.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.36776294708252,
|
|
"epoch": 1.074375955170657,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999582866997913,
|
|
"loss": 6.1428,
|
|
"mean_token_accuracy": 0.13846859708428383,
|
|
"num_tokens": 2488767.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.33557538986206,
|
|
"epoch": 1.0794701986754967,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000499950204688383,
|
|
"loss": 6.2054,
|
|
"mean_token_accuracy": 0.13428086042404175,
|
|
"num_tokens": 2500260.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.31670503616333,
|
|
"epoch": 1.0845644421803362,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999414075647292,
|
|
"loss": 6.1406,
|
|
"mean_token_accuracy": 0.14031125232577324,
|
|
"num_tokens": 2511581.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.433043956756592,
|
|
"epoch": 1.0896586856851758,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499931895356796,
|
|
"loss": 6.325,
|
|
"mean_token_accuracy": 0.12882162407040595,
|
|
"num_tokens": 2524185.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.333220434188843,
|
|
"epoch": 1.0947529291900153,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999216680948223,
|
|
"loss": 6.1196,
|
|
"mean_token_accuracy": 0.13699559718370438,
|
|
"num_tokens": 2535762.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.198199892044068,
|
|
"epoch": 1.0998471726948549,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999107258113204,
|
|
"loss": 6.0817,
|
|
"mean_token_accuracy": 0.1548200875520706,
|
|
"num_tokens": 2547566.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.360767364501953,
|
|
"epoch": 1.1049414161996944,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998990685410759,
|
|
"loss": 6.2467,
|
|
"mean_token_accuracy": 0.12859098017215728,
|
|
"num_tokens": 2559796.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.299922847747803,
|
|
"epoch": 1.110035659704534,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998866963211465,
|
|
"loss": 6.1665,
|
|
"mean_token_accuracy": 0.13427353277802467,
|
|
"num_tokens": 2573137.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.396093559265137,
|
|
"epoch": 1.1151299032093733,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998736091908636,
|
|
"loss": 6.2108,
|
|
"mean_token_accuracy": 0.1331883393228054,
|
|
"num_tokens": 2586550.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.291216564178467,
|
|
"epoch": 1.1202241467142129,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998598071918309,
|
|
"loss": 6.2358,
|
|
"mean_token_accuracy": 0.1343692310154438,
|
|
"num_tokens": 2597666.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.35298638343811,
|
|
"epoch": 1.1253183902190524,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998452903679246,
|
|
"loss": 6.1061,
|
|
"mean_token_accuracy": 0.1420112669467926,
|
|
"num_tokens": 2608518.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.290657424926758,
|
|
"epoch": 1.130412633723892,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998300587652936,
|
|
"loss": 6.1938,
|
|
"mean_token_accuracy": 0.13614270687103272,
|
|
"num_tokens": 2619766.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.3322988986969,
|
|
"epoch": 1.1355068772287316,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998141124323586,
|
|
"loss": 6.2521,
|
|
"mean_token_accuracy": 0.133444906771183,
|
|
"num_tokens": 2632473.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.341130352020263,
|
|
"epoch": 1.1406011207335711,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997974514198126,
|
|
"loss": 6.1981,
|
|
"mean_token_accuracy": 0.13589926660060883,
|
|
"num_tokens": 2644059.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.261853885650635,
|
|
"epoch": 1.1456953642384107,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997800757806209,
|
|
"loss": 6.1346,
|
|
"mean_token_accuracy": 0.1369706079363823,
|
|
"num_tokens": 2655168.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.423413324356079,
|
|
"epoch": 1.15078960774325,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997619855700203,
|
|
"loss": 6.3038,
|
|
"mean_token_accuracy": 0.13099338263273239,
|
|
"num_tokens": 2665431.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.255408096313476,
|
|
"epoch": 1.1558838512480896,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499743180845519,
|
|
"loss": 6.1735,
|
|
"mean_token_accuracy": 0.14822513610124588,
|
|
"num_tokens": 2676942.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.41400785446167,
|
|
"epoch": 1.1609780947529291,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499723661666897,
|
|
"loss": 6.2384,
|
|
"mean_token_accuracy": 0.13193109929561614,
|
|
"num_tokens": 2688510.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.279343891143799,
|
|
"epoch": 1.1660723382577687,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997034280962054,
|
|
"loss": 6.2299,
|
|
"mean_token_accuracy": 0.13077231124043465,
|
|
"num_tokens": 2702140.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.252119684219361,
|
|
"epoch": 1.1711665817626082,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996824801977663,
|
|
"loss": 6.2019,
|
|
"mean_token_accuracy": 0.13781489580869674,
|
|
"num_tokens": 2714358.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.23788161277771,
|
|
"epoch": 1.1762608252674478,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996608180381727,
|
|
"loss": 6.1201,
|
|
"mean_token_accuracy": 0.14271433502435685,
|
|
"num_tokens": 2725477.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.350845098495483,
|
|
"epoch": 1.1813550687722874,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499638441686288,
|
|
"loss": 6.2527,
|
|
"mean_token_accuracy": 0.1335478439927101,
|
|
"num_tokens": 2736696.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.2867883205413815,
|
|
"epoch": 1.186449312277127,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996153512132462,
|
|
"loss": 6.1221,
|
|
"mean_token_accuracy": 0.14645155072212218,
|
|
"num_tokens": 2747655.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.297123384475708,
|
|
"epoch": 1.1915435557819665,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995915466924517,
|
|
"loss": 6.1563,
|
|
"mean_token_accuracy": 0.13962377607822418,
|
|
"num_tokens": 2758955.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.230033016204834,
|
|
"epoch": 1.196637799286806,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995670281995784,
|
|
"loss": 6.0908,
|
|
"mean_token_accuracy": 0.14304591119289398,
|
|
"num_tokens": 2770395.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.255457496643066,
|
|
"epoch": 1.2017320427916454,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995417958125703,
|
|
"loss": 6.1933,
|
|
"mean_token_accuracy": 0.13439659401774406,
|
|
"num_tokens": 2783043.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.337797737121582,
|
|
"epoch": 1.206826286296485,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995158496116402,
|
|
"loss": 6.3264,
|
|
"mean_token_accuracy": 0.127481097728014,
|
|
"num_tokens": 2795937.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.308899641036987,
|
|
"epoch": 1.2119205298013245,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004994891896792712,
|
|
"loss": 6.2637,
|
|
"mean_token_accuracy": 0.13542621061205865,
|
|
"num_tokens": 2807812.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.2987548828125,
|
|
"epoch": 1.217014773306164,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004994618161002142,
|
|
"loss": 6.1658,
|
|
"mean_token_accuracy": 0.14131661579012872,
|
|
"num_tokens": 2819326.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.282303047180176,
|
|
"epoch": 1.2221090168110036,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994337289614893,
|
|
"loss": 6.1073,
|
|
"mean_token_accuracy": 0.1444818675518036,
|
|
"num_tokens": 2830917.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.239668369293213,
|
|
"epoch": 1.2272032603158431,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004994049283523854,
|
|
"loss": 6.2191,
|
|
"mean_token_accuracy": 0.13711530938744546,
|
|
"num_tokens": 2842828.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.2332727909088135,
|
|
"epoch": 1.2322975038206827,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004993754143644585,
|
|
"loss": 6.0111,
|
|
"mean_token_accuracy": 0.1447136551141739,
|
|
"num_tokens": 2854423.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.247496604919434,
|
|
"epoch": 1.237391747325522,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004993451870915333,
|
|
"loss": 6.25,
|
|
"mean_token_accuracy": 0.13692307323217393,
|
|
"num_tokens": 2866032.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.267179250717163,
|
|
"epoch": 1.2424859908303616,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004993142466297016,
|
|
"loss": 6.2069,
|
|
"mean_token_accuracy": 0.14005452692508696,
|
|
"num_tokens": 2877147.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.1810588359832765,
|
|
"epoch": 1.2475802343352012,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004992825930773228,
|
|
"loss": 6.1111,
|
|
"mean_token_accuracy": 0.14460208490490914,
|
|
"num_tokens": 2888723.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.332695770263672,
|
|
"epoch": 1.2526744778400407,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004992502265350224,
|
|
"loss": 6.1751,
|
|
"mean_token_accuracy": 0.1401965655386448,
|
|
"num_tokens": 2901158.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.209536743164063,
|
|
"epoch": 1.2577687213448803,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004992171471056933,
|
|
"loss": 6.1942,
|
|
"mean_token_accuracy": 0.1339487463235855,
|
|
"num_tokens": 2913303.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.271416187286377,
|
|
"epoch": 1.2628629648497198,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004991833548944945,
|
|
"loss": 6.1654,
|
|
"mean_token_accuracy": 0.13918142765760422,
|
|
"num_tokens": 2924036.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.281953620910644,
|
|
"epoch": 1.2679572083545594,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004991488500088505,
|
|
"loss": 6.2536,
|
|
"mean_token_accuracy": 0.13643498942255974,
|
|
"num_tokens": 2936031.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.252941751480103,
|
|
"epoch": 1.273051451859399,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991136325584516,
|
|
"loss": 6.131,
|
|
"mean_token_accuracy": 0.13978536278009415,
|
|
"num_tokens": 2947707.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.175567770004273,
|
|
"epoch": 1.2781456953642385,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004990777026552536,
|
|
"loss": 6.1013,
|
|
"mean_token_accuracy": 0.13597095981240273,
|
|
"num_tokens": 2959212.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.2820676326751705,
|
|
"epoch": 1.283239938869078,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004990410604134769,
|
|
"loss": 6.1171,
|
|
"mean_token_accuracy": 0.13908737003803254,
|
|
"num_tokens": 2969323.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.256225967407227,
|
|
"epoch": 1.2883341823739174,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004990037059496064,
|
|
"loss": 6.2441,
|
|
"mean_token_accuracy": 0.1359136037528515,
|
|
"num_tokens": 2981194.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.2075563907623295,
|
|
"epoch": 1.293428425878757,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004989656393823913,
|
|
"loss": 6.1429,
|
|
"mean_token_accuracy": 0.13628118112683296,
|
|
"num_tokens": 2992373.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.2464573860168455,
|
|
"epoch": 1.2985226693835965,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004989268608328443,
|
|
"loss": 6.1413,
|
|
"mean_token_accuracy": 0.13936529606580733,
|
|
"num_tokens": 3004019.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.257597303390503,
|
|
"epoch": 1.303616912888436,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004988873704242418,
|
|
"loss": 6.1344,
|
|
"mean_token_accuracy": 0.13469365164637565,
|
|
"num_tokens": 3016393.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.1419703483581545,
|
|
"epoch": 1.3087111563932756,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004988471682821229,
|
|
"loss": 6.094,
|
|
"mean_token_accuracy": 0.14097152650356293,
|
|
"num_tokens": 3027062.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.198255729675293,
|
|
"epoch": 1.3138053998981152,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004988062545342894,
|
|
"loss": 6.0993,
|
|
"mean_token_accuracy": 0.13867894187569618,
|
|
"num_tokens": 3039500.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.1662756443023685,
|
|
"epoch": 1.3188996434029547,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004987646293108056,
|
|
"loss": 6.0745,
|
|
"mean_token_accuracy": 0.14618232995271682,
|
|
"num_tokens": 3049296.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.226655530929565,
|
|
"epoch": 1.323993886907794,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004987222927439967,
|
|
"loss": 6.081,
|
|
"mean_token_accuracy": 0.1461057871580124,
|
|
"num_tokens": 3059806.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.330883550643921,
|
|
"epoch": 1.3290881304126336,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004986792449684504,
|
|
"loss": 6.2272,
|
|
"mean_token_accuracy": 0.13858062103390695,
|
|
"num_tokens": 3072769.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.121460962295532,
|
|
"epoch": 1.3341823739174732,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004986354861210144,
|
|
"loss": 6.1846,
|
|
"mean_token_accuracy": 0.13739702403545379,
|
|
"num_tokens": 3084547.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.328854990005493,
|
|
"epoch": 1.3392766174223127,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004985910163407973,
|
|
"loss": 6.1921,
|
|
"mean_token_accuracy": 0.135743211209774,
|
|
"num_tokens": 3095905.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.283027982711792,
|
|
"epoch": 1.3443708609271523,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004985458357691675,
|
|
"loss": 6.1809,
|
|
"mean_token_accuracy": 0.13586296737194062,
|
|
"num_tokens": 3106981.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.245007801055908,
|
|
"epoch": 1.3494651044319919,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004984999445497533,
|
|
"loss": 6.1544,
|
|
"mean_token_accuracy": 0.1399984285235405,
|
|
"num_tokens": 3117916.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.248331689834595,
|
|
"epoch": 1.3545593479368314,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000498453342828442,
|
|
"loss": 6.1771,
|
|
"mean_token_accuracy": 0.13796788901090623,
|
|
"num_tokens": 3129826.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.245303010940551,
|
|
"epoch": 1.359653591441671,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004984060307533794,
|
|
"loss": 6.2508,
|
|
"mean_token_accuracy": 0.13450490757822992,
|
|
"num_tokens": 3142543.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.177638387680053,
|
|
"epoch": 1.3647478349465105,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004983580084749698,
|
|
"loss": 6.1452,
|
|
"mean_token_accuracy": 0.14966057240962982,
|
|
"num_tokens": 3154192.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.176189279556274,
|
|
"epoch": 1.36984207845135,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000498309276145875,
|
|
"loss": 6.0717,
|
|
"mean_token_accuracy": 0.1437835745513439,
|
|
"num_tokens": 3166495.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.133072471618652,
|
|
"epoch": 1.3749363219561896,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004982598339210142,
|
|
"loss": 6.2128,
|
|
"mean_token_accuracy": 0.13311799317598344,
|
|
"num_tokens": 3177798.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.342384338378906,
|
|
"epoch": 1.380030565461029,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004982096819575631,
|
|
"loss": 6.1195,
|
|
"mean_token_accuracy": 0.14083297550678253,
|
|
"num_tokens": 3189625.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.084154319763184,
|
|
"epoch": 1.3851248089658685,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004981588204149538,
|
|
"loss": 6.1154,
|
|
"mean_token_accuracy": 0.14170270413160324,
|
|
"num_tokens": 3202247.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.206786727905273,
|
|
"epoch": 1.390219052470708,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004981072494548743,
|
|
"loss": 6.0851,
|
|
"mean_token_accuracy": 0.14271499440073968,
|
|
"num_tokens": 3214566.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.2440680980682375,
|
|
"epoch": 1.3953132959755477,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004980549692412676,
|
|
"loss": 6.2336,
|
|
"mean_token_accuracy": 0.13270484879612923,
|
|
"num_tokens": 3228692.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.225091123580933,
|
|
"epoch": 1.4004075394803872,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004980019799403313,
|
|
"loss": 6.186,
|
|
"mean_token_accuracy": 0.13683130368590354,
|
|
"num_tokens": 3240933.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.039432382583618,
|
|
"epoch": 1.4055017829852268,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004979482817205176,
|
|
"loss": 5.9728,
|
|
"mean_token_accuracy": 0.14688338935375214,
|
|
"num_tokens": 3252007.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.2535686016082765,
|
|
"epoch": 1.410596026490066,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004978938747525319,
|
|
"loss": 6.0697,
|
|
"mean_token_accuracy": 0.14190152511000634,
|
|
"num_tokens": 3262976.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.027432537078857,
|
|
"epoch": 1.4156902699949057,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004978387592093329,
|
|
"loss": 6.0479,
|
|
"mean_token_accuracy": 0.14016258120536804,
|
|
"num_tokens": 3274352.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.254658222198486,
|
|
"epoch": 1.4207845134997452,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004977829352661318,
|
|
"loss": 6.1054,
|
|
"mean_token_accuracy": 0.1417265459895134,
|
|
"num_tokens": 3287615.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.137201547622681,
|
|
"epoch": 1.4258787570045848,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004977264031003919,
|
|
"loss": 6.1241,
|
|
"mean_token_accuracy": 0.14106439501047136,
|
|
"num_tokens": 3300086.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.068885850906372,
|
|
"epoch": 1.4309730005094243,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004976691628918278,
|
|
"loss": 5.9912,
|
|
"mean_token_accuracy": 0.14750742614269258,
|
|
"num_tokens": 3310961.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.203730583190918,
|
|
"epoch": 1.4360672440142639,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000497611214822405,
|
|
"loss": 6.087,
|
|
"mean_token_accuracy": 0.1431759551167488,
|
|
"num_tokens": 3322077.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.143314933776855,
|
|
"epoch": 1.4411614875191034,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004975525590763393,
|
|
"loss": 6.1219,
|
|
"mean_token_accuracy": 0.14083698242902756,
|
|
"num_tokens": 3332725.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.216067600250244,
|
|
"epoch": 1.446255731023943,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000497493195840096,
|
|
"loss": 6.1492,
|
|
"mean_token_accuracy": 0.13391322121024132,
|
|
"num_tokens": 3344267.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.208370161056519,
|
|
"epoch": 1.4513499745287826,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004974331253023898,
|
|
"loss": 6.188,
|
|
"mean_token_accuracy": 0.14164203256368638,
|
|
"num_tokens": 3356947.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.214238500595092,
|
|
"epoch": 1.4564442180336221,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004973723476541836,
|
|
"loss": 6.0722,
|
|
"mean_token_accuracy": 0.14776080027222632,
|
|
"num_tokens": 3368934.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.135909748077393,
|
|
"epoch": 1.4615384615384617,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004973108630886885,
|
|
"loss": 6.2236,
|
|
"mean_token_accuracy": 0.1354803331196308,
|
|
"num_tokens": 3380806.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.254102373123169,
|
|
"epoch": 1.466632705043301,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004972486718013626,
|
|
"loss": 6.1562,
|
|
"mean_token_accuracy": 0.1398329481482506,
|
|
"num_tokens": 3393469.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.085903930664062,
|
|
"epoch": 1.4717269485481406,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004971857739899109,
|
|
"loss": 6.0091,
|
|
"mean_token_accuracy": 0.14970263689756394,
|
|
"num_tokens": 3405667.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.230799722671509,
|
|
"epoch": 1.4768211920529801,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004971221698542842,
|
|
"loss": 6.1771,
|
|
"mean_token_accuracy": 0.13638693392276763,
|
|
"num_tokens": 3417477.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.099519538879394,
|
|
"epoch": 1.4819154355578197,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004970578595966789,
|
|
"loss": 6.1185,
|
|
"mean_token_accuracy": 0.14716447964310647,
|
|
"num_tokens": 3429395.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.139642143249512,
|
|
"epoch": 1.4870096790626592,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004969928434215358,
|
|
"loss": 6.1276,
|
|
"mean_token_accuracy": 0.14418935626745225,
|
|
"num_tokens": 3441135.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.2753511428833,
|
|
"epoch": 1.4921039225674988,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004969271215355402,
|
|
"loss": 6.1673,
|
|
"mean_token_accuracy": 0.13607125505805015,
|
|
"num_tokens": 3453712.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.181782388687134,
|
|
"epoch": 1.4971981660723381,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004968606941476207,
|
|
"loss": 6.1342,
|
|
"mean_token_accuracy": 0.13905424997210503,
|
|
"num_tokens": 3466346.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.174712610244751,
|
|
"epoch": 1.5022924095771777,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004967935614689485,
|
|
"loss": 6.0985,
|
|
"mean_token_accuracy": 0.1439841754734516,
|
|
"num_tokens": 3478393.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.100687503814697,
|
|
"epoch": 1.5073866530820172,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004967257237129373,
|
|
"loss": 5.9825,
|
|
"mean_token_accuracy": 0.15523604303598404,
|
|
"num_tokens": 3489631.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.101812171936035,
|
|
"epoch": 1.5124808965868568,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004966571810952417,
|
|
"loss": 6.0101,
|
|
"mean_token_accuracy": 0.14085201621055604,
|
|
"num_tokens": 3502403.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.044900608062744,
|
|
"epoch": 1.5175751400916964,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004965879338337577,
|
|
"loss": 5.9947,
|
|
"mean_token_accuracy": 0.15054818540811538,
|
|
"num_tokens": 3513512.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.170858907699585,
|
|
"epoch": 1.522669383596536,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004965179821486205,
|
|
"loss": 6.1728,
|
|
"mean_token_accuracy": 0.13760660737752914,
|
|
"num_tokens": 3525670.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.141346740722656,
|
|
"epoch": 1.5277636271013755,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004964473262622054,
|
|
"loss": 6.0793,
|
|
"mean_token_accuracy": 0.1404211476445198,
|
|
"num_tokens": 3537634.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.092041826248169,
|
|
"epoch": 1.532857870606215,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000496375966399126,
|
|
"loss": 6.0136,
|
|
"mean_token_accuracy": 0.1455122373998165,
|
|
"num_tokens": 3549344.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.159336233139038,
|
|
"epoch": 1.5379521141110546,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000496303902786234,
|
|
"loss": 6.1607,
|
|
"mean_token_accuracy": 0.14189588725566865,
|
|
"num_tokens": 3563193.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.0729227542877195,
|
|
"epoch": 1.5430463576158941,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004962311356526181,
|
|
"loss": 6.0882,
|
|
"mean_token_accuracy": 0.14001245200634002,
|
|
"num_tokens": 3574304.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.226966428756714,
|
|
"epoch": 1.5481406011207337,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004961576652296036,
|
|
"loss": 5.9942,
|
|
"mean_token_accuracy": 0.14970892891287804,
|
|
"num_tokens": 3586396.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.996706819534301,
|
|
"epoch": 1.5532348446255733,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004960834917507516,
|
|
"loss": 5.986,
|
|
"mean_token_accuracy": 0.1510669320821762,
|
|
"num_tokens": 3597945.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.161891603469849,
|
|
"epoch": 1.5583290881304126,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004960086154518578,
|
|
"loss": 6.0987,
|
|
"mean_token_accuracy": 0.14722763895988464,
|
|
"num_tokens": 3610378.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.076696491241455,
|
|
"epoch": 1.5634233316352522,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004959330365709528,
|
|
"loss": 6.0258,
|
|
"mean_token_accuracy": 0.1500461980700493,
|
|
"num_tokens": 3621768.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.105367517471313,
|
|
"epoch": 1.5685175751400917,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004958567553483003,
|
|
"loss": 6.0417,
|
|
"mean_token_accuracy": 0.14288988560438157,
|
|
"num_tokens": 3634473.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.024158382415772,
|
|
"epoch": 1.5736118186449313,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004957797720263967,
|
|
"loss": 5.9167,
|
|
"mean_token_accuracy": 0.1528099149465561,
|
|
"num_tokens": 3644499.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.096716403961182,
|
|
"epoch": 1.5787060621497706,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004957020868499706,
|
|
"loss": 6.0055,
|
|
"mean_token_accuracy": 0.14677556455135346,
|
|
"num_tokens": 3655807.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.121523046493531,
|
|
"epoch": 1.5838003056546102,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004956237000659815,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.1559087172150612,
|
|
"num_tokens": 3667301.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.960540866851806,
|
|
"epoch": 1.5888945491594497,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004955446119236195,
|
|
"loss": 6.0787,
|
|
"mean_token_accuracy": 0.13644454255700111,
|
|
"num_tokens": 3679448.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.185535717010498,
|
|
"epoch": 1.5939887926642893,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004954648226743043,
|
|
"loss": 6.0217,
|
|
"mean_token_accuracy": 0.15028504878282548,
|
|
"num_tokens": 3690985.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.071138954162597,
|
|
"epoch": 1.5990830361691288,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004953843325716841,
|
|
"loss": 6.0883,
|
|
"mean_token_accuracy": 0.14665026739239692,
|
|
"num_tokens": 3703677.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.187681531906128,
|
|
"epoch": 1.6041772796739684,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004953031418716356,
|
|
"loss": 6.0156,
|
|
"mean_token_accuracy": 0.15277430415153503,
|
|
"num_tokens": 3715429.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.00174241065979,
|
|
"epoch": 1.609271523178808,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004952212508322623,
|
|
"loss": 6.0722,
|
|
"mean_token_accuracy": 0.14784569218754767,
|
|
"num_tokens": 3727663.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.087293577194214,
|
|
"epoch": 1.6143657666836475,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004951386597138943,
|
|
"loss": 5.9923,
|
|
"mean_token_accuracy": 0.14931108802556992,
|
|
"num_tokens": 3739585.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.109324932098389,
|
|
"epoch": 1.619460010188487,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004950553687790871,
|
|
"loss": 6.1311,
|
|
"mean_token_accuracy": 0.14180412590503694,
|
|
"num_tokens": 3751382.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.093611574172973,
|
|
"epoch": 1.6245542536933266,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004949713782926208,
|
|
"loss": 6.0117,
|
|
"mean_token_accuracy": 0.15391514897346498,
|
|
"num_tokens": 3762520.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.0834370136260985,
|
|
"epoch": 1.6296484971981662,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004948866885214996,
|
|
"loss": 5.9934,
|
|
"mean_token_accuracy": 0.15164291858673096,
|
|
"num_tokens": 3773821.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.13722333908081,
|
|
"epoch": 1.6347427407030057,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004948012997349506,
|
|
"loss": 6.0664,
|
|
"mean_token_accuracy": 0.1453387849032879,
|
|
"num_tokens": 3785767.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.08644847869873,
|
|
"epoch": 1.6398369842078453,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004947152122044232,
|
|
"loss": 6.0955,
|
|
"mean_token_accuracy": 0.14098975509405137,
|
|
"num_tokens": 3796702.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.107085466384888,
|
|
"epoch": 1.6449312277126846,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004946284262035875,
|
|
"loss": 5.9966,
|
|
"mean_token_accuracy": 0.14310370981693268,
|
|
"num_tokens": 3808757.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.019505548477173,
|
|
"epoch": 1.6500254712175242,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000494540942008335,
|
|
"loss": 5.9833,
|
|
"mean_token_accuracy": 0.14864621683955193,
|
|
"num_tokens": 3819804.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.1659170627594,
|
|
"epoch": 1.6551197147223637,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000494452759896776,
|
|
"loss": 6.0245,
|
|
"mean_token_accuracy": 0.14918728172779083,
|
|
"num_tokens": 3831877.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.029467582702637,
|
|
"epoch": 1.6602139582272033,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004943638801492396,
|
|
"loss": 5.9882,
|
|
"mean_token_accuracy": 0.14464981853961945,
|
|
"num_tokens": 3843356.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.087781381607056,
|
|
"epoch": 1.6653082017320426,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004942743030482728,
|
|
"loss": 6.0172,
|
|
"mean_token_accuracy": 0.14597421437501906,
|
|
"num_tokens": 3854779.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.019084644317627,
|
|
"epoch": 1.6704024452368822,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004941840288786397,
|
|
"loss": 5.9706,
|
|
"mean_token_accuracy": 0.15264465063810348,
|
|
"num_tokens": 3867309.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.110167121887207,
|
|
"epoch": 1.6754966887417218,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004940930579273198,
|
|
"loss": 6.0627,
|
|
"mean_token_accuracy": 0.14825674444437026,
|
|
"num_tokens": 3879341.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.112421607971191,
|
|
"epoch": 1.6805909322465613,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004940013904835079,
|
|
"loss": 5.9514,
|
|
"mean_token_accuracy": 0.15374316275119781,
|
|
"num_tokens": 3890639.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.03355040550232,
|
|
"epoch": 1.6856851757514009,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004939090268386132,
|
|
"loss": 6.0057,
|
|
"mean_token_accuracy": 0.1441144548356533,
|
|
"num_tokens": 3904185.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.034610319137573,
|
|
"epoch": 1.6907794192562404,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004938159672862578,
|
|
"loss": 5.9991,
|
|
"mean_token_accuracy": 0.14242151454091073,
|
|
"num_tokens": 3917137.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.073326921463012,
|
|
"epoch": 1.69587366276108,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004937222121222763,
|
|
"loss": 6.0352,
|
|
"mean_token_accuracy": 0.14397113993763924,
|
|
"num_tokens": 3929911.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.113371801376343,
|
|
"epoch": 1.7009679062659195,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004936277616447144,
|
|
"loss": 6.0705,
|
|
"mean_token_accuracy": 0.13673104122281074,
|
|
"num_tokens": 3941242.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.121015882492065,
|
|
"epoch": 1.706062149770759,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004935326161538281,
|
|
"loss": 6.032,
|
|
"mean_token_accuracy": 0.14769989401102065,
|
|
"num_tokens": 3952464.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.96496729850769,
|
|
"epoch": 1.7111563932755987,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004934367759520834,
|
|
"loss": 5.9828,
|
|
"mean_token_accuracy": 0.14930499494075775,
|
|
"num_tokens": 3965479.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.063432741165161,
|
|
"epoch": 1.7162506367804382,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004933402413441541,
|
|
"loss": 5.9945,
|
|
"mean_token_accuracy": 0.1519572079181671,
|
|
"num_tokens": 3978195.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.137292766571045,
|
|
"epoch": 1.7213448802852778,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004932430126369221,
|
|
"loss": 6.1009,
|
|
"mean_token_accuracy": 0.1380507230758667,
|
|
"num_tokens": 3989373.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.998342800140381,
|
|
"epoch": 1.7264391237901173,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004931450901394753,
|
|
"loss": 6.0126,
|
|
"mean_token_accuracy": 0.14947347193956376,
|
|
"num_tokens": 4000860.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.049466371536255,
|
|
"epoch": 1.7315333672949567,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004930464741631073,
|
|
"loss": 5.9871,
|
|
"mean_token_accuracy": 0.14549367874860764,
|
|
"num_tokens": 4012035.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.02332911491394,
|
|
"epoch": 1.7366276107997962,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004929471650213167,
|
|
"loss": 5.9824,
|
|
"mean_token_accuracy": 0.14759661108255387,
|
|
"num_tokens": 4024490.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.1991688251495365,
|
|
"epoch": 1.7417218543046358,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004928471630298051,
|
|
"loss": 6.0864,
|
|
"mean_token_accuracy": 0.14478200078010559,
|
|
"num_tokens": 4036277.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.91363377571106,
|
|
"epoch": 1.7468160978094753,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004927464685064767,
|
|
"loss": 5.8765,
|
|
"mean_token_accuracy": 0.15802925825119019,
|
|
"num_tokens": 4047734.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.916611814498902,
|
|
"epoch": 1.7519103413143147,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004926450817714376,
|
|
"loss": 5.89,
|
|
"mean_token_accuracy": 0.15395067781209945,
|
|
"num_tokens": 4059037.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.992276859283447,
|
|
"epoch": 1.7570045848191542,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004925430031469941,
|
|
"loss": 5.9661,
|
|
"mean_token_accuracy": 0.14739770889282228,
|
|
"num_tokens": 4070795.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.04740948677063,
|
|
"epoch": 1.7620988283239938,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004924402329576522,
|
|
"loss": 5.815,
|
|
"mean_token_accuracy": 0.16750310882925987,
|
|
"num_tokens": 4082661.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.938603115081787,
|
|
"epoch": 1.7671930718288333,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004923367715301164,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.15013253837823867,
|
|
"num_tokens": 4093151.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.009731960296631,
|
|
"epoch": 1.772287315333673,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004922326191932882,
|
|
"loss": 6.0054,
|
|
"mean_token_accuracy": 0.14514800533652306,
|
|
"num_tokens": 4104380.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.031725835800171,
|
|
"epoch": 1.7773815588385125,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004921277762782661,
|
|
"loss": 5.9102,
|
|
"mean_token_accuracy": 0.1545369416475296,
|
|
"num_tokens": 4115506.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.058172130584717,
|
|
"epoch": 1.782475802343352,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004920222431183437,
|
|
"loss": 6.0688,
|
|
"mean_token_accuracy": 0.14753634482622147,
|
|
"num_tokens": 4126978.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.964846801757813,
|
|
"epoch": 1.7875700458481916,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004919160200490085,
|
|
"loss": 5.8791,
|
|
"mean_token_accuracy": 0.15507214069366454,
|
|
"num_tokens": 4137602.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.084032964706421,
|
|
"epoch": 1.7926642893530311,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004918091074079419,
|
|
"loss": 5.9777,
|
|
"mean_token_accuracy": 0.14689225256443023,
|
|
"num_tokens": 4148617.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.065905427932739,
|
|
"epoch": 1.7977585328578707,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004917015055350167,
|
|
"loss": 6.0774,
|
|
"mean_token_accuracy": 0.15037426650524138,
|
|
"num_tokens": 4162445.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.121287441253662,
|
|
"epoch": 1.8028527763627102,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004915932147722974,
|
|
"loss": 6.0552,
|
|
"mean_token_accuracy": 0.14733544141054153,
|
|
"num_tokens": 4175163.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.065502786636353,
|
|
"epoch": 1.8079470198675498,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004914842354640382,
|
|
"loss": 6.024,
|
|
"mean_token_accuracy": 0.14942564815282822,
|
|
"num_tokens": 4189159.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.119499826431275,
|
|
"epoch": 1.8130412633723894,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000491374567956682,
|
|
"loss": 6.1319,
|
|
"mean_token_accuracy": 0.1392667606472969,
|
|
"num_tokens": 4201989.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.982971239089966,
|
|
"epoch": 1.8181355068772287,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004912642125988597,
|
|
"loss": 5.9005,
|
|
"mean_token_accuracy": 0.15367604941129684,
|
|
"num_tokens": 4214982.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.979884243011474,
|
|
"epoch": 1.8232297503820682,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004911531697413888,
|
|
"loss": 5.8197,
|
|
"mean_token_accuracy": 0.15773751437664033,
|
|
"num_tokens": 4225932.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.0400148868560795,
|
|
"epoch": 1.8283239938869078,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004910414397372724,
|
|
"loss": 6.0547,
|
|
"mean_token_accuracy": 0.14175843968987464,
|
|
"num_tokens": 4238406.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.933195638656616,
|
|
"epoch": 1.8334182373917474,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004909290229416979,
|
|
"loss": 5.8468,
|
|
"mean_token_accuracy": 0.1593022346496582,
|
|
"num_tokens": 4249956.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.020194244384766,
|
|
"epoch": 1.8385124808965867,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000490815919712036,
|
|
"loss": 5.9966,
|
|
"mean_token_accuracy": 0.1435801424086094,
|
|
"num_tokens": 4262367.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.0135705947875975,
|
|
"epoch": 1.8436067244014263,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004907021304078398,
|
|
"loss": 5.9858,
|
|
"mean_token_accuracy": 0.15024446249008178,
|
|
"num_tokens": 4273767.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.017576789855957,
|
|
"epoch": 1.8487009679062658,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004905876553908432,
|
|
"loss": 6.0099,
|
|
"mean_token_accuracy": 0.1527609795331955,
|
|
"num_tokens": 4285545.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.131746530532837,
|
|
"epoch": 1.8537952114111054,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004904724950249599,
|
|
"loss": 6.0007,
|
|
"mean_token_accuracy": 0.14292944744229316,
|
|
"num_tokens": 4296963.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.938863706588745,
|
|
"epoch": 1.858889454915945,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004903566496762824,
|
|
"loss": 5.8591,
|
|
"mean_token_accuracy": 0.1559537261724472,
|
|
"num_tokens": 4307724.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.979634141921997,
|
|
"epoch": 1.8639836984207845,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000490240119713081,
|
|
"loss": 5.9177,
|
|
"mean_token_accuracy": 0.15530329197645187,
|
|
"num_tokens": 4319153.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.925511264801026,
|
|
"epoch": 1.869077941925624,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004901229055058019,
|
|
"loss": 5.8893,
|
|
"mean_token_accuracy": 0.15859820544719697,
|
|
"num_tokens": 4330179.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.000934362411499,
|
|
"epoch": 1.8741721854304636,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004900050074270669,
|
|
"loss": 5.9061,
|
|
"mean_token_accuracy": 0.14837884306907653,
|
|
"num_tokens": 4340747.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.986407136917114,
|
|
"epoch": 1.8792664289353032,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004898864258516717,
|
|
"loss": 5.9356,
|
|
"mean_token_accuracy": 0.1507401555776596,
|
|
"num_tokens": 4352336.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.019035959243775,
|
|
"epoch": 1.8843606724401427,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004897671611565845,
|
|
"loss": 5.8585,
|
|
"mean_token_accuracy": 0.15450628101825714,
|
|
"num_tokens": 4363189.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.9605508804321286,
|
|
"epoch": 1.8894549159449823,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004896472137209457,
|
|
"loss": 6.0114,
|
|
"mean_token_accuracy": 0.15344889163970948,
|
|
"num_tokens": 4375946.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.0401444911956785,
|
|
"epoch": 1.8945491594498218,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004895265839260657,
|
|
"loss": 5.9645,
|
|
"mean_token_accuracy": 0.155985938757658,
|
|
"num_tokens": 4387543.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.9809211730957035,
|
|
"epoch": 1.8996434029546614,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004894052721554242,
|
|
"loss": 5.9741,
|
|
"mean_token_accuracy": 0.15067811906337739,
|
|
"num_tokens": 4399790.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.104261207580566,
|
|
"epoch": 1.9047376464595007,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004892832787946691,
|
|
"loss": 5.9982,
|
|
"mean_token_accuracy": 0.1496062770485878,
|
|
"num_tokens": 4411321.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.987740612030029,
|
|
"epoch": 1.9098318899643403,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004891606042316146,
|
|
"loss": 5.9341,
|
|
"mean_token_accuracy": 0.1538042575120926,
|
|
"num_tokens": 4422712.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.975160217285156,
|
|
"epoch": 1.9149261334691798,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004890372488562408,
|
|
"loss": 5.932,
|
|
"mean_token_accuracy": 0.14831314086914063,
|
|
"num_tokens": 4434941.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.033766174316407,
|
|
"epoch": 1.9200203769740194,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004889132130606921,
|
|
"loss": 5.9845,
|
|
"mean_token_accuracy": 0.15482763946056366,
|
|
"num_tokens": 4447440.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.95661187171936,
|
|
"epoch": 1.9251146204788587,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004887884972392758,
|
|
"loss": 5.9309,
|
|
"mean_token_accuracy": 0.15109919011592865,
|
|
"num_tokens": 4459086.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.010490274429321,
|
|
"epoch": 1.9302088639836983,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000488663101788461,
|
|
"loss": 5.9689,
|
|
"mean_token_accuracy": 0.15499186366796494,
|
|
"num_tokens": 4470463.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.906804370880127,
|
|
"epoch": 1.9353031074885378,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004885370271068773,
|
|
"loss": 5.9099,
|
|
"mean_token_accuracy": 0.15397028774023055,
|
|
"num_tokens": 4481266.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.9154823303222654,
|
|
"epoch": 1.9403973509933774,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004884102735953136,
|
|
"loss": 5.8761,
|
|
"mean_token_accuracy": 0.15611894503235818,
|
|
"num_tokens": 4493367.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.011291933059693,
|
|
"epoch": 1.945491594498217,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000488282841656717,
|
|
"loss": 5.926,
|
|
"mean_token_accuracy": 0.1512049287557602,
|
|
"num_tokens": 4505317.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.954664707183838,
|
|
"epoch": 1.9505858380030565,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048815473169619097,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.15357810258865356,
|
|
"num_tokens": 4516743.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.008725929260254,
|
|
"epoch": 1.955680081507896,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004880259441209945,
|
|
"loss": 5.9257,
|
|
"mean_token_accuracy": 0.15126045495271684,
|
|
"num_tokens": 4529467.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.895899724960327,
|
|
"epoch": 1.9607743250127356,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004878964793405408,
|
|
"loss": 5.8362,
|
|
"mean_token_accuracy": 0.16100727915763854,
|
|
"num_tokens": 4540648.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.9156656742095945,
|
|
"epoch": 1.9658685685175752,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048776633776639584,
|
|
"loss": 5.8651,
|
|
"mean_token_accuracy": 0.15492998957633972,
|
|
"num_tokens": 4552095.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.9102991104125975,
|
|
"epoch": 1.9709628120224147,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048763551981227703,
|
|
"loss": 5.8826,
|
|
"mean_token_accuracy": 0.15725963562726974,
|
|
"num_tokens": 4563286.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.079156112670899,
|
|
"epoch": 1.9760570555272543,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048750402589405224,
|
|
"loss": 5.9678,
|
|
"mean_token_accuracy": 0.1529908835887909,
|
|
"num_tokens": 4575313.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.919543361663818,
|
|
"epoch": 1.9811512990320939,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048737185642973784,
|
|
"loss": 5.9132,
|
|
"mean_token_accuracy": 0.15463192462921144,
|
|
"num_tokens": 4587155.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.0006670475006105,
|
|
"epoch": 1.9862455425369334,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000487239011839498,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.14793393164873123,
|
|
"num_tokens": 4598176.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.0277361392974855,
|
|
"epoch": 1.9913397860417728,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004871054925456431,
|
|
"loss": 5.9973,
|
|
"mean_token_accuracy": 0.1456962801516056,
|
|
"num_tokens": 4610004.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.978898000717163,
|
|
"epoch": 1.9964340295466123,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048697129897262833,
|
|
"loss": 5.915,
|
|
"mean_token_accuracy": 0.15651439279317855,
|
|
"num_tokens": 4620985.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.908947255876329,
|
|
"epoch": 2.001018848700968,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004868364315470523,
|
|
"loss": 5.7981,
|
|
"mean_token_accuracy": 0.16079486078686184,
|
|
"num_tokens": 4630550.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.945104932785034,
|
|
"epoch": 2.0061130922058075,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048670089069765594,
|
|
"loss": 5.5484,
|
|
"mean_token_accuracy": 0.16541721671819687,
|
|
"num_tokens": 4642173.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.870164632797241,
|
|
"epoch": 2.011207335710647,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004865646768553209,
|
|
"loss": 5.5773,
|
|
"mean_token_accuracy": 0.1712649866938591,
|
|
"num_tokens": 4654322.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.85951795578003,
|
|
"epoch": 2.0163015792154866,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004864277904530684,
|
|
"loss": 5.5786,
|
|
"mean_token_accuracy": 0.17355846911668776,
|
|
"num_tokens": 4665716.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.918148946762085,
|
|
"epoch": 2.021395822720326,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048629023192605747,
|
|
"loss": 5.6547,
|
|
"mean_token_accuracy": 0.16473341286182402,
|
|
"num_tokens": 4677148.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.90467095375061,
|
|
"epoch": 2.0264900662251657,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048615200171158395,
|
|
"loss": 5.5792,
|
|
"mean_token_accuracy": 0.16939278990030288,
|
|
"num_tokens": 4690855.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.819876480102539,
|
|
"epoch": 2.0315843097300053,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048601310024907904,
|
|
"loss": 5.5502,
|
|
"mean_token_accuracy": 0.16651132479310035,
|
|
"num_tokens": 4702933.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.930644941329956,
|
|
"epoch": 2.036678553234845,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048587352798010766,
|
|
"loss": 5.6341,
|
|
"mean_token_accuracy": 0.15964097529649734,
|
|
"num_tokens": 4715584.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.036678553234845,
|
|
"eval_entropy": 5.62031678447017,
|
|
"eval_loss": 5.969032287597656,
|
|
"eval_mean_token_accuracy": 0.15611327899458968,
|
|
"eval_num_tokens": 4715584.0,
|
|
"eval_runtime": 2.8116,
|
|
"eval_samples_per_second": 1152.365,
|
|
"eval_steps_per_second": 144.046,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.818984031677246,
|
|
"epoch": 2.0417727967396844,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048573328534836735,
|
|
"loss": 5.601,
|
|
"mean_token_accuracy": 0.1660971850156784,
|
|
"num_tokens": 4726197.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.900150442123413,
|
|
"epoch": 2.0468670402445235,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004855923727996866,
|
|
"loss": 5.5078,
|
|
"mean_token_accuracy": 0.17555468231439592,
|
|
"num_tokens": 4739238.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.8321984767913815,
|
|
"epoch": 2.051961283749363,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004854507907820237,
|
|
"loss": 5.5534,
|
|
"mean_token_accuracy": 0.1598127529025078,
|
|
"num_tokens": 4750658.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.900466394424439,
|
|
"epoch": 2.0570555272542026,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048530853974546495,
|
|
"loss": 5.6333,
|
|
"mean_token_accuracy": 0.1609639659523964,
|
|
"num_tokens": 4763193.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.955789375305176,
|
|
"epoch": 2.062149770759042,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004851656201422237,
|
|
"loss": 5.6704,
|
|
"mean_token_accuracy": 0.162435482442379,
|
|
"num_tokens": 4774701.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.880024576187134,
|
|
"epoch": 2.0672440142638817,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048502203242663846,
|
|
"loss": 5.6025,
|
|
"mean_token_accuracy": 0.16102872639894486,
|
|
"num_tokens": 4786001.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.75847373008728,
|
|
"epoch": 2.0723382577687213,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048487777705517177,
|
|
"loss": 5.5833,
|
|
"mean_token_accuracy": 0.16628326326608658,
|
|
"num_tokens": 4797591.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.928161478042602,
|
|
"epoch": 2.077432501273561,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048473285448640853,
|
|
"loss": 5.6475,
|
|
"mean_token_accuracy": 0.16663917750120164,
|
|
"num_tokens": 4809296.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.871248340606689,
|
|
"epoch": 2.0825267447784004,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004845872651810549,
|
|
"loss": 5.6275,
|
|
"mean_token_accuracy": 0.16441605389118194,
|
|
"num_tokens": 4821141.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.752266454696655,
|
|
"epoch": 2.08762098828324,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048444100960193617,
|
|
"loss": 5.4742,
|
|
"mean_token_accuracy": 0.17370279729366303,
|
|
"num_tokens": 4834043.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.828112459182739,
|
|
"epoch": 2.0927152317880795,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004842940882139961,
|
|
"loss": 5.5404,
|
|
"mean_token_accuracy": 0.16728458404541016,
|
|
"num_tokens": 4845686.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.815605974197387,
|
|
"epoch": 2.097809475292919,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048414650148429465,
|
|
"loss": 5.5796,
|
|
"mean_token_accuracy": 0.1687358096241951,
|
|
"num_tokens": 4857999.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.775838613510132,
|
|
"epoch": 2.1029037187977586,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048399824988200747,
|
|
"loss": 5.5023,
|
|
"mean_token_accuracy": 0.1788402631878853,
|
|
"num_tokens": 4868955.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.832354497909546,
|
|
"epoch": 2.107997962302598,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048384933387842313,
|
|
"loss": 5.5765,
|
|
"mean_token_accuracy": 0.16253895610570906,
|
|
"num_tokens": 4881283.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.747897052764893,
|
|
"epoch": 2.1130922058074377,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004836997539469429,
|
|
"loss": 5.5665,
|
|
"mean_token_accuracy": 0.16654192507267,
|
|
"num_tokens": 4892841.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.843785619735717,
|
|
"epoch": 2.1181864493122773,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048354951056307814,
|
|
"loss": 5.6753,
|
|
"mean_token_accuracy": 0.1653669908642769,
|
|
"num_tokens": 4903926.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.847943210601807,
|
|
"epoch": 2.123280692817117,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048339860420444994,
|
|
"loss": 5.5915,
|
|
"mean_token_accuracy": 0.16666646897792817,
|
|
"num_tokens": 4917163.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.8208503246307375,
|
|
"epoch": 2.128374936321956,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048324703535078633,
|
|
"loss": 5.6041,
|
|
"mean_token_accuracy": 0.1606005996465683,
|
|
"num_tokens": 4930232.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.80724024772644,
|
|
"epoch": 2.1334691798267955,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048309480448392185,
|
|
"loss": 5.578,
|
|
"mean_token_accuracy": 0.17210218608379363,
|
|
"num_tokens": 4941493.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.725574779510498,
|
|
"epoch": 2.138563423331635,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004829419120877954,
|
|
"loss": 5.5223,
|
|
"mean_token_accuracy": 0.17113063484430313,
|
|
"num_tokens": 4953781.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.865182256698608,
|
|
"epoch": 2.1436576668364746,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004827883586484488,
|
|
"loss": 5.6228,
|
|
"mean_token_accuracy": 0.16533854901790618,
|
|
"num_tokens": 4966265.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.8052679061889645,
|
|
"epoch": 2.148751910341314,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004826341446540254,
|
|
"loss": 5.639,
|
|
"mean_token_accuracy": 0.16439810246229172,
|
|
"num_tokens": 4977676.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.874921894073486,
|
|
"epoch": 2.1538461538461537,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048247927059476866,
|
|
"loss": 5.6343,
|
|
"mean_token_accuracy": 0.16010402143001556,
|
|
"num_tokens": 4990818.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.776369380950928,
|
|
"epoch": 2.1589403973509933,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048232373696301985,
|
|
"loss": 5.6298,
|
|
"mean_token_accuracy": 0.1646338015794754,
|
|
"num_tokens": 5003175.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.88043475151062,
|
|
"epoch": 2.164034640855833,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048216754425321734,
|
|
"loss": 5.591,
|
|
"mean_token_accuracy": 0.1656954288482666,
|
|
"num_tokens": 5014619.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.816552972793579,
|
|
"epoch": 2.1691288843606724,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004820106929618948,
|
|
"loss": 5.6167,
|
|
"mean_token_accuracy": 0.16424932479858398,
|
|
"num_tokens": 5026407.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.906729984283447,
|
|
"epoch": 2.174223127865512,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048185318358767926,
|
|
"loss": 5.7523,
|
|
"mean_token_accuracy": 0.15930672287940978,
|
|
"num_tokens": 5039479.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.767476320266724,
|
|
"epoch": 2.1793173713703515,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048169501663129007,
|
|
"loss": 5.5966,
|
|
"mean_token_accuracy": 0.16930720806121827,
|
|
"num_tokens": 5052074.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.996895408630371,
|
|
"epoch": 2.184411614875191,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048153619259553665,
|
|
"loss": 5.6824,
|
|
"mean_token_accuracy": 0.15677667707204818,
|
|
"num_tokens": 5064551.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.810957574844361,
|
|
"epoch": 2.1895058583800306,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048137671198531765,
|
|
"loss": 5.5815,
|
|
"mean_token_accuracy": 0.1658017724752426,
|
|
"num_tokens": 5075616.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.760199737548828,
|
|
"epoch": 2.19460010188487,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048121657530761857,
|
|
"loss": 5.6552,
|
|
"mean_token_accuracy": 0.16322747766971588,
|
|
"num_tokens": 5086531.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.922705745697021,
|
|
"epoch": 2.1996943453897098,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048105578307151094,
|
|
"loss": 5.6916,
|
|
"mean_token_accuracy": 0.16097672730684282,
|
|
"num_tokens": 5097881.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.7204893112182615,
|
|
"epoch": 2.2047885888945493,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048089433578814994,
|
|
"loss": 5.509,
|
|
"mean_token_accuracy": 0.16698236912488937,
|
|
"num_tokens": 5108511.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.797280120849609,
|
|
"epoch": 2.209882832399389,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004807322339707733,
|
|
"loss": 5.6006,
|
|
"mean_token_accuracy": 0.17074504047632216,
|
|
"num_tokens": 5119302.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.7944495677948,
|
|
"epoch": 2.214977075904228,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048056947813469967,
|
|
"loss": 5.6876,
|
|
"mean_token_accuracy": 0.15859579518437386,
|
|
"num_tokens": 5131430.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.811315679550171,
|
|
"epoch": 2.220071319409068,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004804060687973264,
|
|
"loss": 5.6467,
|
|
"mean_token_accuracy": 0.1624533861875534,
|
|
"num_tokens": 5143449.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.841419410705567,
|
|
"epoch": 2.225165562913907,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004802420064781286,
|
|
"loss": 5.6163,
|
|
"mean_token_accuracy": 0.16900259554386138,
|
|
"num_tokens": 5155286.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.791315984725952,
|
|
"epoch": 2.2302598064187467,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004800772916986572,
|
|
"loss": 5.5697,
|
|
"mean_token_accuracy": 0.16355349421501159,
|
|
"num_tokens": 5167624.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.791861534118652,
|
|
"epoch": 2.235354049923586,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047991192498253715,
|
|
"loss": 5.5139,
|
|
"mean_token_accuracy": 0.17494232207536697,
|
|
"num_tokens": 5179206.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.741910982131958,
|
|
"epoch": 2.2404482934284258,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000479745906855466,
|
|
"loss": 5.5742,
|
|
"mean_token_accuracy": 0.16768807172775269,
|
|
"num_tokens": 5190820.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.839303588867187,
|
|
"epoch": 2.2455425369332653,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047957923784521206,
|
|
"loss": 5.6235,
|
|
"mean_token_accuracy": 0.16996713876724243,
|
|
"num_tokens": 5201308.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.734570646286011,
|
|
"epoch": 2.250636780438105,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004794119184816129,
|
|
"loss": 5.5304,
|
|
"mean_token_accuracy": 0.1717229664325714,
|
|
"num_tokens": 5213246.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.81427035331726,
|
|
"epoch": 2.2557310239429444,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004792439492965734,
|
|
"loss": 5.6367,
|
|
"mean_token_accuracy": 0.1608924999833107,
|
|
"num_tokens": 5225484.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.810764932632447,
|
|
"epoch": 2.260825267447784,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004790753308240642,
|
|
"loss": 5.6575,
|
|
"mean_token_accuracy": 0.1606771320104599,
|
|
"num_tokens": 5238728.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.8196601390838625,
|
|
"epoch": 2.2659195109526236,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004789060636001202,
|
|
"loss": 5.6638,
|
|
"mean_token_accuracy": 0.166461843252182,
|
|
"num_tokens": 5250130.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.808871412277222,
|
|
"epoch": 2.271013754457463,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004787361481628385,
|
|
"loss": 5.6044,
|
|
"mean_token_accuracy": 0.16343416571617125,
|
|
"num_tokens": 5262624.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.814700746536255,
|
|
"epoch": 2.2761079979623027,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004785655850523769,
|
|
"loss": 5.6465,
|
|
"mean_token_accuracy": 0.1613625168800354,
|
|
"num_tokens": 5274914.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.738488388061524,
|
|
"epoch": 2.2812022414671422,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047839437481095223,
|
|
"loss": 5.5063,
|
|
"mean_token_accuracy": 0.17593536227941514,
|
|
"num_tokens": 5286836.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.6917564868927,
|
|
"epoch": 2.286296484971982,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047822251798283836,
|
|
"loss": 5.522,
|
|
"mean_token_accuracy": 0.17000024020671844,
|
|
"num_tokens": 5297416.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.722369194030762,
|
|
"epoch": 2.2913907284768213,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004780500151143647,
|
|
"loss": 5.5034,
|
|
"mean_token_accuracy": 0.17222748398780824,
|
|
"num_tokens": 5307852.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.723336887359619,
|
|
"epoch": 2.296484971981661,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004778768667539147,
|
|
"loss": 5.6364,
|
|
"mean_token_accuracy": 0.16439990997314452,
|
|
"num_tokens": 5320500.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.797302150726319,
|
|
"epoch": 2.3015792154865,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047770307345192326,
|
|
"loss": 5.5672,
|
|
"mean_token_accuracy": 0.16709809899330139,
|
|
"num_tokens": 5331867.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.750816011428833,
|
|
"epoch": 2.30667345899134,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047752863576087603,
|
|
"loss": 5.5616,
|
|
"mean_token_accuracy": 0.1664021909236908,
|
|
"num_tokens": 5344181.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.7929909229278564,
|
|
"epoch": 2.311767702496179,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000477353554235307,
|
|
"loss": 5.7328,
|
|
"mean_token_accuracy": 0.16328889578580857,
|
|
"num_tokens": 5356068.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.8103516578674315,
|
|
"epoch": 2.3168619460010187,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047717782943179664,
|
|
"loss": 5.5684,
|
|
"mean_token_accuracy": 0.1686837702989578,
|
|
"num_tokens": 5367456.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.756494569778442,
|
|
"epoch": 2.3219561895058582,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004770014619089709,
|
|
"loss": 5.6092,
|
|
"mean_token_accuracy": 0.16871432214975357,
|
|
"num_tokens": 5379731.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.756763887405396,
|
|
"epoch": 2.327050433010698,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004768244522274985,
|
|
"loss": 5.5532,
|
|
"mean_token_accuracy": 0.16949952095746995,
|
|
"num_tokens": 5391809.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.724135637283325,
|
|
"epoch": 2.3321446765155374,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00047664680095008973,
|
|
"loss": 5.624,
|
|
"mean_token_accuracy": 0.16763131618499755,
|
|
"num_tokens": 5404341.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.779242038726807,
|
|
"epoch": 2.337238920020377,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004764685086414945,
|
|
"loss": 5.6168,
|
|
"mean_token_accuracy": 0.16189232468605042,
|
|
"num_tokens": 5415584.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.798492765426635,
|
|
"epoch": 2.3423331635252165,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047628957586850053,
|
|
"loss": 5.6579,
|
|
"mean_token_accuracy": 0.16597485095262526,
|
|
"num_tokens": 5427980.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.86662712097168,
|
|
"epoch": 2.347427407030056,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004761100031999315,
|
|
"loss": 5.6361,
|
|
"mean_token_accuracy": 0.1711513876914978,
|
|
"num_tokens": 5439782.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.678351545333863,
|
|
"epoch": 2.3525216505348956,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004759297912066454,
|
|
"loss": 5.551,
|
|
"mean_token_accuracy": 0.16436320841312407,
|
|
"num_tokens": 5451715.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.730842161178589,
|
|
"epoch": 2.357615894039735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047574894046153266,
|
|
"loss": 5.4998,
|
|
"mean_token_accuracy": 0.1749166339635849,
|
|
"num_tokens": 5462840.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.619156837463379,
|
|
"epoch": 2.3627101375445747,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004755674515395141,
|
|
"loss": 5.5056,
|
|
"mean_token_accuracy": 0.1761452615261078,
|
|
"num_tokens": 5474694.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.8200681686401365,
|
|
"epoch": 2.3678043810494143,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004753853250175395,
|
|
"loss": 5.6297,
|
|
"mean_token_accuracy": 0.15863382667303086,
|
|
"num_tokens": 5486292.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.758931732177734,
|
|
"epoch": 2.372898624554254,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047520256147458545,
|
|
"loss": 5.5098,
|
|
"mean_token_accuracy": 0.17546423375606537,
|
|
"num_tokens": 5496799.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.729516172409058,
|
|
"epoch": 2.3779928680590934,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004750191614916536,
|
|
"loss": 5.6011,
|
|
"mean_token_accuracy": 0.1696643739938736,
|
|
"num_tokens": 5509520.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.7106858253479,
|
|
"epoch": 2.383087111563933,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004748351256517689,
|
|
"loss": 5.6461,
|
|
"mean_token_accuracy": 0.1626673683524132,
|
|
"num_tokens": 5521993.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.80874400138855,
|
|
"epoch": 2.388181355068772,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047465045453997756,
|
|
"loss": 5.6138,
|
|
"mean_token_accuracy": 0.16498513668775558,
|
|
"num_tokens": 5533878.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.754479932785034,
|
|
"epoch": 2.393275598573612,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004744651487433455,
|
|
"loss": 5.6075,
|
|
"mean_token_accuracy": 0.1706341877579689,
|
|
"num_tokens": 5546492.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.823743581771851,
|
|
"epoch": 2.398369842078451,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004742792088509561,
|
|
"loss": 5.6373,
|
|
"mean_token_accuracy": 0.16473434418439864,
|
|
"num_tokens": 5557540.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.7825061798095705,
|
|
"epoch": 2.4034640855832907,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047409263545390864,
|
|
"loss": 5.638,
|
|
"mean_token_accuracy": 0.16671781837940217,
|
|
"num_tokens": 5569118.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.760390710830689,
|
|
"epoch": 2.4085583290881303,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004739054291453162,
|
|
"loss": 5.6707,
|
|
"mean_token_accuracy": 0.1595264881849289,
|
|
"num_tokens": 5582738.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.718589115142822,
|
|
"epoch": 2.41365257259297,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047371759052030395,
|
|
"loss": 5.5104,
|
|
"mean_token_accuracy": 0.17014646530151367,
|
|
"num_tokens": 5593941.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.775838613510132,
|
|
"epoch": 2.4187468160978094,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004735291201760072,
|
|
"loss": 5.6353,
|
|
"mean_token_accuracy": 0.16346013844013213,
|
|
"num_tokens": 5606473.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.7426496028900145,
|
|
"epoch": 2.423841059602649,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004733400187115694,
|
|
"loss": 5.515,
|
|
"mean_token_accuracy": 0.17410024553537368,
|
|
"num_tokens": 5617355.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.716971206665039,
|
|
"epoch": 2.4289353031074885,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004731502867281404,
|
|
"loss": 5.59,
|
|
"mean_token_accuracy": 0.16747244149446489,
|
|
"num_tokens": 5629062.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.813964700698852,
|
|
"epoch": 2.434029546612328,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047295992482887435,
|
|
"loss": 5.6203,
|
|
"mean_token_accuracy": 0.16134635508060455,
|
|
"num_tokens": 5641529.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.774265432357788,
|
|
"epoch": 2.4391237901171676,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047276893361892803,
|
|
"loss": 5.6608,
|
|
"mean_token_accuracy": 0.16629068851470946,
|
|
"num_tokens": 5654777.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.724766397476197,
|
|
"epoch": 2.444218033622007,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047257731370545865,
|
|
"loss": 5.6079,
|
|
"mean_token_accuracy": 0.16991823315620422,
|
|
"num_tokens": 5667208.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.6530890464782715,
|
|
"epoch": 2.4493122771268467,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004723850656976221,
|
|
"loss": 5.5402,
|
|
"mean_token_accuracy": 0.1727624788880348,
|
|
"num_tokens": 5679148.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.688948631286621,
|
|
"epoch": 2.4544065206316863,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004721921902065711,
|
|
"loss": 5.4202,
|
|
"mean_token_accuracy": 0.18155144155025482,
|
|
"num_tokens": 5690869.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.646128082275391,
|
|
"epoch": 2.459500764136526,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000471998687845453,
|
|
"loss": 5.5877,
|
|
"mean_token_accuracy": 0.17007255852222442,
|
|
"num_tokens": 5702939.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.814693021774292,
|
|
"epoch": 2.4645950076413654,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047180455922940783,
|
|
"loss": 5.6114,
|
|
"mean_token_accuracy": 0.16709734946489335,
|
|
"num_tokens": 5714651.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.680844163894653,
|
|
"epoch": 2.469689251146205,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004716098049755668,
|
|
"loss": 5.5084,
|
|
"mean_token_accuracy": 0.17117357403039932,
|
|
"num_tokens": 5725310.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.730017471313476,
|
|
"epoch": 2.474783494651044,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004714144257030497,
|
|
"loss": 5.6188,
|
|
"mean_token_accuracy": 0.1655963346362114,
|
|
"num_tokens": 5736357.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.731151962280274,
|
|
"epoch": 2.479877738155884,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004712184220329634,
|
|
"loss": 5.6485,
|
|
"mean_token_accuracy": 0.16924342662096023,
|
|
"num_tokens": 5748281.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.815793228149414,
|
|
"epoch": 2.484971981660723,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047102179458839965,
|
|
"loss": 5.5926,
|
|
"mean_token_accuracy": 0.16823556274175644,
|
|
"num_tokens": 5758939.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.754806566238403,
|
|
"epoch": 2.4900662251655628,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004708245439944333,
|
|
"loss": 5.6032,
|
|
"mean_token_accuracy": 0.16816727072000504,
|
|
"num_tokens": 5769750.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.781820678710938,
|
|
"epoch": 2.4951604686704023,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047062667087812007,
|
|
"loss": 5.609,
|
|
"mean_token_accuracy": 0.17051714658737183,
|
|
"num_tokens": 5781917.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.725583934783936,
|
|
"epoch": 2.500254712175242,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004704281758684946,
|
|
"loss": 5.5582,
|
|
"mean_token_accuracy": 0.1663421705365181,
|
|
"num_tokens": 5793665.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.79179630279541,
|
|
"epoch": 2.5053489556800814,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004702290595965686,
|
|
"loss": 5.6533,
|
|
"mean_token_accuracy": 0.164570951461792,
|
|
"num_tokens": 5806364.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.779351615905762,
|
|
"epoch": 2.510443199184921,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004700293226953288,
|
|
"loss": 5.6523,
|
|
"mean_token_accuracy": 0.16515175849199296,
|
|
"num_tokens": 5818693.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.598007869720459,
|
|
"epoch": 2.5155374426897605,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004698289657997348,
|
|
"loss": 5.4234,
|
|
"mean_token_accuracy": 0.1817185491323471,
|
|
"num_tokens": 5828765.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.716284656524659,
|
|
"epoch": 2.5206316861946,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046962798954671727,
|
|
"loss": 5.5848,
|
|
"mean_token_accuracy": 0.16949136853218078,
|
|
"num_tokens": 5840196.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.71624174118042,
|
|
"epoch": 2.5257259296994397,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046942639457517565,
|
|
"loss": 5.5921,
|
|
"mean_token_accuracy": 0.17095975279808046,
|
|
"num_tokens": 5851117.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.702412509918213,
|
|
"epoch": 2.530820173204279,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004692241815259765,
|
|
"loss": 5.5666,
|
|
"mean_token_accuracy": 0.1720548614859581,
|
|
"num_tokens": 5862918.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.698266696929932,
|
|
"epoch": 2.5359144167091188,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000469021351041951,
|
|
"loss": 5.5909,
|
|
"mean_token_accuracy": 0.16819523870944977,
|
|
"num_tokens": 5875620.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.8046821594238285,
|
|
"epoch": 2.5410086602139583,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004688179037678933,
|
|
"loss": 5.6116,
|
|
"mean_token_accuracy": 0.16816546022891998,
|
|
"num_tokens": 5887113.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.7517413139343265,
|
|
"epoch": 2.546102903718798,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004686138403505582,
|
|
"loss": 5.6401,
|
|
"mean_token_accuracy": 0.16600019484758377,
|
|
"num_tokens": 5898707.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.748650217056275,
|
|
"epoch": 2.5511971472236374,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046840916143865913,
|
|
"loss": 5.5659,
|
|
"mean_token_accuracy": 0.16834910959005356,
|
|
"num_tokens": 5909548.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.746601152420044,
|
|
"epoch": 2.556291390728477,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004682038676828666,
|
|
"loss": 5.5268,
|
|
"mean_token_accuracy": 0.17655671685934066,
|
|
"num_tokens": 5920750.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.699986267089844,
|
|
"epoch": 2.561385634233316,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046799795973580515,
|
|
"loss": 5.6674,
|
|
"mean_token_accuracy": 0.16529995203018188,
|
|
"num_tokens": 5932832.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.712204551696777,
|
|
"epoch": 2.566479877738156,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046779143825205217,
|
|
"loss": 5.5856,
|
|
"mean_token_accuracy": 0.16844226121902467,
|
|
"num_tokens": 5945644.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.71591591835022,
|
|
"epoch": 2.5715741212429952,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004675843038881354,
|
|
"loss": 5.5833,
|
|
"mean_token_accuracy": 0.1668347254395485,
|
|
"num_tokens": 5956760.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.749217176437378,
|
|
"epoch": 2.576668364747835,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046737655730253076,
|
|
"loss": 5.5851,
|
|
"mean_token_accuracy": 0.16941590309143068,
|
|
"num_tokens": 5969179.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.751673221588135,
|
|
"epoch": 2.5817626082526743,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046716819915566067,
|
|
"loss": 5.6626,
|
|
"mean_token_accuracy": 0.16343993693590164,
|
|
"num_tokens": 5982051.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.858973407745362,
|
|
"epoch": 2.586856851757514,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046695923010989145,
|
|
"loss": 5.7064,
|
|
"mean_token_accuracy": 0.1570582315325737,
|
|
"num_tokens": 5994038.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.760682821273804,
|
|
"epoch": 2.5919510952623535,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046674965082953157,
|
|
"loss": 5.6142,
|
|
"mean_token_accuracy": 0.1664024993777275,
|
|
"num_tokens": 6005356.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.748070144653321,
|
|
"epoch": 2.597045338767193,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046653946198082943,
|
|
"loss": 5.6498,
|
|
"mean_token_accuracy": 0.16623507738113402,
|
|
"num_tokens": 6018477.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.688044548034668,
|
|
"epoch": 2.6021395822720326,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046632866423197125,
|
|
"loss": 5.5644,
|
|
"mean_token_accuracy": 0.17507824599742888,
|
|
"num_tokens": 6029084.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.751237392425537,
|
|
"epoch": 2.607233825776872,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004661172582530788,
|
|
"loss": 5.553,
|
|
"mean_token_accuracy": 0.1744393900036812,
|
|
"num_tokens": 6040271.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.617528438568115,
|
|
"epoch": 2.6123280692817117,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004659052447162076,
|
|
"loss": 5.4538,
|
|
"mean_token_accuracy": 0.18345775604248046,
|
|
"num_tokens": 6051278.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.655602598190308,
|
|
"epoch": 2.6174223127865512,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004656926242953445,
|
|
"loss": 5.6426,
|
|
"mean_token_accuracy": 0.16686174869537354,
|
|
"num_tokens": 6063397.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.782414865493775,
|
|
"epoch": 2.622516556291391,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004654793976664055,
|
|
"loss": 5.5511,
|
|
"mean_token_accuracy": 0.17096301466226577,
|
|
"num_tokens": 6075105.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.770709800720215,
|
|
"epoch": 2.6276107997962304,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046526556550723395,
|
|
"loss": 5.6077,
|
|
"mean_token_accuracy": 0.16690193563699723,
|
|
"num_tokens": 6086069.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.668476676940918,
|
|
"epoch": 2.63270504330107,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046505112849759794,
|
|
"loss": 5.5334,
|
|
"mean_token_accuracy": 0.17296285480260848,
|
|
"num_tokens": 6096997.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.6466419219970705,
|
|
"epoch": 2.6377992868059095,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046483608731918855,
|
|
"loss": 5.5622,
|
|
"mean_token_accuracy": 0.17210524827241896,
|
|
"num_tokens": 6109510.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.721639251708984,
|
|
"epoch": 2.642893530310749,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046462044265561735,
|
|
"loss": 5.4758,
|
|
"mean_token_accuracy": 0.17866286337375642,
|
|
"num_tokens": 6120111.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.725635290145874,
|
|
"epoch": 2.647987773815588,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046440419519241455,
|
|
"loss": 5.6398,
|
|
"mean_token_accuracy": 0.1649184688925743,
|
|
"num_tokens": 6131558.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.738513660430908,
|
|
"epoch": 2.653082017320428,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004641873456170265,
|
|
"loss": 5.5668,
|
|
"mean_token_accuracy": 0.1745607227087021,
|
|
"num_tokens": 6143687.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.821848821640015,
|
|
"epoch": 2.6581762608252673,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004639698946188137,
|
|
"loss": 5.6725,
|
|
"mean_token_accuracy": 0.16588608771562577,
|
|
"num_tokens": 6153692.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.730045890808105,
|
|
"epoch": 2.663270504330107,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004637518428890486,
|
|
"loss": 5.566,
|
|
"mean_token_accuracy": 0.17797428816556932,
|
|
"num_tokens": 6164689.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.699032831192016,
|
|
"epoch": 2.6683647478349464,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004635331911209133,
|
|
"loss": 5.6323,
|
|
"mean_token_accuracy": 0.16478496938943862,
|
|
"num_tokens": 6176720.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.7133537292480465,
|
|
"epoch": 2.673458991339786,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004633139400094974,
|
|
"loss": 5.5275,
|
|
"mean_token_accuracy": 0.17309380918741227,
|
|
"num_tokens": 6187745.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.686916875839233,
|
|
"epoch": 2.6785532348446255,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000463094090251796,
|
|
"loss": 5.5664,
|
|
"mean_token_accuracy": 0.1765984982252121,
|
|
"num_tokens": 6200403.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.644669437408448,
|
|
"epoch": 2.683647478349465,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046287364254670696,
|
|
"loss": 5.5541,
|
|
"mean_token_accuracy": 0.16958998441696166,
|
|
"num_tokens": 6211412.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.787599802017212,
|
|
"epoch": 2.6887417218543046,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004626525975950292,
|
|
"loss": 5.6695,
|
|
"mean_token_accuracy": 0.16274720877408982,
|
|
"num_tokens": 6222798.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.734703016281128,
|
|
"epoch": 2.693835965359144,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004624309560994603,
|
|
"loss": 5.5418,
|
|
"mean_token_accuracy": 0.17499985545873642,
|
|
"num_tokens": 6234846.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.692458200454712,
|
|
"epoch": 2.6989302088639837,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004622087187645942,
|
|
"loss": 5.5626,
|
|
"mean_token_accuracy": 0.16789343431591988,
|
|
"num_tokens": 6245771.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.6511680603027346,
|
|
"epoch": 2.7040244523688233,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004619858862969188,
|
|
"loss": 5.5038,
|
|
"mean_token_accuracy": 0.17768730074167252,
|
|
"num_tokens": 6258254.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.756880092620849,
|
|
"epoch": 2.709118695873663,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004617624594048143,
|
|
"loss": 5.5661,
|
|
"mean_token_accuracy": 0.16991113126277924,
|
|
"num_tokens": 6269383.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.807396936416626,
|
|
"epoch": 2.7142129393785024,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004615384387985504,
|
|
"loss": 5.5279,
|
|
"mean_token_accuracy": 0.17493753284215927,
|
|
"num_tokens": 6280275.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.601258277893066,
|
|
"epoch": 2.719307182883342,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000461313825190284,
|
|
"loss": 5.5497,
|
|
"mean_token_accuracy": 0.1652575895190239,
|
|
"num_tokens": 6291760.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.6643720149993895,
|
|
"epoch": 2.7244014263881815,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004610886192940574,
|
|
"loss": 5.5138,
|
|
"mean_token_accuracy": 0.17513244450092316,
|
|
"num_tokens": 6303041.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.680649518966675,
|
|
"epoch": 2.729495669893021,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046086282182579566,
|
|
"loss": 5.468,
|
|
"mean_token_accuracy": 0.177858167886734,
|
|
"num_tokens": 6313950.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.587852954864502,
|
|
"epoch": 2.73458991339786,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004606364335033045,
|
|
"loss": 5.6197,
|
|
"mean_token_accuracy": 0.17280302196741104,
|
|
"num_tokens": 6325830.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.832064008712768,
|
|
"epoch": 2.7396841569027,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046040945504626787,
|
|
"loss": 5.649,
|
|
"mean_token_accuracy": 0.16793008893728256,
|
|
"num_tokens": 6337978.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.666331195831299,
|
|
"epoch": 2.7447784004075393,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046018188717624576,
|
|
"loss": 5.5048,
|
|
"mean_token_accuracy": 0.17441839426755906,
|
|
"num_tokens": 6349623.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.606363725662232,
|
|
"epoch": 2.7498726439123793,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000459953730616672,
|
|
"loss": 5.4751,
|
|
"mean_token_accuracy": 0.17461808919906616,
|
|
"num_tokens": 6361149.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.624794578552246,
|
|
"epoch": 2.7549668874172184,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045972498609285175,
|
|
"loss": 5.5114,
|
|
"mean_token_accuracy": 0.17174820154905318,
|
|
"num_tokens": 6372691.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.696760940551758,
|
|
"epoch": 2.760061130922058,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045949565433195925,
|
|
"loss": 5.6292,
|
|
"mean_token_accuracy": 0.15939597636461258,
|
|
"num_tokens": 6385560.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.75138111114502,
|
|
"epoch": 2.7651553744268975,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045926573606303576,
|
|
"loss": 5.5749,
|
|
"mean_token_accuracy": 0.16573688834905625,
|
|
"num_tokens": 6397227.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.742139291763306,
|
|
"epoch": 2.770249617931737,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045903523201698684,
|
|
"loss": 5.5911,
|
|
"mean_token_accuracy": 0.16867348849773406,
|
|
"num_tokens": 6409906.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.766191387176514,
|
|
"epoch": 2.7753438614365766,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004588041429265803,
|
|
"loss": 5.5517,
|
|
"mean_token_accuracy": 0.16983173936605453,
|
|
"num_tokens": 6421922.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.676415681838989,
|
|
"epoch": 2.780438104941416,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004585724695264436,
|
|
"loss": 5.6116,
|
|
"mean_token_accuracy": 0.1660709023475647,
|
|
"num_tokens": 6433135.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.7621002197265625,
|
|
"epoch": 2.7855323484462557,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045834021255306223,
|
|
"loss": 5.6099,
|
|
"mean_token_accuracy": 0.1651996046304703,
|
|
"num_tokens": 6444399.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.748862266540527,
|
|
"epoch": 2.7906265919510953,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004581073727447762,
|
|
"loss": 5.6505,
|
|
"mean_token_accuracy": 0.1635200247168541,
|
|
"num_tokens": 6456227.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.689176321029663,
|
|
"epoch": 2.795720835455935,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004578739508417788,
|
|
"loss": 5.553,
|
|
"mean_token_accuracy": 0.17098861634731294,
|
|
"num_tokens": 6468622.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.697390842437744,
|
|
"epoch": 2.8008150789607744,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004576399475861137,
|
|
"loss": 5.6043,
|
|
"mean_token_accuracy": 0.16807106435298919,
|
|
"num_tokens": 6480408.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.71913161277771,
|
|
"epoch": 2.805909322465614,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004574053637216724,
|
|
"loss": 5.5255,
|
|
"mean_token_accuracy": 0.17133117020130156,
|
|
"num_tokens": 6492864.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.600292348861695,
|
|
"epoch": 2.8110035659704535,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004571701999941926,
|
|
"loss": 5.447,
|
|
"mean_token_accuracy": 0.17601710855960845,
|
|
"num_tokens": 6504103.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.72335729598999,
|
|
"epoch": 2.816097809475293,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000456934457151255,
|
|
"loss": 5.6189,
|
|
"mean_token_accuracy": 0.16324349641799926,
|
|
"num_tokens": 6516560.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.620775604248047,
|
|
"epoch": 2.821192052980132,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045669813594228137,
|
|
"loss": 5.5689,
|
|
"mean_token_accuracy": 0.1696472182869911,
|
|
"num_tokens": 6530002.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.723550462722779,
|
|
"epoch": 2.826286296484972,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045646123711853235,
|
|
"loss": 5.5703,
|
|
"mean_token_accuracy": 0.1710247129201889,
|
|
"num_tokens": 6542454.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.577154636383057,
|
|
"epoch": 2.8313805399898113,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045622376143310435,
|
|
"loss": 5.5084,
|
|
"mean_token_accuracy": 0.17757378816604613,
|
|
"num_tokens": 6553385.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.698977088928222,
|
|
"epoch": 2.8364747834946513,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045598570964092817,
|
|
"loss": 5.5105,
|
|
"mean_token_accuracy": 0.17238980084657668,
|
|
"num_tokens": 6564385.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.653533697128296,
|
|
"epoch": 2.8415690269994904,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045574708249876547,
|
|
"loss": 5.5065,
|
|
"mean_token_accuracy": 0.17230129092931748,
|
|
"num_tokens": 6575369.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.696118783950806,
|
|
"epoch": 2.84666327050433,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045550788076520736,
|
|
"loss": 5.5799,
|
|
"mean_token_accuracy": 0.16681847721338272,
|
|
"num_tokens": 6587173.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.770565414428711,
|
|
"epoch": 2.8517575140091695,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004552681052006713,
|
|
"loss": 5.6447,
|
|
"mean_token_accuracy": 0.16493052691221238,
|
|
"num_tokens": 6598919.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.700059938430786,
|
|
"epoch": 2.856851757514009,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004550277565673992,
|
|
"loss": 5.5952,
|
|
"mean_token_accuracy": 0.1692450761795044,
|
|
"num_tokens": 6610353.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.6513769149780275,
|
|
"epoch": 2.8619460010188487,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045478683562945457,
|
|
"loss": 5.5433,
|
|
"mean_token_accuracy": 0.17437132745981215,
|
|
"num_tokens": 6622420.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.763458251953125,
|
|
"epoch": 2.867040244523688,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004545453431527202,
|
|
"loss": 5.5752,
|
|
"mean_token_accuracy": 0.16620224118232726,
|
|
"num_tokens": 6634234.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.752091741561889,
|
|
"epoch": 2.8721344880285278,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000454303279904896,
|
|
"loss": 5.6764,
|
|
"mean_token_accuracy": 0.1649190068244934,
|
|
"num_tokens": 6648272.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.67063422203064,
|
|
"epoch": 2.8772287315333673,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045406064665549624,
|
|
"loss": 5.51,
|
|
"mean_token_accuracy": 0.1764917865395546,
|
|
"num_tokens": 6660114.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.702568817138672,
|
|
"epoch": 2.882322975038207,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004538174441758471,
|
|
"loss": 5.5603,
|
|
"mean_token_accuracy": 0.1660371392965317,
|
|
"num_tokens": 6670433.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.561418724060059,
|
|
"epoch": 2.8874172185430464,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045357367323908463,
|
|
"loss": 5.4965,
|
|
"mean_token_accuracy": 0.1765263319015503,
|
|
"num_tokens": 6681918.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.8111371994018555,
|
|
"epoch": 2.892511462047886,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045332933462015156,
|
|
"loss": 5.6034,
|
|
"mean_token_accuracy": 0.1653100550174713,
|
|
"num_tokens": 6692681.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.676050043106079,
|
|
"epoch": 2.8976057055527256,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004530844290957957,
|
|
"loss": 5.5709,
|
|
"mean_token_accuracy": 0.16507604271173476,
|
|
"num_tokens": 6704454.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.707255125045776,
|
|
"epoch": 2.902699949057565,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045283895744456694,
|
|
"loss": 5.5462,
|
|
"mean_token_accuracy": 0.1656055971980095,
|
|
"num_tokens": 6718226.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.618920516967774,
|
|
"epoch": 2.9077941925624042,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045259292044681456,
|
|
"loss": 5.503,
|
|
"mean_token_accuracy": 0.17829408198595048,
|
|
"num_tokens": 6730507.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.661051607131958,
|
|
"epoch": 2.9128884360672442,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004523463188846855,
|
|
"loss": 5.5569,
|
|
"mean_token_accuracy": 0.16910663843154908,
|
|
"num_tokens": 6743312.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.619609403610229,
|
|
"epoch": 2.9179826795720833,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004520991535421211,
|
|
"loss": 5.5532,
|
|
"mean_token_accuracy": 0.17489729076623917,
|
|
"num_tokens": 6757945.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.707130575180054,
|
|
"epoch": 2.9230769230769234,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004518514252048552,
|
|
"loss": 5.5982,
|
|
"mean_token_accuracy": 0.1614573195576668,
|
|
"num_tokens": 6770995.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.688896751403808,
|
|
"epoch": 2.9281711665817625,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004516031346604112,
|
|
"loss": 5.5097,
|
|
"mean_token_accuracy": 0.17259131819009782,
|
|
"num_tokens": 6782117.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.613061618804932,
|
|
"epoch": 2.933265410086602,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004513542826980999,
|
|
"loss": 5.6155,
|
|
"mean_token_accuracy": 0.16592106968164444,
|
|
"num_tokens": 6795932.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.778715896606445,
|
|
"epoch": 2.9383596535914416,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004511048701090167,
|
|
"loss": 5.5071,
|
|
"mean_token_accuracy": 0.17740725725889206,
|
|
"num_tokens": 6805708.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.677082109451294,
|
|
"epoch": 2.943453897096281,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004508548976860392,
|
|
"loss": 5.5664,
|
|
"mean_token_accuracy": 0.16885116696357727,
|
|
"num_tokens": 6816964.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.74324221611023,
|
|
"epoch": 2.9485481406011207,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000450604366223825,
|
|
"loss": 5.6029,
|
|
"mean_token_accuracy": 0.16539741307497025,
|
|
"num_tokens": 6828330.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.676993703842163,
|
|
"epoch": 2.9536423841059603,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045035327651880846,
|
|
"loss": 5.5803,
|
|
"mean_token_accuracy": 0.1668591395020485,
|
|
"num_tokens": 6839860.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.698193550109863,
|
|
"epoch": 2.9587366276108,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045010162936919896,
|
|
"loss": 5.5302,
|
|
"mean_token_accuracy": 0.17305754870176315,
|
|
"num_tokens": 6850453.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.620525979995728,
|
|
"epoch": 2.9638308711156394,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004498494255749776,
|
|
"loss": 5.5215,
|
|
"mean_token_accuracy": 0.17544011920690536,
|
|
"num_tokens": 6861661.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.680759716033935,
|
|
"epoch": 2.968925114620479,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004495966659378954,
|
|
"loss": 5.5733,
|
|
"mean_token_accuracy": 0.16637389808893205,
|
|
"num_tokens": 6873474.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.704854440689087,
|
|
"epoch": 2.9740193581253185,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044934335126147016,
|
|
"loss": 5.5253,
|
|
"mean_token_accuracy": 0.17334222048521042,
|
|
"num_tokens": 6885761.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.658196449279785,
|
|
"epoch": 2.979113601630158,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004490894823509843,
|
|
"loss": 5.5551,
|
|
"mean_token_accuracy": 0.1653917744755745,
|
|
"num_tokens": 6895911.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.674737739562988,
|
|
"epoch": 2.9842078451349976,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004488350600134819,
|
|
"loss": 5.4932,
|
|
"mean_token_accuracy": 0.17815617918968202,
|
|
"num_tokens": 6907431.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.640310478210449,
|
|
"epoch": 2.989302088639837,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004485800850577667,
|
|
"loss": 5.5604,
|
|
"mean_token_accuracy": 0.17894753068685532,
|
|
"num_tokens": 6919007.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.745733547210693,
|
|
"epoch": 2.9943963321446763,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004483245582943989,
|
|
"loss": 5.5695,
|
|
"mean_token_accuracy": 0.17720169126987456,
|
|
"num_tokens": 6930194.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.655760669708252,
|
|
"epoch": 2.9994905756495163,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044806848053569314,
|
|
"loss": 5.5319,
|
|
"mean_token_accuracy": 0.17353491336107255,
|
|
"num_tokens": 6941276.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.6295300589667425,
|
|
"epoch": 3.0040753948038716,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044781185259571544,
|
|
"loss": 5.2705,
|
|
"mean_token_accuracy": 0.19175558785597482,
|
|
"num_tokens": 6952351.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.7189483642578125,
|
|
"epoch": 3.009169638308711,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004475546752902811,
|
|
"loss": 5.2839,
|
|
"mean_token_accuracy": 0.1852574020624161,
|
|
"num_tokens": 6965052.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.553421831130981,
|
|
"epoch": 3.0142638818135508,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044729694943695155,
|
|
"loss": 5.1492,
|
|
"mean_token_accuracy": 0.18978555649518966,
|
|
"num_tokens": 6976662.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.652298164367676,
|
|
"epoch": 3.0193581253183903,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004470386758550321,
|
|
"loss": 5.2359,
|
|
"mean_token_accuracy": 0.18836504220962524,
|
|
"num_tokens": 6989787.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.623361349105835,
|
|
"epoch": 3.02445236882323,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004467798553655695,
|
|
"loss": 5.1504,
|
|
"mean_token_accuracy": 0.1876479685306549,
|
|
"num_tokens": 7000866.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.643063926696778,
|
|
"epoch": 3.0295466123280694,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000446520488791349,
|
|
"loss": 5.2655,
|
|
"mean_token_accuracy": 0.18362555503845215,
|
|
"num_tokens": 7012862.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.644259786605835,
|
|
"epoch": 3.034640855832909,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004462605769568915,
|
|
"loss": 5.2276,
|
|
"mean_token_accuracy": 0.185024493932724,
|
|
"num_tokens": 7025434.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.656643486022949,
|
|
"epoch": 3.0397350993377485,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004460001206884519,
|
|
"loss": 5.2497,
|
|
"mean_token_accuracy": 0.18152078241109848,
|
|
"num_tokens": 7037458.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.628327131271362,
|
|
"epoch": 3.0448293428425877,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004457391208140155,
|
|
"loss": 5.2122,
|
|
"mean_token_accuracy": 0.19196427315473558,
|
|
"num_tokens": 7049493.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.568745374679565,
|
|
"epoch": 3.049923586347427,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004454775781632955,
|
|
"loss": 5.2348,
|
|
"mean_token_accuracy": 0.18303260207176208,
|
|
"num_tokens": 7059824.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.575043821334839,
|
|
"epoch": 3.0550178298522668,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044521549356773096,
|
|
"loss": 5.1722,
|
|
"mean_token_accuracy": 0.19056803435087205,
|
|
"num_tokens": 7070604.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.0550178298522668,
|
|
"eval_entropy": 5.41912328225595,
|
|
"eval_loss": 5.802611351013184,
|
|
"eval_mean_token_accuracy": 0.16801401122852608,
|
|
"eval_num_tokens": 7070604.0,
|
|
"eval_runtime": 2.6304,
|
|
"eval_samples_per_second": 1231.749,
|
|
"eval_steps_per_second": 153.969,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9810,
|
|
"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": 1.1276555728896e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|