c375242180
Model: fpadovani/isl-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed455 Source: Original Platform
2046 lines
54 KiB
JSON
2046 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.0183392766174224,
|
|
"eval_steps": 1000,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 3769322303232000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|