bd86e71c20
Model: fpadovani/dan-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.6736401673640167,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.891409587860108,
|
|
"epoch": 0.0041841004184100415,
|
|
"grad_norm": 24.125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.1621,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9848.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.9063023090362545,
|
|
"epoch": 0.008368200836820083,
|
|
"grad_norm": 25.0,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.0066,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 18196.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.956460618972779,
|
|
"epoch": 0.012552301255230125,
|
|
"grad_norm": 27.25,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.9152,
|
|
"mean_token_accuracy": 0.00022191319148987532,
|
|
"num_tokens": 26927.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.071391153335571,
|
|
"epoch": 0.016736401673640166,
|
|
"grad_norm": 34.5,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.5124,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 34987.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.681827545166016,
|
|
"epoch": 0.02092050209205021,
|
|
"grad_norm": 42.5,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 12.7729,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 44936.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.654202032089233,
|
|
"epoch": 0.02510460251046025,
|
|
"grad_norm": 24.5,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 11.6009,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 54562.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.472756958007812,
|
|
"epoch": 0.029288702928870293,
|
|
"grad_norm": 3.703125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.6978,
|
|
"mean_token_accuracy": 0.005889179778750986,
|
|
"num_tokens": 63844.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.730471515655518,
|
|
"epoch": 0.03347280334728033,
|
|
"grad_norm": 3.296875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.5735,
|
|
"mean_token_accuracy": 0.015836346428841352,
|
|
"num_tokens": 74331.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.730430603027344,
|
|
"epoch": 0.03765690376569038,
|
|
"grad_norm": 2.90625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.3807,
|
|
"mean_token_accuracy": 0.019579107593744993,
|
|
"num_tokens": 83028.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.661893463134765,
|
|
"epoch": 0.04184100418410042,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.2168,
|
|
"mean_token_accuracy": 0.020131182763725518,
|
|
"num_tokens": 91395.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.666141891479493,
|
|
"epoch": 0.04602510460251046,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.0907,
|
|
"mean_token_accuracy": 0.016700492519885302,
|
|
"num_tokens": 101868.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.728130531311034,
|
|
"epoch": 0.0502092050209205,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.9346,
|
|
"mean_token_accuracy": 0.02238125018775463,
|
|
"num_tokens": 110403.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.730528163909913,
|
|
"epoch": 0.05439330543933055,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.9073,
|
|
"mean_token_accuracy": 0.02675023004412651,
|
|
"num_tokens": 119557.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.718873310089112,
|
|
"epoch": 0.058577405857740586,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7675,
|
|
"mean_token_accuracy": 0.024789097160100936,
|
|
"num_tokens": 128903.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.718735504150391,
|
|
"epoch": 0.06276150627615062,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7466,
|
|
"mean_token_accuracy": 0.03036366552114487,
|
|
"num_tokens": 137630.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.720433235168457,
|
|
"epoch": 0.06694560669456066,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6143,
|
|
"mean_token_accuracy": 0.03396541755646467,
|
|
"num_tokens": 146335.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.716510105133057,
|
|
"epoch": 0.07112970711297072,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5501,
|
|
"mean_token_accuracy": 0.03693135865032673,
|
|
"num_tokens": 155766.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.7073184967041,
|
|
"epoch": 0.07531380753138076,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4819,
|
|
"mean_token_accuracy": 0.029630407132208346,
|
|
"num_tokens": 165844.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.698113632202148,
|
|
"epoch": 0.0794979079497908,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3976,
|
|
"mean_token_accuracy": 0.040011851117014885,
|
|
"num_tokens": 174858.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.68637809753418,
|
|
"epoch": 0.08368200836820083,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2692,
|
|
"mean_token_accuracy": 0.04454572442919016,
|
|
"num_tokens": 184020.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.661810874938965,
|
|
"epoch": 0.08786610878661087,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1619,
|
|
"mean_token_accuracy": 0.044953469187021255,
|
|
"num_tokens": 193714.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.638691425323486,
|
|
"epoch": 0.09205020920502092,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0919,
|
|
"mean_token_accuracy": 0.04682178944349289,
|
|
"num_tokens": 203139.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.608509731292724,
|
|
"epoch": 0.09623430962343096,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.935,
|
|
"mean_token_accuracy": 0.05361910648643971,
|
|
"num_tokens": 212446.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.535148620605469,
|
|
"epoch": 0.100418410041841,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.7879,
|
|
"mean_token_accuracy": 0.05296131558716297,
|
|
"num_tokens": 220910.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.479767608642579,
|
|
"epoch": 0.10460251046025104,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7201,
|
|
"mean_token_accuracy": 0.05455400906503201,
|
|
"num_tokens": 230522.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.398798561096191,
|
|
"epoch": 0.1087866108786611,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.5632,
|
|
"mean_token_accuracy": 0.05519989021122455,
|
|
"num_tokens": 240092.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.31678285598755,
|
|
"epoch": 0.11297071129707113,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.469,
|
|
"mean_token_accuracy": 0.056049537286162375,
|
|
"num_tokens": 249478.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.232394123077393,
|
|
"epoch": 0.11715481171548117,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.2797,
|
|
"mean_token_accuracy": 0.05960724465548992,
|
|
"num_tokens": 257714.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.128987789154053,
|
|
"epoch": 0.12133891213389121,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.2716,
|
|
"mean_token_accuracy": 0.054746852815151216,
|
|
"num_tokens": 266569.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.016991806030273,
|
|
"epoch": 0.12552301255230125,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.1522,
|
|
"mean_token_accuracy": 0.05324812866747379,
|
|
"num_tokens": 276688.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.930090236663819,
|
|
"epoch": 0.1297071129707113,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.969,
|
|
"mean_token_accuracy": 0.05717452354729176,
|
|
"num_tokens": 285430.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.747330951690675,
|
|
"epoch": 0.13389121338912133,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8918,
|
|
"mean_token_accuracy": 0.05585745945572853,
|
|
"num_tokens": 295727.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.500588703155518,
|
|
"epoch": 0.13807531380753138,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7962,
|
|
"mean_token_accuracy": 0.05966520681977272,
|
|
"num_tokens": 304854.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.284924411773682,
|
|
"epoch": 0.14225941422594143,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7152,
|
|
"mean_token_accuracy": 0.06037968210875988,
|
|
"num_tokens": 313579.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.0167311668396,
|
|
"epoch": 0.14644351464435146,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.7018,
|
|
"mean_token_accuracy": 0.05947771333158016,
|
|
"num_tokens": 323310.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.793067455291748,
|
|
"epoch": 0.1506276150627615,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.5597,
|
|
"mean_token_accuracy": 0.054980911687016486,
|
|
"num_tokens": 331720.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.578739547729493,
|
|
"epoch": 0.15481171548117154,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4836,
|
|
"mean_token_accuracy": 0.05964849032461643,
|
|
"num_tokens": 340331.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.347243404388427,
|
|
"epoch": 0.1589958158995816,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.4873,
|
|
"mean_token_accuracy": 0.05853373594582081,
|
|
"num_tokens": 349677.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.249125862121582,
|
|
"epoch": 0.16317991631799164,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.4457,
|
|
"mean_token_accuracy": 0.05640666857361794,
|
|
"num_tokens": 358674.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.108446025848389,
|
|
"epoch": 0.16736401673640167,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4564,
|
|
"mean_token_accuracy": 0.05889376923441887,
|
|
"num_tokens": 368258.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.07214674949646,
|
|
"epoch": 0.17154811715481172,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.3668,
|
|
"mean_token_accuracy": 0.06832478605210782,
|
|
"num_tokens": 377083.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.066781759262085,
|
|
"epoch": 0.17573221757322174,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.4131,
|
|
"mean_token_accuracy": 0.05864137038588524,
|
|
"num_tokens": 385456.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.966042947769165,
|
|
"epoch": 0.1799163179916318,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3069,
|
|
"mean_token_accuracy": 0.06798864156007767,
|
|
"num_tokens": 395295.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.028728580474853,
|
|
"epoch": 0.18410041841004185,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.3528,
|
|
"mean_token_accuracy": 0.06413010321557522,
|
|
"num_tokens": 405023.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.929877233505249,
|
|
"epoch": 0.18828451882845187,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3556,
|
|
"mean_token_accuracy": 0.06588337235152722,
|
|
"num_tokens": 414224.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.9531862258911135,
|
|
"epoch": 0.19246861924686193,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2507,
|
|
"mean_token_accuracy": 0.06776628717780113,
|
|
"num_tokens": 423951.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.959342336654663,
|
|
"epoch": 0.19665271966527198,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2897,
|
|
"mean_token_accuracy": 0.0697428148239851,
|
|
"num_tokens": 432427.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.9486246585845945,
|
|
"epoch": 0.200836820083682,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1966,
|
|
"mean_token_accuracy": 0.07294891811907292,
|
|
"num_tokens": 441560.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.970178651809692,
|
|
"epoch": 0.20502092050209206,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2917,
|
|
"mean_token_accuracy": 0.06312177963554859,
|
|
"num_tokens": 450364.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.8968164920806885,
|
|
"epoch": 0.20920502092050208,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2249,
|
|
"mean_token_accuracy": 0.07028925977647305,
|
|
"num_tokens": 460076.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.926723957061768,
|
|
"epoch": 0.21338912133891214,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1878,
|
|
"mean_token_accuracy": 0.07347588390111923,
|
|
"num_tokens": 468718.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.886239528656006,
|
|
"epoch": 0.2175732217573222,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.2306,
|
|
"mean_token_accuracy": 0.07341902256011963,
|
|
"num_tokens": 478436.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.901343250274659,
|
|
"epoch": 0.2217573221757322,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.2425,
|
|
"mean_token_accuracy": 0.07287401147186756,
|
|
"num_tokens": 486833.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.8618532657623295,
|
|
"epoch": 0.22594142259414227,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.1112,
|
|
"mean_token_accuracy": 0.08160877451300622,
|
|
"num_tokens": 495119.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.881023740768432,
|
|
"epoch": 0.2301255230125523,
|
|
"grad_norm": 2.8125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2378,
|
|
"mean_token_accuracy": 0.07865233719348907,
|
|
"num_tokens": 505396.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.713179016113282,
|
|
"epoch": 0.23430962343096234,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1126,
|
|
"mean_token_accuracy": 0.07563550472259521,
|
|
"num_tokens": 515259.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.866283369064331,
|
|
"epoch": 0.2384937238493724,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1237,
|
|
"mean_token_accuracy": 0.07651542238891125,
|
|
"num_tokens": 523608.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.718380069732666,
|
|
"epoch": 0.24267782426778242,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.123,
|
|
"mean_token_accuracy": 0.07430066615343094,
|
|
"num_tokens": 534613.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.648086595535278,
|
|
"epoch": 0.24686192468619247,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1218,
|
|
"mean_token_accuracy": 0.08086465299129486,
|
|
"num_tokens": 543347.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.774254369735718,
|
|
"epoch": 0.2510460251046025,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.096,
|
|
"mean_token_accuracy": 0.07647029384970665,
|
|
"num_tokens": 552371.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.655633068084716,
|
|
"epoch": 0.25523012552301255,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0231,
|
|
"mean_token_accuracy": 0.08150922022759914,
|
|
"num_tokens": 560742.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.669692802429199,
|
|
"epoch": 0.2594142259414226,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.1089,
|
|
"mean_token_accuracy": 0.08333353251218796,
|
|
"num_tokens": 569960.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.660837411880493,
|
|
"epoch": 0.26359832635983266,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0674,
|
|
"mean_token_accuracy": 0.07987545132637024,
|
|
"num_tokens": 578859.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.71635251045227,
|
|
"epoch": 0.26778242677824265,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0662,
|
|
"mean_token_accuracy": 0.08209685683250427,
|
|
"num_tokens": 588146.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.540524530410766,
|
|
"epoch": 0.2719665271966527,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.0169,
|
|
"mean_token_accuracy": 0.08419515639543533,
|
|
"num_tokens": 597268.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.600224113464355,
|
|
"epoch": 0.27615062761506276,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9492,
|
|
"mean_token_accuracy": 0.08558865413069724,
|
|
"num_tokens": 606199.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.614625310897827,
|
|
"epoch": 0.2803347280334728,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.0511,
|
|
"mean_token_accuracy": 0.07545995935797692,
|
|
"num_tokens": 615966.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.659131479263306,
|
|
"epoch": 0.28451882845188287,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.026,
|
|
"mean_token_accuracy": 0.08085056543350219,
|
|
"num_tokens": 625689.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.566850900650024,
|
|
"epoch": 0.28870292887029286,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.061,
|
|
"mean_token_accuracy": 0.08259228393435478,
|
|
"num_tokens": 634877.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.583298397064209,
|
|
"epoch": 0.2928870292887029,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9891,
|
|
"mean_token_accuracy": 0.08486707210540771,
|
|
"num_tokens": 644042.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.560001087188721,
|
|
"epoch": 0.29707112970711297,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.959,
|
|
"mean_token_accuracy": 0.08440612144768238,
|
|
"num_tokens": 653072.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.451096248626709,
|
|
"epoch": 0.301255230125523,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9641,
|
|
"mean_token_accuracy": 0.08682873398065567,
|
|
"num_tokens": 663686.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.523260593414307,
|
|
"epoch": 0.3054393305439331,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9034,
|
|
"mean_token_accuracy": 0.08619396463036537,
|
|
"num_tokens": 672293.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.444020557403564,
|
|
"epoch": 0.30962343096234307,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.867,
|
|
"mean_token_accuracy": 0.08871779963374138,
|
|
"num_tokens": 681758.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.431160640716553,
|
|
"epoch": 0.3138075313807531,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8923,
|
|
"mean_token_accuracy": 0.08616945594549179,
|
|
"num_tokens": 690727.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.471702432632446,
|
|
"epoch": 0.3179916317991632,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8543,
|
|
"mean_token_accuracy": 0.08554501757025719,
|
|
"num_tokens": 701210.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.374780702590942,
|
|
"epoch": 0.32217573221757323,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8415,
|
|
"mean_token_accuracy": 0.08815910518169404,
|
|
"num_tokens": 710869.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.413737440109253,
|
|
"epoch": 0.3263598326359833,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.9429,
|
|
"mean_token_accuracy": 0.08294537365436554,
|
|
"num_tokens": 721455.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.397308969497681,
|
|
"epoch": 0.3305439330543933,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8361,
|
|
"mean_token_accuracy": 0.09202742874622345,
|
|
"num_tokens": 730917.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.375412464141846,
|
|
"epoch": 0.33472803347280333,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8569,
|
|
"mean_token_accuracy": 0.09109241589903831,
|
|
"num_tokens": 739302.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.446383142471314,
|
|
"epoch": 0.3389121338912134,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8401,
|
|
"mean_token_accuracy": 0.08787635415792465,
|
|
"num_tokens": 748785.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.324094533920288,
|
|
"epoch": 0.34309623430962344,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7754,
|
|
"mean_token_accuracy": 0.09026021733880044,
|
|
"num_tokens": 757215.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.308840560913086,
|
|
"epoch": 0.3472803347280335,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7076,
|
|
"mean_token_accuracy": 0.09466831386089325,
|
|
"num_tokens": 765682.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.378483152389526,
|
|
"epoch": 0.3514644351464435,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9007,
|
|
"mean_token_accuracy": 0.08436162248253823,
|
|
"num_tokens": 774814.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.392928743362427,
|
|
"epoch": 0.35564853556485354,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8148,
|
|
"mean_token_accuracy": 0.091542549431324,
|
|
"num_tokens": 784019.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.225587844848633,
|
|
"epoch": 0.3598326359832636,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6954,
|
|
"mean_token_accuracy": 0.0938286691904068,
|
|
"num_tokens": 793237.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.288524055480957,
|
|
"epoch": 0.36401673640167365,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7272,
|
|
"mean_token_accuracy": 0.09245252907276154,
|
|
"num_tokens": 801548.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.260332345962524,
|
|
"epoch": 0.3682008368200837,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7725,
|
|
"mean_token_accuracy": 0.09467287361621857,
|
|
"num_tokens": 810974.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.250288343429565,
|
|
"epoch": 0.3723849372384937,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.6705,
|
|
"mean_token_accuracy": 0.09519676417112351,
|
|
"num_tokens": 820023.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.248160552978516,
|
|
"epoch": 0.37656903765690375,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7607,
|
|
"mean_token_accuracy": 0.0901703730225563,
|
|
"num_tokens": 828297.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.292386770248413,
|
|
"epoch": 0.3807531380753138,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.807,
|
|
"mean_token_accuracy": 0.08642732873558998,
|
|
"num_tokens": 838268.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.304117202758789,
|
|
"epoch": 0.38493723849372385,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.7244,
|
|
"mean_token_accuracy": 0.09284975379705429,
|
|
"num_tokens": 846491.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.163941526412964,
|
|
"epoch": 0.3891213389121339,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6779,
|
|
"mean_token_accuracy": 0.09622682482004166,
|
|
"num_tokens": 856501.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.226222562789917,
|
|
"epoch": 0.39330543933054396,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7058,
|
|
"mean_token_accuracy": 0.09297544807195664,
|
|
"num_tokens": 867054.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.1907635688781735,
|
|
"epoch": 0.39748953974895396,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7284,
|
|
"mean_token_accuracy": 0.0978313222527504,
|
|
"num_tokens": 875956.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.051689147949219,
|
|
"epoch": 0.401673640167364,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6452,
|
|
"mean_token_accuracy": 0.09968722686171531,
|
|
"num_tokens": 884427.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.282907915115357,
|
|
"epoch": 0.40585774058577406,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7298,
|
|
"mean_token_accuracy": 0.09669127762317657,
|
|
"num_tokens": 894181.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.166344404220581,
|
|
"epoch": 0.4100418410041841,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.5834,
|
|
"mean_token_accuracy": 0.10224275141954423,
|
|
"num_tokens": 904076.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.067501544952393,
|
|
"epoch": 0.41422594142259417,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7161,
|
|
"mean_token_accuracy": 0.09675642251968383,
|
|
"num_tokens": 913915.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.267646312713623,
|
|
"epoch": 0.41841004184100417,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.7693,
|
|
"mean_token_accuracy": 0.0949087992310524,
|
|
"num_tokens": 922876.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.41841004184100417,
|
|
"eval_entropy": 7.11538662354639,
|
|
"eval_loss": 6.712912559509277,
|
|
"eval_mean_token_accuracy": 0.1014816391915997,
|
|
"eval_num_tokens": 922876.0,
|
|
"eval_runtime": 2.2023,
|
|
"eval_samples_per_second": 1774.078,
|
|
"eval_steps_per_second": 222.044,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.1239762783050535,
|
|
"epoch": 0.4225941422594142,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.5674,
|
|
"mean_token_accuracy": 0.10662517622113228,
|
|
"num_tokens": 931998.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.136424398422241,
|
|
"epoch": 0.42677824267782427,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6209,
|
|
"mean_token_accuracy": 0.09857222437858582,
|
|
"num_tokens": 941387.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.055334186553955,
|
|
"epoch": 0.4309623430962343,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6751,
|
|
"mean_token_accuracy": 0.09733710438013077,
|
|
"num_tokens": 951866.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.04624924659729,
|
|
"epoch": 0.4351464435146444,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6011,
|
|
"mean_token_accuracy": 0.09988755211234093,
|
|
"num_tokens": 961252.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.173402547836304,
|
|
"epoch": 0.4393305439330544,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6201,
|
|
"mean_token_accuracy": 0.10153839737176895,
|
|
"num_tokens": 969857.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.089877700805664,
|
|
"epoch": 0.4435146443514644,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6021,
|
|
"mean_token_accuracy": 0.1013123445212841,
|
|
"num_tokens": 978754.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.053047180175781,
|
|
"epoch": 0.4476987447698745,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5926,
|
|
"mean_token_accuracy": 0.09827639311552047,
|
|
"num_tokens": 987935.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.144248104095459,
|
|
"epoch": 0.45188284518828453,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5749,
|
|
"mean_token_accuracy": 0.10194737911224365,
|
|
"num_tokens": 996873.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.1166425228118895,
|
|
"epoch": 0.4560669456066946,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6455,
|
|
"mean_token_accuracy": 0.0955324687063694,
|
|
"num_tokens": 1006487.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.9321996688842775,
|
|
"epoch": 0.4602510460251046,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.4879,
|
|
"mean_token_accuracy": 0.10409051030874253,
|
|
"num_tokens": 1015715.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.044696474075318,
|
|
"epoch": 0.46443514644351463,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.594,
|
|
"mean_token_accuracy": 0.0993474654853344,
|
|
"num_tokens": 1025091.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.029013824462891,
|
|
"epoch": 0.4686192468619247,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5318,
|
|
"mean_token_accuracy": 0.09725542590022088,
|
|
"num_tokens": 1034174.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.087660264968872,
|
|
"epoch": 0.47280334728033474,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.723,
|
|
"mean_token_accuracy": 0.09320119209587574,
|
|
"num_tokens": 1043777.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.013822841644287,
|
|
"epoch": 0.4769874476987448,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.5692,
|
|
"mean_token_accuracy": 0.09590305835008621,
|
|
"num_tokens": 1052749.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.037209701538086,
|
|
"epoch": 0.4811715481171548,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5847,
|
|
"mean_token_accuracy": 0.09973378553986549,
|
|
"num_tokens": 1061564.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.061286687850952,
|
|
"epoch": 0.48535564853556484,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.511,
|
|
"mean_token_accuracy": 0.1009148709475994,
|
|
"num_tokens": 1070006.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.931887674331665,
|
|
"epoch": 0.4895397489539749,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5634,
|
|
"mean_token_accuracy": 0.10082549154758454,
|
|
"num_tokens": 1079799.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.984090995788574,
|
|
"epoch": 0.49372384937238495,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.4057,
|
|
"mean_token_accuracy": 0.10909034311771393,
|
|
"num_tokens": 1088842.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.9044890880584715,
|
|
"epoch": 0.497907949790795,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.4754,
|
|
"mean_token_accuracy": 0.1099147692322731,
|
|
"num_tokens": 1097960.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.92899260520935,
|
|
"epoch": 0.502092050209205,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4544,
|
|
"mean_token_accuracy": 0.10158643499016762,
|
|
"num_tokens": 1106797.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.944618511199951,
|
|
"epoch": 0.5062761506276151,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6024,
|
|
"mean_token_accuracy": 0.10813010036945343,
|
|
"num_tokens": 1116384.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.791468477249145,
|
|
"epoch": 0.5104602510460251,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.3615,
|
|
"mean_token_accuracy": 0.1100485622882843,
|
|
"num_tokens": 1125206.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.007556772232055,
|
|
"epoch": 0.5146443514644351,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5881,
|
|
"mean_token_accuracy": 0.10688784271478653,
|
|
"num_tokens": 1134110.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.0638329029083256,
|
|
"epoch": 0.5188284518828452,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5964,
|
|
"mean_token_accuracy": 0.10027675852179527,
|
|
"num_tokens": 1142833.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.943243360519409,
|
|
"epoch": 0.5230125523012552,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5036,
|
|
"mean_token_accuracy": 0.10669585466384887,
|
|
"num_tokens": 1152006.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.8594810485839846,
|
|
"epoch": 0.5271966527196653,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4453,
|
|
"mean_token_accuracy": 0.10346821099519729,
|
|
"num_tokens": 1161061.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.976174068450928,
|
|
"epoch": 0.5313807531380753,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.4658,
|
|
"mean_token_accuracy": 0.10728110820055008,
|
|
"num_tokens": 1170754.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.819682836532593,
|
|
"epoch": 0.5355648535564853,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4015,
|
|
"mean_token_accuracy": 0.10926354303956032,
|
|
"num_tokens": 1180854.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.861003255844116,
|
|
"epoch": 0.5397489539748954,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4891,
|
|
"mean_token_accuracy": 0.10695577412843704,
|
|
"num_tokens": 1190567.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.8071857452392575,
|
|
"epoch": 0.5439330543933054,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.3242,
|
|
"mean_token_accuracy": 0.11723242327570915,
|
|
"num_tokens": 1200247.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.8493428230285645,
|
|
"epoch": 0.5481171548117155,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4327,
|
|
"mean_token_accuracy": 0.11054310277104377,
|
|
"num_tokens": 1209358.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.8216430187225345,
|
|
"epoch": 0.5523012552301255,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4453,
|
|
"mean_token_accuracy": 0.10909160152077675,
|
|
"num_tokens": 1217746.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.859017038345337,
|
|
"epoch": 0.5564853556485355,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4093,
|
|
"mean_token_accuracy": 0.10931532606482505,
|
|
"num_tokens": 1226829.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.737533140182495,
|
|
"epoch": 0.5606694560669456,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.3343,
|
|
"mean_token_accuracy": 0.11464144736528396,
|
|
"num_tokens": 1236642.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.9131103515625,
|
|
"epoch": 0.5648535564853556,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4525,
|
|
"mean_token_accuracy": 0.11251221597194672,
|
|
"num_tokens": 1244934.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.66797456741333,
|
|
"epoch": 0.5690376569037657,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3449,
|
|
"mean_token_accuracy": 0.10475735813379287,
|
|
"num_tokens": 1254786.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.805626964569091,
|
|
"epoch": 0.5732217573221757,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.2942,
|
|
"mean_token_accuracy": 0.11413875743746757,
|
|
"num_tokens": 1263745.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.739336395263672,
|
|
"epoch": 0.5774058577405857,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4462,
|
|
"mean_token_accuracy": 0.11418344005942345,
|
|
"num_tokens": 1273101.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.797475862503052,
|
|
"epoch": 0.5815899581589958,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.426,
|
|
"mean_token_accuracy": 0.11143595725297928,
|
|
"num_tokens": 1282547.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.823879671096802,
|
|
"epoch": 0.5857740585774058,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.44,
|
|
"mean_token_accuracy": 0.11051822453737259,
|
|
"num_tokens": 1291594.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.764584350585937,
|
|
"epoch": 0.5899581589958159,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3654,
|
|
"mean_token_accuracy": 0.11968467384576797,
|
|
"num_tokens": 1300282.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.800668144226075,
|
|
"epoch": 0.5941422594142259,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4424,
|
|
"mean_token_accuracy": 0.11207071244716645,
|
|
"num_tokens": 1309304.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.732267570495606,
|
|
"epoch": 0.5983263598326359,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.2876,
|
|
"mean_token_accuracy": 0.11646409556269646,
|
|
"num_tokens": 1317638.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.840695858001709,
|
|
"epoch": 0.602510460251046,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4793,
|
|
"mean_token_accuracy": 0.11017750427126885,
|
|
"num_tokens": 1327242.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.755548000335693,
|
|
"epoch": 0.606694560669456,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.2918,
|
|
"mean_token_accuracy": 0.11189166605472564,
|
|
"num_tokens": 1335880.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.715355634689331,
|
|
"epoch": 0.6108786610878661,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3689,
|
|
"mean_token_accuracy": 0.11269519478082657,
|
|
"num_tokens": 1344909.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.728858232498169,
|
|
"epoch": 0.6150627615062761,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.2341,
|
|
"mean_token_accuracy": 0.12015092521905898,
|
|
"num_tokens": 1353997.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.60258641242981,
|
|
"epoch": 0.6192468619246861,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3813,
|
|
"mean_token_accuracy": 0.11518162414431572,
|
|
"num_tokens": 1363293.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.935403633117676,
|
|
"epoch": 0.6234309623430963,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4142,
|
|
"mean_token_accuracy": 0.10975181236863137,
|
|
"num_tokens": 1372174.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.762114429473877,
|
|
"epoch": 0.6276150627615062,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3011,
|
|
"mean_token_accuracy": 0.11669646203517914,
|
|
"num_tokens": 1381067.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.603954315185547,
|
|
"epoch": 0.6317991631799164,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.2813,
|
|
"mean_token_accuracy": 0.12004445940256118,
|
|
"num_tokens": 1390072.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.682461071014404,
|
|
"epoch": 0.6359832635983264,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.275,
|
|
"mean_token_accuracy": 0.11979681104421616,
|
|
"num_tokens": 1399391.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.677716875076294,
|
|
"epoch": 0.6401673640167364,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.2877,
|
|
"mean_token_accuracy": 0.11631687059998512,
|
|
"num_tokens": 1408054.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.715818881988525,
|
|
"epoch": 0.6443514644351465,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.2935,
|
|
"mean_token_accuracy": 0.1101489320397377,
|
|
"num_tokens": 1417406.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.619862651824951,
|
|
"epoch": 0.6485355648535565,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3465,
|
|
"mean_token_accuracy": 0.11427735313773155,
|
|
"num_tokens": 1426907.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.676693296432495,
|
|
"epoch": 0.6527196652719666,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3893,
|
|
"mean_token_accuracy": 0.11339468657970428,
|
|
"num_tokens": 1436085.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.744153213500977,
|
|
"epoch": 0.6569037656903766,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3009,
|
|
"mean_token_accuracy": 0.11440082043409347,
|
|
"num_tokens": 1444823.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.730417633056641,
|
|
"epoch": 0.6610878661087866,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.346,
|
|
"mean_token_accuracy": 0.11472340002655983,
|
|
"num_tokens": 1454621.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.561496686935425,
|
|
"epoch": 0.6652719665271967,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.2129,
|
|
"mean_token_accuracy": 0.12525596842169762,
|
|
"num_tokens": 1463784.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.660541248321533,
|
|
"epoch": 0.6694560669456067,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3461,
|
|
"mean_token_accuracy": 0.1180668905377388,
|
|
"num_tokens": 1473257.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.669353771209717,
|
|
"epoch": 0.6736401673640168,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2746,
|
|
"mean_token_accuracy": 0.12255375310778618,
|
|
"num_tokens": 1482195.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.5357430934906,
|
|
"epoch": 0.6778242677824268,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3121,
|
|
"mean_token_accuracy": 0.12164309471845627,
|
|
"num_tokens": 1491590.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.670521688461304,
|
|
"epoch": 0.6820083682008368,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.1965,
|
|
"mean_token_accuracy": 0.13060073927044868,
|
|
"num_tokens": 1500208.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.629897880554199,
|
|
"epoch": 0.6861924686192469,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3233,
|
|
"mean_token_accuracy": 0.11592838019132615,
|
|
"num_tokens": 1509412.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.682983350753784,
|
|
"epoch": 0.6903765690376569,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3182,
|
|
"mean_token_accuracy": 0.11884461417794227,
|
|
"num_tokens": 1518426.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.658360767364502,
|
|
"epoch": 0.694560669456067,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.2727,
|
|
"mean_token_accuracy": 0.11531517952680588,
|
|
"num_tokens": 1528605.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.565955686569214,
|
|
"epoch": 0.698744769874477,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.2431,
|
|
"mean_token_accuracy": 0.1226552739739418,
|
|
"num_tokens": 1537449.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.601708698272705,
|
|
"epoch": 0.702928870292887,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.2503,
|
|
"mean_token_accuracy": 0.12718333601951598,
|
|
"num_tokens": 1546743.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.531602811813355,
|
|
"epoch": 0.7071129707112971,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2263,
|
|
"mean_token_accuracy": 0.11939926072955132,
|
|
"num_tokens": 1555691.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.538215208053589,
|
|
"epoch": 0.7112970711297071,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.205,
|
|
"mean_token_accuracy": 0.12462374866008759,
|
|
"num_tokens": 1564957.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.598930406570434,
|
|
"epoch": 0.7154811715481172,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.249,
|
|
"mean_token_accuracy": 0.12943542525172233,
|
|
"num_tokens": 1573901.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.556529903411866,
|
|
"epoch": 0.7196652719665272,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.2396,
|
|
"mean_token_accuracy": 0.12566354051232337,
|
|
"num_tokens": 1582644.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.675654125213623,
|
|
"epoch": 0.7238493723849372,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3693,
|
|
"mean_token_accuracy": 0.11816548332571983,
|
|
"num_tokens": 1592363.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.555848836898804,
|
|
"epoch": 0.7280334728033473,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3051,
|
|
"mean_token_accuracy": 0.12256788238883018,
|
|
"num_tokens": 1602012.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.5441961765289305,
|
|
"epoch": 0.7322175732217573,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.1873,
|
|
"mean_token_accuracy": 0.13024157509207726,
|
|
"num_tokens": 1611625.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.430695104598999,
|
|
"epoch": 0.7364016736401674,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.1534,
|
|
"mean_token_accuracy": 0.1217527948319912,
|
|
"num_tokens": 1620941.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.605828762054443,
|
|
"epoch": 0.7405857740585774,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2518,
|
|
"mean_token_accuracy": 0.12294901311397552,
|
|
"num_tokens": 1629225.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.559442949295044,
|
|
"epoch": 0.7447698744769874,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2177,
|
|
"mean_token_accuracy": 0.12158126160502433,
|
|
"num_tokens": 1639551.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.45328106880188,
|
|
"epoch": 0.7489539748953975,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.2022,
|
|
"mean_token_accuracy": 0.119484893232584,
|
|
"num_tokens": 1648963.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.53769154548645,
|
|
"epoch": 0.7531380753138075,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.1257,
|
|
"mean_token_accuracy": 0.1254202738404274,
|
|
"num_tokens": 1657735.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.590608787536621,
|
|
"epoch": 0.7573221757322176,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.1737,
|
|
"mean_token_accuracy": 0.12596495375037192,
|
|
"num_tokens": 1666349.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.4066187858581545,
|
|
"epoch": 0.7615062761506276,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.0935,
|
|
"mean_token_accuracy": 0.1287170298397541,
|
|
"num_tokens": 1675920.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.37900972366333,
|
|
"epoch": 0.7656903765690377,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.1339,
|
|
"mean_token_accuracy": 0.13005387112498284,
|
|
"num_tokens": 1685626.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.5884452819824215,
|
|
"epoch": 0.7698744769874477,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2512,
|
|
"mean_token_accuracy": 0.13180154114961623,
|
|
"num_tokens": 1694972.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.445847797393799,
|
|
"epoch": 0.7740585774058577,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.1521,
|
|
"mean_token_accuracy": 0.1360354334115982,
|
|
"num_tokens": 1703806.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.4568705558776855,
|
|
"epoch": 0.7782426778242678,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.2401,
|
|
"mean_token_accuracy": 0.1237366445362568,
|
|
"num_tokens": 1712949.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.652240514755249,
|
|
"epoch": 0.7824267782426778,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3513,
|
|
"mean_token_accuracy": 0.12628775238990783,
|
|
"num_tokens": 1722328.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.620529079437256,
|
|
"epoch": 0.7866108786610879,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3163,
|
|
"mean_token_accuracy": 0.12532784938812255,
|
|
"num_tokens": 1731880.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.333013391494751,
|
|
"epoch": 0.7907949790794979,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.1335,
|
|
"mean_token_accuracy": 0.13126946836709977,
|
|
"num_tokens": 1740719.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.458236312866211,
|
|
"epoch": 0.7949790794979079,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1752,
|
|
"mean_token_accuracy": 0.12758501768112182,
|
|
"num_tokens": 1749753.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.5069516658782955,
|
|
"epoch": 0.799163179916318,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.203,
|
|
"mean_token_accuracy": 0.12178897708654404,
|
|
"num_tokens": 1758913.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.565327882766724,
|
|
"epoch": 0.803347280334728,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.2033,
|
|
"mean_token_accuracy": 0.12567698061466218,
|
|
"num_tokens": 1768536.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.401819515228271,
|
|
"epoch": 0.8075313807531381,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.134,
|
|
"mean_token_accuracy": 0.13032307624816894,
|
|
"num_tokens": 1777291.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.449258661270141,
|
|
"epoch": 0.8117154811715481,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2381,
|
|
"mean_token_accuracy": 0.12935445234179496,
|
|
"num_tokens": 1787186.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.481316804885864,
|
|
"epoch": 0.8158995815899581,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.0231,
|
|
"mean_token_accuracy": 0.13392751142382622,
|
|
"num_tokens": 1795581.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.272410821914673,
|
|
"epoch": 0.8200836820083682,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.0317,
|
|
"mean_token_accuracy": 0.1353148899972439,
|
|
"num_tokens": 1804627.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.478474712371826,
|
|
"epoch": 0.8242677824267782,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.1604,
|
|
"mean_token_accuracy": 0.130138298869133,
|
|
"num_tokens": 1813808.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.44154839515686,
|
|
"epoch": 0.8284518828451883,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1622,
|
|
"mean_token_accuracy": 0.13280925750732422,
|
|
"num_tokens": 1823422.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.431913900375366,
|
|
"epoch": 0.8326359832635983,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1945,
|
|
"mean_token_accuracy": 0.12693543881177902,
|
|
"num_tokens": 1832437.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.368013191223144,
|
|
"epoch": 0.8368200836820083,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.1169,
|
|
"mean_token_accuracy": 0.12685068920254708,
|
|
"num_tokens": 1842063.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8368200836820083,
|
|
"eval_entropy": 6.238058360320172,
|
|
"eval_loss": 6.159273147583008,
|
|
"eval_mean_token_accuracy": 0.13784902068978683,
|
|
"eval_num_tokens": 1842063.0,
|
|
"eval_runtime": 2.2943,
|
|
"eval_samples_per_second": 1702.945,
|
|
"eval_steps_per_second": 213.141,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.464809846878052,
|
|
"epoch": 0.8410041841004184,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999998518355666,
|
|
"loss": 6.2652,
|
|
"mean_token_accuracy": 0.12681940272450448,
|
|
"num_tokens": 1851043.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.393350315093994,
|
|
"epoch": 0.8451882845188284,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999992499178905,
|
|
"loss": 6.1087,
|
|
"mean_token_accuracy": 0.132909045368433,
|
|
"num_tokens": 1860268.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.424519777297974,
|
|
"epoch": 0.8493723849372385,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999981849879322,
|
|
"loss": 6.21,
|
|
"mean_token_accuracy": 0.1311655230820179,
|
|
"num_tokens": 1870096.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.392327547073364,
|
|
"epoch": 0.8535564853556485,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999966570478833,
|
|
"loss": 6.1017,
|
|
"mean_token_accuracy": 0.1333017997443676,
|
|
"num_tokens": 1879164.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.3620843410491945,
|
|
"epoch": 0.8577405857740585,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004999946661008878,
|
|
"loss": 6.0509,
|
|
"mean_token_accuracy": 0.13429286554455758,
|
|
"num_tokens": 1888135.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.377326822280883,
|
|
"epoch": 0.8619246861924686,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000499992212151043,
|
|
"loss": 6.1136,
|
|
"mean_token_accuracy": 0.1323494702577591,
|
|
"num_tokens": 1898302.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.301321077346802,
|
|
"epoch": 0.8661087866108786,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999892952033987,
|
|
"loss": 6.088,
|
|
"mean_token_accuracy": 0.13468799963593484,
|
|
"num_tokens": 1907753.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.368103408813477,
|
|
"epoch": 0.8702928870292888,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999859152639573,
|
|
"loss": 6.1351,
|
|
"mean_token_accuracy": 0.13460114374756812,
|
|
"num_tokens": 1917543.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.453844499588013,
|
|
"epoch": 0.8744769874476988,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999820723396745,
|
|
"loss": 6.1638,
|
|
"mean_token_accuracy": 0.12692686915397644,
|
|
"num_tokens": 1927335.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.368426513671875,
|
|
"epoch": 0.8786610878661087,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999777664384581,
|
|
"loss": 6.09,
|
|
"mean_token_accuracy": 0.13307663276791573,
|
|
"num_tokens": 1936381.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.361155986785889,
|
|
"epoch": 0.8828451882845189,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999729975691692,
|
|
"loss": 6.1289,
|
|
"mean_token_accuracy": 0.1341361276805401,
|
|
"num_tokens": 1945559.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.2844610691070555,
|
|
"epoch": 0.8870292887029289,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999677657416212,
|
|
"loss": 6.0049,
|
|
"mean_token_accuracy": 0.12961344420909882,
|
|
"num_tokens": 1954709.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.365634632110596,
|
|
"epoch": 0.891213389121339,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999620709665804,
|
|
"loss": 6.1268,
|
|
"mean_token_accuracy": 0.12884532064199447,
|
|
"num_tokens": 1963664.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.418537521362305,
|
|
"epoch": 0.895397489539749,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999559132557659,
|
|
"loss": 6.197,
|
|
"mean_token_accuracy": 0.12948742732405663,
|
|
"num_tokens": 1974103.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.4297325134277346,
|
|
"epoch": 0.899581589958159,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004999492926218489,
|
|
"loss": 6.0622,
|
|
"mean_token_accuracy": 0.13383897319436072,
|
|
"num_tokens": 1982924.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.259031009674072,
|
|
"epoch": 0.9037656903765691,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499942209078454,
|
|
"loss": 5.9876,
|
|
"mean_token_accuracy": 0.1435515210032463,
|
|
"num_tokens": 1992033.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.2822408199310305,
|
|
"epoch": 0.9079497907949791,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999346626401578,
|
|
"loss": 6.0026,
|
|
"mean_token_accuracy": 0.13485121428966523,
|
|
"num_tokens": 2001218.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.231379890441895,
|
|
"epoch": 0.9121338912133892,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999266533224897,
|
|
"loss": 6.1395,
|
|
"mean_token_accuracy": 0.1337206944823265,
|
|
"num_tokens": 2010463.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.359997653961182,
|
|
"epoch": 0.9163179916317992,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004999181811419314,
|
|
"loss": 6.0216,
|
|
"mean_token_accuracy": 0.1345188781619072,
|
|
"num_tokens": 2019246.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.34965934753418,
|
|
"epoch": 0.9205020920502092,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999092461159177,
|
|
"loss": 6.0877,
|
|
"mean_token_accuracy": 0.13710115626454353,
|
|
"num_tokens": 2028083.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.267071962356567,
|
|
"epoch": 0.9246861924686193,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998998482628351,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.14412403106689453,
|
|
"num_tokens": 2036631.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.291199445724487,
|
|
"epoch": 0.9288702928870293,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004998899876020229,
|
|
"loss": 6.0328,
|
|
"mean_token_accuracy": 0.13799802884459494,
|
|
"num_tokens": 2045910.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.214427423477173,
|
|
"epoch": 0.9330543933054394,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000499879664153773,
|
|
"loss": 6.0909,
|
|
"mean_token_accuracy": 0.14138534739613534,
|
|
"num_tokens": 2055959.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.441401672363281,
|
|
"epoch": 0.9372384937238494,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998688779393291,
|
|
"loss": 6.1434,
|
|
"mean_token_accuracy": 0.13137228786945343,
|
|
"num_tokens": 2065475.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.153232002258301,
|
|
"epoch": 0.9414225941422594,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998576289808877,
|
|
"loss": 5.9181,
|
|
"mean_token_accuracy": 0.1401391714811325,
|
|
"num_tokens": 2074469.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.235647678375244,
|
|
"epoch": 0.9456066945606695,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998459173015972,
|
|
"loss": 6.1075,
|
|
"mean_token_accuracy": 0.13486189842224122,
|
|
"num_tokens": 2084604.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.3427599430084225,
|
|
"epoch": 0.9497907949790795,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004998337429255584,
|
|
"loss": 6.0563,
|
|
"mean_token_accuracy": 0.13504815176129342,
|
|
"num_tokens": 2093377.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.360136318206787,
|
|
"epoch": 0.9539748953974896,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998211058778243,
|
|
"loss": 6.0866,
|
|
"mean_token_accuracy": 0.138973431289196,
|
|
"num_tokens": 2102798.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.287752342224121,
|
|
"epoch": 0.9581589958158996,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998080061843999,
|
|
"loss": 6.0545,
|
|
"mean_token_accuracy": 0.13773418068885804,
|
|
"num_tokens": 2112516.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.360985136032104,
|
|
"epoch": 0.9623430962343096,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499794443872242,
|
|
"loss": 6.1424,
|
|
"mean_token_accuracy": 0.13592344745993615,
|
|
"num_tokens": 2121677.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.212178564071655,
|
|
"epoch": 0.9665271966527197,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997804189692601,
|
|
"loss": 5.973,
|
|
"mean_token_accuracy": 0.1397504687309265,
|
|
"num_tokens": 2131220.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.333057689666748,
|
|
"epoch": 0.9707112970711297,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004997659315043149,
|
|
"loss": 6.1291,
|
|
"mean_token_accuracy": 0.1358908236026764,
|
|
"num_tokens": 2142104.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.252610683441162,
|
|
"epoch": 0.9748953974895398,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997509815072191,
|
|
"loss": 6.0435,
|
|
"mean_token_accuracy": 0.13990800082683563,
|
|
"num_tokens": 2151970.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.158507537841797,
|
|
"epoch": 0.9790794979079498,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997355690087379,
|
|
"loss": 6.0009,
|
|
"mean_token_accuracy": 0.1391149267554283,
|
|
"num_tokens": 2161371.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.218712329864502,
|
|
"epoch": 0.9832635983263598,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997196940405874,
|
|
"loss": 5.9556,
|
|
"mean_token_accuracy": 0.13924067318439484,
|
|
"num_tokens": 2170650.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.2072021484375,
|
|
"epoch": 0.9874476987447699,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997033566354358,
|
|
"loss": 6.0183,
|
|
"mean_token_accuracy": 0.14336950927972794,
|
|
"num_tokens": 2180934.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.268893623352051,
|
|
"epoch": 0.9916317991631799,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996865568269028,
|
|
"loss": 5.9877,
|
|
"mean_token_accuracy": 0.14385268539190293,
|
|
"num_tokens": 2189894.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.245169544219971,
|
|
"epoch": 0.99581589958159,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996692946495597,
|
|
"loss": 5.9351,
|
|
"mean_token_accuracy": 0.15099522173404695,
|
|
"num_tokens": 2200208.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.212211990356446,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004996515701389295,
|
|
"loss": 5.983,
|
|
"mean_token_accuracy": 0.14296597242355347,
|
|
"num_tokens": 2209322.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.20455527305603,
|
|
"epoch": 1.00418410041841,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004996333833314862,
|
|
"loss": 5.8156,
|
|
"mean_token_accuracy": 0.1556951105594635,
|
|
"num_tokens": 2217663.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.1169328689575195,
|
|
"epoch": 1.00836820083682,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004996147342646554,
|
|
"loss": 5.8612,
|
|
"mean_token_accuracy": 0.14581794887781144,
|
|
"num_tokens": 2226559.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.304083585739136,
|
|
"epoch": 1.0125523012552302,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995956229768137,
|
|
"loss": 5.9036,
|
|
"mean_token_accuracy": 0.1477184236049652,
|
|
"num_tokens": 2235468.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.0922456741333,
|
|
"epoch": 1.0167364016736402,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995760495072893,
|
|
"loss": 5.7136,
|
|
"mean_token_accuracy": 0.15675223767757415,
|
|
"num_tokens": 2245176.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.186470460891724,
|
|
"epoch": 1.0209205020920502,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995560138963611,
|
|
"loss": 5.8704,
|
|
"mean_token_accuracy": 0.1509818285703659,
|
|
"num_tokens": 2255123.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.174693536758423,
|
|
"epoch": 1.0251046025104602,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995355161852593,
|
|
"loss": 5.7899,
|
|
"mean_token_accuracy": 0.14881661236286164,
|
|
"num_tokens": 2263596.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.165000247955322,
|
|
"epoch": 1.0292887029288702,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004995145564161648,
|
|
"loss": 5.8368,
|
|
"mean_token_accuracy": 0.1497395634651184,
|
|
"num_tokens": 2272228.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.101798677444458,
|
|
"epoch": 1.0334728033472804,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004994931346322094,
|
|
"loss": 5.8661,
|
|
"mean_token_accuracy": 0.14979260563850402,
|
|
"num_tokens": 2281248.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.252989959716797,
|
|
"epoch": 1.0376569037656904,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994712508774757,
|
|
"loss": 5.8421,
|
|
"mean_token_accuracy": 0.14638827219605446,
|
|
"num_tokens": 2290958.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.087033700942993,
|
|
"epoch": 1.0418410041841004,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499448905196997,
|
|
"loss": 5.7917,
|
|
"mean_token_accuracy": 0.15143862515687942,
|
|
"num_tokens": 2299763.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.1420900344848635,
|
|
"epoch": 1.0460251046025104,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004994260976367573,
|
|
"loss": 5.7923,
|
|
"mean_token_accuracy": 0.15185852721333504,
|
|
"num_tokens": 2309847.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.264729356765747,
|
|
"epoch": 1.0502092050209204,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004994028282436905,
|
|
"loss": 5.832,
|
|
"mean_token_accuracy": 0.14682750254869462,
|
|
"num_tokens": 2317556.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 5.994171810150147,
|
|
"epoch": 1.0543933054393306,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004993790970656814,
|
|
"loss": 5.7699,
|
|
"mean_token_accuracy": 0.15092798620462416,
|
|
"num_tokens": 2327151.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.134915065765381,
|
|
"epoch": 1.0585774058577406,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993549041515652,
|
|
"loss": 5.765,
|
|
"mean_token_accuracy": 0.15306467264890672,
|
|
"num_tokens": 2336119.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.040839529037475,
|
|
"epoch": 1.0627615062761506,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004993302495511268,
|
|
"loss": 5.7985,
|
|
"mean_token_accuracy": 0.14950646013021468,
|
|
"num_tokens": 2345262.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.240773630142212,
|
|
"epoch": 1.0669456066945606,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004993051333151014,
|
|
"loss": 5.9126,
|
|
"mean_token_accuracy": 0.1474372297525406,
|
|
"num_tokens": 2355210.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.9994574069976805,
|
|
"epoch": 1.0711297071129706,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004992795554951742,
|
|
"loss": 5.6565,
|
|
"mean_token_accuracy": 0.1536034032702446,
|
|
"num_tokens": 2364599.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.988495969772339,
|
|
"epoch": 1.0753138075313808,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004992535161439803,
|
|
"loss": 5.8322,
|
|
"mean_token_accuracy": 0.14308930411934853,
|
|
"num_tokens": 2374143.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.214874219894409,
|
|
"epoch": 1.0794979079497908,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004992270153151045,
|
|
"loss": 5.8352,
|
|
"mean_token_accuracy": 0.15135148167610168,
|
|
"num_tokens": 2383155.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.2102819919586185,
|
|
"epoch": 1.0836820083682008,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004992000530630812,
|
|
"loss": 5.8767,
|
|
"mean_token_accuracy": 0.1463464692234993,
|
|
"num_tokens": 2392542.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.0049083709716795,
|
|
"epoch": 1.0878661087866108,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004991726294433943,
|
|
"loss": 5.8446,
|
|
"mean_token_accuracy": 0.14662079960107804,
|
|
"num_tokens": 2401851.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.139134263992309,
|
|
"epoch": 1.0920502092050208,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991447445124774,
|
|
"loss": 5.7869,
|
|
"mean_token_accuracy": 0.15453603789210318,
|
|
"num_tokens": 2410726.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.995836925506592,
|
|
"epoch": 1.096234309623431,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499116398327713,
|
|
"loss": 5.8423,
|
|
"mean_token_accuracy": 0.1490214928984642,
|
|
"num_tokens": 2420702.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.025252819061279,
|
|
"epoch": 1.100418410041841,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004990875909474331,
|
|
"loss": 5.7679,
|
|
"mean_token_accuracy": 0.1548636332154274,
|
|
"num_tokens": 2431335.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.179278898239136,
|
|
"epoch": 1.104602510460251,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004990583224309187,
|
|
"loss": 5.7535,
|
|
"mean_token_accuracy": 0.15248486995697022,
|
|
"num_tokens": 2440722.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.034488916397095,
|
|
"epoch": 1.108786610878661,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004990285928383997,
|
|
"loss": 5.8157,
|
|
"mean_token_accuracy": 0.14678085148334502,
|
|
"num_tokens": 2451230.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.128333854675293,
|
|
"epoch": 1.112970711297071,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004989984022310545,
|
|
"loss": 5.8562,
|
|
"mean_token_accuracy": 0.1468272678554058,
|
|
"num_tokens": 2461442.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.024012660980224,
|
|
"epoch": 1.1171548117154813,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004989677506710109,
|
|
"loss": 5.6766,
|
|
"mean_token_accuracy": 0.15811667293310167,
|
|
"num_tokens": 2471381.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.016521263122558,
|
|
"epoch": 1.1213389121338913,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004989366382213449,
|
|
"loss": 5.7413,
|
|
"mean_token_accuracy": 0.15749377608299256,
|
|
"num_tokens": 2479937.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.027074670791626,
|
|
"epoch": 1.1255230125523012,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004989050649460806,
|
|
"loss": 5.6967,
|
|
"mean_token_accuracy": 0.16208242624998093,
|
|
"num_tokens": 2488974.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.028843021392822,
|
|
"epoch": 1.1297071129707112,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000498873030910191,
|
|
"loss": 5.8483,
|
|
"mean_token_accuracy": 0.15223534554243087,
|
|
"num_tokens": 2498800.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.032519721984864,
|
|
"epoch": 1.1338912133891212,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004988405361795969,
|
|
"loss": 5.745,
|
|
"mean_token_accuracy": 0.14807292744517325,
|
|
"num_tokens": 2506580.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.036730861663818,
|
|
"epoch": 1.1380753138075315,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004988075808211672,
|
|
"loss": 5.7898,
|
|
"mean_token_accuracy": 0.15643058717250824,
|
|
"num_tokens": 2516342.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.992047214508057,
|
|
"epoch": 1.1422594142259415,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004987741649027189,
|
|
"loss": 5.8207,
|
|
"mean_token_accuracy": 0.151021283864975,
|
|
"num_tokens": 2526054.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.131092119216919,
|
|
"epoch": 1.1464435146443515,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004987402884930163,
|
|
"loss": 5.7938,
|
|
"mean_token_accuracy": 0.15176812559366226,
|
|
"num_tokens": 2535556.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.024015760421753,
|
|
"epoch": 1.1506276150627615,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004987059516617717,
|
|
"loss": 5.748,
|
|
"mean_token_accuracy": 0.15519688725471498,
|
|
"num_tokens": 2544930.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.946213817596435,
|
|
"epoch": 1.1548117154811715,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004986711544796449,
|
|
"loss": 5.6594,
|
|
"mean_token_accuracy": 0.15817881226539612,
|
|
"num_tokens": 2554362.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.102620697021484,
|
|
"epoch": 1.1589958158995817,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000498635897018243,
|
|
"loss": 5.793,
|
|
"mean_token_accuracy": 0.15520063936710357,
|
|
"num_tokens": 2563499.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.995561599731445,
|
|
"epoch": 1.1631799163179917,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004986001793501198,
|
|
"loss": 5.7952,
|
|
"mean_token_accuracy": 0.15294883847236634,
|
|
"num_tokens": 2573346.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.056002807617188,
|
|
"epoch": 1.1673640167364017,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004985640015487768,
|
|
"loss": 5.6841,
|
|
"mean_token_accuracy": 0.16190987080335617,
|
|
"num_tokens": 2582541.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.990554094314575,
|
|
"epoch": 1.1715481171548117,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004985273636886621,
|
|
"loss": 5.7632,
|
|
"mean_token_accuracy": 0.15325954258441926,
|
|
"num_tokens": 2592015.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.007609701156616,
|
|
"epoch": 1.1757322175732217,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004984902658451704,
|
|
"loss": 5.7661,
|
|
"mean_token_accuracy": 0.15296930223703384,
|
|
"num_tokens": 2601998.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.997886657714844,
|
|
"epoch": 1.1799163179916319,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004984527080946432,
|
|
"loss": 5.7048,
|
|
"mean_token_accuracy": 0.15458611100912095,
|
|
"num_tokens": 2611435.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.04601354598999,
|
|
"epoch": 1.1841004184100419,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004984146905143683,
|
|
"loss": 5.8114,
|
|
"mean_token_accuracy": 0.1506097808480263,
|
|
"num_tokens": 2621639.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.9846789836883545,
|
|
"epoch": 1.1882845188284519,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004983762131825797,
|
|
"loss": 5.7673,
|
|
"mean_token_accuracy": 0.1566401556134224,
|
|
"num_tokens": 2631390.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.996498966217041,
|
|
"epoch": 1.1924686192468619,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004983372761784576,
|
|
"loss": 5.8091,
|
|
"mean_token_accuracy": 0.15009990185499192,
|
|
"num_tokens": 2640769.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.104054498672485,
|
|
"epoch": 1.196652719665272,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004982978795821281,
|
|
"loss": 5.7988,
|
|
"mean_token_accuracy": 0.1566852107644081,
|
|
"num_tokens": 2651137.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.007593393325806,
|
|
"epoch": 1.200836820083682,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004982580234746632,
|
|
"loss": 5.7281,
|
|
"mean_token_accuracy": 0.1551616907119751,
|
|
"num_tokens": 2660530.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.992457914352417,
|
|
"epoch": 1.205020920502092,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00049821770793808,
|
|
"loss": 5.7829,
|
|
"mean_token_accuracy": 0.15866402983665467,
|
|
"num_tokens": 2669541.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.059764337539673,
|
|
"epoch": 1.209205020920502,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004981769330553418,
|
|
"loss": 5.7885,
|
|
"mean_token_accuracy": 0.15403764247894286,
|
|
"num_tokens": 2678542.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.0553265571594235,
|
|
"epoch": 1.213389121338912,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004981356989103566,
|
|
"loss": 5.8573,
|
|
"mean_token_accuracy": 0.15071797817945481,
|
|
"num_tokens": 2687720.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.043039131164551,
|
|
"epoch": 1.217573221757322,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004980940055879774,
|
|
"loss": 5.7179,
|
|
"mean_token_accuracy": 0.15467721819877625,
|
|
"num_tokens": 2697641.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.938983011245727,
|
|
"epoch": 1.2217573221757323,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004980518531740027,
|
|
"loss": 5.7585,
|
|
"mean_token_accuracy": 0.1521919548511505,
|
|
"num_tokens": 2706600.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.130980491638184,
|
|
"epoch": 1.2259414225941423,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004980092417551752,
|
|
"loss": 5.8652,
|
|
"mean_token_accuracy": 0.15375789552927016,
|
|
"num_tokens": 2716633.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.058181619644165,
|
|
"epoch": 1.2301255230125523,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004979661714191825,
|
|
"loss": 5.6528,
|
|
"mean_token_accuracy": 0.16785621494054795,
|
|
"num_tokens": 2724829.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.955528211593628,
|
|
"epoch": 1.2343096234309623,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004979226422546563,
|
|
"loss": 5.7665,
|
|
"mean_token_accuracy": 0.15509923845529555,
|
|
"num_tokens": 2734276.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.022475242614746,
|
|
"epoch": 1.2384937238493725,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004978786543511726,
|
|
"loss": 5.8575,
|
|
"mean_token_accuracy": 0.1525488644838333,
|
|
"num_tokens": 2744413.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.051792001724243,
|
|
"epoch": 1.2426778242677825,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004978342077992515,
|
|
"loss": 5.7188,
|
|
"mean_token_accuracy": 0.15481539070606232,
|
|
"num_tokens": 2753242.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.854583692550659,
|
|
"epoch": 1.2468619246861925,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004977893026903568,
|
|
"loss": 5.6617,
|
|
"mean_token_accuracy": 0.16461504697799684,
|
|
"num_tokens": 2761811.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.038715505599976,
|
|
"epoch": 1.2510460251046025,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004977439391168963,
|
|
"loss": 5.7574,
|
|
"mean_token_accuracy": 0.1558422937989235,
|
|
"num_tokens": 2770560.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.059652519226074,
|
|
"epoch": 1.2552301255230125,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004976981171722205,
|
|
"loss": 5.7597,
|
|
"mean_token_accuracy": 0.15547596663236618,
|
|
"num_tokens": 2780131.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2552301255230125,
|
|
"eval_entropy": 5.79090596708052,
|
|
"eval_loss": 5.865025043487549,
|
|
"eval_mean_token_accuracy": 0.157731630715612,
|
|
"eval_num_tokens": 2780131.0,
|
|
"eval_runtime": 2.0874,
|
|
"eval_samples_per_second": 1871.74,
|
|
"eval_steps_per_second": 234.267,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.095642948150635,
|
|
"epoch": 1.2594142259414225,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976518369506237,
|
|
"loss": 5.876,
|
|
"mean_token_accuracy": 0.14376280680298806,
|
|
"num_tokens": 2789319.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.941917848587036,
|
|
"epoch": 1.2635983263598327,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004976050985473433,
|
|
"loss": 5.7256,
|
|
"mean_token_accuracy": 0.15156481564044952,
|
|
"num_tokens": 2799253.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.023426866531372,
|
|
"epoch": 1.2677824267782427,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004975579020585593,
|
|
"loss": 5.8262,
|
|
"mean_token_accuracy": 0.15744690522551535,
|
|
"num_tokens": 2808968.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.0533411502838135,
|
|
"epoch": 1.2719665271966527,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004975102475813947,
|
|
"loss": 5.7242,
|
|
"mean_token_accuracy": 0.15870407223701477,
|
|
"num_tokens": 2817282.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.0113911628723145,
|
|
"epoch": 1.2761506276150627,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004974621352139144,
|
|
"loss": 5.7821,
|
|
"mean_token_accuracy": 0.15213275998830794,
|
|
"num_tokens": 2826395.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.044871759414673,
|
|
"epoch": 1.280334728033473,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004974135650551262,
|
|
"loss": 5.8631,
|
|
"mean_token_accuracy": 0.148614738881588,
|
|
"num_tokens": 2836195.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.948652601242065,
|
|
"epoch": 1.284518828451883,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004973645372049796,
|
|
"loss": 5.6917,
|
|
"mean_token_accuracy": 0.16129153966903687,
|
|
"num_tokens": 2846020.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.887058067321777,
|
|
"epoch": 1.288702928870293,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004973150517643659,
|
|
"loss": 5.6387,
|
|
"mean_token_accuracy": 0.1660175383090973,
|
|
"num_tokens": 2854895.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.9782609939575195,
|
|
"epoch": 1.292887029288703,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004972651088351185,
|
|
"loss": 5.6876,
|
|
"mean_token_accuracy": 0.1550069198012352,
|
|
"num_tokens": 2864491.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.003243494033813,
|
|
"epoch": 1.297071129707113,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004972147085200116,
|
|
"loss": 5.7089,
|
|
"mean_token_accuracy": 0.1583276644349098,
|
|
"num_tokens": 2873962.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.021131706237793,
|
|
"epoch": 1.301255230125523,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004971638509227611,
|
|
"loss": 5.7304,
|
|
"mean_token_accuracy": 0.16191184967756272,
|
|
"num_tokens": 2883530.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.935977506637573,
|
|
"epoch": 1.3054393305439331,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004971125361480236,
|
|
"loss": 5.7084,
|
|
"mean_token_accuracy": 0.1528172731399536,
|
|
"num_tokens": 2892867.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.964928483963012,
|
|
"epoch": 1.3096234309623431,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004970607643013969,
|
|
"loss": 5.6739,
|
|
"mean_token_accuracy": 0.16599547117948532,
|
|
"num_tokens": 2902320.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.928283262252807,
|
|
"epoch": 1.3138075313807531,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000497008535489419,
|
|
"loss": 5.6302,
|
|
"mean_token_accuracy": 0.16848794221878052,
|
|
"num_tokens": 2911390.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.889198350906372,
|
|
"epoch": 1.3179916317991631,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004969558498195684,
|
|
"loss": 5.6754,
|
|
"mean_token_accuracy": 0.16107864081859588,
|
|
"num_tokens": 2921295.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.953315258026123,
|
|
"epoch": 1.3221757322175733,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004969027074002638,
|
|
"loss": 5.6972,
|
|
"mean_token_accuracy": 0.158388452231884,
|
|
"num_tokens": 2931328.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.862377786636353,
|
|
"epoch": 1.3263598326359833,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004968491083408636,
|
|
"loss": 5.6311,
|
|
"mean_token_accuracy": 0.1618984967470169,
|
|
"num_tokens": 2940193.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.8608598709106445,
|
|
"epoch": 1.3305439330543933,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004967950527516663,
|
|
"loss": 5.679,
|
|
"mean_token_accuracy": 0.16396260410547256,
|
|
"num_tokens": 2949759.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.889656448364258,
|
|
"epoch": 1.3347280334728033,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004967405407439092,
|
|
"loss": 5.6019,
|
|
"mean_token_accuracy": 0.15576141625642775,
|
|
"num_tokens": 2959185.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.9884919166564945,
|
|
"epoch": 1.3389121338912133,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004966855724297698,
|
|
"loss": 5.7545,
|
|
"mean_token_accuracy": 0.15636038184165954,
|
|
"num_tokens": 2968945.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.905081415176392,
|
|
"epoch": 1.3430962343096233,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004966301479223637,
|
|
"loss": 5.6904,
|
|
"mean_token_accuracy": 0.16265947967767716,
|
|
"num_tokens": 2978395.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.862964487075805,
|
|
"epoch": 1.3472803347280335,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004965742673357456,
|
|
"loss": 5.6066,
|
|
"mean_token_accuracy": 0.17108241766691207,
|
|
"num_tokens": 2987030.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.951865196228027,
|
|
"epoch": 1.3514644351464435,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004965179307849088,
|
|
"loss": 5.6569,
|
|
"mean_token_accuracy": 0.16522777527570726,
|
|
"num_tokens": 2995736.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.986184692382812,
|
|
"epoch": 1.3556485355648535,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000496461138385785,
|
|
"loss": 5.7707,
|
|
"mean_token_accuracy": 0.16053102016448975,
|
|
"num_tokens": 3005105.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.871125841140747,
|
|
"epoch": 1.3598326359832635,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004964038902552438,
|
|
"loss": 5.6772,
|
|
"mean_token_accuracy": 0.15928133577108383,
|
|
"num_tokens": 3013857.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.943917655944825,
|
|
"epoch": 1.3640167364016738,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004963461865110926,
|
|
"loss": 5.7016,
|
|
"mean_token_accuracy": 0.16053634732961655,
|
|
"num_tokens": 3023972.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.834832668304443,
|
|
"epoch": 1.3682008368200838,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004962880272720764,
|
|
"loss": 5.6941,
|
|
"mean_token_accuracy": 0.16110829263925552,
|
|
"num_tokens": 3032705.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.943324041366577,
|
|
"epoch": 1.3723849372384938,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004962294126578775,
|
|
"loss": 5.6149,
|
|
"mean_token_accuracy": 0.1666594684123993,
|
|
"num_tokens": 3040900.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.922968435287475,
|
|
"epoch": 1.3765690376569037,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004961703427891154,
|
|
"loss": 5.6526,
|
|
"mean_token_accuracy": 0.16836990267038346,
|
|
"num_tokens": 3050477.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.960897588729859,
|
|
"epoch": 1.3807531380753137,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004961108177873466,
|
|
"loss": 5.725,
|
|
"mean_token_accuracy": 0.15328031480312349,
|
|
"num_tokens": 3059743.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.852565479278565,
|
|
"epoch": 1.3849372384937237,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004960508377750638,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.16497061252593995,
|
|
"num_tokens": 3067801.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.8161725997924805,
|
|
"epoch": 1.389121338912134,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004959904028756961,
|
|
"loss": 5.704,
|
|
"mean_token_accuracy": 0.15327719002962112,
|
|
"num_tokens": 3077404.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.943395137786865,
|
|
"epoch": 1.393305439330544,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004959295132136091,
|
|
"loss": 5.729,
|
|
"mean_token_accuracy": 0.16609538793563844,
|
|
"num_tokens": 3087130.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.978927564620972,
|
|
"epoch": 1.397489539748954,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004958681689141036,
|
|
"loss": 5.7547,
|
|
"mean_token_accuracy": 0.15868422463536264,
|
|
"num_tokens": 3096038.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.844921827316284,
|
|
"epoch": 1.401673640167364,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004958063701034167,
|
|
"loss": 5.5888,
|
|
"mean_token_accuracy": 0.16769689321517944,
|
|
"num_tokens": 3105668.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.848332071304322,
|
|
"epoch": 1.4058577405857742,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00049574411690872,
|
|
"loss": 5.5847,
|
|
"mean_token_accuracy": 0.16057471334934234,
|
|
"num_tokens": 3114020.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.77159309387207,
|
|
"epoch": 1.4100418410041842,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004956814094581208,
|
|
"loss": 5.611,
|
|
"mean_token_accuracy": 0.1696687251329422,
|
|
"num_tokens": 3124096.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.920372581481933,
|
|
"epoch": 1.4142259414225942,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000495618247880661,
|
|
"loss": 5.6827,
|
|
"mean_token_accuracy": 0.16226903945207596,
|
|
"num_tokens": 3132806.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.8516641616821286,
|
|
"epoch": 1.4184100418410042,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004955546323063168,
|
|
"loss": 5.6113,
|
|
"mean_token_accuracy": 0.17359014451503754,
|
|
"num_tokens": 3142001.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.806506490707397,
|
|
"epoch": 1.4225941422594142,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004954905628659989,
|
|
"loss": 5.5746,
|
|
"mean_token_accuracy": 0.16512429863214492,
|
|
"num_tokens": 3151658.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.879411745071411,
|
|
"epoch": 1.4267782426778242,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004954260396915521,
|
|
"loss": 5.6154,
|
|
"mean_token_accuracy": 0.16517930328845978,
|
|
"num_tokens": 3159573.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.693005704879761,
|
|
"epoch": 1.4309623430962344,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004953610629157544,
|
|
"loss": 5.4799,
|
|
"mean_token_accuracy": 0.17277001291513444,
|
|
"num_tokens": 3168312.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.947100019454956,
|
|
"epoch": 1.4351464435146444,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004952956326723177,
|
|
"loss": 5.5997,
|
|
"mean_token_accuracy": 0.17409175038337707,
|
|
"num_tokens": 3176811.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.838509893417358,
|
|
"epoch": 1.4393305439330544,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000495229749095887,
|
|
"loss": 5.6813,
|
|
"mean_token_accuracy": 0.1600296452641487,
|
|
"num_tokens": 3186599.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.01194658279419,
|
|
"epoch": 1.4435146443514644,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00049516341232204,
|
|
"loss": 5.7782,
|
|
"mean_token_accuracy": 0.16404415518045426,
|
|
"num_tokens": 3195905.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.862938690185547,
|
|
"epoch": 1.4476987447698746,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000495096622487287,
|
|
"loss": 5.7459,
|
|
"mean_token_accuracy": 0.1582975223660469,
|
|
"num_tokens": 3205132.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.993532276153564,
|
|
"epoch": 1.4518828451882846,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000495029379729071,
|
|
"loss": 5.7438,
|
|
"mean_token_accuracy": 0.15568403899669647,
|
|
"num_tokens": 3214362.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.915787315368652,
|
|
"epoch": 1.4560669456066946,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004949616841857666,
|
|
"loss": 5.7358,
|
|
"mean_token_accuracy": 0.15293580591678618,
|
|
"num_tokens": 3224248.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.854905366897583,
|
|
"epoch": 1.4602510460251046,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004948935359966806,
|
|
"loss": 5.6115,
|
|
"mean_token_accuracy": 0.1644871339201927,
|
|
"num_tokens": 3233416.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.758350658416748,
|
|
"epoch": 1.4644351464435146,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004948249353020506,
|
|
"loss": 5.5483,
|
|
"mean_token_accuracy": 0.1677642360329628,
|
|
"num_tokens": 3242354.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.866594886779785,
|
|
"epoch": 1.4686192468619246,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004947558822430461,
|
|
"loss": 5.6246,
|
|
"mean_token_accuracy": 0.16634228825569153,
|
|
"num_tokens": 3251026.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.8610475063323975,
|
|
"epoch": 1.4728033472803348,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004946863769617672,
|
|
"loss": 5.5884,
|
|
"mean_token_accuracy": 0.16393277645111085,
|
|
"num_tokens": 3259429.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.818919134140015,
|
|
"epoch": 1.4769874476987448,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004946164196012445,
|
|
"loss": 5.6387,
|
|
"mean_token_accuracy": 0.16186177730560303,
|
|
"num_tokens": 3269305.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.900064516067505,
|
|
"epoch": 1.4811715481171548,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004945460103054389,
|
|
"loss": 5.7037,
|
|
"mean_token_accuracy": 0.16150607615709306,
|
|
"num_tokens": 3278846.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.884778785705566,
|
|
"epoch": 1.4853556485355648,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004944751492192416,
|
|
"loss": 5.6298,
|
|
"mean_token_accuracy": 0.16663831770420073,
|
|
"num_tokens": 3287200.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.897223949432373,
|
|
"epoch": 1.489539748953975,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004944038364884734,
|
|
"loss": 5.6554,
|
|
"mean_token_accuracy": 0.15921722799539567,
|
|
"num_tokens": 3296651.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.840246629714966,
|
|
"epoch": 1.493723849372385,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004943320722598842,
|
|
"loss": 5.5293,
|
|
"mean_token_accuracy": 0.17194677889347076,
|
|
"num_tokens": 3304887.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.846546983718872,
|
|
"epoch": 1.497907949790795,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004942598566811532,
|
|
"loss": 5.5619,
|
|
"mean_token_accuracy": 0.1705879956483841,
|
|
"num_tokens": 3313925.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.803055047988892,
|
|
"epoch": 1.502092050209205,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004941871899008887,
|
|
"loss": 5.5427,
|
|
"mean_token_accuracy": 0.1690042346715927,
|
|
"num_tokens": 3322943.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.761071348190308,
|
|
"epoch": 1.506276150627615,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000494114072068627,
|
|
"loss": 5.5811,
|
|
"mean_token_accuracy": 0.16471089348196982,
|
|
"num_tokens": 3332742.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.862655353546143,
|
|
"epoch": 1.510460251046025,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004940405033348332,
|
|
"loss": 5.6493,
|
|
"mean_token_accuracy": 0.1646712064743042,
|
|
"num_tokens": 3342228.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.818815994262695,
|
|
"epoch": 1.514644351464435,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004939664838508996,
|
|
"loss": 5.6235,
|
|
"mean_token_accuracy": 0.16362904012203217,
|
|
"num_tokens": 3351854.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.8133251667022705,
|
|
"epoch": 1.5188284518828452,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004938920137691465,
|
|
"loss": 5.6065,
|
|
"mean_token_accuracy": 0.16995372325181962,
|
|
"num_tokens": 3360809.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.821644926071167,
|
|
"epoch": 1.5230125523012552,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004938170932428214,
|
|
"loss": 5.6374,
|
|
"mean_token_accuracy": 0.1564922161400318,
|
|
"num_tokens": 3371033.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.929017734527588,
|
|
"epoch": 1.5271966527196654,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004937417224260984,
|
|
"loss": 5.5169,
|
|
"mean_token_accuracy": 0.17455457597970964,
|
|
"num_tokens": 3379943.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.726641511917114,
|
|
"epoch": 1.5313807531380754,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004936659014740791,
|
|
"loss": 5.6548,
|
|
"mean_token_accuracy": 0.1662447988986969,
|
|
"num_tokens": 3389316.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.924563264846801,
|
|
"epoch": 1.5355648535564854,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004935896305427904,
|
|
"loss": 5.6622,
|
|
"mean_token_accuracy": 0.16375426799058915,
|
|
"num_tokens": 3398914.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.967609214782715,
|
|
"epoch": 1.5397489539748954,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004935129097891857,
|
|
"loss": 5.6454,
|
|
"mean_token_accuracy": 0.15691737979650497,
|
|
"num_tokens": 3407967.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.923986530303955,
|
|
"epoch": 1.5439330543933054,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000493435739371144,
|
|
"loss": 5.6759,
|
|
"mean_token_accuracy": 0.1639568865299225,
|
|
"num_tokens": 3416530.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.827739286422729,
|
|
"epoch": 1.5481171548117154,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004933581194474695,
|
|
"loss": 5.5528,
|
|
"mean_token_accuracy": 0.1630283534526825,
|
|
"num_tokens": 3425762.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.790284538269043,
|
|
"epoch": 1.5523012552301254,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004932800501778918,
|
|
"loss": 5.6263,
|
|
"mean_token_accuracy": 0.16491218209266661,
|
|
"num_tokens": 3434799.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.898684740066528,
|
|
"epoch": 1.5564853556485354,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004932015317230646,
|
|
"loss": 5.6416,
|
|
"mean_token_accuracy": 0.162263023853302,
|
|
"num_tokens": 3442986.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.820795106887817,
|
|
"epoch": 1.5606694560669456,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004931225642445665,
|
|
"loss": 5.6662,
|
|
"mean_token_accuracy": 0.15839100629091263,
|
|
"num_tokens": 3452914.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.805272006988526,
|
|
"epoch": 1.5648535564853556,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004930431479048998,
|
|
"loss": 5.5178,
|
|
"mean_token_accuracy": 0.16989513635635375,
|
|
"num_tokens": 3462093.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.855914783477783,
|
|
"epoch": 1.5690376569037658,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004929632828674904,
|
|
"loss": 5.7165,
|
|
"mean_token_accuracy": 0.1553257957100868,
|
|
"num_tokens": 3471259.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.836420297622681,
|
|
"epoch": 1.5732217573221758,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004928829692966879,
|
|
"loss": 5.6392,
|
|
"mean_token_accuracy": 0.16758336424827575,
|
|
"num_tokens": 3480071.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.965684461593628,
|
|
"epoch": 1.5774058577405858,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004928022073577646,
|
|
"loss": 5.659,
|
|
"mean_token_accuracy": 0.16947642117738723,
|
|
"num_tokens": 3489075.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.76209659576416,
|
|
"epoch": 1.5815899581589958,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004927209972169157,
|
|
"loss": 5.4852,
|
|
"mean_token_accuracy": 0.17077343314886093,
|
|
"num_tokens": 3498495.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.8187662124633786,
|
|
"epoch": 1.5857740585774058,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004926393390412586,
|
|
"loss": 5.629,
|
|
"mean_token_accuracy": 0.16686113625764848,
|
|
"num_tokens": 3508590.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.911009454727173,
|
|
"epoch": 1.5899581589958158,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004925572329988325,
|
|
"loss": 5.6395,
|
|
"mean_token_accuracy": 0.16106612235307693,
|
|
"num_tokens": 3518397.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.768456077575683,
|
|
"epoch": 1.5941422594142258,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004924746792585987,
|
|
"loss": 5.6187,
|
|
"mean_token_accuracy": 0.16009850054979324,
|
|
"num_tokens": 3526620.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.925689744949341,
|
|
"epoch": 1.5983263598326358,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004923916779904394,
|
|
"loss": 5.6344,
|
|
"mean_token_accuracy": 0.16397972404956818,
|
|
"num_tokens": 3535534.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.783263683319092,
|
|
"epoch": 1.602510460251046,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004923082293651578,
|
|
"loss": 5.5887,
|
|
"mean_token_accuracy": 0.16171235889196395,
|
|
"num_tokens": 3544771.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.737907361984253,
|
|
"epoch": 1.606694560669456,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492224333554478,
|
|
"loss": 5.5622,
|
|
"mean_token_accuracy": 0.1592784434556961,
|
|
"num_tokens": 3553970.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.719466972351074,
|
|
"epoch": 1.6108786610878663,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004921399907310437,
|
|
"loss": 5.4701,
|
|
"mean_token_accuracy": 0.16803071945905684,
|
|
"num_tokens": 3563116.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.7726158618927,
|
|
"epoch": 1.6150627615062763,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004920552010684191,
|
|
"loss": 5.6212,
|
|
"mean_token_accuracy": 0.16787808388471603,
|
|
"num_tokens": 3571605.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.770044422149658,
|
|
"epoch": 1.6192468619246863,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004919699647410878,
|
|
"loss": 5.6823,
|
|
"mean_token_accuracy": 0.15952937453985214,
|
|
"num_tokens": 3581837.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.85563097000122,
|
|
"epoch": 1.6234309623430963,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004918842819244523,
|
|
"loss": 5.6331,
|
|
"mean_token_accuracy": 0.161039100587368,
|
|
"num_tokens": 3590853.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.761953830718994,
|
|
"epoch": 1.6276150627615062,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004917981527948342,
|
|
"loss": 5.5308,
|
|
"mean_token_accuracy": 0.16669060438871383,
|
|
"num_tokens": 3599708.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.759849405288696,
|
|
"epoch": 1.6317991631799162,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004917115775294731,
|
|
"loss": 5.5644,
|
|
"mean_token_accuracy": 0.16626980751752854,
|
|
"num_tokens": 3608634.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.808131408691406,
|
|
"epoch": 1.6359832635983262,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004916245563065273,
|
|
"loss": 5.6122,
|
|
"mean_token_accuracy": 0.16667178720235826,
|
|
"num_tokens": 3618776.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.82210340499878,
|
|
"epoch": 1.6401673640167362,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004915370893050722,
|
|
"loss": 5.6226,
|
|
"mean_token_accuracy": 0.16585469096899033,
|
|
"num_tokens": 3628595.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.915043020248413,
|
|
"epoch": 1.6443514644351465,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004914491767051012,
|
|
"loss": 5.7298,
|
|
"mean_token_accuracy": 0.1548106774687767,
|
|
"num_tokens": 3638133.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.846599674224853,
|
|
"epoch": 1.6485355648535565,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004913608186875241,
|
|
"loss": 5.5201,
|
|
"mean_token_accuracy": 0.16792443096637727,
|
|
"num_tokens": 3646679.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.8250938892364506,
|
|
"epoch": 1.6527196652719667,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004912720154341674,
|
|
"loss": 5.6473,
|
|
"mean_token_accuracy": 0.1577279731631279,
|
|
"num_tokens": 3656131.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.823947334289551,
|
|
"epoch": 1.6569037656903767,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004911827671277739,
|
|
"loss": 5.5535,
|
|
"mean_token_accuracy": 0.16799709796905518,
|
|
"num_tokens": 3665187.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.807013511657715,
|
|
"epoch": 1.6610878661087867,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004910930739520025,
|
|
"loss": 5.6161,
|
|
"mean_token_accuracy": 0.16549118906259536,
|
|
"num_tokens": 3674649.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.8873412132263185,
|
|
"epoch": 1.6652719665271967,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004910029360914271,
|
|
"loss": 5.6325,
|
|
"mean_token_accuracy": 0.15765147656202316,
|
|
"num_tokens": 3684680.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.7522777080535885,
|
|
"epoch": 1.6694560669456067,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000490912353731537,
|
|
"loss": 5.5716,
|
|
"mean_token_accuracy": 0.16979746967554094,
|
|
"num_tokens": 3693560.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.745171689987183,
|
|
"epoch": 1.6736401673640167,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004908213270587363,
|
|
"loss": 5.5636,
|
|
"mean_token_accuracy": 0.16735229641199112,
|
|
"num_tokens": 3702632.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6736401673640167,
|
|
"eval_entropy": 5.587893625466126,
|
|
"eval_loss": 5.668591499328613,
|
|
"eval_mean_token_accuracy": 0.16891432853686053,
|
|
"eval_num_tokens": 3702632.0,
|
|
"eval_runtime": 2.272,
|
|
"eval_samples_per_second": 1719.645,
|
|
"eval_steps_per_second": 215.231,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11950,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 799490537717760.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|