bd86e71c20
Model: fpadovani/dan-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
9134 lines
248 KiB
JSON
9134 lines
248 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.7656903765690375,
|
|
"eval_steps": 500,
|
|
"global_step": 4500,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 5.7460853099823,
|
|
"epoch": 1.6778242677824267,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004907298562603431,
|
|
"loss": 5.4493,
|
|
"mean_token_accuracy": 0.17528735250234603,
|
|
"num_tokens": 3711022.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.792792701721192,
|
|
"epoch": 1.6820083682008367,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004906379415245896,
|
|
"loss": 5.5704,
|
|
"mean_token_accuracy": 0.171998530626297,
|
|
"num_tokens": 3719778.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.731085014343262,
|
|
"epoch": 1.6861924686192469,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004905455830406215,
|
|
"loss": 5.5586,
|
|
"mean_token_accuracy": 0.17053967714309692,
|
|
"num_tokens": 3728773.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.7191081047058105,
|
|
"epoch": 1.6903765690376569,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004904527809984978,
|
|
"loss": 5.5965,
|
|
"mean_token_accuracy": 0.16994577497243882,
|
|
"num_tokens": 3737781.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.90582389831543,
|
|
"epoch": 1.694560669456067,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004903595355891902,
|
|
"loss": 5.6436,
|
|
"mean_token_accuracy": 0.15853201001882553,
|
|
"num_tokens": 3748010.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.7872340202331545,
|
|
"epoch": 1.698744769874477,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004902658470045826,
|
|
"loss": 5.6176,
|
|
"mean_token_accuracy": 0.16531070470809936,
|
|
"num_tokens": 3757894.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.865278530120849,
|
|
"epoch": 1.702928870292887,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004901717154374713,
|
|
"loss": 5.7089,
|
|
"mean_token_accuracy": 0.15969219505786897,
|
|
"num_tokens": 3768760.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.770133209228516,
|
|
"epoch": 1.707112970711297,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004900771410815638,
|
|
"loss": 5.5517,
|
|
"mean_token_accuracy": 0.16736289709806443,
|
|
"num_tokens": 3778273.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.716675090789795,
|
|
"epoch": 1.711297071129707,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004899821241314788,
|
|
"loss": 5.5632,
|
|
"mean_token_accuracy": 0.16601912081241607,
|
|
"num_tokens": 3787012.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.80701789855957,
|
|
"epoch": 1.715481171548117,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000489886664782746,
|
|
"loss": 5.5165,
|
|
"mean_token_accuracy": 0.17513245195150376,
|
|
"num_tokens": 3795144.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.846057319641114,
|
|
"epoch": 1.719665271966527,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004897907632318055,
|
|
"loss": 5.7333,
|
|
"mean_token_accuracy": 0.16434134542942047,
|
|
"num_tokens": 3804534.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.767861700057983,
|
|
"epoch": 1.723849372384937,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004896944196760071,
|
|
"loss": 5.5048,
|
|
"mean_token_accuracy": 0.16955313235521316,
|
|
"num_tokens": 3813757.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.684953403472901,
|
|
"epoch": 1.7280334728033473,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004895976343136106,
|
|
"loss": 5.5357,
|
|
"mean_token_accuracy": 0.17599508315324783,
|
|
"num_tokens": 3823157.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.674932622909546,
|
|
"epoch": 1.7322175732217573,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004895004073437844,
|
|
"loss": 5.4152,
|
|
"mean_token_accuracy": 0.17153321206569672,
|
|
"num_tokens": 3831661.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.692170238494873,
|
|
"epoch": 1.7364016736401675,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004894027389666061,
|
|
"loss": 5.4241,
|
|
"mean_token_accuracy": 0.17433399111032485,
|
|
"num_tokens": 3840591.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.754289197921753,
|
|
"epoch": 1.7405857740585775,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004893046293830616,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.16439199000597,
|
|
"num_tokens": 3850094.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.77983512878418,
|
|
"epoch": 1.7447698744769875,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004892060787950447,
|
|
"loss": 5.6104,
|
|
"mean_token_accuracy": 0.1703110530972481,
|
|
"num_tokens": 3858973.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.6613945960998535,
|
|
"epoch": 1.7489539748953975,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004891070874053564,
|
|
"loss": 5.376,
|
|
"mean_token_accuracy": 0.17869551479816437,
|
|
"num_tokens": 3868315.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.716506767272949,
|
|
"epoch": 1.7531380753138075,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004890076554177051,
|
|
"loss": 5.5556,
|
|
"mean_token_accuracy": 0.16812776774168015,
|
|
"num_tokens": 3877605.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.779949760437011,
|
|
"epoch": 1.7573221757322175,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004889077830367061,
|
|
"loss": 5.5524,
|
|
"mean_token_accuracy": 0.1700603887438774,
|
|
"num_tokens": 3886998.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.751646375656128,
|
|
"epoch": 1.7615062761506275,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004888074704678806,
|
|
"loss": 5.5062,
|
|
"mean_token_accuracy": 0.17651222944259642,
|
|
"num_tokens": 3896018.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.757708644866943,
|
|
"epoch": 1.7656903765690377,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004887067179176558,
|
|
"loss": 5.5028,
|
|
"mean_token_accuracy": 0.16861198991537094,
|
|
"num_tokens": 3904957.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.7517915725708,
|
|
"epoch": 1.7698744769874477,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004886055255933639,
|
|
"loss": 5.6139,
|
|
"mean_token_accuracy": 0.1639676995575428,
|
|
"num_tokens": 3913427.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.762232637405395,
|
|
"epoch": 1.7740585774058577,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004885038937032429,
|
|
"loss": 5.5282,
|
|
"mean_token_accuracy": 0.16971725225448608,
|
|
"num_tokens": 3923134.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.759613513946533,
|
|
"epoch": 1.778242677824268,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004884018224564345,
|
|
"loss": 5.4554,
|
|
"mean_token_accuracy": 0.16991421431303025,
|
|
"num_tokens": 3931867.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.760385417938233,
|
|
"epoch": 1.782426778242678,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004882993120629852,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.17172749042510987,
|
|
"num_tokens": 3941057.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.8172108173370365,
|
|
"epoch": 1.786610878661088,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004881963627338449,
|
|
"loss": 5.6718,
|
|
"mean_token_accuracy": 0.16127773523330688,
|
|
"num_tokens": 3950511.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.796531724929809,
|
|
"epoch": 1.790794979079498,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048809297468086654,
|
|
"loss": 5.5748,
|
|
"mean_token_accuracy": 0.1713390588760376,
|
|
"num_tokens": 3959769.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.733460569381714,
|
|
"epoch": 1.794979079497908,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004879891481168063,
|
|
"loss": 5.4987,
|
|
"mean_token_accuracy": 0.16906837970018387,
|
|
"num_tokens": 3968462.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.688917446136474,
|
|
"epoch": 1.799163179916318,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048788488325532235,
|
|
"loss": 5.5487,
|
|
"mean_token_accuracy": 0.1712746575474739,
|
|
"num_tokens": 3977464.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.702218818664551,
|
|
"epoch": 1.803347280334728,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004877801803109753,
|
|
"loss": 5.4576,
|
|
"mean_token_accuracy": 0.17578500658273696,
|
|
"num_tokens": 3986303.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.8299205780029295,
|
|
"epoch": 1.8075313807531381,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048767503949922683,
|
|
"loss": 5.5171,
|
|
"mean_token_accuracy": 0.1743282049894333,
|
|
"num_tokens": 3994970.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.765460681915283,
|
|
"epoch": 1.8117154811715481,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004875694610364398,
|
|
"loss": 5.5958,
|
|
"mean_token_accuracy": 0.1668936088681221,
|
|
"num_tokens": 4004679.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.860610389709473,
|
|
"epoch": 1.8158995815899581,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048746344513987775,
|
|
"loss": 5.6158,
|
|
"mean_token_accuracy": 0.16595034450292587,
|
|
"num_tokens": 4013516.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.791381883621216,
|
|
"epoch": 1.8200836820083683,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004873569920277043,
|
|
"loss": 5.4704,
|
|
"mean_token_accuracy": 0.1770814836025238,
|
|
"num_tokens": 4022335.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.621400976181031,
|
|
"epoch": 1.8242677824267783,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048725010191898296,
|
|
"loss": 5.4644,
|
|
"mean_token_accuracy": 0.17260104715824126,
|
|
"num_tokens": 4031295.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.82382607460022,
|
|
"epoch": 1.8284518828451883,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048714277503367627,
|
|
"loss": 5.6699,
|
|
"mean_token_accuracy": 0.16565849930047988,
|
|
"num_tokens": 4040378.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.80361123085022,
|
|
"epoch": 1.8326359832635983,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048703501159264576,
|
|
"loss": 5.5382,
|
|
"mean_token_accuracy": 0.16324643343687056,
|
|
"num_tokens": 4050217.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.614384746551513,
|
|
"epoch": 1.8368200836820083,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004869268118176512,
|
|
"loss": 5.4113,
|
|
"mean_token_accuracy": 0.1673094302415848,
|
|
"num_tokens": 4059088.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.821628189086914,
|
|
"epoch": 1.8410041841004183,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048681817593135046,
|
|
"loss": 5.5145,
|
|
"mean_token_accuracy": 0.18171378523111342,
|
|
"num_tokens": 4067609.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.755545902252197,
|
|
"epoch": 1.8451882845188283,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048670910415729855,
|
|
"loss": 5.5093,
|
|
"mean_token_accuracy": 0.1711531937122345,
|
|
"num_tokens": 4076763.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.805058765411377,
|
|
"epoch": 1.8493723849372385,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048659959671994786,
|
|
"loss": 5.6575,
|
|
"mean_token_accuracy": 0.1649637520313263,
|
|
"num_tokens": 4086552.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.612725400924683,
|
|
"epoch": 1.8535564853556485,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048648965384464693,
|
|
"loss": 5.4246,
|
|
"mean_token_accuracy": 0.1782457634806633,
|
|
"num_tokens": 4096200.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.6816935539245605,
|
|
"epoch": 1.8577405857740585,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048637927575764057,
|
|
"loss": 5.4601,
|
|
"mean_token_accuracy": 0.173495152592659,
|
|
"num_tokens": 4105080.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.705690526962281,
|
|
"epoch": 1.8619246861924688,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048626846268606913,
|
|
"loss": 5.4225,
|
|
"mean_token_accuracy": 0.17340910136699678,
|
|
"num_tokens": 4113972.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.814240169525147,
|
|
"epoch": 1.8661087866108788,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048615721485796803,
|
|
"loss": 5.6637,
|
|
"mean_token_accuracy": 0.16688043773174285,
|
|
"num_tokens": 4123615.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.752278137207031,
|
|
"epoch": 1.8702928870292888,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004860455325022676,
|
|
"loss": 5.4368,
|
|
"mean_token_accuracy": 0.1718146875500679,
|
|
"num_tokens": 4132128.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.792816925048828,
|
|
"epoch": 1.8744769874476988,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000485933415848792,
|
|
"loss": 5.6275,
|
|
"mean_token_accuracy": 0.16710745096206664,
|
|
"num_tokens": 4141575.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.804529523849487,
|
|
"epoch": 1.8786610878661087,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004858208651282593,
|
|
"loss": 5.4934,
|
|
"mean_token_accuracy": 0.17012665122747422,
|
|
"num_tokens": 4150372.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.680215692520141,
|
|
"epoch": 1.8828451882845187,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000485707880572281,
|
|
"loss": 5.5801,
|
|
"mean_token_accuracy": 0.17325089424848555,
|
|
"num_tokens": 4159007.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.702152252197266,
|
|
"epoch": 1.8870292887029287,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004855944624133608,
|
|
"loss": 5.6195,
|
|
"mean_token_accuracy": 0.1642859920859337,
|
|
"num_tokens": 4168646.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.702015733718872,
|
|
"epoch": 1.891213389121339,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048548061088489523,
|
|
"loss": 5.5344,
|
|
"mean_token_accuracy": 0.1651986286044121,
|
|
"num_tokens": 4177979.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.7645830631256105,
|
|
"epoch": 1.895397489539749,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048536632622117246,
|
|
"loss": 5.4686,
|
|
"mean_token_accuracy": 0.16989308148622512,
|
|
"num_tokens": 4186781.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.687842321395874,
|
|
"epoch": 1.899581589958159,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048525160865737186,
|
|
"loss": 5.518,
|
|
"mean_token_accuracy": 0.17315628230571747,
|
|
"num_tokens": 4195661.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.712222766876221,
|
|
"epoch": 1.9037656903765692,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048513645842956384,
|
|
"loss": 5.4259,
|
|
"mean_token_accuracy": 0.1755947858095169,
|
|
"num_tokens": 4203902.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.666417026519776,
|
|
"epoch": 1.9079497907949792,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048502087577470895,
|
|
"loss": 5.4828,
|
|
"mean_token_accuracy": 0.16765579730272293,
|
|
"num_tokens": 4213663.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.800201034545898,
|
|
"epoch": 1.9121338912133892,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048490486093065774,
|
|
"loss": 5.5266,
|
|
"mean_token_accuracy": 0.1695411831140518,
|
|
"num_tokens": 4222152.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.670174217224121,
|
|
"epoch": 1.9163179916317992,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048478841413615017,
|
|
"loss": 5.5375,
|
|
"mean_token_accuracy": 0.17209279239177705,
|
|
"num_tokens": 4232347.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.787844228744507,
|
|
"epoch": 1.9205020920502092,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048467153563081487,
|
|
"loss": 5.597,
|
|
"mean_token_accuracy": 0.17511929869651793,
|
|
"num_tokens": 4240607.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.761624479293824,
|
|
"epoch": 1.9246861924686192,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004845542256551692,
|
|
"loss": 5.6054,
|
|
"mean_token_accuracy": 0.16762748956680298,
|
|
"num_tokens": 4251131.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.762709379196167,
|
|
"epoch": 1.9288702928870292,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004844364844506181,
|
|
"loss": 5.5404,
|
|
"mean_token_accuracy": 0.17349285185337066,
|
|
"num_tokens": 4259955.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.78967695236206,
|
|
"epoch": 1.9330543933054394,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004843183122594541,
|
|
"loss": 5.5052,
|
|
"mean_token_accuracy": 0.17376520931720735,
|
|
"num_tokens": 4268686.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.728965091705322,
|
|
"epoch": 1.9372384937238494,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004841997093248566,
|
|
"loss": 5.5021,
|
|
"mean_token_accuracy": 0.1730721265077591,
|
|
"num_tokens": 4277377.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.640170955657959,
|
|
"epoch": 1.9414225941422594,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004840806758908913,
|
|
"loss": 5.4896,
|
|
"mean_token_accuracy": 0.17226046323776245,
|
|
"num_tokens": 4287037.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.792774486541748,
|
|
"epoch": 1.9456066945606696,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048396121220250983,
|
|
"loss": 5.6392,
|
|
"mean_token_accuracy": 0.1607039153575897,
|
|
"num_tokens": 4296923.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.654302072525025,
|
|
"epoch": 1.9497907949790796,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004838413185055494,
|
|
"loss": 5.3878,
|
|
"mean_token_accuracy": 0.17721690982580185,
|
|
"num_tokens": 4305873.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.687008857727051,
|
|
"epoch": 1.9539748953974896,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004837209950467318,
|
|
"loss": 5.5519,
|
|
"mean_token_accuracy": 0.1684328004717827,
|
|
"num_tokens": 4314912.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.792973947525025,
|
|
"epoch": 1.9581589958158996,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004836002420736635,
|
|
"loss": 5.552,
|
|
"mean_token_accuracy": 0.16693176329135895,
|
|
"num_tokens": 4324606.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.681936359405517,
|
|
"epoch": 1.9623430962343096,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004834790598348347,
|
|
"loss": 5.4197,
|
|
"mean_token_accuracy": 0.1764517068862915,
|
|
"num_tokens": 4333076.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.771762466430664,
|
|
"epoch": 1.9665271966527196,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004833574485796188,
|
|
"loss": 5.5404,
|
|
"mean_token_accuracy": 0.17310818135738373,
|
|
"num_tokens": 4341893.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.702968120574951,
|
|
"epoch": 1.9707112970711296,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048323540855827244,
|
|
"loss": 5.5232,
|
|
"mean_token_accuracy": 0.1702388569712639,
|
|
"num_tokens": 4352303.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.777474355697632,
|
|
"epoch": 1.9748953974895398,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004831129400219341,
|
|
"loss": 5.4855,
|
|
"mean_token_accuracy": 0.16857484877109527,
|
|
"num_tokens": 4361166.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.7430140495300295,
|
|
"epoch": 1.9790794979079498,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048299004322262457,
|
|
"loss": 5.5282,
|
|
"mean_token_accuracy": 0.169603231549263,
|
|
"num_tokens": 4369999.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.703432226181031,
|
|
"epoch": 1.9832635983263598,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048286671841324565,
|
|
"loss": 5.4836,
|
|
"mean_token_accuracy": 0.17527843713760377,
|
|
"num_tokens": 4379373.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.724852228164673,
|
|
"epoch": 1.98744769874477,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004827429658475799,
|
|
"loss": 5.4793,
|
|
"mean_token_accuracy": 0.1764371246099472,
|
|
"num_tokens": 4389035.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.753543519973755,
|
|
"epoch": 1.99163179916318,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048261878578029026,
|
|
"loss": 5.6417,
|
|
"mean_token_accuracy": 0.1693273589015007,
|
|
"num_tokens": 4399366.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.702381610870361,
|
|
"epoch": 1.99581589958159,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004824941784669193,
|
|
"loss": 5.4853,
|
|
"mean_token_accuracy": 0.17121606469154357,
|
|
"num_tokens": 4408183.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.774022006988526,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004823691441638889,
|
|
"loss": 5.6029,
|
|
"mean_token_accuracy": 0.1578802965581417,
|
|
"num_tokens": 4418644.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.654694890975952,
|
|
"epoch": 2.00418410041841,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004822436831284995,
|
|
"loss": 5.2325,
|
|
"mean_token_accuracy": 0.18315230458974838,
|
|
"num_tokens": 4427828.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.662265491485596,
|
|
"epoch": 2.00836820083682,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004821177956189297,
|
|
"loss": 5.1545,
|
|
"mean_token_accuracy": 0.18890936076641082,
|
|
"num_tokens": 4436194.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.565231800079346,
|
|
"epoch": 2.01255230125523,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004819914818942359,
|
|
"loss": 5.2009,
|
|
"mean_token_accuracy": 0.18946332931518556,
|
|
"num_tokens": 4444468.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.7318864345550535,
|
|
"epoch": 2.01673640167364,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048186474221435127,
|
|
"loss": 5.139,
|
|
"mean_token_accuracy": 0.19455838948488235,
|
|
"num_tokens": 4452760.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.63972840309143,
|
|
"epoch": 2.02092050209205,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048173757684008583,
|
|
"loss": 5.1932,
|
|
"mean_token_accuracy": 0.18992006182670593,
|
|
"num_tokens": 4462154.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.641659164428711,
|
|
"epoch": 2.0251046025104604,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004816099860331255,
|
|
"loss": 5.279,
|
|
"mean_token_accuracy": 0.18541579842567443,
|
|
"num_tokens": 4471987.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.625349521636963,
|
|
"epoch": 2.0292887029288704,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048148197005603145,
|
|
"loss": 5.1935,
|
|
"mean_token_accuracy": 0.1885792851448059,
|
|
"num_tokens": 4481645.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.529834318161011,
|
|
"epoch": 2.0334728033472804,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048135352917224005,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.19121328592300416,
|
|
"num_tokens": 4491166.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.7242209911346436,
|
|
"epoch": 2.0376569037656904,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048122466364606214,
|
|
"loss": 5.1924,
|
|
"mean_token_accuracy": 0.18891631215810775,
|
|
"num_tokens": 4500159.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.514832019805908,
|
|
"epoch": 2.0418410041841004,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048109537374268216,
|
|
"loss": 5.1421,
|
|
"mean_token_accuracy": 0.19749397337436675,
|
|
"num_tokens": 4509737.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.513562107086182,
|
|
"epoch": 2.0460251046025104,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048096565972815793,
|
|
"loss": 5.0655,
|
|
"mean_token_accuracy": 0.19419358521699906,
|
|
"num_tokens": 4518379.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.518503475189209,
|
|
"epoch": 2.0502092050209204,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004808355218694199,
|
|
"loss": 5.2068,
|
|
"mean_token_accuracy": 0.19322666972875596,
|
|
"num_tokens": 4527885.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.639360332489014,
|
|
"epoch": 2.0543933054393304,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048070496043427116,
|
|
"loss": 5.2418,
|
|
"mean_token_accuracy": 0.18575034141540528,
|
|
"num_tokens": 4537894.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.620634078979492,
|
|
"epoch": 2.0585774058577404,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048057397569138596,
|
|
"loss": 5.1391,
|
|
"mean_token_accuracy": 0.19480233788490295,
|
|
"num_tokens": 4546462.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.599647378921508,
|
|
"epoch": 2.062761506276151,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004804425679103099,
|
|
"loss": 5.1685,
|
|
"mean_token_accuracy": 0.1842091366648674,
|
|
"num_tokens": 4555970.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.6098803043365475,
|
|
"epoch": 2.066945606694561,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004803107373614589,
|
|
"loss": 5.2381,
|
|
"mean_token_accuracy": 0.18125001937150956,
|
|
"num_tokens": 4565799.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.657344579696655,
|
|
"epoch": 2.071129707112971,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004801784843161194,
|
|
"loss": 5.2564,
|
|
"mean_token_accuracy": 0.18035603314638138,
|
|
"num_tokens": 4575571.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.588315725326538,
|
|
"epoch": 2.075313807531381,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004800458090464465,
|
|
"loss": 5.1628,
|
|
"mean_token_accuracy": 0.1944723829627037,
|
|
"num_tokens": 4584263.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.5794602394104,
|
|
"epoch": 2.079497907949791,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047991271182546476,
|
|
"loss": 5.1386,
|
|
"mean_token_accuracy": 0.19624482840299606,
|
|
"num_tokens": 4592979.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.549351930618286,
|
|
"epoch": 2.083682008368201,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047977919292706696,
|
|
"loss": 5.156,
|
|
"mean_token_accuracy": 0.19804595708847045,
|
|
"num_tokens": 4602042.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.580633497238159,
|
|
"epoch": 2.087866108786611,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004796452526260134,
|
|
"loss": 5.2306,
|
|
"mean_token_accuracy": 0.1845365434885025,
|
|
"num_tokens": 4611197.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.490824604034424,
|
|
"epoch": 2.092050209205021,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047951089119793164,
|
|
"loss": 5.165,
|
|
"mean_token_accuracy": 0.18662422448396682,
|
|
"num_tokens": 4621612.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.092050209205021,
|
|
"eval_entropy": 5.3659842272721425,
|
|
"eval_loss": 5.560726165771484,
|
|
"eval_mean_token_accuracy": 0.17973723855242896,
|
|
"eval_num_tokens": 4621612.0,
|
|
"eval_runtime": 2.0793,
|
|
"eval_samples_per_second": 1879.017,
|
|
"eval_steps_per_second": 235.178,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.515201997756958,
|
|
"epoch": 2.096234309623431,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000479376108919316,
|
|
"loss": 5.0538,
|
|
"mean_token_accuracy": 0.2010475218296051,
|
|
"num_tokens": 4630198.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.5516763687133786,
|
|
"epoch": 2.100418410041841,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004792409060675267,
|
|
"loss": 5.1562,
|
|
"mean_token_accuracy": 0.18727059811353683,
|
|
"num_tokens": 4638729.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.530217552185059,
|
|
"epoch": 2.104602510460251,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004791052829207895,
|
|
"loss": 5.2642,
|
|
"mean_token_accuracy": 0.1901964485645294,
|
|
"num_tokens": 4648759.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.607113361358643,
|
|
"epoch": 2.1087866108786613,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000478969239758195,
|
|
"loss": 5.2329,
|
|
"mean_token_accuracy": 0.18608419150114058,
|
|
"num_tokens": 4657744.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.520095777511597,
|
|
"epoch": 2.1129707112970713,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047883277685969815,
|
|
"loss": 5.1777,
|
|
"mean_token_accuracy": 0.19294563680887222,
|
|
"num_tokens": 4666154.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.57831826210022,
|
|
"epoch": 2.1171548117154813,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047869589450611773,
|
|
"loss": 5.2101,
|
|
"mean_token_accuracy": 0.1856698289513588,
|
|
"num_tokens": 4675677.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.523758411407471,
|
|
"epoch": 2.1213389121338913,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004785585929791355,
|
|
"loss": 5.2077,
|
|
"mean_token_accuracy": 0.18869074136018754,
|
|
"num_tokens": 4684733.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.580788803100586,
|
|
"epoch": 2.1255230125523012,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000478420872561296,
|
|
"loss": 5.123,
|
|
"mean_token_accuracy": 0.19384075105190277,
|
|
"num_tokens": 4694155.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.504633951187134,
|
|
"epoch": 2.1297071129707112,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047828273353600577,
|
|
"loss": 5.1584,
|
|
"mean_token_accuracy": 0.192996609210968,
|
|
"num_tokens": 4703286.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.593018913269043,
|
|
"epoch": 2.1338912133891212,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047814417618753256,
|
|
"loss": 5.1687,
|
|
"mean_token_accuracy": 0.1865183562040329,
|
|
"num_tokens": 4711666.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.502986288070678,
|
|
"epoch": 2.1380753138075312,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00047800520080100526,
|
|
"loss": 5.1942,
|
|
"mean_token_accuracy": 0.18421708196401596,
|
|
"num_tokens": 4721644.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.565613508224487,
|
|
"epoch": 2.1422594142259412,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047786580766241265,
|
|
"loss": 5.1592,
|
|
"mean_token_accuracy": 0.1934338316321373,
|
|
"num_tokens": 4730885.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.471496152877807,
|
|
"epoch": 2.1464435146443517,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004777259970586036,
|
|
"loss": 5.144,
|
|
"mean_token_accuracy": 0.1901230111718178,
|
|
"num_tokens": 4740871.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.515472507476806,
|
|
"epoch": 2.1506276150627617,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004775857692772859,
|
|
"loss": 5.1129,
|
|
"mean_token_accuracy": 0.18737353384494781,
|
|
"num_tokens": 4750603.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.537063264846802,
|
|
"epoch": 2.1548117154811717,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004774451246070257,
|
|
"loss": 5.2272,
|
|
"mean_token_accuracy": 0.188941852748394,
|
|
"num_tokens": 4760681.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.586339664459229,
|
|
"epoch": 2.1589958158995817,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004773040633372471,
|
|
"loss": 5.2295,
|
|
"mean_token_accuracy": 0.19179176092147826,
|
|
"num_tokens": 4770237.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.5697332382202145,
|
|
"epoch": 2.1631799163179917,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047716258575823147,
|
|
"loss": 5.2384,
|
|
"mean_token_accuracy": 0.1939595341682434,
|
|
"num_tokens": 4779022.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.585575962066651,
|
|
"epoch": 2.1673640167364017,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000477020692161117,
|
|
"loss": 5.2653,
|
|
"mean_token_accuracy": 0.19037812799215317,
|
|
"num_tokens": 4788644.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.486926507949829,
|
|
"epoch": 2.1715481171548117,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004768783828378978,
|
|
"loss": 5.2559,
|
|
"mean_token_accuracy": 0.18055640012025834,
|
|
"num_tokens": 4798628.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.680378341674805,
|
|
"epoch": 2.1757322175732217,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004767356580814237,
|
|
"loss": 5.3521,
|
|
"mean_token_accuracy": 0.18168396949768068,
|
|
"num_tokens": 4807727.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.625668621063232,
|
|
"epoch": 2.1799163179916317,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004765925181853993,
|
|
"loss": 5.2347,
|
|
"mean_token_accuracy": 0.18404139429330826,
|
|
"num_tokens": 4817896.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.5362122535705565,
|
|
"epoch": 2.1841004184100417,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004764489634443834,
|
|
"loss": 5.203,
|
|
"mean_token_accuracy": 0.18753628581762313,
|
|
"num_tokens": 4827135.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.595273685455322,
|
|
"epoch": 2.1882845188284517,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047630499415378863,
|
|
"loss": 5.2749,
|
|
"mean_token_accuracy": 0.18363713175058366,
|
|
"num_tokens": 4837554.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.505596733093261,
|
|
"epoch": 2.192468619246862,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004761606106098806,
|
|
"loss": 5.145,
|
|
"mean_token_accuracy": 0.194529926776886,
|
|
"num_tokens": 4846739.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.535862779617309,
|
|
"epoch": 2.196652719665272,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047601581310977747,
|
|
"loss": 5.2005,
|
|
"mean_token_accuracy": 0.188588510453701,
|
|
"num_tokens": 4855707.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.58413462638855,
|
|
"epoch": 2.200836820083682,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004758706019514493,
|
|
"loss": 5.2108,
|
|
"mean_token_accuracy": 0.18735620379447937,
|
|
"num_tokens": 4865098.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.624204826354981,
|
|
"epoch": 2.205020920502092,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004757249774337172,
|
|
"loss": 5.2181,
|
|
"mean_token_accuracy": 0.19089197367429733,
|
|
"num_tokens": 4874426.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.527350902557373,
|
|
"epoch": 2.209205020920502,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047557893985625293,
|
|
"loss": 5.1922,
|
|
"mean_token_accuracy": 0.19252975583076476,
|
|
"num_tokens": 4883240.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.499300289154053,
|
|
"epoch": 2.213389121338912,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004754324895195786,
|
|
"loss": 5.1637,
|
|
"mean_token_accuracy": 0.19454330205917358,
|
|
"num_tokens": 4891989.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.560925626754761,
|
|
"epoch": 2.217573221757322,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004752856267250652,
|
|
"loss": 5.2581,
|
|
"mean_token_accuracy": 0.18870779126882553,
|
|
"num_tokens": 4901546.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.470449066162109,
|
|
"epoch": 2.221757322175732,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047513835177493286,
|
|
"loss": 5.0904,
|
|
"mean_token_accuracy": 0.19583418667316438,
|
|
"num_tokens": 4910222.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.564888286590576,
|
|
"epoch": 2.225941422594142,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004749906649722498,
|
|
"loss": 5.3175,
|
|
"mean_token_accuracy": 0.1812940463423729,
|
|
"num_tokens": 4920184.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.506103372573852,
|
|
"epoch": 2.2301255230125525,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004748425666209316,
|
|
"loss": 5.2072,
|
|
"mean_token_accuracy": 0.19119634926319123,
|
|
"num_tokens": 4930014.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.560202741622925,
|
|
"epoch": 2.2343096234309625,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004746940570257409,
|
|
"loss": 5.1372,
|
|
"mean_token_accuracy": 0.19535336941480635,
|
|
"num_tokens": 4939556.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.487796211242676,
|
|
"epoch": 2.2384937238493725,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004745451364922867,
|
|
"loss": 5.1854,
|
|
"mean_token_accuracy": 0.1893844336271286,
|
|
"num_tokens": 4949066.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.623820877075195,
|
|
"epoch": 2.2426778242677825,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004743958053270233,
|
|
"loss": 5.2064,
|
|
"mean_token_accuracy": 0.189122474193573,
|
|
"num_tokens": 4957881.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.527834463119507,
|
|
"epoch": 2.2468619246861925,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047424606383725037,
|
|
"loss": 5.2471,
|
|
"mean_token_accuracy": 0.1848796620965004,
|
|
"num_tokens": 4967105.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.530368137359619,
|
|
"epoch": 2.2510460251046025,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004740959123311118,
|
|
"loss": 5.1385,
|
|
"mean_token_accuracy": 0.1938617393374443,
|
|
"num_tokens": 4977676.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.444386339187622,
|
|
"epoch": 2.2552301255230125,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047394535111759526,
|
|
"loss": 5.1743,
|
|
"mean_token_accuracy": 0.19278677254915239,
|
|
"num_tokens": 4986163.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.537687158584594,
|
|
"epoch": 2.2594142259414225,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004737943805065314,
|
|
"loss": 5.1644,
|
|
"mean_token_accuracy": 0.1850258156657219,
|
|
"num_tokens": 4995179.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.583955192565918,
|
|
"epoch": 2.2635983263598325,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047364300080859363,
|
|
"loss": 5.2531,
|
|
"mean_token_accuracy": 0.18397597670555116,
|
|
"num_tokens": 5003856.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.575516128540039,
|
|
"epoch": 2.2677824267782425,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047349121233529695,
|
|
"loss": 5.2776,
|
|
"mean_token_accuracy": 0.1824420765042305,
|
|
"num_tokens": 5013506.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.5048760890960695,
|
|
"epoch": 2.2719665271966525,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004733390153989976,
|
|
"loss": 5.2179,
|
|
"mean_token_accuracy": 0.1937140092253685,
|
|
"num_tokens": 5022159.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.570837497711182,
|
|
"epoch": 2.276150627615063,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047318641031289245,
|
|
"loss": 5.1732,
|
|
"mean_token_accuracy": 0.19195488095283508,
|
|
"num_tokens": 5031869.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.389512300491333,
|
|
"epoch": 2.280334728033473,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004730333973910183,
|
|
"loss": 5.034,
|
|
"mean_token_accuracy": 0.19962729662656784,
|
|
"num_tokens": 5039904.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.5626428604125975,
|
|
"epoch": 2.284518828451883,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004728799769482511,
|
|
"loss": 5.3185,
|
|
"mean_token_accuracy": 0.1873888209462166,
|
|
"num_tokens": 5049144.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.518021774291992,
|
|
"epoch": 2.288702928870293,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047272614930030533,
|
|
"loss": 5.109,
|
|
"mean_token_accuracy": 0.1944963350892067,
|
|
"num_tokens": 5058302.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.448442077636718,
|
|
"epoch": 2.292887029288703,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047257191476373375,
|
|
"loss": 5.1256,
|
|
"mean_token_accuracy": 0.1946272850036621,
|
|
"num_tokens": 5067443.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.495798873901367,
|
|
"epoch": 2.297071129707113,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047241727365592603,
|
|
"loss": 5.2548,
|
|
"mean_token_accuracy": 0.18570697903633118,
|
|
"num_tokens": 5076415.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.58405065536499,
|
|
"epoch": 2.301255230125523,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004722622262951089,
|
|
"loss": 5.2106,
|
|
"mean_token_accuracy": 0.18675605803728104,
|
|
"num_tokens": 5085882.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.580031251907348,
|
|
"epoch": 2.305439330543933,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004721067730003448,
|
|
"loss": 5.1856,
|
|
"mean_token_accuracy": 0.19832148998975754,
|
|
"num_tokens": 5094685.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.506541299819946,
|
|
"epoch": 2.309623430962343,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004719509140915317,
|
|
"loss": 5.199,
|
|
"mean_token_accuracy": 0.18793694525957108,
|
|
"num_tokens": 5103885.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.506344604492187,
|
|
"epoch": 2.3138075313807533,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047179464988940205,
|
|
"loss": 5.1269,
|
|
"mean_token_accuracy": 0.20067960619926453,
|
|
"num_tokens": 5112164.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.4316380500793455,
|
|
"epoch": 2.3179916317991633,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047163798071552253,
|
|
"loss": 5.0883,
|
|
"mean_token_accuracy": 0.19918673783540725,
|
|
"num_tokens": 5121913.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.500161266326904,
|
|
"epoch": 2.3221757322175733,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004714809068922931,
|
|
"loss": 5.1064,
|
|
"mean_token_accuracy": 0.19880214780569078,
|
|
"num_tokens": 5130447.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.523188257217408,
|
|
"epoch": 2.3263598326359833,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004713234287429465,
|
|
"loss": 5.2168,
|
|
"mean_token_accuracy": 0.1912735179066658,
|
|
"num_tokens": 5139863.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.416908836364746,
|
|
"epoch": 2.3305439330543933,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004711655465915473,
|
|
"loss": 5.1859,
|
|
"mean_token_accuracy": 0.1934982568025589,
|
|
"num_tokens": 5149945.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.580909681320191,
|
|
"epoch": 2.3347280334728033,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004710072607629917,
|
|
"loss": 5.1424,
|
|
"mean_token_accuracy": 0.1887284204363823,
|
|
"num_tokens": 5160293.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.422379064559936,
|
|
"epoch": 2.3389121338912133,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047084857158300646,
|
|
"loss": 5.1573,
|
|
"mean_token_accuracy": 0.1847264811396599,
|
|
"num_tokens": 5169896.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.501246500015259,
|
|
"epoch": 2.3430962343096233,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047068947937814844,
|
|
"loss": 5.1988,
|
|
"mean_token_accuracy": 0.19358194768428802,
|
|
"num_tokens": 5178898.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.546908760070801,
|
|
"epoch": 2.3472803347280333,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047052998447580374,
|
|
"loss": 5.2638,
|
|
"mean_token_accuracy": 0.18672089725732804,
|
|
"num_tokens": 5188733.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.460029363632202,
|
|
"epoch": 2.3514644351464433,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047037008720418724,
|
|
"loss": 5.1662,
|
|
"mean_token_accuracy": 0.19014161676168442,
|
|
"num_tokens": 5198117.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.557684230804443,
|
|
"epoch": 2.3556485355648533,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004702097878923419,
|
|
"loss": 5.2136,
|
|
"mean_token_accuracy": 0.18946583718061447,
|
|
"num_tokens": 5206791.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.467631816864014,
|
|
"epoch": 2.3598326359832638,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004700490868701377,
|
|
"loss": 5.1099,
|
|
"mean_token_accuracy": 0.1978949174284935,
|
|
"num_tokens": 5216427.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.372316217422485,
|
|
"epoch": 2.3640167364016738,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004698879844682718,
|
|
"loss": 5.1004,
|
|
"mean_token_accuracy": 0.20074017643928527,
|
|
"num_tokens": 5225983.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.44347653388977,
|
|
"epoch": 2.3682008368200838,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046972648101826695,
|
|
"loss": 5.1872,
|
|
"mean_token_accuracy": 0.19352349787950515,
|
|
"num_tokens": 5234547.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.638831281661988,
|
|
"epoch": 2.3723849372384938,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004695645768524712,
|
|
"loss": 5.2427,
|
|
"mean_token_accuracy": 0.1966871440410614,
|
|
"num_tokens": 5243281.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.517685079574585,
|
|
"epoch": 2.3765690376569037,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004694022723040573,
|
|
"loss": 5.1317,
|
|
"mean_token_accuracy": 0.18907570391893386,
|
|
"num_tokens": 5252318.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.516454982757568,
|
|
"epoch": 2.3807531380753137,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046923956770702206,
|
|
"loss": 5.2246,
|
|
"mean_token_accuracy": 0.19897218346595763,
|
|
"num_tokens": 5261781.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.54458532333374,
|
|
"epoch": 2.3849372384937237,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004690764633961852,
|
|
"loss": 5.1144,
|
|
"mean_token_accuracy": 0.19382914155721664,
|
|
"num_tokens": 5270409.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.356353092193603,
|
|
"epoch": 2.3891213389121337,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004689129597071893,
|
|
"loss": 5.1348,
|
|
"mean_token_accuracy": 0.20556528568267823,
|
|
"num_tokens": 5278766.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.553450059890747,
|
|
"epoch": 2.393305439330544,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046874905697649857,
|
|
"loss": 5.2965,
|
|
"mean_token_accuracy": 0.1770468220114708,
|
|
"num_tokens": 5287572.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.499509668350219,
|
|
"epoch": 2.397489539748954,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046858475554139857,
|
|
"loss": 5.1927,
|
|
"mean_token_accuracy": 0.18967376947402953,
|
|
"num_tokens": 5297348.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.550220346450805,
|
|
"epoch": 2.401673640167364,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004684200557399952,
|
|
"loss": 5.2215,
|
|
"mean_token_accuracy": 0.18880867063999177,
|
|
"num_tokens": 5305748.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.474714517593384,
|
|
"epoch": 2.405857740585774,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046825495791121414,
|
|
"loss": 5.1422,
|
|
"mean_token_accuracy": 0.19921002089977263,
|
|
"num_tokens": 5314482.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.510020160675049,
|
|
"epoch": 2.410041841004184,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046808946239480017,
|
|
"loss": 5.2323,
|
|
"mean_token_accuracy": 0.1875918075442314,
|
|
"num_tokens": 5325161.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.456033229827881,
|
|
"epoch": 2.414225941422594,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004679235695313165,
|
|
"loss": 5.2239,
|
|
"mean_token_accuracy": 0.18275309801101686,
|
|
"num_tokens": 5335262.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.48456506729126,
|
|
"epoch": 2.418410041841004,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046775727966214393,
|
|
"loss": 5.1232,
|
|
"mean_token_accuracy": 0.18791476339101792,
|
|
"num_tokens": 5345145.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.384779596328736,
|
|
"epoch": 2.422594142259414,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004675905931294803,
|
|
"loss": 5.1144,
|
|
"mean_token_accuracy": 0.19320453554391862,
|
|
"num_tokens": 5354715.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.4337364673614506,
|
|
"epoch": 2.426778242677824,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004674235102763398,
|
|
"loss": 5.17,
|
|
"mean_token_accuracy": 0.19619799852371217,
|
|
"num_tokens": 5364180.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.475883054733276,
|
|
"epoch": 2.430962343096234,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046725603144655195,
|
|
"loss": 5.1472,
|
|
"mean_token_accuracy": 0.1947048380970955,
|
|
"num_tokens": 5372919.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.427154874801635,
|
|
"epoch": 2.435146443514644,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004670881569847612,
|
|
"loss": 5.1247,
|
|
"mean_token_accuracy": 0.1995283156633377,
|
|
"num_tokens": 5383002.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.548765087127686,
|
|
"epoch": 2.439330543933054,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046691988723642636,
|
|
"loss": 5.3135,
|
|
"mean_token_accuracy": 0.17792350202798843,
|
|
"num_tokens": 5393622.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.5228126525878904,
|
|
"epoch": 2.4435146443514646,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004667512225478195,
|
|
"loss": 5.2065,
|
|
"mean_token_accuracy": 0.1899525359272957,
|
|
"num_tokens": 5403106.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.52912859916687,
|
|
"epoch": 2.4476987447698746,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046658216326602526,
|
|
"loss": 5.2036,
|
|
"mean_token_accuracy": 0.19226169139146804,
|
|
"num_tokens": 5412596.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.5384259700775145,
|
|
"epoch": 2.4518828451882846,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004664127097389406,
|
|
"loss": 5.2422,
|
|
"mean_token_accuracy": 0.19120582193136215,
|
|
"num_tokens": 5422301.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.562191295623779,
|
|
"epoch": 2.4560669456066946,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046624286231527367,
|
|
"loss": 5.1506,
|
|
"mean_token_accuracy": 0.19070896953344346,
|
|
"num_tokens": 5432178.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.396261882781983,
|
|
"epoch": 2.4602510460251046,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004660726213445431,
|
|
"loss": 5.1828,
|
|
"mean_token_accuracy": 0.19135988652706146,
|
|
"num_tokens": 5441162.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.462901020050049,
|
|
"epoch": 2.4644351464435146,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004659019871770774,
|
|
"loss": 5.1087,
|
|
"mean_token_accuracy": 0.20298400074243544,
|
|
"num_tokens": 5450566.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.497792959213257,
|
|
"epoch": 2.4686192468619246,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046573096016401435,
|
|
"loss": 5.1597,
|
|
"mean_token_accuracy": 0.2034789353609085,
|
|
"num_tokens": 5460289.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.438758134841919,
|
|
"epoch": 2.4728033472803346,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004655595406572999,
|
|
"loss": 5.126,
|
|
"mean_token_accuracy": 0.19774290025234223,
|
|
"num_tokens": 5469581.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.4851494312286375,
|
|
"epoch": 2.476987447698745,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000465387729009688,
|
|
"loss": 5.2395,
|
|
"mean_token_accuracy": 0.1940693289041519,
|
|
"num_tokens": 5478005.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.497672700881958,
|
|
"epoch": 2.481171548117155,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046521552557473927,
|
|
"loss": 5.2229,
|
|
"mean_token_accuracy": 0.18679770082235336,
|
|
"num_tokens": 5486963.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.449537849426269,
|
|
"epoch": 2.485355648535565,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046504293070682073,
|
|
"loss": 5.1075,
|
|
"mean_token_accuracy": 0.18713995665311814,
|
|
"num_tokens": 5496368.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.40260009765625,
|
|
"epoch": 2.489539748953975,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046486994476110493,
|
|
"loss": 5.1081,
|
|
"mean_token_accuracy": 0.19878703057765962,
|
|
"num_tokens": 5505706.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.391598129272461,
|
|
"epoch": 2.493723849372385,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046469656809356916,
|
|
"loss": 5.0926,
|
|
"mean_token_accuracy": 0.2008656695485115,
|
|
"num_tokens": 5513945.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.414861249923706,
|
|
"epoch": 2.497907949790795,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046452280106099454,
|
|
"loss": 5.1364,
|
|
"mean_token_accuracy": 0.19161589443683624,
|
|
"num_tokens": 5523831.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.534905290603637,
|
|
"epoch": 2.502092050209205,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004643486440209659,
|
|
"loss": 5.2825,
|
|
"mean_token_accuracy": 0.1891816258430481,
|
|
"num_tokens": 5532705.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.540836668014526,
|
|
"epoch": 2.506276150627615,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004641740973318702,
|
|
"loss": 5.1902,
|
|
"mean_token_accuracy": 0.19652958065271378,
|
|
"num_tokens": 5542793.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.538711309432983,
|
|
"epoch": 2.510460251046025,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004639991613528967,
|
|
"loss": 5.186,
|
|
"mean_token_accuracy": 0.19217222332954406,
|
|
"num_tokens": 5551938.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.510460251046025,
|
|
"eval_entropy": 5.210564913681436,
|
|
"eval_loss": 5.479666709899902,
|
|
"eval_mean_token_accuracy": 0.18513260146408725,
|
|
"eval_num_tokens": 5551938.0,
|
|
"eval_runtime": 2.2843,
|
|
"eval_samples_per_second": 1710.342,
|
|
"eval_steps_per_second": 214.066,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.45005521774292,
|
|
"epoch": 2.514644351464435,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046382383644403543,
|
|
"loss": 5.2161,
|
|
"mean_token_accuracy": 0.18677466958761216,
|
|
"num_tokens": 5561374.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.422075986862183,
|
|
"epoch": 2.518828451882845,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046364812296607676,
|
|
"loss": 5.0773,
|
|
"mean_token_accuracy": 0.19821320623159408,
|
|
"num_tokens": 5570459.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.410913896560669,
|
|
"epoch": 2.523012552301255,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046347202128061094,
|
|
"loss": 5.1824,
|
|
"mean_token_accuracy": 0.19033672660589218,
|
|
"num_tokens": 5579692.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.525054979324341,
|
|
"epoch": 2.5271966527196654,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004632955317500268,
|
|
"loss": 5.1501,
|
|
"mean_token_accuracy": 0.18968389183282852,
|
|
"num_tokens": 5588338.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.479957532882691,
|
|
"epoch": 2.5313807531380754,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046311865473751154,
|
|
"loss": 5.2301,
|
|
"mean_token_accuracy": 0.1925045669078827,
|
|
"num_tokens": 5598589.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.525213527679443,
|
|
"epoch": 2.5355648535564854,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004629413906070495,
|
|
"loss": 5.1945,
|
|
"mean_token_accuracy": 0.1976826325058937,
|
|
"num_tokens": 5608127.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.537798976898193,
|
|
"epoch": 2.5397489539748954,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046276373972342177,
|
|
"loss": 5.2441,
|
|
"mean_token_accuracy": 0.19765941500663758,
|
|
"num_tokens": 5617728.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.436708641052246,
|
|
"epoch": 2.5439330543933054,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004625857024522054,
|
|
"loss": 5.1957,
|
|
"mean_token_accuracy": 0.19046255946159363,
|
|
"num_tokens": 5626823.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.460533714294433,
|
|
"epoch": 2.5481171548117154,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046240727915977236,
|
|
"loss": 5.1682,
|
|
"mean_token_accuracy": 0.1939646139740944,
|
|
"num_tokens": 5636284.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.513768196105957,
|
|
"epoch": 2.5523012552301254,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004622284702132893,
|
|
"loss": 5.3019,
|
|
"mean_token_accuracy": 0.18235254138708115,
|
|
"num_tokens": 5645721.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.4662353515625,
|
|
"epoch": 2.5564853556485354,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046204927598071595,
|
|
"loss": 5.1989,
|
|
"mean_token_accuracy": 0.1968150407075882,
|
|
"num_tokens": 5654734.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.508523845672608,
|
|
"epoch": 2.560669456066946,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046186969683080547,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.2051626041531563,
|
|
"num_tokens": 5664010.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.4541003704071045,
|
|
"epoch": 2.564853556485356,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046168973313310294,
|
|
"loss": 5.1742,
|
|
"mean_token_accuracy": 0.18989036828279496,
|
|
"num_tokens": 5673028.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.436137247085571,
|
|
"epoch": 2.569037656903766,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004615093852579445,
|
|
"loss": 5.1519,
|
|
"mean_token_accuracy": 0.19903206080198288,
|
|
"num_tokens": 5682517.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.3691307544708256,
|
|
"epoch": 2.573221757322176,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046132865357645717,
|
|
"loss": 5.0647,
|
|
"mean_token_accuracy": 0.20081644952297212,
|
|
"num_tokens": 5690920.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.42713737487793,
|
|
"epoch": 2.577405857740586,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046114753846055774,
|
|
"loss": 5.1463,
|
|
"mean_token_accuracy": 0.19409082531929017,
|
|
"num_tokens": 5700336.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.528608608245849,
|
|
"epoch": 2.581589958158996,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000460966040282952,
|
|
"loss": 5.2896,
|
|
"mean_token_accuracy": 0.18710611760616302,
|
|
"num_tokens": 5711085.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.475103759765625,
|
|
"epoch": 2.585774058577406,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046078415941713384,
|
|
"loss": 5.0917,
|
|
"mean_token_accuracy": 0.20023749768733978,
|
|
"num_tokens": 5719560.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.401956367492676,
|
|
"epoch": 2.589958158995816,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004606018962373852,
|
|
"loss": 5.0567,
|
|
"mean_token_accuracy": 0.1984439268708229,
|
|
"num_tokens": 5727865.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.4202680587768555,
|
|
"epoch": 2.594142259414226,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004604192511187741,
|
|
"loss": 5.1923,
|
|
"mean_token_accuracy": 0.18652990013360976,
|
|
"num_tokens": 5737705.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.521626329421997,
|
|
"epoch": 2.598326359832636,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046023622443715483,
|
|
"loss": 5.208,
|
|
"mean_token_accuracy": 0.1806870549917221,
|
|
"num_tokens": 5747057.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.451520776748657,
|
|
"epoch": 2.602510460251046,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046005281656916707,
|
|
"loss": 5.1665,
|
|
"mean_token_accuracy": 0.19002385288476945,
|
|
"num_tokens": 5755793.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.531509876251221,
|
|
"epoch": 2.606694560669456,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004598690278922346,
|
|
"loss": 5.17,
|
|
"mean_token_accuracy": 0.18862478584051132,
|
|
"num_tokens": 5765385.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.31077880859375,
|
|
"epoch": 2.6108786610878663,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045968485878456497,
|
|
"loss": 5.0843,
|
|
"mean_token_accuracy": 0.19352277368307114,
|
|
"num_tokens": 5773947.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.450425624847412,
|
|
"epoch": 2.6150627615062763,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004595003096251486,
|
|
"loss": 5.0826,
|
|
"mean_token_accuracy": 0.19837626814842224,
|
|
"num_tokens": 5782700.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.48174557685852,
|
|
"epoch": 2.6192468619246863,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045931538079375803,
|
|
"loss": 5.2418,
|
|
"mean_token_accuracy": 0.18418097943067552,
|
|
"num_tokens": 5792452.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.499843740463257,
|
|
"epoch": 2.6234309623430963,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045913007267094704,
|
|
"loss": 5.192,
|
|
"mean_token_accuracy": 0.19567284286022185,
|
|
"num_tokens": 5801270.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.493938255310058,
|
|
"epoch": 2.6276150627615062,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004589443856380499,
|
|
"loss": 5.2236,
|
|
"mean_token_accuracy": 0.19769749641418458,
|
|
"num_tokens": 5810704.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.447635078430176,
|
|
"epoch": 2.6317991631799162,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004587583200771807,
|
|
"loss": 5.1646,
|
|
"mean_token_accuracy": 0.1921568065881729,
|
|
"num_tokens": 5819690.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.4061689376831055,
|
|
"epoch": 2.6359832635983262,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004585718763712325,
|
|
"loss": 5.0801,
|
|
"mean_token_accuracy": 0.19526297897100447,
|
|
"num_tokens": 5828912.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.323699903488159,
|
|
"epoch": 2.6401673640167362,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004583850549038764,
|
|
"loss": 5.0257,
|
|
"mean_token_accuracy": 0.1997259423136711,
|
|
"num_tokens": 5836914.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.418633270263672,
|
|
"epoch": 2.6443514644351467,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045819785605956105,
|
|
"loss": 5.1758,
|
|
"mean_token_accuracy": 0.19801366329193115,
|
|
"num_tokens": 5846268.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.426020288467408,
|
|
"epoch": 2.6485355648535567,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004580102802235115,
|
|
"loss": 5.1909,
|
|
"mean_token_accuracy": 0.19660556465387344,
|
|
"num_tokens": 5855029.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.480724096298218,
|
|
"epoch": 2.6527196652719667,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004578223277817286,
|
|
"loss": 5.1801,
|
|
"mean_token_accuracy": 0.18064286708831787,
|
|
"num_tokens": 5865049.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.496839189529419,
|
|
"epoch": 2.6569037656903767,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045763399912098835,
|
|
"loss": 5.1568,
|
|
"mean_token_accuracy": 0.1926622837781906,
|
|
"num_tokens": 5873593.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.351538276672363,
|
|
"epoch": 2.6610878661087867,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045744529462884086,
|
|
"loss": 5.0961,
|
|
"mean_token_accuracy": 0.1953709453344345,
|
|
"num_tokens": 5882562.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.4730151176452635,
|
|
"epoch": 2.6652719665271967,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045725621469360964,
|
|
"loss": 5.2689,
|
|
"mean_token_accuracy": 0.17878875732421876,
|
|
"num_tokens": 5892516.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.48150668144226,
|
|
"epoch": 2.6694560669456067,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004570667597043907,
|
|
"loss": 5.1594,
|
|
"mean_token_accuracy": 0.1930798038840294,
|
|
"num_tokens": 5901494.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.4825843334197994,
|
|
"epoch": 2.6736401673640167,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000456876930051052,
|
|
"loss": 5.1544,
|
|
"mean_token_accuracy": 0.19203907400369644,
|
|
"num_tokens": 5910750.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.530600595474243,
|
|
"epoch": 2.6778242677824267,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004566867261242325,
|
|
"loss": 5.1934,
|
|
"mean_token_accuracy": 0.19413925409317018,
|
|
"num_tokens": 5920277.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.438911008834839,
|
|
"epoch": 2.6820083682008367,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045649614831534127,
|
|
"loss": 5.167,
|
|
"mean_token_accuracy": 0.1914822891354561,
|
|
"num_tokens": 5929880.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.541011190414428,
|
|
"epoch": 2.6861924686192467,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045630519701655677,
|
|
"loss": 5.3199,
|
|
"mean_token_accuracy": 0.18979873806238173,
|
|
"num_tokens": 5939538.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.455945158004761,
|
|
"epoch": 2.6903765690376567,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045611387262082604,
|
|
"loss": 5.1538,
|
|
"mean_token_accuracy": 0.19412111788988112,
|
|
"num_tokens": 5948747.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.419193363189697,
|
|
"epoch": 2.694560669456067,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045592217552186407,
|
|
"loss": 5.1549,
|
|
"mean_token_accuracy": 0.19655723720788956,
|
|
"num_tokens": 5958209.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.483190965652466,
|
|
"epoch": 2.698744769874477,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004557301061141525,
|
|
"loss": 5.1734,
|
|
"mean_token_accuracy": 0.19360368549823762,
|
|
"num_tokens": 5967184.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.387605953216553,
|
|
"epoch": 2.702928870292887,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045553766479293945,
|
|
"loss": 5.1906,
|
|
"mean_token_accuracy": 0.18894344568252563,
|
|
"num_tokens": 5976006.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.314581108093262,
|
|
"epoch": 2.707112970711297,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045534485195423806,
|
|
"loss": 5.0517,
|
|
"mean_token_accuracy": 0.1985267385840416,
|
|
"num_tokens": 5984671.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.575276041030884,
|
|
"epoch": 2.711297071129707,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045515166799482633,
|
|
"loss": 5.1647,
|
|
"mean_token_accuracy": 0.19555417001247405,
|
|
"num_tokens": 5993190.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.459084606170654,
|
|
"epoch": 2.715481171548117,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00045495811331224564,
|
|
"loss": 5.1691,
|
|
"mean_token_accuracy": 0.19051553010940553,
|
|
"num_tokens": 6004080.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.418127870559692,
|
|
"epoch": 2.719665271966527,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004547641883048005,
|
|
"loss": 5.1721,
|
|
"mean_token_accuracy": 0.19425502121448518,
|
|
"num_tokens": 6013212.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.461031866073609,
|
|
"epoch": 2.723849372384937,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004545698933715573,
|
|
"loss": 5.1086,
|
|
"mean_token_accuracy": 0.1971532315015793,
|
|
"num_tokens": 6023444.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.430924892425537,
|
|
"epoch": 2.7280334728033475,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045437522891234386,
|
|
"loss": 5.1171,
|
|
"mean_token_accuracy": 0.19316145032644272,
|
|
"num_tokens": 6032217.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.418096113204956,
|
|
"epoch": 2.7322175732217575,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045418019532774827,
|
|
"loss": 5.0691,
|
|
"mean_token_accuracy": 0.19897682666778566,
|
|
"num_tokens": 6040246.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.444777917861939,
|
|
"epoch": 2.7364016736401675,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045398479301911836,
|
|
"loss": 5.1919,
|
|
"mean_token_accuracy": 0.18612636774778366,
|
|
"num_tokens": 6049447.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.293983793258667,
|
|
"epoch": 2.7405857740585775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045378902238856064,
|
|
"loss": 5.0544,
|
|
"mean_token_accuracy": 0.1953136995434761,
|
|
"num_tokens": 6058294.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.4795948505401615,
|
|
"epoch": 2.7447698744769875,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004535928838389396,
|
|
"loss": 5.1946,
|
|
"mean_token_accuracy": 0.19363477379083632,
|
|
"num_tokens": 6067237.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.530374145507812,
|
|
"epoch": 2.7489539748953975,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004533963777738768,
|
|
"loss": 5.1929,
|
|
"mean_token_accuracy": 0.1908920481801033,
|
|
"num_tokens": 6075599.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.471873998641968,
|
|
"epoch": 2.7531380753138075,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045319950459775024,
|
|
"loss": 5.1788,
|
|
"mean_token_accuracy": 0.18758751302957535,
|
|
"num_tokens": 6085131.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.497594976425171,
|
|
"epoch": 2.7573221757322175,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004530022647156931,
|
|
"loss": 5.131,
|
|
"mean_token_accuracy": 0.1962914362549782,
|
|
"num_tokens": 6094745.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.405278253555298,
|
|
"epoch": 2.7615062761506275,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004528046585335935,
|
|
"loss": 5.189,
|
|
"mean_token_accuracy": 0.1876565173268318,
|
|
"num_tokens": 6104182.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.404895973205567,
|
|
"epoch": 2.7656903765690375,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045260668645809304,
|
|
"loss": 5.063,
|
|
"mean_token_accuracy": 0.20144966542720794,
|
|
"num_tokens": 6113370.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.355315923690796,
|
|
"epoch": 2.7698744769874475,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004524083488965866,
|
|
"loss": 5.1491,
|
|
"mean_token_accuracy": 0.19548142552375794,
|
|
"num_tokens": 6122202.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.440937900543213,
|
|
"epoch": 2.7740585774058575,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004522096462572208,
|
|
"loss": 5.1909,
|
|
"mean_token_accuracy": 0.19046046435832978,
|
|
"num_tokens": 6131081.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.584582901000976,
|
|
"epoch": 2.778242677824268,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004520105789488939,
|
|
"loss": 5.2682,
|
|
"mean_token_accuracy": 0.18785250335931777,
|
|
"num_tokens": 6141707.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.4165846824646,
|
|
"epoch": 2.782426778242678,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045181114738125427,
|
|
"loss": 5.1162,
|
|
"mean_token_accuracy": 0.19086143225431443,
|
|
"num_tokens": 6151651.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.411024045944214,
|
|
"epoch": 2.786610878661088,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045161135196470015,
|
|
"loss": 5.0809,
|
|
"mean_token_accuracy": 0.20366580188274383,
|
|
"num_tokens": 6159928.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.386791753768921,
|
|
"epoch": 2.790794979079498,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004514111931103783,
|
|
"loss": 5.177,
|
|
"mean_token_accuracy": 0.19021530151367189,
|
|
"num_tokens": 6169115.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.482845830917358,
|
|
"epoch": 2.794979079497908,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045121067123018353,
|
|
"loss": 5.1917,
|
|
"mean_token_accuracy": 0.1912764087319374,
|
|
"num_tokens": 6177995.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.464848613739013,
|
|
"epoch": 2.799163179916318,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045100978673675755,
|
|
"loss": 5.1827,
|
|
"mean_token_accuracy": 0.19271366745233537,
|
|
"num_tokens": 6188081.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.504720783233642,
|
|
"epoch": 2.803347280334728,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045080854004348836,
|
|
"loss": 5.1425,
|
|
"mean_token_accuracy": 0.19421978443861007,
|
|
"num_tokens": 6198010.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.370794725418091,
|
|
"epoch": 2.8075313807531384,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045060693156450933,
|
|
"loss": 5.1303,
|
|
"mean_token_accuracy": 0.19802175015211104,
|
|
"num_tokens": 6207164.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.480674743652344,
|
|
"epoch": 2.8117154811715483,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004504049617146983,
|
|
"loss": 5.2537,
|
|
"mean_token_accuracy": 0.1849811553955078,
|
|
"num_tokens": 6216512.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.4581897258758545,
|
|
"epoch": 2.8158995815899583,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045020263090967674,
|
|
"loss": 5.1974,
|
|
"mean_token_accuracy": 0.19351045340299605,
|
|
"num_tokens": 6226196.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.414691305160522,
|
|
"epoch": 2.8200836820083683,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004499999395658088,
|
|
"loss": 5.0801,
|
|
"mean_token_accuracy": 0.1901591569185257,
|
|
"num_tokens": 6234820.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.434601163864135,
|
|
"epoch": 2.8242677824267783,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000449796888100201,
|
|
"loss": 5.1325,
|
|
"mean_token_accuracy": 0.1923603594303131,
|
|
"num_tokens": 6243813.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.550749826431274,
|
|
"epoch": 2.8284518828451883,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044959347693070026,
|
|
"loss": 5.2262,
|
|
"mean_token_accuracy": 0.19507645517587663,
|
|
"num_tokens": 6253376.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.5274067401885985,
|
|
"epoch": 2.8326359832635983,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004493897064758943,
|
|
"loss": 5.2693,
|
|
"mean_token_accuracy": 0.18676251471042632,
|
|
"num_tokens": 6262640.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.539704513549805,
|
|
"epoch": 2.8368200836820083,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004491855771551098,
|
|
"loss": 5.2023,
|
|
"mean_token_accuracy": 0.19597030878067018,
|
|
"num_tokens": 6270908.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.4151534080505375,
|
|
"epoch": 2.8410041841004183,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004489810893884121,
|
|
"loss": 5.1131,
|
|
"mean_token_accuracy": 0.1924937844276428,
|
|
"num_tokens": 6280361.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.441484069824218,
|
|
"epoch": 2.8451882845188283,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004487762435966042,
|
|
"loss": 5.1343,
|
|
"mean_token_accuracy": 0.1965727686882019,
|
|
"num_tokens": 6288832.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.383919286727905,
|
|
"epoch": 2.8493723849372383,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044857104020122585,
|
|
"loss": 5.11,
|
|
"mean_token_accuracy": 0.1955936536192894,
|
|
"num_tokens": 6298553.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.3417943000793455,
|
|
"epoch": 2.8535564853556483,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004483654796245526,
|
|
"loss": 5.0686,
|
|
"mean_token_accuracy": 0.20062152594327926,
|
|
"num_tokens": 6307692.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.5038236618042,
|
|
"epoch": 2.8577405857740583,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000448159562289595,
|
|
"loss": 5.1596,
|
|
"mean_token_accuracy": 0.19322860687971116,
|
|
"num_tokens": 6316729.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.509893608093262,
|
|
"epoch": 2.8619246861924688,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00044795328862009794,
|
|
"loss": 5.2424,
|
|
"mean_token_accuracy": 0.18641323447227479,
|
|
"num_tokens": 6326210.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.31737790107727,
|
|
"epoch": 2.8661087866108788,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044774665904053943,
|
|
"loss": 5.1378,
|
|
"mean_token_accuracy": 0.19454777538776397,
|
|
"num_tokens": 6335156.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.3467284679412845,
|
|
"epoch": 2.8702928870292888,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004475396739761299,
|
|
"loss": 5.0384,
|
|
"mean_token_accuracy": 0.20037504583597182,
|
|
"num_tokens": 6344264.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.44615797996521,
|
|
"epoch": 2.8744769874476988,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044733233385281117,
|
|
"loss": 5.0896,
|
|
"mean_token_accuracy": 0.20376382470130922,
|
|
"num_tokens": 6352997.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.399966907501221,
|
|
"epoch": 2.8786610878661087,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044712463909725626,
|
|
"loss": 5.0983,
|
|
"mean_token_accuracy": 0.1949354276061058,
|
|
"num_tokens": 6362312.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.484350872039795,
|
|
"epoch": 2.8828451882845187,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044691659013686727,
|
|
"loss": 5.2254,
|
|
"mean_token_accuracy": 0.19254852831363678,
|
|
"num_tokens": 6372107.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.407816648483276,
|
|
"epoch": 2.8870292887029287,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004467081873997756,
|
|
"loss": 5.0676,
|
|
"mean_token_accuracy": 0.20338770598173142,
|
|
"num_tokens": 6381306.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.39616961479187,
|
|
"epoch": 2.891213389121339,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044649943131484056,
|
|
"loss": 5.0182,
|
|
"mean_token_accuracy": 0.19823280423879625,
|
|
"num_tokens": 6390404.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.526677751541138,
|
|
"epoch": 2.895397489539749,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044629032231164874,
|
|
"loss": 5.2386,
|
|
"mean_token_accuracy": 0.1838893249630928,
|
|
"num_tokens": 6400366.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.406765794754028,
|
|
"epoch": 2.899581589958159,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004460808608205127,
|
|
"loss": 5.0485,
|
|
"mean_token_accuracy": 0.19383162409067153,
|
|
"num_tokens": 6408955.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.321411371231079,
|
|
"epoch": 2.903765690376569,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004458710472724706,
|
|
"loss": 5.086,
|
|
"mean_token_accuracy": 0.19688250869512558,
|
|
"num_tokens": 6417172.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.438844347000122,
|
|
"epoch": 2.907949790794979,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044566088209928485,
|
|
"loss": 5.1085,
|
|
"mean_token_accuracy": 0.1957206279039383,
|
|
"num_tokens": 6425898.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.357139158248901,
|
|
"epoch": 2.912133891213389,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044545036573344175,
|
|
"loss": 5.1482,
|
|
"mean_token_accuracy": 0.1937362015247345,
|
|
"num_tokens": 6435940.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.37000880241394,
|
|
"epoch": 2.916317991631799,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044523949860815015,
|
|
"loss": 5.0823,
|
|
"mean_token_accuracy": 0.19690680950880052,
|
|
"num_tokens": 6445119.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.419875144958496,
|
|
"epoch": 2.920502092050209,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004450282811573405,
|
|
"loss": 5.1241,
|
|
"mean_token_accuracy": 0.19647807627916336,
|
|
"num_tokens": 6453986.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.394122457504272,
|
|
"epoch": 2.924686192468619,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044481671381566446,
|
|
"loss": 5.099,
|
|
"mean_token_accuracy": 0.19239669740200044,
|
|
"num_tokens": 6463658.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.402914381027221,
|
|
"epoch": 2.928870292887029,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004446047970184937,
|
|
"loss": 5.1334,
|
|
"mean_token_accuracy": 0.20131623446941377,
|
|
"num_tokens": 6472370.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.928870292887029,
|
|
"eval_entropy": 5.18738692718537,
|
|
"eval_loss": 5.393954277038574,
|
|
"eval_mean_token_accuracy": 0.19046805642072895,
|
|
"eval_num_tokens": 6472370.0,
|
|
"eval_runtime": 2.0802,
|
|
"eval_samples_per_second": 1878.19,
|
|
"eval_steps_per_second": 235.074,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.441805744171143,
|
|
"epoch": 2.933054393305439,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044439253120191877,
|
|
"loss": 5.2471,
|
|
"mean_token_accuracy": 0.18391723930835724,
|
|
"num_tokens": 6482642.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.390582752227783,
|
|
"epoch": 2.937238493723849,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044417991680274874,
|
|
"loss": 5.1616,
|
|
"mean_token_accuracy": 0.19022503942251207,
|
|
"num_tokens": 6491818.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.455659961700439,
|
|
"epoch": 2.941422594142259,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004439669542585097,
|
|
"loss": 5.1416,
|
|
"mean_token_accuracy": 0.1938461348414421,
|
|
"num_tokens": 6500908.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.357761192321777,
|
|
"epoch": 2.9456066945606696,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004437536440074444,
|
|
"loss": 5.1143,
|
|
"mean_token_accuracy": 0.19709693044424056,
|
|
"num_tokens": 6510182.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.349863958358765,
|
|
"epoch": 2.9497907949790796,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004435399864885111,
|
|
"loss": 5.097,
|
|
"mean_token_accuracy": 0.20174257606267929,
|
|
"num_tokens": 6519051.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.350916957855224,
|
|
"epoch": 2.9539748953974896,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004433259821413825,
|
|
"loss": 5.0424,
|
|
"mean_token_accuracy": 0.20692880153656007,
|
|
"num_tokens": 6527949.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.486267566680908,
|
|
"epoch": 2.9581589958158996,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044311163140644524,
|
|
"loss": 5.3049,
|
|
"mean_token_accuracy": 0.18809596300125123,
|
|
"num_tokens": 6537227.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.402765083312988,
|
|
"epoch": 2.9623430962343096,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044289693472479867,
|
|
"loss": 5.1028,
|
|
"mean_token_accuracy": 0.18896447569131852,
|
|
"num_tokens": 6546938.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.378043985366821,
|
|
"epoch": 2.9665271966527196,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044268189253825397,
|
|
"loss": 5.1246,
|
|
"mean_token_accuracy": 0.19479287266731263,
|
|
"num_tokens": 6555750.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.3190099716186525,
|
|
"epoch": 2.9707112970711296,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004424665052893334,
|
|
"loss": 5.0364,
|
|
"mean_token_accuracy": 0.20484523475170135,
|
|
"num_tokens": 6565680.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.2810431003570555,
|
|
"epoch": 2.97489539748954,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044225077342126935,
|
|
"loss": 5.0195,
|
|
"mean_token_accuracy": 0.2079988196492195,
|
|
"num_tokens": 6574595.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.455760955810547,
|
|
"epoch": 2.97907949790795,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004420346973780033,
|
|
"loss": 5.166,
|
|
"mean_token_accuracy": 0.1946111351251602,
|
|
"num_tokens": 6583333.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.461076164245606,
|
|
"epoch": 2.98326359832636,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000441818277604185,
|
|
"loss": 5.1826,
|
|
"mean_token_accuracy": 0.19705627113580704,
|
|
"num_tokens": 6592252.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.359698820114136,
|
|
"epoch": 2.98744769874477,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004416015145451715,
|
|
"loss": 5.0176,
|
|
"mean_token_accuracy": 0.1966521769762039,
|
|
"num_tokens": 6600675.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.342548513412476,
|
|
"epoch": 2.99163179916318,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004413844086470264,
|
|
"loss": 5.1093,
|
|
"mean_token_accuracy": 0.196390962600708,
|
|
"num_tokens": 6609722.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.399357604980469,
|
|
"epoch": 2.99581589958159,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004411669603565187,
|
|
"loss": 5.0973,
|
|
"mean_token_accuracy": 0.20202213078737258,
|
|
"num_tokens": 6618415.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.416395568847657,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000440949170121122,
|
|
"loss": 5.1098,
|
|
"mean_token_accuracy": 0.19941746145486833,
|
|
"num_tokens": 6627966.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.414464902877808,
|
|
"epoch": 3.00418410041841,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044073103838901373,
|
|
"loss": 4.7804,
|
|
"mean_token_accuracy": 0.22372852861881257,
|
|
"num_tokens": 6636820.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.332126426696777,
|
|
"epoch": 3.00836820083682,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044051256560907374,
|
|
"loss": 4.8259,
|
|
"mean_token_accuracy": 0.20850012451410294,
|
|
"num_tokens": 6645792.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.421600341796875,
|
|
"epoch": 3.01255230125523,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044029375223088394,
|
|
"loss": 4.9167,
|
|
"mean_token_accuracy": 0.21081550866365434,
|
|
"num_tokens": 6654850.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.304796457290649,
|
|
"epoch": 3.01673640167364,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044007459870472706,
|
|
"loss": 4.8515,
|
|
"mean_token_accuracy": 0.21035699248313905,
|
|
"num_tokens": 6664524.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.293784236907959,
|
|
"epoch": 3.02092050209205,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004398551054815859,
|
|
"loss": 4.7804,
|
|
"mean_token_accuracy": 0.21237717866897582,
|
|
"num_tokens": 6674330.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.3872912406921385,
|
|
"epoch": 3.0251046025104604,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004396352730131422,
|
|
"loss": 4.8619,
|
|
"mean_token_accuracy": 0.2090701013803482,
|
|
"num_tokens": 6683121.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.331309509277344,
|
|
"epoch": 3.0292887029288704,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004394151017517758,
|
|
"loss": 4.8129,
|
|
"mean_token_accuracy": 0.2161374032497406,
|
|
"num_tokens": 6692566.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.409944725036621,
|
|
"epoch": 3.0334728033472804,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004391945921505638,
|
|
"loss": 4.9198,
|
|
"mean_token_accuracy": 0.20837588161230086,
|
|
"num_tokens": 6701577.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.290264987945557,
|
|
"epoch": 3.0376569037656904,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043897374466327944,
|
|
"loss": 4.7792,
|
|
"mean_token_accuracy": 0.22140164375305177,
|
|
"num_tokens": 6710672.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.284568881988525,
|
|
"epoch": 3.0418410041841004,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043875255974439145,
|
|
"loss": 4.7743,
|
|
"mean_token_accuracy": 0.21317554414272308,
|
|
"num_tokens": 6719412.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.272785949707031,
|
|
"epoch": 3.0460251046025104,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004385310378490627,
|
|
"loss": 4.7736,
|
|
"mean_token_accuracy": 0.216214157640934,
|
|
"num_tokens": 6729169.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.296234035491944,
|
|
"epoch": 3.0502092050209204,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043830917943314977,
|
|
"loss": 4.7357,
|
|
"mean_token_accuracy": 0.2184191182255745,
|
|
"num_tokens": 6738549.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.287523412704468,
|
|
"epoch": 3.0543933054393304,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043808698495320154,
|
|
"loss": 4.7959,
|
|
"mean_token_accuracy": 0.20930714756250382,
|
|
"num_tokens": 6747909.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.279695892333985,
|
|
"epoch": 3.0585774058577404,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043786445486645846,
|
|
"loss": 4.8465,
|
|
"mean_token_accuracy": 0.21694675385951995,
|
|
"num_tokens": 6757125.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.335519552230835,
|
|
"epoch": 3.062761506276151,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004376415896308518,
|
|
"loss": 4.83,
|
|
"mean_token_accuracy": 0.21452222168445587,
|
|
"num_tokens": 6765774.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.338800573348999,
|
|
"epoch": 3.066945606694561,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043741838970500234,
|
|
"loss": 4.8693,
|
|
"mean_token_accuracy": 0.21994654089212418,
|
|
"num_tokens": 6774980.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.320406293869018,
|
|
"epoch": 3.071129707112971,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004371948555482196,
|
|
"loss": 4.8588,
|
|
"mean_token_accuracy": 0.21087666153907775,
|
|
"num_tokens": 6784187.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.286651706695556,
|
|
"epoch": 3.075313807531381,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000436970987620501,
|
|
"loss": 4.8877,
|
|
"mean_token_accuracy": 0.21355757117271423,
|
|
"num_tokens": 6792953.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.331783580780029,
|
|
"epoch": 3.079497907949791,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043674678638253066,
|
|
"loss": 4.776,
|
|
"mean_token_accuracy": 0.2178946927189827,
|
|
"num_tokens": 6803004.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.280574321746826,
|
|
"epoch": 3.083682008368201,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004365222522956788,
|
|
"loss": 4.8191,
|
|
"mean_token_accuracy": 0.21685658246278763,
|
|
"num_tokens": 6811883.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.3175147533416744,
|
|
"epoch": 3.087866108786611,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004362973858220004,
|
|
"loss": 4.8368,
|
|
"mean_token_accuracy": 0.2200515538454056,
|
|
"num_tokens": 6820327.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.277409982681275,
|
|
"epoch": 3.092050209205021,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043607218742423447,
|
|
"loss": 4.8446,
|
|
"mean_token_accuracy": 0.2149869292974472,
|
|
"num_tokens": 6829412.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.2733382225036625,
|
|
"epoch": 3.096234309623431,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043584665756580323,
|
|
"loss": 4.8164,
|
|
"mean_token_accuracy": 0.2121194928884506,
|
|
"num_tokens": 6838997.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.3417370319366455,
|
|
"epoch": 3.100418410041841,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004356207967108106,
|
|
"loss": 4.8625,
|
|
"mean_token_accuracy": 0.20787467509508134,
|
|
"num_tokens": 6848957.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.3111471176147464,
|
|
"epoch": 3.104602510460251,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004353946053240422,
|
|
"loss": 4.915,
|
|
"mean_token_accuracy": 0.20897810757160187,
|
|
"num_tokens": 6859688.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.354472303390503,
|
|
"epoch": 3.1087866108786613,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004351680838709635,
|
|
"loss": 4.9535,
|
|
"mean_token_accuracy": 0.19767368584871292,
|
|
"num_tokens": 6869149.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.369803953170776,
|
|
"epoch": 3.1129707112970713,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043494123281771915,
|
|
"loss": 4.9031,
|
|
"mean_token_accuracy": 0.20806351751089097,
|
|
"num_tokens": 6879432.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.284275817871094,
|
|
"epoch": 3.1171548117154813,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043471405263113214,
|
|
"loss": 4.8834,
|
|
"mean_token_accuracy": 0.21226709634065627,
|
|
"num_tokens": 6888398.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.283653306961059,
|
|
"epoch": 3.1213389121338913,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004344865437787027,
|
|
"loss": 4.7997,
|
|
"mean_token_accuracy": 0.2118497833609581,
|
|
"num_tokens": 6898082.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.260298728942871,
|
|
"epoch": 3.1255230125523012,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043425870672860774,
|
|
"loss": 4.7946,
|
|
"mean_token_accuracy": 0.21875353306531906,
|
|
"num_tokens": 6907693.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.411351203918457,
|
|
"epoch": 3.1297071129707112,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004340305419496989,
|
|
"loss": 4.9902,
|
|
"mean_token_accuracy": 0.20211635231971742,
|
|
"num_tokens": 6917095.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.325109052658081,
|
|
"epoch": 3.1338912133891212,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004338020499115029,
|
|
"loss": 4.9084,
|
|
"mean_token_accuracy": 0.20705681294202805,
|
|
"num_tokens": 6926952.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.317350387573242,
|
|
"epoch": 3.1380753138075312,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004335732310842196,
|
|
"loss": 4.846,
|
|
"mean_token_accuracy": 0.21386293172836304,
|
|
"num_tokens": 6936704.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.285439491271973,
|
|
"epoch": 3.1422594142259412,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004333440859387212,
|
|
"loss": 4.8618,
|
|
"mean_token_accuracy": 0.21324080526828765,
|
|
"num_tokens": 6946614.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.3763164520263675,
|
|
"epoch": 3.1464435146443517,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043311461494655165,
|
|
"loss": 4.8903,
|
|
"mean_token_accuracy": 0.21385957598686217,
|
|
"num_tokens": 6955980.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.363683700561523,
|
|
"epoch": 3.1506276150627617,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004328848185799254,
|
|
"loss": 4.8501,
|
|
"mean_token_accuracy": 0.2117926448583603,
|
|
"num_tokens": 6964108.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.298487567901612,
|
|
"epoch": 3.1548117154811717,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004326546973117264,
|
|
"loss": 4.8625,
|
|
"mean_token_accuracy": 0.21872201561927795,
|
|
"num_tokens": 6972895.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.322932004928589,
|
|
"epoch": 3.1589958158995817,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043242425161550723,
|
|
"loss": 4.8606,
|
|
"mean_token_accuracy": 0.2135120615363121,
|
|
"num_tokens": 6982222.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.381505346298217,
|
|
"epoch": 3.1631799163179917,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043219348196548807,
|
|
"loss": 4.8733,
|
|
"mean_token_accuracy": 0.2146986186504364,
|
|
"num_tokens": 6990820.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.283068037033081,
|
|
"epoch": 3.1673640167364017,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004319623888365557,
|
|
"loss": 4.8873,
|
|
"mean_token_accuracy": 0.21023990958929062,
|
|
"num_tokens": 7000846.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.256968259811401,
|
|
"epoch": 3.1715481171548117,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043173097270426275,
|
|
"loss": 4.8268,
|
|
"mean_token_accuracy": 0.21127672195434571,
|
|
"num_tokens": 7009086.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.414501047134399,
|
|
"epoch": 3.1757322175732217,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043149923404482634,
|
|
"loss": 4.8872,
|
|
"mean_token_accuracy": 0.20702462494373322,
|
|
"num_tokens": 7017752.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.287156391143799,
|
|
"epoch": 3.1799163179916317,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043126717333512733,
|
|
"loss": 4.8882,
|
|
"mean_token_accuracy": 0.21704576760530472,
|
|
"num_tokens": 7026535.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.325661373138428,
|
|
"epoch": 3.1841004184100417,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004310347910527094,
|
|
"loss": 4.918,
|
|
"mean_token_accuracy": 0.21344293653964996,
|
|
"num_tokens": 7035462.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.311455535888672,
|
|
"epoch": 3.1882845188284517,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004308020876757778,
|
|
"loss": 4.8211,
|
|
"mean_token_accuracy": 0.21726108193397523,
|
|
"num_tokens": 7043978.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.263794183731079,
|
|
"epoch": 3.192468619246862,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000430569063683199,
|
|
"loss": 4.8337,
|
|
"mean_token_accuracy": 0.21879592537879944,
|
|
"num_tokens": 7053757.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.202440595626831,
|
|
"epoch": 3.196652719665272,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004303357195544986,
|
|
"loss": 4.7548,
|
|
"mean_token_accuracy": 0.21192261278629304,
|
|
"num_tokens": 7062491.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.273592615127564,
|
|
"epoch": 3.200836820083682,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004301020557698615,
|
|
"loss": 4.8129,
|
|
"mean_token_accuracy": 0.21283192783594132,
|
|
"num_tokens": 7072157.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.259838247299195,
|
|
"epoch": 3.205020920502092,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042986807281013013,
|
|
"loss": 4.7874,
|
|
"mean_token_accuracy": 0.21128478795289993,
|
|
"num_tokens": 7080631.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.326618814468384,
|
|
"epoch": 3.209205020920502,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004296337711568037,
|
|
"loss": 4.8604,
|
|
"mean_token_accuracy": 0.21791129559278488,
|
|
"num_tokens": 7089944.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.309530115127563,
|
|
"epoch": 3.213389121338912,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042939915129203745,
|
|
"loss": 4.8937,
|
|
"mean_token_accuracy": 0.21427640467882156,
|
|
"num_tokens": 7099190.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.134705352783203,
|
|
"epoch": 3.217573221757322,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004291642136986413,
|
|
"loss": 4.8565,
|
|
"mean_token_accuracy": 0.20714019536972045,
|
|
"num_tokens": 7107944.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.2829009056091305,
|
|
"epoch": 3.221757322175732,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000428928958860079,
|
|
"loss": 4.7174,
|
|
"mean_token_accuracy": 0.21646514683961868,
|
|
"num_tokens": 7116953.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.21854944229126,
|
|
"epoch": 3.225941422594142,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004286933872604673,
|
|
"loss": 4.7637,
|
|
"mean_token_accuracy": 0.2164536491036415,
|
|
"num_tokens": 7126236.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.293243455886841,
|
|
"epoch": 3.2301255230125525,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004284574993845746,
|
|
"loss": 4.854,
|
|
"mean_token_accuracy": 0.21248100847005844,
|
|
"num_tokens": 7136286.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.309260702133178,
|
|
"epoch": 3.2343096234309625,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00042822129571782025,
|
|
"loss": 4.9048,
|
|
"mean_token_accuracy": 0.2081179141998291,
|
|
"num_tokens": 7145551.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.30060772895813,
|
|
"epoch": 3.2384937238493725,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042798477674627343,
|
|
"loss": 4.8514,
|
|
"mean_token_accuracy": 0.21206263303756714,
|
|
"num_tokens": 7155486.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.304602098464966,
|
|
"epoch": 3.2426778242677825,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042774794295665213,
|
|
"loss": 4.8445,
|
|
"mean_token_accuracy": 0.21740451604127883,
|
|
"num_tokens": 7163984.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.24478759765625,
|
|
"epoch": 3.2468619246861925,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004275107948363223,
|
|
"loss": 4.8142,
|
|
"mean_token_accuracy": 0.2093614861369133,
|
|
"num_tokens": 7172854.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.242204427719116,
|
|
"epoch": 3.2510460251046025,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004272733328732965,
|
|
"loss": 4.8092,
|
|
"mean_token_accuracy": 0.20978806614875795,
|
|
"num_tokens": 7182202.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.31148624420166,
|
|
"epoch": 3.2552301255230125,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042703555755623337,
|
|
"loss": 4.9276,
|
|
"mean_token_accuracy": 0.2063053011894226,
|
|
"num_tokens": 7192062.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.362660026550293,
|
|
"epoch": 3.2594142259414225,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004267974693744362,
|
|
"loss": 4.9359,
|
|
"mean_token_accuracy": 0.2064971521496773,
|
|
"num_tokens": 7201766.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.307424163818359,
|
|
"epoch": 3.2635983263598325,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004265590688178523,
|
|
"loss": 4.844,
|
|
"mean_token_accuracy": 0.21115203946828842,
|
|
"num_tokens": 7211361.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.297022676467895,
|
|
"epoch": 3.2677824267782425,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004263203563770716,
|
|
"loss": 4.8321,
|
|
"mean_token_accuracy": 0.21695517599582673,
|
|
"num_tokens": 7220221.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.317932367324829,
|
|
"epoch": 3.2719665271966525,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000426081332543326,
|
|
"loss": 4.9365,
|
|
"mean_token_accuracy": 0.20593331307172774,
|
|
"num_tokens": 7230134.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.190858840942383,
|
|
"epoch": 3.276150627615063,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004258419978084881,
|
|
"loss": 4.7534,
|
|
"mean_token_accuracy": 0.22077322751283646,
|
|
"num_tokens": 7238713.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.333839082717896,
|
|
"epoch": 3.280334728033473,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004256023526650702,
|
|
"loss": 4.8618,
|
|
"mean_token_accuracy": 0.21216015964746476,
|
|
"num_tokens": 7247603.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.25458173751831,
|
|
"epoch": 3.284518828451883,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004253623976062236,
|
|
"loss": 4.8944,
|
|
"mean_token_accuracy": 0.21344434916973115,
|
|
"num_tokens": 7257452.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.431085109710693,
|
|
"epoch": 3.288702928870293,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004251221331257372,
|
|
"loss": 4.9043,
|
|
"mean_token_accuracy": 0.21185788959264756,
|
|
"num_tokens": 7266401.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.264184141159058,
|
|
"epoch": 3.292887029288703,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004248815597180366,
|
|
"loss": 4.8599,
|
|
"mean_token_accuracy": 0.21911235749721528,
|
|
"num_tokens": 7275836.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.309340286254883,
|
|
"epoch": 3.297071129707113,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042464067787818333,
|
|
"loss": 4.8962,
|
|
"mean_token_accuracy": 0.21380368918180465,
|
|
"num_tokens": 7284854.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.305110931396484,
|
|
"epoch": 3.301255230125523,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004243994881018735,
|
|
"loss": 4.8685,
|
|
"mean_token_accuracy": 0.20804680585861207,
|
|
"num_tokens": 7293421.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.26351146697998,
|
|
"epoch": 3.305439330543933,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004241579908854367,
|
|
"loss": 4.8075,
|
|
"mean_token_accuracy": 0.21481467485427858,
|
|
"num_tokens": 7302184.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.330760288238525,
|
|
"epoch": 3.309623430962343,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042391618672583546,
|
|
"loss": 4.9541,
|
|
"mean_token_accuracy": 0.20623802691698073,
|
|
"num_tokens": 7311596.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.332799100875855,
|
|
"epoch": 3.3138075313807533,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00042367407612066383,
|
|
"loss": 4.8008,
|
|
"mean_token_accuracy": 0.21611622720956802,
|
|
"num_tokens": 7319866.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.261944818496704,
|
|
"epoch": 3.3179916317991633,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004234316595681468,
|
|
"loss": 4.8497,
|
|
"mean_token_accuracy": 0.2179259404540062,
|
|
"num_tokens": 7329327.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.361376476287842,
|
|
"epoch": 3.3221757322175733,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004231889375671382,
|
|
"loss": 5.0105,
|
|
"mean_token_accuracy": 0.20290058553218843,
|
|
"num_tokens": 7339717.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.333584928512574,
|
|
"epoch": 3.3263598326359833,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042294591061712124,
|
|
"loss": 4.8488,
|
|
"mean_token_accuracy": 0.21299046277999878,
|
|
"num_tokens": 7348850.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.2334037780761715,
|
|
"epoch": 3.3305439330543933,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042270257921820623,
|
|
"loss": 4.7989,
|
|
"mean_token_accuracy": 0.21840860098600387,
|
|
"num_tokens": 7357081.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.168159866333008,
|
|
"epoch": 3.3347280334728033,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00042245894387112987,
|
|
"loss": 4.814,
|
|
"mean_token_accuracy": 0.21887375861406327,
|
|
"num_tokens": 7365762.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.300364351272583,
|
|
"epoch": 3.3389121338912133,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004222150050772548,
|
|
"loss": 4.8654,
|
|
"mean_token_accuracy": 0.21250252425670624,
|
|
"num_tokens": 7374870.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.235871648788452,
|
|
"epoch": 3.3430962343096233,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004219707633385677,
|
|
"loss": 4.756,
|
|
"mean_token_accuracy": 0.21517035812139512,
|
|
"num_tokens": 7383836.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.22374997138977,
|
|
"epoch": 3.3472803347280333,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042172621915767886,
|
|
"loss": 4.8484,
|
|
"mean_token_accuracy": 0.219151970744133,
|
|
"num_tokens": 7394059.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.3472803347280333,
|
|
"eval_entropy": 5.051907828005302,
|
|
"eval_loss": 5.386562824249268,
|
|
"eval_mean_token_accuracy": 0.193265353700866,
|
|
"eval_num_tokens": 7394059.0,
|
|
"eval_runtime": 2.2796,
|
|
"eval_samples_per_second": 1713.908,
|
|
"eval_steps_per_second": 214.513,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.202338838577271,
|
|
"epoch": 3.3514644351464433,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042148137303782077,
|
|
"loss": 4.8031,
|
|
"mean_token_accuracy": 0.21724270582199096,
|
|
"num_tokens": 7402947.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.286719417572021,
|
|
"epoch": 3.3556485355648533,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004212362254828475,
|
|
"loss": 4.8796,
|
|
"mean_token_accuracy": 0.2181011125445366,
|
|
"num_tokens": 7413430.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.321802473068237,
|
|
"epoch": 3.3598326359832638,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042099077699723326,
|
|
"loss": 4.8959,
|
|
"mean_token_accuracy": 0.21520495861768724,
|
|
"num_tokens": 7423239.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.284657669067383,
|
|
"epoch": 3.3640167364016738,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004207450280860716,
|
|
"loss": 4.8676,
|
|
"mean_token_accuracy": 0.2169347435235977,
|
|
"num_tokens": 7432519.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.210376834869384,
|
|
"epoch": 3.3682008368200838,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004204989792550742,
|
|
"loss": 4.8413,
|
|
"mean_token_accuracy": 0.21180400550365447,
|
|
"num_tokens": 7442089.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.303813219070435,
|
|
"epoch": 3.3723849372384938,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00042025263101057016,
|
|
"loss": 4.8772,
|
|
"mean_token_accuracy": 0.20973510295152664,
|
|
"num_tokens": 7451059.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.296484899520874,
|
|
"epoch": 3.3765690376569037,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042000598385950427,
|
|
"loss": 4.8182,
|
|
"mean_token_accuracy": 0.21249048113822938,
|
|
"num_tokens": 7458960.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.262817811965943,
|
|
"epoch": 3.3807531380753137,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000419759038309437,
|
|
"loss": 4.9834,
|
|
"mean_token_accuracy": 0.20212535709142684,
|
|
"num_tokens": 7468863.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.334522342681884,
|
|
"epoch": 3.3849372384937237,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041951179486854253,
|
|
"loss": 4.9335,
|
|
"mean_token_accuracy": 0.2125263035297394,
|
|
"num_tokens": 7477204.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.220955085754395,
|
|
"epoch": 3.3891213389121337,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000419264254045608,
|
|
"loss": 4.7658,
|
|
"mean_token_accuracy": 0.21290932595729828,
|
|
"num_tokens": 7487367.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.098130941390991,
|
|
"epoch": 3.393305439330544,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004190164163500326,
|
|
"loss": 4.7714,
|
|
"mean_token_accuracy": 0.2265939474105835,
|
|
"num_tokens": 7495808.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.292756223678589,
|
|
"epoch": 3.397489539748954,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004187682822918266,
|
|
"loss": 4.9079,
|
|
"mean_token_accuracy": 0.203317891061306,
|
|
"num_tokens": 7506119.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.277822875976563,
|
|
"epoch": 3.401673640167364,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041851985238161,
|
|
"loss": 4.8788,
|
|
"mean_token_accuracy": 0.21251827031373977,
|
|
"num_tokens": 7514931.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.152722930908203,
|
|
"epoch": 3.405857740585774,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041827112713061155,
|
|
"loss": 4.6992,
|
|
"mean_token_accuracy": 0.2283136159181595,
|
|
"num_tokens": 7523753.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.227513647079467,
|
|
"epoch": 3.410041841004184,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041802210705066775,
|
|
"loss": 4.7742,
|
|
"mean_token_accuracy": 0.22093135267496108,
|
|
"num_tokens": 7532447.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.23858380317688,
|
|
"epoch": 3.414225941422594,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000417772792654222,
|
|
"loss": 4.9182,
|
|
"mean_token_accuracy": 0.2094281166791916,
|
|
"num_tokens": 7541388.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.281600093841552,
|
|
"epoch": 3.418410041841004,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041752318445432323,
|
|
"loss": 4.9237,
|
|
"mean_token_accuracy": 0.20162102431058884,
|
|
"num_tokens": 7550422.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.274779033660889,
|
|
"epoch": 3.422594142259414,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000417273282964625,
|
|
"loss": 4.7818,
|
|
"mean_token_accuracy": 0.2208615854382515,
|
|
"num_tokens": 7559522.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.27416615486145,
|
|
"epoch": 3.426778242677824,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004170230886993844,
|
|
"loss": 4.8902,
|
|
"mean_token_accuracy": 0.2027203246951103,
|
|
"num_tokens": 7568785.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.265643739700318,
|
|
"epoch": 3.430962343096234,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041677260217346093,
|
|
"loss": 4.8246,
|
|
"mean_token_accuracy": 0.2129284158349037,
|
|
"num_tokens": 7579013.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.233200311660767,
|
|
"epoch": 3.435146443514644,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041652182390231556,
|
|
"loss": 4.8392,
|
|
"mean_token_accuracy": 0.21611061096191406,
|
|
"num_tokens": 7588577.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.239192771911621,
|
|
"epoch": 3.439330543933054,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041627075440200977,
|
|
"loss": 4.7774,
|
|
"mean_token_accuracy": 0.224619160592556,
|
|
"num_tokens": 7597702.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.211343383789062,
|
|
"epoch": 3.4435146443514646,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004160193941892043,
|
|
"loss": 4.8391,
|
|
"mean_token_accuracy": 0.21327215433120728,
|
|
"num_tokens": 7607156.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.236083889007569,
|
|
"epoch": 3.4476987447698746,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004157677437811578,
|
|
"loss": 4.916,
|
|
"mean_token_accuracy": 0.2173907831311226,
|
|
"num_tokens": 7615936.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.33146824836731,
|
|
"epoch": 3.4518828451882846,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004155158036957265,
|
|
"loss": 4.9006,
|
|
"mean_token_accuracy": 0.2133211687207222,
|
|
"num_tokens": 7626568.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.267955875396728,
|
|
"epoch": 3.4560669456066946,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004152635744513626,
|
|
"loss": 4.8364,
|
|
"mean_token_accuracy": 0.2181462198495865,
|
|
"num_tokens": 7635102.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.200731468200684,
|
|
"epoch": 3.4602510460251046,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004150110565671133,
|
|
"loss": 4.7696,
|
|
"mean_token_accuracy": 0.22591931670904158,
|
|
"num_tokens": 7644204.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.310220718383789,
|
|
"epoch": 3.4644351464435146,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041475825056261985,
|
|
"loss": 4.8897,
|
|
"mean_token_accuracy": 0.20997055619955063,
|
|
"num_tokens": 7653585.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.2633789539337155,
|
|
"epoch": 3.4686192468619246,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004145051569581162,
|
|
"loss": 4.882,
|
|
"mean_token_accuracy": 0.2115129977464676,
|
|
"num_tokens": 7662734.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.3008036613464355,
|
|
"epoch": 3.4728033472803346,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004142517762744286,
|
|
"loss": 4.831,
|
|
"mean_token_accuracy": 0.21699218600988388,
|
|
"num_tokens": 7671682.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.32148118019104,
|
|
"epoch": 3.476987447698745,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004139981090329735,
|
|
"loss": 4.9342,
|
|
"mean_token_accuracy": 0.2123778536915779,
|
|
"num_tokens": 7681551.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.290265035629273,
|
|
"epoch": 3.481171548117155,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004137441557557575,
|
|
"loss": 4.8843,
|
|
"mean_token_accuracy": 0.21044880598783494,
|
|
"num_tokens": 7690943.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.2038647651672365,
|
|
"epoch": 3.485355648535565,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041348991696537533,
|
|
"loss": 4.7949,
|
|
"mean_token_accuracy": 0.21960133761167527,
|
|
"num_tokens": 7700376.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.318165302276611,
|
|
"epoch": 3.489539748953975,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004132353931850098,
|
|
"loss": 4.8773,
|
|
"mean_token_accuracy": 0.20765067934989928,
|
|
"num_tokens": 7709845.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.326406478881836,
|
|
"epoch": 3.493723849372385,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004129805849384299,
|
|
"loss": 4.9837,
|
|
"mean_token_accuracy": 0.2085929498076439,
|
|
"num_tokens": 7719908.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.24226746559143,
|
|
"epoch": 3.497907949790795,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004127254927499898,
|
|
"loss": 4.887,
|
|
"mean_token_accuracy": 0.20964450985193253,
|
|
"num_tokens": 7729543.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.336486577987671,
|
|
"epoch": 3.502092050209205,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004124701171446286,
|
|
"loss": 4.9032,
|
|
"mean_token_accuracy": 0.21419214457273483,
|
|
"num_tokens": 7739371.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.297622776031494,
|
|
"epoch": 3.506276150627615,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00041221445864786794,
|
|
"loss": 4.8308,
|
|
"mean_token_accuracy": 0.21566258519887924,
|
|
"num_tokens": 7748027.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.132001256942749,
|
|
"epoch": 3.510460251046025,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000411958517785812,
|
|
"loss": 4.7654,
|
|
"mean_token_accuracy": 0.22347344905138017,
|
|
"num_tokens": 7756588.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.275112247467041,
|
|
"epoch": 3.514644351464435,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041170229508514584,
|
|
"loss": 4.7491,
|
|
"mean_token_accuracy": 0.2179200455546379,
|
|
"num_tokens": 7765222.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.271496677398682,
|
|
"epoch": 3.518828451882845,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004114457910731347,
|
|
"loss": 4.85,
|
|
"mean_token_accuracy": 0.2155054032802582,
|
|
"num_tokens": 7774139.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.169318151473999,
|
|
"epoch": 3.523012552301255,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004111890062776226,
|
|
"loss": 4.744,
|
|
"mean_token_accuracy": 0.21903010606765747,
|
|
"num_tokens": 7783597.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.1600666522979735,
|
|
"epoch": 3.5271966527196654,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004109319412270314,
|
|
"loss": 4.8414,
|
|
"mean_token_accuracy": 0.22158835977315902,
|
|
"num_tokens": 7792711.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.369886016845703,
|
|
"epoch": 3.5313807531380754,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004106745964503594,
|
|
"loss": 4.8963,
|
|
"mean_token_accuracy": 0.20904409289360046,
|
|
"num_tokens": 7801878.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.245414447784424,
|
|
"epoch": 3.5355648535564854,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000410416972477181,
|
|
"loss": 4.7968,
|
|
"mean_token_accuracy": 0.21746836453676224,
|
|
"num_tokens": 7810484.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.310872888565063,
|
|
"epoch": 3.5397489539748954,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041015906983764495,
|
|
"loss": 4.9239,
|
|
"mean_token_accuracy": 0.21283266842365264,
|
|
"num_tokens": 7820521.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.271335601806641,
|
|
"epoch": 3.5439330543933054,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040990088906247326,
|
|
"loss": 4.8406,
|
|
"mean_token_accuracy": 0.21531397104263306,
|
|
"num_tokens": 7829218.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.240310716629028,
|
|
"epoch": 3.5481171548117154,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004096424306829605,
|
|
"loss": 4.8542,
|
|
"mean_token_accuracy": 0.21008332669734955,
|
|
"num_tokens": 7837795.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.302878284454346,
|
|
"epoch": 3.5523012552301254,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004093836952309726,
|
|
"loss": 4.8631,
|
|
"mean_token_accuracy": 0.21237823814153672,
|
|
"num_tokens": 7846973.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.320414733886719,
|
|
"epoch": 3.5564853556485354,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004091246832389453,
|
|
"loss": 4.9025,
|
|
"mean_token_accuracy": 0.2142421141266823,
|
|
"num_tokens": 7856627.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.210345602035522,
|
|
"epoch": 3.560669456066946,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004088653952398837,
|
|
"loss": 4.8151,
|
|
"mean_token_accuracy": 0.220051047205925,
|
|
"num_tokens": 7866308.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.238493013381958,
|
|
"epoch": 3.564853556485356,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040860583176736097,
|
|
"loss": 4.8187,
|
|
"mean_token_accuracy": 0.21831425726413728,
|
|
"num_tokens": 7875350.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.240096807479858,
|
|
"epoch": 3.569037656903766,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004083459933555169,
|
|
"loss": 4.8181,
|
|
"mean_token_accuracy": 0.22090701907873153,
|
|
"num_tokens": 7884276.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.310153388977051,
|
|
"epoch": 3.573221757322176,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004080858805390571,
|
|
"loss": 4.8762,
|
|
"mean_token_accuracy": 0.212080979347229,
|
|
"num_tokens": 7893836.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.272569799423218,
|
|
"epoch": 3.577405857740586,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000407825493853252,
|
|
"loss": 4.8585,
|
|
"mean_token_accuracy": 0.22039470225572586,
|
|
"num_tokens": 7903249.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.229804944992066,
|
|
"epoch": 3.581589958158996,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040756483383393556,
|
|
"loss": 4.9034,
|
|
"mean_token_accuracy": 0.20768687576055528,
|
|
"num_tokens": 7912217.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.298969030380249,
|
|
"epoch": 3.585774058577406,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004073039010175041,
|
|
"loss": 4.9443,
|
|
"mean_token_accuracy": 0.20172094106674193,
|
|
"num_tokens": 7922241.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.264969396591186,
|
|
"epoch": 3.589958158995816,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040704269594091556,
|
|
"loss": 4.8692,
|
|
"mean_token_accuracy": 0.21229377686977385,
|
|
"num_tokens": 7931866.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.310373973846436,
|
|
"epoch": 3.594142259414226,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004067812191416878,
|
|
"loss": 4.8899,
|
|
"mean_token_accuracy": 0.2093264251947403,
|
|
"num_tokens": 7941787.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.33265495300293,
|
|
"epoch": 3.598326359832636,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004065194711578982,
|
|
"loss": 4.9061,
|
|
"mean_token_accuracy": 0.20772757083177568,
|
|
"num_tokens": 7950832.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.255284929275513,
|
|
"epoch": 3.602510460251046,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004062574525281819,
|
|
"loss": 4.9137,
|
|
"mean_token_accuracy": 0.20651647597551345,
|
|
"num_tokens": 7960059.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.298674392700195,
|
|
"epoch": 3.606694560669456,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040599516379173113,
|
|
"loss": 4.8672,
|
|
"mean_token_accuracy": 0.2038848727941513,
|
|
"num_tokens": 7969411.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.276816272735596,
|
|
"epoch": 3.6108786610878663,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040573260548829417,
|
|
"loss": 4.8797,
|
|
"mean_token_accuracy": 0.21113304942846298,
|
|
"num_tokens": 7978446.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.154929685592651,
|
|
"epoch": 3.6150627615062763,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004054697781581734,
|
|
"loss": 4.7622,
|
|
"mean_token_accuracy": 0.21304932683706285,
|
|
"num_tokens": 7987117.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.279610872268677,
|
|
"epoch": 3.6192468619246863,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004052066823422255,
|
|
"loss": 4.8512,
|
|
"mean_token_accuracy": 0.21258513927459716,
|
|
"num_tokens": 7995989.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.244248676300049,
|
|
"epoch": 3.6234309623430963,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004049433185818593,
|
|
"loss": 4.9066,
|
|
"mean_token_accuracy": 0.2103349357843399,
|
|
"num_tokens": 8005486.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.230851507186889,
|
|
"epoch": 3.6276150627615062,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040467968741903507,
|
|
"loss": 4.7966,
|
|
"mean_token_accuracy": 0.21647756546735764,
|
|
"num_tokens": 8014853.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.298684930801391,
|
|
"epoch": 3.6317991631799162,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004044157893962633,
|
|
"loss": 4.8401,
|
|
"mean_token_accuracy": 0.2138268381357193,
|
|
"num_tokens": 8023931.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.265063667297364,
|
|
"epoch": 3.6359832635983262,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004041516250566037,
|
|
"loss": 4.8694,
|
|
"mean_token_accuracy": 0.21532629281282425,
|
|
"num_tokens": 8032659.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.293608379364014,
|
|
"epoch": 3.6401673640167362,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040388719494366404,
|
|
"loss": 4.946,
|
|
"mean_token_accuracy": 0.20897147953510284,
|
|
"num_tokens": 8041857.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.338675832748413,
|
|
"epoch": 3.6443514644351467,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00040362249960159904,
|
|
"loss": 4.9525,
|
|
"mean_token_accuracy": 0.21012792587280274,
|
|
"num_tokens": 8051186.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.257472801208496,
|
|
"epoch": 3.6485355648535567,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040335753957510906,
|
|
"loss": 4.8833,
|
|
"mean_token_accuracy": 0.20727422088384628,
|
|
"num_tokens": 8060396.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.238701248168946,
|
|
"epoch": 3.6527196652719667,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040309231540943937,
|
|
"loss": 4.8148,
|
|
"mean_token_accuracy": 0.21855190247297288,
|
|
"num_tokens": 8068106.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.2845056533813475,
|
|
"epoch": 3.6569037656903767,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004028268276503784,
|
|
"loss": 4.8326,
|
|
"mean_token_accuracy": 0.21222348064184188,
|
|
"num_tokens": 8077878.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.316188764572144,
|
|
"epoch": 3.6610878661087867,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004025610768442576,
|
|
"loss": 4.9678,
|
|
"mean_token_accuracy": 0.20575987547636032,
|
|
"num_tokens": 8086968.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.2372612953186035,
|
|
"epoch": 3.6652719665271967,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004022950635379492,
|
|
"loss": 4.7753,
|
|
"mean_token_accuracy": 0.2245744690299034,
|
|
"num_tokens": 8095999.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.1967182636260985,
|
|
"epoch": 3.6694560669456067,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040202878827886587,
|
|
"loss": 4.7874,
|
|
"mean_token_accuracy": 0.2240486264228821,
|
|
"num_tokens": 8105376.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.2550302028656,
|
|
"epoch": 3.6736401673640167,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004017622516149592,
|
|
"loss": 4.8705,
|
|
"mean_token_accuracy": 0.2172391325235367,
|
|
"num_tokens": 8115024.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.280132818222046,
|
|
"epoch": 3.6778242677824267,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004014954540947189,
|
|
"loss": 4.8785,
|
|
"mean_token_accuracy": 0.21235982477664947,
|
|
"num_tokens": 8124306.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.249938917160034,
|
|
"epoch": 3.6820083682008367,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040122839626717137,
|
|
"loss": 4.8771,
|
|
"mean_token_accuracy": 0.218519726395607,
|
|
"num_tokens": 8133151.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.263608169555664,
|
|
"epoch": 3.6861924686192467,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004009610786818786,
|
|
"loss": 4.8806,
|
|
"mean_token_accuracy": 0.21643056273460387,
|
|
"num_tokens": 8141745.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.186485481262207,
|
|
"epoch": 3.6903765690376567,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040069350188893735,
|
|
"loss": 4.7341,
|
|
"mean_token_accuracy": 0.2246642902493477,
|
|
"num_tokens": 8150884.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.261892366409302,
|
|
"epoch": 3.694560669456067,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040042566643897745,
|
|
"loss": 4.8395,
|
|
"mean_token_accuracy": 0.20635552406311036,
|
|
"num_tokens": 8160138.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.3060273170471195,
|
|
"epoch": 3.698744769874477,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004001575728831613,
|
|
"loss": 4.9175,
|
|
"mean_token_accuracy": 0.21187147349119187,
|
|
"num_tokens": 8169655.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.369554948806763,
|
|
"epoch": 3.702928870292887,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003998892217731824,
|
|
"loss": 4.898,
|
|
"mean_token_accuracy": 0.2194460242986679,
|
|
"num_tokens": 8178846.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.219706726074219,
|
|
"epoch": 3.707112970711297,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000399620613661264,
|
|
"loss": 4.8171,
|
|
"mean_token_accuracy": 0.2108996778726578,
|
|
"num_tokens": 8187525.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.215937185287475,
|
|
"epoch": 3.711297071129707,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003993517491001586,
|
|
"loss": 4.9193,
|
|
"mean_token_accuracy": 0.207285076379776,
|
|
"num_tokens": 8196855.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.190539264678955,
|
|
"epoch": 3.715481171548117,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003990826286431462,
|
|
"loss": 4.7836,
|
|
"mean_token_accuracy": 0.21064182221889496,
|
|
"num_tokens": 8205530.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.294638204574585,
|
|
"epoch": 3.719665271966527,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003988132528440333,
|
|
"loss": 4.8794,
|
|
"mean_token_accuracy": 0.21264730840921403,
|
|
"num_tokens": 8215066.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.304563760757446,
|
|
"epoch": 3.723849372384937,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003985436222571521,
|
|
"loss": 4.8502,
|
|
"mean_token_accuracy": 0.21065464615821838,
|
|
"num_tokens": 8224585.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.266892147064209,
|
|
"epoch": 3.7280334728033475,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000398273737437359,
|
|
"loss": 4.931,
|
|
"mean_token_accuracy": 0.20827907919883729,
|
|
"num_tokens": 8233334.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.267903804779053,
|
|
"epoch": 3.7322175732217575,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039800359894003355,
|
|
"loss": 4.866,
|
|
"mean_token_accuracy": 0.2217654287815094,
|
|
"num_tokens": 8241636.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.294774961471558,
|
|
"epoch": 3.7364016736401675,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003977332073210774,
|
|
"loss": 4.8845,
|
|
"mean_token_accuracy": 0.2120090126991272,
|
|
"num_tokens": 8250273.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.2656162738800045,
|
|
"epoch": 3.7405857740585775,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003974625631369129,
|
|
"loss": 4.9193,
|
|
"mean_token_accuracy": 0.2052695035934448,
|
|
"num_tokens": 8260349.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.314387035369873,
|
|
"epoch": 3.7447698744769875,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003971916669444824,
|
|
"loss": 4.9275,
|
|
"mean_token_accuracy": 0.2096450522542,
|
|
"num_tokens": 8270505.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.338422012329102,
|
|
"epoch": 3.7489539748953975,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039692051930124663,
|
|
"loss": 4.911,
|
|
"mean_token_accuracy": 0.210771045088768,
|
|
"num_tokens": 8278651.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.274332332611084,
|
|
"epoch": 3.7531380753138075,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003966491207651838,
|
|
"loss": 4.8474,
|
|
"mean_token_accuracy": 0.21943927705287933,
|
|
"num_tokens": 8287173.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.338146591186524,
|
|
"epoch": 3.7573221757322175,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039637747189478866,
|
|
"loss": 4.8438,
|
|
"mean_token_accuracy": 0.21564748585224153,
|
|
"num_tokens": 8295785.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.277880859375,
|
|
"epoch": 3.7615062761506275,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003961055732490708,
|
|
"loss": 4.9632,
|
|
"mean_token_accuracy": 0.21075915098190307,
|
|
"num_tokens": 8305534.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.2556273460388185,
|
|
"epoch": 3.7656903765690375,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039583342538755374,
|
|
"loss": 4.8452,
|
|
"mean_token_accuracy": 0.21637595742940902,
|
|
"num_tokens": 8314232.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.7656903765690375,
|
|
"eval_entropy": 5.044442587347362,
|
|
"eval_loss": 5.333103179931641,
|
|
"eval_mean_token_accuracy": 0.19628965473735746,
|
|
"eval_num_tokens": 8314232.0,
|
|
"eval_runtime": 2.0866,
|
|
"eval_samples_per_second": 1872.455,
|
|
"eval_steps_per_second": 234.356,
|
|
"step": 4500
|
|
}
|
|
],
|
|
"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": 1795255463116800.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|