6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.3210444646583552,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691989517211914,
|
|
"epoch": 0.0005350741077639254,
|
|
"grad_norm": 10.75,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8201,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10038.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69193468093872,
|
|
"epoch": 0.0010701482155278507,
|
|
"grad_norm": 11.9375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.789,
|
|
"mean_token_accuracy": 0.00010810811072587967,
|
|
"num_tokens": 21916.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690756225585938,
|
|
"epoch": 0.0016052223232917758,
|
|
"grad_norm": 8.875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5855,
|
|
"mean_token_accuracy": 0.029181174421682953,
|
|
"num_tokens": 33434.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.660381984710693,
|
|
"epoch": 0.0021402964310557014,
|
|
"grad_norm": 5.71875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.3113,
|
|
"mean_token_accuracy": 0.04592931531369686,
|
|
"num_tokens": 44216.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.51073760986328,
|
|
"epoch": 0.0026753705388196263,
|
|
"grad_norm": 3.71875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.1036,
|
|
"mean_token_accuracy": 0.04351861346513033,
|
|
"num_tokens": 54874.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.5190824508667,
|
|
"epoch": 0.0032104446465835517,
|
|
"grad_norm": 3.09375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.9872,
|
|
"mean_token_accuracy": 0.041112912446260454,
|
|
"num_tokens": 65085.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.547387790679931,
|
|
"epoch": 0.003745518754347477,
|
|
"grad_norm": 2.859375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.904,
|
|
"mean_token_accuracy": 0.04395376648753881,
|
|
"num_tokens": 75601.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.474061203002929,
|
|
"epoch": 0.004280592862111403,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.8879,
|
|
"mean_token_accuracy": 0.04307105913758278,
|
|
"num_tokens": 86193.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.531705379486084,
|
|
"epoch": 0.004815666969875327,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8417,
|
|
"mean_token_accuracy": 0.043867765367031096,
|
|
"num_tokens": 96957.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.478270435333252,
|
|
"epoch": 0.005350741077639253,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.8114,
|
|
"mean_token_accuracy": 0.046108495071530343,
|
|
"num_tokens": 108408.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.473760986328125,
|
|
"epoch": 0.005885815185403178,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.7686,
|
|
"mean_token_accuracy": 0.05027508921921253,
|
|
"num_tokens": 119273.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.462549209594727,
|
|
"epoch": 0.006420889293167103,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.6954,
|
|
"mean_token_accuracy": 0.05541543699800968,
|
|
"num_tokens": 130817.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.302016353607177,
|
|
"epoch": 0.006955963400931029,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.6448,
|
|
"mean_token_accuracy": 0.05586838573217392,
|
|
"num_tokens": 142599.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.404022407531738,
|
|
"epoch": 0.007491037508694954,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.5697,
|
|
"mean_token_accuracy": 0.05663685463368893,
|
|
"num_tokens": 153617.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.37060546875,
|
|
"epoch": 0.00802611161645888,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.5239,
|
|
"mean_token_accuracy": 0.05575243458151817,
|
|
"num_tokens": 165280.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.341149616241456,
|
|
"epoch": 0.008561185724222806,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.4643,
|
|
"mean_token_accuracy": 0.055300182476639745,
|
|
"num_tokens": 176896.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.367101669311523,
|
|
"epoch": 0.00909625983198673,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.4019,
|
|
"mean_token_accuracy": 0.05629164129495621,
|
|
"num_tokens": 188000.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.27455005645752,
|
|
"epoch": 0.009631333939750655,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.3493,
|
|
"mean_token_accuracy": 0.057158011198043826,
|
|
"num_tokens": 199360.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.319551277160645,
|
|
"epoch": 0.01016640804751458,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.2957,
|
|
"mean_token_accuracy": 0.052791529521346095,
|
|
"num_tokens": 210214.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.196423625946045,
|
|
"epoch": 0.010701482155278505,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.1506,
|
|
"mean_token_accuracy": 0.06025413721799851,
|
|
"num_tokens": 220567.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.14912462234497,
|
|
"epoch": 0.011236556263042432,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.057,
|
|
"mean_token_accuracy": 0.06051998771727085,
|
|
"num_tokens": 230585.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.099914455413819,
|
|
"epoch": 0.011771630370806356,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.9907,
|
|
"mean_token_accuracy": 0.06339392885565757,
|
|
"num_tokens": 241996.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.083064746856689,
|
|
"epoch": 0.012306704478570282,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9558,
|
|
"mean_token_accuracy": 0.05828723199665546,
|
|
"num_tokens": 253348.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.981218147277833,
|
|
"epoch": 0.012841778586334207,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8376,
|
|
"mean_token_accuracy": 0.05671278983354568,
|
|
"num_tokens": 264239.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.828715324401855,
|
|
"epoch": 0.013376852694098133,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7502,
|
|
"mean_token_accuracy": 0.056628919392824176,
|
|
"num_tokens": 274522.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.77067985534668,
|
|
"epoch": 0.013911926801862057,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6474,
|
|
"mean_token_accuracy": 0.06091504544019699,
|
|
"num_tokens": 284492.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.592880916595458,
|
|
"epoch": 0.014447000909625984,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.6035,
|
|
"mean_token_accuracy": 0.05946716032922268,
|
|
"num_tokens": 294796.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.45958080291748,
|
|
"epoch": 0.014982075017389908,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.574,
|
|
"mean_token_accuracy": 0.05749143101274967,
|
|
"num_tokens": 305252.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.359578514099121,
|
|
"epoch": 0.015517149125153834,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.4804,
|
|
"mean_token_accuracy": 0.056437050178647044,
|
|
"num_tokens": 315505.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.158096408843994,
|
|
"epoch": 0.01605222323291776,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.4819,
|
|
"mean_token_accuracy": 0.05990221984684467,
|
|
"num_tokens": 326997.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.094014644622803,
|
|
"epoch": 0.016587297340681685,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.4165,
|
|
"mean_token_accuracy": 0.05523004196584225,
|
|
"num_tokens": 338089.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.880174732208252,
|
|
"epoch": 0.01712237144844561,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.4077,
|
|
"mean_token_accuracy": 0.05682347491383553,
|
|
"num_tokens": 348536.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.93060073852539,
|
|
"epoch": 0.017657445556209534,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.4181,
|
|
"mean_token_accuracy": 0.058796605467796324,
|
|
"num_tokens": 359776.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.809493732452392,
|
|
"epoch": 0.01819251966397346,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.3472,
|
|
"mean_token_accuracy": 0.06165213547646999,
|
|
"num_tokens": 370810.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.747115898132325,
|
|
"epoch": 0.018727593771737387,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.3511,
|
|
"mean_token_accuracy": 0.05888295993208885,
|
|
"num_tokens": 381788.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.742884063720703,
|
|
"epoch": 0.01926266787950131,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.3376,
|
|
"mean_token_accuracy": 0.0602983683347702,
|
|
"num_tokens": 391984.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.610486602783203,
|
|
"epoch": 0.019797741987265235,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.3311,
|
|
"mean_token_accuracy": 0.058925900235772134,
|
|
"num_tokens": 403430.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.670144844055176,
|
|
"epoch": 0.02033281609502916,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.318,
|
|
"mean_token_accuracy": 0.05627542734146118,
|
|
"num_tokens": 414467.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.56268243789673,
|
|
"epoch": 0.020867890202793088,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.2838,
|
|
"mean_token_accuracy": 0.06462269835174084,
|
|
"num_tokens": 425752.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.562543296813965,
|
|
"epoch": 0.02140296431055701,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.2618,
|
|
"mean_token_accuracy": 0.06471644267439843,
|
|
"num_tokens": 435908.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.58950891494751,
|
|
"epoch": 0.021938038418320937,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.3051,
|
|
"mean_token_accuracy": 0.05973440445959568,
|
|
"num_tokens": 446424.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.575794887542724,
|
|
"epoch": 0.022473112526084863,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.2691,
|
|
"mean_token_accuracy": 0.0648487914353609,
|
|
"num_tokens": 456854.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.506811428070069,
|
|
"epoch": 0.02300818663384879,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.2861,
|
|
"mean_token_accuracy": 0.06673729792237282,
|
|
"num_tokens": 467899.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.497941493988037,
|
|
"epoch": 0.023543260741612712,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.2297,
|
|
"mean_token_accuracy": 0.06340378113090991,
|
|
"num_tokens": 478683.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.574148082733155,
|
|
"epoch": 0.02407833484937664,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.2648,
|
|
"mean_token_accuracy": 0.06170066706836223,
|
|
"num_tokens": 488745.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.4907470703125,
|
|
"epoch": 0.024613408957140565,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.3431,
|
|
"mean_token_accuracy": 0.059930100291967395,
|
|
"num_tokens": 499448.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.599916458129883,
|
|
"epoch": 0.02514848306490449,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.27,
|
|
"mean_token_accuracy": 0.0650645636022091,
|
|
"num_tokens": 510148.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.483675956726074,
|
|
"epoch": 0.025683557172668414,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.1965,
|
|
"mean_token_accuracy": 0.06112907975912094,
|
|
"num_tokens": 522093.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.460341262817384,
|
|
"epoch": 0.02621863128043234,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.2277,
|
|
"mean_token_accuracy": 0.06530554480850696,
|
|
"num_tokens": 532598.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.49462070465088,
|
|
"epoch": 0.026753705388196266,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.1956,
|
|
"mean_token_accuracy": 0.06578520573675632,
|
|
"num_tokens": 542652.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.460633659362793,
|
|
"epoch": 0.027288779495960192,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.2328,
|
|
"mean_token_accuracy": 0.06693280301988125,
|
|
"num_tokens": 552526.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.399180793762207,
|
|
"epoch": 0.027823853603724115,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.2517,
|
|
"mean_token_accuracy": 0.06344069167971611,
|
|
"num_tokens": 563143.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.48379201889038,
|
|
"epoch": 0.02835892771148804,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.2149,
|
|
"mean_token_accuracy": 0.06581672765314579,
|
|
"num_tokens": 573814.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.457896327972412,
|
|
"epoch": 0.028894001819251967,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.1733,
|
|
"mean_token_accuracy": 0.07218663729727268,
|
|
"num_tokens": 585367.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.507985496520996,
|
|
"epoch": 0.02942907592701589,
|
|
"grad_norm": 3.078125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.2318,
|
|
"mean_token_accuracy": 0.06521918512880802,
|
|
"num_tokens": 595902.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.41034107208252,
|
|
"epoch": 0.029964150034779816,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.138,
|
|
"mean_token_accuracy": 0.0642226304858923,
|
|
"num_tokens": 607178.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.431483745574951,
|
|
"epoch": 0.030499224142543743,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.1309,
|
|
"mean_token_accuracy": 0.06817701198160649,
|
|
"num_tokens": 618604.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.404507827758788,
|
|
"epoch": 0.03103429825030767,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.2,
|
|
"mean_token_accuracy": 0.0719299040734768,
|
|
"num_tokens": 628111.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.490602397918702,
|
|
"epoch": 0.03156937235807159,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.2289,
|
|
"mean_token_accuracy": 0.06933129876852036,
|
|
"num_tokens": 639358.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.35900354385376,
|
|
"epoch": 0.03210444646583552,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.0903,
|
|
"mean_token_accuracy": 0.07086234539747238,
|
|
"num_tokens": 649504.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.400328350067138,
|
|
"epoch": 0.032639520573599444,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.1111,
|
|
"mean_token_accuracy": 0.0681851863861084,
|
|
"num_tokens": 659924.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.422282886505126,
|
|
"epoch": 0.03317459468136337,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.15,
|
|
"mean_token_accuracy": 0.06455899253487588,
|
|
"num_tokens": 671120.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.319589424133301,
|
|
"epoch": 0.033709668789127296,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.1985,
|
|
"mean_token_accuracy": 0.06430512070655822,
|
|
"num_tokens": 681956.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.48956241607666,
|
|
"epoch": 0.03424474289689122,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.182,
|
|
"mean_token_accuracy": 0.06503869034349918,
|
|
"num_tokens": 693226.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.33993911743164,
|
|
"epoch": 0.03477981700465514,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.1812,
|
|
"mean_token_accuracy": 0.07134119421243668,
|
|
"num_tokens": 704691.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.362220954895019,
|
|
"epoch": 0.03531489111241907,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.1304,
|
|
"mean_token_accuracy": 0.062357674539089206,
|
|
"num_tokens": 716000.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.332426834106446,
|
|
"epoch": 0.035849965220182994,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.0529,
|
|
"mean_token_accuracy": 0.06962560564279556,
|
|
"num_tokens": 726723.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.32654685974121,
|
|
"epoch": 0.03638503932794692,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.0646,
|
|
"mean_token_accuracy": 0.07072275690734386,
|
|
"num_tokens": 736740.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.333600425720215,
|
|
"epoch": 0.03692011343571085,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.1131,
|
|
"mean_token_accuracy": 0.068069913610816,
|
|
"num_tokens": 747567.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.398985290527344,
|
|
"epoch": 0.03745518754347477,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.0626,
|
|
"mean_token_accuracy": 0.06914932392537594,
|
|
"num_tokens": 757937.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.163579368591309,
|
|
"epoch": 0.0379902616512387,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.0955,
|
|
"mean_token_accuracy": 0.07165054567158222,
|
|
"num_tokens": 768003.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.49596710205078,
|
|
"epoch": 0.03852533575900262,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.1288,
|
|
"mean_token_accuracy": 0.06314317509531975,
|
|
"num_tokens": 779098.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.285042762756348,
|
|
"epoch": 0.039060409866766545,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.099,
|
|
"mean_token_accuracy": 0.06513171158730983,
|
|
"num_tokens": 790002.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.28923921585083,
|
|
"epoch": 0.03959548397453047,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.0887,
|
|
"mean_token_accuracy": 0.06843880005180836,
|
|
"num_tokens": 800444.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.375462627410888,
|
|
"epoch": 0.0401305580822944,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.0323,
|
|
"mean_token_accuracy": 0.0713271751999855,
|
|
"num_tokens": 810912.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.272585010528564,
|
|
"epoch": 0.04066563219005832,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.1198,
|
|
"mean_token_accuracy": 0.0620170421898365,
|
|
"num_tokens": 822809.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.293894195556641,
|
|
"epoch": 0.04120070629782225,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.0433,
|
|
"mean_token_accuracy": 0.07052243873476982,
|
|
"num_tokens": 833362.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.276584053039551,
|
|
"epoch": 0.041735780405586176,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.0465,
|
|
"mean_token_accuracy": 0.07357921116054059,
|
|
"num_tokens": 844239.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.217984676361084,
|
|
"epoch": 0.0422708545133501,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.0493,
|
|
"mean_token_accuracy": 0.06498224660754204,
|
|
"num_tokens": 855178.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.19730110168457,
|
|
"epoch": 0.04280592862111402,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.0369,
|
|
"mean_token_accuracy": 0.07206248976290226,
|
|
"num_tokens": 866120.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.337959480285644,
|
|
"epoch": 0.04334100272887795,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.0361,
|
|
"mean_token_accuracy": 0.07243039160966873,
|
|
"num_tokens": 876569.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.284227752685547,
|
|
"epoch": 0.043876076836641874,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.0951,
|
|
"mean_token_accuracy": 0.06462065242230892,
|
|
"num_tokens": 887580.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.235477828979493,
|
|
"epoch": 0.0444111509444058,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.0165,
|
|
"mean_token_accuracy": 0.0666106827557087,
|
|
"num_tokens": 897922.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.222909069061279,
|
|
"epoch": 0.044946225052169726,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.9827,
|
|
"mean_token_accuracy": 0.07159848175942898,
|
|
"num_tokens": 908522.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.27180290222168,
|
|
"epoch": 0.04548129915993365,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.9946,
|
|
"mean_token_accuracy": 0.06654324233531952,
|
|
"num_tokens": 918253.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.194285202026368,
|
|
"epoch": 0.04601637326769758,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.0034,
|
|
"mean_token_accuracy": 0.07207662984728813,
|
|
"num_tokens": 929150.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.239371013641357,
|
|
"epoch": 0.0465514473754615,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.0511,
|
|
"mean_token_accuracy": 0.07106035277247429,
|
|
"num_tokens": 941041.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.180944156646728,
|
|
"epoch": 0.047086521483225424,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.9757,
|
|
"mean_token_accuracy": 0.07760139927268028,
|
|
"num_tokens": 952714.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.150847339630127,
|
|
"epoch": 0.04762159559098935,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.0514,
|
|
"mean_token_accuracy": 0.06907343231141568,
|
|
"num_tokens": 964330.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.246770858764648,
|
|
"epoch": 0.04815666969875328,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.9805,
|
|
"mean_token_accuracy": 0.06862297914922237,
|
|
"num_tokens": 975257.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.260761165618897,
|
|
"epoch": 0.0486917438065172,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.9876,
|
|
"mean_token_accuracy": 0.07001541927456856,
|
|
"num_tokens": 986335.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.069970512390137,
|
|
"epoch": 0.04922681791428113,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.8883,
|
|
"mean_token_accuracy": 0.07471956759691238,
|
|
"num_tokens": 997521.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.172566795349121,
|
|
"epoch": 0.049761892022045055,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.941,
|
|
"mean_token_accuracy": 0.0669164728373289,
|
|
"num_tokens": 1008383.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.124609184265136,
|
|
"epoch": 0.05029696612980898,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.998,
|
|
"mean_token_accuracy": 0.06806416064500809,
|
|
"num_tokens": 1020552.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.196929836273194,
|
|
"epoch": 0.0508320402375729,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.9718,
|
|
"mean_token_accuracy": 0.07053619101643563,
|
|
"num_tokens": 1031127.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.022743558883667,
|
|
"epoch": 0.05136711434533683,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.8837,
|
|
"mean_token_accuracy": 0.07386223524808884,
|
|
"num_tokens": 1043209.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.2311222076416,
|
|
"epoch": 0.05190218845310075,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 8.0182,
|
|
"mean_token_accuracy": 0.06782832555472851,
|
|
"num_tokens": 1054032.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.030213356018066,
|
|
"epoch": 0.05243726256086468,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.8628,
|
|
"mean_token_accuracy": 0.07288614809513091,
|
|
"num_tokens": 1064565.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.178169107437133,
|
|
"epoch": 0.052972336668628606,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.9975,
|
|
"mean_token_accuracy": 0.0721826508641243,
|
|
"num_tokens": 1076352.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.112149047851563,
|
|
"epoch": 0.05350741077639253,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.9381,
|
|
"mean_token_accuracy": 0.07542271986603737,
|
|
"num_tokens": 1087697.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.137825202941894,
|
|
"epoch": 0.05404248488415646,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.876,
|
|
"mean_token_accuracy": 0.07182962782680988,
|
|
"num_tokens": 1098056.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.017078876495361,
|
|
"epoch": 0.054577558991920384,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.8477,
|
|
"mean_token_accuracy": 0.07691125422716141,
|
|
"num_tokens": 1109219.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.091530561447144,
|
|
"epoch": 0.055112633099684304,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.9312,
|
|
"mean_token_accuracy": 0.0700716007500887,
|
|
"num_tokens": 1120248.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.096777963638306,
|
|
"epoch": 0.05564770720744823,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.9366,
|
|
"mean_token_accuracy": 0.07186660170555115,
|
|
"num_tokens": 1131278.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.089863491058349,
|
|
"epoch": 0.056182781315212156,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.9537,
|
|
"mean_token_accuracy": 0.06224438659846783,
|
|
"num_tokens": 1143147.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.162030553817749,
|
|
"epoch": 0.05671785542297608,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.84,
|
|
"mean_token_accuracy": 0.0739860113710165,
|
|
"num_tokens": 1154316.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.977855777740478,
|
|
"epoch": 0.05725292953074001,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.8967,
|
|
"mean_token_accuracy": 0.06953345276415349,
|
|
"num_tokens": 1164696.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.048009157180786,
|
|
"epoch": 0.057788003638503935,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.8769,
|
|
"mean_token_accuracy": 0.07776758968830108,
|
|
"num_tokens": 1176129.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.093161487579346,
|
|
"epoch": 0.05832307774626786,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.8881,
|
|
"mean_token_accuracy": 0.07334664314985276,
|
|
"num_tokens": 1186499.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.04116907119751,
|
|
"epoch": 0.05885815185403178,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.8557,
|
|
"mean_token_accuracy": 0.07062512040138244,
|
|
"num_tokens": 1196841.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.01404366493225,
|
|
"epoch": 0.059393225961795706,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.8108,
|
|
"mean_token_accuracy": 0.07467551231384277,
|
|
"num_tokens": 1207647.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.952892255783081,
|
|
"epoch": 0.05992830006955963,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.801,
|
|
"mean_token_accuracy": 0.07650438733398915,
|
|
"num_tokens": 1217072.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.061871433258057,
|
|
"epoch": 0.06046337417732356,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.9129,
|
|
"mean_token_accuracy": 0.0698221705853939,
|
|
"num_tokens": 1228564.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.987078428268433,
|
|
"epoch": 0.060998448285087485,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8458,
|
|
"mean_token_accuracy": 0.07497354932129383,
|
|
"num_tokens": 1238755.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.964116954803467,
|
|
"epoch": 0.06153352239285141,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.8004,
|
|
"mean_token_accuracy": 0.0752273216843605,
|
|
"num_tokens": 1248968.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.962353372573853,
|
|
"epoch": 0.06206859650061534,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.7714,
|
|
"mean_token_accuracy": 0.07172017395496369,
|
|
"num_tokens": 1259256.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.952462911605835,
|
|
"epoch": 0.06260367060837926,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.8067,
|
|
"mean_token_accuracy": 0.07487296611070633,
|
|
"num_tokens": 1270469.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.963806390762329,
|
|
"epoch": 0.06313874471614318,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.7694,
|
|
"mean_token_accuracy": 0.07279012463986874,
|
|
"num_tokens": 1280558.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.986854600906372,
|
|
"epoch": 0.06367381882390712,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.8307,
|
|
"mean_token_accuracy": 0.0718784749507904,
|
|
"num_tokens": 1291372.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.02498688697815,
|
|
"epoch": 0.06420889293167104,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.9061,
|
|
"mean_token_accuracy": 0.0674042422324419,
|
|
"num_tokens": 1302521.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.8433699131011965,
|
|
"epoch": 0.06474396703943497,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.7274,
|
|
"mean_token_accuracy": 0.07444682195782662,
|
|
"num_tokens": 1313632.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.005569124221802,
|
|
"epoch": 0.06527904114719889,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.8118,
|
|
"mean_token_accuracy": 0.07571293860673904,
|
|
"num_tokens": 1325247.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.937390947341919,
|
|
"epoch": 0.06581411525496281,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.8015,
|
|
"mean_token_accuracy": 0.075490290671587,
|
|
"num_tokens": 1336026.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.93022665977478,
|
|
"epoch": 0.06634918936272674,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.7122,
|
|
"mean_token_accuracy": 0.0769732192158699,
|
|
"num_tokens": 1346920.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.828203010559082,
|
|
"epoch": 0.06688426347049066,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.8585,
|
|
"mean_token_accuracy": 0.0724661260843277,
|
|
"num_tokens": 1359752.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.050937986373901,
|
|
"epoch": 0.06741933757825459,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.8615,
|
|
"mean_token_accuracy": 0.07357175201177597,
|
|
"num_tokens": 1370874.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.897751331329346,
|
|
"epoch": 0.06795441168601851,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.7624,
|
|
"mean_token_accuracy": 0.07756051793694496,
|
|
"num_tokens": 1382316.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.840310096740723,
|
|
"epoch": 0.06848948579378245,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.7878,
|
|
"mean_token_accuracy": 0.07577594220638276,
|
|
"num_tokens": 1394656.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.928963756561279,
|
|
"epoch": 0.06902455990154636,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.7522,
|
|
"mean_token_accuracy": 0.07719769552350045,
|
|
"num_tokens": 1405061.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.838529491424561,
|
|
"epoch": 0.06955963400931028,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.7183,
|
|
"mean_token_accuracy": 0.08189134374260902,
|
|
"num_tokens": 1415573.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.810403823852539,
|
|
"epoch": 0.07009470811707422,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.7436,
|
|
"mean_token_accuracy": 0.07435674965381622,
|
|
"num_tokens": 1426103.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.886046600341797,
|
|
"epoch": 0.07062978222483814,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.7599,
|
|
"mean_token_accuracy": 0.07234186008572578,
|
|
"num_tokens": 1437256.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.909786796569824,
|
|
"epoch": 0.07116485633260207,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.7241,
|
|
"mean_token_accuracy": 0.0786049198359251,
|
|
"num_tokens": 1446975.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.831602573394775,
|
|
"epoch": 0.07169993044036599,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.7425,
|
|
"mean_token_accuracy": 0.07444198578596115,
|
|
"num_tokens": 1457751.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.829366731643677,
|
|
"epoch": 0.07223500454812992,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.7049,
|
|
"mean_token_accuracy": 0.0754860796034336,
|
|
"num_tokens": 1468186.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.825147390365601,
|
|
"epoch": 0.07277007865589384,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.7508,
|
|
"mean_token_accuracy": 0.07570655383169651,
|
|
"num_tokens": 1479780.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.8603636741638185,
|
|
"epoch": 0.07330515276365776,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.6712,
|
|
"mean_token_accuracy": 0.0832133986055851,
|
|
"num_tokens": 1490565.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.892835903167724,
|
|
"epoch": 0.0738402268714217,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.7227,
|
|
"mean_token_accuracy": 0.07809135504066944,
|
|
"num_tokens": 1501552.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.798111057281494,
|
|
"epoch": 0.07437530097918561,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.7035,
|
|
"mean_token_accuracy": 0.07376831583678722,
|
|
"num_tokens": 1513311.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.795570516586304,
|
|
"epoch": 0.07491037508694955,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.7428,
|
|
"mean_token_accuracy": 0.08057990670204163,
|
|
"num_tokens": 1524142.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.8165356636047365,
|
|
"epoch": 0.07544544919471347,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.7188,
|
|
"mean_token_accuracy": 0.07599906846880913,
|
|
"num_tokens": 1535571.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.865922546386718,
|
|
"epoch": 0.0759805233024774,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.7225,
|
|
"mean_token_accuracy": 0.07798022851347923,
|
|
"num_tokens": 1546528.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.727042579650879,
|
|
"epoch": 0.07651559741024132,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.7386,
|
|
"mean_token_accuracy": 0.07202219888567925,
|
|
"num_tokens": 1557264.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.867890882492065,
|
|
"epoch": 0.07705067151800524,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.6785,
|
|
"mean_token_accuracy": 0.08276100754737854,
|
|
"num_tokens": 1566773.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.699678039550781,
|
|
"epoch": 0.07758574562576917,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.6459,
|
|
"mean_token_accuracy": 0.08662735894322396,
|
|
"num_tokens": 1578173.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.779191970825195,
|
|
"epoch": 0.07812081973353309,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.6221,
|
|
"mean_token_accuracy": 0.08249625936150551,
|
|
"num_tokens": 1588537.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.7231770038604735,
|
|
"epoch": 0.07865589384129702,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.6355,
|
|
"mean_token_accuracy": 0.08746174871921539,
|
|
"num_tokens": 1599159.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.7618378639221195,
|
|
"epoch": 0.07919096794906094,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.6989,
|
|
"mean_token_accuracy": 0.07925031632184983,
|
|
"num_tokens": 1610378.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.720492362976074,
|
|
"epoch": 0.07972604205682488,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.705,
|
|
"mean_token_accuracy": 0.07612368240952491,
|
|
"num_tokens": 1621215.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.786748743057251,
|
|
"epoch": 0.0802611161645888,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.6622,
|
|
"mean_token_accuracy": 0.0790612380951643,
|
|
"num_tokens": 1631170.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.78691840171814,
|
|
"epoch": 0.08079619027235273,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.6574,
|
|
"mean_token_accuracy": 0.07428538724780083,
|
|
"num_tokens": 1641897.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.766008996963501,
|
|
"epoch": 0.08133126438011665,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.7353,
|
|
"mean_token_accuracy": 0.07569916620850563,
|
|
"num_tokens": 1653592.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.616559743881226,
|
|
"epoch": 0.08186633848788057,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6528,
|
|
"mean_token_accuracy": 0.07474047541618348,
|
|
"num_tokens": 1665760.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.830330228805542,
|
|
"epoch": 0.0824014125956445,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.6454,
|
|
"mean_token_accuracy": 0.08570240736007691,
|
|
"num_tokens": 1676211.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.737586927413941,
|
|
"epoch": 0.08293648670340842,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.6666,
|
|
"mean_token_accuracy": 0.0797475405037403,
|
|
"num_tokens": 1686405.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.75378794670105,
|
|
"epoch": 0.08347156081117235,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.6699,
|
|
"mean_token_accuracy": 0.07950277514755726,
|
|
"num_tokens": 1697340.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.726286888122559,
|
|
"epoch": 0.08400663491893627,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.5787,
|
|
"mean_token_accuracy": 0.0861763522028923,
|
|
"num_tokens": 1707250.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.63549427986145,
|
|
"epoch": 0.0845417090267002,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.5653,
|
|
"mean_token_accuracy": 0.0784637302160263,
|
|
"num_tokens": 1718333.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.709765481948852,
|
|
"epoch": 0.08507678313446412,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.6627,
|
|
"mean_token_accuracy": 0.07198781482875347,
|
|
"num_tokens": 1730520.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.7255151748657225,
|
|
"epoch": 0.08561185724222804,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.7149,
|
|
"mean_token_accuracy": 0.07679420076310635,
|
|
"num_tokens": 1741898.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.718593311309815,
|
|
"epoch": 0.08614693134999198,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.6508,
|
|
"mean_token_accuracy": 0.08017718270421029,
|
|
"num_tokens": 1751671.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.577520227432251,
|
|
"epoch": 0.0866820054577559,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.6669,
|
|
"mean_token_accuracy": 0.08265335634350776,
|
|
"num_tokens": 1762076.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.796188259124756,
|
|
"epoch": 0.08721707956551983,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.642,
|
|
"mean_token_accuracy": 0.07752098590135574,
|
|
"num_tokens": 1772666.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.724516534805298,
|
|
"epoch": 0.08775215367328375,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.6572,
|
|
"mean_token_accuracy": 0.08227546997368336,
|
|
"num_tokens": 1783621.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.700971364974976,
|
|
"epoch": 0.08828722778104768,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.7147,
|
|
"mean_token_accuracy": 0.07807004414498805,
|
|
"num_tokens": 1794654.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.651264762878418,
|
|
"epoch": 0.0888223018888116,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.6008,
|
|
"mean_token_accuracy": 0.0848226711153984,
|
|
"num_tokens": 1805853.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.536796283721924,
|
|
"epoch": 0.08935737599657552,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.5896,
|
|
"mean_token_accuracy": 0.0828265130519867,
|
|
"num_tokens": 1817006.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.733750724792481,
|
|
"epoch": 0.08989245010433945,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.583,
|
|
"mean_token_accuracy": 0.07734453342854977,
|
|
"num_tokens": 1827166.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.7373639106750485,
|
|
"epoch": 0.09042752421210337,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.6809,
|
|
"mean_token_accuracy": 0.07751412987709046,
|
|
"num_tokens": 1837182.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.567351484298706,
|
|
"epoch": 0.0909625983198673,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.6388,
|
|
"mean_token_accuracy": 0.07895995602011681,
|
|
"num_tokens": 1849034.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.595164203643799,
|
|
"epoch": 0.09149767242763122,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.5075,
|
|
"mean_token_accuracy": 0.08833360373973846,
|
|
"num_tokens": 1859033.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.610419750213623,
|
|
"epoch": 0.09203274653539516,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.6121,
|
|
"mean_token_accuracy": 0.08088177517056465,
|
|
"num_tokens": 1870274.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.5731220722198485,
|
|
"epoch": 0.09256782064315908,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.5611,
|
|
"mean_token_accuracy": 0.08315053284168243,
|
|
"num_tokens": 1881306.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.561473751068116,
|
|
"epoch": 0.093102894750923,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.5356,
|
|
"mean_token_accuracy": 0.08015808910131454,
|
|
"num_tokens": 1892464.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.661314296722412,
|
|
"epoch": 0.09363796885868693,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.6219,
|
|
"mean_token_accuracy": 0.07585933655500413,
|
|
"num_tokens": 1903743.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.607653522491455,
|
|
"epoch": 0.09417304296645085,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.5902,
|
|
"mean_token_accuracy": 0.08468717783689499,
|
|
"num_tokens": 1915253.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.622669744491577,
|
|
"epoch": 0.09470811707421478,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.5473,
|
|
"mean_token_accuracy": 0.08187699615955353,
|
|
"num_tokens": 1926113.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.57468581199646,
|
|
"epoch": 0.0952431911819787,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.6456,
|
|
"mean_token_accuracy": 0.08173620626330376,
|
|
"num_tokens": 1937718.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.625470066070557,
|
|
"epoch": 0.09577826528974263,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.6334,
|
|
"mean_token_accuracy": 0.0791581004858017,
|
|
"num_tokens": 1948307.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.575066232681275,
|
|
"epoch": 0.09631333939750655,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.4358,
|
|
"mean_token_accuracy": 0.09009880647063255,
|
|
"num_tokens": 1957861.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.480437755584717,
|
|
"epoch": 0.09684841350527049,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.4911,
|
|
"mean_token_accuracy": 0.08970233350992203,
|
|
"num_tokens": 1967984.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.55365891456604,
|
|
"epoch": 0.0973834876130344,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.4885,
|
|
"mean_token_accuracy": 0.08413884043693542,
|
|
"num_tokens": 1978608.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.550483322143554,
|
|
"epoch": 0.09791856172079832,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.547,
|
|
"mean_token_accuracy": 0.08248279690742492,
|
|
"num_tokens": 1989355.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.514042758941651,
|
|
"epoch": 0.09845363582856226,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.5303,
|
|
"mean_token_accuracy": 0.08344742506742478,
|
|
"num_tokens": 1999754.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.540969705581665,
|
|
"epoch": 0.09898870993632618,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.5232,
|
|
"mean_token_accuracy": 0.0832397285848856,
|
|
"num_tokens": 2010877.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.635095930099487,
|
|
"epoch": 0.09952378404409011,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.6286,
|
|
"mean_token_accuracy": 0.08003128916025162,
|
|
"num_tokens": 2022301.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.549521160125733,
|
|
"epoch": 0.10005885815185403,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.5119,
|
|
"mean_token_accuracy": 0.08251530304551125,
|
|
"num_tokens": 2033537.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.477602958679199,
|
|
"epoch": 0.10059393225961796,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.4502,
|
|
"mean_token_accuracy": 0.0781034879386425,
|
|
"num_tokens": 2044360.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.583854675292969,
|
|
"epoch": 0.10112900636738188,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.5601,
|
|
"mean_token_accuracy": 0.08421726487576961,
|
|
"num_tokens": 2055000.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.492854499816895,
|
|
"epoch": 0.1016640804751458,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.5467,
|
|
"mean_token_accuracy": 0.08681007251143455,
|
|
"num_tokens": 2064807.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.5536168098449705,
|
|
"epoch": 0.10219915458290973,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.5102,
|
|
"mean_token_accuracy": 0.08256960362195968,
|
|
"num_tokens": 2075618.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.505359792709351,
|
|
"epoch": 0.10273422869067365,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.5116,
|
|
"mean_token_accuracy": 0.08244576305150986,
|
|
"num_tokens": 2086226.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.541278457641601,
|
|
"epoch": 0.10326930279843759,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.5116,
|
|
"mean_token_accuracy": 0.08005514740943909,
|
|
"num_tokens": 2097222.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.481921100616455,
|
|
"epoch": 0.1038043769062015,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.541,
|
|
"mean_token_accuracy": 0.08661947920918464,
|
|
"num_tokens": 2108919.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.46715989112854,
|
|
"epoch": 0.10433945101396544,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.4939,
|
|
"mean_token_accuracy": 0.08317564725875855,
|
|
"num_tokens": 2120215.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.5228071212768555,
|
|
"epoch": 0.10487452512172936,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.5188,
|
|
"mean_token_accuracy": 0.0818349651992321,
|
|
"num_tokens": 2131503.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.487857866287231,
|
|
"epoch": 0.10540959922949328,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.4588,
|
|
"mean_token_accuracy": 0.08936587274074555,
|
|
"num_tokens": 2141551.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.4559633255004885,
|
|
"epoch": 0.10594467333725721,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.4615,
|
|
"mean_token_accuracy": 0.08929954171180725,
|
|
"num_tokens": 2151821.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.451591444015503,
|
|
"epoch": 0.10647974744502113,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.5065,
|
|
"mean_token_accuracy": 0.08539062738418579,
|
|
"num_tokens": 2162474.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.53413724899292,
|
|
"epoch": 0.10701482155278506,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.5602,
|
|
"mean_token_accuracy": 0.08457510098814965,
|
|
"num_tokens": 2173985.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.5422203063964846,
|
|
"epoch": 0.10754989566054898,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999999979210101,
|
|
"loss": 7.5944,
|
|
"mean_token_accuracy": 0.07782966643571854,
|
|
"num_tokens": 2186432.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.4759071350097654,
|
|
"epoch": 0.10808496976831292,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999999894751138,
|
|
"loss": 7.4463,
|
|
"mean_token_accuracy": 0.0834141232073307,
|
|
"num_tokens": 2197085.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.424153137207031,
|
|
"epoch": 0.10862004387607684,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999999745323744,
|
|
"loss": 7.4486,
|
|
"mean_token_accuracy": 0.08975227214396,
|
|
"num_tokens": 2207162.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.464861583709717,
|
|
"epoch": 0.10915511798384077,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999999530927923,
|
|
"loss": 7.5469,
|
|
"mean_token_accuracy": 0.07938519641757011,
|
|
"num_tokens": 2218714.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.515410757064819,
|
|
"epoch": 0.10969019209160469,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999999251563682,
|
|
"loss": 7.4879,
|
|
"mean_token_accuracy": 0.08599804565310479,
|
|
"num_tokens": 2229711.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.45625114440918,
|
|
"epoch": 0.11022526619936861,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999998907231029,
|
|
"loss": 7.4912,
|
|
"mean_token_accuracy": 0.08116911202669144,
|
|
"num_tokens": 2239885.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.481421899795532,
|
|
"epoch": 0.11076034030713254,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999998497929974,
|
|
"loss": 7.4923,
|
|
"mean_token_accuracy": 0.08827452957630158,
|
|
"num_tokens": 2250221.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.4760716438293455,
|
|
"epoch": 0.11129541441489646,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999998023660527,
|
|
"loss": 7.5526,
|
|
"mean_token_accuracy": 0.07815601006150245,
|
|
"num_tokens": 2261059.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.454020214080811,
|
|
"epoch": 0.11183048852266039,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999997484422705,
|
|
"loss": 7.4273,
|
|
"mean_token_accuracy": 0.08656688034534454,
|
|
"num_tokens": 2271174.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.404057025909424,
|
|
"epoch": 0.11236556263042431,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999996880216521,
|
|
"loss": 7.5037,
|
|
"mean_token_accuracy": 0.08764296993613244,
|
|
"num_tokens": 2282431.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.506435585021973,
|
|
"epoch": 0.11290063673818825,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999996211041994,
|
|
"loss": 7.4922,
|
|
"mean_token_accuracy": 0.08441931083798408,
|
|
"num_tokens": 2292891.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.472446441650391,
|
|
"epoch": 0.11343571084595216,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999995476899141,
|
|
"loss": 7.5049,
|
|
"mean_token_accuracy": 0.08200813457369804,
|
|
"num_tokens": 2303764.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.319696569442749,
|
|
"epoch": 0.11397078495371608,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999994677787987,
|
|
"loss": 7.4002,
|
|
"mean_token_accuracy": 0.09418007209897042,
|
|
"num_tokens": 2314172.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.449784660339356,
|
|
"epoch": 0.11450585906148002,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999993813708551,
|
|
"loss": 7.3939,
|
|
"mean_token_accuracy": 0.08950763419270516,
|
|
"num_tokens": 2324848.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.320974636077881,
|
|
"epoch": 0.11504093316924394,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000499999288466086,
|
|
"loss": 7.4332,
|
|
"mean_token_accuracy": 0.08251803517341613,
|
|
"num_tokens": 2336601.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.46982364654541,
|
|
"epoch": 0.11557600727700787,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999991890644941,
|
|
"loss": 7.4167,
|
|
"mean_token_accuracy": 0.0846226740628481,
|
|
"num_tokens": 2347281.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.453939628601074,
|
|
"epoch": 0.11611108138477179,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999990831660822,
|
|
"loss": 7.4224,
|
|
"mean_token_accuracy": 0.0908889777958393,
|
|
"num_tokens": 2358199.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.446761226654052,
|
|
"epoch": 0.11664615549253572,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999989707708534,
|
|
"loss": 7.4877,
|
|
"mean_token_accuracy": 0.08379835970699787,
|
|
"num_tokens": 2369092.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.3978382587432865,
|
|
"epoch": 0.11718122960029964,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999988518788111,
|
|
"loss": 7.458,
|
|
"mean_token_accuracy": 0.08341244608163834,
|
|
"num_tokens": 2378849.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.428147649765014,
|
|
"epoch": 0.11771630370806356,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999987264899583,
|
|
"loss": 7.4385,
|
|
"mean_token_accuracy": 0.09159169495105743,
|
|
"num_tokens": 2389587.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.492716360092163,
|
|
"epoch": 0.1182513778158275,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000499998594604299,
|
|
"loss": 7.434,
|
|
"mean_token_accuracy": 0.08649549037218093,
|
|
"num_tokens": 2400442.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.284411334991455,
|
|
"epoch": 0.11878645192359141,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000499998456221837,
|
|
"loss": 7.4154,
|
|
"mean_token_accuracy": 0.08677608072757721,
|
|
"num_tokens": 2410771.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.476975584030152,
|
|
"epoch": 0.11932152603135535,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999983113425762,
|
|
"loss": 7.4299,
|
|
"mean_token_accuracy": 0.09064068272709846,
|
|
"num_tokens": 2421112.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.400354480743408,
|
|
"epoch": 0.11985660013911927,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999981599665207,
|
|
"loss": 7.4163,
|
|
"mean_token_accuracy": 0.08564619868993759,
|
|
"num_tokens": 2431891.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.315796279907227,
|
|
"epoch": 0.1203916742468832,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999980020936748,
|
|
"loss": 7.3479,
|
|
"mean_token_accuracy": 0.08722568973898888,
|
|
"num_tokens": 2442987.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.354978418350219,
|
|
"epoch": 0.12092674835464712,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999978377240434,
|
|
"loss": 7.4002,
|
|
"mean_token_accuracy": 0.08564592450857163,
|
|
"num_tokens": 2452703.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.375062036514282,
|
|
"epoch": 0.12146182246241104,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499997666857631,
|
|
"loss": 7.4598,
|
|
"mean_token_accuracy": 0.08786297589540482,
|
|
"num_tokens": 2462975.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.394401025772095,
|
|
"epoch": 0.12199689657017497,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999974894944426,
|
|
"loss": 7.4003,
|
|
"mean_token_accuracy": 0.08598766103386879,
|
|
"num_tokens": 2473853.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.384897661209107,
|
|
"epoch": 0.12253197067793889,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999973056344832,
|
|
"loss": 7.3822,
|
|
"mean_token_accuracy": 0.08470982015132904,
|
|
"num_tokens": 2485633.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.337519359588623,
|
|
"epoch": 0.12306704478570282,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999971152777583,
|
|
"loss": 7.3414,
|
|
"mean_token_accuracy": 0.08668373227119446,
|
|
"num_tokens": 2496922.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.443722057342529,
|
|
"epoch": 0.12360211889346674,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999969184242733,
|
|
"loss": 7.4147,
|
|
"mean_token_accuracy": 0.0896741084754467,
|
|
"num_tokens": 2507890.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.319026279449463,
|
|
"epoch": 0.12413719300123068,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499996715074034,
|
|
"loss": 7.4076,
|
|
"mean_token_accuracy": 0.09130077436566353,
|
|
"num_tokens": 2519749.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.3418297290802,
|
|
"epoch": 0.1246722671089946,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999965052270461,
|
|
"loss": 7.4121,
|
|
"mean_token_accuracy": 0.09187431186437607,
|
|
"num_tokens": 2530666.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.364394998550415,
|
|
"epoch": 0.12520734121675853,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999962888833157,
|
|
"loss": 7.4155,
|
|
"mean_token_accuracy": 0.08915347754955291,
|
|
"num_tokens": 2541594.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.363469171524048,
|
|
"epoch": 0.12574241532452243,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999960660428491,
|
|
"loss": 7.3946,
|
|
"mean_token_accuracy": 0.0930194191634655,
|
|
"num_tokens": 2552330.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.449467992782592,
|
|
"epoch": 0.12627748943228637,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999958367056526,
|
|
"loss": 7.4494,
|
|
"mean_token_accuracy": 0.08116971924901009,
|
|
"num_tokens": 2564601.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.422946262359619,
|
|
"epoch": 0.1268125635400503,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499995600871733,
|
|
"loss": 7.4242,
|
|
"mean_token_accuracy": 0.08717223256826401,
|
|
"num_tokens": 2575539.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.284857511520386,
|
|
"epoch": 0.12734763764781423,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999953585410971,
|
|
"loss": 7.3588,
|
|
"mean_token_accuracy": 0.08616523109376431,
|
|
"num_tokens": 2586635.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.405594778060913,
|
|
"epoch": 0.12788271175557814,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999951097137518,
|
|
"loss": 7.388,
|
|
"mean_token_accuracy": 0.08991223797202111,
|
|
"num_tokens": 2596814.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.325789880752564,
|
|
"epoch": 0.12841778586334207,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999948543897044,
|
|
"loss": 7.4089,
|
|
"mean_token_accuracy": 0.08616380542516708,
|
|
"num_tokens": 2608087.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.498183393478394,
|
|
"epoch": 0.128952859971106,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999945925689621,
|
|
"loss": 7.41,
|
|
"mean_token_accuracy": 0.08784973174333573,
|
|
"num_tokens": 2619377.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.285972166061401,
|
|
"epoch": 0.12948793407886994,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999943242515325,
|
|
"loss": 7.314,
|
|
"mean_token_accuracy": 0.08880106881260871,
|
|
"num_tokens": 2629089.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.320152521133423,
|
|
"epoch": 0.13002300818663384,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999940494374236,
|
|
"loss": 7.3835,
|
|
"mean_token_accuracy": 0.08867338374257087,
|
|
"num_tokens": 2639743.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.45218596458435,
|
|
"epoch": 0.13055808229439778,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499993768126643,
|
|
"loss": 7.4254,
|
|
"mean_token_accuracy": 0.082533248513937,
|
|
"num_tokens": 2650631.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.268480491638184,
|
|
"epoch": 0.1310931564021617,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000499993480319199,
|
|
"loss": 7.3811,
|
|
"mean_token_accuracy": 0.0892926201224327,
|
|
"num_tokens": 2660197.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.394870710372925,
|
|
"epoch": 0.13162823050992561,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999931860150999,
|
|
"loss": 7.3556,
|
|
"mean_token_accuracy": 0.08826216235756874,
|
|
"num_tokens": 2670162.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.390807247161865,
|
|
"epoch": 0.13216330461768955,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999928852143541,
|
|
"loss": 7.3554,
|
|
"mean_token_accuracy": 0.08857011944055557,
|
|
"num_tokens": 2681233.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.200170421600342,
|
|
"epoch": 0.13269837872545348,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999925779169703,
|
|
"loss": 7.3187,
|
|
"mean_token_accuracy": 0.09283828288316727,
|
|
"num_tokens": 2692145.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.413662004470825,
|
|
"epoch": 0.13323345283321741,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999922641229575,
|
|
"loss": 7.3447,
|
|
"mean_token_accuracy": 0.09158495515584945,
|
|
"num_tokens": 2703192.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.324686527252197,
|
|
"epoch": 0.13376852694098132,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999919438323248,
|
|
"loss": 7.3597,
|
|
"mean_token_accuracy": 0.08890319019556045,
|
|
"num_tokens": 2714804.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.333380794525146,
|
|
"epoch": 0.13430360104874525,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999916170450812,
|
|
"loss": 7.3259,
|
|
"mean_token_accuracy": 0.09193522408604622,
|
|
"num_tokens": 2724340.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.259483861923218,
|
|
"epoch": 0.13483867515650919,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999912837612364,
|
|
"loss": 7.3341,
|
|
"mean_token_accuracy": 0.09227988123893738,
|
|
"num_tokens": 2735767.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.455391693115234,
|
|
"epoch": 0.1353737492642731,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999909439807997,
|
|
"loss": 7.3851,
|
|
"mean_token_accuracy": 0.08482631966471672,
|
|
"num_tokens": 2747658.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.264850521087647,
|
|
"epoch": 0.13590882337203702,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999905977037813,
|
|
"loss": 7.2465,
|
|
"mean_token_accuracy": 0.09336961582303047,
|
|
"num_tokens": 2757901.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.284139823913574,
|
|
"epoch": 0.13644389747980096,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999902449301909,
|
|
"loss": 7.2989,
|
|
"mean_token_accuracy": 0.0874991238117218,
|
|
"num_tokens": 2768945.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.24199595451355,
|
|
"epoch": 0.1369789715875649,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004999898856600387,
|
|
"loss": 7.271,
|
|
"mean_token_accuracy": 0.103138717263937,
|
|
"num_tokens": 2780276.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.350469398498535,
|
|
"epoch": 0.1375140456953288,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999895198933354,
|
|
"loss": 7.3416,
|
|
"mean_token_accuracy": 0.09537098631262779,
|
|
"num_tokens": 2791183.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.342680644989014,
|
|
"epoch": 0.13804911980309273,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999891476300911,
|
|
"loss": 7.2607,
|
|
"mean_token_accuracy": 0.08925738111138344,
|
|
"num_tokens": 2801928.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.263091659545898,
|
|
"epoch": 0.13858419391085666,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000499988768870317,
|
|
"loss": 7.3525,
|
|
"mean_token_accuracy": 0.08583549410104752,
|
|
"num_tokens": 2813415.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.329908657073974,
|
|
"epoch": 0.13911926801862057,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999883836140236,
|
|
"loss": 7.3533,
|
|
"mean_token_accuracy": 0.08882890194654465,
|
|
"num_tokens": 2825288.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.312420797348023,
|
|
"epoch": 0.1396543421263845,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999879918612224,
|
|
"loss": 7.3068,
|
|
"mean_token_accuracy": 0.09304987639188766,
|
|
"num_tokens": 2836298.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.313601350784301,
|
|
"epoch": 0.14018941623414843,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999875936119245,
|
|
"loss": 7.2898,
|
|
"mean_token_accuracy": 0.0945392332971096,
|
|
"num_tokens": 2846367.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.240101766586304,
|
|
"epoch": 0.14072449034191237,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999871888661415,
|
|
"loss": 7.3047,
|
|
"mean_token_accuracy": 0.08923600688576698,
|
|
"num_tokens": 2856832.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.309039974212647,
|
|
"epoch": 0.14125956444967627,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000499986777623885,
|
|
"loss": 7.3245,
|
|
"mean_token_accuracy": 0.08692366257309914,
|
|
"num_tokens": 2867954.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.439533376693726,
|
|
"epoch": 0.1417946385574402,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999863598851669,
|
|
"loss": 7.3611,
|
|
"mean_token_accuracy": 0.09297569468617439,
|
|
"num_tokens": 2878714.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.053972816467285,
|
|
"epoch": 0.14232971266520414,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004999859356499992,
|
|
"loss": 7.298,
|
|
"mean_token_accuracy": 0.10052636489272118,
|
|
"num_tokens": 2890697.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.382122421264649,
|
|
"epoch": 0.14286478677296804,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999855049183943,
|
|
"loss": 7.3219,
|
|
"mean_token_accuracy": 0.08626203685998916,
|
|
"num_tokens": 2901468.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.25765290260315,
|
|
"epoch": 0.14339986088073198,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999850676903646,
|
|
"loss": 7.2671,
|
|
"mean_token_accuracy": 0.09408968985080719,
|
|
"num_tokens": 2913466.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.231286239624024,
|
|
"epoch": 0.1439349349884959,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999846239659226,
|
|
"loss": 7.2505,
|
|
"mean_token_accuracy": 0.09706522151827812,
|
|
"num_tokens": 2923638.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.269954109191895,
|
|
"epoch": 0.14447000909625984,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999841737450812,
|
|
"loss": 7.3357,
|
|
"mean_token_accuracy": 0.09442631006240845,
|
|
"num_tokens": 2934790.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.268723440170288,
|
|
"epoch": 0.14500508320402375,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999837170278535,
|
|
"loss": 7.2757,
|
|
"mean_token_accuracy": 0.08723422512412071,
|
|
"num_tokens": 2946320.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.3566694259643555,
|
|
"epoch": 0.14554015731178768,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999832538142526,
|
|
"loss": 7.2369,
|
|
"mean_token_accuracy": 0.09091865047812461,
|
|
"num_tokens": 2957077.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.177463865280151,
|
|
"epoch": 0.14607523141955162,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999827841042917,
|
|
"loss": 7.2678,
|
|
"mean_token_accuracy": 0.08816000148653984,
|
|
"num_tokens": 2967286.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.2939177513122555,
|
|
"epoch": 0.14661030552731552,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004999823078979846,
|
|
"loss": 7.3057,
|
|
"mean_token_accuracy": 0.09016561508178711,
|
|
"num_tokens": 2979516.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.379639339447022,
|
|
"epoch": 0.14714537963507945,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999818251953449,
|
|
"loss": 7.419,
|
|
"mean_token_accuracy": 0.08294185698032379,
|
|
"num_tokens": 2991594.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.246149206161499,
|
|
"epoch": 0.1476804537428434,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999813359963867,
|
|
"loss": 7.2972,
|
|
"mean_token_accuracy": 0.08775367885828018,
|
|
"num_tokens": 3002487.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.372300243377685,
|
|
"epoch": 0.14821552785060732,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999808403011241,
|
|
"loss": 7.3157,
|
|
"mean_token_accuracy": 0.09396110102534294,
|
|
"num_tokens": 3013197.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.142706060409546,
|
|
"epoch": 0.14875060195837123,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999803381095712,
|
|
"loss": 7.3451,
|
|
"mean_token_accuracy": 0.09063652530312538,
|
|
"num_tokens": 3024512.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.415092658996582,
|
|
"epoch": 0.14928567606613516,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999798294217428,
|
|
"loss": 7.3046,
|
|
"mean_token_accuracy": 0.08789012357592582,
|
|
"num_tokens": 3035621.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.229216432571411,
|
|
"epoch": 0.1498207501738991,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999793142376533,
|
|
"loss": 7.2876,
|
|
"mean_token_accuracy": 0.09253833740949631,
|
|
"num_tokens": 3046492.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.263148021697998,
|
|
"epoch": 0.150355824281663,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999787925573177,
|
|
"loss": 7.3086,
|
|
"mean_token_accuracy": 0.09204896241426468,
|
|
"num_tokens": 3057715.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.3932922840118405,
|
|
"epoch": 0.15089089838942693,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999782643807513,
|
|
"loss": 7.2654,
|
|
"mean_token_accuracy": 0.09031828343868256,
|
|
"num_tokens": 3068344.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.193799686431885,
|
|
"epoch": 0.15142597249719086,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999777297079689,
|
|
"loss": 7.2547,
|
|
"mean_token_accuracy": 0.09031877219676972,
|
|
"num_tokens": 3079576.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.342895412445069,
|
|
"epoch": 0.1519610466049548,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999771885389863,
|
|
"loss": 7.2153,
|
|
"mean_token_accuracy": 0.09486196860671044,
|
|
"num_tokens": 3090201.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.1453712463378904,
|
|
"epoch": 0.1524961207127187,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999766408738189,
|
|
"loss": 7.1936,
|
|
"mean_token_accuracy": 0.10047192424535752,
|
|
"num_tokens": 3099824.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.24136872291565,
|
|
"epoch": 0.15303119482048264,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999760867124827,
|
|
"loss": 7.2738,
|
|
"mean_token_accuracy": 0.09082015976309776,
|
|
"num_tokens": 3109948.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.299263906478882,
|
|
"epoch": 0.15356626892824657,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999755260549937,
|
|
"loss": 7.2626,
|
|
"mean_token_accuracy": 0.0974075585603714,
|
|
"num_tokens": 3121492.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.21744818687439,
|
|
"epoch": 0.15410134303601047,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0004999749589013677,
|
|
"loss": 7.0545,
|
|
"mean_token_accuracy": 0.10996845588088036,
|
|
"num_tokens": 3131990.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.1235129833221436,
|
|
"epoch": 0.1546364171437744,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999743852516217,
|
|
"loss": 7.2409,
|
|
"mean_token_accuracy": 0.0921150028705597,
|
|
"num_tokens": 3142567.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.32010293006897,
|
|
"epoch": 0.15517149125153834,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004999738051057717,
|
|
"loss": 7.2522,
|
|
"mean_token_accuracy": 0.09435679912567138,
|
|
"num_tokens": 3153083.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.188344669342041,
|
|
"epoch": 0.15570656535930227,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999732184638347,
|
|
"loss": 7.2555,
|
|
"mean_token_accuracy": 0.09004457890987397,
|
|
"num_tokens": 3163985.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.262719058990479,
|
|
"epoch": 0.15624163946706618,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999726253258278,
|
|
"loss": 7.2392,
|
|
"mean_token_accuracy": 0.09183276668190957,
|
|
"num_tokens": 3175259.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.254732751846314,
|
|
"epoch": 0.1567767135748301,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999720256917678,
|
|
"loss": 7.206,
|
|
"mean_token_accuracy": 0.0977108657360077,
|
|
"num_tokens": 3185487.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.127510118484497,
|
|
"epoch": 0.15731178768259405,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999714195616723,
|
|
"loss": 7.1831,
|
|
"mean_token_accuracy": 0.09826227650046349,
|
|
"num_tokens": 3196200.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.333015441894531,
|
|
"epoch": 0.15784686179035795,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999708069355586,
|
|
"loss": 7.3054,
|
|
"mean_token_accuracy": 0.0843054249882698,
|
|
"num_tokens": 3207675.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.17571177482605,
|
|
"epoch": 0.15838193589812188,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999701878134445,
|
|
"loss": 7.1815,
|
|
"mean_token_accuracy": 0.09501941055059433,
|
|
"num_tokens": 3218117.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.304254627227783,
|
|
"epoch": 0.15891701000588582,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999695621953477,
|
|
"loss": 7.2505,
|
|
"mean_token_accuracy": 0.10208047479391098,
|
|
"num_tokens": 3229208.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.173984098434448,
|
|
"epoch": 0.15945208411364975,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999689300812866,
|
|
"loss": 7.1584,
|
|
"mean_token_accuracy": 0.101242895424366,
|
|
"num_tokens": 3239586.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.24401364326477,
|
|
"epoch": 0.15998715822141366,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999682914712791,
|
|
"loss": 7.2254,
|
|
"mean_token_accuracy": 0.09680869728326798,
|
|
"num_tokens": 3250103.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.228427124023438,
|
|
"epoch": 0.1605222323291776,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999676463653439,
|
|
"loss": 7.2027,
|
|
"mean_token_accuracy": 0.10232653468847275,
|
|
"num_tokens": 3261844.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.227877807617188,
|
|
"epoch": 0.16105730643694152,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999669947634996,
|
|
"loss": 7.1639,
|
|
"mean_token_accuracy": 0.09760255515575408,
|
|
"num_tokens": 3272073.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.178577184677124,
|
|
"epoch": 0.16159238054470546,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999663366657647,
|
|
"loss": 7.1667,
|
|
"mean_token_accuracy": 0.09620498046278954,
|
|
"num_tokens": 3281725.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.214940547943115,
|
|
"epoch": 0.16212745465246936,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999656720721586,
|
|
"loss": 7.3059,
|
|
"mean_token_accuracy": 0.094396660476923,
|
|
"num_tokens": 3292708.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.296448183059693,
|
|
"epoch": 0.1626625287602333,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999650009827004,
|
|
"loss": 7.2773,
|
|
"mean_token_accuracy": 0.09127842709422111,
|
|
"num_tokens": 3303548.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.19203143119812,
|
|
"epoch": 0.16319760286799723,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999643233974092,
|
|
"loss": 7.1486,
|
|
"mean_token_accuracy": 0.09975013583898544,
|
|
"num_tokens": 3314261.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.1947274684906,
|
|
"epoch": 0.16373267697576113,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999636393163049,
|
|
"loss": 7.1714,
|
|
"mean_token_accuracy": 0.09445247575640678,
|
|
"num_tokens": 3324833.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.053837537765503,
|
|
"epoch": 0.16426775108352507,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999629487394071,
|
|
"loss": 7.0568,
|
|
"mean_token_accuracy": 0.10237468853592872,
|
|
"num_tokens": 3334790.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.266930866241455,
|
|
"epoch": 0.164802825191289,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999622516667358,
|
|
"loss": 7.205,
|
|
"mean_token_accuracy": 0.09691800698637962,
|
|
"num_tokens": 3345115.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.133483982086181,
|
|
"epoch": 0.16533789929905293,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499961548098311,
|
|
"loss": 7.089,
|
|
"mean_token_accuracy": 0.10441275909543038,
|
|
"num_tokens": 3356187.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.117019701004028,
|
|
"epoch": 0.16587297340681684,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999608380341533,
|
|
"loss": 7.1282,
|
|
"mean_token_accuracy": 0.09887576475739479,
|
|
"num_tokens": 3367218.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.160897636413575,
|
|
"epoch": 0.16640804751458077,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999601214742829,
|
|
"loss": 7.27,
|
|
"mean_token_accuracy": 0.08936030119657516,
|
|
"num_tokens": 3377557.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.371627759933472,
|
|
"epoch": 0.1669431216223447,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999593984187206,
|
|
"loss": 7.2482,
|
|
"mean_token_accuracy": 0.09751093834638595,
|
|
"num_tokens": 3387402.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.0916835308074955,
|
|
"epoch": 0.1674781957301086,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999586688674872,
|
|
"loss": 7.1346,
|
|
"mean_token_accuracy": 0.09648581743240356,
|
|
"num_tokens": 3397924.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.216659879684448,
|
|
"epoch": 0.16801326983787254,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000499957932820604,
|
|
"loss": 7.2118,
|
|
"mean_token_accuracy": 0.08759412840008736,
|
|
"num_tokens": 3408622.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.262946891784668,
|
|
"epoch": 0.16854834394563648,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499957190278092,
|
|
"loss": 7.1622,
|
|
"mean_token_accuracy": 0.09248490408062934,
|
|
"num_tokens": 3419692.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.098281383514404,
|
|
"epoch": 0.1690834180534004,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999564412399728,
|
|
"loss": 7.2154,
|
|
"mean_token_accuracy": 0.09459864124655723,
|
|
"num_tokens": 3430986.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.176830434799195,
|
|
"epoch": 0.1696184921611643,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499955685706268,
|
|
"loss": 7.1354,
|
|
"mean_token_accuracy": 0.09504674524068832,
|
|
"num_tokens": 3442211.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.03007402420044,
|
|
"epoch": 0.17015356626892825,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999549236769993,
|
|
"loss": 7.1229,
|
|
"mean_token_accuracy": 0.10297105386853218,
|
|
"num_tokens": 3453585.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.225273466110229,
|
|
"epoch": 0.17068864037669218,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999541551521888,
|
|
"loss": 7.1695,
|
|
"mean_token_accuracy": 0.0996770367026329,
|
|
"num_tokens": 3464156.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.249066162109375,
|
|
"epoch": 0.17122371448445609,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999533801318587,
|
|
"loss": 7.2238,
|
|
"mean_token_accuracy": 0.08952004536986351,
|
|
"num_tokens": 3474853.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.262774181365967,
|
|
"epoch": 0.17175878859222002,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999525986160313,
|
|
"loss": 7.194,
|
|
"mean_token_accuracy": 0.10073793306946754,
|
|
"num_tokens": 3486841.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.140778827667236,
|
|
"epoch": 0.17229386269998395,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999518106047293,
|
|
"loss": 7.2169,
|
|
"mean_token_accuracy": 0.09210691601037979,
|
|
"num_tokens": 3498735.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.175544404983521,
|
|
"epoch": 0.17282893680774788,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999510160979754,
|
|
"loss": 7.2352,
|
|
"mean_token_accuracy": 0.10104434639215469,
|
|
"num_tokens": 3508829.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.1884317874908445,
|
|
"epoch": 0.1733640109155118,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999502150957925,
|
|
"loss": 7.141,
|
|
"mean_token_accuracy": 0.09438429847359657,
|
|
"num_tokens": 3519465.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.188577127456665,
|
|
"epoch": 0.17389908502327572,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999494075982037,
|
|
"loss": 7.1086,
|
|
"mean_token_accuracy": 0.10255491808056831,
|
|
"num_tokens": 3530401.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.12376561164856,
|
|
"epoch": 0.17443415913103966,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999485936052324,
|
|
"loss": 7.1856,
|
|
"mean_token_accuracy": 0.09849482700228691,
|
|
"num_tokens": 3541136.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.288824605941772,
|
|
"epoch": 0.17496923323880356,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499947773116902,
|
|
"loss": 7.1753,
|
|
"mean_token_accuracy": 0.09568259268999099,
|
|
"num_tokens": 3552525.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.090163278579712,
|
|
"epoch": 0.1755043073465675,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999469461332365,
|
|
"loss": 7.1414,
|
|
"mean_token_accuracy": 0.09975111782550812,
|
|
"num_tokens": 3563028.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.199271249771118,
|
|
"epoch": 0.17603938145433143,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999461126542592,
|
|
"loss": 7.0912,
|
|
"mean_token_accuracy": 0.10047454982995987,
|
|
"num_tokens": 3573354.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.979637813568115,
|
|
"epoch": 0.17657445556209536,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999452726799947,
|
|
"loss": 7.0802,
|
|
"mean_token_accuracy": 0.09618928134441376,
|
|
"num_tokens": 3585576.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.297661590576172,
|
|
"epoch": 0.17710952966985927,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999444262104671,
|
|
"loss": 7.2058,
|
|
"mean_token_accuracy": 0.08774002119898797,
|
|
"num_tokens": 3596478.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.110372161865234,
|
|
"epoch": 0.1776446037776232,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999435732457007,
|
|
"loss": 7.153,
|
|
"mean_token_accuracy": 0.09257297441363335,
|
|
"num_tokens": 3608014.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.221296691894532,
|
|
"epoch": 0.17817967788538713,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999427137857203,
|
|
"loss": 7.2102,
|
|
"mean_token_accuracy": 0.09210594072937965,
|
|
"num_tokens": 3620120.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.107110977172852,
|
|
"epoch": 0.17871475199315104,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999418478305506,
|
|
"loss": 7.1096,
|
|
"mean_token_accuracy": 0.09546202942728996,
|
|
"num_tokens": 3632578.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.156751394271851,
|
|
"epoch": 0.17924982610091497,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999409753802167,
|
|
"loss": 7.131,
|
|
"mean_token_accuracy": 0.10362366437911988,
|
|
"num_tokens": 3643245.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.167398452758789,
|
|
"epoch": 0.1797849002086789,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999400964347437,
|
|
"loss": 7.1251,
|
|
"mean_token_accuracy": 0.09799086153507233,
|
|
"num_tokens": 3655832.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.151060581207275,
|
|
"epoch": 0.18031997431644284,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999392109941571,
|
|
"loss": 7.1001,
|
|
"mean_token_accuracy": 0.09830698594450951,
|
|
"num_tokens": 3666566.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.172510385513306,
|
|
"epoch": 0.18085504842420674,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999383190584822,
|
|
"loss": 7.1298,
|
|
"mean_token_accuracy": 0.10040193051099777,
|
|
"num_tokens": 3677572.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.112444877624512,
|
|
"epoch": 0.18139012253197068,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999374206277451,
|
|
"loss": 7.0549,
|
|
"mean_token_accuracy": 0.09736391827464104,
|
|
"num_tokens": 3687116.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.057082557678223,
|
|
"epoch": 0.1819251966397346,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999365157019717,
|
|
"loss": 7.1658,
|
|
"mean_token_accuracy": 0.09722192957997322,
|
|
"num_tokens": 3697961.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 7.233665180206299,
|
|
"epoch": 0.18246027074749852,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999356042811878,
|
|
"loss": 7.0802,
|
|
"mean_token_accuracy": 0.10210576206445694,
|
|
"num_tokens": 3708834.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.087706089019775,
|
|
"epoch": 0.18299534485526245,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00049993468636542,
|
|
"loss": 7.1472,
|
|
"mean_token_accuracy": 0.09556594043970108,
|
|
"num_tokens": 3719835.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.116140556335449,
|
|
"epoch": 0.18353041896302638,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999337619546948,
|
|
"loss": 7.0339,
|
|
"mean_token_accuracy": 0.10466360598802567,
|
|
"num_tokens": 3730408.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.085261631011963,
|
|
"epoch": 0.18406549307079031,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999328310490387,
|
|
"loss": 7.1275,
|
|
"mean_token_accuracy": 0.09871943295001984,
|
|
"num_tokens": 3741257.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.190493202209472,
|
|
"epoch": 0.18460056717855422,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999318936484789,
|
|
"loss": 7.1175,
|
|
"mean_token_accuracy": 0.10177163109183311,
|
|
"num_tokens": 3752615.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.106344223022461,
|
|
"epoch": 0.18513564128631815,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999309497530422,
|
|
"loss": 7.1395,
|
|
"mean_token_accuracy": 0.09094236344099045,
|
|
"num_tokens": 3763656.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.083023738861084,
|
|
"epoch": 0.1856707153940821,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999299993627558,
|
|
"loss": 7.1148,
|
|
"mean_token_accuracy": 0.10153383314609528,
|
|
"num_tokens": 3775257.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.0514672756195065,
|
|
"epoch": 0.186205789501846,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999290424776475,
|
|
"loss": 7.0758,
|
|
"mean_token_accuracy": 0.1056319996714592,
|
|
"num_tokens": 3786094.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.330498886108399,
|
|
"epoch": 0.18674086360960993,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999280790977445,
|
|
"loss": 7.1657,
|
|
"mean_token_accuracy": 0.09170655235648155,
|
|
"num_tokens": 3796672.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.131537580490113,
|
|
"epoch": 0.18727593771737386,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499927109223075,
|
|
"loss": 7.1508,
|
|
"mean_token_accuracy": 0.09670756980776787,
|
|
"num_tokens": 3807910.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.103532934188843,
|
|
"epoch": 0.1878110118251378,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999261328536667,
|
|
"loss": 7.1028,
|
|
"mean_token_accuracy": 0.1024494618177414,
|
|
"num_tokens": 3820792.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.02842001914978,
|
|
"epoch": 0.1883460859329017,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999251499895479,
|
|
"loss": 7.0234,
|
|
"mean_token_accuracy": 0.1013932503759861,
|
|
"num_tokens": 3831126.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.213843870162964,
|
|
"epoch": 0.18888116004066563,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999241606307471,
|
|
"loss": 7.1353,
|
|
"mean_token_accuracy": 0.10057510435581207,
|
|
"num_tokens": 3841923.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.119020175933838,
|
|
"epoch": 0.18941623414842956,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999231647772927,
|
|
"loss": 7.0491,
|
|
"mean_token_accuracy": 0.10542590543627739,
|
|
"num_tokens": 3852504.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.992453765869141,
|
|
"epoch": 0.1899513082561935,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999221624292136,
|
|
"loss": 7.0559,
|
|
"mean_token_accuracy": 0.10485710874199868,
|
|
"num_tokens": 3862761.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.134696340560913,
|
|
"epoch": 0.1904863823639574,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999211535865385,
|
|
"loss": 7.0556,
|
|
"mean_token_accuracy": 0.09678238332271576,
|
|
"num_tokens": 3873662.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.063473129272461,
|
|
"epoch": 0.19102145647172133,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999201382492969,
|
|
"loss": 7.0088,
|
|
"mean_token_accuracy": 0.10047155022621154,
|
|
"num_tokens": 3884835.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 7.110291051864624,
|
|
"epoch": 0.19155653057948527,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999191164175178,
|
|
"loss": 7.154,
|
|
"mean_token_accuracy": 0.09878607764840126,
|
|
"num_tokens": 3895073.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.209331798553467,
|
|
"epoch": 0.19209160468724917,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999180880912309,
|
|
"loss": 7.1115,
|
|
"mean_token_accuracy": 0.09884128645062447,
|
|
"num_tokens": 3905440.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 7.072672271728516,
|
|
"epoch": 0.1926266787950131,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999170532704657,
|
|
"loss": 7.1145,
|
|
"mean_token_accuracy": 0.0968818336725235,
|
|
"num_tokens": 3917135.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.13597002029419,
|
|
"epoch": 0.19316175290277704,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999160119552523,
|
|
"loss": 7.117,
|
|
"mean_token_accuracy": 0.10278327167034149,
|
|
"num_tokens": 3928247.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 7.091355848312378,
|
|
"epoch": 0.19369682701054097,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999149641456206,
|
|
"loss": 7.1009,
|
|
"mean_token_accuracy": 0.09872358813881874,
|
|
"num_tokens": 3938609.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.21585783958435,
|
|
"epoch": 0.19423190111830488,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499913909841601,
|
|
"loss": 7.0861,
|
|
"mean_token_accuracy": 0.10127425193786621,
|
|
"num_tokens": 3949221.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 7.009944200515747,
|
|
"epoch": 0.1947669752260688,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999128490432238,
|
|
"loss": 7.0484,
|
|
"mean_token_accuracy": 0.09548230320215226,
|
|
"num_tokens": 3959475.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.083078098297119,
|
|
"epoch": 0.19530204933383274,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999117817505196,
|
|
"loss": 7.0515,
|
|
"mean_token_accuracy": 0.1084510900080204,
|
|
"num_tokens": 3969638.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.053667259216309,
|
|
"epoch": 0.19583712344159665,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999107079635194,
|
|
"loss": 7.1079,
|
|
"mean_token_accuracy": 0.10137292668223381,
|
|
"num_tokens": 3980543.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.355273485183716,
|
|
"epoch": 0.19637219754936058,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000499909627682254,
|
|
"loss": 7.162,
|
|
"mean_token_accuracy": 0.08682658635079861,
|
|
"num_tokens": 3993518.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.068727111816406,
|
|
"epoch": 0.19690727165712452,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999085409067549,
|
|
"loss": 7.1487,
|
|
"mean_token_accuracy": 0.09789205938577653,
|
|
"num_tokens": 4002940.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 7.013206481933594,
|
|
"epoch": 0.19744234576488845,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999074476370531,
|
|
"loss": 7.0792,
|
|
"mean_token_accuracy": 0.09683184549212456,
|
|
"num_tokens": 4013982.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.100980806350708,
|
|
"epoch": 0.19797741987265235,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999063478731805,
|
|
"loss": 7.0638,
|
|
"mean_token_accuracy": 0.10558338239789009,
|
|
"num_tokens": 4025607.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.0951310157775875,
|
|
"epoch": 0.1985124939804163,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999052416151687,
|
|
"loss": 7.1104,
|
|
"mean_token_accuracy": 0.10141080170869828,
|
|
"num_tokens": 4037467.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 7.149590158462525,
|
|
"epoch": 0.19904756808818022,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999041288630496,
|
|
"loss": 7.1127,
|
|
"mean_token_accuracy": 0.10126180425286294,
|
|
"num_tokens": 4049146.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.181366395950318,
|
|
"epoch": 0.19958264219594413,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999030096168553,
|
|
"loss": 7.1485,
|
|
"mean_token_accuracy": 0.09781184569001197,
|
|
"num_tokens": 4060500.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 7.157667875289917,
|
|
"epoch": 0.20011771630370806,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999018838766182,
|
|
"loss": 7.0987,
|
|
"mean_token_accuracy": 0.0985781691968441,
|
|
"num_tokens": 4070553.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 7.14162769317627,
|
|
"epoch": 0.200652790411472,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499900751642371,
|
|
"loss": 7.0343,
|
|
"mean_token_accuracy": 0.11338024884462357,
|
|
"num_tokens": 4079653.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.015338134765625,
|
|
"epoch": 0.20118786451923593,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998996129141461,
|
|
"loss": 7.123,
|
|
"mean_token_accuracy": 0.09923376441001892,
|
|
"num_tokens": 4089912.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.212585592269898,
|
|
"epoch": 0.20172293862699983,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998984676919764,
|
|
"loss": 7.0772,
|
|
"mean_token_accuracy": 0.09714810699224471,
|
|
"num_tokens": 4100577.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 7.073903751373291,
|
|
"epoch": 0.20225801273476376,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998973159758952,
|
|
"loss": 7.1033,
|
|
"mean_token_accuracy": 0.09901956990361213,
|
|
"num_tokens": 4111752.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 7.06131501197815,
|
|
"epoch": 0.2027930868425277,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998961577659355,
|
|
"loss": 7.0616,
|
|
"mean_token_accuracy": 0.10182834565639495,
|
|
"num_tokens": 4122616.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 7.192514705657959,
|
|
"epoch": 0.2033281609502916,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998949930621309,
|
|
"loss": 7.1547,
|
|
"mean_token_accuracy": 0.10196215733885765,
|
|
"num_tokens": 4134275.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 7.112344741821289,
|
|
"epoch": 0.20386323505805554,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998938218645149,
|
|
"loss": 7.0506,
|
|
"mean_token_accuracy": 0.10170819535851479,
|
|
"num_tokens": 4146412.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 7.029837560653687,
|
|
"epoch": 0.20439830916581947,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004998926441731215,
|
|
"loss": 7.0512,
|
|
"mean_token_accuracy": 0.1002343863248825,
|
|
"num_tokens": 4157592.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 7.06212477684021,
|
|
"epoch": 0.2049333832735834,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998914599879845,
|
|
"loss": 7.0269,
|
|
"mean_token_accuracy": 0.09835589304566383,
|
|
"num_tokens": 4168637.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 7.0664630889892575,
|
|
"epoch": 0.2054684573813473,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998902693091382,
|
|
"loss": 7.0736,
|
|
"mean_token_accuracy": 0.09925776198506356,
|
|
"num_tokens": 4179951.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 7.0451087474823,
|
|
"epoch": 0.20600353148911124,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998890721366171,
|
|
"loss": 6.9616,
|
|
"mean_token_accuracy": 0.10666931569576263,
|
|
"num_tokens": 4189895.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 7.097507953643799,
|
|
"epoch": 0.20653860559687517,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998878684704556,
|
|
"loss": 7.0214,
|
|
"mean_token_accuracy": 0.10524726510047913,
|
|
"num_tokens": 4199796.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 7.036600160598755,
|
|
"epoch": 0.20707367970463908,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998866583106884,
|
|
"loss": 7.0461,
|
|
"mean_token_accuracy": 0.09883956611156464,
|
|
"num_tokens": 4210428.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 7.218980216979981,
|
|
"epoch": 0.207608753812403,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998854416573506,
|
|
"loss": 7.1246,
|
|
"mean_token_accuracy": 0.10078881680965424,
|
|
"num_tokens": 4221515.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 7.1100975513458256,
|
|
"epoch": 0.20814382792016695,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998842185104772,
|
|
"loss": 7.1387,
|
|
"mean_token_accuracy": 0.09548087790608406,
|
|
"num_tokens": 4232580.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 7.041958284378052,
|
|
"epoch": 0.20867890202793088,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998829888701036,
|
|
"loss": 7.0484,
|
|
"mean_token_accuracy": 0.10664152652025223,
|
|
"num_tokens": 4242871.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 7.147633504867554,
|
|
"epoch": 0.20921397613569478,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998817527362654,
|
|
"loss": 7.0773,
|
|
"mean_token_accuracy": 0.09752313494682312,
|
|
"num_tokens": 4253419.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 7.032255792617798,
|
|
"epoch": 0.20974905024345872,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998805101089982,
|
|
"loss": 7.0528,
|
|
"mean_token_accuracy": 0.10345240533351899,
|
|
"num_tokens": 4264327.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 7.170184564590454,
|
|
"epoch": 0.21028412435122265,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004998792609883378,
|
|
"loss": 7.0428,
|
|
"mean_token_accuracy": 0.10343464836478233,
|
|
"num_tokens": 4274568.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 7.056762933731079,
|
|
"epoch": 0.21081919845898656,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998780053743203,
|
|
"loss": 7.0324,
|
|
"mean_token_accuracy": 0.10682544857263565,
|
|
"num_tokens": 4284895.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 7.12225079536438,
|
|
"epoch": 0.2113542725667505,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998767432669822,
|
|
"loss": 7.0637,
|
|
"mean_token_accuracy": 0.0994573712348938,
|
|
"num_tokens": 4297258.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 7.026572132110596,
|
|
"epoch": 0.21188934667451442,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998754746663595,
|
|
"loss": 6.9842,
|
|
"mean_token_accuracy": 0.10585973560810089,
|
|
"num_tokens": 4308307.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 7.050811910629273,
|
|
"epoch": 0.21242442078227836,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998741995724892,
|
|
"loss": 7.0491,
|
|
"mean_token_accuracy": 0.10263351649045944,
|
|
"num_tokens": 4318801.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 7.070459413528442,
|
|
"epoch": 0.21295949489004226,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998729179854079,
|
|
"loss": 7.0252,
|
|
"mean_token_accuracy": 0.10016985535621643,
|
|
"num_tokens": 4329531.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 7.054882478713989,
|
|
"epoch": 0.2134945689978062,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004998716299051527,
|
|
"loss": 7.0253,
|
|
"mean_token_accuracy": 0.10470812693238259,
|
|
"num_tokens": 4341203.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 7.026881456375122,
|
|
"epoch": 0.21402964310557013,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998703353317609,
|
|
"loss": 7.0788,
|
|
"mean_token_accuracy": 0.0973646655678749,
|
|
"num_tokens": 4352426.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 7.11639552116394,
|
|
"epoch": 0.21456471721333403,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998690342652697,
|
|
"loss": 7.0103,
|
|
"mean_token_accuracy": 0.10563170909881592,
|
|
"num_tokens": 4363413.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 7.054438829421997,
|
|
"epoch": 0.21509979132109797,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004998677267057166,
|
|
"loss": 6.951,
|
|
"mean_token_accuracy": 0.10644056648015976,
|
|
"num_tokens": 4373470.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.9533806324005125,
|
|
"epoch": 0.2156348654288619,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998664126531397,
|
|
"loss": 7.0164,
|
|
"mean_token_accuracy": 0.10371329113841057,
|
|
"num_tokens": 4384438.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 7.1470112800598145,
|
|
"epoch": 0.21616993953662583,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998650921075766,
|
|
"loss": 6.9978,
|
|
"mean_token_accuracy": 0.10406329333782197,
|
|
"num_tokens": 4395100.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.9813416481018065,
|
|
"epoch": 0.21670501364438974,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998637650690656,
|
|
"loss": 7.1179,
|
|
"mean_token_accuracy": 0.09519804865121842,
|
|
"num_tokens": 4407131.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 7.10913496017456,
|
|
"epoch": 0.21724008775215367,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499862431537645,
|
|
"loss": 6.9569,
|
|
"mean_token_accuracy": 0.10057392492890357,
|
|
"num_tokens": 4417701.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.981591272354126,
|
|
"epoch": 0.2177751618599176,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998610915133533,
|
|
"loss": 7.0173,
|
|
"mean_token_accuracy": 0.10014636293053628,
|
|
"num_tokens": 4429309.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 7.007551097869873,
|
|
"epoch": 0.21831023596768154,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998597449962292,
|
|
"loss": 7.0663,
|
|
"mean_token_accuracy": 0.10122283399105073,
|
|
"num_tokens": 4440850.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 7.092380857467651,
|
|
"epoch": 0.21884531007544544,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998583919863115,
|
|
"loss": 7.0501,
|
|
"mean_token_accuracy": 0.09975898936390877,
|
|
"num_tokens": 4451995.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.99484076499939,
|
|
"epoch": 0.21938038418320938,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998570324836393,
|
|
"loss": 6.959,
|
|
"mean_token_accuracy": 0.10816834419965744,
|
|
"num_tokens": 4461916.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 7.084143257141113,
|
|
"epoch": 0.2199154582909733,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499855666488252,
|
|
"loss": 7.0571,
|
|
"mean_token_accuracy": 0.10147327557206154,
|
|
"num_tokens": 4471603.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 7.105906391143799,
|
|
"epoch": 0.22045053239873721,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998542940001889,
|
|
"loss": 7.083,
|
|
"mean_token_accuracy": 0.10265670791268348,
|
|
"num_tokens": 4483166.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 7.020576333999633,
|
|
"epoch": 0.22098560650650115,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998529150194895,
|
|
"loss": 7.0022,
|
|
"mean_token_accuracy": 0.09936799854040146,
|
|
"num_tokens": 4493824.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 7.097088718414307,
|
|
"epoch": 0.22152068061426508,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998515295461939,
|
|
"loss": 6.9225,
|
|
"mean_token_accuracy": 0.1188635177910328,
|
|
"num_tokens": 4505674.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.968977689743042,
|
|
"epoch": 0.22205575472202901,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000499850137580342,
|
|
"loss": 7.0049,
|
|
"mean_token_accuracy": 0.10718596279621125,
|
|
"num_tokens": 4516516.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 7.039025449752808,
|
|
"epoch": 0.22259082882979292,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998487391219739,
|
|
"loss": 6.9963,
|
|
"mean_token_accuracy": 0.09949744194746017,
|
|
"num_tokens": 4527811.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 7.0311308860778805,
|
|
"epoch": 0.22312590293755685,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998473341711301,
|
|
"loss": 6.9388,
|
|
"mean_token_accuracy": 0.09995686337351799,
|
|
"num_tokens": 4538988.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 7.018882703781128,
|
|
"epoch": 0.22366097704532079,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998459227278512,
|
|
"loss": 6.9733,
|
|
"mean_token_accuracy": 0.1059156596660614,
|
|
"num_tokens": 4550049.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 7.072498369216919,
|
|
"epoch": 0.2241960511530847,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998445047921777,
|
|
"loss": 7.0719,
|
|
"mean_token_accuracy": 0.10480737760663032,
|
|
"num_tokens": 4559784.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 7.031924390792847,
|
|
"epoch": 0.22473112526084862,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998430803641509,
|
|
"loss": 6.9337,
|
|
"mean_token_accuracy": 0.10361162722110748,
|
|
"num_tokens": 4569982.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.939174699783325,
|
|
"epoch": 0.22526619936861256,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998416494438118,
|
|
"loss": 6.9534,
|
|
"mean_token_accuracy": 0.10496753230690956,
|
|
"num_tokens": 4581117.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 7.033869123458862,
|
|
"epoch": 0.2258012734763765,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998402120312014,
|
|
"loss": 7.0198,
|
|
"mean_token_accuracy": 0.10706626474857331,
|
|
"num_tokens": 4592012.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 7.034424209594727,
|
|
"epoch": 0.2263363475841404,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998387681263616,
|
|
"loss": 7.0581,
|
|
"mean_token_accuracy": 0.10076023638248444,
|
|
"num_tokens": 4602591.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 7.038776493072509,
|
|
"epoch": 0.22687142169190433,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499837317729334,
|
|
"loss": 6.985,
|
|
"mean_token_accuracy": 0.10343138575553894,
|
|
"num_tokens": 4613418.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.944168281555176,
|
|
"epoch": 0.22740649579966826,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998358608401603,
|
|
"loss": 6.9051,
|
|
"mean_token_accuracy": 0.11338085606694222,
|
|
"num_tokens": 4625156.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 7.1117607116699215,
|
|
"epoch": 0.22794156990743217,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000499834397458883,
|
|
"loss": 6.937,
|
|
"mean_token_accuracy": 0.10821644887328148,
|
|
"num_tokens": 4634523.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.958322238922119,
|
|
"epoch": 0.2284766440151961,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998329275855438,
|
|
"loss": 6.929,
|
|
"mean_token_accuracy": 0.11250298544764518,
|
|
"num_tokens": 4646292.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.985101842880249,
|
|
"epoch": 0.22901171812296003,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998314512201855,
|
|
"loss": 6.9099,
|
|
"mean_token_accuracy": 0.10238043665885925,
|
|
"num_tokens": 4656417.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.924033832550049,
|
|
"epoch": 0.22954679223072397,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998299683628507,
|
|
"loss": 6.9282,
|
|
"mean_token_accuracy": 0.10534189045429229,
|
|
"num_tokens": 4666933.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 7.059316158294678,
|
|
"epoch": 0.23008186633848787,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998284790135821,
|
|
"loss": 7.1084,
|
|
"mean_token_accuracy": 0.09782184064388275,
|
|
"num_tokens": 4678482.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 7.053033971786499,
|
|
"epoch": 0.2306169404462518,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998269831724228,
|
|
"loss": 6.9481,
|
|
"mean_token_accuracy": 0.1064787782728672,
|
|
"num_tokens": 4688454.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.9519795894622805,
|
|
"epoch": 0.23115201455401574,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998254808394159,
|
|
"loss": 6.9839,
|
|
"mean_token_accuracy": 0.1061179980635643,
|
|
"num_tokens": 4699545.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 7.083680486679077,
|
|
"epoch": 0.23168708866177964,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004998239720146048,
|
|
"loss": 6.9986,
|
|
"mean_token_accuracy": 0.10272667407989503,
|
|
"num_tokens": 4711149.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.961511278152466,
|
|
"epoch": 0.23222216276954358,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004998224566980332,
|
|
"loss": 6.9935,
|
|
"mean_token_accuracy": 0.10200226753950119,
|
|
"num_tokens": 4722651.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 7.031105852127075,
|
|
"epoch": 0.2327572368773075,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998209348897447,
|
|
"loss": 6.9926,
|
|
"mean_token_accuracy": 0.09458313658833503,
|
|
"num_tokens": 4733883.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 7.135159015655518,
|
|
"epoch": 0.23329231098507144,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998194065897833,
|
|
"loss": 6.955,
|
|
"mean_token_accuracy": 0.10691621899604797,
|
|
"num_tokens": 4744517.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.9776819229125975,
|
|
"epoch": 0.23382738509283535,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499817871798193,
|
|
"loss": 6.9804,
|
|
"mean_token_accuracy": 0.11214952319860458,
|
|
"num_tokens": 4755481.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.972070980072021,
|
|
"epoch": 0.23436245920059928,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998163305150185,
|
|
"loss": 6.9039,
|
|
"mean_token_accuracy": 0.11068079173564911,
|
|
"num_tokens": 4765535.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.973287200927734,
|
|
"epoch": 0.23489753330836322,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998147827403038,
|
|
"loss": 7.0273,
|
|
"mean_token_accuracy": 0.10471541583538055,
|
|
"num_tokens": 4777045.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 7.077732086181641,
|
|
"epoch": 0.23543260741612712,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499813228474094,
|
|
"loss": 6.9064,
|
|
"mean_token_accuracy": 0.10825864002108573,
|
|
"num_tokens": 4787958.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.917084503173828,
|
|
"epoch": 0.23596768152389105,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998116677164338,
|
|
"loss": 6.9445,
|
|
"mean_token_accuracy": 0.10730226412415504,
|
|
"num_tokens": 4798502.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.931448936462402,
|
|
"epoch": 0.236502755631655,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998101004673682,
|
|
"loss": 6.9413,
|
|
"mean_token_accuracy": 0.11088823229074478,
|
|
"num_tokens": 4809179.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 7.135161733627319,
|
|
"epoch": 0.23703782973941892,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998085267269426,
|
|
"loss": 7.0215,
|
|
"mean_token_accuracy": 0.09467339664697647,
|
|
"num_tokens": 4819803.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 7.004361200332641,
|
|
"epoch": 0.23757290384718283,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998069464952025,
|
|
"loss": 6.9916,
|
|
"mean_token_accuracy": 0.10000614672899247,
|
|
"num_tokens": 4831226.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 7.049906778335571,
|
|
"epoch": 0.23810797795494676,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004998053597721932,
|
|
"loss": 7.0124,
|
|
"mean_token_accuracy": 0.10538152158260346,
|
|
"num_tokens": 4842168.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 7.019608783721924,
|
|
"epoch": 0.2386430520627107,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998037665579608,
|
|
"loss": 6.9621,
|
|
"mean_token_accuracy": 0.10858180820941925,
|
|
"num_tokens": 4852655.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.915230321884155,
|
|
"epoch": 0.2391781261704746,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998021668525513,
|
|
"loss": 6.8626,
|
|
"mean_token_accuracy": 0.10984659045934678,
|
|
"num_tokens": 4862164.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 7.046167945861816,
|
|
"epoch": 0.23971320027823853,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998005606560107,
|
|
"loss": 6.9964,
|
|
"mean_token_accuracy": 0.10990610420703888,
|
|
"num_tokens": 4874233.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.992388391494751,
|
|
"epoch": 0.24024827438600246,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004997989479683856,
|
|
"loss": 6.9876,
|
|
"mean_token_accuracy": 0.1066631942987442,
|
|
"num_tokens": 4885948.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.963246202468872,
|
|
"epoch": 0.2407833484937664,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004997973287897224,
|
|
"loss": 6.9667,
|
|
"mean_token_accuracy": 0.10817388147115707,
|
|
"num_tokens": 4897128.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.980781888961792,
|
|
"epoch": 0.2413184226015303,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997957031200681,
|
|
"loss": 6.939,
|
|
"mean_token_accuracy": 0.11121672838926315,
|
|
"num_tokens": 4907699.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.887354183197021,
|
|
"epoch": 0.24185349670929424,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997940709594692,
|
|
"loss": 6.8796,
|
|
"mean_token_accuracy": 0.10674800127744674,
|
|
"num_tokens": 4918772.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 7.091795778274536,
|
|
"epoch": 0.24238857081705817,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997924323079733,
|
|
"loss": 6.9368,
|
|
"mean_token_accuracy": 0.10729291215538979,
|
|
"num_tokens": 4930046.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.9101183891296385,
|
|
"epoch": 0.24292364492482207,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997907871656273,
|
|
"loss": 6.9568,
|
|
"mean_token_accuracy": 0.10801581889390946,
|
|
"num_tokens": 4940609.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 7.002806997299194,
|
|
"epoch": 0.243458719032586,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997891355324791,
|
|
"loss": 6.9658,
|
|
"mean_token_accuracy": 0.10598139539361,
|
|
"num_tokens": 4951339.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.983994722366333,
|
|
"epoch": 0.24399379314034994,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004997874774085761,
|
|
"loss": 6.878,
|
|
"mean_token_accuracy": 0.10262224972248077,
|
|
"num_tokens": 4961083.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.895718050003052,
|
|
"epoch": 0.24452886724811387,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997858127939662,
|
|
"loss": 6.8908,
|
|
"mean_token_accuracy": 0.10761211439967155,
|
|
"num_tokens": 4971399.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 7.041612005233764,
|
|
"epoch": 0.24506394135587778,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997841416886976,
|
|
"loss": 6.9715,
|
|
"mean_token_accuracy": 0.1061974436044693,
|
|
"num_tokens": 4982429.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 7.0298323154449465,
|
|
"epoch": 0.2455990154636417,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997824640928186,
|
|
"loss": 7.0184,
|
|
"mean_token_accuracy": 0.10754595324397087,
|
|
"num_tokens": 4993706.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.988226366043091,
|
|
"epoch": 0.24613408957140565,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004997807800063774,
|
|
"loss": 6.9127,
|
|
"mean_token_accuracy": 0.10858980864286423,
|
|
"num_tokens": 5005411.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 7.023472309112549,
|
|
"epoch": 0.24666916367916958,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000499779089429423,
|
|
"loss": 7.0413,
|
|
"mean_token_accuracy": 0.10208934471011162,
|
|
"num_tokens": 5016011.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 7.013689231872559,
|
|
"epoch": 0.24720423778693348,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997773923620037,
|
|
"loss": 6.9954,
|
|
"mean_token_accuracy": 0.10329191759228706,
|
|
"num_tokens": 5027841.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 7.100406980514526,
|
|
"epoch": 0.24773931189469742,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997756888041689,
|
|
"loss": 6.9853,
|
|
"mean_token_accuracy": 0.09798330888152122,
|
|
"num_tokens": 5038640.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.95166802406311,
|
|
"epoch": 0.24827438600246135,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997739787559677,
|
|
"loss": 6.9356,
|
|
"mean_token_accuracy": 0.10357877835631371,
|
|
"num_tokens": 5050423.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.978884506225586,
|
|
"epoch": 0.24880946011022526,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997722622174494,
|
|
"loss": 6.954,
|
|
"mean_token_accuracy": 0.10356095060706139,
|
|
"num_tokens": 5060654.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.934212970733642,
|
|
"epoch": 0.2493445342179892,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997705391886636,
|
|
"loss": 6.8582,
|
|
"mean_token_accuracy": 0.11322998628020287,
|
|
"num_tokens": 5070165.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.913174200057983,
|
|
"epoch": 0.24987960832575312,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997688096696601,
|
|
"loss": 6.8995,
|
|
"mean_token_accuracy": 0.10103722065687179,
|
|
"num_tokens": 5080937.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.911728572845459,
|
|
"epoch": 0.25041468243351706,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997670736604888,
|
|
"loss": 6.8744,
|
|
"mean_token_accuracy": 0.11288927420973778,
|
|
"num_tokens": 5090955.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 7.074804210662842,
|
|
"epoch": 0.250949756541281,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997653311611998,
|
|
"loss": 7.0112,
|
|
"mean_token_accuracy": 0.09920887500047684,
|
|
"num_tokens": 5102201.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 7.0164875984191895,
|
|
"epoch": 0.25148483064904487,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997635821718435,
|
|
"loss": 6.9312,
|
|
"mean_token_accuracy": 0.11026841551065444,
|
|
"num_tokens": 5112235.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.8925614833831785,
|
|
"epoch": 0.2520199047568088,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004997618266924704,
|
|
"loss": 6.8219,
|
|
"mean_token_accuracy": 0.10299434512853622,
|
|
"num_tokens": 5122360.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 7.002241230010986,
|
|
"epoch": 0.25255497886457273,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499760064723131,
|
|
"loss": 6.9801,
|
|
"mean_token_accuracy": 0.1024145670235157,
|
|
"num_tokens": 5133084.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.938703870773315,
|
|
"epoch": 0.25309005297233667,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997582962638764,
|
|
"loss": 6.9133,
|
|
"mean_token_accuracy": 0.09837969020009041,
|
|
"num_tokens": 5144495.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 7.031873512268066,
|
|
"epoch": 0.2536251270801006,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997565213147575,
|
|
"loss": 6.9238,
|
|
"mean_token_accuracy": 0.10646962001919746,
|
|
"num_tokens": 5154824.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.939056253433227,
|
|
"epoch": 0.25416020118786453,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997547398758257,
|
|
"loss": 6.9111,
|
|
"mean_token_accuracy": 0.10193499475717545,
|
|
"num_tokens": 5165061.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.981441783905029,
|
|
"epoch": 0.25469527529562846,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997529519471323,
|
|
"loss": 6.8679,
|
|
"mean_token_accuracy": 0.11417916044592857,
|
|
"num_tokens": 5174964.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.884857940673828,
|
|
"epoch": 0.25523034940339234,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997511575287291,
|
|
"loss": 6.9267,
|
|
"mean_token_accuracy": 0.10812475010752678,
|
|
"num_tokens": 5185458.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.991512250900269,
|
|
"epoch": 0.2557654235111563,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997493566206677,
|
|
"loss": 6.9025,
|
|
"mean_token_accuracy": 0.10954718366265297,
|
|
"num_tokens": 5196193.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.966927862167358,
|
|
"epoch": 0.2563004976189202,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004997475492230003,
|
|
"loss": 7.0198,
|
|
"mean_token_accuracy": 0.09881971776485443,
|
|
"num_tokens": 5208541.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.989241456985473,
|
|
"epoch": 0.25683557172668414,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499745735335779,
|
|
"loss": 6.8501,
|
|
"mean_token_accuracy": 0.10620964914560319,
|
|
"num_tokens": 5218567.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.943647241592407,
|
|
"epoch": 0.2573706458344481,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004997439149590561,
|
|
"loss": 6.9666,
|
|
"mean_token_accuracy": 0.10247603580355644,
|
|
"num_tokens": 5228293.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.982013940811157,
|
|
"epoch": 0.257905719942212,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997420880928843,
|
|
"loss": 6.913,
|
|
"mean_token_accuracy": 0.10574470162391662,
|
|
"num_tokens": 5238880.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.9576988697052,
|
|
"epoch": 0.25844079404997594,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997402547373162,
|
|
"loss": 6.9947,
|
|
"mean_token_accuracy": 0.10212295204401016,
|
|
"num_tokens": 5250104.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 7.100748538970947,
|
|
"epoch": 0.2589758681577399,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004997384148924049,
|
|
"loss": 7.0243,
|
|
"mean_token_accuracy": 0.10289878994226456,
|
|
"num_tokens": 5260355.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.962227201461792,
|
|
"epoch": 0.25951094226550375,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997365685582035,
|
|
"loss": 6.9296,
|
|
"mean_token_accuracy": 0.10283196046948433,
|
|
"num_tokens": 5272117.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.9883177280426025,
|
|
"epoch": 0.2600460163732677,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004997347157347651,
|
|
"loss": 7.0307,
|
|
"mean_token_accuracy": 0.10037739872932434,
|
|
"num_tokens": 5284583.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 7.049625587463379,
|
|
"epoch": 0.2605810904810316,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997328564221435,
|
|
"loss": 6.8241,
|
|
"mean_token_accuracy": 0.12059795558452606,
|
|
"num_tokens": 5295846.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.871410655975342,
|
|
"epoch": 0.26111616458879555,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004997309906203922,
|
|
"loss": 6.9662,
|
|
"mean_token_accuracy": 0.10526571422815323,
|
|
"num_tokens": 5307802.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 7.033421468734741,
|
|
"epoch": 0.2616512386965595,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997291183295652,
|
|
"loss": 6.8913,
|
|
"mean_token_accuracy": 0.10737156346440316,
|
|
"num_tokens": 5318263.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.998748159408569,
|
|
"epoch": 0.2621863128043234,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997272395497164,
|
|
"loss": 6.9388,
|
|
"mean_token_accuracy": 0.10455406904220581,
|
|
"num_tokens": 5328555.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.912611675262451,
|
|
"epoch": 0.26272138691208735,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997253542809002,
|
|
"loss": 6.9152,
|
|
"mean_token_accuracy": 0.10682450905442238,
|
|
"num_tokens": 5339954.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 7.079062366485596,
|
|
"epoch": 0.26325646101985123,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997234625231709,
|
|
"loss": 6.9637,
|
|
"mean_token_accuracy": 0.10808332860469819,
|
|
"num_tokens": 5350919.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.93997278213501,
|
|
"epoch": 0.26379153512761516,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997215642765832,
|
|
"loss": 6.925,
|
|
"mean_token_accuracy": 0.10664721727371215,
|
|
"num_tokens": 5361844.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.987752532958984,
|
|
"epoch": 0.2643266092353791,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499719659541192,
|
|
"loss": 6.9562,
|
|
"mean_token_accuracy": 0.10334968268871307,
|
|
"num_tokens": 5372026.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.954167985916138,
|
|
"epoch": 0.26486168334314303,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004997177483170521,
|
|
"loss": 6.9927,
|
|
"mean_token_accuracy": 0.10936494544148445,
|
|
"num_tokens": 5382965.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 7.007135391235352,
|
|
"epoch": 0.26539675745090696,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997158306042188,
|
|
"loss": 6.9458,
|
|
"mean_token_accuracy": 0.10711643844842911,
|
|
"num_tokens": 5393738.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.897781848907471,
|
|
"epoch": 0.2659318315586709,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997139064027476,
|
|
"loss": 6.8237,
|
|
"mean_token_accuracy": 0.1119477018713951,
|
|
"num_tokens": 5404059.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.912676382064819,
|
|
"epoch": 0.26646690566643483,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997119757126938,
|
|
"loss": 6.855,
|
|
"mean_token_accuracy": 0.10871710032224655,
|
|
"num_tokens": 5414152.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 7.022440433502197,
|
|
"epoch": 0.2670019797741987,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997100385341133,
|
|
"loss": 7.0235,
|
|
"mean_token_accuracy": 0.10726772770285606,
|
|
"num_tokens": 5425227.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 7.004488468170166,
|
|
"epoch": 0.26753705388196264,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997080948670621,
|
|
"loss": 6.8663,
|
|
"mean_token_accuracy": 0.11122208759188652,
|
|
"num_tokens": 5435716.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.936031723022461,
|
|
"epoch": 0.26807212798972657,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997061447115962,
|
|
"loss": 6.9347,
|
|
"mean_token_accuracy": 0.10375397354364395,
|
|
"num_tokens": 5446519.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.975105237960816,
|
|
"epoch": 0.2686072020974905,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997041880677719,
|
|
"loss": 6.925,
|
|
"mean_token_accuracy": 0.10548185482621193,
|
|
"num_tokens": 5457496.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.880755996704101,
|
|
"epoch": 0.26914227620525444,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997022249356457,
|
|
"loss": 6.823,
|
|
"mean_token_accuracy": 0.11008255481719971,
|
|
"num_tokens": 5469059.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.923517560958862,
|
|
"epoch": 0.26967735031301837,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004997002553152745,
|
|
"loss": 6.9138,
|
|
"mean_token_accuracy": 0.10938023030757904,
|
|
"num_tokens": 5480324.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.9742035388946535,
|
|
"epoch": 0.2702124244207823,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004996982792067148,
|
|
"loss": 6.9339,
|
|
"mean_token_accuracy": 0.09864719212055206,
|
|
"num_tokens": 5491513.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.8970160484313965,
|
|
"epoch": 0.2707474985285462,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499696296610024,
|
|
"loss": 6.8394,
|
|
"mean_token_accuracy": 0.1051232136785984,
|
|
"num_tokens": 5502414.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.9252660274505615,
|
|
"epoch": 0.2712825726363101,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004996943075252592,
|
|
"loss": 6.9715,
|
|
"mean_token_accuracy": 0.1062911570072174,
|
|
"num_tokens": 5512962.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.977799654006958,
|
|
"epoch": 0.27181764674407405,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996923119524777,
|
|
"loss": 6.8355,
|
|
"mean_token_accuracy": 0.10766077339649201,
|
|
"num_tokens": 5524648.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.855292940139771,
|
|
"epoch": 0.272352720851838,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996903098917375,
|
|
"loss": 6.9214,
|
|
"mean_token_accuracy": 0.0998130515217781,
|
|
"num_tokens": 5536158.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.939780187606812,
|
|
"epoch": 0.2728877949596019,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499688301343096,
|
|
"loss": 6.8405,
|
|
"mean_token_accuracy": 0.10747457072138786,
|
|
"num_tokens": 5547374.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 7.020946025848389,
|
|
"epoch": 0.27342286906736585,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996862863066114,
|
|
"loss": 6.9137,
|
|
"mean_token_accuracy": 0.10850636437535285,
|
|
"num_tokens": 5559212.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.907329320907593,
|
|
"epoch": 0.2739579431751298,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996842647823419,
|
|
"loss": 6.8823,
|
|
"mean_token_accuracy": 0.10653341263532638,
|
|
"num_tokens": 5569302.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.910869598388672,
|
|
"epoch": 0.27449301728289366,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996822367703458,
|
|
"loss": 6.8905,
|
|
"mean_token_accuracy": 0.10348111093044281,
|
|
"num_tokens": 5581688.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.955097198486328,
|
|
"epoch": 0.2750280913906576,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996802022706817,
|
|
"loss": 6.8211,
|
|
"mean_token_accuracy": 0.10833865478634834,
|
|
"num_tokens": 5591376.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.861729669570923,
|
|
"epoch": 0.2755631654984215,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996781612834083,
|
|
"loss": 6.8798,
|
|
"mean_token_accuracy": 0.10732169449329376,
|
|
"num_tokens": 5601510.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 7.005447006225586,
|
|
"epoch": 0.27609823960618546,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996761138085846,
|
|
"loss": 6.9316,
|
|
"mean_token_accuracy": 0.1021797627210617,
|
|
"num_tokens": 5611436.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.92574257850647,
|
|
"epoch": 0.2766333137139494,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004996740598462697,
|
|
"loss": 6.9341,
|
|
"mean_token_accuracy": 0.10267650112509727,
|
|
"num_tokens": 5620470.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.860880899429321,
|
|
"epoch": 0.2771683878217133,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996719993965229,
|
|
"loss": 6.8297,
|
|
"mean_token_accuracy": 0.10769467577338218,
|
|
"num_tokens": 5630740.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.8753687858581545,
|
|
"epoch": 0.27770346192947726,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996699324594036,
|
|
"loss": 6.8303,
|
|
"mean_token_accuracy": 0.11734023317694664,
|
|
"num_tokens": 5641138.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.985844898223877,
|
|
"epoch": 0.27823853603724114,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004996678590349717,
|
|
"loss": 6.974,
|
|
"mean_token_accuracy": 0.1085183672606945,
|
|
"num_tokens": 5652474.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.9662446022033695,
|
|
"epoch": 0.27877361014500507,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996657791232868,
|
|
"loss": 6.8709,
|
|
"mean_token_accuracy": 0.11153203919529915,
|
|
"num_tokens": 5663178.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.891348218917846,
|
|
"epoch": 0.279308684252769,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996636927244092,
|
|
"loss": 6.8963,
|
|
"mean_token_accuracy": 0.10582477301359176,
|
|
"num_tokens": 5673565.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 7.040777587890625,
|
|
"epoch": 0.27984375836053293,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499661599838399,
|
|
"loss": 6.9389,
|
|
"mean_token_accuracy": 0.10204790085554123,
|
|
"num_tokens": 5684651.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.929867219924927,
|
|
"epoch": 0.28037883246829687,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996595004653167,
|
|
"loss": 6.8204,
|
|
"mean_token_accuracy": 0.10971119105815888,
|
|
"num_tokens": 5696081.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.925084161758423,
|
|
"epoch": 0.2809139065760608,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004996573946052229,
|
|
"loss": 6.855,
|
|
"mean_token_accuracy": 0.10883698239922523,
|
|
"num_tokens": 5706830.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.910267782211304,
|
|
"epoch": 0.28144898068382473,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996552822581783,
|
|
"loss": 6.8815,
|
|
"mean_token_accuracy": 0.11361035332083702,
|
|
"num_tokens": 5717261.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.91580810546875,
|
|
"epoch": 0.2819840547915886,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000499653163424244,
|
|
"loss": 6.8481,
|
|
"mean_token_accuracy": 0.10841184929013252,
|
|
"num_tokens": 5727280.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.938942956924438,
|
|
"epoch": 0.28251912889935255,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004996510381034814,
|
|
"loss": 6.9083,
|
|
"mean_token_accuracy": 0.10522805228829384,
|
|
"num_tokens": 5737485.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.926429605484008,
|
|
"epoch": 0.2830542030071165,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004996489062959515,
|
|
"loss": 6.8735,
|
|
"mean_token_accuracy": 0.10467999801039696,
|
|
"num_tokens": 5747683.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.867719650268555,
|
|
"epoch": 0.2835892771148804,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996467680017159,
|
|
"loss": 6.9325,
|
|
"mean_token_accuracy": 0.10280923023819924,
|
|
"num_tokens": 5758148.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.917259216308594,
|
|
"epoch": 0.28412435122264434,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004996446232208365,
|
|
"loss": 6.8627,
|
|
"mean_token_accuracy": 0.11170612499117852,
|
|
"num_tokens": 5769109.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.9372947216033936,
|
|
"epoch": 0.2846594253304083,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996424719533753,
|
|
"loss": 6.8939,
|
|
"mean_token_accuracy": 0.1052110217511654,
|
|
"num_tokens": 5780344.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.913125228881836,
|
|
"epoch": 0.2851944994381722,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996403141993941,
|
|
"loss": 6.6962,
|
|
"mean_token_accuracy": 0.11824454516172409,
|
|
"num_tokens": 5790167.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.92453293800354,
|
|
"epoch": 0.2857295735459361,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996381499589555,
|
|
"loss": 6.8624,
|
|
"mean_token_accuracy": 0.10613423809409142,
|
|
"num_tokens": 5800076.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.944131517410279,
|
|
"epoch": 0.2862646476537,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004996359792321219,
|
|
"loss": 6.8113,
|
|
"mean_token_accuracy": 0.11141969487071038,
|
|
"num_tokens": 5810547.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.8161341667175295,
|
|
"epoch": 0.28679972176146395,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499633802018956,
|
|
"loss": 6.8478,
|
|
"mean_token_accuracy": 0.1100471280515194,
|
|
"num_tokens": 5821210.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.917617702484131,
|
|
"epoch": 0.2873347958692279,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996316183195204,
|
|
"loss": 6.8869,
|
|
"mean_token_accuracy": 0.10786092206835747,
|
|
"num_tokens": 5832361.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.962878274917602,
|
|
"epoch": 0.2878698699769918,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996294281338786,
|
|
"loss": 6.8717,
|
|
"mean_token_accuracy": 0.10681624412536621,
|
|
"num_tokens": 5843242.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.976676988601684,
|
|
"epoch": 0.28840494408475575,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004996272314620936,
|
|
"loss": 6.9215,
|
|
"mean_token_accuracy": 0.10194345191121101,
|
|
"num_tokens": 5855163.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.837163925170898,
|
|
"epoch": 0.2889400181925197,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004996250283042288,
|
|
"loss": 6.8839,
|
|
"mean_token_accuracy": 0.11228486746549607,
|
|
"num_tokens": 5866908.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.974035596847534,
|
|
"epoch": 0.28947509230028357,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499622818660348,
|
|
"loss": 6.9045,
|
|
"mean_token_accuracy": 0.10939288139343262,
|
|
"num_tokens": 5877342.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.942460250854492,
|
|
"epoch": 0.2900101664080475,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996206025305147,
|
|
"loss": 6.8573,
|
|
"mean_token_accuracy": 0.10582418739795685,
|
|
"num_tokens": 5887960.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.87635645866394,
|
|
"epoch": 0.29054524051581143,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996183799147932,
|
|
"loss": 6.8397,
|
|
"mean_token_accuracy": 0.1095063179731369,
|
|
"num_tokens": 5899887.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.917806816101074,
|
|
"epoch": 0.29108031462357536,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996161508132475,
|
|
"loss": 6.8583,
|
|
"mean_token_accuracy": 0.10231703668832778,
|
|
"num_tokens": 5911022.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 7.016377210617065,
|
|
"epoch": 0.2916153887313393,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499613915225942,
|
|
"loss": 6.9222,
|
|
"mean_token_accuracy": 0.11102863773703575,
|
|
"num_tokens": 5922780.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.881858253479004,
|
|
"epoch": 0.29215046283910323,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996116731529412,
|
|
"loss": 6.9251,
|
|
"mean_token_accuracy": 0.10592872574925423,
|
|
"num_tokens": 5934112.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.8940752983093265,
|
|
"epoch": 0.29268553694686716,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00049960942459431,
|
|
"loss": 6.8866,
|
|
"mean_token_accuracy": 0.1067856416106224,
|
|
"num_tokens": 5944386.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.870763492584229,
|
|
"epoch": 0.29322061105463104,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996071695501132,
|
|
"loss": 6.813,
|
|
"mean_token_accuracy": 0.10948670953512192,
|
|
"num_tokens": 5955552.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.921037864685059,
|
|
"epoch": 0.293755685162395,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996049080204159,
|
|
"loss": 6.8137,
|
|
"mean_token_accuracy": 0.1098986655473709,
|
|
"num_tokens": 5966790.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.925626182556153,
|
|
"epoch": 0.2942907592701589,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996026400052835,
|
|
"loss": 6.8214,
|
|
"mean_token_accuracy": 0.10917853116989136,
|
|
"num_tokens": 5977704.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.853207015991211,
|
|
"epoch": 0.29482583337792284,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004996003655047814,
|
|
"loss": 6.8364,
|
|
"mean_token_accuracy": 0.11162171512842178,
|
|
"num_tokens": 5989221.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.919997644424439,
|
|
"epoch": 0.2953609074856868,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004995980845189753,
|
|
"loss": 6.871,
|
|
"mean_token_accuracy": 0.10931915193796157,
|
|
"num_tokens": 6000906.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.931192445755005,
|
|
"epoch": 0.2958959815934507,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995957970479312,
|
|
"loss": 6.8945,
|
|
"mean_token_accuracy": 0.10708941668272018,
|
|
"num_tokens": 6012039.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.8547321319580075,
|
|
"epoch": 0.29643105570121464,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995935030917148,
|
|
"loss": 6.776,
|
|
"mean_token_accuracy": 0.11066783592104912,
|
|
"num_tokens": 6024383.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.994161558151245,
|
|
"epoch": 0.2969661298089785,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995912026503926,
|
|
"loss": 6.8508,
|
|
"mean_token_accuracy": 0.10267031714320182,
|
|
"num_tokens": 6034904.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.838885879516601,
|
|
"epoch": 0.29750120391674245,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499588895724031,
|
|
"loss": 6.8488,
|
|
"mean_token_accuracy": 0.10749915912747383,
|
|
"num_tokens": 6044449.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.962449741363526,
|
|
"epoch": 0.2980362780245064,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004995865823126968,
|
|
"loss": 6.8765,
|
|
"mean_token_accuracy": 0.10825972631573677,
|
|
"num_tokens": 6054465.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.8947694301605225,
|
|
"epoch": 0.2985713521322703,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995842624164564,
|
|
"loss": 6.8962,
|
|
"mean_token_accuracy": 0.10824007838964463,
|
|
"num_tokens": 6065539.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.909287405014038,
|
|
"epoch": 0.29910642624003425,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499581936035377,
|
|
"loss": 6.8459,
|
|
"mean_token_accuracy": 0.10840294510126114,
|
|
"num_tokens": 6076135.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.903677034378052,
|
|
"epoch": 0.2996415003477982,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995796031695257,
|
|
"loss": 6.8641,
|
|
"mean_token_accuracy": 0.10931005999445915,
|
|
"num_tokens": 6086604.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.90281114578247,
|
|
"epoch": 0.3001765744555621,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00049957726381897,
|
|
"loss": 6.8781,
|
|
"mean_token_accuracy": 0.10950748696923256,
|
|
"num_tokens": 6095672.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.90290675163269,
|
|
"epoch": 0.300711648563326,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995749179837773,
|
|
"loss": 6.7498,
|
|
"mean_token_accuracy": 0.11466853022575378,
|
|
"num_tokens": 6105652.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.826155042648315,
|
|
"epoch": 0.30124672267108993,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004995725656640155,
|
|
"loss": 6.7856,
|
|
"mean_token_accuracy": 0.11099454537034034,
|
|
"num_tokens": 6116930.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.844819593429565,
|
|
"epoch": 0.30178179677885386,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004995702068597524,
|
|
"loss": 6.7894,
|
|
"mean_token_accuracy": 0.11423054486513137,
|
|
"num_tokens": 6126659.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.890478515625,
|
|
"epoch": 0.3023168708866178,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995678415710561,
|
|
"loss": 6.8288,
|
|
"mean_token_accuracy": 0.11060848459601402,
|
|
"num_tokens": 6137227.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.962517929077149,
|
|
"epoch": 0.3028519449943817,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499565469797995,
|
|
"loss": 6.8821,
|
|
"mean_token_accuracy": 0.10931277051568031,
|
|
"num_tokens": 6147800.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.822924327850342,
|
|
"epoch": 0.30338701910214566,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995630915406374,
|
|
"loss": 6.7326,
|
|
"mean_token_accuracy": 0.12161283493041992,
|
|
"num_tokens": 6157671.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.918770694732666,
|
|
"epoch": 0.3039220932099096,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995607067990521,
|
|
"loss": 6.8698,
|
|
"mean_token_accuracy": 0.10645673274993897,
|
|
"num_tokens": 6168651.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.873465251922608,
|
|
"epoch": 0.30445716731767347,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000499558315573308,
|
|
"loss": 6.8487,
|
|
"mean_token_accuracy": 0.1086043007671833,
|
|
"num_tokens": 6179925.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.818239831924439,
|
|
"epoch": 0.3049922414254374,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004995559178634741,
|
|
"loss": 6.8438,
|
|
"mean_token_accuracy": 0.10887812077999115,
|
|
"num_tokens": 6190541.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 7.041949510574341,
|
|
"epoch": 0.30552731553320134,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995535136696196,
|
|
"loss": 6.8866,
|
|
"mean_token_accuracy": 0.09955592602491378,
|
|
"num_tokens": 6201491.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.8139080047607425,
|
|
"epoch": 0.30606238964096527,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499551102991814,
|
|
"loss": 6.6991,
|
|
"mean_token_accuracy": 0.11249880269169807,
|
|
"num_tokens": 6212443.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.817849111557007,
|
|
"epoch": 0.3065974637487292,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995486858301268,
|
|
"loss": 6.8981,
|
|
"mean_token_accuracy": 0.1030767761170864,
|
|
"num_tokens": 6222623.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.925403165817261,
|
|
"epoch": 0.30713253785649314,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004995462621846279,
|
|
"loss": 6.7852,
|
|
"mean_token_accuracy": 0.10933116525411606,
|
|
"num_tokens": 6233712.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.854152202606201,
|
|
"epoch": 0.30766761196425707,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995438320553871,
|
|
"loss": 6.8534,
|
|
"mean_token_accuracy": 0.10412434935569763,
|
|
"num_tokens": 6244254.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.823256492614746,
|
|
"epoch": 0.30820268607202095,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004995413954424748,
|
|
"loss": 6.7504,
|
|
"mean_token_accuracy": 0.116096231341362,
|
|
"num_tokens": 6254202.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.877709770202637,
|
|
"epoch": 0.3087377601797849,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995389523459612,
|
|
"loss": 6.8231,
|
|
"mean_token_accuracy": 0.10538254305720329,
|
|
"num_tokens": 6264496.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.839356851577759,
|
|
"epoch": 0.3092728342875488,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499536502765917,
|
|
"loss": 6.828,
|
|
"mean_token_accuracy": 0.1125398851931095,
|
|
"num_tokens": 6275835.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.893553924560547,
|
|
"epoch": 0.30980790839531275,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004995340467024128,
|
|
"loss": 6.7842,
|
|
"mean_token_accuracy": 0.10834744572639465,
|
|
"num_tokens": 6286805.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.906386661529541,
|
|
"epoch": 0.3103429825030767,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995315841555195,
|
|
"loss": 6.7916,
|
|
"mean_token_accuracy": 0.10366136953234673,
|
|
"num_tokens": 6298444.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.783759641647339,
|
|
"epoch": 0.3108780566108406,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995291151253083,
|
|
"loss": 6.7348,
|
|
"mean_token_accuracy": 0.11380124613642692,
|
|
"num_tokens": 6308259.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.846135902404785,
|
|
"epoch": 0.31141313071860455,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995266396118505,
|
|
"loss": 6.8013,
|
|
"mean_token_accuracy": 0.10161566883325576,
|
|
"num_tokens": 6320058.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.900366973876953,
|
|
"epoch": 0.3119482048263684,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004995241576152174,
|
|
"loss": 6.7723,
|
|
"mean_token_accuracy": 0.10676236972212791,
|
|
"num_tokens": 6331203.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.897499227523804,
|
|
"epoch": 0.31248327893413236,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499521669135481,
|
|
"loss": 6.8882,
|
|
"mean_token_accuracy": 0.10588050186634064,
|
|
"num_tokens": 6342027.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.894981956481933,
|
|
"epoch": 0.3130183530418963,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995191741727128,
|
|
"loss": 6.7873,
|
|
"mean_token_accuracy": 0.10809844583272935,
|
|
"num_tokens": 6352580.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.887385463714599,
|
|
"epoch": 0.3135534271496602,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000499516672726985,
|
|
"loss": 6.8106,
|
|
"mean_token_accuracy": 0.11217835023999215,
|
|
"num_tokens": 6363997.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.83791708946228,
|
|
"epoch": 0.31408850125742416,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004995141647983699,
|
|
"loss": 6.8464,
|
|
"mean_token_accuracy": 0.11019290536642075,
|
|
"num_tokens": 6375284.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.8710222244262695,
|
|
"epoch": 0.3146235753651881,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995116503869398,
|
|
"loss": 6.7437,
|
|
"mean_token_accuracy": 0.11758178621530532,
|
|
"num_tokens": 6386698.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.864429473876953,
|
|
"epoch": 0.315158649472952,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004995091294927673,
|
|
"loss": 6.818,
|
|
"mean_token_accuracy": 0.10886727124452591,
|
|
"num_tokens": 6397237.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.973788785934448,
|
|
"epoch": 0.3156937235807159,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995066021159253,
|
|
"loss": 6.9364,
|
|
"mean_token_accuracy": 0.10483678579330444,
|
|
"num_tokens": 6408266.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.842577266693115,
|
|
"epoch": 0.31622879768847983,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995040682564867,
|
|
"loss": 6.8199,
|
|
"mean_token_accuracy": 0.11405593305826187,
|
|
"num_tokens": 6419563.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.8359404563903805,
|
|
"epoch": 0.31676387179624377,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995015279145247,
|
|
"loss": 6.7863,
|
|
"mean_token_accuracy": 0.11180907115340233,
|
|
"num_tokens": 6430220.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.857833385467529,
|
|
"epoch": 0.3172989459040077,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004994989810901126,
|
|
"loss": 6.7961,
|
|
"mean_token_accuracy": 0.10685641467571258,
|
|
"num_tokens": 6440157.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.856301593780517,
|
|
"epoch": 0.31783402001177163,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004994964277833239,
|
|
"loss": 6.8463,
|
|
"mean_token_accuracy": 0.11110184714198112,
|
|
"num_tokens": 6449511.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.923713493347168,
|
|
"epoch": 0.31836909411953557,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004994938679942324,
|
|
"loss": 6.8365,
|
|
"mean_token_accuracy": 0.10282818749547004,
|
|
"num_tokens": 6460608.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.919619989395142,
|
|
"epoch": 0.3189041682272995,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004994913017229121,
|
|
"loss": 6.8372,
|
|
"mean_token_accuracy": 0.1131968080997467,
|
|
"num_tokens": 6472737.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.817192554473877,
|
|
"epoch": 0.31943924233506343,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004994887289694371,
|
|
"loss": 6.7184,
|
|
"mean_token_accuracy": 0.11385754495859146,
|
|
"num_tokens": 6482993.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.960534572601318,
|
|
"epoch": 0.3199743164428273,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994861497338814,
|
|
"loss": 6.9207,
|
|
"mean_token_accuracy": 0.11038238108158112,
|
|
"num_tokens": 6494530.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.894915056228638,
|
|
"epoch": 0.32050939055059124,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004994835640163198,
|
|
"loss": 6.7626,
|
|
"mean_token_accuracy": 0.10633266866207122,
|
|
"num_tokens": 6504691.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.885722494125366,
|
|
"epoch": 0.3210444646583552,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004994809718168267,
|
|
"loss": 6.7865,
|
|
"mean_token_accuracy": 0.10957603380084038,
|
|
"num_tokens": 6514924.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.3210444646583552,
|
|
"eval_entropy": 6.6982880289021205,
|
|
"eval_loss": 6.9056596755981445,
|
|
"eval_mean_token_accuracy": 0.11213191206946045,
|
|
"eval_num_tokens": 6514924.0,
|
|
"eval_runtime": 22.6381,
|
|
"eval_samples_per_second": 1311.022,
|
|
"eval_steps_per_second": 163.883,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 93440,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 3000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 9543356688384000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|