e1fa97edea
Model: fpadovani/heb-hebr-10mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407 Source: Original Platform
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": 4.759765005111694,
|
|
"epoch": 0.0005350741077639254,
|
|
"grad_norm": 24.375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.199,
|
|
"mean_token_accuracy": 0.004748707497492432,
|
|
"num_tokens": 10038.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.763779735565185,
|
|
"epoch": 0.0010701482155278507,
|
|
"grad_norm": 25.125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.2064,
|
|
"mean_token_accuracy": 0.0057549867779016495,
|
|
"num_tokens": 21916.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.7975164413452145,
|
|
"epoch": 0.0016052223232917758,
|
|
"grad_norm": 29.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.1506,
|
|
"mean_token_accuracy": 0.0034938069467898457,
|
|
"num_tokens": 33434.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 4.968814086914063,
|
|
"epoch": 0.0021402964310557014,
|
|
"grad_norm": 38.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.7109,
|
|
"mean_token_accuracy": 0.003357043443247676,
|
|
"num_tokens": 44216.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.636525678634643,
|
|
"epoch": 0.0026753705388196263,
|
|
"grad_norm": 51.0,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 12.7453,
|
|
"mean_token_accuracy": 0.0032943477854132652,
|
|
"num_tokens": 54874.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 8.125221157073975,
|
|
"epoch": 0.0032104446465835517,
|
|
"grad_norm": 17.875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 11.4214,
|
|
"mean_token_accuracy": 0.0011823114822618664,
|
|
"num_tokens": 65085.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.552507495880127,
|
|
"epoch": 0.003745518754347477,
|
|
"grad_norm": 4.125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.7363,
|
|
"mean_token_accuracy": 0.0020443544257432223,
|
|
"num_tokens": 75601.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.720990562438965,
|
|
"epoch": 0.004280592862111403,
|
|
"grad_norm": 3.328125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.6406,
|
|
"mean_token_accuracy": 0.013470453396439552,
|
|
"num_tokens": 86193.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.704570293426514,
|
|
"epoch": 0.004815666969875327,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.4959,
|
|
"mean_token_accuracy": 0.015227390732616187,
|
|
"num_tokens": 96957.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.638626861572266,
|
|
"epoch": 0.005350741077639253,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.3582,
|
|
"mean_token_accuracy": 0.01454832088202238,
|
|
"num_tokens": 108408.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.683897018432617,
|
|
"epoch": 0.005885815185403178,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.2245,
|
|
"mean_token_accuracy": 0.015560841839760541,
|
|
"num_tokens": 119273.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.71973533630371,
|
|
"epoch": 0.006420889293167103,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.1245,
|
|
"mean_token_accuracy": 0.03457072209566832,
|
|
"num_tokens": 130817.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.70310287475586,
|
|
"epoch": 0.006955963400931029,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.0512,
|
|
"mean_token_accuracy": 0.04912868961691856,
|
|
"num_tokens": 142599.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.679985427856446,
|
|
"epoch": 0.007491037508694954,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.9832,
|
|
"mean_token_accuracy": 0.04829649589955807,
|
|
"num_tokens": 153617.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.683323860168457,
|
|
"epoch": 0.00802611161645888,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.9422,
|
|
"mean_token_accuracy": 0.04904289618134498,
|
|
"num_tokens": 165280.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.676106262207032,
|
|
"epoch": 0.008561185724222806,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.8806,
|
|
"mean_token_accuracy": 0.04988635517656803,
|
|
"num_tokens": 176896.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.664983177185059,
|
|
"epoch": 0.00909625983198673,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.8275,
|
|
"mean_token_accuracy": 0.05339982435107231,
|
|
"num_tokens": 188000.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.663267135620117,
|
|
"epoch": 0.009631333939750655,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.7684,
|
|
"mean_token_accuracy": 0.051240837574005126,
|
|
"num_tokens": 199360.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.648833179473877,
|
|
"epoch": 0.01016640804751458,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.716,
|
|
"mean_token_accuracy": 0.04947503618896008,
|
|
"num_tokens": 210214.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.61328296661377,
|
|
"epoch": 0.010701482155278505,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.5866,
|
|
"mean_token_accuracy": 0.05695506259799003,
|
|
"num_tokens": 220567.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.584807205200196,
|
|
"epoch": 0.011236556263042432,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.4837,
|
|
"mean_token_accuracy": 0.05784616619348526,
|
|
"num_tokens": 230585.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.56409158706665,
|
|
"epoch": 0.011771630370806356,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.4355,
|
|
"mean_token_accuracy": 0.06093316040933132,
|
|
"num_tokens": 241996.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.524872207641602,
|
|
"epoch": 0.012306704478570282,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.3966,
|
|
"mean_token_accuracy": 0.0568605650216341,
|
|
"num_tokens": 253348.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.46424674987793,
|
|
"epoch": 0.012841778586334207,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.2988,
|
|
"mean_token_accuracy": 0.05472910478711128,
|
|
"num_tokens": 264239.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.440286445617676,
|
|
"epoch": 0.013376852694098133,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.1996,
|
|
"mean_token_accuracy": 0.05646762736141682,
|
|
"num_tokens": 274522.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.376360607147216,
|
|
"epoch": 0.013911926801862057,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.0874,
|
|
"mean_token_accuracy": 0.056457117572426795,
|
|
"num_tokens": 284492.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.292502307891846,
|
|
"epoch": 0.014447000909625984,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 9.0249,
|
|
"mean_token_accuracy": 0.057459867745637896,
|
|
"num_tokens": 294796.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.18876838684082,
|
|
"epoch": 0.014982075017389908,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.9839,
|
|
"mean_token_accuracy": 0.05458516478538513,
|
|
"num_tokens": 305252.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.138203525543213,
|
|
"epoch": 0.015517149125153834,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.863,
|
|
"mean_token_accuracy": 0.05204447396099567,
|
|
"num_tokens": 315505.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.065136623382568,
|
|
"epoch": 0.01605222323291776,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.8148,
|
|
"mean_token_accuracy": 0.05895447283983231,
|
|
"num_tokens": 326997.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.919741535186768,
|
|
"epoch": 0.016587297340681685,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.7167,
|
|
"mean_token_accuracy": 0.05453297197818756,
|
|
"num_tokens": 338089.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.827782726287841,
|
|
"epoch": 0.01712237144844561,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.6645,
|
|
"mean_token_accuracy": 0.054577385634183885,
|
|
"num_tokens": 348536.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.709816837310791,
|
|
"epoch": 0.017657445556209534,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.6434,
|
|
"mean_token_accuracy": 0.05497662127017975,
|
|
"num_tokens": 359776.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.536125087738037,
|
|
"epoch": 0.01819251966397346,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.5416,
|
|
"mean_token_accuracy": 0.05734114237129688,
|
|
"num_tokens": 370810.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.37319803237915,
|
|
"epoch": 0.018727593771737387,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.512,
|
|
"mean_token_accuracy": 0.05475095845758915,
|
|
"num_tokens": 381788.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.286770725250244,
|
|
"epoch": 0.01926266787950131,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.4773,
|
|
"mean_token_accuracy": 0.05117404013872147,
|
|
"num_tokens": 391984.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.142317199707032,
|
|
"epoch": 0.019797741987265235,
|
|
"grad_norm": 0.609375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.462,
|
|
"mean_token_accuracy": 0.05436502285301685,
|
|
"num_tokens": 403430.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 9.018431186676025,
|
|
"epoch": 0.02033281609502916,
|
|
"grad_norm": 0.6484375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.4383,
|
|
"mean_token_accuracy": 0.052350400015711784,
|
|
"num_tokens": 414467.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.938930225372314,
|
|
"epoch": 0.020867890202793088,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.4015,
|
|
"mean_token_accuracy": 0.05855946056544781,
|
|
"num_tokens": 425752.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.85777645111084,
|
|
"epoch": 0.02140296431055701,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.3693,
|
|
"mean_token_accuracy": 0.05885430946946144,
|
|
"num_tokens": 435908.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.804098224639892,
|
|
"epoch": 0.021938038418320937,
|
|
"grad_norm": 0.66796875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.4059,
|
|
"mean_token_accuracy": 0.0562417134642601,
|
|
"num_tokens": 446424.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.765947341918945,
|
|
"epoch": 0.022473112526084863,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.3715,
|
|
"mean_token_accuracy": 0.06059782654047012,
|
|
"num_tokens": 456854.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.768339824676513,
|
|
"epoch": 0.02300818663384879,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.3964,
|
|
"mean_token_accuracy": 0.06077548712491989,
|
|
"num_tokens": 467899.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.720343017578125,
|
|
"epoch": 0.023543260741612712,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.3382,
|
|
"mean_token_accuracy": 0.057309944182634354,
|
|
"num_tokens": 478683.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.685015773773193,
|
|
"epoch": 0.02407833484937664,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.3726,
|
|
"mean_token_accuracy": 0.0574649728834629,
|
|
"num_tokens": 488745.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.69738130569458,
|
|
"epoch": 0.024613408957140565,
|
|
"grad_norm": 0.640625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.4486,
|
|
"mean_token_accuracy": 0.05442229248583317,
|
|
"num_tokens": 499448.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.748478317260743,
|
|
"epoch": 0.02514848306490449,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.3757,
|
|
"mean_token_accuracy": 0.05678051896393299,
|
|
"num_tokens": 510148.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.693877410888671,
|
|
"epoch": 0.025683557172668414,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.3156,
|
|
"mean_token_accuracy": 0.053839444741606714,
|
|
"num_tokens": 522093.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.693435382843017,
|
|
"epoch": 0.02621863128043234,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.3365,
|
|
"mean_token_accuracy": 0.060893227159976956,
|
|
"num_tokens": 532598.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.625317096710205,
|
|
"epoch": 0.026753705388196266,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.3073,
|
|
"mean_token_accuracy": 0.05914956107735634,
|
|
"num_tokens": 542652.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.666687965393066,
|
|
"epoch": 0.027288779495960192,
|
|
"grad_norm": 0.62890625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.3343,
|
|
"mean_token_accuracy": 0.05921483002603054,
|
|
"num_tokens": 552526.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.59899835586548,
|
|
"epoch": 0.027823853603724115,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.357,
|
|
"mean_token_accuracy": 0.059534988924860956,
|
|
"num_tokens": 563143.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.635274982452392,
|
|
"epoch": 0.02835892771148804,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.3285,
|
|
"mean_token_accuracy": 0.05812331661581993,
|
|
"num_tokens": 573814.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.651222038269044,
|
|
"epoch": 0.028894001819251967,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.2886,
|
|
"mean_token_accuracy": 0.06336523443460465,
|
|
"num_tokens": 585367.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.534408664703369,
|
|
"epoch": 0.02942907592701589,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.3858,
|
|
"mean_token_accuracy": 0.06117042489349842,
|
|
"num_tokens": 595902.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.656291198730468,
|
|
"epoch": 0.029964150034779816,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.264,
|
|
"mean_token_accuracy": 0.05527368187904358,
|
|
"num_tokens": 607178.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.634803867340088,
|
|
"epoch": 0.030499224142543743,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.2565,
|
|
"mean_token_accuracy": 0.05927442014217377,
|
|
"num_tokens": 618604.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.526193332672118,
|
|
"epoch": 0.03103429825030767,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.3182,
|
|
"mean_token_accuracy": 0.06472528986632824,
|
|
"num_tokens": 628111.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.551859188079835,
|
|
"epoch": 0.03156937235807159,
|
|
"grad_norm": 0.62109375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.3524,
|
|
"mean_token_accuracy": 0.06350056529045105,
|
|
"num_tokens": 639358.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.559032249450684,
|
|
"epoch": 0.03210444646583552,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.2369,
|
|
"mean_token_accuracy": 0.061780452728271484,
|
|
"num_tokens": 649504.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.562529945373536,
|
|
"epoch": 0.032639520573599444,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.2503,
|
|
"mean_token_accuracy": 0.06190255731344223,
|
|
"num_tokens": 659924.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.53670949935913,
|
|
"epoch": 0.03317459468136337,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.2818,
|
|
"mean_token_accuracy": 0.060764011740684507,
|
|
"num_tokens": 671120.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.538024520874023,
|
|
"epoch": 0.033709668789127296,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.3386,
|
|
"mean_token_accuracy": 0.05993119850754738,
|
|
"num_tokens": 681956.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.596301746368407,
|
|
"epoch": 0.03424474289689122,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.3111,
|
|
"mean_token_accuracy": 0.060398251935839656,
|
|
"num_tokens": 693226.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.568628025054931,
|
|
"epoch": 0.03477981700465514,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.3253,
|
|
"mean_token_accuracy": 0.06589160449802875,
|
|
"num_tokens": 704691.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.47495059967041,
|
|
"epoch": 0.03531489111241907,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.2856,
|
|
"mean_token_accuracy": 0.05669932141900062,
|
|
"num_tokens": 716000.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.474805164337159,
|
|
"epoch": 0.035849965220182994,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.1999,
|
|
"mean_token_accuracy": 0.06121533028781414,
|
|
"num_tokens": 726723.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.498011589050293,
|
|
"epoch": 0.03638503932794692,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.218,
|
|
"mean_token_accuracy": 0.06442321352660656,
|
|
"num_tokens": 736740.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.463557243347168,
|
|
"epoch": 0.03692011343571085,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.2888,
|
|
"mean_token_accuracy": 0.06197960674762726,
|
|
"num_tokens": 747567.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.562387371063233,
|
|
"epoch": 0.03745518754347477,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.2184,
|
|
"mean_token_accuracy": 0.06554635129868984,
|
|
"num_tokens": 757937.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.381116771697998,
|
|
"epoch": 0.0379902616512387,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.2329,
|
|
"mean_token_accuracy": 0.06743667759001255,
|
|
"num_tokens": 768003.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.57433042526245,
|
|
"epoch": 0.03852533575900262,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.2918,
|
|
"mean_token_accuracy": 0.060314114391803744,
|
|
"num_tokens": 779098.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.524942874908447,
|
|
"epoch": 0.039060409866766545,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.2617,
|
|
"mean_token_accuracy": 0.05991893112659454,
|
|
"num_tokens": 790002.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.520723247528077,
|
|
"epoch": 0.03959548397453047,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.24,
|
|
"mean_token_accuracy": 0.0622980535030365,
|
|
"num_tokens": 800444.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.49396734237671,
|
|
"epoch": 0.0401305580822944,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.2068,
|
|
"mean_token_accuracy": 0.0646934848278761,
|
|
"num_tokens": 810912.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.529602336883546,
|
|
"epoch": 0.04066563219005832,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.2892,
|
|
"mean_token_accuracy": 0.061168758943676946,
|
|
"num_tokens": 822809.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.436344432830811,
|
|
"epoch": 0.04120070629782225,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.1843,
|
|
"mean_token_accuracy": 0.06499753147363663,
|
|
"num_tokens": 833362.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.464655780792237,
|
|
"epoch": 0.041735780405586176,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.2136,
|
|
"mean_token_accuracy": 0.06534497700631618,
|
|
"num_tokens": 844239.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.412539768218995,
|
|
"epoch": 0.0422708545133501,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.21,
|
|
"mean_token_accuracy": 0.05918620526790619,
|
|
"num_tokens": 855178.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.480862522125244,
|
|
"epoch": 0.04280592862111402,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.2122,
|
|
"mean_token_accuracy": 0.060220999643206596,
|
|
"num_tokens": 866120.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.434346771240234,
|
|
"epoch": 0.04334100272887795,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.1945,
|
|
"mean_token_accuracy": 0.06288457810878753,
|
|
"num_tokens": 876569.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.53380298614502,
|
|
"epoch": 0.043876076836641874,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.2451,
|
|
"mean_token_accuracy": 0.05817388035356998,
|
|
"num_tokens": 887580.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.417744541168213,
|
|
"epoch": 0.0444111509444058,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.1963,
|
|
"mean_token_accuracy": 0.06088313870131969,
|
|
"num_tokens": 897922.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.480956935882569,
|
|
"epoch": 0.044946225052169726,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.154,
|
|
"mean_token_accuracy": 0.062450488656759263,
|
|
"num_tokens": 908522.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.501266574859619,
|
|
"epoch": 0.04548129915993365,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.1712,
|
|
"mean_token_accuracy": 0.06173374205827713,
|
|
"num_tokens": 918253.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.375071716308593,
|
|
"epoch": 0.04601637326769758,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.1577,
|
|
"mean_token_accuracy": 0.06654028706252575,
|
|
"num_tokens": 929150.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.457318782806396,
|
|
"epoch": 0.0465514473754615,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.2037,
|
|
"mean_token_accuracy": 0.06640968844294548,
|
|
"num_tokens": 941041.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.420125675201415,
|
|
"epoch": 0.047086521483225424,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.1506,
|
|
"mean_token_accuracy": 0.06981723047792912,
|
|
"num_tokens": 952714.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.38680305480957,
|
|
"epoch": 0.04762159559098935,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.2206,
|
|
"mean_token_accuracy": 0.056760451197624205,
|
|
"num_tokens": 964330.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.4091046333313,
|
|
"epoch": 0.04815666969875328,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.1359,
|
|
"mean_token_accuracy": 0.06024715341627598,
|
|
"num_tokens": 975257.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.439432144165039,
|
|
"epoch": 0.0486917438065172,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.173,
|
|
"mean_token_accuracy": 0.06317405998706818,
|
|
"num_tokens": 986335.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.334716606140137,
|
|
"epoch": 0.04922681791428113,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.0587,
|
|
"mean_token_accuracy": 0.06826802715659142,
|
|
"num_tokens": 997521.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.383316135406494,
|
|
"epoch": 0.049761892022045055,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 8.0987,
|
|
"mean_token_accuracy": 0.061755566671490666,
|
|
"num_tokens": 1008383.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.42426586151123,
|
|
"epoch": 0.05029696612980898,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.1665,
|
|
"mean_token_accuracy": 0.06531071439385414,
|
|
"num_tokens": 1020552.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.319906425476074,
|
|
"epoch": 0.0508320402375729,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 8.1349,
|
|
"mean_token_accuracy": 0.06562352441251278,
|
|
"num_tokens": 1031127.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.294714164733886,
|
|
"epoch": 0.05136711434533683,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 8.0537,
|
|
"mean_token_accuracy": 0.06683347783982754,
|
|
"num_tokens": 1043209.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.477115249633789,
|
|
"epoch": 0.05190218845310075,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 8.182,
|
|
"mean_token_accuracy": 0.06099085658788681,
|
|
"num_tokens": 1054032.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.271202278137206,
|
|
"epoch": 0.05243726256086468,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 8.0176,
|
|
"mean_token_accuracy": 0.06989072859287263,
|
|
"num_tokens": 1064565.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.429040908813477,
|
|
"epoch": 0.052972336668628606,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 8.1709,
|
|
"mean_token_accuracy": 0.06545415259897709,
|
|
"num_tokens": 1076352.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.389126300811768,
|
|
"epoch": 0.05350741077639253,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 8.0995,
|
|
"mean_token_accuracy": 0.06924427561461925,
|
|
"num_tokens": 1087697.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.324423885345459,
|
|
"epoch": 0.05404248488415646,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 8.0191,
|
|
"mean_token_accuracy": 0.06561592519283295,
|
|
"num_tokens": 1098056.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.339075374603272,
|
|
"epoch": 0.054577558991920384,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 8.0061,
|
|
"mean_token_accuracy": 0.07367064505815506,
|
|
"num_tokens": 1109219.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.308711051940918,
|
|
"epoch": 0.055112633099684304,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 8.0719,
|
|
"mean_token_accuracy": 0.06625799126923085,
|
|
"num_tokens": 1120248.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.255969905853272,
|
|
"epoch": 0.05564770720744823,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 8.0894,
|
|
"mean_token_accuracy": 0.06586733907461166,
|
|
"num_tokens": 1131278.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.364176273345947,
|
|
"epoch": 0.056182781315212156,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 8.1062,
|
|
"mean_token_accuracy": 0.05967978574335575,
|
|
"num_tokens": 1143147.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.352798652648925,
|
|
"epoch": 0.05671785542297608,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 8.0109,
|
|
"mean_token_accuracy": 0.06417375653982163,
|
|
"num_tokens": 1154316.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.29659538269043,
|
|
"epoch": 0.05725292953074001,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 8.0685,
|
|
"mean_token_accuracy": 0.06354658231139183,
|
|
"num_tokens": 1164696.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.285994625091552,
|
|
"epoch": 0.057788003638503935,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 8.0421,
|
|
"mean_token_accuracy": 0.07412712648510933,
|
|
"num_tokens": 1176129.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.28906831741333,
|
|
"epoch": 0.05832307774626786,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 8.071,
|
|
"mean_token_accuracy": 0.06631042063236237,
|
|
"num_tokens": 1186499.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.289284801483154,
|
|
"epoch": 0.05885815185403178,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 8.0173,
|
|
"mean_token_accuracy": 0.06871062219142914,
|
|
"num_tokens": 1196841.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.295639991760254,
|
|
"epoch": 0.059393225961795706,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.9761,
|
|
"mean_token_accuracy": 0.07033407911658288,
|
|
"num_tokens": 1207647.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.243279266357423,
|
|
"epoch": 0.05992830006955963,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.963,
|
|
"mean_token_accuracy": 0.072792012616992,
|
|
"num_tokens": 1217072.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.300912570953368,
|
|
"epoch": 0.06046337417732356,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 8.073,
|
|
"mean_token_accuracy": 0.06520008333027363,
|
|
"num_tokens": 1228564.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.203814697265624,
|
|
"epoch": 0.060998448285087485,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.9942,
|
|
"mean_token_accuracy": 0.06911226697266101,
|
|
"num_tokens": 1238755.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.239333248138427,
|
|
"epoch": 0.06153352239285141,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.9646,
|
|
"mean_token_accuracy": 0.07128063514828682,
|
|
"num_tokens": 1248968.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.241281127929687,
|
|
"epoch": 0.06206859650061534,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.9258,
|
|
"mean_token_accuracy": 0.0690683875232935,
|
|
"num_tokens": 1259256.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.180652809143066,
|
|
"epoch": 0.06260367060837926,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.9653,
|
|
"mean_token_accuracy": 0.07212716266512871,
|
|
"num_tokens": 1270469.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.237191200256348,
|
|
"epoch": 0.06313874471614318,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.9286,
|
|
"mean_token_accuracy": 0.06839926056563854,
|
|
"num_tokens": 1280558.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.18700132369995,
|
|
"epoch": 0.06367381882390712,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.9695,
|
|
"mean_token_accuracy": 0.06805149577558041,
|
|
"num_tokens": 1291372.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.287422943115235,
|
|
"epoch": 0.06420889293167104,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 8.0554,
|
|
"mean_token_accuracy": 0.06435842141509056,
|
|
"num_tokens": 1302521.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.17969207763672,
|
|
"epoch": 0.06474396703943497,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.8954,
|
|
"mean_token_accuracy": 0.07011312432587147,
|
|
"num_tokens": 1313632.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.187294721603394,
|
|
"epoch": 0.06527904114719889,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.9765,
|
|
"mean_token_accuracy": 0.06986295953392982,
|
|
"num_tokens": 1325247.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.145281457901001,
|
|
"epoch": 0.06581411525496281,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.9495,
|
|
"mean_token_accuracy": 0.07135364897549153,
|
|
"num_tokens": 1336026.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.15605869293213,
|
|
"epoch": 0.06634918936272674,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.8724,
|
|
"mean_token_accuracy": 0.07117584012448788,
|
|
"num_tokens": 1346920.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.162275314331055,
|
|
"epoch": 0.06688426347049066,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 8.0208,
|
|
"mean_token_accuracy": 0.06915087811648846,
|
|
"num_tokens": 1359752.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.272911071777344,
|
|
"epoch": 0.06741933757825459,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 8.0125,
|
|
"mean_token_accuracy": 0.06714854501187802,
|
|
"num_tokens": 1370874.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.166405773162841,
|
|
"epoch": 0.06795441168601851,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.933,
|
|
"mean_token_accuracy": 0.07350475676357746,
|
|
"num_tokens": 1382316.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 8.12221031188965,
|
|
"epoch": 0.06848948579378245,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.9287,
|
|
"mean_token_accuracy": 0.06466298960149289,
|
|
"num_tokens": 1394656.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.132702732086182,
|
|
"epoch": 0.06902455990154636,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.904,
|
|
"mean_token_accuracy": 0.07535643540322781,
|
|
"num_tokens": 1405061.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 8.1514066696167,
|
|
"epoch": 0.06955963400931028,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.8574,
|
|
"mean_token_accuracy": 0.07722640819847584,
|
|
"num_tokens": 1415573.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.076834058761596,
|
|
"epoch": 0.07009470811707422,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.8971,
|
|
"mean_token_accuracy": 0.06901067160069943,
|
|
"num_tokens": 1426103.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 8.042639923095702,
|
|
"epoch": 0.07062978222483814,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.9236,
|
|
"mean_token_accuracy": 0.06627062112092971,
|
|
"num_tokens": 1437256.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 8.271612071990967,
|
|
"epoch": 0.07116485633260207,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.8667,
|
|
"mean_token_accuracy": 0.07184043861925601,
|
|
"num_tokens": 1446975.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 8.110226726531982,
|
|
"epoch": 0.07169993044036599,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.8856,
|
|
"mean_token_accuracy": 0.07188437804579735,
|
|
"num_tokens": 1457751.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.960531234741211,
|
|
"epoch": 0.07223500454812992,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.8516,
|
|
"mean_token_accuracy": 0.07059676125645638,
|
|
"num_tokens": 1468186.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 8.202023792266846,
|
|
"epoch": 0.07277007865589384,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.896,
|
|
"mean_token_accuracy": 0.0678649365901947,
|
|
"num_tokens": 1479780.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 8.120744943618774,
|
|
"epoch": 0.07330515276365776,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.841,
|
|
"mean_token_accuracy": 0.07492000721395016,
|
|
"num_tokens": 1490565.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 8.100931549072266,
|
|
"epoch": 0.0738402268714217,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.8612,
|
|
"mean_token_accuracy": 0.07190988808870316,
|
|
"num_tokens": 1501552.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 8.065852403640747,
|
|
"epoch": 0.07437530097918561,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.8713,
|
|
"mean_token_accuracy": 0.06612264662981034,
|
|
"num_tokens": 1513311.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 8.115069341659545,
|
|
"epoch": 0.07491037508694955,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.895,
|
|
"mean_token_accuracy": 0.07125801891088486,
|
|
"num_tokens": 1524142.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 8.081360721588135,
|
|
"epoch": 0.07544544919471347,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.8645,
|
|
"mean_token_accuracy": 0.07241233922541142,
|
|
"num_tokens": 1535571.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 8.069763135910033,
|
|
"epoch": 0.0759805233024774,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.8535,
|
|
"mean_token_accuracy": 0.07254451140761375,
|
|
"num_tokens": 1546528.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 8.03998236656189,
|
|
"epoch": 0.07651559741024132,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.8782,
|
|
"mean_token_accuracy": 0.06598596908152103,
|
|
"num_tokens": 1557264.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 8.095155239105225,
|
|
"epoch": 0.07705067151800524,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.8277,
|
|
"mean_token_accuracy": 0.07438212521374225,
|
|
"num_tokens": 1566773.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.952378559112549,
|
|
"epoch": 0.07758574562576917,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.7892,
|
|
"mean_token_accuracy": 0.07861022278666496,
|
|
"num_tokens": 1578173.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 8.084946393966675,
|
|
"epoch": 0.07812081973353309,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.7521,
|
|
"mean_token_accuracy": 0.07678151875734329,
|
|
"num_tokens": 1588537.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.978636407852173,
|
|
"epoch": 0.07865589384129702,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.7674,
|
|
"mean_token_accuracy": 0.08002844266593456,
|
|
"num_tokens": 1599159.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 8.004686641693116,
|
|
"epoch": 0.07919096794906094,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.8519,
|
|
"mean_token_accuracy": 0.07209457606077194,
|
|
"num_tokens": 1610378.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 8.040560579299926,
|
|
"epoch": 0.07972604205682488,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.861,
|
|
"mean_token_accuracy": 0.07267312295734882,
|
|
"num_tokens": 1621215.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.936432123184204,
|
|
"epoch": 0.0802611161645888,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.79,
|
|
"mean_token_accuracy": 0.07331596687436104,
|
|
"num_tokens": 1631170.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 8.133660507202148,
|
|
"epoch": 0.08079619027235273,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.7989,
|
|
"mean_token_accuracy": 0.07404692210257054,
|
|
"num_tokens": 1641897.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.98411865234375,
|
|
"epoch": 0.08133126438011665,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.8606,
|
|
"mean_token_accuracy": 0.07070158869028091,
|
|
"num_tokens": 1653592.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.9289227485656735,
|
|
"epoch": 0.08186633848788057,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.7832,
|
|
"mean_token_accuracy": 0.06835788786411286,
|
|
"num_tokens": 1665760.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 8.100461387634278,
|
|
"epoch": 0.0824014125956445,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.7963,
|
|
"mean_token_accuracy": 0.07924762107431889,
|
|
"num_tokens": 1676211.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 8.04494333267212,
|
|
"epoch": 0.08293648670340842,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.8154,
|
|
"mean_token_accuracy": 0.07205268889665603,
|
|
"num_tokens": 1686405.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 8.116489696502686,
|
|
"epoch": 0.08347156081117235,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.816,
|
|
"mean_token_accuracy": 0.07210099287331104,
|
|
"num_tokens": 1697340.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 8.015373849868775,
|
|
"epoch": 0.08400663491893627,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.716,
|
|
"mean_token_accuracy": 0.0766339547932148,
|
|
"num_tokens": 1707250.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.935082626342774,
|
|
"epoch": 0.0845417090267002,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.6759,
|
|
"mean_token_accuracy": 0.07099579833447933,
|
|
"num_tokens": 1718333.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 8.006236696243286,
|
|
"epoch": 0.08507678313446412,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.7727,
|
|
"mean_token_accuracy": 0.07233709581196308,
|
|
"num_tokens": 1730520.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.998106575012207,
|
|
"epoch": 0.08561185724222804,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.8245,
|
|
"mean_token_accuracy": 0.0690892156213522,
|
|
"num_tokens": 1741898.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.937285470962524,
|
|
"epoch": 0.08614693134999198,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.7619,
|
|
"mean_token_accuracy": 0.07777528837323189,
|
|
"num_tokens": 1751671.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.906507158279419,
|
|
"epoch": 0.0866820054577559,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.7796,
|
|
"mean_token_accuracy": 0.07674273923039436,
|
|
"num_tokens": 1762076.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 8.027475690841674,
|
|
"epoch": 0.08721707956551983,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.7598,
|
|
"mean_token_accuracy": 0.07528497837483883,
|
|
"num_tokens": 1772666.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.961116218566895,
|
|
"epoch": 0.08775215367328375,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.7747,
|
|
"mean_token_accuracy": 0.08033046908676625,
|
|
"num_tokens": 1783621.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.981131458282471,
|
|
"epoch": 0.08828722778104768,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.8234,
|
|
"mean_token_accuracy": 0.07162153050303459,
|
|
"num_tokens": 1794654.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.881627893447876,
|
|
"epoch": 0.0888223018888116,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.7166,
|
|
"mean_token_accuracy": 0.08037983030080795,
|
|
"num_tokens": 1805853.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.805557489395142,
|
|
"epoch": 0.08935737599657552,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.6985,
|
|
"mean_token_accuracy": 0.07677435502409935,
|
|
"num_tokens": 1817006.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 8.030841255187989,
|
|
"epoch": 0.08989245010433945,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.6996,
|
|
"mean_token_accuracy": 0.07248782590031624,
|
|
"num_tokens": 1827166.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 8.012809896469117,
|
|
"epoch": 0.09042752421210337,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.781,
|
|
"mean_token_accuracy": 0.07629549242556095,
|
|
"num_tokens": 1837182.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.861330270767212,
|
|
"epoch": 0.0909625983198673,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.7371,
|
|
"mean_token_accuracy": 0.07596477270126342,
|
|
"num_tokens": 1849034.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.900086402893066,
|
|
"epoch": 0.09149767242763122,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.6039,
|
|
"mean_token_accuracy": 0.08392401710152626,
|
|
"num_tokens": 1859033.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.819063472747803,
|
|
"epoch": 0.09203274653539516,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.7286,
|
|
"mean_token_accuracy": 0.07808272317051887,
|
|
"num_tokens": 1870274.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.888581132888794,
|
|
"epoch": 0.09256782064315908,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.6714,
|
|
"mean_token_accuracy": 0.08056179471313954,
|
|
"num_tokens": 1881306.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.833876037597657,
|
|
"epoch": 0.093102894750923,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.6631,
|
|
"mean_token_accuracy": 0.07485895380377769,
|
|
"num_tokens": 1892464.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.968428325653076,
|
|
"epoch": 0.09363796885868693,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.7325,
|
|
"mean_token_accuracy": 0.07241618894040584,
|
|
"num_tokens": 1903743.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.845100402832031,
|
|
"epoch": 0.09417304296645085,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.6869,
|
|
"mean_token_accuracy": 0.08143132627010345,
|
|
"num_tokens": 1915253.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.928981351852417,
|
|
"epoch": 0.09470811707421478,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.6399,
|
|
"mean_token_accuracy": 0.08035231530666351,
|
|
"num_tokens": 1926113.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.823946809768676,
|
|
"epoch": 0.0952431911819787,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.735,
|
|
"mean_token_accuracy": 0.07786822356283665,
|
|
"num_tokens": 1937718.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.9432416439056395,
|
|
"epoch": 0.09577826528974263,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.7198,
|
|
"mean_token_accuracy": 0.07977167144417763,
|
|
"num_tokens": 1948307.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.745344495773315,
|
|
"epoch": 0.09631333939750655,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.5286,
|
|
"mean_token_accuracy": 0.08684809878468513,
|
|
"num_tokens": 1957861.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.771445035934448,
|
|
"epoch": 0.09684841350527049,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.5873,
|
|
"mean_token_accuracy": 0.08686093017458915,
|
|
"num_tokens": 1967984.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.84833459854126,
|
|
"epoch": 0.0973834876130344,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.5976,
|
|
"mean_token_accuracy": 0.08169375136494636,
|
|
"num_tokens": 1978608.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.825530004501343,
|
|
"epoch": 0.09791856172079832,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.6436,
|
|
"mean_token_accuracy": 0.08153785914182662,
|
|
"num_tokens": 1989355.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.696963882446289,
|
|
"epoch": 0.09845363582856226,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.6287,
|
|
"mean_token_accuracy": 0.08496365323662758,
|
|
"num_tokens": 1999754.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.85645866394043,
|
|
"epoch": 0.09898870993632618,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.638,
|
|
"mean_token_accuracy": 0.08035108707845211,
|
|
"num_tokens": 2010877.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.902437019348144,
|
|
"epoch": 0.09952378404409011,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.7244,
|
|
"mean_token_accuracy": 0.0775139220058918,
|
|
"num_tokens": 2022301.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.757477712631226,
|
|
"epoch": 0.10005885815185403,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.6194,
|
|
"mean_token_accuracy": 0.08088013455271721,
|
|
"num_tokens": 2033537.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.801491928100586,
|
|
"epoch": 0.10059393225961796,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.5671,
|
|
"mean_token_accuracy": 0.0867505919188261,
|
|
"num_tokens": 2044360.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.825805759429931,
|
|
"epoch": 0.10112900636738188,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.6443,
|
|
"mean_token_accuracy": 0.08098645284771919,
|
|
"num_tokens": 2055000.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.741448211669922,
|
|
"epoch": 0.1016640804751458,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.6416,
|
|
"mean_token_accuracy": 0.08422183245420456,
|
|
"num_tokens": 2064807.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.827320861816406,
|
|
"epoch": 0.10219915458290973,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.6029,
|
|
"mean_token_accuracy": 0.07696961686015129,
|
|
"num_tokens": 2075618.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.788719272613525,
|
|
"epoch": 0.10273422869067365,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.6011,
|
|
"mean_token_accuracy": 0.0814262680709362,
|
|
"num_tokens": 2086226.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.8088513851165775,
|
|
"epoch": 0.10326930279843759,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.5913,
|
|
"mean_token_accuracy": 0.078329161927104,
|
|
"num_tokens": 2097222.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.7249979972839355,
|
|
"epoch": 0.1038043769062015,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.6347,
|
|
"mean_token_accuracy": 0.08402503952383995,
|
|
"num_tokens": 2108919.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.705420303344726,
|
|
"epoch": 0.10433945101396544,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.5907,
|
|
"mean_token_accuracy": 0.07948351129889489,
|
|
"num_tokens": 2120215.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.832537698745727,
|
|
"epoch": 0.10487452512172936,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.6102,
|
|
"mean_token_accuracy": 0.07788611873984337,
|
|
"num_tokens": 2131503.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.718122386932373,
|
|
"epoch": 0.10540959922949328,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.5579,
|
|
"mean_token_accuracy": 0.08647179454565049,
|
|
"num_tokens": 2141551.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.673153209686279,
|
|
"epoch": 0.10594467333725721,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.5469,
|
|
"mean_token_accuracy": 0.08761123642325401,
|
|
"num_tokens": 2151821.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.752789545059204,
|
|
"epoch": 0.10647974744502113,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.5979,
|
|
"mean_token_accuracy": 0.0813947595655918,
|
|
"num_tokens": 2162474.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.811589527130127,
|
|
"epoch": 0.10701482155278506,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.6235,
|
|
"mean_token_accuracy": 0.08300729878246785,
|
|
"num_tokens": 2173985.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.779503297805786,
|
|
"epoch": 0.10754989566054898,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004999999979210101,
|
|
"loss": 7.6631,
|
|
"mean_token_accuracy": 0.07762378752231598,
|
|
"num_tokens": 2186432.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.744147109985351,
|
|
"epoch": 0.10808496976831292,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999999894751138,
|
|
"loss": 7.5197,
|
|
"mean_token_accuracy": 0.0851745329797268,
|
|
"num_tokens": 2197085.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.667597770690918,
|
|
"epoch": 0.10862004387607684,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999999745323744,
|
|
"loss": 7.539,
|
|
"mean_token_accuracy": 0.0865162130445242,
|
|
"num_tokens": 2207162.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.788536834716797,
|
|
"epoch": 0.10915511798384077,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999999530927923,
|
|
"loss": 7.6289,
|
|
"mean_token_accuracy": 0.07713088616728783,
|
|
"num_tokens": 2218714.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.642853784561157,
|
|
"epoch": 0.10969019209160469,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004999999251563682,
|
|
"loss": 7.5508,
|
|
"mean_token_accuracy": 0.08075930066406727,
|
|
"num_tokens": 2229711.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.76542911529541,
|
|
"epoch": 0.11022526619936861,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004999998907231029,
|
|
"loss": 7.559,
|
|
"mean_token_accuracy": 0.08542619198560715,
|
|
"num_tokens": 2239885.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.702492570877075,
|
|
"epoch": 0.11076034030713254,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999998497929974,
|
|
"loss": 7.5658,
|
|
"mean_token_accuracy": 0.08461903110146522,
|
|
"num_tokens": 2250221.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.676487588882447,
|
|
"epoch": 0.11129541441489646,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004999998023660527,
|
|
"loss": 7.6434,
|
|
"mean_token_accuracy": 0.07421790175139904,
|
|
"num_tokens": 2261059.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.728254747390747,
|
|
"epoch": 0.11183048852266039,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004999997484422705,
|
|
"loss": 7.4963,
|
|
"mean_token_accuracy": 0.08282011076807975,
|
|
"num_tokens": 2271174.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.690599012374878,
|
|
"epoch": 0.11236556263042431,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999996880216521,
|
|
"loss": 7.5629,
|
|
"mean_token_accuracy": 0.08049203157424926,
|
|
"num_tokens": 2282431.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.683617353439331,
|
|
"epoch": 0.11290063673818825,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999996211041994,
|
|
"loss": 7.5512,
|
|
"mean_token_accuracy": 0.08223441652953625,
|
|
"num_tokens": 2292891.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.760586452484131,
|
|
"epoch": 0.11343571084595216,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999995476899141,
|
|
"loss": 7.5679,
|
|
"mean_token_accuracy": 0.08535757660865784,
|
|
"num_tokens": 2303764.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.620778036117554,
|
|
"epoch": 0.11397078495371608,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999994677787987,
|
|
"loss": 7.4748,
|
|
"mean_token_accuracy": 0.08901284039020538,
|
|
"num_tokens": 2314172.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.693076801300049,
|
|
"epoch": 0.11450585906148002,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999993813708551,
|
|
"loss": 7.4564,
|
|
"mean_token_accuracy": 0.08755525574088097,
|
|
"num_tokens": 2324848.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.582232046127319,
|
|
"epoch": 0.11504093316924394,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000499999288466086,
|
|
"loss": 7.4952,
|
|
"mean_token_accuracy": 0.08118128105998039,
|
|
"num_tokens": 2336601.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.6848491668701175,
|
|
"epoch": 0.11557600727700787,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004999991890644941,
|
|
"loss": 7.468,
|
|
"mean_token_accuracy": 0.08362995944917202,
|
|
"num_tokens": 2347281.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.642825555801392,
|
|
"epoch": 0.11611108138477179,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999990831660822,
|
|
"loss": 7.5125,
|
|
"mean_token_accuracy": 0.08922351896762848,
|
|
"num_tokens": 2358199.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.686036920547485,
|
|
"epoch": 0.11664615549253572,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999989707708534,
|
|
"loss": 7.5507,
|
|
"mean_token_accuracy": 0.08290127031505108,
|
|
"num_tokens": 2369092.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.613842344284057,
|
|
"epoch": 0.11718122960029964,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999988518788111,
|
|
"loss": 7.5223,
|
|
"mean_token_accuracy": 0.08195193111896515,
|
|
"num_tokens": 2378849.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.688374471664429,
|
|
"epoch": 0.11771630370806356,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004999987264899583,
|
|
"loss": 7.5234,
|
|
"mean_token_accuracy": 0.08619830012321472,
|
|
"num_tokens": 2389587.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.740108346939087,
|
|
"epoch": 0.1182513778158275,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000499998594604299,
|
|
"loss": 7.4909,
|
|
"mean_token_accuracy": 0.08550030142068862,
|
|
"num_tokens": 2400442.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.534623289108277,
|
|
"epoch": 0.11878645192359141,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000499998456221837,
|
|
"loss": 7.4883,
|
|
"mean_token_accuracy": 0.08606440126895905,
|
|
"num_tokens": 2410771.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.644052982330322,
|
|
"epoch": 0.11932152603135535,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999983113425762,
|
|
"loss": 7.4918,
|
|
"mean_token_accuracy": 0.09239444956183433,
|
|
"num_tokens": 2421112.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.656321287155151,
|
|
"epoch": 0.11985660013911927,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999981599665207,
|
|
"loss": 7.4718,
|
|
"mean_token_accuracy": 0.08363690003752708,
|
|
"num_tokens": 2431891.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.589166831970215,
|
|
"epoch": 0.1203916742468832,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999980020936748,
|
|
"loss": 7.4192,
|
|
"mean_token_accuracy": 0.09008971899747849,
|
|
"num_tokens": 2442987.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.620885610580444,
|
|
"epoch": 0.12092674835464712,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999978377240434,
|
|
"loss": 7.4522,
|
|
"mean_token_accuracy": 0.08518644496798515,
|
|
"num_tokens": 2452703.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.634619951248169,
|
|
"epoch": 0.12146182246241104,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000499997666857631,
|
|
"loss": 7.5143,
|
|
"mean_token_accuracy": 0.08547966852784157,
|
|
"num_tokens": 2462975.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.556209659576416,
|
|
"epoch": 0.12199689657017497,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004999974894944426,
|
|
"loss": 7.465,
|
|
"mean_token_accuracy": 0.08037517294287681,
|
|
"num_tokens": 2473853.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.648143863677978,
|
|
"epoch": 0.12253197067793889,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004999973056344832,
|
|
"loss": 7.4356,
|
|
"mean_token_accuracy": 0.08240287378430367,
|
|
"num_tokens": 2485633.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.544134950637817,
|
|
"epoch": 0.12306704478570282,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004999971152777583,
|
|
"loss": 7.4008,
|
|
"mean_token_accuracy": 0.0854021079838276,
|
|
"num_tokens": 2496922.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.654058313369751,
|
|
"epoch": 0.12360211889346674,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004999969184242733,
|
|
"loss": 7.4612,
|
|
"mean_token_accuracy": 0.08849633410573006,
|
|
"num_tokens": 2507890.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.603252363204956,
|
|
"epoch": 0.12413719300123068,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000499996715074034,
|
|
"loss": 7.4782,
|
|
"mean_token_accuracy": 0.08564752526581287,
|
|
"num_tokens": 2519749.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.571427631378174,
|
|
"epoch": 0.1246722671089946,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999965052270461,
|
|
"loss": 7.4656,
|
|
"mean_token_accuracy": 0.08699908182024955,
|
|
"num_tokens": 2530666.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.64540319442749,
|
|
"epoch": 0.12520734121675853,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004999962888833157,
|
|
"loss": 7.471,
|
|
"mean_token_accuracy": 0.08744210675358773,
|
|
"num_tokens": 2541594.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.585118579864502,
|
|
"epoch": 0.12574241532452243,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999960660428491,
|
|
"loss": 7.4757,
|
|
"mean_token_accuracy": 0.08734978139400482,
|
|
"num_tokens": 2552330.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.640705585479736,
|
|
"epoch": 0.12627748943228637,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999958367056526,
|
|
"loss": 7.5049,
|
|
"mean_token_accuracy": 0.08555901274085045,
|
|
"num_tokens": 2564601.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.65483775138855,
|
|
"epoch": 0.1268125635400503,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000499995600871733,
|
|
"loss": 7.4721,
|
|
"mean_token_accuracy": 0.08904926031827927,
|
|
"num_tokens": 2575539.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.503138256072998,
|
|
"epoch": 0.12734763764781423,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004999953585410971,
|
|
"loss": 7.4135,
|
|
"mean_token_accuracy": 0.08864182159304619,
|
|
"num_tokens": 2586635.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.650179767608643,
|
|
"epoch": 0.12788271175557814,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999951097137518,
|
|
"loss": 7.4342,
|
|
"mean_token_accuracy": 0.09002284705638885,
|
|
"num_tokens": 2596814.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.510055303573608,
|
|
"epoch": 0.12841778586334207,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999948543897044,
|
|
"loss": 7.4695,
|
|
"mean_token_accuracy": 0.08551743030548095,
|
|
"num_tokens": 2608087.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.754941749572754,
|
|
"epoch": 0.128952859971106,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999945925689621,
|
|
"loss": 7.4764,
|
|
"mean_token_accuracy": 0.09063678905367852,
|
|
"num_tokens": 2619377.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.45523681640625,
|
|
"epoch": 0.12948793407886994,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999943242515325,
|
|
"loss": 7.3686,
|
|
"mean_token_accuracy": 0.08870491459965706,
|
|
"num_tokens": 2629089.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.6135406494140625,
|
|
"epoch": 0.13002300818663384,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999940494374236,
|
|
"loss": 7.4405,
|
|
"mean_token_accuracy": 0.0854348972439766,
|
|
"num_tokens": 2639743.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.64836049079895,
|
|
"epoch": 0.13055808229439778,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000499993768126643,
|
|
"loss": 7.4719,
|
|
"mean_token_accuracy": 0.08272441625595092,
|
|
"num_tokens": 2650631.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.567201042175293,
|
|
"epoch": 0.1310931564021617,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499993480319199,
|
|
"loss": 7.4509,
|
|
"mean_token_accuracy": 0.08986086025834084,
|
|
"num_tokens": 2660197.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.526758337020874,
|
|
"epoch": 0.13162823050992561,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999931860150999,
|
|
"loss": 7.4156,
|
|
"mean_token_accuracy": 0.08493227660655975,
|
|
"num_tokens": 2670162.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.630824851989746,
|
|
"epoch": 0.13216330461768955,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004999928852143541,
|
|
"loss": 7.4031,
|
|
"mean_token_accuracy": 0.09029604569077491,
|
|
"num_tokens": 2681233.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.431710243225098,
|
|
"epoch": 0.13269837872545348,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004999925779169703,
|
|
"loss": 7.3671,
|
|
"mean_token_accuracy": 0.09119424894452095,
|
|
"num_tokens": 2692145.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.588590002059936,
|
|
"epoch": 0.13323345283321741,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004999922641229575,
|
|
"loss": 7.4026,
|
|
"mean_token_accuracy": 0.09271374940872193,
|
|
"num_tokens": 2703192.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.56281681060791,
|
|
"epoch": 0.13376852694098132,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004999919438323248,
|
|
"loss": 7.4117,
|
|
"mean_token_accuracy": 0.08688192069530487,
|
|
"num_tokens": 2714804.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.49661545753479,
|
|
"epoch": 0.13430360104874525,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999916170450812,
|
|
"loss": 7.3701,
|
|
"mean_token_accuracy": 0.0922422744333744,
|
|
"num_tokens": 2724340.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.538181924819947,
|
|
"epoch": 0.13483867515650919,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004999912837612364,
|
|
"loss": 7.3767,
|
|
"mean_token_accuracy": 0.09194523319602013,
|
|
"num_tokens": 2735767.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.622563457489013,
|
|
"epoch": 0.1353737492642731,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999909439807997,
|
|
"loss": 7.4463,
|
|
"mean_token_accuracy": 0.08266032785177231,
|
|
"num_tokens": 2747658.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.48611364364624,
|
|
"epoch": 0.13590882337203702,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999905977037813,
|
|
"loss": 7.312,
|
|
"mean_token_accuracy": 0.09179738909006119,
|
|
"num_tokens": 2757901.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.4827721118927,
|
|
"epoch": 0.13644389747980096,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999902449301909,
|
|
"loss": 7.3539,
|
|
"mean_token_accuracy": 0.08355544060468674,
|
|
"num_tokens": 2768945.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.4961072444915775,
|
|
"epoch": 0.1369789715875649,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004999898856600387,
|
|
"loss": 7.3173,
|
|
"mean_token_accuracy": 0.10180329754948617,
|
|
"num_tokens": 2780276.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.553973293304443,
|
|
"epoch": 0.1375140456953288,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999895198933354,
|
|
"loss": 7.3965,
|
|
"mean_token_accuracy": 0.0919288918375969,
|
|
"num_tokens": 2791183.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.49184365272522,
|
|
"epoch": 0.13804911980309273,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999891476300911,
|
|
"loss": 7.3122,
|
|
"mean_token_accuracy": 0.08812543600797654,
|
|
"num_tokens": 2801928.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.552814531326294,
|
|
"epoch": 0.13858419391085666,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000499988768870317,
|
|
"loss": 7.3956,
|
|
"mean_token_accuracy": 0.08465510159730912,
|
|
"num_tokens": 2813415.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.482685708999634,
|
|
"epoch": 0.13911926801862057,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999883836140236,
|
|
"loss": 7.4172,
|
|
"mean_token_accuracy": 0.08735132664442062,
|
|
"num_tokens": 2825288.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.536992979049683,
|
|
"epoch": 0.1396543421263845,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999879918612224,
|
|
"loss": 7.3522,
|
|
"mean_token_accuracy": 0.09331145882606506,
|
|
"num_tokens": 2836298.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.510290431976318,
|
|
"epoch": 0.14018941623414843,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999875936119245,
|
|
"loss": 7.3316,
|
|
"mean_token_accuracy": 0.09056525379419327,
|
|
"num_tokens": 2846367.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.458161878585815,
|
|
"epoch": 0.14072449034191237,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999871888661415,
|
|
"loss": 7.3488,
|
|
"mean_token_accuracy": 0.08874687254428863,
|
|
"num_tokens": 2856832.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.506878137588501,
|
|
"epoch": 0.14125956444967627,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000499986777623885,
|
|
"loss": 7.3681,
|
|
"mean_token_accuracy": 0.08353512287139893,
|
|
"num_tokens": 2867954.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.641098499298096,
|
|
"epoch": 0.1417946385574402,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999863598851669,
|
|
"loss": 7.4016,
|
|
"mean_token_accuracy": 0.09097437039017678,
|
|
"num_tokens": 2878714.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.356536340713501,
|
|
"epoch": 0.14232971266520414,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999859356499992,
|
|
"loss": 7.3319,
|
|
"mean_token_accuracy": 0.09417055919766426,
|
|
"num_tokens": 2890697.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.514051151275635,
|
|
"epoch": 0.14286478677296804,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999855049183943,
|
|
"loss": 7.3504,
|
|
"mean_token_accuracy": 0.08786595612764359,
|
|
"num_tokens": 2901468.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.498105907440186,
|
|
"epoch": 0.14339986088073198,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004999850676903646,
|
|
"loss": 7.3144,
|
|
"mean_token_accuracy": 0.09161052480340004,
|
|
"num_tokens": 2913466.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.440569972991943,
|
|
"epoch": 0.1439349349884959,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004999846239659226,
|
|
"loss": 7.2868,
|
|
"mean_token_accuracy": 0.09555015340447426,
|
|
"num_tokens": 2923638.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.524512910842896,
|
|
"epoch": 0.14447000909625984,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999841737450812,
|
|
"loss": 7.3678,
|
|
"mean_token_accuracy": 0.0967521421611309,
|
|
"num_tokens": 2934790.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.4529260158538815,
|
|
"epoch": 0.14500508320402375,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999837170278535,
|
|
"loss": 7.3126,
|
|
"mean_token_accuracy": 0.0890039324760437,
|
|
"num_tokens": 2946320.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.528018951416016,
|
|
"epoch": 0.14554015731178768,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999832538142526,
|
|
"loss": 7.2832,
|
|
"mean_token_accuracy": 0.08959799520671367,
|
|
"num_tokens": 2957077.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.395437908172608,
|
|
"epoch": 0.14607523141955162,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999827841042917,
|
|
"loss": 7.305,
|
|
"mean_token_accuracy": 0.08984916433691978,
|
|
"num_tokens": 2967286.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.475974082946777,
|
|
"epoch": 0.14661030552731552,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999823078979846,
|
|
"loss": 7.342,
|
|
"mean_token_accuracy": 0.0904863603413105,
|
|
"num_tokens": 2979516.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.595507574081421,
|
|
"epoch": 0.14714537963507945,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999818251953449,
|
|
"loss": 7.4628,
|
|
"mean_token_accuracy": 0.08827103711664677,
|
|
"num_tokens": 2991594.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.429067468643188,
|
|
"epoch": 0.1476804537428434,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999813359963867,
|
|
"loss": 7.3428,
|
|
"mean_token_accuracy": 0.08966336324810982,
|
|
"num_tokens": 3002487.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.569780778884888,
|
|
"epoch": 0.14821552785060732,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999808403011241,
|
|
"loss": 7.3603,
|
|
"mean_token_accuracy": 0.09212637171149254,
|
|
"num_tokens": 3013197.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.384968948364258,
|
|
"epoch": 0.14875060195837123,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999803381095712,
|
|
"loss": 7.3808,
|
|
"mean_token_accuracy": 0.08996171355247498,
|
|
"num_tokens": 3024512.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.576485824584961,
|
|
"epoch": 0.14928567606613516,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999798294217428,
|
|
"loss": 7.346,
|
|
"mean_token_accuracy": 0.08968236446380615,
|
|
"num_tokens": 3035621.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.431740236282349,
|
|
"epoch": 0.1498207501738991,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999793142376533,
|
|
"loss": 7.3432,
|
|
"mean_token_accuracy": 0.08854975476861,
|
|
"num_tokens": 3046492.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.440773344039917,
|
|
"epoch": 0.150355824281663,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999787925573177,
|
|
"loss": 7.3546,
|
|
"mean_token_accuracy": 0.09137408435344696,
|
|
"num_tokens": 3057715.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.582657051086426,
|
|
"epoch": 0.15089089838942693,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999782643807513,
|
|
"loss": 7.3135,
|
|
"mean_token_accuracy": 0.08860881105065346,
|
|
"num_tokens": 3068344.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.376219367980957,
|
|
"epoch": 0.15142597249719086,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004999777297079689,
|
|
"loss": 7.2929,
|
|
"mean_token_accuracy": 0.08662425726652145,
|
|
"num_tokens": 3079576.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.48095703125,
|
|
"epoch": 0.1519610466049548,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999771885389863,
|
|
"loss": 7.2626,
|
|
"mean_token_accuracy": 0.09326665848493576,
|
|
"num_tokens": 3090201.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.417229890823364,
|
|
"epoch": 0.1524961207127187,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999766408738189,
|
|
"loss": 7.2349,
|
|
"mean_token_accuracy": 0.09925753399729728,
|
|
"num_tokens": 3099824.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.380178737640381,
|
|
"epoch": 0.15303119482048264,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999760867124827,
|
|
"loss": 7.3052,
|
|
"mean_token_accuracy": 0.09141960963606835,
|
|
"num_tokens": 3109948.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.5375657081604,
|
|
"epoch": 0.15356626892824657,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999755260549937,
|
|
"loss": 7.3051,
|
|
"mean_token_accuracy": 0.09628649353981018,
|
|
"num_tokens": 3121492.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.324652862548828,
|
|
"epoch": 0.15410134303601047,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004999749589013677,
|
|
"loss": 7.0668,
|
|
"mean_token_accuracy": 0.11169986054301262,
|
|
"num_tokens": 3131990.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.332792663574219,
|
|
"epoch": 0.1546364171437744,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999743852516217,
|
|
"loss": 7.2535,
|
|
"mean_token_accuracy": 0.09437919408082962,
|
|
"num_tokens": 3142567.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.527145719528198,
|
|
"epoch": 0.15517149125153834,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999738051057717,
|
|
"loss": 7.3012,
|
|
"mean_token_accuracy": 0.09176478385925294,
|
|
"num_tokens": 3153083.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.429166221618653,
|
|
"epoch": 0.15570656535930227,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999732184638347,
|
|
"loss": 7.2866,
|
|
"mean_token_accuracy": 0.09161894842982292,
|
|
"num_tokens": 3163985.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.386832141876221,
|
|
"epoch": 0.15624163946706618,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999726253258278,
|
|
"loss": 7.2769,
|
|
"mean_token_accuracy": 0.09116858392953872,
|
|
"num_tokens": 3175259.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.479492425918579,
|
|
"epoch": 0.1567767135748301,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999720256917678,
|
|
"loss": 7.2395,
|
|
"mean_token_accuracy": 0.09686134830117225,
|
|
"num_tokens": 3185487.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.282340335845947,
|
|
"epoch": 0.15731178768259405,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999714195616723,
|
|
"loss": 7.2391,
|
|
"mean_token_accuracy": 0.09826455116271973,
|
|
"num_tokens": 3196200.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.583243417739868,
|
|
"epoch": 0.15784686179035795,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999708069355586,
|
|
"loss": 7.3415,
|
|
"mean_token_accuracy": 0.08557748645544053,
|
|
"num_tokens": 3207675.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.284139347076416,
|
|
"epoch": 0.15838193589812188,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999701878134445,
|
|
"loss": 7.2096,
|
|
"mean_token_accuracy": 0.09773985520005227,
|
|
"num_tokens": 3218117.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.4924530506134035,
|
|
"epoch": 0.15891701000588582,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004999695621953477,
|
|
"loss": 7.2714,
|
|
"mean_token_accuracy": 0.10123906880617142,
|
|
"num_tokens": 3229208.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.381686782836914,
|
|
"epoch": 0.15945208411364975,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999689300812866,
|
|
"loss": 7.1971,
|
|
"mean_token_accuracy": 0.10069452524185181,
|
|
"num_tokens": 3239586.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.443973064422607,
|
|
"epoch": 0.15998715822141366,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004999682914712791,
|
|
"loss": 7.257,
|
|
"mean_token_accuracy": 0.0957643635571003,
|
|
"num_tokens": 3250103.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.41712498664856,
|
|
"epoch": 0.1605222323291776,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004999676463653439,
|
|
"loss": 7.2518,
|
|
"mean_token_accuracy": 0.09973225593566895,
|
|
"num_tokens": 3261844.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.393565559387207,
|
|
"epoch": 0.16105730643694152,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999669947634996,
|
|
"loss": 7.2099,
|
|
"mean_token_accuracy": 0.09588405340909958,
|
|
"num_tokens": 3272073.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.391902828216553,
|
|
"epoch": 0.16159238054470546,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999663366657647,
|
|
"loss": 7.2034,
|
|
"mean_token_accuracy": 0.09728947803378105,
|
|
"num_tokens": 3281725.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.398671960830688,
|
|
"epoch": 0.16212745465246936,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999656720721586,
|
|
"loss": 7.3271,
|
|
"mean_token_accuracy": 0.09106058180332184,
|
|
"num_tokens": 3292708.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.4855554580688475,
|
|
"epoch": 0.1626625287602333,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999650009827004,
|
|
"loss": 7.3079,
|
|
"mean_token_accuracy": 0.094503453373909,
|
|
"num_tokens": 3303548.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.324214076995849,
|
|
"epoch": 0.16319760286799723,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999643233974092,
|
|
"loss": 7.1854,
|
|
"mean_token_accuracy": 0.10153073742985726,
|
|
"num_tokens": 3314261.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.4001932621002195,
|
|
"epoch": 0.16373267697576113,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004999636393163049,
|
|
"loss": 7.1979,
|
|
"mean_token_accuracy": 0.09524251073598862,
|
|
"num_tokens": 3324833.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.291304874420166,
|
|
"epoch": 0.16426775108352507,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999629487394071,
|
|
"loss": 7.1051,
|
|
"mean_token_accuracy": 0.10538214668631554,
|
|
"num_tokens": 3334790.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.413803243637085,
|
|
"epoch": 0.164802825191289,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999622516667358,
|
|
"loss": 7.2313,
|
|
"mean_token_accuracy": 0.09709590896964074,
|
|
"num_tokens": 3345115.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.333492660522461,
|
|
"epoch": 0.16533789929905293,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.000499961548098311,
|
|
"loss": 7.1207,
|
|
"mean_token_accuracy": 0.10122803598642349,
|
|
"num_tokens": 3356187.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.272494459152222,
|
|
"epoch": 0.16587297340681684,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999608380341533,
|
|
"loss": 7.1359,
|
|
"mean_token_accuracy": 0.10306615009903908,
|
|
"num_tokens": 3367218.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.344357633590699,
|
|
"epoch": 0.16640804751458077,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999601214742829,
|
|
"loss": 7.2837,
|
|
"mean_token_accuracy": 0.09136597141623497,
|
|
"num_tokens": 3377557.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.438312625885009,
|
|
"epoch": 0.1669431216223447,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999593984187206,
|
|
"loss": 7.2729,
|
|
"mean_token_accuracy": 0.09725563302636146,
|
|
"num_tokens": 3387402.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.341435861587525,
|
|
"epoch": 0.1674781957301086,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004999586688674872,
|
|
"loss": 7.1538,
|
|
"mean_token_accuracy": 0.10022934228181839,
|
|
"num_tokens": 3397924.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.39786114692688,
|
|
"epoch": 0.16801326983787254,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000499957932820604,
|
|
"loss": 7.2342,
|
|
"mean_token_accuracy": 0.09389038234949112,
|
|
"num_tokens": 3408622.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.419701194763183,
|
|
"epoch": 0.16854834394563648,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499957190278092,
|
|
"loss": 7.1859,
|
|
"mean_token_accuracy": 0.09724239408969879,
|
|
"num_tokens": 3419692.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.312865161895752,
|
|
"epoch": 0.1690834180534004,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999564412399728,
|
|
"loss": 7.2472,
|
|
"mean_token_accuracy": 0.093363007158041,
|
|
"num_tokens": 3430986.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.353525066375733,
|
|
"epoch": 0.1696184921611643,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000499955685706268,
|
|
"loss": 7.1695,
|
|
"mean_token_accuracy": 0.096306062489748,
|
|
"num_tokens": 3442211.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.285671186447144,
|
|
"epoch": 0.17015356626892825,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999549236769993,
|
|
"loss": 7.1595,
|
|
"mean_token_accuracy": 0.10178262889385223,
|
|
"num_tokens": 3453585.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.279556846618652,
|
|
"epoch": 0.17068864037669218,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999541551521888,
|
|
"loss": 7.1921,
|
|
"mean_token_accuracy": 0.09866762459278107,
|
|
"num_tokens": 3464156.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.484783601760864,
|
|
"epoch": 0.17122371448445609,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999533801318587,
|
|
"loss": 7.2645,
|
|
"mean_token_accuracy": 0.08954625204205513,
|
|
"num_tokens": 3474853.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.384049606323242,
|
|
"epoch": 0.17175878859222002,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999525986160313,
|
|
"loss": 7.2346,
|
|
"mean_token_accuracy": 0.1001165673136711,
|
|
"num_tokens": 3486841.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.336293649673462,
|
|
"epoch": 0.17229386269998395,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999518106047293,
|
|
"loss": 7.2373,
|
|
"mean_token_accuracy": 0.09608884081244469,
|
|
"num_tokens": 3498735.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.364135122299194,
|
|
"epoch": 0.17282893680774788,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004999510160979754,
|
|
"loss": 7.2662,
|
|
"mean_token_accuracy": 0.09740655794739723,
|
|
"num_tokens": 3508829.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.306561088562011,
|
|
"epoch": 0.1733640109155118,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999502150957925,
|
|
"loss": 7.189,
|
|
"mean_token_accuracy": 0.09756208583712578,
|
|
"num_tokens": 3519465.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.395194959640503,
|
|
"epoch": 0.17389908502327572,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999494075982037,
|
|
"loss": 7.1279,
|
|
"mean_token_accuracy": 0.10642676651477814,
|
|
"num_tokens": 3530401.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.308092880249023,
|
|
"epoch": 0.17443415913103966,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999485936052324,
|
|
"loss": 7.1994,
|
|
"mean_token_accuracy": 0.09804261550307274,
|
|
"num_tokens": 3541136.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.413141489028931,
|
|
"epoch": 0.17496923323880356,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000499947773116902,
|
|
"loss": 7.1996,
|
|
"mean_token_accuracy": 0.09641246423125267,
|
|
"num_tokens": 3552525.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.236075019836425,
|
|
"epoch": 0.1755043073465675,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999469461332365,
|
|
"loss": 7.1501,
|
|
"mean_token_accuracy": 0.10242327377200126,
|
|
"num_tokens": 3563028.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.417467212677002,
|
|
"epoch": 0.17603938145433143,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999461126542592,
|
|
"loss": 7.1158,
|
|
"mean_token_accuracy": 0.10092604756355286,
|
|
"num_tokens": 3573354.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.150946855545044,
|
|
"epoch": 0.17657445556209536,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004999452726799947,
|
|
"loss": 7.104,
|
|
"mean_token_accuracy": 0.09954961612820626,
|
|
"num_tokens": 3585576.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.43422303199768,
|
|
"epoch": 0.17710952966985927,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999444262104671,
|
|
"loss": 7.2255,
|
|
"mean_token_accuracy": 0.09162149503827095,
|
|
"num_tokens": 3596478.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.3217792987823485,
|
|
"epoch": 0.1776446037776232,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999435732457007,
|
|
"loss": 7.1781,
|
|
"mean_token_accuracy": 0.0928413562476635,
|
|
"num_tokens": 3608014.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.3171303272247314,
|
|
"epoch": 0.17817967788538713,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999427137857203,
|
|
"loss": 7.231,
|
|
"mean_token_accuracy": 0.09090021923184395,
|
|
"num_tokens": 3620120.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.364529132843018,
|
|
"epoch": 0.17871475199315104,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999418478305506,
|
|
"loss": 7.1385,
|
|
"mean_token_accuracy": 0.10106266811490058,
|
|
"num_tokens": 3632578.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.233688068389893,
|
|
"epoch": 0.17924982610091497,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999409753802167,
|
|
"loss": 7.1471,
|
|
"mean_token_accuracy": 0.10476156324148178,
|
|
"num_tokens": 3643245.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.329768943786621,
|
|
"epoch": 0.1797849002086789,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004999400964347437,
|
|
"loss": 7.1406,
|
|
"mean_token_accuracy": 0.09999557435512543,
|
|
"num_tokens": 3655832.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.324776554107666,
|
|
"epoch": 0.18031997431644284,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999392109941571,
|
|
"loss": 7.1195,
|
|
"mean_token_accuracy": 0.10120863169431686,
|
|
"num_tokens": 3666566.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.377109861373901,
|
|
"epoch": 0.18085504842420674,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999383190584822,
|
|
"loss": 7.1558,
|
|
"mean_token_accuracy": 0.09975881800055504,
|
|
"num_tokens": 3677572.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.258830785751343,
|
|
"epoch": 0.18139012253197068,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999374206277451,
|
|
"loss": 7.0797,
|
|
"mean_token_accuracy": 0.10004075467586518,
|
|
"num_tokens": 3687116.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.235633325576782,
|
|
"epoch": 0.1819251966397346,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004999365157019717,
|
|
"loss": 7.1698,
|
|
"mean_token_accuracy": 0.10185194239020348,
|
|
"num_tokens": 3697961.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 7.383031129837036,
|
|
"epoch": 0.18246027074749852,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999356042811878,
|
|
"loss": 7.0962,
|
|
"mean_token_accuracy": 0.10583037212491035,
|
|
"num_tokens": 3708834.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.2048969745635985,
|
|
"epoch": 0.18299534485526245,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.00049993468636542,
|
|
"loss": 7.168,
|
|
"mean_token_accuracy": 0.09880446568131447,
|
|
"num_tokens": 3719835.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.3344207286834715,
|
|
"epoch": 0.18353041896302638,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999337619546948,
|
|
"loss": 7.0614,
|
|
"mean_token_accuracy": 0.10437633171677589,
|
|
"num_tokens": 3730408.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.20396728515625,
|
|
"epoch": 0.18406549307079031,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004999328310490387,
|
|
"loss": 7.1371,
|
|
"mean_token_accuracy": 0.10319086164236069,
|
|
"num_tokens": 3741257.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.309871482849121,
|
|
"epoch": 0.18460056717855422,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999318936484789,
|
|
"loss": 7.1195,
|
|
"mean_token_accuracy": 0.10339285656809807,
|
|
"num_tokens": 3752615.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.274545431137085,
|
|
"epoch": 0.18513564128631815,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004999309497530422,
|
|
"loss": 7.1536,
|
|
"mean_token_accuracy": 0.09540738239884376,
|
|
"num_tokens": 3763656.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.17682466506958,
|
|
"epoch": 0.1856707153940821,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004999299993627558,
|
|
"loss": 7.1274,
|
|
"mean_token_accuracy": 0.10024873688817024,
|
|
"num_tokens": 3775257.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.275924921035767,
|
|
"epoch": 0.186205789501846,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999290424776475,
|
|
"loss": 7.1085,
|
|
"mean_token_accuracy": 0.10480740368366241,
|
|
"num_tokens": 3786094.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.383003568649292,
|
|
"epoch": 0.18674086360960993,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999280790977445,
|
|
"loss": 7.1905,
|
|
"mean_token_accuracy": 0.09309849962592125,
|
|
"num_tokens": 3796672.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.339628458023071,
|
|
"epoch": 0.18727593771737386,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000499927109223075,
|
|
"loss": 7.1565,
|
|
"mean_token_accuracy": 0.10013005062937737,
|
|
"num_tokens": 3807910.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.22684063911438,
|
|
"epoch": 0.1878110118251378,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999261328536667,
|
|
"loss": 7.1184,
|
|
"mean_token_accuracy": 0.10545954629778861,
|
|
"num_tokens": 3820792.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.222686910629273,
|
|
"epoch": 0.1883460859329017,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999251499895479,
|
|
"loss": 7.0348,
|
|
"mean_token_accuracy": 0.10182796642184258,
|
|
"num_tokens": 3831126.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.290640163421631,
|
|
"epoch": 0.18888116004066563,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999241606307471,
|
|
"loss": 7.1585,
|
|
"mean_token_accuracy": 0.0980623982846737,
|
|
"num_tokens": 3841923.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.286150789260864,
|
|
"epoch": 0.18941623414842956,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999231647772927,
|
|
"loss": 7.0679,
|
|
"mean_token_accuracy": 0.10358999967575074,
|
|
"num_tokens": 3852504.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 7.19472336769104,
|
|
"epoch": 0.1899513082561935,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999221624292136,
|
|
"loss": 7.0699,
|
|
"mean_token_accuracy": 0.10635367557406425,
|
|
"num_tokens": 3862761.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.289284896850586,
|
|
"epoch": 0.1904863823639574,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999211535865385,
|
|
"loss": 7.0759,
|
|
"mean_token_accuracy": 0.10067537054419518,
|
|
"num_tokens": 3873662.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.168746614456177,
|
|
"epoch": 0.19102145647172133,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999201382492969,
|
|
"loss": 7.0188,
|
|
"mean_token_accuracy": 0.10261877253651619,
|
|
"num_tokens": 3884835.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 7.28413405418396,
|
|
"epoch": 0.19155653057948527,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999191164175178,
|
|
"loss": 7.1592,
|
|
"mean_token_accuracy": 0.09962226003408432,
|
|
"num_tokens": 3895073.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.337110424041748,
|
|
"epoch": 0.19209160468724917,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999180880912309,
|
|
"loss": 7.1158,
|
|
"mean_token_accuracy": 0.10559302195906639,
|
|
"num_tokens": 3905440.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 7.21136155128479,
|
|
"epoch": 0.1926266787950131,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999170532704657,
|
|
"loss": 7.1183,
|
|
"mean_token_accuracy": 0.09755382165312768,
|
|
"num_tokens": 3917135.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.304487466812134,
|
|
"epoch": 0.19316175290277704,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004999160119552523,
|
|
"loss": 7.1225,
|
|
"mean_token_accuracy": 0.10279463827610016,
|
|
"num_tokens": 3928247.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 7.196545743942261,
|
|
"epoch": 0.19369682701054097,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999149641456206,
|
|
"loss": 7.1084,
|
|
"mean_token_accuracy": 0.10402424335479736,
|
|
"num_tokens": 3938609.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.373271131515503,
|
|
"epoch": 0.19423190111830488,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000499913909841601,
|
|
"loss": 7.0897,
|
|
"mean_token_accuracy": 0.10455763563513756,
|
|
"num_tokens": 3949221.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 7.1703612327575685,
|
|
"epoch": 0.1947669752260688,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999128490432238,
|
|
"loss": 7.0609,
|
|
"mean_token_accuracy": 0.09710687473416328,
|
|
"num_tokens": 3959475.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.2145233154296875,
|
|
"epoch": 0.19530204933383274,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999117817505196,
|
|
"loss": 7.0592,
|
|
"mean_token_accuracy": 0.1138175070285797,
|
|
"num_tokens": 3969638.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.178833293914795,
|
|
"epoch": 0.19583712344159665,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999107079635194,
|
|
"loss": 7.1232,
|
|
"mean_token_accuracy": 0.10179126933217049,
|
|
"num_tokens": 3980543.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.44319224357605,
|
|
"epoch": 0.19637219754936058,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000499909627682254,
|
|
"loss": 7.1773,
|
|
"mean_token_accuracy": 0.08765889480710029,
|
|
"num_tokens": 3993518.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.2637087345123295,
|
|
"epoch": 0.19690727165712452,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999085409067549,
|
|
"loss": 7.1375,
|
|
"mean_token_accuracy": 0.10476782023906708,
|
|
"num_tokens": 4002940.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 7.169889783859253,
|
|
"epoch": 0.19744234576488845,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004999074476370531,
|
|
"loss": 7.0939,
|
|
"mean_token_accuracy": 0.09844619929790496,
|
|
"num_tokens": 4013982.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.24789080619812,
|
|
"epoch": 0.19797741987265235,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999063478731805,
|
|
"loss": 7.0485,
|
|
"mean_token_accuracy": 0.10713553130626678,
|
|
"num_tokens": 4025607.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.222923517227173,
|
|
"epoch": 0.1985124939804163,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004999052416151687,
|
|
"loss": 7.1148,
|
|
"mean_token_accuracy": 0.10136601626873017,
|
|
"num_tokens": 4037467.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 7.270085382461548,
|
|
"epoch": 0.19904756808818022,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004999041288630496,
|
|
"loss": 7.1147,
|
|
"mean_token_accuracy": 0.10222559496760368,
|
|
"num_tokens": 4049146.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.300042200088501,
|
|
"epoch": 0.19958264219594413,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004999030096168553,
|
|
"loss": 7.1579,
|
|
"mean_token_accuracy": 0.09921248778700828,
|
|
"num_tokens": 4060500.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 7.314883184432984,
|
|
"epoch": 0.20011771630370806,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999018838766182,
|
|
"loss": 7.0989,
|
|
"mean_token_accuracy": 0.0998796634376049,
|
|
"num_tokens": 4070553.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 7.253989124298096,
|
|
"epoch": 0.200652790411472,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000499900751642371,
|
|
"loss": 7.0327,
|
|
"mean_token_accuracy": 0.11456628292798995,
|
|
"num_tokens": 4079653.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.165084648132324,
|
|
"epoch": 0.20118786451923593,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004998996129141461,
|
|
"loss": 7.1197,
|
|
"mean_token_accuracy": 0.09987151175737381,
|
|
"num_tokens": 4089912.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.321857929229736,
|
|
"epoch": 0.20172293862699983,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004998984676919764,
|
|
"loss": 7.0693,
|
|
"mean_token_accuracy": 0.09889922812581062,
|
|
"num_tokens": 4100577.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 7.251675939559936,
|
|
"epoch": 0.20225801273476376,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004998973159758952,
|
|
"loss": 7.116,
|
|
"mean_token_accuracy": 0.1035578764975071,
|
|
"num_tokens": 4111752.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 7.180023431777954,
|
|
"epoch": 0.2027930868425277,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998961577659355,
|
|
"loss": 7.0633,
|
|
"mean_token_accuracy": 0.10196540728211403,
|
|
"num_tokens": 4122616.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 7.318381118774414,
|
|
"epoch": 0.2033281609502916,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004998949930621309,
|
|
"loss": 7.1603,
|
|
"mean_token_accuracy": 0.10213067010045052,
|
|
"num_tokens": 4134275.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 7.190880680084229,
|
|
"epoch": 0.20386323505805554,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998938218645149,
|
|
"loss": 7.0717,
|
|
"mean_token_accuracy": 0.10233089625835419,
|
|
"num_tokens": 4146412.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 7.185058450698852,
|
|
"epoch": 0.20439830916581947,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998926441731215,
|
|
"loss": 7.0594,
|
|
"mean_token_accuracy": 0.10094341784715652,
|
|
"num_tokens": 4157592.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 7.2327782154083256,
|
|
"epoch": 0.2049333832735834,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998914599879845,
|
|
"loss": 7.0388,
|
|
"mean_token_accuracy": 0.10173192545771599,
|
|
"num_tokens": 4168637.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 7.216303873062134,
|
|
"epoch": 0.2054684573813473,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004998902693091382,
|
|
"loss": 7.073,
|
|
"mean_token_accuracy": 0.10192683935165406,
|
|
"num_tokens": 4179951.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 7.214565849304199,
|
|
"epoch": 0.20600353148911124,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004998890721366171,
|
|
"loss": 6.9886,
|
|
"mean_token_accuracy": 0.1056469924747944,
|
|
"num_tokens": 4189895.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 7.218944406509399,
|
|
"epoch": 0.20653860559687517,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998878684704556,
|
|
"loss": 7.0203,
|
|
"mean_token_accuracy": 0.10886224210262299,
|
|
"num_tokens": 4199796.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 7.145756816864013,
|
|
"epoch": 0.20707367970463908,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998866583106884,
|
|
"loss": 7.0497,
|
|
"mean_token_accuracy": 0.1005377635359764,
|
|
"num_tokens": 4210428.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 7.31018853187561,
|
|
"epoch": 0.207608753812403,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998854416573506,
|
|
"loss": 7.1162,
|
|
"mean_token_accuracy": 0.10497400760650635,
|
|
"num_tokens": 4221515.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 7.217876052856445,
|
|
"epoch": 0.20814382792016695,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998842185104772,
|
|
"loss": 7.1294,
|
|
"mean_token_accuracy": 0.10152154713869095,
|
|
"num_tokens": 4232580.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 7.19381012916565,
|
|
"epoch": 0.20867890202793088,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998829888701036,
|
|
"loss": 7.0433,
|
|
"mean_token_accuracy": 0.10803714469075203,
|
|
"num_tokens": 4242871.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 7.283778476715088,
|
|
"epoch": 0.20921397613569478,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998817527362654,
|
|
"loss": 7.0692,
|
|
"mean_token_accuracy": 0.09783539399504662,
|
|
"num_tokens": 4253419.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 7.182916593551636,
|
|
"epoch": 0.20974905024345872,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004998805101089982,
|
|
"loss": 7.0529,
|
|
"mean_token_accuracy": 0.10560425892472267,
|
|
"num_tokens": 4264327.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 7.240354061126709,
|
|
"epoch": 0.21028412435122265,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004998792609883378,
|
|
"loss": 7.0356,
|
|
"mean_token_accuracy": 0.10485936850309371,
|
|
"num_tokens": 4274568.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 7.218569707870484,
|
|
"epoch": 0.21081919845898656,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998780053743203,
|
|
"loss": 7.0283,
|
|
"mean_token_accuracy": 0.10580286756157875,
|
|
"num_tokens": 4284895.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 7.211511325836182,
|
|
"epoch": 0.2113542725667505,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004998767432669822,
|
|
"loss": 7.0579,
|
|
"mean_token_accuracy": 0.10149188414216041,
|
|
"num_tokens": 4297258.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 7.180872201919556,
|
|
"epoch": 0.21188934667451442,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004998754746663595,
|
|
"loss": 6.9933,
|
|
"mean_token_accuracy": 0.1065586656332016,
|
|
"num_tokens": 4308307.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 7.1800261497497555,
|
|
"epoch": 0.21242442078227836,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004998741995724892,
|
|
"loss": 7.0546,
|
|
"mean_token_accuracy": 0.10103240609169006,
|
|
"num_tokens": 4318801.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 7.196444845199585,
|
|
"epoch": 0.21295949489004226,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004998729179854079,
|
|
"loss": 7.0325,
|
|
"mean_token_accuracy": 0.10279674157500267,
|
|
"num_tokens": 4329531.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 7.1866326332092285,
|
|
"epoch": 0.2134945689978062,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004998716299051527,
|
|
"loss": 7.0133,
|
|
"mean_token_accuracy": 0.10350802689790725,
|
|
"num_tokens": 4341203.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 7.124855422973633,
|
|
"epoch": 0.21402964310557013,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004998703353317609,
|
|
"loss": 7.0584,
|
|
"mean_token_accuracy": 0.10162648260593414,
|
|
"num_tokens": 4352426.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 7.259735202789306,
|
|
"epoch": 0.21456471721333403,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004998690342652697,
|
|
"loss": 7.0121,
|
|
"mean_token_accuracy": 0.10889948382973672,
|
|
"num_tokens": 4363413.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 7.1536116123199465,
|
|
"epoch": 0.21509979132109797,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004998677267057166,
|
|
"loss": 6.9458,
|
|
"mean_token_accuracy": 0.10910924375057221,
|
|
"num_tokens": 4373470.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 7.114294338226318,
|
|
"epoch": 0.2156348654288619,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004998664126531397,
|
|
"loss": 7.0155,
|
|
"mean_token_accuracy": 0.10767255127429962,
|
|
"num_tokens": 4384438.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 7.260488796234131,
|
|
"epoch": 0.21616993953662583,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004998650921075766,
|
|
"loss": 6.9928,
|
|
"mean_token_accuracy": 0.10748248249292373,
|
|
"num_tokens": 4395100.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 7.089138031005859,
|
|
"epoch": 0.21670501364438974,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004998637650690656,
|
|
"loss": 7.1018,
|
|
"mean_token_accuracy": 0.09796615988016129,
|
|
"num_tokens": 4407131.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 7.218321037292481,
|
|
"epoch": 0.21724008775215367,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000499862431537645,
|
|
"loss": 6.9445,
|
|
"mean_token_accuracy": 0.10602100789546967,
|
|
"num_tokens": 4417701.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 7.156389379501343,
|
|
"epoch": 0.2177751618599176,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998610915133533,
|
|
"loss": 7.0019,
|
|
"mean_token_accuracy": 0.10421578660607338,
|
|
"num_tokens": 4429309.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 7.151696300506591,
|
|
"epoch": 0.21831023596768154,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004998597449962292,
|
|
"loss": 7.0604,
|
|
"mean_token_accuracy": 0.1074062243103981,
|
|
"num_tokens": 4440850.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 7.173864507675171,
|
|
"epoch": 0.21884531007544544,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004998583919863115,
|
|
"loss": 7.0455,
|
|
"mean_token_accuracy": 0.10184828117489815,
|
|
"num_tokens": 4451995.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 7.095217037200928,
|
|
"epoch": 0.21938038418320938,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998570324836393,
|
|
"loss": 6.9646,
|
|
"mean_token_accuracy": 0.10822436287999153,
|
|
"num_tokens": 4461916.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 7.188471364974975,
|
|
"epoch": 0.2199154582909733,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000499855666488252,
|
|
"loss": 7.0417,
|
|
"mean_token_accuracy": 0.10385469049215316,
|
|
"num_tokens": 4471603.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 7.260391330718994,
|
|
"epoch": 0.22045053239873721,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004998542940001889,
|
|
"loss": 7.0733,
|
|
"mean_token_accuracy": 0.10464386790990829,
|
|
"num_tokens": 4483166.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 7.165126323699951,
|
|
"epoch": 0.22098560650650115,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004998529150194895,
|
|
"loss": 6.984,
|
|
"mean_token_accuracy": 0.10515503585338593,
|
|
"num_tokens": 4493824.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 7.1778943061828615,
|
|
"epoch": 0.22152068061426508,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004998515295461939,
|
|
"loss": 6.8879,
|
|
"mean_token_accuracy": 0.12077882513403893,
|
|
"num_tokens": 4505674.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 7.045021867752075,
|
|
"epoch": 0.22205575472202901,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000499850137580342,
|
|
"loss": 6.9944,
|
|
"mean_token_accuracy": 0.11151223108172417,
|
|
"num_tokens": 4516516.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 7.183830261230469,
|
|
"epoch": 0.22259082882979292,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004998487391219739,
|
|
"loss": 6.9922,
|
|
"mean_token_accuracy": 0.10238008573651314,
|
|
"num_tokens": 4527811.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 7.124439287185669,
|
|
"epoch": 0.22312590293755685,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004998473341711301,
|
|
"loss": 6.9376,
|
|
"mean_token_accuracy": 0.10194609165191651,
|
|
"num_tokens": 4538988.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 7.129032325744629,
|
|
"epoch": 0.22366097704532079,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998459227278512,
|
|
"loss": 6.964,
|
|
"mean_token_accuracy": 0.10870476886630058,
|
|
"num_tokens": 4550049.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 7.269771909713745,
|
|
"epoch": 0.2241960511530847,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004998445047921777,
|
|
"loss": 7.0811,
|
|
"mean_token_accuracy": 0.1076191745698452,
|
|
"num_tokens": 4559784.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 7.1074059009552,
|
|
"epoch": 0.22473112526084862,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998430803641509,
|
|
"loss": 6.9183,
|
|
"mean_token_accuracy": 0.10604787841439248,
|
|
"num_tokens": 4569982.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 7.101476860046387,
|
|
"epoch": 0.22526619936861256,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004998416494438118,
|
|
"loss": 6.9299,
|
|
"mean_token_accuracy": 0.10904891490936279,
|
|
"num_tokens": 4581117.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 7.105316066741944,
|
|
"epoch": 0.2258012734763765,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998402120312014,
|
|
"loss": 6.9953,
|
|
"mean_token_accuracy": 0.11073005571961403,
|
|
"num_tokens": 4592012.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 7.180981636047363,
|
|
"epoch": 0.2263363475841404,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004998387681263616,
|
|
"loss": 7.0385,
|
|
"mean_token_accuracy": 0.10344724059104919,
|
|
"num_tokens": 4602591.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 7.147944641113281,
|
|
"epoch": 0.22687142169190433,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000499837317729334,
|
|
"loss": 6.9648,
|
|
"mean_token_accuracy": 0.10446924120187759,
|
|
"num_tokens": 4613418.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 7.088868284225464,
|
|
"epoch": 0.22740649579966826,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004998358608401603,
|
|
"loss": 6.8946,
|
|
"mean_token_accuracy": 0.11337370723485947,
|
|
"num_tokens": 4625156.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 7.165902519226075,
|
|
"epoch": 0.22794156990743217,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000499834397458883,
|
|
"loss": 6.9344,
|
|
"mean_token_accuracy": 0.11145322620868683,
|
|
"num_tokens": 4634523.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 7.076514863967896,
|
|
"epoch": 0.2284766440151961,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004998329275855438,
|
|
"loss": 6.9174,
|
|
"mean_token_accuracy": 0.11302619948983192,
|
|
"num_tokens": 4646292.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 7.122889423370362,
|
|
"epoch": 0.22901171812296003,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004998314512201855,
|
|
"loss": 6.9004,
|
|
"mean_token_accuracy": 0.10689050704240799,
|
|
"num_tokens": 4656417.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 7.06037015914917,
|
|
"epoch": 0.22954679223072397,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004998299683628507,
|
|
"loss": 6.912,
|
|
"mean_token_accuracy": 0.11160982474684715,
|
|
"num_tokens": 4666933.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 7.187684774398804,
|
|
"epoch": 0.23008186633848787,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004998284790135821,
|
|
"loss": 7.1033,
|
|
"mean_token_accuracy": 0.1016832061111927,
|
|
"num_tokens": 4678482.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 7.139988660812378,
|
|
"epoch": 0.2306169404462518,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998269831724228,
|
|
"loss": 6.9424,
|
|
"mean_token_accuracy": 0.10893533229827881,
|
|
"num_tokens": 4688454.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 7.114541387557983,
|
|
"epoch": 0.23115201455401574,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998254808394159,
|
|
"loss": 6.9693,
|
|
"mean_token_accuracy": 0.1090779609978199,
|
|
"num_tokens": 4699545.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 7.173770380020142,
|
|
"epoch": 0.23168708866177964,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004998239720146048,
|
|
"loss": 6.9818,
|
|
"mean_token_accuracy": 0.10480888038873673,
|
|
"num_tokens": 4711149.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 7.0449871063232425,
|
|
"epoch": 0.23222216276954358,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004998224566980332,
|
|
"loss": 6.9846,
|
|
"mean_token_accuracy": 0.10382255315780639,
|
|
"num_tokens": 4722651.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 7.14964804649353,
|
|
"epoch": 0.2327572368773075,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998209348897447,
|
|
"loss": 6.9818,
|
|
"mean_token_accuracy": 0.10008153393864631,
|
|
"num_tokens": 4733883.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 7.267318153381348,
|
|
"epoch": 0.23329231098507144,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998194065897833,
|
|
"loss": 6.9546,
|
|
"mean_token_accuracy": 0.10759458690881729,
|
|
"num_tokens": 4744517.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 7.039386034011841,
|
|
"epoch": 0.23382738509283535,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000499817871798193,
|
|
"loss": 6.9613,
|
|
"mean_token_accuracy": 0.11343310549855232,
|
|
"num_tokens": 4755481.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 7.088238620758057,
|
|
"epoch": 0.23436245920059928,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998163305150185,
|
|
"loss": 6.8813,
|
|
"mean_token_accuracy": 0.1131195604801178,
|
|
"num_tokens": 4765535.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 7.1311625957489015,
|
|
"epoch": 0.23489753330836322,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998147827403038,
|
|
"loss": 7.0102,
|
|
"mean_token_accuracy": 0.1072376735508442,
|
|
"num_tokens": 4777045.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 7.156279706954956,
|
|
"epoch": 0.23543260741612712,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000499813228474094,
|
|
"loss": 6.8885,
|
|
"mean_token_accuracy": 0.11066517904400826,
|
|
"num_tokens": 4787958.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 7.07044153213501,
|
|
"epoch": 0.23596768152389105,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004998116677164338,
|
|
"loss": 6.9279,
|
|
"mean_token_accuracy": 0.10897202789783478,
|
|
"num_tokens": 4798502.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 7.007950210571289,
|
|
"epoch": 0.236502755631655,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998101004673682,
|
|
"loss": 6.9386,
|
|
"mean_token_accuracy": 0.11299166306853295,
|
|
"num_tokens": 4809179.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 7.225791692733765,
|
|
"epoch": 0.23703782973941892,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004998085267269426,
|
|
"loss": 7.0021,
|
|
"mean_token_accuracy": 0.1013013631105423,
|
|
"num_tokens": 4819803.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 7.156044769287109,
|
|
"epoch": 0.23757290384718283,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004998069464952025,
|
|
"loss": 6.9661,
|
|
"mean_token_accuracy": 0.10483337342739105,
|
|
"num_tokens": 4831226.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 7.1146551132202145,
|
|
"epoch": 0.23810797795494676,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004998053597721932,
|
|
"loss": 6.9805,
|
|
"mean_token_accuracy": 0.10689718723297119,
|
|
"num_tokens": 4842168.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 7.112945604324341,
|
|
"epoch": 0.2386430520627107,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004998037665579608,
|
|
"loss": 6.9683,
|
|
"mean_token_accuracy": 0.10687630698084831,
|
|
"num_tokens": 4852655.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 7.031319570541382,
|
|
"epoch": 0.2391781261704746,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004998021668525513,
|
|
"loss": 6.8648,
|
|
"mean_token_accuracy": 0.11213752254843712,
|
|
"num_tokens": 4862164.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 7.138062334060669,
|
|
"epoch": 0.23971320027823853,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004998005606560107,
|
|
"loss": 6.9837,
|
|
"mean_token_accuracy": 0.10962075069546699,
|
|
"num_tokens": 4874233.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 7.145175266265869,
|
|
"epoch": 0.24024827438600246,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004997989479683856,
|
|
"loss": 6.9847,
|
|
"mean_token_accuracy": 0.1075660951435566,
|
|
"num_tokens": 4885948.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 7.035645341873169,
|
|
"epoch": 0.2407833484937664,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004997973287897224,
|
|
"loss": 6.9566,
|
|
"mean_token_accuracy": 0.10865945667028427,
|
|
"num_tokens": 4897128.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 7.135949563980103,
|
|
"epoch": 0.2413184226015303,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004997957031200681,
|
|
"loss": 6.9369,
|
|
"mean_token_accuracy": 0.11560432091355324,
|
|
"num_tokens": 4907699.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 7.00885157585144,
|
|
"epoch": 0.24185349670929424,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004997940709594692,
|
|
"loss": 6.8547,
|
|
"mean_token_accuracy": 0.11122522950172424,
|
|
"num_tokens": 4918772.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 7.157874059677124,
|
|
"epoch": 0.24238857081705817,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997924323079733,
|
|
"loss": 6.9092,
|
|
"mean_token_accuracy": 0.10879954397678375,
|
|
"num_tokens": 4930046.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 7.033572673797607,
|
|
"epoch": 0.24292364492482207,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004997907871656273,
|
|
"loss": 6.9361,
|
|
"mean_token_accuracy": 0.11303869113326073,
|
|
"num_tokens": 4940609.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 7.085981941223144,
|
|
"epoch": 0.243458719032586,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997891355324791,
|
|
"loss": 6.9537,
|
|
"mean_token_accuracy": 0.11050518602132797,
|
|
"num_tokens": 4951339.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 7.1102917194366455,
|
|
"epoch": 0.24399379314034994,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004997874774085761,
|
|
"loss": 6.8694,
|
|
"mean_token_accuracy": 0.10475466623902321,
|
|
"num_tokens": 4961083.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 7.02321424484253,
|
|
"epoch": 0.24452886724811387,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997858127939662,
|
|
"loss": 6.8815,
|
|
"mean_token_accuracy": 0.11015179753303528,
|
|
"num_tokens": 4971399.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 7.133182764053345,
|
|
"epoch": 0.24506394135587778,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004997841416886976,
|
|
"loss": 6.9557,
|
|
"mean_token_accuracy": 0.10723848342895508,
|
|
"num_tokens": 4982429.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 7.134330749511719,
|
|
"epoch": 0.2455990154636417,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004997824640928186,
|
|
"loss": 6.9981,
|
|
"mean_token_accuracy": 0.10663792341947556,
|
|
"num_tokens": 4993706.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 7.106071710586548,
|
|
"epoch": 0.24613408957140565,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004997807800063774,
|
|
"loss": 6.8969,
|
|
"mean_token_accuracy": 0.11038667410612106,
|
|
"num_tokens": 5005411.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 7.113573980331421,
|
|
"epoch": 0.24666916367916958,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000499779089429423,
|
|
"loss": 7.0192,
|
|
"mean_token_accuracy": 0.10511812567710876,
|
|
"num_tokens": 5016011.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 7.082342147827148,
|
|
"epoch": 0.24720423778693348,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004997773923620037,
|
|
"loss": 6.9766,
|
|
"mean_token_accuracy": 0.1077098935842514,
|
|
"num_tokens": 5027841.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 7.156000328063965,
|
|
"epoch": 0.24773931189469742,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004997756888041689,
|
|
"loss": 6.9674,
|
|
"mean_token_accuracy": 0.09967386648058892,
|
|
"num_tokens": 5038640.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 7.031879425048828,
|
|
"epoch": 0.24827438600246135,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004997739787559677,
|
|
"loss": 6.9064,
|
|
"mean_token_accuracy": 0.1101887047290802,
|
|
"num_tokens": 5050423.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 7.084950113296509,
|
|
"epoch": 0.24880946011022526,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004997722622174494,
|
|
"loss": 6.9261,
|
|
"mean_token_accuracy": 0.10869491174817085,
|
|
"num_tokens": 5060654.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 7.073810243606568,
|
|
"epoch": 0.2493445342179892,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004997705391886636,
|
|
"loss": 6.8532,
|
|
"mean_token_accuracy": 0.1153720997273922,
|
|
"num_tokens": 5070165.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.990339374542236,
|
|
"epoch": 0.24987960832575312,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004997688096696601,
|
|
"loss": 6.8889,
|
|
"mean_token_accuracy": 0.10511518865823746,
|
|
"num_tokens": 5080937.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 7.040760135650634,
|
|
"epoch": 0.25041468243351706,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004997670736604888,
|
|
"loss": 6.8536,
|
|
"mean_token_accuracy": 0.11363203302025796,
|
|
"num_tokens": 5090955.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 7.100010633468628,
|
|
"epoch": 0.250949756541281,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004997653311611998,
|
|
"loss": 6.9896,
|
|
"mean_token_accuracy": 0.1013368934392929,
|
|
"num_tokens": 5102201.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 7.11283597946167,
|
|
"epoch": 0.25148483064904487,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004997635821718435,
|
|
"loss": 6.912,
|
|
"mean_token_accuracy": 0.11232623681426049,
|
|
"num_tokens": 5112235.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 7.029507255554199,
|
|
"epoch": 0.2520199047568088,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997618266924704,
|
|
"loss": 6.7832,
|
|
"mean_token_accuracy": 0.10930138230323791,
|
|
"num_tokens": 5122360.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 7.053276777267456,
|
|
"epoch": 0.25255497886457273,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000499760064723131,
|
|
"loss": 6.9599,
|
|
"mean_token_accuracy": 0.10386712178587913,
|
|
"num_tokens": 5133084.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 7.001993083953858,
|
|
"epoch": 0.25309005297233667,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004997582962638764,
|
|
"loss": 6.9092,
|
|
"mean_token_accuracy": 0.10329946726560593,
|
|
"num_tokens": 5144495.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 7.14462685585022,
|
|
"epoch": 0.2536251270801006,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004997565213147575,
|
|
"loss": 6.9045,
|
|
"mean_token_accuracy": 0.11057875901460648,
|
|
"num_tokens": 5154824.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 7.04837121963501,
|
|
"epoch": 0.25416020118786453,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997547398758257,
|
|
"loss": 6.8976,
|
|
"mean_token_accuracy": 0.10639405995607376,
|
|
"num_tokens": 5165061.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 7.022197437286377,
|
|
"epoch": 0.25469527529562846,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004997529519471323,
|
|
"loss": 6.8471,
|
|
"mean_token_accuracy": 0.11550139263272285,
|
|
"num_tokens": 5174964.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 7.051410436630249,
|
|
"epoch": 0.25523034940339234,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997511575287291,
|
|
"loss": 6.8892,
|
|
"mean_token_accuracy": 0.11107081025838852,
|
|
"num_tokens": 5185458.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 7.052774667739868,
|
|
"epoch": 0.2557654235111563,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997493566206677,
|
|
"loss": 6.878,
|
|
"mean_token_accuracy": 0.11274964585900307,
|
|
"num_tokens": 5196193.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 7.087007999420166,
|
|
"epoch": 0.2563004976189202,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004997475492230003,
|
|
"loss": 7.0016,
|
|
"mean_token_accuracy": 0.10122420638799667,
|
|
"num_tokens": 5208541.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 7.037209749221802,
|
|
"epoch": 0.25683557172668414,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000499745735335779,
|
|
"loss": 6.8283,
|
|
"mean_token_accuracy": 0.11095813065767288,
|
|
"num_tokens": 5218567.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 7.078734445571899,
|
|
"epoch": 0.2573706458344481,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004997439149590561,
|
|
"loss": 6.9304,
|
|
"mean_token_accuracy": 0.10828692242503166,
|
|
"num_tokens": 5228293.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 7.043195676803589,
|
|
"epoch": 0.257905719942212,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997420880928843,
|
|
"loss": 6.8715,
|
|
"mean_token_accuracy": 0.11050004214048385,
|
|
"num_tokens": 5238880.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 7.082090950012207,
|
|
"epoch": 0.25844079404997594,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004997402547373162,
|
|
"loss": 6.9731,
|
|
"mean_token_accuracy": 0.10441731810569763,
|
|
"num_tokens": 5250104.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 7.139094257354737,
|
|
"epoch": 0.2589758681577399,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004997384148924049,
|
|
"loss": 6.9951,
|
|
"mean_token_accuracy": 0.10500225946307182,
|
|
"num_tokens": 5260355.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 7.020048427581787,
|
|
"epoch": 0.25951094226550375,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004997365685582035,
|
|
"loss": 6.9181,
|
|
"mean_token_accuracy": 0.10640754848718643,
|
|
"num_tokens": 5272117.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 7.079926252365112,
|
|
"epoch": 0.2600460163732677,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004997347157347651,
|
|
"loss": 6.9946,
|
|
"mean_token_accuracy": 0.10754604786634445,
|
|
"num_tokens": 5284583.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 7.123588037490845,
|
|
"epoch": 0.2605810904810316,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004997328564221435,
|
|
"loss": 6.7935,
|
|
"mean_token_accuracy": 0.12066915556788445,
|
|
"num_tokens": 5295846.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.995146322250366,
|
|
"epoch": 0.26111616458879555,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004997309906203922,
|
|
"loss": 6.9324,
|
|
"mean_token_accuracy": 0.10952212065458297,
|
|
"num_tokens": 5307802.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 7.15656852722168,
|
|
"epoch": 0.2616512386965595,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004997291183295652,
|
|
"loss": 6.8757,
|
|
"mean_token_accuracy": 0.10969269722700119,
|
|
"num_tokens": 5318263.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 7.028390216827392,
|
|
"epoch": 0.2621863128043234,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004997272395497164,
|
|
"loss": 6.9303,
|
|
"mean_token_accuracy": 0.10968295708298684,
|
|
"num_tokens": 5328555.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.9868518829345705,
|
|
"epoch": 0.26272138691208735,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004997253542809002,
|
|
"loss": 6.8927,
|
|
"mean_token_accuracy": 0.10809791460633278,
|
|
"num_tokens": 5339954.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 7.193512296676635,
|
|
"epoch": 0.26325646101985123,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004997234625231709,
|
|
"loss": 6.9324,
|
|
"mean_token_accuracy": 0.10953526422381402,
|
|
"num_tokens": 5350919.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.973828077316284,
|
|
"epoch": 0.26379153512761516,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004997215642765832,
|
|
"loss": 6.895,
|
|
"mean_token_accuracy": 0.11250809505581856,
|
|
"num_tokens": 5361844.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 7.0807849884033205,
|
|
"epoch": 0.2643266092353791,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000499719659541192,
|
|
"loss": 6.9167,
|
|
"mean_token_accuracy": 0.10835950300097466,
|
|
"num_tokens": 5372026.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 7.093844032287597,
|
|
"epoch": 0.26486168334314303,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004997177483170521,
|
|
"loss": 6.9599,
|
|
"mean_token_accuracy": 0.11532194167375565,
|
|
"num_tokens": 5382965.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 7.004912328720093,
|
|
"epoch": 0.26539675745090696,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004997158306042188,
|
|
"loss": 6.9202,
|
|
"mean_token_accuracy": 0.11223710849881172,
|
|
"num_tokens": 5393738.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 7.040785694122315,
|
|
"epoch": 0.2659318315586709,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004997139064027476,
|
|
"loss": 6.7822,
|
|
"mean_token_accuracy": 0.1194936640560627,
|
|
"num_tokens": 5404059.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.9767382621765135,
|
|
"epoch": 0.26646690566643483,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004997119757126938,
|
|
"loss": 6.8286,
|
|
"mean_token_accuracy": 0.11358326822519302,
|
|
"num_tokens": 5414152.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 7.081306219100952,
|
|
"epoch": 0.2670019797741987,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004997100385341133,
|
|
"loss": 7.0047,
|
|
"mean_token_accuracy": 0.11303253397345543,
|
|
"num_tokens": 5425227.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 7.00633659362793,
|
|
"epoch": 0.26753705388196264,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997080948670621,
|
|
"loss": 6.8416,
|
|
"mean_token_accuracy": 0.11532852947711944,
|
|
"num_tokens": 5435716.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 7.059450101852417,
|
|
"epoch": 0.26807212798972657,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004997061447115962,
|
|
"loss": 6.8843,
|
|
"mean_token_accuracy": 0.10835974663496017,
|
|
"num_tokens": 5446519.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 7.0279497623443605,
|
|
"epoch": 0.2686072020974905,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004997041880677719,
|
|
"loss": 6.888,
|
|
"mean_token_accuracy": 0.11044454723596572,
|
|
"num_tokens": 5457496.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.935314702987671,
|
|
"epoch": 0.26914227620525444,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004997022249356457,
|
|
"loss": 6.7806,
|
|
"mean_token_accuracy": 0.11524100303649902,
|
|
"num_tokens": 5469059.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 7.044953393936157,
|
|
"epoch": 0.26967735031301837,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997002553152745,
|
|
"loss": 6.8806,
|
|
"mean_token_accuracy": 0.11304009705781937,
|
|
"num_tokens": 5480324.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 7.024924993515015,
|
|
"epoch": 0.2702124244207823,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004996982792067148,
|
|
"loss": 6.889,
|
|
"mean_token_accuracy": 0.10333864018321037,
|
|
"num_tokens": 5491513.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.982390928268432,
|
|
"epoch": 0.2707474985285462,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000499696296610024,
|
|
"loss": 6.8084,
|
|
"mean_token_accuracy": 0.1128513678908348,
|
|
"num_tokens": 5502414.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.966418361663818,
|
|
"epoch": 0.2712825726363101,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004996943075252592,
|
|
"loss": 6.944,
|
|
"mean_token_accuracy": 0.11141985207796097,
|
|
"num_tokens": 5512962.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 7.015233278274536,
|
|
"epoch": 0.27181764674407405,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004996923119524777,
|
|
"loss": 6.8026,
|
|
"mean_token_accuracy": 0.11136919856071473,
|
|
"num_tokens": 5524648.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.970895910263062,
|
|
"epoch": 0.272352720851838,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004996903098917375,
|
|
"loss": 6.8846,
|
|
"mean_token_accuracy": 0.10550413206219673,
|
|
"num_tokens": 5536158.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.9682692527771,
|
|
"epoch": 0.2728877949596019,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000499688301343096,
|
|
"loss": 6.7946,
|
|
"mean_token_accuracy": 0.11154654324054718,
|
|
"num_tokens": 5547374.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 7.100108480453491,
|
|
"epoch": 0.27342286906736585,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004996862863066114,
|
|
"loss": 6.8914,
|
|
"mean_token_accuracy": 0.11287438571453094,
|
|
"num_tokens": 5559212.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.999144172668457,
|
|
"epoch": 0.2739579431751298,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996842647823419,
|
|
"loss": 6.8469,
|
|
"mean_token_accuracy": 0.11468030512332916,
|
|
"num_tokens": 5569302.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.963889455795288,
|
|
"epoch": 0.27449301728289366,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004996822367703458,
|
|
"loss": 6.8464,
|
|
"mean_token_accuracy": 0.10833697766065598,
|
|
"num_tokens": 5581688.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 7.065867328643799,
|
|
"epoch": 0.2750280913906576,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004996802022706817,
|
|
"loss": 6.7811,
|
|
"mean_token_accuracy": 0.11610291451215744,
|
|
"num_tokens": 5591376.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.911547708511352,
|
|
"epoch": 0.2755631654984215,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004996781612834083,
|
|
"loss": 6.8206,
|
|
"mean_token_accuracy": 0.11811210587620735,
|
|
"num_tokens": 5601510.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 7.062700414657593,
|
|
"epoch": 0.27609823960618546,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004996761138085846,
|
|
"loss": 6.9144,
|
|
"mean_token_accuracy": 0.10678515806794167,
|
|
"num_tokens": 5611436.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 7.025034236907959,
|
|
"epoch": 0.2766333137139494,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996740598462697,
|
|
"loss": 6.9045,
|
|
"mean_token_accuracy": 0.10823121890425683,
|
|
"num_tokens": 5620470.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.927712631225586,
|
|
"epoch": 0.2771683878217133,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004996719993965229,
|
|
"loss": 6.803,
|
|
"mean_token_accuracy": 0.1145929142832756,
|
|
"num_tokens": 5630740.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.955496549606323,
|
|
"epoch": 0.27770346192947726,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004996699324594036,
|
|
"loss": 6.8039,
|
|
"mean_token_accuracy": 0.1161543108522892,
|
|
"num_tokens": 5641138.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 7.075690460205078,
|
|
"epoch": 0.27823853603724114,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004996678590349717,
|
|
"loss": 6.9325,
|
|
"mean_token_accuracy": 0.11235370934009552,
|
|
"num_tokens": 5652474.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.996727037429809,
|
|
"epoch": 0.27877361014500507,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004996657791232868,
|
|
"loss": 6.8411,
|
|
"mean_token_accuracy": 0.11515266224741935,
|
|
"num_tokens": 5663178.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 7.003625869750977,
|
|
"epoch": 0.279308684252769,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996636927244092,
|
|
"loss": 6.8484,
|
|
"mean_token_accuracy": 0.10914000719785691,
|
|
"num_tokens": 5673565.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 7.075772523880005,
|
|
"epoch": 0.27984375836053293,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000499661599838399,
|
|
"loss": 6.9003,
|
|
"mean_token_accuracy": 0.10649769231677056,
|
|
"num_tokens": 5684651.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.911233377456665,
|
|
"epoch": 0.28037883246829687,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004996595004653167,
|
|
"loss": 6.7904,
|
|
"mean_token_accuracy": 0.11583374589681625,
|
|
"num_tokens": 5696081.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.993151569366455,
|
|
"epoch": 0.2809139065760608,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004996573946052229,
|
|
"loss": 6.8294,
|
|
"mean_token_accuracy": 0.1125739298760891,
|
|
"num_tokens": 5706830.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.982648468017578,
|
|
"epoch": 0.28144898068382473,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004996552822581783,
|
|
"loss": 6.8391,
|
|
"mean_token_accuracy": 0.1167409248650074,
|
|
"num_tokens": 5717261.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 7.002830171585083,
|
|
"epoch": 0.2819840547915886,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499653163424244,
|
|
"loss": 6.8305,
|
|
"mean_token_accuracy": 0.11221889033913612,
|
|
"num_tokens": 5727280.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.9843268394470215,
|
|
"epoch": 0.28251912889935255,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004996510381034814,
|
|
"loss": 6.8746,
|
|
"mean_token_accuracy": 0.10521704554557801,
|
|
"num_tokens": 5737485.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 7.011574792861938,
|
|
"epoch": 0.2830542030071165,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004996489062959515,
|
|
"loss": 6.8336,
|
|
"mean_token_accuracy": 0.11000596359372139,
|
|
"num_tokens": 5747683.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.939910221099853,
|
|
"epoch": 0.2835892771148804,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004996467680017159,
|
|
"loss": 6.9016,
|
|
"mean_token_accuracy": 0.10526101067662239,
|
|
"num_tokens": 5758148.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.925260782241821,
|
|
"epoch": 0.28412435122264434,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004996446232208365,
|
|
"loss": 6.8311,
|
|
"mean_token_accuracy": 0.11652650088071823,
|
|
"num_tokens": 5769109.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 7.011774253845215,
|
|
"epoch": 0.2846594253304083,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004996424719533753,
|
|
"loss": 6.8467,
|
|
"mean_token_accuracy": 0.11380974426865578,
|
|
"num_tokens": 5780344.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.951174783706665,
|
|
"epoch": 0.2851944994381722,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004996403141993941,
|
|
"loss": 6.6666,
|
|
"mean_token_accuracy": 0.12389900684356689,
|
|
"num_tokens": 5790167.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.946682929992676,
|
|
"epoch": 0.2857295735459361,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004996381499589555,
|
|
"loss": 6.8417,
|
|
"mean_token_accuracy": 0.11240337193012237,
|
|
"num_tokens": 5800076.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 7.035910606384277,
|
|
"epoch": 0.2862646476537,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004996359792321219,
|
|
"loss": 6.7903,
|
|
"mean_token_accuracy": 0.11802778467535972,
|
|
"num_tokens": 5810547.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.917967891693115,
|
|
"epoch": 0.28679972176146395,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000499633802018956,
|
|
"loss": 6.8173,
|
|
"mean_token_accuracy": 0.11811300665140152,
|
|
"num_tokens": 5821210.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.960421514511109,
|
|
"epoch": 0.2873347958692279,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004996316183195204,
|
|
"loss": 6.8489,
|
|
"mean_token_accuracy": 0.11026589795947075,
|
|
"num_tokens": 5832361.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 7.026253318786621,
|
|
"epoch": 0.2878698699769918,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004996294281338786,
|
|
"loss": 6.8443,
|
|
"mean_token_accuracy": 0.10899497270584106,
|
|
"num_tokens": 5843242.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 7.042272329330444,
|
|
"epoch": 0.28840494408475575,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004996272314620936,
|
|
"loss": 6.8897,
|
|
"mean_token_accuracy": 0.10638803094625474,
|
|
"num_tokens": 5855163.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.922289705276489,
|
|
"epoch": 0.2889400181925197,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004996250283042288,
|
|
"loss": 6.8265,
|
|
"mean_token_accuracy": 0.1135579526424408,
|
|
"num_tokens": 5866908.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 7.018845796585083,
|
|
"epoch": 0.28947509230028357,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000499622818660348,
|
|
"loss": 6.8806,
|
|
"mean_token_accuracy": 0.10942182540893555,
|
|
"num_tokens": 5877342.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.949248218536377,
|
|
"epoch": 0.2900101664080475,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004996206025305147,
|
|
"loss": 6.8217,
|
|
"mean_token_accuracy": 0.10982293486595154,
|
|
"num_tokens": 5887960.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.947164630889892,
|
|
"epoch": 0.29054524051581143,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004996183799147932,
|
|
"loss": 6.7974,
|
|
"mean_token_accuracy": 0.11249916031956672,
|
|
"num_tokens": 5899887.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 7.010842323303223,
|
|
"epoch": 0.29108031462357536,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004996161508132475,
|
|
"loss": 6.8142,
|
|
"mean_token_accuracy": 0.10648874044418336,
|
|
"num_tokens": 5911022.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 7.063547801971436,
|
|
"epoch": 0.2916153887313393,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000499613915225942,
|
|
"loss": 6.8805,
|
|
"mean_token_accuracy": 0.11401300653815269,
|
|
"num_tokens": 5922780.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.957791137695312,
|
|
"epoch": 0.29215046283910323,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996116731529412,
|
|
"loss": 6.9079,
|
|
"mean_token_accuracy": 0.11033277809619904,
|
|
"num_tokens": 5934112.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.973647975921631,
|
|
"epoch": 0.29268553694686716,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00049960942459431,
|
|
"loss": 6.8414,
|
|
"mean_token_accuracy": 0.10922785475850105,
|
|
"num_tokens": 5944386.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.913372564315796,
|
|
"epoch": 0.29322061105463104,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004996071695501132,
|
|
"loss": 6.7672,
|
|
"mean_token_accuracy": 0.11577581316232681,
|
|
"num_tokens": 5955552.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.972257375717163,
|
|
"epoch": 0.293755685162395,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004996049080204159,
|
|
"loss": 6.7788,
|
|
"mean_token_accuracy": 0.11552264243364334,
|
|
"num_tokens": 5966790.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.993252277374268,
|
|
"epoch": 0.2942907592701589,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004996026400052835,
|
|
"loss": 6.7831,
|
|
"mean_token_accuracy": 0.11271130293607712,
|
|
"num_tokens": 5977704.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.870125818252563,
|
|
"epoch": 0.29482583337792284,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004996003655047814,
|
|
"loss": 6.8028,
|
|
"mean_token_accuracy": 0.11466444432735443,
|
|
"num_tokens": 5989221.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.9888458251953125,
|
|
"epoch": 0.2953609074856868,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004995980845189753,
|
|
"loss": 6.834,
|
|
"mean_token_accuracy": 0.1083465002477169,
|
|
"num_tokens": 6000906.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.996977138519287,
|
|
"epoch": 0.2958959815934507,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004995957970479312,
|
|
"loss": 6.8506,
|
|
"mean_token_accuracy": 0.11252145394682884,
|
|
"num_tokens": 6012039.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.935248565673828,
|
|
"epoch": 0.29643105570121464,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004995935030917148,
|
|
"loss": 6.7372,
|
|
"mean_token_accuracy": 0.11361273676156998,
|
|
"num_tokens": 6024383.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 7.046666049957276,
|
|
"epoch": 0.2969661298089785,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004995912026503926,
|
|
"loss": 6.8044,
|
|
"mean_token_accuracy": 0.1105383276939392,
|
|
"num_tokens": 6034904.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.906943941116333,
|
|
"epoch": 0.29750120391674245,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000499588895724031,
|
|
"loss": 6.8156,
|
|
"mean_token_accuracy": 0.1101146712899208,
|
|
"num_tokens": 6044449.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.984871864318848,
|
|
"epoch": 0.2980362780245064,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004995865823126968,
|
|
"loss": 6.8266,
|
|
"mean_token_accuracy": 0.11390623450279236,
|
|
"num_tokens": 6054465.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.957597541809082,
|
|
"epoch": 0.2985713521322703,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004995842624164564,
|
|
"loss": 6.8403,
|
|
"mean_token_accuracy": 0.11375648602843284,
|
|
"num_tokens": 6065539.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.96545295715332,
|
|
"epoch": 0.29910642624003425,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000499581936035377,
|
|
"loss": 6.8012,
|
|
"mean_token_accuracy": 0.11368539929389954,
|
|
"num_tokens": 6076135.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.972835540771484,
|
|
"epoch": 0.2996415003477982,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004995796031695257,
|
|
"loss": 6.8043,
|
|
"mean_token_accuracy": 0.11199271976947785,
|
|
"num_tokens": 6086604.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.936370754241944,
|
|
"epoch": 0.3001765744555621,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00049957726381897,
|
|
"loss": 6.8275,
|
|
"mean_token_accuracy": 0.11501410380005836,
|
|
"num_tokens": 6095672.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.921350908279419,
|
|
"epoch": 0.300711648563326,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004995749179837773,
|
|
"loss": 6.7088,
|
|
"mean_token_accuracy": 0.12310401126742362,
|
|
"num_tokens": 6105652.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.871649312973022,
|
|
"epoch": 0.30124672267108993,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004995725656640155,
|
|
"loss": 6.7453,
|
|
"mean_token_accuracy": 0.11668358147144317,
|
|
"num_tokens": 6116930.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.905279350280762,
|
|
"epoch": 0.30178179677885386,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004995702068597524,
|
|
"loss": 6.7557,
|
|
"mean_token_accuracy": 0.11802282258868217,
|
|
"num_tokens": 6126659.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.936028718948364,
|
|
"epoch": 0.3023168708866178,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004995678415710561,
|
|
"loss": 6.7841,
|
|
"mean_token_accuracy": 0.11382660195231438,
|
|
"num_tokens": 6137227.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 7.028936529159546,
|
|
"epoch": 0.3028519449943817,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000499565469797995,
|
|
"loss": 6.8442,
|
|
"mean_token_accuracy": 0.11068922504782677,
|
|
"num_tokens": 6147800.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.8801487445831295,
|
|
"epoch": 0.30338701910214566,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004995630915406374,
|
|
"loss": 6.6811,
|
|
"mean_token_accuracy": 0.1217758022248745,
|
|
"num_tokens": 6157671.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.933578777313232,
|
|
"epoch": 0.3039220932099096,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004995607067990521,
|
|
"loss": 6.8252,
|
|
"mean_token_accuracy": 0.11223353296518326,
|
|
"num_tokens": 6168651.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.9156494140625,
|
|
"epoch": 0.30445716731767347,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000499558315573308,
|
|
"loss": 6.7999,
|
|
"mean_token_accuracy": 0.1103752687573433,
|
|
"num_tokens": 6179925.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.876695585250855,
|
|
"epoch": 0.3049922414254374,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995559178634741,
|
|
"loss": 6.7933,
|
|
"mean_token_accuracy": 0.1192070983350277,
|
|
"num_tokens": 6190541.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 7.041800832748413,
|
|
"epoch": 0.30552731553320134,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004995535136696196,
|
|
"loss": 6.838,
|
|
"mean_token_accuracy": 0.10699132680892945,
|
|
"num_tokens": 6201491.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.847650003433228,
|
|
"epoch": 0.30606238964096527,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.000499551102991814,
|
|
"loss": 6.6434,
|
|
"mean_token_accuracy": 0.12001704573631286,
|
|
"num_tokens": 6212443.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.952008056640625,
|
|
"epoch": 0.3065974637487292,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004995486858301268,
|
|
"loss": 6.8451,
|
|
"mean_token_accuracy": 0.10546251088380813,
|
|
"num_tokens": 6222623.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.9607843399047855,
|
|
"epoch": 0.30713253785649314,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004995462621846279,
|
|
"loss": 6.7437,
|
|
"mean_token_accuracy": 0.11431733518838882,
|
|
"num_tokens": 6233712.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.921725225448609,
|
|
"epoch": 0.30766761196425707,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004995438320553871,
|
|
"loss": 6.8108,
|
|
"mean_token_accuracy": 0.11184044554829597,
|
|
"num_tokens": 6244254.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.933862113952637,
|
|
"epoch": 0.30820268607202095,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004995413954424748,
|
|
"loss": 6.7041,
|
|
"mean_token_accuracy": 0.12413104623556137,
|
|
"num_tokens": 6254202.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.896720361709595,
|
|
"epoch": 0.3087377601797849,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004995389523459612,
|
|
"loss": 6.7875,
|
|
"mean_token_accuracy": 0.10941583663225174,
|
|
"num_tokens": 6264496.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.89367413520813,
|
|
"epoch": 0.3092728342875488,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000499536502765917,
|
|
"loss": 6.7968,
|
|
"mean_token_accuracy": 0.11605312377214431,
|
|
"num_tokens": 6275835.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.961062669754028,
|
|
"epoch": 0.30980790839531275,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995340467024128,
|
|
"loss": 6.7421,
|
|
"mean_token_accuracy": 0.11403903961181641,
|
|
"num_tokens": 6286805.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.9369104385375975,
|
|
"epoch": 0.3103429825030767,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004995315841555195,
|
|
"loss": 6.7534,
|
|
"mean_token_accuracy": 0.10973872393369674,
|
|
"num_tokens": 6298444.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.821942901611328,
|
|
"epoch": 0.3108780566108406,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004995291151253083,
|
|
"loss": 6.6866,
|
|
"mean_token_accuracy": 0.11732677444815635,
|
|
"num_tokens": 6308259.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.903902912139893,
|
|
"epoch": 0.31141313071860455,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004995266396118505,
|
|
"loss": 6.762,
|
|
"mean_token_accuracy": 0.10594131126999855,
|
|
"num_tokens": 6320058.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.915698528289795,
|
|
"epoch": 0.3119482048263684,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004995241576152174,
|
|
"loss": 6.7316,
|
|
"mean_token_accuracy": 0.10760388672351837,
|
|
"num_tokens": 6331203.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.955277776718139,
|
|
"epoch": 0.31248327893413236,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000499521669135481,
|
|
"loss": 6.8324,
|
|
"mean_token_accuracy": 0.11208058670163154,
|
|
"num_tokens": 6342027.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.928550672531128,
|
|
"epoch": 0.3130183530418963,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004995191741727128,
|
|
"loss": 6.7431,
|
|
"mean_token_accuracy": 0.11125858277082443,
|
|
"num_tokens": 6352580.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.969546937942505,
|
|
"epoch": 0.3135534271496602,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.000499516672726985,
|
|
"loss": 6.7556,
|
|
"mean_token_accuracy": 0.1169343538582325,
|
|
"num_tokens": 6363997.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.8640326976776125,
|
|
"epoch": 0.31408850125742416,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004995141647983699,
|
|
"loss": 6.8027,
|
|
"mean_token_accuracy": 0.11516904607415199,
|
|
"num_tokens": 6375284.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.883784198760987,
|
|
"epoch": 0.3146235753651881,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004995116503869398,
|
|
"loss": 6.6913,
|
|
"mean_token_accuracy": 0.12319560199975968,
|
|
"num_tokens": 6386698.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.950956916809082,
|
|
"epoch": 0.315158649472952,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004995091294927673,
|
|
"loss": 6.7806,
|
|
"mean_token_accuracy": 0.11270102486014366,
|
|
"num_tokens": 6397237.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.972240686416626,
|
|
"epoch": 0.3156937235807159,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004995066021159253,
|
|
"loss": 6.8847,
|
|
"mean_token_accuracy": 0.11340105310082435,
|
|
"num_tokens": 6408266.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.8899555683135985,
|
|
"epoch": 0.31622879768847983,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004995040682564867,
|
|
"loss": 6.775,
|
|
"mean_token_accuracy": 0.11872687861323357,
|
|
"num_tokens": 6419563.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.864490795135498,
|
|
"epoch": 0.31676387179624377,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004995015279145247,
|
|
"loss": 6.7399,
|
|
"mean_token_accuracy": 0.11859864667057991,
|
|
"num_tokens": 6430220.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.928368997573853,
|
|
"epoch": 0.3172989459040077,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004994989810901126,
|
|
"loss": 6.7558,
|
|
"mean_token_accuracy": 0.11230956315994263,
|
|
"num_tokens": 6440157.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.9050641536712645,
|
|
"epoch": 0.31783402001177163,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004994964277833239,
|
|
"loss": 6.7907,
|
|
"mean_token_accuracy": 0.1221705548465252,
|
|
"num_tokens": 6449511.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.942772340774536,
|
|
"epoch": 0.31836909411953557,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004994938679942324,
|
|
"loss": 6.7836,
|
|
"mean_token_accuracy": 0.11035550981760026,
|
|
"num_tokens": 6460608.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.934708499908448,
|
|
"epoch": 0.3189041682272995,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004994913017229121,
|
|
"loss": 6.77,
|
|
"mean_token_accuracy": 0.11715537756681442,
|
|
"num_tokens": 6472737.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.8332374572753904,
|
|
"epoch": 0.31943924233506343,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004994887289694371,
|
|
"loss": 6.6674,
|
|
"mean_token_accuracy": 0.11837019473314285,
|
|
"num_tokens": 6482993.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 7.0178844928741455,
|
|
"epoch": 0.3199743164428273,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004994861497338814,
|
|
"loss": 6.8735,
|
|
"mean_token_accuracy": 0.11568538472056389,
|
|
"num_tokens": 6494530.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.916510963439942,
|
|
"epoch": 0.32050939055059124,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004994835640163198,
|
|
"loss": 6.6973,
|
|
"mean_token_accuracy": 0.1126147136092186,
|
|
"num_tokens": 6504691.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.894684362411499,
|
|
"epoch": 0.3210444646583552,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004994809718168267,
|
|
"loss": 6.7383,
|
|
"mean_token_accuracy": 0.11528538763523102,
|
|
"num_tokens": 6514924.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.3210444646583552,
|
|
"eval_entropy": 6.626076072759705,
|
|
"eval_loss": 6.841334342956543,
|
|
"eval_mean_token_accuracy": 0.11807519927460228,
|
|
"eval_num_tokens": 6514924.0,
|
|
"eval_runtime": 17.6886,
|
|
"eval_samples_per_second": 1677.858,
|
|
"eval_steps_per_second": 209.739,
|
|
"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": 1414915727523840.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|