6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.3369461447745269,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742550659179688,
|
|
"epoch": 0.0005615769079575448,
|
|
"grad_norm": 5.46875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7833,
|
|
"mean_token_accuracy": 9.587727836333216e-05,
|
|
"num_tokens": 11659.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742562961578368,
|
|
"epoch": 0.0011231538159150896,
|
|
"grad_norm": 5.96875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7559,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 21794.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742598533630371,
|
|
"epoch": 0.0016847307238726343,
|
|
"grad_norm": 5.78125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7489,
|
|
"mean_token_accuracy": 6.915629492141307e-05,
|
|
"num_tokens": 33206.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742581844329834,
|
|
"epoch": 0.0022463076318301792,
|
|
"grad_norm": 5.3125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.694,
|
|
"mean_token_accuracy": 0.0004815136082470417,
|
|
"num_tokens": 44487.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742487525939941,
|
|
"epoch": 0.002807884539787724,
|
|
"grad_norm": 4.875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5854,
|
|
"mean_token_accuracy": 0.006200016639195382,
|
|
"num_tokens": 56062.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742135620117187,
|
|
"epoch": 0.0033694614477452686,
|
|
"grad_norm": 3.984375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4464,
|
|
"mean_token_accuracy": 0.028756615333259107,
|
|
"num_tokens": 67453.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741005325317383,
|
|
"epoch": 0.003931038355702813,
|
|
"grad_norm": 3.109375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3034,
|
|
"mean_token_accuracy": 0.03541945908218622,
|
|
"num_tokens": 78527.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.738347434997559,
|
|
"epoch": 0.0044926152636603585,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1364,
|
|
"mean_token_accuracy": 0.04151417501270771,
|
|
"num_tokens": 89267.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.734824466705323,
|
|
"epoch": 0.005054192171617903,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.0434,
|
|
"mean_token_accuracy": 0.04415128082036972,
|
|
"num_tokens": 102035.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.732510948181153,
|
|
"epoch": 0.005615769079575448,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9562,
|
|
"mean_token_accuracy": 0.03771504815667868,
|
|
"num_tokens": 112734.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.731938362121582,
|
|
"epoch": 0.006177345987532993,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.8743,
|
|
"mean_token_accuracy": 0.045871196687221526,
|
|
"num_tokens": 123854.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.73061008453369,
|
|
"epoch": 0.006738922895490537,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8175,
|
|
"mean_token_accuracy": 0.04188961498439312,
|
|
"num_tokens": 133444.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.72856788635254,
|
|
"epoch": 0.007300499803448083,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.7416,
|
|
"mean_token_accuracy": 0.04224029406905174,
|
|
"num_tokens": 144678.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.725998210906983,
|
|
"epoch": 0.007862076711405626,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7079,
|
|
"mean_token_accuracy": 0.04524468667805195,
|
|
"num_tokens": 155866.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.722082805633544,
|
|
"epoch": 0.008423653619363172,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.6415,
|
|
"mean_token_accuracy": 0.04917885698378086,
|
|
"num_tokens": 166965.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.717686748504638,
|
|
"epoch": 0.008985230527320717,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.5656,
|
|
"mean_token_accuracy": 0.04601088799536228,
|
|
"num_tokens": 177423.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.713061332702637,
|
|
"epoch": 0.00954680743527826,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.4879,
|
|
"mean_token_accuracy": 0.05287941135466099,
|
|
"num_tokens": 188687.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.705609130859376,
|
|
"epoch": 0.010108384343235806,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4583,
|
|
"mean_token_accuracy": 0.04864194244146347,
|
|
"num_tokens": 200173.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.695794582366943,
|
|
"epoch": 0.010669961251193351,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3535,
|
|
"mean_token_accuracy": 0.051952024549245836,
|
|
"num_tokens": 210877.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.683987998962403,
|
|
"epoch": 0.011231538159150897,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2644,
|
|
"mean_token_accuracy": 0.04764242954552174,
|
|
"num_tokens": 221770.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.668627834320068,
|
|
"epoch": 0.01179311506710844,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1557,
|
|
"mean_token_accuracy": 0.0619601234793663,
|
|
"num_tokens": 232140.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.643933582305909,
|
|
"epoch": 0.012354691975065986,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1009,
|
|
"mean_token_accuracy": 0.0608074888586998,
|
|
"num_tokens": 244186.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.613188362121582,
|
|
"epoch": 0.012916268883023531,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9502,
|
|
"mean_token_accuracy": 0.06373463608324528,
|
|
"num_tokens": 254805.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.57162914276123,
|
|
"epoch": 0.013477845790981074,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.7802,
|
|
"mean_token_accuracy": 0.06789718717336654,
|
|
"num_tokens": 265055.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.509242630004882,
|
|
"epoch": 0.01403942269893862,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.6595,
|
|
"mean_token_accuracy": 0.06531061716377735,
|
|
"num_tokens": 275850.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.434095478057861,
|
|
"epoch": 0.014600999606896165,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.5608,
|
|
"mean_token_accuracy": 0.06572793014347553,
|
|
"num_tokens": 286802.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.361983966827392,
|
|
"epoch": 0.015162576514853709,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.4719,
|
|
"mean_token_accuracy": 0.06788074560463428,
|
|
"num_tokens": 297070.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.27025728225708,
|
|
"epoch": 0.015724153422811252,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.3615,
|
|
"mean_token_accuracy": 0.0678123127669096,
|
|
"num_tokens": 309617.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.180762577056885,
|
|
"epoch": 0.0162857303307688,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.2203,
|
|
"mean_token_accuracy": 0.06984737552702427,
|
|
"num_tokens": 321310.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.010982036590576,
|
|
"epoch": 0.016847307238726343,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.0541,
|
|
"mean_token_accuracy": 0.07262194454669953,
|
|
"num_tokens": 332243.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.835519409179687,
|
|
"epoch": 0.017408884146683887,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.9575,
|
|
"mean_token_accuracy": 0.06847953647375107,
|
|
"num_tokens": 342847.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.67606430053711,
|
|
"epoch": 0.017970461054641434,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8107,
|
|
"mean_token_accuracy": 0.07214009724557399,
|
|
"num_tokens": 353570.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.435340785980225,
|
|
"epoch": 0.018532037962598977,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.6786,
|
|
"mean_token_accuracy": 0.08545113056898117,
|
|
"num_tokens": 364330.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.163984489440917,
|
|
"epoch": 0.01909361487055652,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.637,
|
|
"mean_token_accuracy": 0.08047130256891251,
|
|
"num_tokens": 374930.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.023346519470214,
|
|
"epoch": 0.019655191778514068,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.558,
|
|
"mean_token_accuracy": 0.08092962652444839,
|
|
"num_tokens": 384690.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.729331684112548,
|
|
"epoch": 0.020216768686471612,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.5022,
|
|
"mean_token_accuracy": 0.08200787231326104,
|
|
"num_tokens": 395753.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.560184669494628,
|
|
"epoch": 0.020778345594429155,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4189,
|
|
"mean_token_accuracy": 0.0850575651973486,
|
|
"num_tokens": 407049.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.38232650756836,
|
|
"epoch": 0.021339922502386702,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3876,
|
|
"mean_token_accuracy": 0.08972770720720291,
|
|
"num_tokens": 417942.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.248396492004394,
|
|
"epoch": 0.021901499410344246,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3787,
|
|
"mean_token_accuracy": 0.09045429974794388,
|
|
"num_tokens": 428151.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.15650987625122,
|
|
"epoch": 0.022463076318301793,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4337,
|
|
"mean_token_accuracy": 0.07940280884504318,
|
|
"num_tokens": 439837.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.114033794403076,
|
|
"epoch": 0.023024653226259337,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2614,
|
|
"mean_token_accuracy": 0.08433199226856232,
|
|
"num_tokens": 450357.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.024269819259644,
|
|
"epoch": 0.02358623013421688,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2624,
|
|
"mean_token_accuracy": 0.08984772339463235,
|
|
"num_tokens": 461174.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.967411947250366,
|
|
"epoch": 0.024147807042174427,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.307,
|
|
"mean_token_accuracy": 0.08454840555787087,
|
|
"num_tokens": 472785.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.925465250015259,
|
|
"epoch": 0.02470938395013197,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2511,
|
|
"mean_token_accuracy": 0.09478851184248924,
|
|
"num_tokens": 483248.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.875831556320191,
|
|
"epoch": 0.025270960858089515,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.142,
|
|
"mean_token_accuracy": 0.09188979342579842,
|
|
"num_tokens": 494690.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.84493317604065,
|
|
"epoch": 0.025832537766047062,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.3296,
|
|
"mean_token_accuracy": 0.08929432481527329,
|
|
"num_tokens": 506192.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.927040433883667,
|
|
"epoch": 0.026394114674004605,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1847,
|
|
"mean_token_accuracy": 0.09605624824762345,
|
|
"num_tokens": 517152.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.799702072143555,
|
|
"epoch": 0.02695569158196215,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1141,
|
|
"mean_token_accuracy": 0.09965193793177604,
|
|
"num_tokens": 527836.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.821067762374878,
|
|
"epoch": 0.027517268489919696,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2301,
|
|
"mean_token_accuracy": 0.09543063268065452,
|
|
"num_tokens": 539491.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.813785552978516,
|
|
"epoch": 0.02807884539787724,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2104,
|
|
"mean_token_accuracy": 0.09490795731544495,
|
|
"num_tokens": 550816.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.805890893936157,
|
|
"epoch": 0.028640422305834783,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.2396,
|
|
"mean_token_accuracy": 0.09041734337806702,
|
|
"num_tokens": 561566.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.751222324371338,
|
|
"epoch": 0.02920199921379233,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1521,
|
|
"mean_token_accuracy": 0.09656212627887725,
|
|
"num_tokens": 573032.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.7648115158081055,
|
|
"epoch": 0.029763576121749874,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0837,
|
|
"mean_token_accuracy": 0.10204266905784606,
|
|
"num_tokens": 584140.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.7348565578460695,
|
|
"epoch": 0.030325153029707418,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.1494,
|
|
"mean_token_accuracy": 0.10442616268992425,
|
|
"num_tokens": 595097.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.7251269817352295,
|
|
"epoch": 0.030886729937664965,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1622,
|
|
"mean_token_accuracy": 0.097444436699152,
|
|
"num_tokens": 606797.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.689198732376099,
|
|
"epoch": 0.031448306845622505,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1191,
|
|
"mean_token_accuracy": 0.09910647869110108,
|
|
"num_tokens": 618187.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.718263101577759,
|
|
"epoch": 0.032009883753580055,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.0568,
|
|
"mean_token_accuracy": 0.10645604655146598,
|
|
"num_tokens": 629829.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.608854055404663,
|
|
"epoch": 0.0325714606615376,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.0604,
|
|
"mean_token_accuracy": 0.10817909017205238,
|
|
"num_tokens": 640914.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.597271966934204,
|
|
"epoch": 0.03313303756949514,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0713,
|
|
"mean_token_accuracy": 0.09871483817696572,
|
|
"num_tokens": 653057.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.650600433349609,
|
|
"epoch": 0.033694614477452686,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9991,
|
|
"mean_token_accuracy": 0.10946055650711059,
|
|
"num_tokens": 663437.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.538260412216187,
|
|
"epoch": 0.03425619138541023,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0616,
|
|
"mean_token_accuracy": 0.10260984227061272,
|
|
"num_tokens": 675416.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.554474115371704,
|
|
"epoch": 0.03481776829336777,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.8611,
|
|
"mean_token_accuracy": 0.1113228090107441,
|
|
"num_tokens": 686633.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.547433662414551,
|
|
"epoch": 0.035379345201325324,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9481,
|
|
"mean_token_accuracy": 0.10943943783640861,
|
|
"num_tokens": 698065.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.557992935180664,
|
|
"epoch": 0.03594092210928287,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9576,
|
|
"mean_token_accuracy": 0.10788579881191254,
|
|
"num_tokens": 708652.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.549759817123413,
|
|
"epoch": 0.03650249901724041,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1014,
|
|
"mean_token_accuracy": 0.104249157756567,
|
|
"num_tokens": 720199.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.516535854339599,
|
|
"epoch": 0.037064075925197955,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.0884,
|
|
"mean_token_accuracy": 0.10238429754972458,
|
|
"num_tokens": 732964.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.490964794158936,
|
|
"epoch": 0.0376256528331555,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.8492,
|
|
"mean_token_accuracy": 0.11204343438148498,
|
|
"num_tokens": 743528.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.517170190811157,
|
|
"epoch": 0.03818722974111304,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9139,
|
|
"mean_token_accuracy": 0.11466537714004517,
|
|
"num_tokens": 754365.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.3772297382354735,
|
|
"epoch": 0.03874880664907059,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8306,
|
|
"mean_token_accuracy": 0.10668750330805779,
|
|
"num_tokens": 766094.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.388029289245606,
|
|
"epoch": 0.039310383557028136,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.89,
|
|
"mean_token_accuracy": 0.11124317348003387,
|
|
"num_tokens": 777275.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.42672815322876,
|
|
"epoch": 0.03987196046498568,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9507,
|
|
"mean_token_accuracy": 0.11006278768181801,
|
|
"num_tokens": 789436.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.445900392532349,
|
|
"epoch": 0.040433537372943223,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9722,
|
|
"mean_token_accuracy": 0.11323394179344178,
|
|
"num_tokens": 800164.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.275260972976684,
|
|
"epoch": 0.04099511428090077,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.8431,
|
|
"mean_token_accuracy": 0.11253381669521331,
|
|
"num_tokens": 811797.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.396295213699341,
|
|
"epoch": 0.04155669118885831,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.8549,
|
|
"mean_token_accuracy": 0.11424426287412644,
|
|
"num_tokens": 824158.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.365560102462768,
|
|
"epoch": 0.04211826809681586,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8502,
|
|
"mean_token_accuracy": 0.11723987609148026,
|
|
"num_tokens": 835803.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.352114200592041,
|
|
"epoch": 0.042679845004773405,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.913,
|
|
"mean_token_accuracy": 0.1121498741209507,
|
|
"num_tokens": 848106.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.327092075347901,
|
|
"epoch": 0.04324142191273095,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.886,
|
|
"mean_token_accuracy": 0.10864873975515366,
|
|
"num_tokens": 859101.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.300347328186035,
|
|
"epoch": 0.04380299882068849,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7609,
|
|
"mean_token_accuracy": 0.12088448479771614,
|
|
"num_tokens": 870309.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.309509944915772,
|
|
"epoch": 0.044364575728646036,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8472,
|
|
"mean_token_accuracy": 0.11545197889208794,
|
|
"num_tokens": 881515.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.336116790771484,
|
|
"epoch": 0.044926152636603586,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8,
|
|
"mean_token_accuracy": 0.11816975474357605,
|
|
"num_tokens": 892129.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.230046129226684,
|
|
"epoch": 0.04548772954456113,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8228,
|
|
"mean_token_accuracy": 0.11072085574269294,
|
|
"num_tokens": 903186.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.260617780685425,
|
|
"epoch": 0.046049306452518673,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7811,
|
|
"mean_token_accuracy": 0.12182695344090462,
|
|
"num_tokens": 913471.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.292573118209839,
|
|
"epoch": 0.04661088336047622,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7853,
|
|
"mean_token_accuracy": 0.11431619748473168,
|
|
"num_tokens": 924283.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.2130533218383786,
|
|
"epoch": 0.04717246026843376,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.8177,
|
|
"mean_token_accuracy": 0.12043273225426673,
|
|
"num_tokens": 934975.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.203190803527832,
|
|
"epoch": 0.047734037176391304,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6934,
|
|
"mean_token_accuracy": 0.12043462619185448,
|
|
"num_tokens": 945587.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.19977822303772,
|
|
"epoch": 0.048295614084348855,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7526,
|
|
"mean_token_accuracy": 0.11818815246224404,
|
|
"num_tokens": 955935.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.208397436141968,
|
|
"epoch": 0.0488571909923064,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7148,
|
|
"mean_token_accuracy": 0.11614556312561035,
|
|
"num_tokens": 967960.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.185926675796509,
|
|
"epoch": 0.04941876790026394,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7518,
|
|
"mean_token_accuracy": 0.12180673331022263,
|
|
"num_tokens": 979680.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.172959804534912,
|
|
"epoch": 0.049980344808221486,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8161,
|
|
"mean_token_accuracy": 0.11992981359362602,
|
|
"num_tokens": 990600.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.308594560623169,
|
|
"epoch": 0.05054192171617903,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.8481,
|
|
"mean_token_accuracy": 0.12150532379746437,
|
|
"num_tokens": 1002331.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.118796253204346,
|
|
"epoch": 0.05110349862413657,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.6594,
|
|
"mean_token_accuracy": 0.12037151604890824,
|
|
"num_tokens": 1013473.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.1910758972167965,
|
|
"epoch": 0.051665075532094124,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.749,
|
|
"mean_token_accuracy": 0.11884191855788231,
|
|
"num_tokens": 1024434.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.1801292419433596,
|
|
"epoch": 0.05222665244005167,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7697,
|
|
"mean_token_accuracy": 0.1166395016014576,
|
|
"num_tokens": 1035452.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.1035239696502686,
|
|
"epoch": 0.05278822934800921,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6438,
|
|
"mean_token_accuracy": 0.11835466921329499,
|
|
"num_tokens": 1046213.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.088221454620362,
|
|
"epoch": 0.053349806255966754,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.666,
|
|
"mean_token_accuracy": 0.12225791588425636,
|
|
"num_tokens": 1057469.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.244529485702515,
|
|
"epoch": 0.0539113831639243,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8169,
|
|
"mean_token_accuracy": 0.11744396165013313,
|
|
"num_tokens": 1069543.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.995753431320191,
|
|
"epoch": 0.05447296007188184,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6818,
|
|
"mean_token_accuracy": 0.12118739560246468,
|
|
"num_tokens": 1080476.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.242927026748657,
|
|
"epoch": 0.05503453697983939,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7896,
|
|
"mean_token_accuracy": 0.12034596726298333,
|
|
"num_tokens": 1091700.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.122195816040039,
|
|
"epoch": 0.055596113887796936,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.677,
|
|
"mean_token_accuracy": 0.12182887271046638,
|
|
"num_tokens": 1103056.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.137054586410523,
|
|
"epoch": 0.05615769079575448,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.7363,
|
|
"mean_token_accuracy": 0.12379441931843757,
|
|
"num_tokens": 1114836.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.997432088851928,
|
|
"epoch": 0.05671926770371202,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.6501,
|
|
"mean_token_accuracy": 0.12145527005195618,
|
|
"num_tokens": 1125993.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.124313068389893,
|
|
"epoch": 0.05728084461166957,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7208,
|
|
"mean_token_accuracy": 0.12192488759756089,
|
|
"num_tokens": 1136719.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.03917121887207,
|
|
"epoch": 0.05784242151962711,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5859,
|
|
"mean_token_accuracy": 0.1255400411784649,
|
|
"num_tokens": 1146947.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.102051591873169,
|
|
"epoch": 0.05840399842758466,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.7183,
|
|
"mean_token_accuracy": 0.12346506491303444,
|
|
"num_tokens": 1157974.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.9870562076568605,
|
|
"epoch": 0.058965575335542204,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.646,
|
|
"mean_token_accuracy": 0.11965501159429551,
|
|
"num_tokens": 1168573.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.062936162948608,
|
|
"epoch": 0.05952715224349975,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.5636,
|
|
"mean_token_accuracy": 0.12382580935955048,
|
|
"num_tokens": 1178446.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.034807634353638,
|
|
"epoch": 0.06008872915145729,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5794,
|
|
"mean_token_accuracy": 0.1255895234644413,
|
|
"num_tokens": 1190558.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.959788131713867,
|
|
"epoch": 0.060650306059414835,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.746,
|
|
"mean_token_accuracy": 0.12257594242691994,
|
|
"num_tokens": 1201929.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.939215898513794,
|
|
"epoch": 0.06121188296737238,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5748,
|
|
"mean_token_accuracy": 0.12955221608281137,
|
|
"num_tokens": 1213155.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.138561248779297,
|
|
"epoch": 0.06177345987532993,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.656,
|
|
"mean_token_accuracy": 0.12121416926383972,
|
|
"num_tokens": 1223961.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.852568626403809,
|
|
"epoch": 0.06233503678328747,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5532,
|
|
"mean_token_accuracy": 0.12889384627342224,
|
|
"num_tokens": 1233888.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.934140396118164,
|
|
"epoch": 0.06289661369124501,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6656,
|
|
"mean_token_accuracy": 0.11890221461653709,
|
|
"num_tokens": 1244405.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.04783000946045,
|
|
"epoch": 0.06345819059920256,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6081,
|
|
"mean_token_accuracy": 0.12794038355350495,
|
|
"num_tokens": 1257099.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.002995395660401,
|
|
"epoch": 0.06401976750716011,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7054,
|
|
"mean_token_accuracy": 0.1265850067138672,
|
|
"num_tokens": 1267965.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.995109415054321,
|
|
"epoch": 0.06458134441511765,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6099,
|
|
"mean_token_accuracy": 0.12399538457393647,
|
|
"num_tokens": 1280075.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.007772064208984,
|
|
"epoch": 0.0651429213230752,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6188,
|
|
"mean_token_accuracy": 0.12613040655851365,
|
|
"num_tokens": 1290574.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.981190586090088,
|
|
"epoch": 0.06570449823103273,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5851,
|
|
"mean_token_accuracy": 0.12990300580859185,
|
|
"num_tokens": 1302680.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.931596183776856,
|
|
"epoch": 0.06626607513899029,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5766,
|
|
"mean_token_accuracy": 0.12965173944830893,
|
|
"num_tokens": 1314577.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.966735315322876,
|
|
"epoch": 0.06682765204694784,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6977,
|
|
"mean_token_accuracy": 0.1269223101437092,
|
|
"num_tokens": 1326560.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.021285438537598,
|
|
"epoch": 0.06738922895490537,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.7678,
|
|
"mean_token_accuracy": 0.11841874867677689,
|
|
"num_tokens": 1339105.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.793500185012817,
|
|
"epoch": 0.06795080586286292,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4295,
|
|
"mean_token_accuracy": 0.1327479861676693,
|
|
"num_tokens": 1349914.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.940507316589356,
|
|
"epoch": 0.06851238277082046,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6528,
|
|
"mean_token_accuracy": 0.12645572870969773,
|
|
"num_tokens": 1361700.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.953217029571533,
|
|
"epoch": 0.06907395967877801,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.6735,
|
|
"mean_token_accuracy": 0.12332428023219108,
|
|
"num_tokens": 1373103.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.914129304885864,
|
|
"epoch": 0.06963553658673555,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.459,
|
|
"mean_token_accuracy": 0.13300212621688842,
|
|
"num_tokens": 1383742.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.799388074874878,
|
|
"epoch": 0.0701971134946931,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6126,
|
|
"mean_token_accuracy": 0.12680527344346046,
|
|
"num_tokens": 1396656.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.921048307418824,
|
|
"epoch": 0.07075869040265065,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4886,
|
|
"mean_token_accuracy": 0.14117394238710404,
|
|
"num_tokens": 1406709.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.875065326690674,
|
|
"epoch": 0.07132026731060818,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5279,
|
|
"mean_token_accuracy": 0.13476523160934448,
|
|
"num_tokens": 1417793.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.934533977508545,
|
|
"epoch": 0.07188184421856574,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.6046,
|
|
"mean_token_accuracy": 0.1279690831899643,
|
|
"num_tokens": 1429093.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.780666828155518,
|
|
"epoch": 0.07244342112652327,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5326,
|
|
"mean_token_accuracy": 0.13261018618941306,
|
|
"num_tokens": 1440907.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.917075872421265,
|
|
"epoch": 0.07300499803448082,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.6025,
|
|
"mean_token_accuracy": 0.12796421274542807,
|
|
"num_tokens": 1452089.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.888136577606201,
|
|
"epoch": 0.07356657494243837,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6206,
|
|
"mean_token_accuracy": 0.12372568026185035,
|
|
"num_tokens": 1464378.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.848985958099365,
|
|
"epoch": 0.07412815185039591,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5431,
|
|
"mean_token_accuracy": 0.1299239970743656,
|
|
"num_tokens": 1476678.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.798602628707886,
|
|
"epoch": 0.07468972875835346,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.3923,
|
|
"mean_token_accuracy": 0.13382024466991424,
|
|
"num_tokens": 1486808.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.9307825565338135,
|
|
"epoch": 0.075251305666311,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6765,
|
|
"mean_token_accuracy": 0.12494136691093445,
|
|
"num_tokens": 1499281.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.8795429229736325,
|
|
"epoch": 0.07581288257426855,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5764,
|
|
"mean_token_accuracy": 0.12798068225383757,
|
|
"num_tokens": 1510202.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.831705522537232,
|
|
"epoch": 0.07637445948222608,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4526,
|
|
"mean_token_accuracy": 0.1334339514374733,
|
|
"num_tokens": 1521492.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.733158874511719,
|
|
"epoch": 0.07693603639018363,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4612,
|
|
"mean_token_accuracy": 0.13912885189056395,
|
|
"num_tokens": 1532856.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.818010997772217,
|
|
"epoch": 0.07749761329814119,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4673,
|
|
"mean_token_accuracy": 0.1312962956726551,
|
|
"num_tokens": 1543408.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.730507946014404,
|
|
"epoch": 0.07805919020609872,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5037,
|
|
"mean_token_accuracy": 0.12911636829376222,
|
|
"num_tokens": 1555474.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.848674011230469,
|
|
"epoch": 0.07862076711405627,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5921,
|
|
"mean_token_accuracy": 0.13230398222804068,
|
|
"num_tokens": 1567484.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.715642690658569,
|
|
"epoch": 0.07918234402201381,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4247,
|
|
"mean_token_accuracy": 0.14034333527088166,
|
|
"num_tokens": 1578204.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.863985157012939,
|
|
"epoch": 0.07974392092997136,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4245,
|
|
"mean_token_accuracy": 0.1318908639252186,
|
|
"num_tokens": 1589012.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.696931076049805,
|
|
"epoch": 0.08030549783792891,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4363,
|
|
"mean_token_accuracy": 0.13300200998783113,
|
|
"num_tokens": 1601099.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.80228910446167,
|
|
"epoch": 0.08086707474588645,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5967,
|
|
"mean_token_accuracy": 0.12784573882818223,
|
|
"num_tokens": 1612814.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.878298664093018,
|
|
"epoch": 0.081428651653844,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4838,
|
|
"mean_token_accuracy": 0.13338544070720673,
|
|
"num_tokens": 1624953.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.6930359363555905,
|
|
"epoch": 0.08199022856180153,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.5261,
|
|
"mean_token_accuracy": 0.13288487046957015,
|
|
"num_tokens": 1635673.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.79468789100647,
|
|
"epoch": 0.08255180546975908,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.4303,
|
|
"mean_token_accuracy": 0.13556234017014504,
|
|
"num_tokens": 1647282.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.7501686096191404,
|
|
"epoch": 0.08311338237771662,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.4432,
|
|
"mean_token_accuracy": 0.13580124080181122,
|
|
"num_tokens": 1658426.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.72845983505249,
|
|
"epoch": 0.08367495928567417,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4705,
|
|
"mean_token_accuracy": 0.13703056201338767,
|
|
"num_tokens": 1669958.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.7487999439239506,
|
|
"epoch": 0.08423653619363172,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5243,
|
|
"mean_token_accuracy": 0.13213849067687988,
|
|
"num_tokens": 1682409.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.751234817504883,
|
|
"epoch": 0.08479811310158926,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4479,
|
|
"mean_token_accuracy": 0.1415566846728325,
|
|
"num_tokens": 1692409.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.7592613220214846,
|
|
"epoch": 0.08535969000954681,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.5309,
|
|
"mean_token_accuracy": 0.13545768857002258,
|
|
"num_tokens": 1705018.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.738600540161133,
|
|
"epoch": 0.08592126691750435,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4573,
|
|
"mean_token_accuracy": 0.13337523862719536,
|
|
"num_tokens": 1715625.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.6791502952575685,
|
|
"epoch": 0.0864828438254619,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.537,
|
|
"mean_token_accuracy": 0.12219079732894897,
|
|
"num_tokens": 1727045.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.9133740901947025,
|
|
"epoch": 0.08704442073341945,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5994,
|
|
"mean_token_accuracy": 0.12983450889587403,
|
|
"num_tokens": 1737374.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.78628134727478,
|
|
"epoch": 0.08760599764137698,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5867,
|
|
"mean_token_accuracy": 0.13110264837741853,
|
|
"num_tokens": 1747788.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.590769243240357,
|
|
"epoch": 0.08816757454933453,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2882,
|
|
"mean_token_accuracy": 0.1412363886833191,
|
|
"num_tokens": 1757769.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.66961555480957,
|
|
"epoch": 0.08872915145729207,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.4112,
|
|
"mean_token_accuracy": 0.13004442900419236,
|
|
"num_tokens": 1769246.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.755950880050659,
|
|
"epoch": 0.08929072836524962,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5278,
|
|
"mean_token_accuracy": 0.13320355713367463,
|
|
"num_tokens": 1781306.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.751245784759521,
|
|
"epoch": 0.08985230527320717,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.429,
|
|
"mean_token_accuracy": 0.1355741389095783,
|
|
"num_tokens": 1792888.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.691725397109986,
|
|
"epoch": 0.09041388218116471,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.437,
|
|
"mean_token_accuracy": 0.13696239069104194,
|
|
"num_tokens": 1802938.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.722809076309204,
|
|
"epoch": 0.09097545908912226,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4553,
|
|
"mean_token_accuracy": 0.1335028439760208,
|
|
"num_tokens": 1814738.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.6880824089050295,
|
|
"epoch": 0.0915370359970798,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5314,
|
|
"mean_token_accuracy": 0.1331937789916992,
|
|
"num_tokens": 1825100.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.560724306106567,
|
|
"epoch": 0.09209861290503735,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3725,
|
|
"mean_token_accuracy": 0.13867289274930955,
|
|
"num_tokens": 1835879.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.843134069442749,
|
|
"epoch": 0.09266018981299488,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.5243,
|
|
"mean_token_accuracy": 0.125446654856205,
|
|
"num_tokens": 1848352.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.709258794784546,
|
|
"epoch": 0.09322176672095243,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4662,
|
|
"mean_token_accuracy": 0.13462386131286622,
|
|
"num_tokens": 1859687.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.656016445159912,
|
|
"epoch": 0.09378334362890998,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4593,
|
|
"mean_token_accuracy": 0.13432704880833626,
|
|
"num_tokens": 1871438.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.657505941390991,
|
|
"epoch": 0.09434492053686752,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3499,
|
|
"mean_token_accuracy": 0.1395242303609848,
|
|
"num_tokens": 1881519.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.7550750255584715,
|
|
"epoch": 0.09490649744482507,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.519,
|
|
"mean_token_accuracy": 0.1305337093770504,
|
|
"num_tokens": 1892428.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.742611217498779,
|
|
"epoch": 0.09546807435278261,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4219,
|
|
"mean_token_accuracy": 0.13647260293364524,
|
|
"num_tokens": 1903760.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.685311698913575,
|
|
"epoch": 0.09602965126074016,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.47,
|
|
"mean_token_accuracy": 0.130086899548769,
|
|
"num_tokens": 1915430.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.657859754562378,
|
|
"epoch": 0.09659122816869771,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4318,
|
|
"mean_token_accuracy": 0.13017144575715064,
|
|
"num_tokens": 1925709.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.760464239120483,
|
|
"epoch": 0.09715280507665525,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4697,
|
|
"mean_token_accuracy": 0.1342718742787838,
|
|
"num_tokens": 1938148.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.464514970779419,
|
|
"epoch": 0.0977143819846128,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3438,
|
|
"mean_token_accuracy": 0.13634837418794632,
|
|
"num_tokens": 1949028.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.70568528175354,
|
|
"epoch": 0.09827595889257033,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4016,
|
|
"mean_token_accuracy": 0.13514200374484062,
|
|
"num_tokens": 1960353.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.69066276550293,
|
|
"epoch": 0.09883753580052788,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5016,
|
|
"mean_token_accuracy": 0.13418799117207528,
|
|
"num_tokens": 1972335.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.591297006607055,
|
|
"epoch": 0.09939911270848542,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.3568,
|
|
"mean_token_accuracy": 0.136691115796566,
|
|
"num_tokens": 1982851.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.583767509460449,
|
|
"epoch": 0.09996068961644297,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3297,
|
|
"mean_token_accuracy": 0.1408061757683754,
|
|
"num_tokens": 1993931.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.68323655128479,
|
|
"epoch": 0.10052226652440052,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4678,
|
|
"mean_token_accuracy": 0.135716450214386,
|
|
"num_tokens": 2004838.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.533196067810058,
|
|
"epoch": 0.10108384343235806,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2726,
|
|
"mean_token_accuracy": 0.1382671043276787,
|
|
"num_tokens": 2015620.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.655929517745972,
|
|
"epoch": 0.10164542034031561,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4045,
|
|
"mean_token_accuracy": 0.14243076518177986,
|
|
"num_tokens": 2027489.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.497221660614014,
|
|
"epoch": 0.10220699724827315,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2791,
|
|
"mean_token_accuracy": 0.1438898392021656,
|
|
"num_tokens": 2038475.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.563181304931641,
|
|
"epoch": 0.1027685741562307,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2808,
|
|
"mean_token_accuracy": 0.14109615236520767,
|
|
"num_tokens": 2048938.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.534266376495362,
|
|
"epoch": 0.10333015106418825,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3659,
|
|
"mean_token_accuracy": 0.13268538564443588,
|
|
"num_tokens": 2060372.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.537831211090088,
|
|
"epoch": 0.10389172797214578,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3626,
|
|
"mean_token_accuracy": 0.13345034644007683,
|
|
"num_tokens": 2071558.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.619134902954102,
|
|
"epoch": 0.10445330488010333,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3985,
|
|
"mean_token_accuracy": 0.13327228128910065,
|
|
"num_tokens": 2082350.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.5742902755737305,
|
|
"epoch": 0.10501488178806087,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4442,
|
|
"mean_token_accuracy": 0.1369822859764099,
|
|
"num_tokens": 2093448.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.717497825622559,
|
|
"epoch": 0.10557645869601842,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.4786,
|
|
"mean_token_accuracy": 0.135139499604702,
|
|
"num_tokens": 2104171.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.5645914554595945,
|
|
"epoch": 0.10613803560397596,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.3644,
|
|
"mean_token_accuracy": 0.13385832235217093,
|
|
"num_tokens": 2114496.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.612611770629883,
|
|
"epoch": 0.10669961251193351,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3628,
|
|
"mean_token_accuracy": 0.1407134100794792,
|
|
"num_tokens": 2125930.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.603974390029907,
|
|
"epoch": 0.10726118941989106,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3126,
|
|
"mean_token_accuracy": 0.1430915966629982,
|
|
"num_tokens": 2136452.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.581751966476441,
|
|
"epoch": 0.1078227663278486,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4209,
|
|
"mean_token_accuracy": 0.13645661920309066,
|
|
"num_tokens": 2148233.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.4400779724121096,
|
|
"epoch": 0.10838434323580615,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.209,
|
|
"mean_token_accuracy": 0.1424497328698635,
|
|
"num_tokens": 2159475.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.48789873123169,
|
|
"epoch": 0.10894592014376368,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3354,
|
|
"mean_token_accuracy": 0.13974294438958168,
|
|
"num_tokens": 2170042.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.54668779373169,
|
|
"epoch": 0.10950749705172123,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3332,
|
|
"mean_token_accuracy": 0.13519977256655694,
|
|
"num_tokens": 2182234.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.461383056640625,
|
|
"epoch": 0.11006907395967878,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.3302,
|
|
"mean_token_accuracy": 0.1389101855456829,
|
|
"num_tokens": 2192477.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.4945393085479735,
|
|
"epoch": 0.11063065086763632,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3111,
|
|
"mean_token_accuracy": 0.13645306155085563,
|
|
"num_tokens": 2203082.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.488329744338989,
|
|
"epoch": 0.11119222777559387,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.2808,
|
|
"mean_token_accuracy": 0.1423295982182026,
|
|
"num_tokens": 2213684.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.481782674789429,
|
|
"epoch": 0.11175380468355141,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2903,
|
|
"mean_token_accuracy": 0.13810332641005515,
|
|
"num_tokens": 2225524.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.548757648468017,
|
|
"epoch": 0.11231538159150896,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2541,
|
|
"mean_token_accuracy": 0.14044561684131623,
|
|
"num_tokens": 2236004.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.52207841873169,
|
|
"epoch": 0.1128769584994665,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000499999997707492,
|
|
"loss": 6.4111,
|
|
"mean_token_accuracy": 0.12956544756889343,
|
|
"num_tokens": 2247730.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.574723386764527,
|
|
"epoch": 0.11343853540742405,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999999883941787,
|
|
"loss": 6.3719,
|
|
"mean_token_accuracy": 0.14130636155605317,
|
|
"num_tokens": 2260132.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.472715663909912,
|
|
"epoch": 0.1140001123153816,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999999719167785,
|
|
"loss": 6.332,
|
|
"mean_token_accuracy": 0.14104451537132262,
|
|
"num_tokens": 2270060.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.53025188446045,
|
|
"epoch": 0.11456168922333913,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999999482752917,
|
|
"loss": 6.3458,
|
|
"mean_token_accuracy": 0.1401911422610283,
|
|
"num_tokens": 2281568.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.55501389503479,
|
|
"epoch": 0.11512326613129668,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999999174697195,
|
|
"loss": 6.3655,
|
|
"mean_token_accuracy": 0.12993661910295487,
|
|
"num_tokens": 2293325.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.545099115371704,
|
|
"epoch": 0.11568484303925422,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999998795000627,
|
|
"loss": 6.3734,
|
|
"mean_token_accuracy": 0.1424901679158211,
|
|
"num_tokens": 2304780.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.4766498565673825,
|
|
"epoch": 0.11624641994721177,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999998343663223,
|
|
"loss": 6.3002,
|
|
"mean_token_accuracy": 0.13595817238092422,
|
|
"num_tokens": 2315999.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.538008546829223,
|
|
"epoch": 0.11680799685516932,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999997820685,
|
|
"loss": 6.4523,
|
|
"mean_token_accuracy": 0.1396072454750538,
|
|
"num_tokens": 2326695.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.518211507797242,
|
|
"epoch": 0.11736957376312686,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999997226065974,
|
|
"loss": 6.2754,
|
|
"mean_token_accuracy": 0.1473626583814621,
|
|
"num_tokens": 2336925.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.457152557373047,
|
|
"epoch": 0.11793115067108441,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999996559806163,
|
|
"loss": 6.1916,
|
|
"mean_token_accuracy": 0.14323539957404136,
|
|
"num_tokens": 2347431.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.450267267227173,
|
|
"epoch": 0.11849272757904195,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999995821905589,
|
|
"loss": 6.2976,
|
|
"mean_token_accuracy": 0.14234493300318718,
|
|
"num_tokens": 2359894.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.444874811172485,
|
|
"epoch": 0.1190543044869995,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999995012364276,
|
|
"loss": 6.3301,
|
|
"mean_token_accuracy": 0.1389567509293556,
|
|
"num_tokens": 2371181.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.532542181015015,
|
|
"epoch": 0.11961588139495703,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999994131182248,
|
|
"loss": 6.2249,
|
|
"mean_token_accuracy": 0.14046408236026764,
|
|
"num_tokens": 2381531.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.54881477355957,
|
|
"epoch": 0.12017745830291458,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999993178359535,
|
|
"loss": 6.3564,
|
|
"mean_token_accuracy": 0.14078477621078492,
|
|
"num_tokens": 2391573.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.4431471824646,
|
|
"epoch": 0.12073903521087213,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999992153896166,
|
|
"loss": 6.3146,
|
|
"mean_token_accuracy": 0.1424903988838196,
|
|
"num_tokens": 2402822.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.424335241317749,
|
|
"epoch": 0.12130061211882967,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999991057792175,
|
|
"loss": 6.2338,
|
|
"mean_token_accuracy": 0.14391232281923294,
|
|
"num_tokens": 2414322.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.573728752136231,
|
|
"epoch": 0.12186218902678722,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999989890047596,
|
|
"loss": 6.4169,
|
|
"mean_token_accuracy": 0.14226323813199998,
|
|
"num_tokens": 2426560.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.537834119796753,
|
|
"epoch": 0.12242376593474476,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999988650662467,
|
|
"loss": 6.3298,
|
|
"mean_token_accuracy": 0.14009772390127181,
|
|
"num_tokens": 2437723.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.393655395507812,
|
|
"epoch": 0.12298534284270231,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999987339636825,
|
|
"loss": 6.2686,
|
|
"mean_token_accuracy": 0.1454232707619667,
|
|
"num_tokens": 2448497.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.39782018661499,
|
|
"epoch": 0.12354691975065986,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999985956970713,
|
|
"loss": 6.1461,
|
|
"mean_token_accuracy": 0.15012311190366745,
|
|
"num_tokens": 2459089.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.399022197723388,
|
|
"epoch": 0.1241084966586174,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999984502664178,
|
|
"loss": 6.232,
|
|
"mean_token_accuracy": 0.14566392898559571,
|
|
"num_tokens": 2469882.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.414362478256225,
|
|
"epoch": 0.12467007356657495,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999982976717261,
|
|
"loss": 6.2702,
|
|
"mean_token_accuracy": 0.14421137124300004,
|
|
"num_tokens": 2482120.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.482989406585693,
|
|
"epoch": 0.1252316504745325,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999981379130016,
|
|
"loss": 6.281,
|
|
"mean_token_accuracy": 0.14298845529556276,
|
|
"num_tokens": 2492626.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.556837034225464,
|
|
"epoch": 0.12579322738249002,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000499997970990249,
|
|
"loss": 6.3345,
|
|
"mean_token_accuracy": 0.1420891359448433,
|
|
"num_tokens": 2506347.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.283041858673096,
|
|
"epoch": 0.12635480429044757,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999977969034737,
|
|
"loss": 6.2195,
|
|
"mean_token_accuracy": 0.14754280745983123,
|
|
"num_tokens": 2516584.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.48365969657898,
|
|
"epoch": 0.12691638119840512,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999976156526814,
|
|
"loss": 6.2571,
|
|
"mean_token_accuracy": 0.14406370520591735,
|
|
"num_tokens": 2526731.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.371777677536011,
|
|
"epoch": 0.12747795810636267,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999974272378776,
|
|
"loss": 6.1664,
|
|
"mean_token_accuracy": 0.14858057275414466,
|
|
"num_tokens": 2537180.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.46185154914856,
|
|
"epoch": 0.12803953501432022,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999972316590685,
|
|
"loss": 6.2588,
|
|
"mean_token_accuracy": 0.14099827483296395,
|
|
"num_tokens": 2547617.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.342129230499268,
|
|
"epoch": 0.12860111192227774,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999970289162604,
|
|
"loss": 6.1259,
|
|
"mean_token_accuracy": 0.1511496439576149,
|
|
"num_tokens": 2557743.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.364615345001221,
|
|
"epoch": 0.1291626888302353,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999968190094595,
|
|
"loss": 6.2275,
|
|
"mean_token_accuracy": 0.14248820543289184,
|
|
"num_tokens": 2568351.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.346706676483154,
|
|
"epoch": 0.12972426573819285,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999966019386727,
|
|
"loss": 6.284,
|
|
"mean_token_accuracy": 0.1438751205801964,
|
|
"num_tokens": 2579703.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.507516384124756,
|
|
"epoch": 0.1302858426461504,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999963777039067,
|
|
"loss": 6.2616,
|
|
"mean_token_accuracy": 0.14255546554923057,
|
|
"num_tokens": 2591197.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.382220983505249,
|
|
"epoch": 0.13084741955410795,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499996146305169,
|
|
"loss": 6.2282,
|
|
"mean_token_accuracy": 0.14040454253554344,
|
|
"num_tokens": 2601836.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.333654737472534,
|
|
"epoch": 0.13140899646206547,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999959077424665,
|
|
"loss": 6.1738,
|
|
"mean_token_accuracy": 0.14583537429571153,
|
|
"num_tokens": 2613513.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.3707410335540775,
|
|
"epoch": 0.13197057337002302,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999956620158072,
|
|
"loss": 6.2076,
|
|
"mean_token_accuracy": 0.14694963172078132,
|
|
"num_tokens": 2625246.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.359919452667237,
|
|
"epoch": 0.13253215027798057,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999954091251987,
|
|
"loss": 6.1798,
|
|
"mean_token_accuracy": 0.14660840332508088,
|
|
"num_tokens": 2636375.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.385069990158081,
|
|
"epoch": 0.13309372718593812,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999951490706491,
|
|
"loss": 6.2716,
|
|
"mean_token_accuracy": 0.1412426635622978,
|
|
"num_tokens": 2648586.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.442723178863526,
|
|
"epoch": 0.13365530409389567,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999948818521667,
|
|
"loss": 6.2785,
|
|
"mean_token_accuracy": 0.1502460092306137,
|
|
"num_tokens": 2660847.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.351391172409057,
|
|
"epoch": 0.1342168810018532,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00049999460746976,
|
|
"loss": 6.2095,
|
|
"mean_token_accuracy": 0.14432434290647506,
|
|
"num_tokens": 2671160.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.561935091018677,
|
|
"epoch": 0.13477845790981074,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999943259234377,
|
|
"loss": 6.3171,
|
|
"mean_token_accuracy": 0.14210065752267836,
|
|
"num_tokens": 2681622.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.271227216720581,
|
|
"epoch": 0.1353400348177683,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999940372132089,
|
|
"loss": 6.1086,
|
|
"mean_token_accuracy": 0.1514277994632721,
|
|
"num_tokens": 2691503.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.338787794113159,
|
|
"epoch": 0.13590161172572585,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999937413390826,
|
|
"loss": 6.2185,
|
|
"mean_token_accuracy": 0.14308153241872787,
|
|
"num_tokens": 2702207.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.401724863052368,
|
|
"epoch": 0.1364631886336834,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999934383010683,
|
|
"loss": 6.2105,
|
|
"mean_token_accuracy": 0.14515243843197823,
|
|
"num_tokens": 2713178.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.416276550292968,
|
|
"epoch": 0.13702476554164092,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999931280991757,
|
|
"loss": 6.2213,
|
|
"mean_token_accuracy": 0.15066151767969133,
|
|
"num_tokens": 2723717.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.349833822250366,
|
|
"epoch": 0.13758634244959847,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999928107334146,
|
|
"loss": 6.2798,
|
|
"mean_token_accuracy": 0.1415386199951172,
|
|
"num_tokens": 2735669.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.367289066314697,
|
|
"epoch": 0.13814791935755602,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999924862037952,
|
|
"loss": 6.1856,
|
|
"mean_token_accuracy": 0.1438096582889557,
|
|
"num_tokens": 2746087.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.332089853286743,
|
|
"epoch": 0.13870949626551357,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999921545103278,
|
|
"loss": 6.2014,
|
|
"mean_token_accuracy": 0.14393291622400284,
|
|
"num_tokens": 2757143.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.428728485107422,
|
|
"epoch": 0.1392710731734711,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999918156530229,
|
|
"loss": 6.3014,
|
|
"mean_token_accuracy": 0.14128980934619903,
|
|
"num_tokens": 2769340.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.446348237991333,
|
|
"epoch": 0.13983265008142864,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999914696318913,
|
|
"loss": 6.2146,
|
|
"mean_token_accuracy": 0.14424442052841185,
|
|
"num_tokens": 2779514.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.338409376144409,
|
|
"epoch": 0.1403942269893862,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999911164469441,
|
|
"loss": 6.1603,
|
|
"mean_token_accuracy": 0.14622590690851212,
|
|
"num_tokens": 2790801.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.357332468032837,
|
|
"epoch": 0.14095580389734375,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999907560981925,
|
|
"loss": 6.1487,
|
|
"mean_token_accuracy": 0.146085686981678,
|
|
"num_tokens": 2801082.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.335656404495239,
|
|
"epoch": 0.1415173808053013,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999903885856479,
|
|
"loss": 6.1549,
|
|
"mean_token_accuracy": 0.14812700152397157,
|
|
"num_tokens": 2812719.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.2651628971099855,
|
|
"epoch": 0.14207895771325882,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499990013909322,
|
|
"loss": 6.1605,
|
|
"mean_token_accuracy": 0.14404110759496688,
|
|
"num_tokens": 2823766.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.331585931777954,
|
|
"epoch": 0.14264053462121637,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499989632069227,
|
|
"loss": 6.2544,
|
|
"mean_token_accuracy": 0.14194617569446563,
|
|
"num_tokens": 2834801.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.331230735778808,
|
|
"epoch": 0.14320211152917392,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999892430653747,
|
|
"loss": 6.067,
|
|
"mean_token_accuracy": 0.1510644406080246,
|
|
"num_tokens": 2845769.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.218266439437866,
|
|
"epoch": 0.14376368843713147,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999888468977776,
|
|
"loss": 6.1092,
|
|
"mean_token_accuracy": 0.147148197889328,
|
|
"num_tokens": 2856001.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.341158008575439,
|
|
"epoch": 0.14432526534508902,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999884435664485,
|
|
"loss": 6.1444,
|
|
"mean_token_accuracy": 0.14794962406158446,
|
|
"num_tokens": 2867703.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.199103784561157,
|
|
"epoch": 0.14488684225304654,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999880330714,
|
|
"loss": 6.0821,
|
|
"mean_token_accuracy": 0.1512378692626953,
|
|
"num_tokens": 2878671.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.32013258934021,
|
|
"epoch": 0.1454484191610041,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999876154126453,
|
|
"loss": 6.2049,
|
|
"mean_token_accuracy": 0.14154347777366638,
|
|
"num_tokens": 2890137.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.280849170684815,
|
|
"epoch": 0.14600999606896164,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999871905901977,
|
|
"loss": 6.108,
|
|
"mean_token_accuracy": 0.14886464178562164,
|
|
"num_tokens": 2900344.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.263293170928955,
|
|
"epoch": 0.1465715729769192,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999867586040706,
|
|
"loss": 6.1274,
|
|
"mean_token_accuracy": 0.14697801023721696,
|
|
"num_tokens": 2912270.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.3712437629699705,
|
|
"epoch": 0.14713314988487675,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999863194542779,
|
|
"loss": 6.206,
|
|
"mean_token_accuracy": 0.13799021914601325,
|
|
"num_tokens": 2924677.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.215903329849243,
|
|
"epoch": 0.14769472679283427,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999858731408336,
|
|
"loss": 6.1221,
|
|
"mean_token_accuracy": 0.15069050639867781,
|
|
"num_tokens": 2935794.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.380605936050415,
|
|
"epoch": 0.14825630370079182,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999854196637518,
|
|
"loss": 6.2322,
|
|
"mean_token_accuracy": 0.14286753982305528,
|
|
"num_tokens": 2945958.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.292030143737793,
|
|
"epoch": 0.14881788060874937,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999849590230469,
|
|
"loss": 6.1825,
|
|
"mean_token_accuracy": 0.14438321217894554,
|
|
"num_tokens": 2957536.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.38133692741394,
|
|
"epoch": 0.14937945751670692,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999844912187336,
|
|
"loss": 6.1535,
|
|
"mean_token_accuracy": 0.14794423878192903,
|
|
"num_tokens": 2967787.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.29340763092041,
|
|
"epoch": 0.14994103442466447,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499984016250827,
|
|
"loss": 6.1824,
|
|
"mean_token_accuracy": 0.14508575797080994,
|
|
"num_tokens": 2980522.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.2017439842224125,
|
|
"epoch": 0.150502611332622,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499983534119342,
|
|
"loss": 6.124,
|
|
"mean_token_accuracy": 0.15016201436519622,
|
|
"num_tokens": 2990497.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.38787727355957,
|
|
"epoch": 0.15106418824057954,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999830448242941,
|
|
"loss": 6.2135,
|
|
"mean_token_accuracy": 0.14598810970783233,
|
|
"num_tokens": 3002931.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.314008617401123,
|
|
"epoch": 0.1516257651485371,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999825483656986,
|
|
"loss": 6.134,
|
|
"mean_token_accuracy": 0.1540141612291336,
|
|
"num_tokens": 3014137.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.420076322555542,
|
|
"epoch": 0.15218734205649465,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999820447435716,
|
|
"loss": 6.309,
|
|
"mean_token_accuracy": 0.13888537734746934,
|
|
"num_tokens": 3026145.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.252240705490112,
|
|
"epoch": 0.15274891896445217,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999815339579291,
|
|
"loss": 6.1566,
|
|
"mean_token_accuracy": 0.1455765649676323,
|
|
"num_tokens": 3037068.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.313040637969971,
|
|
"epoch": 0.15331049587240972,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999810160087872,
|
|
"loss": 6.0889,
|
|
"mean_token_accuracy": 0.1504852518439293,
|
|
"num_tokens": 3047836.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.320889616012574,
|
|
"epoch": 0.15387207278036727,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999804908961626,
|
|
"loss": 6.2165,
|
|
"mean_token_accuracy": 0.14922351911664009,
|
|
"num_tokens": 3060309.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.3300347328186035,
|
|
"epoch": 0.15443364968832482,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999799586200719,
|
|
"loss": 6.1815,
|
|
"mean_token_accuracy": 0.14712618961930274,
|
|
"num_tokens": 3071789.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.247071981430054,
|
|
"epoch": 0.15499522659628237,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499979419180532,
|
|
"loss": 6.0938,
|
|
"mean_token_accuracy": 0.15286702811717987,
|
|
"num_tokens": 3081756.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.298181962966919,
|
|
"epoch": 0.1555568035042399,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999788725775601,
|
|
"loss": 6.21,
|
|
"mean_token_accuracy": 0.14156180024147033,
|
|
"num_tokens": 3093456.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.382714557647705,
|
|
"epoch": 0.15611838041219744,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999783188111737,
|
|
"loss": 6.1381,
|
|
"mean_token_accuracy": 0.14954148083925248,
|
|
"num_tokens": 3104565.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.256937789916992,
|
|
"epoch": 0.156679957320155,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999777578813904,
|
|
"loss": 6.1531,
|
|
"mean_token_accuracy": 0.14698541909456253,
|
|
"num_tokens": 3115577.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.24382963180542,
|
|
"epoch": 0.15724153422811254,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499977189788228,
|
|
"loss": 6.0549,
|
|
"mean_token_accuracy": 0.14838047921657563,
|
|
"num_tokens": 3126019.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.324401712417602,
|
|
"epoch": 0.1578031111360701,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999766145317046,
|
|
"loss": 6.2172,
|
|
"mean_token_accuracy": 0.13645764887332917,
|
|
"num_tokens": 3136191.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.369932317733765,
|
|
"epoch": 0.15836468804402762,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999760321118385,
|
|
"loss": 6.0737,
|
|
"mean_token_accuracy": 0.1490643858909607,
|
|
"num_tokens": 3147737.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.144622421264648,
|
|
"epoch": 0.15892626495198517,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999754425286483,
|
|
"loss": 6.1337,
|
|
"mean_token_accuracy": 0.14620147198438643,
|
|
"num_tokens": 3158494.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.409546136856079,
|
|
"epoch": 0.15948784185994272,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999748457821529,
|
|
"loss": 6.1659,
|
|
"mean_token_accuracy": 0.14886748641729355,
|
|
"num_tokens": 3170550.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.171326398849487,
|
|
"epoch": 0.16004941876790027,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000499974241872371,
|
|
"loss": 6.1544,
|
|
"mean_token_accuracy": 0.15117276310920716,
|
|
"num_tokens": 3182860.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.3079229354858395,
|
|
"epoch": 0.16061099567585782,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999736307993221,
|
|
"loss": 6.1532,
|
|
"mean_token_accuracy": 0.14341338276863097,
|
|
"num_tokens": 3193807.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.1183257579803465,
|
|
"epoch": 0.16117257258381534,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999730125630255,
|
|
"loss": 6.0216,
|
|
"mean_token_accuracy": 0.14781626164913178,
|
|
"num_tokens": 3203959.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.274162673950196,
|
|
"epoch": 0.1617341494917729,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999723871635008,
|
|
"loss": 6.039,
|
|
"mean_token_accuracy": 0.15461266785860062,
|
|
"num_tokens": 3215416.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.329146146774292,
|
|
"epoch": 0.16229572639973044,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999717546007681,
|
|
"loss": 6.1911,
|
|
"mean_token_accuracy": 0.1445060230791569,
|
|
"num_tokens": 3226938.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.1372476577758786,
|
|
"epoch": 0.162857303307688,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999711148748476,
|
|
"loss": 5.9971,
|
|
"mean_token_accuracy": 0.1559592068195343,
|
|
"num_tokens": 3238406.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.184646844863892,
|
|
"epoch": 0.16341888021564555,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999704679857594,
|
|
"loss": 6.023,
|
|
"mean_token_accuracy": 0.1487284079194069,
|
|
"num_tokens": 3249695.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.337225914001465,
|
|
"epoch": 0.16398045712360307,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999698139335243,
|
|
"loss": 6.1617,
|
|
"mean_token_accuracy": 0.15224283635616304,
|
|
"num_tokens": 3259656.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.184023857116699,
|
|
"epoch": 0.16454203403156062,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999691527181632,
|
|
"loss": 6.0,
|
|
"mean_token_accuracy": 0.15234700739383697,
|
|
"num_tokens": 3269712.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.068046569824219,
|
|
"epoch": 0.16510361093951817,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999684843396968,
|
|
"loss": 5.9915,
|
|
"mean_token_accuracy": 0.15861952006816865,
|
|
"num_tokens": 3280343.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.2015856266021725,
|
|
"epoch": 0.16566518784747572,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999678087981467,
|
|
"loss": 6.0279,
|
|
"mean_token_accuracy": 0.1526598796248436,
|
|
"num_tokens": 3291649.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.3056028366088865,
|
|
"epoch": 0.16622676475543324,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004999671260935343,
|
|
"loss": 6.2112,
|
|
"mean_token_accuracy": 0.14913778901100158,
|
|
"num_tokens": 3304190.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.126730823516846,
|
|
"epoch": 0.1667883416633908,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999664362258814,
|
|
"loss": 5.9306,
|
|
"mean_token_accuracy": 0.16392721086740494,
|
|
"num_tokens": 3313966.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.21044430732727,
|
|
"epoch": 0.16734991857134834,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999657391952099,
|
|
"loss": 6.1883,
|
|
"mean_token_accuracy": 0.1507849559187889,
|
|
"num_tokens": 3326017.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.236493635177612,
|
|
"epoch": 0.1679114954793059,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000499965035001542,
|
|
"loss": 6.0788,
|
|
"mean_token_accuracy": 0.14554568827152253,
|
|
"num_tokens": 3338264.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.257958507537841,
|
|
"epoch": 0.16847307238726345,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999643236449001,
|
|
"loss": 6.1231,
|
|
"mean_token_accuracy": 0.1576220154762268,
|
|
"num_tokens": 3348901.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.217171573638916,
|
|
"epoch": 0.16903464929522097,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999636051253069,
|
|
"loss": 6.0476,
|
|
"mean_token_accuracy": 0.15184800177812577,
|
|
"num_tokens": 3358441.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.293072080612182,
|
|
"epoch": 0.16959622620317852,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999628794427854,
|
|
"loss": 6.133,
|
|
"mean_token_accuracy": 0.15614369064569472,
|
|
"num_tokens": 3369334.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.208935117721557,
|
|
"epoch": 0.17015780311113607,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999621465973583,
|
|
"loss": 6.1283,
|
|
"mean_token_accuracy": 0.15123554393649102,
|
|
"num_tokens": 3380698.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.426233530044556,
|
|
"epoch": 0.17071938001909362,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999614065890495,
|
|
"loss": 6.1402,
|
|
"mean_token_accuracy": 0.1447037622332573,
|
|
"num_tokens": 3391061.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.1783136367797855,
|
|
"epoch": 0.17128095692705117,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999606594178821,
|
|
"loss": 6.1831,
|
|
"mean_token_accuracy": 0.15091484934091567,
|
|
"num_tokens": 3403983.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.342173147201538,
|
|
"epoch": 0.1718425338350087,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00049995990508388,
|
|
"loss": 6.1231,
|
|
"mean_token_accuracy": 0.15125393867492676,
|
|
"num_tokens": 3413838.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.076597023010254,
|
|
"epoch": 0.17240411074296624,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999591435870674,
|
|
"loss": 5.9419,
|
|
"mean_token_accuracy": 0.15306319296360016,
|
|
"num_tokens": 3423978.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.309637880325317,
|
|
"epoch": 0.1729656876509238,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999583749274684,
|
|
"loss": 6.1192,
|
|
"mean_token_accuracy": 0.15553057491779326,
|
|
"num_tokens": 3433416.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.0907206535339355,
|
|
"epoch": 0.17352726455888134,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999575991051074,
|
|
"loss": 5.97,
|
|
"mean_token_accuracy": 0.15673767775297165,
|
|
"num_tokens": 3444983.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.13786096572876,
|
|
"epoch": 0.1740888414668389,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999568161200093,
|
|
"loss": 5.9552,
|
|
"mean_token_accuracy": 0.16142054796218872,
|
|
"num_tokens": 3455808.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.1685872077941895,
|
|
"epoch": 0.17465041837479642,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999560259721989,
|
|
"loss": 6.1104,
|
|
"mean_token_accuracy": 0.14527395665645598,
|
|
"num_tokens": 3467503.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.331913900375366,
|
|
"epoch": 0.17521199528275397,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999552286617014,
|
|
"loss": 6.2931,
|
|
"mean_token_accuracy": 0.14554888010025024,
|
|
"num_tokens": 3479941.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.151180124282837,
|
|
"epoch": 0.17577357219071152,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999544241885421,
|
|
"loss": 5.9826,
|
|
"mean_token_accuracy": 0.15955932140350343,
|
|
"num_tokens": 3491069.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.212237930297851,
|
|
"epoch": 0.17633514909866907,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999536125527467,
|
|
"loss": 6.1606,
|
|
"mean_token_accuracy": 0.1459153823554516,
|
|
"num_tokens": 3502065.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.338152885437012,
|
|
"epoch": 0.17689672600662662,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000499952793754341,
|
|
"loss": 6.1222,
|
|
"mean_token_accuracy": 0.15344193577766418,
|
|
"num_tokens": 3514228.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.192629480361939,
|
|
"epoch": 0.17745830291458414,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999519677933512,
|
|
"loss": 6.0525,
|
|
"mean_token_accuracy": 0.15235214978456496,
|
|
"num_tokens": 3525801.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.1850494861602785,
|
|
"epoch": 0.1780198798225417,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999511346698035,
|
|
"loss": 5.9837,
|
|
"mean_token_accuracy": 0.1552681341767311,
|
|
"num_tokens": 3536431.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.141394138336182,
|
|
"epoch": 0.17858145673049924,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999502943837244,
|
|
"loss": 6.0957,
|
|
"mean_token_accuracy": 0.15150652602314948,
|
|
"num_tokens": 3549098.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.296343088150024,
|
|
"epoch": 0.1791430336384568,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999494469351406,
|
|
"loss": 6.0648,
|
|
"mean_token_accuracy": 0.14865811467170714,
|
|
"num_tokens": 3560532.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.116127634048462,
|
|
"epoch": 0.17970461054641435,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999485923240793,
|
|
"loss": 6.0503,
|
|
"mean_token_accuracy": 0.15432106032967569,
|
|
"num_tokens": 3572175.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.233989953994751,
|
|
"epoch": 0.18026618745437187,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999477305505675,
|
|
"loss": 5.9729,
|
|
"mean_token_accuracy": 0.17390876933932303,
|
|
"num_tokens": 3582862.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.209008359909058,
|
|
"epoch": 0.18082776436232942,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999468616146328,
|
|
"loss": 6.1604,
|
|
"mean_token_accuracy": 0.1523948520421982,
|
|
"num_tokens": 3594685.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.118362140655518,
|
|
"epoch": 0.18138934127028697,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999459855163026,
|
|
"loss": 5.9363,
|
|
"mean_token_accuracy": 0.1629515752196312,
|
|
"num_tokens": 3604847.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.185347604751587,
|
|
"epoch": 0.18195091817824452,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499945102255605,
|
|
"loss": 6.045,
|
|
"mean_token_accuracy": 0.15654354467988013,
|
|
"num_tokens": 3615549.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.171912479400635,
|
|
"epoch": 0.18251249508620204,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999442118325682,
|
|
"loss": 5.9686,
|
|
"mean_token_accuracy": 0.16261090338230133,
|
|
"num_tokens": 3626300.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.0599151134490965,
|
|
"epoch": 0.1830740719941596,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999433142472205,
|
|
"loss": 5.9836,
|
|
"mean_token_accuracy": 0.15348479002714158,
|
|
"num_tokens": 3637832.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.173618030548096,
|
|
"epoch": 0.18363564890211714,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999424094995902,
|
|
"loss": 6.0249,
|
|
"mean_token_accuracy": 0.1517447590827942,
|
|
"num_tokens": 3649179.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.176636791229248,
|
|
"epoch": 0.1841972258100747,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999414975897066,
|
|
"loss": 6.0489,
|
|
"mean_token_accuracy": 0.15198502838611602,
|
|
"num_tokens": 3660011.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.254629039764405,
|
|
"epoch": 0.18475880271803224,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999405785175983,
|
|
"loss": 6.1367,
|
|
"mean_token_accuracy": 0.14882190227508546,
|
|
"num_tokens": 3672791.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.142079782485962,
|
|
"epoch": 0.18532037962598977,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999396522832949,
|
|
"loss": 5.9599,
|
|
"mean_token_accuracy": 0.15162472128868104,
|
|
"num_tokens": 3684709.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.06971001625061,
|
|
"epoch": 0.18588195653394732,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999387188868256,
|
|
"loss": 5.9376,
|
|
"mean_token_accuracy": 0.1539404273033142,
|
|
"num_tokens": 3694675.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.053702163696289,
|
|
"epoch": 0.18644353344190487,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999377783282203,
|
|
"loss": 5.9411,
|
|
"mean_token_accuracy": 0.154044945538044,
|
|
"num_tokens": 3706384.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.222718238830566,
|
|
"epoch": 0.18700511034986242,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999368306075089,
|
|
"loss": 6.013,
|
|
"mean_token_accuracy": 0.15491660088300704,
|
|
"num_tokens": 3717419.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.1200758934021,
|
|
"epoch": 0.18756668725781997,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999358757247215,
|
|
"loss": 6.0313,
|
|
"mean_token_accuracy": 0.1560240775346756,
|
|
"num_tokens": 3727306.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.299317455291748,
|
|
"epoch": 0.1881282641657775,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999349136798886,
|
|
"loss": 6.1666,
|
|
"mean_token_accuracy": 0.15007805228233337,
|
|
"num_tokens": 3739476.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.097905826568604,
|
|
"epoch": 0.18868984107373504,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999339444730409,
|
|
"loss": 5.9178,
|
|
"mean_token_accuracy": 0.16339074969291686,
|
|
"num_tokens": 3751035.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.131561183929444,
|
|
"epoch": 0.1892514179816926,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999329681042092,
|
|
"loss": 6.0198,
|
|
"mean_token_accuracy": 0.15711974799633027,
|
|
"num_tokens": 3762881.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.139648151397705,
|
|
"epoch": 0.18981299488965014,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999319845734243,
|
|
"loss": 6.0316,
|
|
"mean_token_accuracy": 0.15230293124914168,
|
|
"num_tokens": 3774180.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.116710186004639,
|
|
"epoch": 0.1903745717976077,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999309938807181,
|
|
"loss": 6.0333,
|
|
"mean_token_accuracy": 0.1552741691470146,
|
|
"num_tokens": 3784864.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.22691388130188,
|
|
"epoch": 0.19093614870556522,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999299960261216,
|
|
"loss": 6.0085,
|
|
"mean_token_accuracy": 0.15241112262010575,
|
|
"num_tokens": 3797693.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.1396164894104,
|
|
"epoch": 0.19149772561352277,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999289910096669,
|
|
"loss": 6.0236,
|
|
"mean_token_accuracy": 0.15405184477567674,
|
|
"num_tokens": 3809379.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.053248691558838,
|
|
"epoch": 0.19205930252148032,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999279788313858,
|
|
"loss": 5.8776,
|
|
"mean_token_accuracy": 0.16238332390785218,
|
|
"num_tokens": 3819473.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.033798170089722,
|
|
"epoch": 0.19262087942943787,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999269594913107,
|
|
"loss": 5.8739,
|
|
"mean_token_accuracy": 0.1607612520456314,
|
|
"num_tokens": 3831037.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.053096103668213,
|
|
"epoch": 0.19318245633739542,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999259329894739,
|
|
"loss": 6.0182,
|
|
"mean_token_accuracy": 0.15346318036317824,
|
|
"num_tokens": 3841767.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.032930660247803,
|
|
"epoch": 0.19374403324535294,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999248993259081,
|
|
"loss": 5.8539,
|
|
"mean_token_accuracy": 0.16085348576307296,
|
|
"num_tokens": 3851629.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.0723738193511965,
|
|
"epoch": 0.1943056101533105,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999238585006463,
|
|
"loss": 5.9286,
|
|
"mean_token_accuracy": 0.15617268905043602,
|
|
"num_tokens": 3863247.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.095740032196045,
|
|
"epoch": 0.19486718706126804,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999228105137218,
|
|
"loss": 5.9025,
|
|
"mean_token_accuracy": 0.15338031351566314,
|
|
"num_tokens": 3873949.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.136980056762695,
|
|
"epoch": 0.1954287639692256,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999217553651676,
|
|
"loss": 6.0144,
|
|
"mean_token_accuracy": 0.15692252218723296,
|
|
"num_tokens": 3884992.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.215441608428955,
|
|
"epoch": 0.19599034087718312,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999206930550175,
|
|
"loss": 6.1379,
|
|
"mean_token_accuracy": 0.1581611804664135,
|
|
"num_tokens": 3897029.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.093951034545898,
|
|
"epoch": 0.19655191778514067,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999196235833054,
|
|
"loss": 5.9791,
|
|
"mean_token_accuracy": 0.1530992403626442,
|
|
"num_tokens": 3909510.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.112313222885132,
|
|
"epoch": 0.19711349469309822,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999185469500651,
|
|
"loss": 5.8794,
|
|
"mean_token_accuracy": 0.1585567131638527,
|
|
"num_tokens": 3920593.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.073207998275757,
|
|
"epoch": 0.19767507160105577,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000499917463155331,
|
|
"loss": 6.008,
|
|
"mean_token_accuracy": 0.15076811462640763,
|
|
"num_tokens": 3932215.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.162782859802246,
|
|
"epoch": 0.19823664850901332,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999163721991378,
|
|
"loss": 5.9614,
|
|
"mean_token_accuracy": 0.1571688860654831,
|
|
"num_tokens": 3943876.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.025495481491089,
|
|
"epoch": 0.19879822541697084,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00049991527408152,
|
|
"loss": 5.9544,
|
|
"mean_token_accuracy": 0.15912806242704391,
|
|
"num_tokens": 3954533.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.150787830352783,
|
|
"epoch": 0.1993598023249284,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999141688025127,
|
|
"loss": 6.0102,
|
|
"mean_token_accuracy": 0.1601330891251564,
|
|
"num_tokens": 3965722.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.076108074188232,
|
|
"epoch": 0.19992137923288594,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499913056362151,
|
|
"loss": 5.9457,
|
|
"mean_token_accuracy": 0.15454069674015045,
|
|
"num_tokens": 3976737.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.150193881988526,
|
|
"epoch": 0.2004829561408435,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999119367604703,
|
|
"loss": 6.0181,
|
|
"mean_token_accuracy": 0.1597037598490715,
|
|
"num_tokens": 3986102.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.126420879364014,
|
|
"epoch": 0.20104453304880104,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999108099975064,
|
|
"loss": 5.9904,
|
|
"mean_token_accuracy": 0.15712075233459472,
|
|
"num_tokens": 3996534.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.039429378509522,
|
|
"epoch": 0.20160610995675857,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499909676073295,
|
|
"loss": 5.9055,
|
|
"mean_token_accuracy": 0.1583872377872467,
|
|
"num_tokens": 4006947.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.060782194137573,
|
|
"epoch": 0.20216768686471612,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999085349878723,
|
|
"loss": 5.9449,
|
|
"mean_token_accuracy": 0.16321475505828859,
|
|
"num_tokens": 4017723.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.0788586139678955,
|
|
"epoch": 0.20272926377267367,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999073867412746,
|
|
"loss": 6.0245,
|
|
"mean_token_accuracy": 0.1539995141327381,
|
|
"num_tokens": 4028347.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.088930892944336,
|
|
"epoch": 0.20329084068063122,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999062313335385,
|
|
"loss": 5.8783,
|
|
"mean_token_accuracy": 0.16769046783447267,
|
|
"num_tokens": 4039648.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.099359846115112,
|
|
"epoch": 0.20385241758858877,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999050687647008,
|
|
"loss": 5.9335,
|
|
"mean_token_accuracy": 0.1550333634018898,
|
|
"num_tokens": 4050786.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.12097282409668,
|
|
"epoch": 0.2044139944965463,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999038990347984,
|
|
"loss": 5.9806,
|
|
"mean_token_accuracy": 0.1614360526204109,
|
|
"num_tokens": 4061184.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.204589939117431,
|
|
"epoch": 0.20497557140450384,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999027221438688,
|
|
"loss": 6.1202,
|
|
"mean_token_accuracy": 0.15479563772678376,
|
|
"num_tokens": 4073607.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.022753477096558,
|
|
"epoch": 0.2055371483124614,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499901538091949,
|
|
"loss": 5.9277,
|
|
"mean_token_accuracy": 0.1573045566678047,
|
|
"num_tokens": 4083924.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.086099147796631,
|
|
"epoch": 0.20609872522041894,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999003468790773,
|
|
"loss": 5.9193,
|
|
"mean_token_accuracy": 0.16215644925832748,
|
|
"num_tokens": 4094083.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.078646850585938,
|
|
"epoch": 0.2066603021283765,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004998991485052911,
|
|
"loss": 5.9532,
|
|
"mean_token_accuracy": 0.16525087803602218,
|
|
"num_tokens": 4105124.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.219698572158814,
|
|
"epoch": 0.20722187903633402,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000499897942970629,
|
|
"loss": 6.1184,
|
|
"mean_token_accuracy": 0.15077613070607185,
|
|
"num_tokens": 4116792.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.221433782577515,
|
|
"epoch": 0.20778345594429157,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000499896730275129,
|
|
"loss": 6.096,
|
|
"mean_token_accuracy": 0.15168196707963943,
|
|
"num_tokens": 4127441.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.097982692718506,
|
|
"epoch": 0.20834503285224912,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00049989551041883,
|
|
"loss": 6.0898,
|
|
"mean_token_accuracy": 0.15172615349292756,
|
|
"num_tokens": 4138696.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.1136155128479,
|
|
"epoch": 0.20890660976020667,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998942834017707,
|
|
"loss": 5.9484,
|
|
"mean_token_accuracy": 0.1588933140039444,
|
|
"num_tokens": 4151254.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.051373052597046,
|
|
"epoch": 0.2094681866681642,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998930492239901,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.1542994648218155,
|
|
"num_tokens": 4162046.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.029537487030029,
|
|
"epoch": 0.21002976357612174,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998918078855277,
|
|
"loss": 5.8608,
|
|
"mean_token_accuracy": 0.16153852343559266,
|
|
"num_tokens": 4172409.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.062843751907349,
|
|
"epoch": 0.2105913404840793,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004998905593864227,
|
|
"loss": 6.0243,
|
|
"mean_token_accuracy": 0.16137329041957854,
|
|
"num_tokens": 4182949.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.069054412841797,
|
|
"epoch": 0.21115291739203684,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998893037267153,
|
|
"loss": 5.9882,
|
|
"mean_token_accuracy": 0.1569446697831154,
|
|
"num_tokens": 4194164.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.163977527618409,
|
|
"epoch": 0.2117144942999944,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004998880409064452,
|
|
"loss": 5.9555,
|
|
"mean_token_accuracy": 0.15890365242958068,
|
|
"num_tokens": 4206056.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.018830442428589,
|
|
"epoch": 0.21227607120795192,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998867709256527,
|
|
"loss": 6.0183,
|
|
"mean_token_accuracy": 0.1555065929889679,
|
|
"num_tokens": 4216660.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.058361387252807,
|
|
"epoch": 0.21283764811590947,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000499885493784378,
|
|
"loss": 5.8838,
|
|
"mean_token_accuracy": 0.1596136897802353,
|
|
"num_tokens": 4228495.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.068961715698242,
|
|
"epoch": 0.21339922502386702,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004998842094826621,
|
|
"loss": 5.8992,
|
|
"mean_token_accuracy": 0.15846104472875594,
|
|
"num_tokens": 4238560.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.199825143814087,
|
|
"epoch": 0.21396080193182457,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998829180205458,
|
|
"loss": 6.1062,
|
|
"mean_token_accuracy": 0.1520417958498001,
|
|
"num_tokens": 4250553.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.102573680877685,
|
|
"epoch": 0.21452237883978212,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998816193980701,
|
|
"loss": 6.0304,
|
|
"mean_token_accuracy": 0.1546974703669548,
|
|
"num_tokens": 4262050.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.224831295013428,
|
|
"epoch": 0.21508395574773964,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998803136152764,
|
|
"loss": 5.975,
|
|
"mean_token_accuracy": 0.14990944266319275,
|
|
"num_tokens": 4273019.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.058222246170044,
|
|
"epoch": 0.2156455326556972,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998790006722063,
|
|
"loss": 5.8647,
|
|
"mean_token_accuracy": 0.1674592226743698,
|
|
"num_tokens": 4283585.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.047630691528321,
|
|
"epoch": 0.21620710956365474,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998776805689016,
|
|
"loss": 5.937,
|
|
"mean_token_accuracy": 0.16005693525075912,
|
|
"num_tokens": 4294905.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.1609485149383545,
|
|
"epoch": 0.2167686864716123,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998763533054045,
|
|
"loss": 6.0415,
|
|
"mean_token_accuracy": 0.15440894961357116,
|
|
"num_tokens": 4305926.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.985680294036865,
|
|
"epoch": 0.21733026337956984,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998750188817568,
|
|
"loss": 6.0186,
|
|
"mean_token_accuracy": 0.15445369929075242,
|
|
"num_tokens": 4318297.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.032767105102539,
|
|
"epoch": 0.21789184028752737,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998736772980015,
|
|
"loss": 5.84,
|
|
"mean_token_accuracy": 0.16766211837530137,
|
|
"num_tokens": 4328795.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.012192440032959,
|
|
"epoch": 0.21845341719548492,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000499872328554181,
|
|
"loss": 5.8628,
|
|
"mean_token_accuracy": 0.15800613760948182,
|
|
"num_tokens": 4340279.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.122261905670166,
|
|
"epoch": 0.21901499410344247,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998709726503383,
|
|
"loss": 6.0275,
|
|
"mean_token_accuracy": 0.15114919617772102,
|
|
"num_tokens": 4352157.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.07682991027832,
|
|
"epoch": 0.21957657101140002,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998696095865168,
|
|
"loss": 5.8643,
|
|
"mean_token_accuracy": 0.16682649552822112,
|
|
"num_tokens": 4361909.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.879463386535645,
|
|
"epoch": 0.22013814791935757,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998682393627595,
|
|
"loss": 5.7878,
|
|
"mean_token_accuracy": 0.1570300579071045,
|
|
"num_tokens": 4372177.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.987885189056397,
|
|
"epoch": 0.2206997248273151,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998668619791104,
|
|
"loss": 5.8391,
|
|
"mean_token_accuracy": 0.16166026890277863,
|
|
"num_tokens": 4382872.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.05077543258667,
|
|
"epoch": 0.22126130173527264,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998654774356131,
|
|
"loss": 5.9305,
|
|
"mean_token_accuracy": 0.15829528868198395,
|
|
"num_tokens": 4393623.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.013175201416016,
|
|
"epoch": 0.2218228786432302,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004998640857323118,
|
|
"loss": 5.8682,
|
|
"mean_token_accuracy": 0.16012260168790818,
|
|
"num_tokens": 4404215.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.117317819595337,
|
|
"epoch": 0.22238445555118774,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998626868692509,
|
|
"loss": 6.1271,
|
|
"mean_token_accuracy": 0.156798305362463,
|
|
"num_tokens": 4415978.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.133135509490967,
|
|
"epoch": 0.2229460324591453,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998612808464748,
|
|
"loss": 5.8725,
|
|
"mean_token_accuracy": 0.16460159718990325,
|
|
"num_tokens": 4427060.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.928405904769898,
|
|
"epoch": 0.22350760936710282,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004998598676640283,
|
|
"loss": 5.9522,
|
|
"mean_token_accuracy": 0.15562185198068618,
|
|
"num_tokens": 4438145.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.232151079177856,
|
|
"epoch": 0.22406918627506037,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004998584473219563,
|
|
"loss": 6.0299,
|
|
"mean_token_accuracy": 0.15950280129909516,
|
|
"num_tokens": 4449324.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.970883798599243,
|
|
"epoch": 0.22463076318301792,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998570198203043,
|
|
"loss": 5.7709,
|
|
"mean_token_accuracy": 0.16454007625579833,
|
|
"num_tokens": 4459728.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.013013935089111,
|
|
"epoch": 0.22519234009097547,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998555851591175,
|
|
"loss": 5.9703,
|
|
"mean_token_accuracy": 0.1580568253993988,
|
|
"num_tokens": 4471164.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.116335105895996,
|
|
"epoch": 0.225753916998933,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004998541433384416,
|
|
"loss": 6.0193,
|
|
"mean_token_accuracy": 0.1572705999016762,
|
|
"num_tokens": 4483349.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.119635820388794,
|
|
"epoch": 0.22631549390689054,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998526943583226,
|
|
"loss": 5.986,
|
|
"mean_token_accuracy": 0.15683266669511794,
|
|
"num_tokens": 4495311.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.996360874176025,
|
|
"epoch": 0.2268770708148481,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004998512382188066,
|
|
"loss": 5.9179,
|
|
"mean_token_accuracy": 0.15888708531856538,
|
|
"num_tokens": 4507017.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.032978343963623,
|
|
"epoch": 0.22743864772280564,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998497749199401,
|
|
"loss": 5.9484,
|
|
"mean_token_accuracy": 0.15785446017980576,
|
|
"num_tokens": 4517746.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.110689783096314,
|
|
"epoch": 0.2280002246307632,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998483044617694,
|
|
"loss": 6.0512,
|
|
"mean_token_accuracy": 0.14857769906520843,
|
|
"num_tokens": 4529149.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.109778976440429,
|
|
"epoch": 0.22856180153872072,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998468268443416,
|
|
"loss": 5.9216,
|
|
"mean_token_accuracy": 0.15968868732452393,
|
|
"num_tokens": 4540630.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.904194164276123,
|
|
"epoch": 0.22912337844667827,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998453420677034,
|
|
"loss": 5.8583,
|
|
"mean_token_accuracy": 0.1592124655842781,
|
|
"num_tokens": 4552877.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.08227162361145,
|
|
"epoch": 0.22968495535463582,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998438501319025,
|
|
"loss": 5.953,
|
|
"mean_token_accuracy": 0.16537819653749466,
|
|
"num_tokens": 4563634.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.079555511474609,
|
|
"epoch": 0.23024653226259337,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998423510369862,
|
|
"loss": 5.8719,
|
|
"mean_token_accuracy": 0.16541080325841903,
|
|
"num_tokens": 4574007.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.959904050827026,
|
|
"epoch": 0.23080810917055092,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998408447830022,
|
|
"loss": 5.8867,
|
|
"mean_token_accuracy": 0.1608034625649452,
|
|
"num_tokens": 4584146.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.00870270729065,
|
|
"epoch": 0.23136968607850844,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998393313699985,
|
|
"loss": 5.9059,
|
|
"mean_token_accuracy": 0.1573877900838852,
|
|
"num_tokens": 4594839.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.191771459579468,
|
|
"epoch": 0.231931262986466,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004998378107980233,
|
|
"loss": 6.0678,
|
|
"mean_token_accuracy": 0.1526424765586853,
|
|
"num_tokens": 4608229.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.032376861572265,
|
|
"epoch": 0.23249283989442354,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998362830671249,
|
|
"loss": 5.8767,
|
|
"mean_token_accuracy": 0.16054777950048446,
|
|
"num_tokens": 4619206.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.951002788543701,
|
|
"epoch": 0.2330544168023811,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998347481773522,
|
|
"loss": 5.9053,
|
|
"mean_token_accuracy": 0.16458992809057235,
|
|
"num_tokens": 4628966.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.079168605804443,
|
|
"epoch": 0.23361599371033864,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998332061287538,
|
|
"loss": 5.7883,
|
|
"mean_token_accuracy": 0.1636739507317543,
|
|
"num_tokens": 4639935.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.051212406158447,
|
|
"epoch": 0.23417757061829617,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499831656921379,
|
|
"loss": 6.0266,
|
|
"mean_token_accuracy": 0.15532584488391876,
|
|
"num_tokens": 4651998.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.094482707977295,
|
|
"epoch": 0.23473914752625372,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499830100555277,
|
|
"loss": 6.0291,
|
|
"mean_token_accuracy": 0.15595776289701463,
|
|
"num_tokens": 4664002.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.037249612808227,
|
|
"epoch": 0.23530072443421127,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004998285370304974,
|
|
"loss": 5.8896,
|
|
"mean_token_accuracy": 0.16437548100948335,
|
|
"num_tokens": 4675343.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.085287809371948,
|
|
"epoch": 0.23586230134216882,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00049982696634709,
|
|
"loss": 5.9351,
|
|
"mean_token_accuracy": 0.16324877589941025,
|
|
"num_tokens": 4686410.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.030001497268676,
|
|
"epoch": 0.23642387825012637,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004998253885051048,
|
|
"loss": 5.9697,
|
|
"mean_token_accuracy": 0.16037132889032363,
|
|
"num_tokens": 4697768.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.1078695297241214,
|
|
"epoch": 0.2369854551580839,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499823803504592,
|
|
"loss": 5.9731,
|
|
"mean_token_accuracy": 0.1577034831047058,
|
|
"num_tokens": 4709933.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.090886545181275,
|
|
"epoch": 0.23754703206604144,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998222113456022,
|
|
"loss": 6.014,
|
|
"mean_token_accuracy": 0.15369939506053926,
|
|
"num_tokens": 4721385.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.996338558197022,
|
|
"epoch": 0.238108608973999,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004998206120281858,
|
|
"loss": 5.8288,
|
|
"mean_token_accuracy": 0.16938644796609878,
|
|
"num_tokens": 4732562.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.999813508987427,
|
|
"epoch": 0.23867018588195654,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499819005552394,
|
|
"loss": 5.8352,
|
|
"mean_token_accuracy": 0.16146625876426696,
|
|
"num_tokens": 4742794.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.974365329742431,
|
|
"epoch": 0.23923176278991407,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998173919182779,
|
|
"loss": 5.814,
|
|
"mean_token_accuracy": 0.16607148945331573,
|
|
"num_tokens": 4754903.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.948705959320068,
|
|
"epoch": 0.23979333969787162,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000499815771125889,
|
|
"loss": 5.8377,
|
|
"mean_token_accuracy": 0.16147608906030655,
|
|
"num_tokens": 4766165.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.005882406234742,
|
|
"epoch": 0.24035491660582917,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998141431752785,
|
|
"loss": 5.8587,
|
|
"mean_token_accuracy": 0.16653385609388352,
|
|
"num_tokens": 4776551.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.128078126907349,
|
|
"epoch": 0.24091649351378672,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004998125080664987,
|
|
"loss": 6.0525,
|
|
"mean_token_accuracy": 0.1545305222272873,
|
|
"num_tokens": 4788344.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.060748529434204,
|
|
"epoch": 0.24147807042174427,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998108657996012,
|
|
"loss": 5.8475,
|
|
"mean_token_accuracy": 0.15840872079133989,
|
|
"num_tokens": 4799314.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.97811918258667,
|
|
"epoch": 0.2420396473297018,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998092163746387,
|
|
"loss": 5.8964,
|
|
"mean_token_accuracy": 0.16318009197711944,
|
|
"num_tokens": 4809588.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.96418137550354,
|
|
"epoch": 0.24260122423765934,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998075597916636,
|
|
"loss": 5.7132,
|
|
"mean_token_accuracy": 0.1681472808122635,
|
|
"num_tokens": 4820076.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.002331733703613,
|
|
"epoch": 0.2431628011456169,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998058960507286,
|
|
"loss": 5.992,
|
|
"mean_token_accuracy": 0.15780160278081895,
|
|
"num_tokens": 4831504.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.020361518859863,
|
|
"epoch": 0.24372437805357444,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998042251518866,
|
|
"loss": 5.8343,
|
|
"mean_token_accuracy": 0.16251724809408188,
|
|
"num_tokens": 4842826.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.909654855728149,
|
|
"epoch": 0.244285954961532,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998025470951908,
|
|
"loss": 5.824,
|
|
"mean_token_accuracy": 0.16593606919050216,
|
|
"num_tokens": 4853279.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.0372607707977295,
|
|
"epoch": 0.24484753186948952,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998008618806949,
|
|
"loss": 5.8828,
|
|
"mean_token_accuracy": 0.16533545404672623,
|
|
"num_tokens": 4863332.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.999047374725341,
|
|
"epoch": 0.24540910877744707,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997991695084523,
|
|
"loss": 6.0049,
|
|
"mean_token_accuracy": 0.15240172892808915,
|
|
"num_tokens": 4874217.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.115704011917114,
|
|
"epoch": 0.24597068568540462,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997974699785169,
|
|
"loss": 5.8562,
|
|
"mean_token_accuracy": 0.1665976122021675,
|
|
"num_tokens": 4885322.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.871498727798462,
|
|
"epoch": 0.24653226259336217,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997957632909429,
|
|
"loss": 5.8132,
|
|
"mean_token_accuracy": 0.16305921524763106,
|
|
"num_tokens": 4895546.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.153412675857544,
|
|
"epoch": 0.24709383950131972,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997940494457846,
|
|
"loss": 6.0032,
|
|
"mean_token_accuracy": 0.15596965700387955,
|
|
"num_tokens": 4906957.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.124114465713501,
|
|
"epoch": 0.24765541640927724,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997923284430967,
|
|
"loss": 5.9452,
|
|
"mean_token_accuracy": 0.1516578808426857,
|
|
"num_tokens": 4917965.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.93087248802185,
|
|
"epoch": 0.2482169933172348,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997906002829338,
|
|
"loss": 5.8238,
|
|
"mean_token_accuracy": 0.1671748161315918,
|
|
"num_tokens": 4928585.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.9913819313049315,
|
|
"epoch": 0.24877857022519234,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000499788864965351,
|
|
"loss": 5.833,
|
|
"mean_token_accuracy": 0.16251434683799743,
|
|
"num_tokens": 4939818.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.075236415863037,
|
|
"epoch": 0.2493401471331499,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997871224904036,
|
|
"loss": 5.9279,
|
|
"mean_token_accuracy": 0.160538013279438,
|
|
"num_tokens": 4951752.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.8859203338623045,
|
|
"epoch": 0.24990172404110744,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997853728581468,
|
|
"loss": 5.7528,
|
|
"mean_token_accuracy": 0.1742950603365898,
|
|
"num_tokens": 4961846.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.972510623931885,
|
|
"epoch": 0.250463300949065,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004997836160686368,
|
|
"loss": 5.9287,
|
|
"mean_token_accuracy": 0.15616793632507325,
|
|
"num_tokens": 4973693.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.999047088623047,
|
|
"epoch": 0.25102487785702254,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997818521219291,
|
|
"loss": 5.8239,
|
|
"mean_token_accuracy": 0.16658675223588942,
|
|
"num_tokens": 4986156.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.945030450820923,
|
|
"epoch": 0.25158645476498004,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004997800810180803,
|
|
"loss": 5.9059,
|
|
"mean_token_accuracy": 0.16279772520065308,
|
|
"num_tokens": 4997615.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.058555555343628,
|
|
"epoch": 0.2521480316729376,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997783027571464,
|
|
"loss": 5.8432,
|
|
"mean_token_accuracy": 0.16085562109947205,
|
|
"num_tokens": 5007655.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.952951717376709,
|
|
"epoch": 0.25270960858089514,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997765173391841,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.1652421697974205,
|
|
"num_tokens": 5018933.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.1129766464233395,
|
|
"epoch": 0.2532711854888527,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997747247642505,
|
|
"loss": 5.9081,
|
|
"mean_token_accuracy": 0.16212444603443146,
|
|
"num_tokens": 5030495.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.000020503997803,
|
|
"epoch": 0.25383276239681024,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004997729250324023,
|
|
"loss": 5.96,
|
|
"mean_token_accuracy": 0.15744963735342027,
|
|
"num_tokens": 5041553.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.949009561538697,
|
|
"epoch": 0.2543943393047678,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997711181436971,
|
|
"loss": 5.8867,
|
|
"mean_token_accuracy": 0.16389947086572648,
|
|
"num_tokens": 5052312.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.130438995361328,
|
|
"epoch": 0.25495591621272534,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004997693040981922,
|
|
"loss": 6.0774,
|
|
"mean_token_accuracy": 0.15607239678502083,
|
|
"num_tokens": 5063445.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.100005340576172,
|
|
"epoch": 0.2555174931206829,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997674828959456,
|
|
"loss": 5.8637,
|
|
"mean_token_accuracy": 0.1615953415632248,
|
|
"num_tokens": 5073052.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.964355039596557,
|
|
"epoch": 0.25607907002864044,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997656545370151,
|
|
"loss": 5.8996,
|
|
"mean_token_accuracy": 0.162807098031044,
|
|
"num_tokens": 5084174.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.064599895477295,
|
|
"epoch": 0.256640646936598,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499763819021459,
|
|
"loss": 5.8753,
|
|
"mean_token_accuracy": 0.162973290681839,
|
|
"num_tokens": 5094106.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.088547229766846,
|
|
"epoch": 0.2572022238445555,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004997619763493357,
|
|
"loss": 5.9606,
|
|
"mean_token_accuracy": 0.16254740357398986,
|
|
"num_tokens": 5106963.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.932668828964234,
|
|
"epoch": 0.25776380075251304,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000499760126520704,
|
|
"loss": 5.7586,
|
|
"mean_token_accuracy": 0.15899893045425414,
|
|
"num_tokens": 5119476.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.0435731410980225,
|
|
"epoch": 0.2583253776604706,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997582695356225,
|
|
"loss": 6.0162,
|
|
"mean_token_accuracy": 0.15558764040470124,
|
|
"num_tokens": 5132169.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.967125463485718,
|
|
"epoch": 0.25888695456842814,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997564053941507,
|
|
"loss": 5.8449,
|
|
"mean_token_accuracy": 0.1600498989224434,
|
|
"num_tokens": 5143354.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.956926536560059,
|
|
"epoch": 0.2594485314763857,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997545340963478,
|
|
"loss": 5.9234,
|
|
"mean_token_accuracy": 0.15473302602767944,
|
|
"num_tokens": 5154624.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.064026832580566,
|
|
"epoch": 0.26001010838434324,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004997526556422733,
|
|
"loss": 5.805,
|
|
"mean_token_accuracy": 0.1624491482973099,
|
|
"num_tokens": 5165349.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.973732566833496,
|
|
"epoch": 0.2605716852923008,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997507700319872,
|
|
"loss": 5.8869,
|
|
"mean_token_accuracy": 0.16123015880584718,
|
|
"num_tokens": 5175874.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.963538599014282,
|
|
"epoch": 0.26113326220025834,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004997488772655492,
|
|
"loss": 5.7897,
|
|
"mean_token_accuracy": 0.16631451696157457,
|
|
"num_tokens": 5186591.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.951702833175659,
|
|
"epoch": 0.2616948391082159,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997469773430199,
|
|
"loss": 5.8832,
|
|
"mean_token_accuracy": 0.16798330396413802,
|
|
"num_tokens": 5197307.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.961882829666138,
|
|
"epoch": 0.2622564160161734,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004997450702644596,
|
|
"loss": 5.8451,
|
|
"mean_token_accuracy": 0.16027480661869048,
|
|
"num_tokens": 5209868.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.992185497283936,
|
|
"epoch": 0.26281799292413094,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997431560299292,
|
|
"loss": 5.8627,
|
|
"mean_token_accuracy": 0.1646285265684128,
|
|
"num_tokens": 5220365.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.977267789840698,
|
|
"epoch": 0.2633795698320885,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004997412346394894,
|
|
"loss": 5.7728,
|
|
"mean_token_accuracy": 0.1643584579229355,
|
|
"num_tokens": 5231575.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.980089855194092,
|
|
"epoch": 0.26394114674004604,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997393060932017,
|
|
"loss": 5.8646,
|
|
"mean_token_accuracy": 0.16169388443231583,
|
|
"num_tokens": 5241852.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.0129913806915285,
|
|
"epoch": 0.2645027236480036,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004997373703911273,
|
|
"loss": 5.9098,
|
|
"mean_token_accuracy": 0.16636008620262147,
|
|
"num_tokens": 5254536.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.019440126419068,
|
|
"epoch": 0.26506430055596114,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997354275333277,
|
|
"loss": 5.8246,
|
|
"mean_token_accuracy": 0.16354746818542482,
|
|
"num_tokens": 5264946.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.750850582122803,
|
|
"epoch": 0.2656258774639187,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000499733477519865,
|
|
"loss": 5.6565,
|
|
"mean_token_accuracy": 0.16620689481496811,
|
|
"num_tokens": 5275527.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.940898036956787,
|
|
"epoch": 0.26618745437187624,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997315203508013,
|
|
"loss": 5.8218,
|
|
"mean_token_accuracy": 0.16179233640432358,
|
|
"num_tokens": 5285471.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.861508893966675,
|
|
"epoch": 0.2667490312798338,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997295560261988,
|
|
"loss": 5.7549,
|
|
"mean_token_accuracy": 0.1623705729842186,
|
|
"num_tokens": 5295743.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.0062055587768555,
|
|
"epoch": 0.26731060818779134,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00049972758454612,
|
|
"loss": 5.8892,
|
|
"mean_token_accuracy": 0.15389246344566346,
|
|
"num_tokens": 5307963.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.945818948745727,
|
|
"epoch": 0.26787218509574884,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997256059106279,
|
|
"loss": 5.7893,
|
|
"mean_token_accuracy": 0.16452015787363053,
|
|
"num_tokens": 5318394.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.984850311279297,
|
|
"epoch": 0.2684337620037064,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004997236201197853,
|
|
"loss": 6.0264,
|
|
"mean_token_accuracy": 0.1485483929514885,
|
|
"num_tokens": 5331728.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.003773021697998,
|
|
"epoch": 0.26899533891166394,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997216271736554,
|
|
"loss": 5.8979,
|
|
"mean_token_accuracy": 0.17206651121377944,
|
|
"num_tokens": 5342825.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.011975288391113,
|
|
"epoch": 0.2695569158196215,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004997196270723018,
|
|
"loss": 5.8037,
|
|
"mean_token_accuracy": 0.16357291787862777,
|
|
"num_tokens": 5354232.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.938725376129151,
|
|
"epoch": 0.27011849272757904,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997176198157881,
|
|
"loss": 5.8948,
|
|
"mean_token_accuracy": 0.15622272118926048,
|
|
"num_tokens": 5364807.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.948577547073365,
|
|
"epoch": 0.2706800696355366,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997156054041781,
|
|
"loss": 5.8751,
|
|
"mean_token_accuracy": 0.15974455922842026,
|
|
"num_tokens": 5375593.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.011623287200928,
|
|
"epoch": 0.27124164654349414,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997135838375362,
|
|
"loss": 5.8221,
|
|
"mean_token_accuracy": 0.16726389527320862,
|
|
"num_tokens": 5386740.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.8543072700500485,
|
|
"epoch": 0.2718032234514517,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997115551159267,
|
|
"loss": 5.7997,
|
|
"mean_token_accuracy": 0.16655707061290742,
|
|
"num_tokens": 5397732.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.040705680847168,
|
|
"epoch": 0.27236480035940924,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000499709519239414,
|
|
"loss": 5.87,
|
|
"mean_token_accuracy": 0.15803890973329543,
|
|
"num_tokens": 5409392.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.810342502593994,
|
|
"epoch": 0.2729263772673668,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004997074762080632,
|
|
"loss": 5.8158,
|
|
"mean_token_accuracy": 0.17186738848686217,
|
|
"num_tokens": 5420903.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.911894273757935,
|
|
"epoch": 0.2734879541753243,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000499705426021939,
|
|
"loss": 5.761,
|
|
"mean_token_accuracy": 0.16358311474323273,
|
|
"num_tokens": 5432631.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.893311023712158,
|
|
"epoch": 0.27404953108328184,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000499703368681107,
|
|
"loss": 5.7113,
|
|
"mean_token_accuracy": 0.17483898848295212,
|
|
"num_tokens": 5443463.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.013640213012695,
|
|
"epoch": 0.2746111079912394,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004997013041856326,
|
|
"loss": 6.0427,
|
|
"mean_token_accuracy": 0.15905722007155418,
|
|
"num_tokens": 5455903.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.00086669921875,
|
|
"epoch": 0.27517268489919694,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004996992325355815,
|
|
"loss": 5.8598,
|
|
"mean_token_accuracy": 0.16078447848558425,
|
|
"num_tokens": 5467201.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.952702760696411,
|
|
"epoch": 0.2757342618071545,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004996971537310198,
|
|
"loss": 5.7355,
|
|
"mean_token_accuracy": 0.16615803241729737,
|
|
"num_tokens": 5478109.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.957705736160278,
|
|
"epoch": 0.27629583871511204,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004996950677720133,
|
|
"loss": 5.8551,
|
|
"mean_token_accuracy": 0.16423224210739135,
|
|
"num_tokens": 5489621.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.93759388923645,
|
|
"epoch": 0.2768574156230696,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499692974658629,
|
|
"loss": 5.815,
|
|
"mean_token_accuracy": 0.16715700626373292,
|
|
"num_tokens": 5500606.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.0008186340332035,
|
|
"epoch": 0.27741899253102714,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996908743909331,
|
|
"loss": 5.7597,
|
|
"mean_token_accuracy": 0.1725642591714859,
|
|
"num_tokens": 5510768.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.799460363388062,
|
|
"epoch": 0.2779805694389847,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004996887669689926,
|
|
"loss": 5.781,
|
|
"mean_token_accuracy": 0.16516642272472382,
|
|
"num_tokens": 5522210.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.052443265914917,
|
|
"epoch": 0.2785421463469422,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004996866523928747,
|
|
"loss": 5.8989,
|
|
"mean_token_accuracy": 0.1589887335896492,
|
|
"num_tokens": 5533055.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.951822566986084,
|
|
"epoch": 0.27910372325489974,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004996845306626467,
|
|
"loss": 5.9103,
|
|
"mean_token_accuracy": 0.1590309724211693,
|
|
"num_tokens": 5544345.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.969441175460815,
|
|
"epoch": 0.2796653001628573,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004996824017783759,
|
|
"loss": 5.8232,
|
|
"mean_token_accuracy": 0.16280512511730194,
|
|
"num_tokens": 5556734.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.991024303436279,
|
|
"epoch": 0.28022687707081484,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996802657401304,
|
|
"loss": 5.8603,
|
|
"mean_token_accuracy": 0.162888403236866,
|
|
"num_tokens": 5566874.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.989664602279663,
|
|
"epoch": 0.2807884539787724,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004996781225479781,
|
|
"loss": 5.8781,
|
|
"mean_token_accuracy": 0.1601580023765564,
|
|
"num_tokens": 5578812.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.951496219635009,
|
|
"epoch": 0.28135003088672994,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996759722019872,
|
|
"loss": 5.9058,
|
|
"mean_token_accuracy": 0.16278692781925203,
|
|
"num_tokens": 5590021.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.0353498458862305,
|
|
"epoch": 0.2819116077946875,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004996738147022262,
|
|
"loss": 5.9117,
|
|
"mean_token_accuracy": 0.1564626947045326,
|
|
"num_tokens": 5600921.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.905371189117432,
|
|
"epoch": 0.28247318470264504,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996716500487638,
|
|
"loss": 5.825,
|
|
"mean_token_accuracy": 0.16611457020044326,
|
|
"num_tokens": 5610615.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.86320858001709,
|
|
"epoch": 0.2830347616106026,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000499669478241669,
|
|
"loss": 5.7587,
|
|
"mean_token_accuracy": 0.17136051654815673,
|
|
"num_tokens": 5621596.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.044602966308593,
|
|
"epoch": 0.28359633851856014,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996672992810108,
|
|
"loss": 5.8193,
|
|
"mean_token_accuracy": 0.16327339112758638,
|
|
"num_tokens": 5633047.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.8826415061950685,
|
|
"epoch": 0.28415791542651764,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004996651131668587,
|
|
"loss": 5.7823,
|
|
"mean_token_accuracy": 0.1643436849117279,
|
|
"num_tokens": 5644970.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.896431255340576,
|
|
"epoch": 0.2847194923344752,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004996629198992823,
|
|
"loss": 5.9127,
|
|
"mean_token_accuracy": 0.15781734585762025,
|
|
"num_tokens": 5656631.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.949198150634766,
|
|
"epoch": 0.28528106924243274,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004996607194783513,
|
|
"loss": 5.8638,
|
|
"mean_token_accuracy": 0.15979290157556533,
|
|
"num_tokens": 5668808.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.890338563919068,
|
|
"epoch": 0.2858426461503903,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004996585119041359,
|
|
"loss": 5.7855,
|
|
"mean_token_accuracy": 0.16485803872346877,
|
|
"num_tokens": 5679493.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.0119789123535154,
|
|
"epoch": 0.28640422305834784,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996562971767063,
|
|
"loss": 5.8199,
|
|
"mean_token_accuracy": 0.1721026435494423,
|
|
"num_tokens": 5689879.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.97424168586731,
|
|
"epoch": 0.2869657999663054,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000499654075296133,
|
|
"loss": 5.9128,
|
|
"mean_token_accuracy": 0.16606319844722747,
|
|
"num_tokens": 5699934.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.907690334320068,
|
|
"epoch": 0.28752737687426294,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499651846262487,
|
|
"loss": 5.8203,
|
|
"mean_token_accuracy": 0.16842011064291001,
|
|
"num_tokens": 5710171.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.016073274612427,
|
|
"epoch": 0.2880889537822205,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996496100758389,
|
|
"loss": 5.8119,
|
|
"mean_token_accuracy": 0.15843043178319932,
|
|
"num_tokens": 5721190.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.969920349121094,
|
|
"epoch": 0.28865053069017804,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996473667362604,
|
|
"loss": 5.9343,
|
|
"mean_token_accuracy": 0.15901276618242263,
|
|
"num_tokens": 5733706.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.938461780548096,
|
|
"epoch": 0.28921210759813554,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004996451162438224,
|
|
"loss": 5.89,
|
|
"mean_token_accuracy": 0.1645979881286621,
|
|
"num_tokens": 5745712.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.821240472793579,
|
|
"epoch": 0.2897736845060931,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996428585985967,
|
|
"loss": 5.7068,
|
|
"mean_token_accuracy": 0.17919964641332625,
|
|
"num_tokens": 5757915.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.995480871200561,
|
|
"epoch": 0.29033526141405064,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004996405938006553,
|
|
"loss": 5.8531,
|
|
"mean_token_accuracy": 0.16649403274059296,
|
|
"num_tokens": 5768788.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.961299180984497,
|
|
"epoch": 0.2908968383220082,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996383218500703,
|
|
"loss": 5.7885,
|
|
"mean_token_accuracy": 0.16060467511415483,
|
|
"num_tokens": 5779806.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.804096269607544,
|
|
"epoch": 0.29145841522996574,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499636042746914,
|
|
"loss": 5.7532,
|
|
"mean_token_accuracy": 0.1604410633444786,
|
|
"num_tokens": 5790177.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.949581050872803,
|
|
"epoch": 0.2920199921379233,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499633756491259,
|
|
"loss": 5.7803,
|
|
"mean_token_accuracy": 0.17098623216152192,
|
|
"num_tokens": 5800251.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.966438150405883,
|
|
"epoch": 0.29258156904588084,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004996314630831781,
|
|
"loss": 5.8654,
|
|
"mean_token_accuracy": 0.16320379972457885,
|
|
"num_tokens": 5812210.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.956333351135254,
|
|
"epoch": 0.2931431459538384,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004996291625227443,
|
|
"loss": 5.8503,
|
|
"mean_token_accuracy": 0.16726463437080383,
|
|
"num_tokens": 5823280.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.0089202404022215,
|
|
"epoch": 0.29370472286179594,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004996268548100307,
|
|
"loss": 5.9115,
|
|
"mean_token_accuracy": 0.1580904632806778,
|
|
"num_tokens": 5835076.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.815197277069092,
|
|
"epoch": 0.2942662997697535,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000499624539945111,
|
|
"loss": 5.7054,
|
|
"mean_token_accuracy": 0.17123110741376876,
|
|
"num_tokens": 5845376.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.862633275985718,
|
|
"epoch": 0.294827876677711,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004996222179280589,
|
|
"loss": 5.7421,
|
|
"mean_token_accuracy": 0.16225717961788177,
|
|
"num_tokens": 5857124.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.990662527084351,
|
|
"epoch": 0.29538945358566854,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004996198887589481,
|
|
"loss": 5.8554,
|
|
"mean_token_accuracy": 0.15759045258164406,
|
|
"num_tokens": 5868681.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.914901399612427,
|
|
"epoch": 0.2959510304936261,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004996175524378531,
|
|
"loss": 5.838,
|
|
"mean_token_accuracy": 0.16877058297395706,
|
|
"num_tokens": 5880532.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.950133657455444,
|
|
"epoch": 0.29651260740158364,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499615208964848,
|
|
"loss": 5.7916,
|
|
"mean_token_accuracy": 0.1670111671090126,
|
|
"num_tokens": 5891582.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.908686256408691,
|
|
"epoch": 0.2970741843095412,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004996128583400077,
|
|
"loss": 5.7954,
|
|
"mean_token_accuracy": 0.16432597637176513,
|
|
"num_tokens": 5903543.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.854915571212769,
|
|
"epoch": 0.29763576121749874,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996105005634066,
|
|
"loss": 5.7566,
|
|
"mean_token_accuracy": 0.17550766468048096,
|
|
"num_tokens": 5913140.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.953357362747193,
|
|
"epoch": 0.2981973381254563,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004996081356351202,
|
|
"loss": 5.7344,
|
|
"mean_token_accuracy": 0.16562701612710953,
|
|
"num_tokens": 5924010.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.931583070755005,
|
|
"epoch": 0.29875891503341384,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996057635552237,
|
|
"loss": 5.9304,
|
|
"mean_token_accuracy": 0.16155474483966828,
|
|
"num_tokens": 5936246.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.950146579742432,
|
|
"epoch": 0.2993204919413714,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004996033843237925,
|
|
"loss": 5.8149,
|
|
"mean_token_accuracy": 0.1572188138961792,
|
|
"num_tokens": 5947594.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.957934617996216,
|
|
"epoch": 0.29988206884932894,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004996009979409025,
|
|
"loss": 5.7606,
|
|
"mean_token_accuracy": 0.16158466041088104,
|
|
"num_tokens": 5957660.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.738403224945069,
|
|
"epoch": 0.30044364575728644,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004995986044066295,
|
|
"loss": 5.5848,
|
|
"mean_token_accuracy": 0.17791350483894347,
|
|
"num_tokens": 5968881.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.894114017486572,
|
|
"epoch": 0.301005222665244,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00049959620372105,
|
|
"loss": 5.7614,
|
|
"mean_token_accuracy": 0.16772934794425964,
|
|
"num_tokens": 5980032.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.77871789932251,
|
|
"epoch": 0.30156679957320154,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004995937958842401,
|
|
"loss": 5.7557,
|
|
"mean_token_accuracy": 0.16537490487098694,
|
|
"num_tokens": 5991781.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.914988374710083,
|
|
"epoch": 0.3021283764811591,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004995913808962767,
|
|
"loss": 5.7879,
|
|
"mean_token_accuracy": 0.1649462789297104,
|
|
"num_tokens": 6002303.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.009264516830444,
|
|
"epoch": 0.30268995338911664,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995889587572366,
|
|
"loss": 5.8814,
|
|
"mean_token_accuracy": 0.1677633598446846,
|
|
"num_tokens": 6011523.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.906981372833252,
|
|
"epoch": 0.3032515302970742,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000499586529467197,
|
|
"loss": 5.6844,
|
|
"mean_token_accuracy": 0.16914817541837693,
|
|
"num_tokens": 6022169.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.74225082397461,
|
|
"epoch": 0.30381310720503174,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004995840930262351,
|
|
"loss": 5.7428,
|
|
"mean_token_accuracy": 0.16678209528326987,
|
|
"num_tokens": 6033393.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.971656274795532,
|
|
"epoch": 0.3043746841129893,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004995816494344287,
|
|
"loss": 5.8107,
|
|
"mean_token_accuracy": 0.15811437368392944,
|
|
"num_tokens": 6045365.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.846801328659057,
|
|
"epoch": 0.30493626102094684,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004995791986918554,
|
|
"loss": 5.8361,
|
|
"mean_token_accuracy": 0.1662740647792816,
|
|
"num_tokens": 6057482.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.941861152648926,
|
|
"epoch": 0.30549783792890434,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004995767407985933,
|
|
"loss": 5.7375,
|
|
"mean_token_accuracy": 0.17222627848386765,
|
|
"num_tokens": 6068194.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.903083658218383,
|
|
"epoch": 0.3060594148368619,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004995742757547207,
|
|
"loss": 5.7946,
|
|
"mean_token_accuracy": 0.1700889363884926,
|
|
"num_tokens": 6080540.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.854546451568604,
|
|
"epoch": 0.30662099174481944,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000499571803560316,
|
|
"loss": 5.8443,
|
|
"mean_token_accuracy": 0.16272160336375235,
|
|
"num_tokens": 6092464.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.971442317962646,
|
|
"epoch": 0.307182568652777,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004995693242154581,
|
|
"loss": 5.7935,
|
|
"mean_token_accuracy": 0.16596091389656067,
|
|
"num_tokens": 6103242.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.849540948867798,
|
|
"epoch": 0.30774414556073454,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004995668377202258,
|
|
"loss": 5.8144,
|
|
"mean_token_accuracy": 0.1654301702976227,
|
|
"num_tokens": 6116383.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.819859552383423,
|
|
"epoch": 0.3083057224686921,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004995643440746982,
|
|
"loss": 5.6267,
|
|
"mean_token_accuracy": 0.17252051830291748,
|
|
"num_tokens": 6126883.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.9390562057495115,
|
|
"epoch": 0.30886729937664964,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004995618432789548,
|
|
"loss": 5.8212,
|
|
"mean_token_accuracy": 0.16041162610054016,
|
|
"num_tokens": 6139613.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.920775699615478,
|
|
"epoch": 0.3094288762846072,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004995593353330753,
|
|
"loss": 5.883,
|
|
"mean_token_accuracy": 0.1595902293920517,
|
|
"num_tokens": 6151950.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.84655532836914,
|
|
"epoch": 0.30999045319256474,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004995568202371393,
|
|
"loss": 5.6638,
|
|
"mean_token_accuracy": 0.17385415136814117,
|
|
"num_tokens": 6161563.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.809947443008423,
|
|
"epoch": 0.3105520301005223,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004995542979912272,
|
|
"loss": 5.6793,
|
|
"mean_token_accuracy": 0.17475582361221315,
|
|
"num_tokens": 6170966.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.866269159317016,
|
|
"epoch": 0.3111136070084798,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004995517685954192,
|
|
"loss": 5.7548,
|
|
"mean_token_accuracy": 0.16865830570459367,
|
|
"num_tokens": 6182563.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.905898571014404,
|
|
"epoch": 0.31167518391643734,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004995492320497958,
|
|
"loss": 5.8659,
|
|
"mean_token_accuracy": 0.1606164261698723,
|
|
"num_tokens": 6194495.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.8810731887817385,
|
|
"epoch": 0.3122367608243949,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004995466883544376,
|
|
"loss": 5.7628,
|
|
"mean_token_accuracy": 0.16512676775455476,
|
|
"num_tokens": 6206594.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.955170059204102,
|
|
"epoch": 0.31279833773235244,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004995441375094258,
|
|
"loss": 5.7886,
|
|
"mean_token_accuracy": 0.16353470236063003,
|
|
"num_tokens": 6218371.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.889581632614136,
|
|
"epoch": 0.31335991464031,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004995415795148417,
|
|
"loss": 5.7207,
|
|
"mean_token_accuracy": 0.17165483981370927,
|
|
"num_tokens": 6228723.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.917896366119384,
|
|
"epoch": 0.31392149154826754,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004995390143707665,
|
|
"loss": 5.908,
|
|
"mean_token_accuracy": 0.16421120017766952,
|
|
"num_tokens": 6240535.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.842968416213989,
|
|
"epoch": 0.3144830684562251,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004995364420772821,
|
|
"loss": 5.6054,
|
|
"mean_token_accuracy": 0.1732317879796028,
|
|
"num_tokens": 6250643.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.829766130447387,
|
|
"epoch": 0.31504464536418264,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995338626344702,
|
|
"loss": 5.7341,
|
|
"mean_token_accuracy": 0.16942535787820817,
|
|
"num_tokens": 6259569.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.917622089385986,
|
|
"epoch": 0.3156062222721402,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000499531276042413,
|
|
"loss": 5.9022,
|
|
"mean_token_accuracy": 0.17028167098760605,
|
|
"num_tokens": 6271108.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.926321697235108,
|
|
"epoch": 0.31616779918009774,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995286823011929,
|
|
"loss": 5.738,
|
|
"mean_token_accuracy": 0.15809677839279174,
|
|
"num_tokens": 6282352.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.87783522605896,
|
|
"epoch": 0.31672937608805524,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004995260814108926,
|
|
"loss": 5.8509,
|
|
"mean_token_accuracy": 0.1608995482325554,
|
|
"num_tokens": 6294946.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.803282737731934,
|
|
"epoch": 0.3172909529960128,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004995234733715947,
|
|
"loss": 5.6583,
|
|
"mean_token_accuracy": 0.17535417079925536,
|
|
"num_tokens": 6305552.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.835207843780518,
|
|
"epoch": 0.31785252990397034,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004995208581833824,
|
|
"loss": 5.7333,
|
|
"mean_token_accuracy": 0.1743067905306816,
|
|
"num_tokens": 6315815.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.907580804824829,
|
|
"epoch": 0.3184141068119279,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995182358463388,
|
|
"loss": 5.7671,
|
|
"mean_token_accuracy": 0.16423598527908326,
|
|
"num_tokens": 6327384.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.8704465389251705,
|
|
"epoch": 0.31897568371988544,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004995156063605475,
|
|
"loss": 5.7459,
|
|
"mean_token_accuracy": 0.15973282158374785,
|
|
"num_tokens": 6338083.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.971950387954712,
|
|
"epoch": 0.319537260627843,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004995129697260922,
|
|
"loss": 5.7529,
|
|
"mean_token_accuracy": 0.17029481679201125,
|
|
"num_tokens": 6349058.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.814688491821289,
|
|
"epoch": 0.32009883753580054,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004995103259430569,
|
|
"loss": 5.7498,
|
|
"mean_token_accuracy": 0.1704786464571953,
|
|
"num_tokens": 6359655.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.95844259262085,
|
|
"epoch": 0.3206604144437581,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004995076750115258,
|
|
"loss": 5.8915,
|
|
"mean_token_accuracy": 0.16460290402173997,
|
|
"num_tokens": 6372088.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.860590982437134,
|
|
"epoch": 0.32122199135171564,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004995050169315832,
|
|
"loss": 5.7776,
|
|
"mean_token_accuracy": 0.16560305953025817,
|
|
"num_tokens": 6382946.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.871022176742554,
|
|
"epoch": 0.32178356825967314,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004995023517033138,
|
|
"loss": 5.7859,
|
|
"mean_token_accuracy": 0.16105389073491097,
|
|
"num_tokens": 6392591.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.835945558547974,
|
|
"epoch": 0.3223451451676307,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004994996793268024,
|
|
"loss": 5.722,
|
|
"mean_token_accuracy": 0.17206769585609435,
|
|
"num_tokens": 6403159.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.90753059387207,
|
|
"epoch": 0.32290672207558824,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004994969998021342,
|
|
"loss": 5.7444,
|
|
"mean_token_accuracy": 0.16411917731165887,
|
|
"num_tokens": 6414232.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.856598854064941,
|
|
"epoch": 0.3234682989835458,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004994943131293944,
|
|
"loss": 5.7139,
|
|
"mean_token_accuracy": 0.17788330912590028,
|
|
"num_tokens": 6425762.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.817793512344361,
|
|
"epoch": 0.32402987589150334,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004994916193086687,
|
|
"loss": 5.7755,
|
|
"mean_token_accuracy": 0.16134170442819595,
|
|
"num_tokens": 6439077.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.9469287395477295,
|
|
"epoch": 0.3245914527994609,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004994889183400428,
|
|
"loss": 5.831,
|
|
"mean_token_accuracy": 0.16600467413663864,
|
|
"num_tokens": 6450425.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.849802923202515,
|
|
"epoch": 0.32515302970741844,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004994862102236025,
|
|
"loss": 5.7945,
|
|
"mean_token_accuracy": 0.16656314879655837,
|
|
"num_tokens": 6461347.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.884488153457641,
|
|
"epoch": 0.325714606615376,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004994834949594342,
|
|
"loss": 5.7833,
|
|
"mean_token_accuracy": 0.1678066909313202,
|
|
"num_tokens": 6472428.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.844870376586914,
|
|
"epoch": 0.32627618352333354,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004994807725476246,
|
|
"loss": 5.7445,
|
|
"mean_token_accuracy": 0.17545183449983598,
|
|
"num_tokens": 6483393.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.855503273010254,
|
|
"epoch": 0.3268377604312911,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004994780429882599,
|
|
"loss": 5.7243,
|
|
"mean_token_accuracy": 0.16932949125766755,
|
|
"num_tokens": 6493520.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.791838979721069,
|
|
"epoch": 0.3273993373392486,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004994753062814274,
|
|
"loss": 5.6659,
|
|
"mean_token_accuracy": 0.1761859968304634,
|
|
"num_tokens": 6505282.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.974535512924194,
|
|
"epoch": 0.32796091424720614,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004994725624272139,
|
|
"loss": 5.838,
|
|
"mean_token_accuracy": 0.17055148035287857,
|
|
"num_tokens": 6516808.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.885997676849366,
|
|
"epoch": 0.3285224911551637,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004994698114257071,
|
|
"loss": 5.8016,
|
|
"mean_token_accuracy": 0.16254309713840484,
|
|
"num_tokens": 6529184.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.788518714904785,
|
|
"epoch": 0.32908406806312124,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004994670532769944,
|
|
"loss": 5.6459,
|
|
"mean_token_accuracy": 0.17723568826913833,
|
|
"num_tokens": 6539909.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.893859624862671,
|
|
"epoch": 0.3296456449710788,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004994642879811637,
|
|
"loss": 5.6959,
|
|
"mean_token_accuracy": 0.1725839927792549,
|
|
"num_tokens": 6550870.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.777626419067383,
|
|
"epoch": 0.33020722187903634,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499461515538303,
|
|
"loss": 5.7488,
|
|
"mean_token_accuracy": 0.17353251129388808,
|
|
"num_tokens": 6561784.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.808140802383423,
|
|
"epoch": 0.3307687987869939,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004994587359485006,
|
|
"loss": 5.7035,
|
|
"mean_token_accuracy": 0.16959256380796434,
|
|
"num_tokens": 6573422.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.940847444534302,
|
|
"epoch": 0.33133037569495144,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000499455949211845,
|
|
"loss": 5.7666,
|
|
"mean_token_accuracy": 0.17316267341375352,
|
|
"num_tokens": 6584565.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.905264854431152,
|
|
"epoch": 0.331891952602909,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004994531553284248,
|
|
"loss": 5.6778,
|
|
"mean_token_accuracy": 0.1691496714949608,
|
|
"num_tokens": 6594238.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.750309991836548,
|
|
"epoch": 0.3324535295108665,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004994503542983293,
|
|
"loss": 5.7469,
|
|
"mean_token_accuracy": 0.1695124924182892,
|
|
"num_tokens": 6606309.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.996778583526611,
|
|
"epoch": 0.33301510641882404,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004994475461216472,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.16220774352550507,
|
|
"num_tokens": 6617916.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.843599700927735,
|
|
"epoch": 0.3335766833267816,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004994447307984684,
|
|
"loss": 5.7021,
|
|
"mean_token_accuracy": 0.1709466338157654,
|
|
"num_tokens": 6627951.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.760891056060791,
|
|
"epoch": 0.33413826023473914,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004994419083288823,
|
|
"loss": 5.7622,
|
|
"mean_token_accuracy": 0.16756141930818558,
|
|
"num_tokens": 6639285.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.866618871688843,
|
|
"epoch": 0.3346998371426967,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004994390787129787,
|
|
"loss": 5.6267,
|
|
"mean_token_accuracy": 0.17637762576341628,
|
|
"num_tokens": 6649973.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.75209150314331,
|
|
"epoch": 0.33526141405065424,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004994362419508478,
|
|
"loss": 5.648,
|
|
"mean_token_accuracy": 0.1709219753742218,
|
|
"num_tokens": 6660625.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.846449995040894,
|
|
"epoch": 0.3358229909586118,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00049943339804258,
|
|
"loss": 5.7373,
|
|
"mean_token_accuracy": 0.16657833606004716,
|
|
"num_tokens": 6673191.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.888789987564087,
|
|
"epoch": 0.33638456786656934,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004994305469882657,
|
|
"loss": 5.7279,
|
|
"mean_token_accuracy": 0.17310751378536224,
|
|
"num_tokens": 6683993.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.8018608570098875,
|
|
"epoch": 0.3369461447745269,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004994276887879957,
|
|
"loss": 5.7611,
|
|
"mean_token_accuracy": 0.1704024627804756,
|
|
"num_tokens": 6694881.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.3369461447745269,
|
|
"eval_entropy": 5.657197658690343,
|
|
"eval_loss": 5.7566962242126465,
|
|
"eval_mean_token_accuracy": 0.17553335150354513,
|
|
"eval_num_tokens": 6694881.0,
|
|
"eval_runtime": 18.5774,
|
|
"eval_samples_per_second": 1669.392,
|
|
"eval_steps_per_second": 208.694,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 89030,
|
|
"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": 1549103788523520.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|