ee0d80c219
Model: fpadovani/eus-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.2887838418564677,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.491108751296997,
|
|
"epoch": 0.004297378599054577,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.3189,
|
|
"mean_token_accuracy": 0.09141870960593224,
|
|
"num_tokens": 10107.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.515214681625366,
|
|
"epoch": 0.008594757198109154,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.3203,
|
|
"mean_token_accuracy": 0.09664802104234696,
|
|
"num_tokens": 18391.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.51798095703125,
|
|
"epoch": 0.01289213579716373,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.3336,
|
|
"mean_token_accuracy": 0.09290757700800896,
|
|
"num_tokens": 27061.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.496134233474732,
|
|
"epoch": 0.017189514396218308,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.3792,
|
|
"mean_token_accuracy": 0.10020544826984405,
|
|
"num_tokens": 36339.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.491674518585205,
|
|
"epoch": 0.021486892995272882,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.3608,
|
|
"mean_token_accuracy": 0.09223981052637101,
|
|
"num_tokens": 45770.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.550447177886963,
|
|
"epoch": 0.02578427159432746,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.3766,
|
|
"mean_token_accuracy": 0.09226111099123954,
|
|
"num_tokens": 54575.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.5649909496307375,
|
|
"epoch": 0.030081650193382038,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.4503,
|
|
"mean_token_accuracy": 0.09087274968624115,
|
|
"num_tokens": 66403.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.568733358383179,
|
|
"epoch": 0.034379028792436615,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.3758,
|
|
"mean_token_accuracy": 0.09583879709243774,
|
|
"num_tokens": 76510.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.536895418167115,
|
|
"epoch": 0.03867640739149119,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.2707,
|
|
"mean_token_accuracy": 0.09973574429750443,
|
|
"num_tokens": 84836.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.539294195175171,
|
|
"epoch": 0.042973785990545764,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 7.2456,
|
|
"mean_token_accuracy": 0.10200604945421218,
|
|
"num_tokens": 93197.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.543957090377807,
|
|
"epoch": 0.047271164589600345,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.245,
|
|
"mean_token_accuracy": 0.09889603033661842,
|
|
"num_tokens": 101546.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.518790864944458,
|
|
"epoch": 0.05156854318865492,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.3185,
|
|
"mean_token_accuracy": 0.0964392140507698,
|
|
"num_tokens": 111703.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.569249010086059,
|
|
"epoch": 0.055865921787709494,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.2462,
|
|
"mean_token_accuracy": 0.10106701627373696,
|
|
"num_tokens": 119894.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.529278564453125,
|
|
"epoch": 0.060163300386764075,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.212,
|
|
"mean_token_accuracy": 0.10316545292735099,
|
|
"num_tokens": 128885.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.5085443496704105,
|
|
"epoch": 0.06446067898581866,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.3341,
|
|
"mean_token_accuracy": 0.09605684131383896,
|
|
"num_tokens": 138106.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.545536041259766,
|
|
"epoch": 0.06875805758487323,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.2453,
|
|
"mean_token_accuracy": 0.1033376321196556,
|
|
"num_tokens": 146691.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.557823944091797,
|
|
"epoch": 0.0730554361839278,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.2809,
|
|
"mean_token_accuracy": 0.10218598842620849,
|
|
"num_tokens": 155792.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.59003210067749,
|
|
"epoch": 0.07735281478298238,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.358,
|
|
"mean_token_accuracy": 0.09359859973192215,
|
|
"num_tokens": 166944.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.617101383209229,
|
|
"epoch": 0.08165019338203695,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.1989,
|
|
"mean_token_accuracy": 0.1046324297785759,
|
|
"num_tokens": 175303.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.528326416015625,
|
|
"epoch": 0.08594757198109153,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.2591,
|
|
"mean_token_accuracy": 0.10362932458519936,
|
|
"num_tokens": 184708.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.538442516326905,
|
|
"epoch": 0.09024495058014612,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.2175,
|
|
"mean_token_accuracy": 0.10121936872601509,
|
|
"num_tokens": 193835.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.487286758422852,
|
|
"epoch": 0.09454232917920069,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.2608,
|
|
"mean_token_accuracy": 0.10149957984685898,
|
|
"num_tokens": 203344.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.519776153564453,
|
|
"epoch": 0.09883970777825526,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 7.365,
|
|
"mean_token_accuracy": 0.090200637280941,
|
|
"num_tokens": 213048.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.641110944747925,
|
|
"epoch": 0.10313708637730984,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.2283,
|
|
"mean_token_accuracy": 0.10701292306184769,
|
|
"num_tokens": 221784.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.459906339645386,
|
|
"epoch": 0.10743446497636441,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.2232,
|
|
"mean_token_accuracy": 0.0979800947010517,
|
|
"num_tokens": 230971.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.543297910690308,
|
|
"epoch": 0.11173184357541899,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.1925,
|
|
"mean_token_accuracy": 0.10740587189793586,
|
|
"num_tokens": 240524.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.5613665103912355,
|
|
"epoch": 0.11602922217447358,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.1644,
|
|
"mean_token_accuracy": 0.10438360348343849,
|
|
"num_tokens": 249220.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.50055251121521,
|
|
"epoch": 0.12032660077352815,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.2801,
|
|
"mean_token_accuracy": 0.10291781276464462,
|
|
"num_tokens": 258934.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.580437517166137,
|
|
"epoch": 0.12462397937258272,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.1981,
|
|
"mean_token_accuracy": 0.10456435978412629,
|
|
"num_tokens": 267680.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.493255138397217,
|
|
"epoch": 0.1289213579716373,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.2271,
|
|
"mean_token_accuracy": 0.10516949817538261,
|
|
"num_tokens": 276227.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.456282997131348,
|
|
"epoch": 0.1332187365706919,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.1356,
|
|
"mean_token_accuracy": 0.10240317732095719,
|
|
"num_tokens": 286342.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.478529405593872,
|
|
"epoch": 0.13751611516974646,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.2034,
|
|
"mean_token_accuracy": 0.10278507694602013,
|
|
"num_tokens": 294994.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.560914468765259,
|
|
"epoch": 0.14181349376880104,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.1728,
|
|
"mean_token_accuracy": 0.10567683354020119,
|
|
"num_tokens": 303882.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.53267731666565,
|
|
"epoch": 0.1461108723678556,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.2185,
|
|
"mean_token_accuracy": 0.10334216356277466,
|
|
"num_tokens": 312515.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.510133743286133,
|
|
"epoch": 0.15040825096691018,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.1819,
|
|
"mean_token_accuracy": 0.1053494967520237,
|
|
"num_tokens": 320801.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.564500951766968,
|
|
"epoch": 0.15470562956596476,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.1958,
|
|
"mean_token_accuracy": 0.10476180166006088,
|
|
"num_tokens": 329382.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.523471403121948,
|
|
"epoch": 0.15900300816501933,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.1811,
|
|
"mean_token_accuracy": 0.1035641834139824,
|
|
"num_tokens": 337894.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.490967750549316,
|
|
"epoch": 0.1633003867640739,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.1711,
|
|
"mean_token_accuracy": 0.10333409234881401,
|
|
"num_tokens": 346380.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.575121831893921,
|
|
"epoch": 0.16759776536312848,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.2686,
|
|
"mean_token_accuracy": 0.09736321717500687,
|
|
"num_tokens": 356305.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.543933534622193,
|
|
"epoch": 0.17189514396218306,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.2081,
|
|
"mean_token_accuracy": 0.10070987790822983,
|
|
"num_tokens": 364899.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.537495565414429,
|
|
"epoch": 0.17619252256123766,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.1983,
|
|
"mean_token_accuracy": 0.10088053569197655,
|
|
"num_tokens": 373663.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.536725664138794,
|
|
"epoch": 0.18048990116029223,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.1549,
|
|
"mean_token_accuracy": 0.10402223244309425,
|
|
"num_tokens": 382730.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.5419622421264645,
|
|
"epoch": 0.1847872797593468,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2121,
|
|
"mean_token_accuracy": 0.09917423650622367,
|
|
"num_tokens": 392676.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.498674964904785,
|
|
"epoch": 0.18908465835840138,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.1809,
|
|
"mean_token_accuracy": 0.09803537428379058,
|
|
"num_tokens": 401050.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.47620153427124,
|
|
"epoch": 0.19338203695745596,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.1693,
|
|
"mean_token_accuracy": 0.10769422501325607,
|
|
"num_tokens": 410009.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.477664566040039,
|
|
"epoch": 0.19767941555651053,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.1813,
|
|
"mean_token_accuracy": 0.1022846557199955,
|
|
"num_tokens": 419302.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.504415702819824,
|
|
"epoch": 0.2019767941555651,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.0897,
|
|
"mean_token_accuracy": 0.11041160002350807,
|
|
"num_tokens": 427296.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.440302228927612,
|
|
"epoch": 0.20627417275461968,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.2172,
|
|
"mean_token_accuracy": 0.10320142880082131,
|
|
"num_tokens": 436368.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.594078588485718,
|
|
"epoch": 0.21057155135367425,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1116,
|
|
"mean_token_accuracy": 0.10851684138178826,
|
|
"num_tokens": 445535.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.354097700119018,
|
|
"epoch": 0.21486892995272883,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1184,
|
|
"mean_token_accuracy": 0.10696499496698379,
|
|
"num_tokens": 454769.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.493505668640137,
|
|
"epoch": 0.2191663085517834,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1528,
|
|
"mean_token_accuracy": 0.10274568796157837,
|
|
"num_tokens": 463975.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.474530839920044,
|
|
"epoch": 0.22346368715083798,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1706,
|
|
"mean_token_accuracy": 0.10459161475300789,
|
|
"num_tokens": 472899.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.493699789047241,
|
|
"epoch": 0.22776106574989258,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1087,
|
|
"mean_token_accuracy": 0.10158660039305686,
|
|
"num_tokens": 481556.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.483793687820435,
|
|
"epoch": 0.23205844434894715,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.1274,
|
|
"mean_token_accuracy": 0.10895142033696174,
|
|
"num_tokens": 490253.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.462975406646729,
|
|
"epoch": 0.23635582294800173,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1663,
|
|
"mean_token_accuracy": 0.10515937954187393,
|
|
"num_tokens": 498444.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.426873064041137,
|
|
"epoch": 0.2406532015470563,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.093,
|
|
"mean_token_accuracy": 0.11008984223008156,
|
|
"num_tokens": 508330.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.484303283691406,
|
|
"epoch": 0.24495058014611087,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.2074,
|
|
"mean_token_accuracy": 0.10344811007380486,
|
|
"num_tokens": 517900.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.455639600753784,
|
|
"epoch": 0.24924795874516545,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1342,
|
|
"mean_token_accuracy": 0.10256287604570388,
|
|
"num_tokens": 527808.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.514905118942261,
|
|
"epoch": 0.25354533734422,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1506,
|
|
"mean_token_accuracy": 0.10260518342256546,
|
|
"num_tokens": 536931.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.504227924346924,
|
|
"epoch": 0.2578427159432746,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.1732,
|
|
"mean_token_accuracy": 0.10155779272317886,
|
|
"num_tokens": 545758.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.49675178527832,
|
|
"epoch": 0.26214009454232917,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.1899,
|
|
"mean_token_accuracy": 0.10567427054047585,
|
|
"num_tokens": 555165.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.394629907608032,
|
|
"epoch": 0.2664374731413838,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0983,
|
|
"mean_token_accuracy": 0.10148947015404701,
|
|
"num_tokens": 564719.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.432728815078735,
|
|
"epoch": 0.2707348517404383,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0682,
|
|
"mean_token_accuracy": 0.10739247426390648,
|
|
"num_tokens": 573572.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.586349010467529,
|
|
"epoch": 0.2750322303394929,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.2456,
|
|
"mean_token_accuracy": 0.10255748108029365,
|
|
"num_tokens": 581497.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.318369674682617,
|
|
"epoch": 0.27932960893854747,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1311,
|
|
"mean_token_accuracy": 0.10708568394184112,
|
|
"num_tokens": 591107.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.487468528747558,
|
|
"epoch": 0.28362698753760207,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1639,
|
|
"mean_token_accuracy": 0.10406824648380279,
|
|
"num_tokens": 600241.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.458619403839111,
|
|
"epoch": 0.2879243661366566,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.0833,
|
|
"mean_token_accuracy": 0.10933952927589416,
|
|
"num_tokens": 608697.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.387399578094483,
|
|
"epoch": 0.2922217447357112,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0767,
|
|
"mean_token_accuracy": 0.11132601723074913,
|
|
"num_tokens": 617275.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.476171684265137,
|
|
"epoch": 0.29651912333476577,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1697,
|
|
"mean_token_accuracy": 0.1055254265666008,
|
|
"num_tokens": 626644.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.3781880855560305,
|
|
"epoch": 0.30081650193382037,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.067,
|
|
"mean_token_accuracy": 0.10572395622730255,
|
|
"num_tokens": 635110.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.468985176086425,
|
|
"epoch": 0.30511388053287497,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.2351,
|
|
"mean_token_accuracy": 0.10162713378667831,
|
|
"num_tokens": 644746.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.523141098022461,
|
|
"epoch": 0.3094112591319295,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.1415,
|
|
"mean_token_accuracy": 0.10366995185613632,
|
|
"num_tokens": 654281.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.358317518234253,
|
|
"epoch": 0.3137086377309841,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.1342,
|
|
"mean_token_accuracy": 0.1051877960562706,
|
|
"num_tokens": 663174.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.420715570449829,
|
|
"epoch": 0.31800601633003867,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.0567,
|
|
"mean_token_accuracy": 0.1060252383351326,
|
|
"num_tokens": 672178.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.440627765655518,
|
|
"epoch": 0.32230339492909327,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0829,
|
|
"mean_token_accuracy": 0.10851024910807609,
|
|
"num_tokens": 681323.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.312611055374146,
|
|
"epoch": 0.3266007735281478,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0134,
|
|
"mean_token_accuracy": 0.1070731021463871,
|
|
"num_tokens": 690461.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.485116195678711,
|
|
"epoch": 0.3308981521272024,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.038,
|
|
"mean_token_accuracy": 0.10665726512670518,
|
|
"num_tokens": 699199.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.286711549758911,
|
|
"epoch": 0.33519553072625696,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.0332,
|
|
"mean_token_accuracy": 0.11690814569592475,
|
|
"num_tokens": 707949.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.411632871627807,
|
|
"epoch": 0.33949290932531156,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0682,
|
|
"mean_token_accuracy": 0.10877912119030952,
|
|
"num_tokens": 715752.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.451524162292481,
|
|
"epoch": 0.3437902879243661,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.0304,
|
|
"mean_token_accuracy": 0.11193935424089432,
|
|
"num_tokens": 724416.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.3398294925689695,
|
|
"epoch": 0.3480876665234207,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0979,
|
|
"mean_token_accuracy": 0.10706080496311188,
|
|
"num_tokens": 733116.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.373892545700073,
|
|
"epoch": 0.3523850451224753,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.0823,
|
|
"mean_token_accuracy": 0.10571770668029785,
|
|
"num_tokens": 742093.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.35628867149353,
|
|
"epoch": 0.35668242372152986,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.0485,
|
|
"mean_token_accuracy": 0.11085971966385841,
|
|
"num_tokens": 750402.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.272590160369873,
|
|
"epoch": 0.36097980232058446,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0175,
|
|
"mean_token_accuracy": 0.11432302817702293,
|
|
"num_tokens": 760961.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.479759454727173,
|
|
"epoch": 0.365277180919639,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.1296,
|
|
"mean_token_accuracy": 0.10654517561197281,
|
|
"num_tokens": 770554.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.259108257293701,
|
|
"epoch": 0.3695745595186936,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0763,
|
|
"mean_token_accuracy": 0.11016570776700974,
|
|
"num_tokens": 779172.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.46284122467041,
|
|
"epoch": 0.37387193811774816,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.1502,
|
|
"mean_token_accuracy": 0.10204855352640152,
|
|
"num_tokens": 788040.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.377580165863037,
|
|
"epoch": 0.37816931671680276,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.0556,
|
|
"mean_token_accuracy": 0.10719484761357308,
|
|
"num_tokens": 796786.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.2963978290557865,
|
|
"epoch": 0.3824666953158573,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0257,
|
|
"mean_token_accuracy": 0.11093981638550758,
|
|
"num_tokens": 805520.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.230100870132446,
|
|
"epoch": 0.3867640739149119,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9865,
|
|
"mean_token_accuracy": 0.10948721244931221,
|
|
"num_tokens": 814939.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.344197130203247,
|
|
"epoch": 0.39106145251396646,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.9985,
|
|
"mean_token_accuracy": 0.11251249313354492,
|
|
"num_tokens": 823862.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.369870948791504,
|
|
"epoch": 0.39535883111302106,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.0169,
|
|
"mean_token_accuracy": 0.10975069999694824,
|
|
"num_tokens": 832820.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.322132015228272,
|
|
"epoch": 0.39965620971207566,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0044,
|
|
"mean_token_accuracy": 0.11229758933186532,
|
|
"num_tokens": 841538.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.337738561630249,
|
|
"epoch": 0.4039535883111302,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.1229,
|
|
"mean_token_accuracy": 0.10578583925962448,
|
|
"num_tokens": 851123.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.333683729171753,
|
|
"epoch": 0.4082509669101848,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.0822,
|
|
"mean_token_accuracy": 0.1119972050189972,
|
|
"num_tokens": 860357.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.331111764907837,
|
|
"epoch": 0.41254834550923936,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.9903,
|
|
"mean_token_accuracy": 0.11009360924363136,
|
|
"num_tokens": 869980.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.231062650680542,
|
|
"epoch": 0.41684572410829396,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.9683,
|
|
"mean_token_accuracy": 0.11301724538207054,
|
|
"num_tokens": 878250.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.38806791305542,
|
|
"epoch": 0.4211431027073485,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.0717,
|
|
"mean_token_accuracy": 0.10711101293563843,
|
|
"num_tokens": 887624.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.216950559616089,
|
|
"epoch": 0.4254404813064031,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.0432,
|
|
"mean_token_accuracy": 0.11447345688939095,
|
|
"num_tokens": 897120.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.223309516906738,
|
|
"epoch": 0.42973785990545765,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.0009,
|
|
"mean_token_accuracy": 0.11249879151582717,
|
|
"num_tokens": 906215.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.42973785990545765,
|
|
"eval_entropy": 7.237139492421536,
|
|
"eval_loss": 7.02439546585083,
|
|
"eval_mean_token_accuracy": 0.11452774357822565,
|
|
"eval_num_tokens": 906215.0,
|
|
"eval_runtime": 2.6841,
|
|
"eval_samples_per_second": 1322.226,
|
|
"eval_steps_per_second": 165.418,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.341209077835083,
|
|
"epoch": 0.43403523850451226,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.0128,
|
|
"mean_token_accuracy": 0.11152525022625923,
|
|
"num_tokens": 915181.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.27630581855774,
|
|
"epoch": 0.4383326171035668,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.0197,
|
|
"mean_token_accuracy": 0.1118933878839016,
|
|
"num_tokens": 924377.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.359603357315064,
|
|
"epoch": 0.4426299957026214,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.1061,
|
|
"mean_token_accuracy": 0.10802508145570755,
|
|
"num_tokens": 933114.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.359967756271362,
|
|
"epoch": 0.44692737430167595,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.0622,
|
|
"mean_token_accuracy": 0.10966231301426888,
|
|
"num_tokens": 943306.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.2014100551605225,
|
|
"epoch": 0.45122475290073055,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.9652,
|
|
"mean_token_accuracy": 0.1156508482992649,
|
|
"num_tokens": 951515.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.308715724945069,
|
|
"epoch": 0.45552213149978515,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.0944,
|
|
"mean_token_accuracy": 0.10246431827545166,
|
|
"num_tokens": 962686.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.312524938583374,
|
|
"epoch": 0.4598195100988397,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.0842,
|
|
"mean_token_accuracy": 0.11101078242063522,
|
|
"num_tokens": 972136.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.238412475585937,
|
|
"epoch": 0.4641168886978943,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.9806,
|
|
"mean_token_accuracy": 0.11057559326291085,
|
|
"num_tokens": 981301.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.246242666244507,
|
|
"epoch": 0.46841426729694885,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.0099,
|
|
"mean_token_accuracy": 0.10596411451697349,
|
|
"num_tokens": 990360.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.30305585861206,
|
|
"epoch": 0.47271164589600345,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.0565,
|
|
"mean_token_accuracy": 0.10379971489310265,
|
|
"num_tokens": 999415.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.228947544097901,
|
|
"epoch": 0.477009024495058,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.9819,
|
|
"mean_token_accuracy": 0.10675028637051583,
|
|
"num_tokens": 1008762.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.223627710342408,
|
|
"epoch": 0.4813064030941126,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.9712,
|
|
"mean_token_accuracy": 0.11833404675126076,
|
|
"num_tokens": 1017704.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.206258535385132,
|
|
"epoch": 0.48560378169316715,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.9319,
|
|
"mean_token_accuracy": 0.11355781629681587,
|
|
"num_tokens": 1026251.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.172951602935791,
|
|
"epoch": 0.48990116029222175,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.9471,
|
|
"mean_token_accuracy": 0.11119012758135796,
|
|
"num_tokens": 1036191.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.232996034622192,
|
|
"epoch": 0.4941985388912763,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.9737,
|
|
"mean_token_accuracy": 0.11196622401475906,
|
|
"num_tokens": 1044936.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.271992301940918,
|
|
"epoch": 0.4984959174903309,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.9757,
|
|
"mean_token_accuracy": 0.11317787989974022,
|
|
"num_tokens": 1053683.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.157695055007935,
|
|
"epoch": 0.5027932960893855,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.9486,
|
|
"mean_token_accuracy": 0.11400579363107681,
|
|
"num_tokens": 1062932.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.19578332901001,
|
|
"epoch": 0.50709067468844,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.9565,
|
|
"mean_token_accuracy": 0.11118854060769082,
|
|
"num_tokens": 1072313.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.2023499488830565,
|
|
"epoch": 0.5113880532874946,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.9656,
|
|
"mean_token_accuracy": 0.10927657783031464,
|
|
"num_tokens": 1081675.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.180068778991699,
|
|
"epoch": 0.5156854318865493,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.9584,
|
|
"mean_token_accuracy": 0.10982275977730752,
|
|
"num_tokens": 1091541.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.253817129135132,
|
|
"epoch": 0.5199828104856038,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.9824,
|
|
"mean_token_accuracy": 0.11794964224100113,
|
|
"num_tokens": 1100724.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.123758220672608,
|
|
"epoch": 0.5242801890846583,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.9407,
|
|
"mean_token_accuracy": 0.11821772009134293,
|
|
"num_tokens": 1108869.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.2373988151550295,
|
|
"epoch": 0.5285775676837129,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.968,
|
|
"mean_token_accuracy": 0.1125386081635952,
|
|
"num_tokens": 1117314.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.098644876480103,
|
|
"epoch": 0.5328749462827675,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.9317,
|
|
"mean_token_accuracy": 0.11509137749671935,
|
|
"num_tokens": 1126786.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.08965916633606,
|
|
"epoch": 0.5371723248818221,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.9196,
|
|
"mean_token_accuracy": 0.11359397768974304,
|
|
"num_tokens": 1136013.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.193327474594116,
|
|
"epoch": 0.5414697034808766,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.9515,
|
|
"mean_token_accuracy": 0.11913771256804466,
|
|
"num_tokens": 1144970.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.144840240478516,
|
|
"epoch": 0.5457670820799312,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.9041,
|
|
"mean_token_accuracy": 0.12084828615188599,
|
|
"num_tokens": 1153810.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.152580881118775,
|
|
"epoch": 0.5500644606789858,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.9033,
|
|
"mean_token_accuracy": 0.11283695176243783,
|
|
"num_tokens": 1162498.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.078862094879151,
|
|
"epoch": 0.5543618392780404,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.9844,
|
|
"mean_token_accuracy": 0.11267361715435982,
|
|
"num_tokens": 1172091.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.244156694412231,
|
|
"epoch": 0.5586592178770949,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.9711,
|
|
"mean_token_accuracy": 0.1131252221763134,
|
|
"num_tokens": 1181400.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.147411584854126,
|
|
"epoch": 0.5629565964761496,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.9429,
|
|
"mean_token_accuracy": 0.11570452973246574,
|
|
"num_tokens": 1189780.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.069279909133911,
|
|
"epoch": 0.5672539750752041,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.9177,
|
|
"mean_token_accuracy": 0.11673617735505104,
|
|
"num_tokens": 1198671.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.186899328231812,
|
|
"epoch": 0.5715513536742587,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.9346,
|
|
"mean_token_accuracy": 0.11273332387208938,
|
|
"num_tokens": 1207173.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.0509740829467775,
|
|
"epoch": 0.5758487322733132,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.8633,
|
|
"mean_token_accuracy": 0.11901918649673462,
|
|
"num_tokens": 1216387.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.1644703388214115,
|
|
"epoch": 0.5801461108723679,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.861,
|
|
"mean_token_accuracy": 0.11411493569612503,
|
|
"num_tokens": 1224461.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.09780912399292,
|
|
"epoch": 0.5844434894714224,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.9331,
|
|
"mean_token_accuracy": 0.11777294874191284,
|
|
"num_tokens": 1233774.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.0529937744140625,
|
|
"epoch": 0.588740868070477,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.8872,
|
|
"mean_token_accuracy": 0.11746734082698822,
|
|
"num_tokens": 1242812.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.235918474197388,
|
|
"epoch": 0.5930382466695315,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.0025,
|
|
"mean_token_accuracy": 0.10898117944598198,
|
|
"num_tokens": 1252872.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.88110318183899,
|
|
"epoch": 0.5973356252685862,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.7799,
|
|
"mean_token_accuracy": 0.11457612216472626,
|
|
"num_tokens": 1260852.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.073757219314575,
|
|
"epoch": 0.6016330038676407,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.8843,
|
|
"mean_token_accuracy": 0.11701429635286331,
|
|
"num_tokens": 1268925.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.184703063964844,
|
|
"epoch": 0.6059303824666953,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.9396,
|
|
"mean_token_accuracy": 0.10898746475577355,
|
|
"num_tokens": 1278994.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.032673788070679,
|
|
"epoch": 0.6102277610657499,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.9216,
|
|
"mean_token_accuracy": 0.11066435426473617,
|
|
"num_tokens": 1287698.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.109692239761353,
|
|
"epoch": 0.6145251396648045,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.853,
|
|
"mean_token_accuracy": 0.11746916100382805,
|
|
"num_tokens": 1297475.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.045703029632568,
|
|
"epoch": 0.618822518263859,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.8945,
|
|
"mean_token_accuracy": 0.11652352586388588,
|
|
"num_tokens": 1306836.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.154710102081299,
|
|
"epoch": 0.6231198968629136,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.9426,
|
|
"mean_token_accuracy": 0.10900169834494591,
|
|
"num_tokens": 1315872.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.087551069259644,
|
|
"epoch": 0.6274172754619682,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.8936,
|
|
"mean_token_accuracy": 0.1230031467974186,
|
|
"num_tokens": 1324624.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.031505298614502,
|
|
"epoch": 0.6317146540610228,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.8703,
|
|
"mean_token_accuracy": 0.12103518173098564,
|
|
"num_tokens": 1333058.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.979029607772827,
|
|
"epoch": 0.6360120326600773,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.8463,
|
|
"mean_token_accuracy": 0.11296493709087371,
|
|
"num_tokens": 1342376.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.05972146987915,
|
|
"epoch": 0.6403094112591319,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.8815,
|
|
"mean_token_accuracy": 0.10966399237513542,
|
|
"num_tokens": 1351386.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.994411182403565,
|
|
"epoch": 0.6446067898581865,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.7743,
|
|
"mean_token_accuracy": 0.12411837130784989,
|
|
"num_tokens": 1358958.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.1177832126617435,
|
|
"epoch": 0.6489041684572411,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.9423,
|
|
"mean_token_accuracy": 0.1123286470770836,
|
|
"num_tokens": 1368599.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.095702123641968,
|
|
"epoch": 0.6532015470562956,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.9805,
|
|
"mean_token_accuracy": 0.10967258512973785,
|
|
"num_tokens": 1378529.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.926534938812256,
|
|
"epoch": 0.6574989256553503,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.7622,
|
|
"mean_token_accuracy": 0.12447424307465553,
|
|
"num_tokens": 1386993.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.022530889511108,
|
|
"epoch": 0.6617963042544048,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.8542,
|
|
"mean_token_accuracy": 0.1206921823322773,
|
|
"num_tokens": 1395790.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.937219333648682,
|
|
"epoch": 0.6660936828534594,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.8662,
|
|
"mean_token_accuracy": 0.11413793414831161,
|
|
"num_tokens": 1405587.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.998539781570434,
|
|
"epoch": 0.6703910614525139,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.8418,
|
|
"mean_token_accuracy": 0.11912463083863259,
|
|
"num_tokens": 1414478.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.073210048675537,
|
|
"epoch": 0.6746884400515686,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.9072,
|
|
"mean_token_accuracy": 0.11900003850460053,
|
|
"num_tokens": 1423791.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.071209669113159,
|
|
"epoch": 0.6789858186506231,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.9546,
|
|
"mean_token_accuracy": 0.10661662891507148,
|
|
"num_tokens": 1432955.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.010345840454102,
|
|
"epoch": 0.6832831972496777,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.9059,
|
|
"mean_token_accuracy": 0.11467154771089554,
|
|
"num_tokens": 1441907.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.940794563293457,
|
|
"epoch": 0.6875805758487322,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.7741,
|
|
"mean_token_accuracy": 0.12223436459898948,
|
|
"num_tokens": 1451062.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.077725887298584,
|
|
"epoch": 0.6918779544477869,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.8575,
|
|
"mean_token_accuracy": 0.11200195476412773,
|
|
"num_tokens": 1460132.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.851260232925415,
|
|
"epoch": 0.6961753330468414,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.7443,
|
|
"mean_token_accuracy": 0.12217162400484086,
|
|
"num_tokens": 1469582.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.973898649215698,
|
|
"epoch": 0.700472711645896,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.8383,
|
|
"mean_token_accuracy": 0.11406316757202148,
|
|
"num_tokens": 1479053.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.982749509811401,
|
|
"epoch": 0.7047700902449506,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.8506,
|
|
"mean_token_accuracy": 0.1121716171503067,
|
|
"num_tokens": 1488189.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.829049015045166,
|
|
"epoch": 0.7090674688440052,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.7189,
|
|
"mean_token_accuracy": 0.12386269792914391,
|
|
"num_tokens": 1497324.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.92303500175476,
|
|
"epoch": 0.7133648474430597,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.8317,
|
|
"mean_token_accuracy": 0.1222977451980114,
|
|
"num_tokens": 1506543.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.996142625808716,
|
|
"epoch": 0.7176622260421143,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.8217,
|
|
"mean_token_accuracy": 0.1265745148062706,
|
|
"num_tokens": 1516737.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.881067752838135,
|
|
"epoch": 0.7219596046411689,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.6835,
|
|
"mean_token_accuracy": 0.12508728802204133,
|
|
"num_tokens": 1525561.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.863363933563233,
|
|
"epoch": 0.7262569832402235,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.7097,
|
|
"mean_token_accuracy": 0.1254148416221142,
|
|
"num_tokens": 1533323.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.951838445663452,
|
|
"epoch": 0.730554361839278,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.7733,
|
|
"mean_token_accuracy": 0.12557526156306267,
|
|
"num_tokens": 1542632.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.895253658294678,
|
|
"epoch": 0.7348517404383326,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.736,
|
|
"mean_token_accuracy": 0.12572805732488632,
|
|
"num_tokens": 1551236.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.809239673614502,
|
|
"epoch": 0.7391491190373872,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.7388,
|
|
"mean_token_accuracy": 0.11835454404354095,
|
|
"num_tokens": 1559674.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.876753854751587,
|
|
"epoch": 0.7434464976364418,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.8232,
|
|
"mean_token_accuracy": 0.12170301601290703,
|
|
"num_tokens": 1569481.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.016146326065064,
|
|
"epoch": 0.7477438762354963,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.8623,
|
|
"mean_token_accuracy": 0.11574919894337654,
|
|
"num_tokens": 1578488.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.8354627132415775,
|
|
"epoch": 0.752041254834551,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.751,
|
|
"mean_token_accuracy": 0.12034946754574775,
|
|
"num_tokens": 1586675.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.949285078048706,
|
|
"epoch": 0.7563386334336055,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.8707,
|
|
"mean_token_accuracy": 0.11235570311546325,
|
|
"num_tokens": 1595411.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.988899230957031,
|
|
"epoch": 0.7606360120326601,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.8369,
|
|
"mean_token_accuracy": 0.11851750463247299,
|
|
"num_tokens": 1604046.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.915553617477417,
|
|
"epoch": 0.7649333906317146,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.8301,
|
|
"mean_token_accuracy": 0.1143190972507,
|
|
"num_tokens": 1613759.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.809361219406128,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.8007,
|
|
"mean_token_accuracy": 0.1207757331430912,
|
|
"num_tokens": 1623323.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.934272289276123,
|
|
"epoch": 0.7735281478298238,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.7856,
|
|
"mean_token_accuracy": 0.11662489771842957,
|
|
"num_tokens": 1631727.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.889858341217041,
|
|
"epoch": 0.7778255264288784,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.703,
|
|
"mean_token_accuracy": 0.12306550592184066,
|
|
"num_tokens": 1639544.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.9098219871521,
|
|
"epoch": 0.7821229050279329,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.8408,
|
|
"mean_token_accuracy": 0.11140461266040802,
|
|
"num_tokens": 1648931.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.821422386169433,
|
|
"epoch": 0.7864202836269876,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.711,
|
|
"mean_token_accuracy": 0.1199341893196106,
|
|
"num_tokens": 1657688.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.889161062240601,
|
|
"epoch": 0.7907176622260421,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.8008,
|
|
"mean_token_accuracy": 0.12064535990357399,
|
|
"num_tokens": 1666879.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.840409898757935,
|
|
"epoch": 0.7950150408250967,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.7666,
|
|
"mean_token_accuracy": 0.12245901748538017,
|
|
"num_tokens": 1676773.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.8722686767578125,
|
|
"epoch": 0.7993124194241513,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.8727,
|
|
"mean_token_accuracy": 0.11736220642924308,
|
|
"num_tokens": 1686144.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.927791166305542,
|
|
"epoch": 0.8036097980232059,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.8363,
|
|
"mean_token_accuracy": 0.11194746792316437,
|
|
"num_tokens": 1695476.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.838814640045166,
|
|
"epoch": 0.8079071766222604,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.7542,
|
|
"mean_token_accuracy": 0.11981326714158058,
|
|
"num_tokens": 1704907.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.837986898422241,
|
|
"epoch": 0.812204555221315,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.7911,
|
|
"mean_token_accuracy": 0.1210789442062378,
|
|
"num_tokens": 1714564.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.931927728652954,
|
|
"epoch": 0.8165019338203696,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.8032,
|
|
"mean_token_accuracy": 0.11883696243166923,
|
|
"num_tokens": 1725285.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.8115462303161625,
|
|
"epoch": 0.8207993124194242,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.7991,
|
|
"mean_token_accuracy": 0.11632968410849572,
|
|
"num_tokens": 1734331.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.857008695602417,
|
|
"epoch": 0.8250966910184787,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.726,
|
|
"mean_token_accuracy": 0.1166811965405941,
|
|
"num_tokens": 1742340.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.806628656387329,
|
|
"epoch": 0.8293940696175333,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.7189,
|
|
"mean_token_accuracy": 0.12429693192243577,
|
|
"num_tokens": 1751725.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.705836391448974,
|
|
"epoch": 0.8336914482165879,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.6562,
|
|
"mean_token_accuracy": 0.12673527225852013,
|
|
"num_tokens": 1760294.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.776405715942383,
|
|
"epoch": 0.8379888268156425,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.7301,
|
|
"mean_token_accuracy": 0.12227423414587975,
|
|
"num_tokens": 1768912.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.929677724838257,
|
|
"epoch": 0.842286205414697,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.9124,
|
|
"mean_token_accuracy": 0.11897997409105301,
|
|
"num_tokens": 1778633.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.877961349487305,
|
|
"epoch": 0.8465835840137517,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.7875,
|
|
"mean_token_accuracy": 0.125636800378561,
|
|
"num_tokens": 1787275.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.902551507949829,
|
|
"epoch": 0.8508809626128062,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.7787,
|
|
"mean_token_accuracy": 0.11841877847909928,
|
|
"num_tokens": 1795994.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.722687721252441,
|
|
"epoch": 0.8551783412118608,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.7049,
|
|
"mean_token_accuracy": 0.12368022575974465,
|
|
"num_tokens": 1806379.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.836816644668579,
|
|
"epoch": 0.8594757198109153,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.7419,
|
|
"mean_token_accuracy": 0.12570958435535431,
|
|
"num_tokens": 1816135.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8594757198109153,
|
|
"eval_entropy": 6.472366559612858,
|
|
"eval_loss": 6.754819869995117,
|
|
"eval_mean_token_accuracy": 0.12417131071692114,
|
|
"eval_num_tokens": 1816135.0,
|
|
"eval_runtime": 2.6416,
|
|
"eval_samples_per_second": 1343.488,
|
|
"eval_steps_per_second": 168.078,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.819082975387573,
|
|
"epoch": 0.86377309840997,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999998427807679,
|
|
"loss": 6.7428,
|
|
"mean_token_accuracy": 0.12198886126279831,
|
|
"num_tokens": 1824777.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.709800720214844,
|
|
"epoch": 0.8680704770090245,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004999992040780138,
|
|
"loss": 6.7973,
|
|
"mean_token_accuracy": 0.1257079139351845,
|
|
"num_tokens": 1833807.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.860920858383179,
|
|
"epoch": 0.8723678556080791,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999980740669294,
|
|
"loss": 6.7525,
|
|
"mean_token_accuracy": 0.12449089586734771,
|
|
"num_tokens": 1843375.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.839705562591552,
|
|
"epoch": 0.8766652342071336,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999964527499823,
|
|
"loss": 6.8141,
|
|
"mean_token_accuracy": 0.11705794557929039,
|
|
"num_tokens": 1853036.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.833159351348877,
|
|
"epoch": 0.8809626128061883,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999943401307127,
|
|
"loss": 6.7628,
|
|
"mean_token_accuracy": 0.11710175424814225,
|
|
"num_tokens": 1862041.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.74225492477417,
|
|
"epoch": 0.8852599914052428,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999917362137337,
|
|
"loss": 6.6787,
|
|
"mean_token_accuracy": 0.12101037502288818,
|
|
"num_tokens": 1870707.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.785744285583496,
|
|
"epoch": 0.8895573700042974,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999886410047312,
|
|
"loss": 6.6862,
|
|
"mean_token_accuracy": 0.12277518585324287,
|
|
"num_tokens": 1879787.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.834198951721191,
|
|
"epoch": 0.8938547486033519,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999850545104638,
|
|
"loss": 6.7421,
|
|
"mean_token_accuracy": 0.12042063474655151,
|
|
"num_tokens": 1889413.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.72258677482605,
|
|
"epoch": 0.8981521272024066,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999809767387633,
|
|
"loss": 6.7197,
|
|
"mean_token_accuracy": 0.11978086978197097,
|
|
"num_tokens": 1898283.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.7828127384185795,
|
|
"epoch": 0.9024495058014611,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999764076985337,
|
|
"loss": 6.7543,
|
|
"mean_token_accuracy": 0.12366160079836845,
|
|
"num_tokens": 1907175.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.719510412216186,
|
|
"epoch": 0.9067468844005157,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999713473997519,
|
|
"loss": 6.8002,
|
|
"mean_token_accuracy": 0.12121418863534927,
|
|
"num_tokens": 1918223.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.836523628234863,
|
|
"epoch": 0.9110442629995703,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999657958534677,
|
|
"loss": 6.7295,
|
|
"mean_token_accuracy": 0.11709223091602325,
|
|
"num_tokens": 1928801.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.680103588104248,
|
|
"epoch": 0.9153416415986249,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999597530718034,
|
|
"loss": 6.7014,
|
|
"mean_token_accuracy": 0.12364157661795616,
|
|
"num_tokens": 1937406.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.803171920776367,
|
|
"epoch": 0.9196390201976794,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499953219067954,
|
|
"loss": 6.6924,
|
|
"mean_token_accuracy": 0.12274096831679344,
|
|
"num_tokens": 1947184.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.794332647323609,
|
|
"epoch": 0.923936398796734,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999461938561873,
|
|
"loss": 6.719,
|
|
"mean_token_accuracy": 0.12224036455154419,
|
|
"num_tokens": 1956293.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.758011388778686,
|
|
"epoch": 0.9282337773957886,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999386774518432,
|
|
"loss": 6.7009,
|
|
"mean_token_accuracy": 0.12048266977071762,
|
|
"num_tokens": 1964791.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.697661304473877,
|
|
"epoch": 0.9325311559948432,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999306698713349,
|
|
"loss": 6.6068,
|
|
"mean_token_accuracy": 0.1272837094962597,
|
|
"num_tokens": 1973754.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.7667817115783695,
|
|
"epoch": 0.9368285345938977,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999221711321477,
|
|
"loss": 6.6887,
|
|
"mean_token_accuracy": 0.12394253239035606,
|
|
"num_tokens": 1983035.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.704556512832641,
|
|
"epoch": 0.9411259131929522,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999131812528393,
|
|
"loss": 6.7139,
|
|
"mean_token_accuracy": 0.12714729234576225,
|
|
"num_tokens": 1992584.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.736967897415161,
|
|
"epoch": 0.9454232917920069,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00049990370025304,
|
|
"loss": 6.7395,
|
|
"mean_token_accuracy": 0.1247944861650467,
|
|
"num_tokens": 2001876.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.762415361404419,
|
|
"epoch": 0.9497206703910615,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998937281534526,
|
|
"loss": 6.6302,
|
|
"mean_token_accuracy": 0.13029390051960946,
|
|
"num_tokens": 2011067.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.732195615768433,
|
|
"epoch": 0.954018048990116,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998832649758521,
|
|
"loss": 6.7277,
|
|
"mean_token_accuracy": 0.11973504275083542,
|
|
"num_tokens": 2020763.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.740197038650512,
|
|
"epoch": 0.9583154275891707,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004998723107430862,
|
|
"loss": 6.6922,
|
|
"mean_token_accuracy": 0.12620774358510972,
|
|
"num_tokens": 2029534.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.712693786621093,
|
|
"epoch": 0.9626128061882252,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998608654790741,
|
|
"loss": 6.6464,
|
|
"mean_token_accuracy": 0.12175018936395646,
|
|
"num_tokens": 2039143.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.659943866729736,
|
|
"epoch": 0.9669101847872797,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499848929208808,
|
|
"loss": 6.6024,
|
|
"mean_token_accuracy": 0.13268646001815795,
|
|
"num_tokens": 2048253.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.671807098388672,
|
|
"epoch": 0.9712075633863343,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998365019583519,
|
|
"loss": 6.6707,
|
|
"mean_token_accuracy": 0.13119613155722618,
|
|
"num_tokens": 2057234.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.749271821975708,
|
|
"epoch": 0.975504941985389,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998235837548417,
|
|
"loss": 6.7138,
|
|
"mean_token_accuracy": 0.1277630090713501,
|
|
"num_tokens": 2065431.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.724707078933716,
|
|
"epoch": 0.9798023205844435,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000499810174626486,
|
|
"loss": 6.7137,
|
|
"mean_token_accuracy": 0.12455461621284485,
|
|
"num_tokens": 2074723.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.69525032043457,
|
|
"epoch": 0.984099699183498,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997962746025646,
|
|
"loss": 6.5975,
|
|
"mean_token_accuracy": 0.13215404078364373,
|
|
"num_tokens": 2084509.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.723079824447632,
|
|
"epoch": 0.9883970777825526,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997818837134298,
|
|
"loss": 6.7185,
|
|
"mean_token_accuracy": 0.12199998944997788,
|
|
"num_tokens": 2093110.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.635384130477905,
|
|
"epoch": 0.9926944563816072,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997670019905057,
|
|
"loss": 6.5836,
|
|
"mean_token_accuracy": 0.12477853298187255,
|
|
"num_tokens": 2102355.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.679490280151367,
|
|
"epoch": 0.9969918349806618,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997516294662876,
|
|
"loss": 6.6247,
|
|
"mean_token_accuracy": 0.1251891292631626,
|
|
"num_tokens": 2110418.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.690038098229302,
|
|
"epoch": 1.0008594757198108,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004997357661743433,
|
|
"loss": 6.5994,
|
|
"mean_token_accuracy": 0.13212493724293178,
|
|
"num_tokens": 2117866.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.635151147842407,
|
|
"epoch": 1.0051568543188656,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997194121493118,
|
|
"loss": 6.4157,
|
|
"mean_token_accuracy": 0.1288985140621662,
|
|
"num_tokens": 2126082.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.5850855827331545,
|
|
"epoch": 1.0094542329179201,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997025674269037,
|
|
"loss": 6.3647,
|
|
"mean_token_accuracy": 0.13999661505222322,
|
|
"num_tokens": 2134042.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.608558988571167,
|
|
"epoch": 1.0137516115169747,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996852320439013,
|
|
"loss": 6.4719,
|
|
"mean_token_accuracy": 0.13355429843068123,
|
|
"num_tokens": 2142570.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.5898195743560795,
|
|
"epoch": 1.0180489901160292,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996674060381578,
|
|
"loss": 6.408,
|
|
"mean_token_accuracy": 0.13155465126037597,
|
|
"num_tokens": 2151310.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.65716142654419,
|
|
"epoch": 1.0223463687150838,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996490894485985,
|
|
"loss": 6.4649,
|
|
"mean_token_accuracy": 0.1299987778067589,
|
|
"num_tokens": 2160662.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.589327764511109,
|
|
"epoch": 1.0266437473141383,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996302823152193,
|
|
"loss": 6.4217,
|
|
"mean_token_accuracy": 0.1353737846016884,
|
|
"num_tokens": 2170067.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.548572969436646,
|
|
"epoch": 1.0309411259131929,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996109846790873,
|
|
"loss": 6.3812,
|
|
"mean_token_accuracy": 0.13803449124097825,
|
|
"num_tokens": 2178850.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.5560492992401125,
|
|
"epoch": 1.0352385045122476,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004995911965823412,
|
|
"loss": 6.4036,
|
|
"mean_token_accuracy": 0.13800363689661027,
|
|
"num_tokens": 2188307.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.570348930358887,
|
|
"epoch": 1.0395358831113022,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004995709180681899,
|
|
"loss": 6.4088,
|
|
"mean_token_accuracy": 0.14277965128421782,
|
|
"num_tokens": 2197026.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.5593053817749025,
|
|
"epoch": 1.0438332617103567,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499550149180914,
|
|
"loss": 6.3881,
|
|
"mean_token_accuracy": 0.1331556722521782,
|
|
"num_tokens": 2205537.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.566903448104858,
|
|
"epoch": 1.0481306403094113,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995288899658641,
|
|
"loss": 6.3982,
|
|
"mean_token_accuracy": 0.13826436847448348,
|
|
"num_tokens": 2214508.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.582273006439209,
|
|
"epoch": 1.0524280189084658,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004995071404694619,
|
|
"loss": 6.529,
|
|
"mean_token_accuracy": 0.12735923379659653,
|
|
"num_tokens": 2223084.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.660481309890747,
|
|
"epoch": 1.0567253975075204,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004994849007391996,
|
|
"loss": 6.4492,
|
|
"mean_token_accuracy": 0.1277565985918045,
|
|
"num_tokens": 2231406.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.501922702789306,
|
|
"epoch": 1.061022776106575,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004994621708236401,
|
|
"loss": 6.363,
|
|
"mean_token_accuracy": 0.1401769794523716,
|
|
"num_tokens": 2239867.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.563766050338745,
|
|
"epoch": 1.0653201547056295,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000499438950772416,
|
|
"loss": 6.4318,
|
|
"mean_token_accuracy": 0.1355653516948223,
|
|
"num_tokens": 2248844.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.487766885757447,
|
|
"epoch": 1.0696175333046842,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004994152406362311,
|
|
"loss": 6.3553,
|
|
"mean_token_accuracy": 0.14220800772309303,
|
|
"num_tokens": 2257599.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.559006404876709,
|
|
"epoch": 1.0739149119037388,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004993910404668586,
|
|
"loss": 6.3984,
|
|
"mean_token_accuracy": 0.1350975051522255,
|
|
"num_tokens": 2266510.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.513402509689331,
|
|
"epoch": 1.0782122905027933,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499366350317142,
|
|
"loss": 6.3973,
|
|
"mean_token_accuracy": 0.134781850874424,
|
|
"num_tokens": 2275462.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.512863445281982,
|
|
"epoch": 1.0825096691018479,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993411702409948,
|
|
"loss": 6.3671,
|
|
"mean_token_accuracy": 0.136747994273901,
|
|
"num_tokens": 2283826.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.527937221527099,
|
|
"epoch": 1.0868070477009024,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004993155002934002,
|
|
"loss": 6.3742,
|
|
"mean_token_accuracy": 0.13640071898698808,
|
|
"num_tokens": 2292967.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.6068323135375975,
|
|
"epoch": 1.091104426299957,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0004992893405304111,
|
|
"loss": 6.5223,
|
|
"mean_token_accuracy": 0.13157014027237893,
|
|
"num_tokens": 2302336.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.5013409614562985,
|
|
"epoch": 1.0954018048990115,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00049926269100915,
|
|
"loss": 6.4184,
|
|
"mean_token_accuracy": 0.13976007997989653,
|
|
"num_tokens": 2311465.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.582226037979126,
|
|
"epoch": 1.0996991834980663,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004992355517878087,
|
|
"loss": 6.5322,
|
|
"mean_token_accuracy": 0.13058929219841958,
|
|
"num_tokens": 2320281.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.528811502456665,
|
|
"epoch": 1.1039965620971208,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004992079229256484,
|
|
"loss": 6.4264,
|
|
"mean_token_accuracy": 0.12976702973246573,
|
|
"num_tokens": 2329755.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.4910472393035885,
|
|
"epoch": 1.1082939406961754,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991798044829996,
|
|
"loss": 6.3767,
|
|
"mean_token_accuracy": 0.13432964831590652,
|
|
"num_tokens": 2338807.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.610663795471192,
|
|
"epoch": 1.11259131929523,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004991511965212618,
|
|
"loss": 6.4672,
|
|
"mean_token_accuracy": 0.13277963325381278,
|
|
"num_tokens": 2348056.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.493175792694092,
|
|
"epoch": 1.1168886978942845,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991220991029032,
|
|
"loss": 6.4859,
|
|
"mean_token_accuracy": 0.13005012422800064,
|
|
"num_tokens": 2357780.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.564814758300781,
|
|
"epoch": 1.121186076493339,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000499092512291461,
|
|
"loss": 6.4499,
|
|
"mean_token_accuracy": 0.13376415893435478,
|
|
"num_tokens": 2367060.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.5433674335479735,
|
|
"epoch": 1.1254834550923936,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499062436151541,
|
|
"loss": 6.4314,
|
|
"mean_token_accuracy": 0.13260492384433747,
|
|
"num_tokens": 2375751.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.533551168441773,
|
|
"epoch": 1.129780833691448,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004990318707488173,
|
|
"loss": 6.5099,
|
|
"mean_token_accuracy": 0.1266904965043068,
|
|
"num_tokens": 2385013.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.557472038269043,
|
|
"epoch": 1.1340782122905029,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004990008161500327,
|
|
"loss": 6.3801,
|
|
"mean_token_accuracy": 0.14031620547175408,
|
|
"num_tokens": 2392935.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.503352212905884,
|
|
"epoch": 1.1383755908895574,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000498969272422998,
|
|
"loss": 6.4049,
|
|
"mean_token_accuracy": 0.14203781262040138,
|
|
"num_tokens": 2401560.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.473355054855347,
|
|
"epoch": 1.142672969488612,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004989372396365921,
|
|
"loss": 6.3262,
|
|
"mean_token_accuracy": 0.13685949817299842,
|
|
"num_tokens": 2410050.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.538544416427612,
|
|
"epoch": 1.1469703480876665,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989047178607618,
|
|
"loss": 6.4325,
|
|
"mean_token_accuracy": 0.136402727663517,
|
|
"num_tokens": 2418980.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.51959867477417,
|
|
"epoch": 1.151267726686721,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004988717071665215,
|
|
"loss": 6.4253,
|
|
"mean_token_accuracy": 0.1328268252313137,
|
|
"num_tokens": 2427992.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.489779472351074,
|
|
"epoch": 1.1555651052857756,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004988382076259537,
|
|
"loss": 6.3409,
|
|
"mean_token_accuracy": 0.13519960269331932,
|
|
"num_tokens": 2436368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.419038581848144,
|
|
"epoch": 1.1598624838848304,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004988042193122077,
|
|
"loss": 6.3336,
|
|
"mean_token_accuracy": 0.14399233236908912,
|
|
"num_tokens": 2445499.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.545976400375366,
|
|
"epoch": 1.164159862483885,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004987697422995005,
|
|
"loss": 6.3652,
|
|
"mean_token_accuracy": 0.13806355968117714,
|
|
"num_tokens": 2454312.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.509993267059326,
|
|
"epoch": 1.1684572410829395,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004987347766631161,
|
|
"loss": 6.4259,
|
|
"mean_token_accuracy": 0.13582523986697198,
|
|
"num_tokens": 2462922.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.490610218048095,
|
|
"epoch": 1.172754619681994,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004986993224794055,
|
|
"loss": 6.4833,
|
|
"mean_token_accuracy": 0.12911048233509065,
|
|
"num_tokens": 2472195.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.525020790100098,
|
|
"epoch": 1.1770519982810486,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004986633798257865,
|
|
"loss": 6.3819,
|
|
"mean_token_accuracy": 0.13833871781826018,
|
|
"num_tokens": 2481021.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.4724901676177975,
|
|
"epoch": 1.181349376880103,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004986269487807434,
|
|
"loss": 6.3939,
|
|
"mean_token_accuracy": 0.13369214460253714,
|
|
"num_tokens": 2490250.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.521751642227173,
|
|
"epoch": 1.1856467554791577,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000498590029423827,
|
|
"loss": 6.4626,
|
|
"mean_token_accuracy": 0.1330279603600502,
|
|
"num_tokens": 2499122.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.55017466545105,
|
|
"epoch": 1.1899441340782122,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004985526218356546,
|
|
"loss": 6.4269,
|
|
"mean_token_accuracy": 0.13100680485367774,
|
|
"num_tokens": 2508454.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.4670604228973385,
|
|
"epoch": 1.1942415126772667,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004985147260979093,
|
|
"loss": 6.3665,
|
|
"mean_token_accuracy": 0.1456628531217575,
|
|
"num_tokens": 2517353.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.5169178485870365,
|
|
"epoch": 1.1985388912763215,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004984763422933402,
|
|
"loss": 6.3758,
|
|
"mean_token_accuracy": 0.14275188893079757,
|
|
"num_tokens": 2526321.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.4620006561279295,
|
|
"epoch": 1.202836269875376,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004984374705057623,
|
|
"loss": 6.4184,
|
|
"mean_token_accuracy": 0.13711269348859786,
|
|
"num_tokens": 2535924.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.438448047637939,
|
|
"epoch": 1.2071336484744306,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004983981108200561,
|
|
"loss": 6.4106,
|
|
"mean_token_accuracy": 0.13143872320652009,
|
|
"num_tokens": 2545606.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.424742078781128,
|
|
"epoch": 1.2114310270734852,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004983582633221672,
|
|
"loss": 6.3839,
|
|
"mean_token_accuracy": 0.13689693436026573,
|
|
"num_tokens": 2554947.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.620230293273925,
|
|
"epoch": 1.2157284056725397,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983179280991068,
|
|
"loss": 6.5392,
|
|
"mean_token_accuracy": 0.12959686368703843,
|
|
"num_tokens": 2564462.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.524547338485718,
|
|
"epoch": 1.2200257842715942,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004982771052389508,
|
|
"loss": 6.3797,
|
|
"mean_token_accuracy": 0.141092449426651,
|
|
"num_tokens": 2573124.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.482010555267334,
|
|
"epoch": 1.224323162870649,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004982357948308401,
|
|
"loss": 6.4728,
|
|
"mean_token_accuracy": 0.12631098851561545,
|
|
"num_tokens": 2581829.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.566068410873413,
|
|
"epoch": 1.2286205414697036,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004981939969649799,
|
|
"loss": 6.3285,
|
|
"mean_token_accuracy": 0.13993503898382187,
|
|
"num_tokens": 2590631.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.452121257781982,
|
|
"epoch": 1.232917920068758,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004981517117326404,
|
|
"loss": 6.4557,
|
|
"mean_token_accuracy": 0.13728617280721664,
|
|
"num_tokens": 2600684.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.486893844604492,
|
|
"epoch": 1.2372152986678127,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981089392261553,
|
|
"loss": 6.3508,
|
|
"mean_token_accuracy": 0.14346109181642533,
|
|
"num_tokens": 2609667.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.4833098411560055,
|
|
"epoch": 1.2415126772668672,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000498065679538923,
|
|
"loss": 6.447,
|
|
"mean_token_accuracy": 0.13537913784384728,
|
|
"num_tokens": 2620025.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.500622749328613,
|
|
"epoch": 1.2458100558659218,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004980219327654049,
|
|
"loss": 6.3454,
|
|
"mean_token_accuracy": 0.13805984035134317,
|
|
"num_tokens": 2629032.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.423587512969971,
|
|
"epoch": 1.2501074344649763,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000497977699001127,
|
|
"loss": 6.3492,
|
|
"mean_token_accuracy": 0.13877770975232123,
|
|
"num_tokens": 2638303.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.512550354003906,
|
|
"epoch": 1.2544048130640308,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004979329783426778,
|
|
"loss": 6.3561,
|
|
"mean_token_accuracy": 0.1412653423845768,
|
|
"num_tokens": 2647902.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.393660068511963,
|
|
"epoch": 1.2587021916630854,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004978877708877094,
|
|
"loss": 6.4141,
|
|
"mean_token_accuracy": 0.13558436930179596,
|
|
"num_tokens": 2657902.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.485628700256347,
|
|
"epoch": 1.2629995702621402,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004978420767349368,
|
|
"loss": 6.3533,
|
|
"mean_token_accuracy": 0.13661525994539261,
|
|
"num_tokens": 2667082.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.461724853515625,
|
|
"epoch": 1.2672969488611947,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004977958959841379,
|
|
"loss": 6.4176,
|
|
"mean_token_accuracy": 0.1369339980185032,
|
|
"num_tokens": 2676855.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.50676383972168,
|
|
"epoch": 1.2715943274602493,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000497749228736153,
|
|
"loss": 6.3514,
|
|
"mean_token_accuracy": 0.13876376375555993,
|
|
"num_tokens": 2685750.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.416869592666626,
|
|
"epoch": 1.2758917060593038,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004977020750928845,
|
|
"loss": 6.433,
|
|
"mean_token_accuracy": 0.1434354990720749,
|
|
"num_tokens": 2695272.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.51352972984314,
|
|
"epoch": 1.2801890846583583,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976544351572973,
|
|
"loss": 6.3404,
|
|
"mean_token_accuracy": 0.14171317517757415,
|
|
"num_tokens": 2704806.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.346848726272583,
|
|
"epoch": 1.2844864632574131,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004976063090334179,
|
|
"loss": 6.4255,
|
|
"mean_token_accuracy": 0.13577363267540932,
|
|
"num_tokens": 2713521.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.52107515335083,
|
|
"epoch": 1.2887838418564677,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004975576968263346,
|
|
"loss": 6.3883,
|
|
"mean_token_accuracy": 0.13231359645724297,
|
|
"num_tokens": 2721848.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2887838418564677,
|
|
"eval_entropy": 6.345255539223954,
|
|
"eval_loss": 6.523230075836182,
|
|
"eval_mean_token_accuracy": 0.13721702877130057,
|
|
"eval_num_tokens": 2721848.0,
|
|
"eval_runtime": 2.6548,
|
|
"eval_samples_per_second": 1336.804,
|
|
"eval_steps_per_second": 167.242,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11630,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 4136966316288000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|