5035 lines
137 KiB
JSON
5035 lines
137 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.05434310059994783,
|
|
"eval_steps": 500,
|
|
"global_step": 2500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692033958435058,
|
|
"epoch": 0.00010868620119989566,
|
|
"grad_norm": 14.125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7755,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 7936.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691999912261963,
|
|
"epoch": 0.00021737240239979132,
|
|
"grad_norm": 15.1875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.6915,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 15561.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690576362609864,
|
|
"epoch": 0.000326058603599687,
|
|
"grad_norm": 10.0,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4237,
|
|
"mean_token_accuracy": 0.03714140746742487,
|
|
"num_tokens": 23101.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.662552452087402,
|
|
"epoch": 0.00043474480479958265,
|
|
"grad_norm": 6.4375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.13,
|
|
"mean_token_accuracy": 0.053733503073453905,
|
|
"num_tokens": 31122.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.53134651184082,
|
|
"epoch": 0.0005434310059994783,
|
|
"grad_norm": 4.21875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8666,
|
|
"mean_token_accuracy": 0.049097494408488276,
|
|
"num_tokens": 39424.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.499347591400147,
|
|
"epoch": 0.000652117207199374,
|
|
"grad_norm": 3.40625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.6573,
|
|
"mean_token_accuracy": 0.05333031602203846,
|
|
"num_tokens": 46113.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.45414752960205,
|
|
"epoch": 0.0007608034083992696,
|
|
"grad_norm": 4.03125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.5644,
|
|
"mean_token_accuracy": 0.050568538904190066,
|
|
"num_tokens": 53459.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.488454818725586,
|
|
"epoch": 0.0008694896095991653,
|
|
"grad_norm": 2.875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.5746,
|
|
"mean_token_accuracy": 0.054808919131755826,
|
|
"num_tokens": 61325.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.44335298538208,
|
|
"epoch": 0.0009781758107990609,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.5243,
|
|
"mean_token_accuracy": 0.05896916501224041,
|
|
"num_tokens": 68741.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.389917469024658,
|
|
"epoch": 0.0010868620119989566,
|
|
"grad_norm": 2.703125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.4498,
|
|
"mean_token_accuracy": 0.07500801645219327,
|
|
"num_tokens": 76827.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.32972011566162,
|
|
"epoch": 0.0011955482131988523,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4108,
|
|
"mean_token_accuracy": 0.06764259599149228,
|
|
"num_tokens": 84543.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.223662376403809,
|
|
"epoch": 0.001304234414398748,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.3763,
|
|
"mean_token_accuracy": 0.06775642447173595,
|
|
"num_tokens": 92505.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.253492069244384,
|
|
"epoch": 0.0014129206155986437,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.292,
|
|
"mean_token_accuracy": 0.07278058938682079,
|
|
"num_tokens": 100004.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.179506492614745,
|
|
"epoch": 0.0015216068167985392,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.1464,
|
|
"mean_token_accuracy": 0.07609819024801254,
|
|
"num_tokens": 106865.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.167201709747314,
|
|
"epoch": 0.0016302930179984349,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.184,
|
|
"mean_token_accuracy": 0.07288527600467205,
|
|
"num_tokens": 115745.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.253772735595703,
|
|
"epoch": 0.0017389792191983306,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.1019,
|
|
"mean_token_accuracy": 0.07272576205432416,
|
|
"num_tokens": 124724.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.141531181335449,
|
|
"epoch": 0.0018476654203982263,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.9728,
|
|
"mean_token_accuracy": 0.07559517920017242,
|
|
"num_tokens": 131933.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.05212688446045,
|
|
"epoch": 0.0019563516215981218,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9554,
|
|
"mean_token_accuracy": 0.07529733926057816,
|
|
"num_tokens": 139726.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.118945980072022,
|
|
"epoch": 0.0020650378227980177,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8937,
|
|
"mean_token_accuracy": 0.07433236017823219,
|
|
"num_tokens": 147276.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.070981788635255,
|
|
"epoch": 0.002173724023997913,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8032,
|
|
"mean_token_accuracy": 0.07382776252925397,
|
|
"num_tokens": 155449.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.931837844848634,
|
|
"epoch": 0.002282410225197809,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.6788,
|
|
"mean_token_accuracy": 0.06880230605602264,
|
|
"num_tokens": 163547.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.92315607070923,
|
|
"epoch": 0.0023910964263977046,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.5975,
|
|
"mean_token_accuracy": 0.0769424244761467,
|
|
"num_tokens": 171777.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.855748653411865,
|
|
"epoch": 0.0024997826275976,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.4659,
|
|
"mean_token_accuracy": 0.0743311032652855,
|
|
"num_tokens": 179337.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.745928478240966,
|
|
"epoch": 0.002608468828797496,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.3905,
|
|
"mean_token_accuracy": 0.07685245871543885,
|
|
"num_tokens": 186939.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.60959987640381,
|
|
"epoch": 0.0027171550299973915,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.2612,
|
|
"mean_token_accuracy": 0.07991581708192826,
|
|
"num_tokens": 194755.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.507084274291993,
|
|
"epoch": 0.0028258412311972874,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.1596,
|
|
"mean_token_accuracy": 0.07951611280441284,
|
|
"num_tokens": 202610.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.229420471191407,
|
|
"epoch": 0.002934527432397183,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.0589,
|
|
"mean_token_accuracy": 0.07813627794384956,
|
|
"num_tokens": 211305.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.083441066741944,
|
|
"epoch": 0.0030432136335970784,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.9516,
|
|
"mean_token_accuracy": 0.07972251251339912,
|
|
"num_tokens": 218447.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.977557182312012,
|
|
"epoch": 0.0031518998347969743,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.9735,
|
|
"mean_token_accuracy": 0.07708753384649754,
|
|
"num_tokens": 226108.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.854303741455078,
|
|
"epoch": 0.0032605860359968698,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.9018,
|
|
"mean_token_accuracy": 0.0749433733522892,
|
|
"num_tokens": 233213.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.737026977539063,
|
|
"epoch": 0.0033692722371967657,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.9657,
|
|
"mean_token_accuracy": 0.07061286456882954,
|
|
"num_tokens": 241483.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.660515594482423,
|
|
"epoch": 0.003477958438396661,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8216,
|
|
"mean_token_accuracy": 0.08022509105503559,
|
|
"num_tokens": 249537.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.48324089050293,
|
|
"epoch": 0.0035866446395965567,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7574,
|
|
"mean_token_accuracy": 0.08265799134969712,
|
|
"num_tokens": 256775.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.382307147979736,
|
|
"epoch": 0.0036953308407964526,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7628,
|
|
"mean_token_accuracy": 0.08266141265630722,
|
|
"num_tokens": 263931.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.312588787078857,
|
|
"epoch": 0.003804017041996348,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.7557,
|
|
"mean_token_accuracy": 0.08323334604501724,
|
|
"num_tokens": 272293.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.368490505218507,
|
|
"epoch": 0.0039127032431962435,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7964,
|
|
"mean_token_accuracy": 0.07600408792495728,
|
|
"num_tokens": 280599.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.208444499969483,
|
|
"epoch": 0.00402138944439614,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7604,
|
|
"mean_token_accuracy": 0.07719066366553307,
|
|
"num_tokens": 287961.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.142381763458252,
|
|
"epoch": 0.004130075645596035,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.684,
|
|
"mean_token_accuracy": 0.08424306884407998,
|
|
"num_tokens": 297280.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.162713718414306,
|
|
"epoch": 0.004238761846795931,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.7526,
|
|
"mean_token_accuracy": 0.08235807195305825,
|
|
"num_tokens": 304699.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.149117755889893,
|
|
"epoch": 0.004347448047995826,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.6864,
|
|
"mean_token_accuracy": 0.08296664133667946,
|
|
"num_tokens": 313026.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.115727186203003,
|
|
"epoch": 0.004456134249195722,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.7092,
|
|
"mean_token_accuracy": 0.08166395723819733,
|
|
"num_tokens": 320738.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.076230096817017,
|
|
"epoch": 0.004564820450395618,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.6861,
|
|
"mean_token_accuracy": 0.07836256250739097,
|
|
"num_tokens": 328740.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.025853300094605,
|
|
"epoch": 0.004673506651595514,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.6627,
|
|
"mean_token_accuracy": 0.08186743408441544,
|
|
"num_tokens": 336819.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.132151794433593,
|
|
"epoch": 0.004782192852795409,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.5969,
|
|
"mean_token_accuracy": 0.08840056881308556,
|
|
"num_tokens": 343978.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.009299278259277,
|
|
"epoch": 0.004890879053995305,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.6897,
|
|
"mean_token_accuracy": 0.07886088080704212,
|
|
"num_tokens": 353783.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.969748306274414,
|
|
"epoch": 0.0049995652551952,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.5909,
|
|
"mean_token_accuracy": 0.08582306802272796,
|
|
"num_tokens": 361371.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.9828167915344235,
|
|
"epoch": 0.0051082514563950965,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.6185,
|
|
"mean_token_accuracy": 0.07745145447552204,
|
|
"num_tokens": 367717.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.051587057113647,
|
|
"epoch": 0.005216937657594992,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.6177,
|
|
"mean_token_accuracy": 0.08022010251879692,
|
|
"num_tokens": 375529.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.905487108230591,
|
|
"epoch": 0.0053256238587948874,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.5073,
|
|
"mean_token_accuracy": 0.08242316544055939,
|
|
"num_tokens": 383031.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.966064023971557,
|
|
"epoch": 0.005434310059994783,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.6472,
|
|
"mean_token_accuracy": 0.0840654157102108,
|
|
"num_tokens": 391914.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.939682960510254,
|
|
"epoch": 0.005542996261194678,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.6111,
|
|
"mean_token_accuracy": 0.08445862084627151,
|
|
"num_tokens": 399447.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.95916223526001,
|
|
"epoch": 0.005651682462394575,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.6053,
|
|
"mean_token_accuracy": 0.08623345196247101,
|
|
"num_tokens": 407636.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.863192081451416,
|
|
"epoch": 0.00576036866359447,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.5498,
|
|
"mean_token_accuracy": 0.08252564668655396,
|
|
"num_tokens": 416650.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.940821886062622,
|
|
"epoch": 0.005869054864794366,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.4851,
|
|
"mean_token_accuracy": 0.08535419404506683,
|
|
"num_tokens": 424619.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.923966264724731,
|
|
"epoch": 0.005977741065994261,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.5731,
|
|
"mean_token_accuracy": 0.08694756105542183,
|
|
"num_tokens": 432854.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.861378288269043,
|
|
"epoch": 0.006086427267194157,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.5646,
|
|
"mean_token_accuracy": 0.08049703910946845,
|
|
"num_tokens": 440472.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.979065895080566,
|
|
"epoch": 0.006195113468394053,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.4921,
|
|
"mean_token_accuracy": 0.08637023121118545,
|
|
"num_tokens": 447843.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.926018571853637,
|
|
"epoch": 0.0063037996695939486,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.6005,
|
|
"mean_token_accuracy": 0.08329573571681977,
|
|
"num_tokens": 455272.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.811395597457886,
|
|
"epoch": 0.006412485870793844,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.4287,
|
|
"mean_token_accuracy": 0.08994799107313156,
|
|
"num_tokens": 462190.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.937566328048706,
|
|
"epoch": 0.0065211720719937395,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.5291,
|
|
"mean_token_accuracy": 0.08383448868989944,
|
|
"num_tokens": 470562.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.770080280303955,
|
|
"epoch": 0.006629858273193635,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.433,
|
|
"mean_token_accuracy": 0.09061657935380936,
|
|
"num_tokens": 478523.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.844554090499878,
|
|
"epoch": 0.006738544474393531,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.517,
|
|
"mean_token_accuracy": 0.08662905022501946,
|
|
"num_tokens": 486759.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.828848934173584,
|
|
"epoch": 0.006847230675593427,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.5283,
|
|
"mean_token_accuracy": 0.08165703043341636,
|
|
"num_tokens": 495168.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.823043441772461,
|
|
"epoch": 0.006955916876793322,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.4207,
|
|
"mean_token_accuracy": 0.08956636488437653,
|
|
"num_tokens": 503490.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.771451330184936,
|
|
"epoch": 0.007064603077993218,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.5539,
|
|
"mean_token_accuracy": 0.08244621530175208,
|
|
"num_tokens": 512522.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.877453756332398,
|
|
"epoch": 0.007173289279193113,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.4913,
|
|
"mean_token_accuracy": 0.08120876513421535,
|
|
"num_tokens": 521100.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.771992826461792,
|
|
"epoch": 0.00728197548039301,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.396,
|
|
"mean_token_accuracy": 0.0948154278099537,
|
|
"num_tokens": 528027.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.7505473613739015,
|
|
"epoch": 0.007390661681592905,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.3994,
|
|
"mean_token_accuracy": 0.08797064200043678,
|
|
"num_tokens": 535541.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.820106267929077,
|
|
"epoch": 0.007499347882792801,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.4177,
|
|
"mean_token_accuracy": 0.08866829052567482,
|
|
"num_tokens": 543253.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.8131262302398685,
|
|
"epoch": 0.007608034083992696,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.4312,
|
|
"mean_token_accuracy": 0.08869538456201553,
|
|
"num_tokens": 550414.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.678452491760254,
|
|
"epoch": 0.007716720285192592,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.4866,
|
|
"mean_token_accuracy": 0.08567069470882416,
|
|
"num_tokens": 558554.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.860786962509155,
|
|
"epoch": 0.007825406486392487,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.4562,
|
|
"mean_token_accuracy": 0.0866042397916317,
|
|
"num_tokens": 566297.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.691314792633056,
|
|
"epoch": 0.007934092687592383,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.399,
|
|
"mean_token_accuracy": 0.09093262627720833,
|
|
"num_tokens": 574600.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.761751461029053,
|
|
"epoch": 0.00804277888879228,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.4278,
|
|
"mean_token_accuracy": 0.08792822286486626,
|
|
"num_tokens": 583349.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.841218614578247,
|
|
"epoch": 0.008151465089992175,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.3269,
|
|
"mean_token_accuracy": 0.08868191167712211,
|
|
"num_tokens": 590169.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.639244318008423,
|
|
"epoch": 0.00826015129119207,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.2957,
|
|
"mean_token_accuracy": 0.09457754641771317,
|
|
"num_tokens": 597508.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.685401201248169,
|
|
"epoch": 0.008368837492391966,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.375,
|
|
"mean_token_accuracy": 0.09122001230716706,
|
|
"num_tokens": 605412.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.795647668838501,
|
|
"epoch": 0.008477523693591862,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.3872,
|
|
"mean_token_accuracy": 0.09148699790239334,
|
|
"num_tokens": 613512.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.641701507568359,
|
|
"epoch": 0.008586209894791757,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.3275,
|
|
"mean_token_accuracy": 0.09618531838059426,
|
|
"num_tokens": 621117.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.689551591873169,
|
|
"epoch": 0.008694896095991653,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.3832,
|
|
"mean_token_accuracy": 0.08407204449176789,
|
|
"num_tokens": 629970.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.637691450119019,
|
|
"epoch": 0.008803582297191548,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.3219,
|
|
"mean_token_accuracy": 0.09131594821810722,
|
|
"num_tokens": 638184.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.5977555274963375,
|
|
"epoch": 0.008912268498391444,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.2837,
|
|
"mean_token_accuracy": 0.08929966911673545,
|
|
"num_tokens": 645953.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.6464704990386965,
|
|
"epoch": 0.00902095469959134,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.3305,
|
|
"mean_token_accuracy": 0.08722574934363365,
|
|
"num_tokens": 654291.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.672793436050415,
|
|
"epoch": 0.009129640900791236,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.3299,
|
|
"mean_token_accuracy": 0.08851959705352783,
|
|
"num_tokens": 661393.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.531088209152221,
|
|
"epoch": 0.009238327101991132,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.3139,
|
|
"mean_token_accuracy": 0.0963095597922802,
|
|
"num_tokens": 669913.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.74607515335083,
|
|
"epoch": 0.009347013303191027,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.3663,
|
|
"mean_token_accuracy": 0.09074652194976807,
|
|
"num_tokens": 678745.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.417545986175537,
|
|
"epoch": 0.009455699504390923,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.2468,
|
|
"mean_token_accuracy": 0.09117906540632248,
|
|
"num_tokens": 686191.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.690976524353028,
|
|
"epoch": 0.009564385705590818,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.2921,
|
|
"mean_token_accuracy": 0.08830784261226654,
|
|
"num_tokens": 694776.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.471565675735474,
|
|
"epoch": 0.009673071906790714,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.2498,
|
|
"mean_token_accuracy": 0.08958808556199074,
|
|
"num_tokens": 702481.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.601398086547851,
|
|
"epoch": 0.00978175810799061,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.2892,
|
|
"mean_token_accuracy": 0.08943363651633263,
|
|
"num_tokens": 710304.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.6107948303222654,
|
|
"epoch": 0.009890444309190505,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.2655,
|
|
"mean_token_accuracy": 0.0950858935713768,
|
|
"num_tokens": 718462.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.4872644424438475,
|
|
"epoch": 0.0099991305103904,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.2699,
|
|
"mean_token_accuracy": 0.09967489093542099,
|
|
"num_tokens": 726798.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.636318254470825,
|
|
"epoch": 0.010107816711590296,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.2289,
|
|
"mean_token_accuracy": 0.09415075033903123,
|
|
"num_tokens": 735054.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.467661046981812,
|
|
"epoch": 0.010216502912790193,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.2689,
|
|
"mean_token_accuracy": 0.08854644820094108,
|
|
"num_tokens": 743701.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.5553264141082765,
|
|
"epoch": 0.010325189113990088,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.3436,
|
|
"mean_token_accuracy": 0.08918146193027496,
|
|
"num_tokens": 751548.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.606575393676758,
|
|
"epoch": 0.010433875315189984,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.1245,
|
|
"mean_token_accuracy": 0.09939143806695938,
|
|
"num_tokens": 758580.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.376218938827515,
|
|
"epoch": 0.01054256151638988,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.1472,
|
|
"mean_token_accuracy": 0.09824580624699593,
|
|
"num_tokens": 766240.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.460130596160889,
|
|
"epoch": 0.010651247717589775,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.1908,
|
|
"mean_token_accuracy": 0.09887742102146149,
|
|
"num_tokens": 774206.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.499665117263794,
|
|
"epoch": 0.01075993391878967,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.135,
|
|
"mean_token_accuracy": 0.10001260414719582,
|
|
"num_tokens": 782005.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.313957262039184,
|
|
"epoch": 0.010868620119989566,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.1614,
|
|
"mean_token_accuracy": 0.10134897753596306,
|
|
"num_tokens": 790043.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.518388891220093,
|
|
"epoch": 0.010977306321189461,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.1951,
|
|
"mean_token_accuracy": 0.10190339386463165,
|
|
"num_tokens": 798248.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.401588821411133,
|
|
"epoch": 0.011085992522389357,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.1494,
|
|
"mean_token_accuracy": 0.10203814506530762,
|
|
"num_tokens": 806023.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.4617040157318115,
|
|
"epoch": 0.011194678723589252,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.26,
|
|
"mean_token_accuracy": 0.08982759192585946,
|
|
"num_tokens": 813719.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.584583950042725,
|
|
"epoch": 0.01130336492478915,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.2398,
|
|
"mean_token_accuracy": 0.09505607783794404,
|
|
"num_tokens": 821938.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.490258741378784,
|
|
"epoch": 0.011412051125989045,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.1423,
|
|
"mean_token_accuracy": 0.09996162354946136,
|
|
"num_tokens": 830263.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.332801628112793,
|
|
"epoch": 0.01152073732718894,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.1478,
|
|
"mean_token_accuracy": 0.10072804763913154,
|
|
"num_tokens": 838040.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.352591896057129,
|
|
"epoch": 0.011629423528388836,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.1318,
|
|
"mean_token_accuracy": 0.09845451191067696,
|
|
"num_tokens": 846265.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.524213409423828,
|
|
"epoch": 0.011738109729588731,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.2088,
|
|
"mean_token_accuracy": 0.09776617884635926,
|
|
"num_tokens": 853815.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.406747961044312,
|
|
"epoch": 0.011846795930788627,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.1712,
|
|
"mean_token_accuracy": 0.1003396026790142,
|
|
"num_tokens": 861186.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.5143406867980955,
|
|
"epoch": 0.011955482131988522,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.1261,
|
|
"mean_token_accuracy": 0.09955452755093575,
|
|
"num_tokens": 868619.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.397789716720581,
|
|
"epoch": 0.012064168333188418,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.1921,
|
|
"mean_token_accuracy": 0.1024585336446762,
|
|
"num_tokens": 876521.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.418723487854004,
|
|
"epoch": 0.012172854534388313,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.1915,
|
|
"mean_token_accuracy": 0.09342663809657097,
|
|
"num_tokens": 884060.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.401181697845459,
|
|
"epoch": 0.012281540735588209,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.1436,
|
|
"mean_token_accuracy": 0.09810893088579178,
|
|
"num_tokens": 892218.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.3739972591400145,
|
|
"epoch": 0.012390226936788106,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.153,
|
|
"mean_token_accuracy": 0.09526406824588776,
|
|
"num_tokens": 899472.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.431902933120727,
|
|
"epoch": 0.012498913137988002,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.1138,
|
|
"mean_token_accuracy": 0.10175292342901229,
|
|
"num_tokens": 907371.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.30218768119812,
|
|
"epoch": 0.012607599339187897,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.118,
|
|
"mean_token_accuracy": 0.10189053043723106,
|
|
"num_tokens": 914376.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.398109769821167,
|
|
"epoch": 0.012716285540387793,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.1022,
|
|
"mean_token_accuracy": 0.09777733609080315,
|
|
"num_tokens": 921992.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.318228054046631,
|
|
"epoch": 0.012824971741587688,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.1565,
|
|
"mean_token_accuracy": 0.10031871050596237,
|
|
"num_tokens": 929966.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.380307865142822,
|
|
"epoch": 0.012933657942787584,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.0322,
|
|
"mean_token_accuracy": 0.10295850262045861,
|
|
"num_tokens": 936500.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.2398364543914795,
|
|
"epoch": 0.013042344143987479,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.0636,
|
|
"mean_token_accuracy": 0.10201629102230073,
|
|
"num_tokens": 944685.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.3245521068573,
|
|
"epoch": 0.013151030345187375,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.09,
|
|
"mean_token_accuracy": 0.10841693356633186,
|
|
"num_tokens": 952379.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.226641845703125,
|
|
"epoch": 0.01325971654638727,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.0067,
|
|
"mean_token_accuracy": 0.10042177364230156,
|
|
"num_tokens": 960310.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.356974458694458,
|
|
"epoch": 0.013368402747587165,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.1009,
|
|
"mean_token_accuracy": 0.09788779467344284,
|
|
"num_tokens": 967581.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.381576633453369,
|
|
"epoch": 0.013477088948787063,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.0731,
|
|
"mean_token_accuracy": 0.10214511007070541,
|
|
"num_tokens": 975351.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.133514547348023,
|
|
"epoch": 0.013585775149986958,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.0016,
|
|
"mean_token_accuracy": 0.10031676739454269,
|
|
"num_tokens": 982962.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.36269965171814,
|
|
"epoch": 0.013694461351186854,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.0818,
|
|
"mean_token_accuracy": 0.0940178707242012,
|
|
"num_tokens": 991081.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.196866226196289,
|
|
"epoch": 0.01380314755238675,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.0304,
|
|
"mean_token_accuracy": 0.0977426715195179,
|
|
"num_tokens": 999169.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.23091402053833,
|
|
"epoch": 0.013911833753586645,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.0051,
|
|
"mean_token_accuracy": 0.10395268872380256,
|
|
"num_tokens": 1007086.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.237727499008178,
|
|
"epoch": 0.01402051995478654,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.0415,
|
|
"mean_token_accuracy": 0.10429325178265572,
|
|
"num_tokens": 1014937.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.289627265930176,
|
|
"epoch": 0.014129206155986436,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.0527,
|
|
"mean_token_accuracy": 0.10648775473237038,
|
|
"num_tokens": 1022271.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.281875371932983,
|
|
"epoch": 0.014237892357186331,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.0394,
|
|
"mean_token_accuracy": 0.09872807115316391,
|
|
"num_tokens": 1029946.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.201374578475952,
|
|
"epoch": 0.014346578558386227,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.0564,
|
|
"mean_token_accuracy": 0.09937692657113076,
|
|
"num_tokens": 1037659.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.247886657714844,
|
|
"epoch": 0.014455264759586122,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.0678,
|
|
"mean_token_accuracy": 0.10430879592895508,
|
|
"num_tokens": 1045420.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.278404426574707,
|
|
"epoch": 0.01456395096078602,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.0594,
|
|
"mean_token_accuracy": 0.10522538945078849,
|
|
"num_tokens": 1052681.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.258583641052246,
|
|
"epoch": 0.014672637161985915,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.9977,
|
|
"mean_token_accuracy": 0.10505606159567833,
|
|
"num_tokens": 1060114.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.2787620544433596,
|
|
"epoch": 0.01478132336318581,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.0666,
|
|
"mean_token_accuracy": 0.10130468308925629,
|
|
"num_tokens": 1068833.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.07578821182251,
|
|
"epoch": 0.014890009564385706,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.0062,
|
|
"mean_token_accuracy": 0.10653988644480705,
|
|
"num_tokens": 1075850.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.184060382843017,
|
|
"epoch": 0.014998695765585601,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.9698,
|
|
"mean_token_accuracy": 0.10846218839287758,
|
|
"num_tokens": 1083956.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.173658752441407,
|
|
"epoch": 0.015107381966785497,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.003,
|
|
"mean_token_accuracy": 0.09950482398271561,
|
|
"num_tokens": 1091744.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.154714155197143,
|
|
"epoch": 0.015216068167985392,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.9642,
|
|
"mean_token_accuracy": 0.10058322846889496,
|
|
"num_tokens": 1099029.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.211764144897461,
|
|
"epoch": 0.015324754369185288,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.0826,
|
|
"mean_token_accuracy": 0.10248045325279236,
|
|
"num_tokens": 1106329.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.2008462905883786,
|
|
"epoch": 0.015433440570385183,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.0031,
|
|
"mean_token_accuracy": 0.10641120374202728,
|
|
"num_tokens": 1114672.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.307254886627197,
|
|
"epoch": 0.01554212677158508,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.122,
|
|
"mean_token_accuracy": 0.09702686741948127,
|
|
"num_tokens": 1123782.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.048694181442261,
|
|
"epoch": 0.015650812972784974,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.9479,
|
|
"mean_token_accuracy": 0.10803859531879426,
|
|
"num_tokens": 1131211.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.091199684143066,
|
|
"epoch": 0.01575949917398487,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.9317,
|
|
"mean_token_accuracy": 0.10954350158572197,
|
|
"num_tokens": 1138955.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.106390285491943,
|
|
"epoch": 0.015868185375184765,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.9803,
|
|
"mean_token_accuracy": 0.10243654400110244,
|
|
"num_tokens": 1147376.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.204882383346558,
|
|
"epoch": 0.015976871576384662,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.9549,
|
|
"mean_token_accuracy": 0.10707944706082344,
|
|
"num_tokens": 1155196.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.020097827911377,
|
|
"epoch": 0.01608555777758456,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.9581,
|
|
"mean_token_accuracy": 0.10678377524018287,
|
|
"num_tokens": 1162503.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.215338659286499,
|
|
"epoch": 0.016194243978784453,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.006,
|
|
"mean_token_accuracy": 0.10291345119476318,
|
|
"num_tokens": 1170387.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.980220603942871,
|
|
"epoch": 0.01630293017998435,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.8741,
|
|
"mean_token_accuracy": 0.11537401303648949,
|
|
"num_tokens": 1176921.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.128726530075073,
|
|
"epoch": 0.016411616381184244,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.0252,
|
|
"mean_token_accuracy": 0.1050336092710495,
|
|
"num_tokens": 1185063.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.0691938400268555,
|
|
"epoch": 0.01652030258238414,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.9361,
|
|
"mean_token_accuracy": 0.1111590951681137,
|
|
"num_tokens": 1193008.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.12714900970459,
|
|
"epoch": 0.016628988783584035,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.9554,
|
|
"mean_token_accuracy": 0.1107974775135517,
|
|
"num_tokens": 1200868.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.084636926651001,
|
|
"epoch": 0.016737674984783932,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.9061,
|
|
"mean_token_accuracy": 0.11364880874752999,
|
|
"num_tokens": 1208687.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.034700918197632,
|
|
"epoch": 0.016846361185983826,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.9637,
|
|
"mean_token_accuracy": 0.10925066992640495,
|
|
"num_tokens": 1216804.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.038867473602295,
|
|
"epoch": 0.016955047387183723,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.8871,
|
|
"mean_token_accuracy": 0.11413683071732521,
|
|
"num_tokens": 1224505.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.978211402893066,
|
|
"epoch": 0.017063733588383617,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.8853,
|
|
"mean_token_accuracy": 0.11565895602107049,
|
|
"num_tokens": 1232706.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.117547416687012,
|
|
"epoch": 0.017172419789583514,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.8904,
|
|
"mean_token_accuracy": 0.11539890021085739,
|
|
"num_tokens": 1239551.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.993913698196411,
|
|
"epoch": 0.01728110599078341,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.8834,
|
|
"mean_token_accuracy": 0.11389343962073326,
|
|
"num_tokens": 1247909.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.153699827194214,
|
|
"epoch": 0.017389792191983305,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.0588,
|
|
"mean_token_accuracy": 0.10906191244721412,
|
|
"num_tokens": 1256523.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.072479391098023,
|
|
"epoch": 0.017498478393183203,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.8826,
|
|
"mean_token_accuracy": 0.11302874237298965,
|
|
"num_tokens": 1263879.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.110428047180176,
|
|
"epoch": 0.017607164594383096,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.9637,
|
|
"mean_token_accuracy": 0.1068289540708065,
|
|
"num_tokens": 1271287.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.977171850204468,
|
|
"epoch": 0.017715850795582994,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.8514,
|
|
"mean_token_accuracy": 0.1114407129585743,
|
|
"num_tokens": 1278828.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.056746530532837,
|
|
"epoch": 0.017824536996782887,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.9732,
|
|
"mean_token_accuracy": 0.10797528550028801,
|
|
"num_tokens": 1286827.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.063602304458618,
|
|
"epoch": 0.017933223197982785,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.9201,
|
|
"mean_token_accuracy": 0.10415855050086975,
|
|
"num_tokens": 1294546.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.962296867370606,
|
|
"epoch": 0.01804190939918268,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.8398,
|
|
"mean_token_accuracy": 0.1096323385834694,
|
|
"num_tokens": 1301574.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.106445598602295,
|
|
"epoch": 0.018150595600382576,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.8577,
|
|
"mean_token_accuracy": 0.11122734472155571,
|
|
"num_tokens": 1308224.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.868801546096802,
|
|
"epoch": 0.018259281801582473,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.923,
|
|
"mean_token_accuracy": 0.10960033088922501,
|
|
"num_tokens": 1317335.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.096353387832641,
|
|
"epoch": 0.018367968002782366,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.8735,
|
|
"mean_token_accuracy": 0.10573652610182763,
|
|
"num_tokens": 1324825.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.913944911956787,
|
|
"epoch": 0.018476654203982264,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.9635,
|
|
"mean_token_accuracy": 0.10770962685346604,
|
|
"num_tokens": 1332818.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.95415940284729,
|
|
"epoch": 0.018585340405182157,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.769,
|
|
"mean_token_accuracy": 0.12482169717550277,
|
|
"num_tokens": 1339858.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.976625156402588,
|
|
"epoch": 0.018694026606382055,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.9342,
|
|
"mean_token_accuracy": 0.10149514004588127,
|
|
"num_tokens": 1348542.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.979064416885376,
|
|
"epoch": 0.01880271280758195,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.8578,
|
|
"mean_token_accuracy": 0.11146032214164733,
|
|
"num_tokens": 1355851.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.964614057540894,
|
|
"epoch": 0.018911399008781846,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.867,
|
|
"mean_token_accuracy": 0.11173160448670387,
|
|
"num_tokens": 1363844.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.057435846328735,
|
|
"epoch": 0.01902008520998174,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.8731,
|
|
"mean_token_accuracy": 0.11148432195186615,
|
|
"num_tokens": 1372525.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.942585229873657,
|
|
"epoch": 0.019128771411181637,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.9121,
|
|
"mean_token_accuracy": 0.10839233919978142,
|
|
"num_tokens": 1381371.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.933096694946289,
|
|
"epoch": 0.01923745761238153,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.8116,
|
|
"mean_token_accuracy": 0.11815176531672478,
|
|
"num_tokens": 1389140.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.891059732437133,
|
|
"epoch": 0.019346143813581428,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.8372,
|
|
"mean_token_accuracy": 0.11450439095497131,
|
|
"num_tokens": 1396397.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.9327497482299805,
|
|
"epoch": 0.019454830014781325,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.9194,
|
|
"mean_token_accuracy": 0.11007297709584236,
|
|
"num_tokens": 1404600.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.999571323394775,
|
|
"epoch": 0.01956351621598122,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.8758,
|
|
"mean_token_accuracy": 0.11412210613489152,
|
|
"num_tokens": 1413063.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.883196830749512,
|
|
"epoch": 0.019672202417181116,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.7573,
|
|
"mean_token_accuracy": 0.11876690089702606,
|
|
"num_tokens": 1419875.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.874788761138916,
|
|
"epoch": 0.01978088861838101,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.8001,
|
|
"mean_token_accuracy": 0.10891184434294701,
|
|
"num_tokens": 1427405.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.950111055374146,
|
|
"epoch": 0.019889574819580907,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.9241,
|
|
"mean_token_accuracy": 0.10468844398856163,
|
|
"num_tokens": 1435721.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.985974359512329,
|
|
"epoch": 0.0199982610207808,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.8313,
|
|
"mean_token_accuracy": 0.10862488150596619,
|
|
"num_tokens": 1443091.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.866302967071533,
|
|
"epoch": 0.020106947221980698,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.8088,
|
|
"mean_token_accuracy": 0.1172551192343235,
|
|
"num_tokens": 1450452.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.967225980758667,
|
|
"epoch": 0.02021563342318059,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.9108,
|
|
"mean_token_accuracy": 0.11415560767054558,
|
|
"num_tokens": 1459130.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.973930406570434,
|
|
"epoch": 0.02032431962438049,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.8542,
|
|
"mean_token_accuracy": 0.10872401073575019,
|
|
"num_tokens": 1466950.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.964308834075927,
|
|
"epoch": 0.020433005825580386,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.9507,
|
|
"mean_token_accuracy": 0.10664009302854538,
|
|
"num_tokens": 1476679.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.870850515365601,
|
|
"epoch": 0.02054169202678028,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.7456,
|
|
"mean_token_accuracy": 0.11636312678456306,
|
|
"num_tokens": 1484000.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.836710262298584,
|
|
"epoch": 0.020650378227980177,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.8694,
|
|
"mean_token_accuracy": 0.11150090396404266,
|
|
"num_tokens": 1491563.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.927568483352661,
|
|
"epoch": 0.02075906442918007,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.8676,
|
|
"mean_token_accuracy": 0.1152837723493576,
|
|
"num_tokens": 1499879.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.948094701766967,
|
|
"epoch": 0.020867750630379968,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.8011,
|
|
"mean_token_accuracy": 0.1114250123500824,
|
|
"num_tokens": 1507764.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.80685830116272,
|
|
"epoch": 0.02097643683157986,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.8263,
|
|
"mean_token_accuracy": 0.1110868975520134,
|
|
"num_tokens": 1516098.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.883790922164917,
|
|
"epoch": 0.02108512303277976,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.7996,
|
|
"mean_token_accuracy": 0.11565712839365005,
|
|
"num_tokens": 1523800.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.985372495651245,
|
|
"epoch": 0.021193809233979653,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.9104,
|
|
"mean_token_accuracy": 0.10778488591313362,
|
|
"num_tokens": 1531435.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.863921737670898,
|
|
"epoch": 0.02130249543517955,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.797,
|
|
"mean_token_accuracy": 0.11670528054237365,
|
|
"num_tokens": 1539757.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.790615463256836,
|
|
"epoch": 0.021411181636379444,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.7501,
|
|
"mean_token_accuracy": 0.11390289589762688,
|
|
"num_tokens": 1547272.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.816875362396241,
|
|
"epoch": 0.02151986783757934,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.7661,
|
|
"mean_token_accuracy": 0.11140070185065269,
|
|
"num_tokens": 1555286.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.8604302406311035,
|
|
"epoch": 0.021628554038779238,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.7601,
|
|
"mean_token_accuracy": 0.11161275953054428,
|
|
"num_tokens": 1563762.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.862444543838501,
|
|
"epoch": 0.021737240239979132,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.8113,
|
|
"mean_token_accuracy": 0.10896943360567093,
|
|
"num_tokens": 1572996.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.944311189651489,
|
|
"epoch": 0.02184592644117903,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.8662,
|
|
"mean_token_accuracy": 0.10712903067469597,
|
|
"num_tokens": 1581009.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.861703729629516,
|
|
"epoch": 0.021954612642378923,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.839,
|
|
"mean_token_accuracy": 0.11141621619462967,
|
|
"num_tokens": 1589133.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.793341779708863,
|
|
"epoch": 0.02206329884357882,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.7883,
|
|
"mean_token_accuracy": 0.1176340863108635,
|
|
"num_tokens": 1596843.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.771137332916259,
|
|
"epoch": 0.022171985044778714,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.6697,
|
|
"mean_token_accuracy": 0.12174412459135056,
|
|
"num_tokens": 1604287.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.842767333984375,
|
|
"epoch": 0.02228067124597861,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.7256,
|
|
"mean_token_accuracy": 0.11848887726664543,
|
|
"num_tokens": 1612995.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.747415065765381,
|
|
"epoch": 0.022389357447178505,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.7856,
|
|
"mean_token_accuracy": 0.11309662461280823,
|
|
"num_tokens": 1622090.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.809288883209229,
|
|
"epoch": 0.022498043648378402,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.7228,
|
|
"mean_token_accuracy": 0.1198489598929882,
|
|
"num_tokens": 1630697.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.7372705936431885,
|
|
"epoch": 0.0226067298495783,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.7425,
|
|
"mean_token_accuracy": 0.11601528003811837,
|
|
"num_tokens": 1638711.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.753994655609131,
|
|
"epoch": 0.022715416050778193,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.6434,
|
|
"mean_token_accuracy": 0.12457199022173882,
|
|
"num_tokens": 1646536.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.740097522735596,
|
|
"epoch": 0.02282410225197809,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.7043,
|
|
"mean_token_accuracy": 0.11510038226842881,
|
|
"num_tokens": 1654633.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.73102216720581,
|
|
"epoch": 0.022932788453177984,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.7956,
|
|
"mean_token_accuracy": 0.11422985568642616,
|
|
"num_tokens": 1662844.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.789561605453491,
|
|
"epoch": 0.02304147465437788,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.7318,
|
|
"mean_token_accuracy": 0.12340603768825531,
|
|
"num_tokens": 1670771.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.733522748947143,
|
|
"epoch": 0.023150160855577775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.6846,
|
|
"mean_token_accuracy": 0.11971172913908959,
|
|
"num_tokens": 1678136.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.7756870746612545,
|
|
"epoch": 0.023258847056777672,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.6925,
|
|
"mean_token_accuracy": 0.11457556933164596,
|
|
"num_tokens": 1686262.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.733363199234009,
|
|
"epoch": 0.023367533257977566,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.7026,
|
|
"mean_token_accuracy": 0.12492116540670395,
|
|
"num_tokens": 1693674.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.771343755722046,
|
|
"epoch": 0.023476219459177463,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.7377,
|
|
"mean_token_accuracy": 0.11399596929550171,
|
|
"num_tokens": 1701470.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.687263298034668,
|
|
"epoch": 0.02358490566037736,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.6237,
|
|
"mean_token_accuracy": 0.12396905422210694,
|
|
"num_tokens": 1708849.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.770354318618774,
|
|
"epoch": 0.023693591861577254,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.6882,
|
|
"mean_token_accuracy": 0.12379547283053398,
|
|
"num_tokens": 1716360.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.719215774536133,
|
|
"epoch": 0.02380227806277715,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.7194,
|
|
"mean_token_accuracy": 0.11905450150370597,
|
|
"num_tokens": 1724218.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.804384231567383,
|
|
"epoch": 0.023910964263977045,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.7272,
|
|
"mean_token_accuracy": 0.11393096148967743,
|
|
"num_tokens": 1732431.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.6787989139556885,
|
|
"epoch": 0.024019650465176942,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.7156,
|
|
"mean_token_accuracy": 0.11123198568820954,
|
|
"num_tokens": 1740466.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.843336868286133,
|
|
"epoch": 0.024128336666376836,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.7396,
|
|
"mean_token_accuracy": 0.1168380431830883,
|
|
"num_tokens": 1747952.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.669610834121704,
|
|
"epoch": 0.024237022867576733,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.7167,
|
|
"mean_token_accuracy": 0.12077742591500282,
|
|
"num_tokens": 1755903.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.817348575592041,
|
|
"epoch": 0.024345709068776627,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.8025,
|
|
"mean_token_accuracy": 0.11085736006498337,
|
|
"num_tokens": 1764451.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.821982717514038,
|
|
"epoch": 0.024454395269976524,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.8682,
|
|
"mean_token_accuracy": 0.11098600178956985,
|
|
"num_tokens": 1773573.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.814310121536255,
|
|
"epoch": 0.024563081471176418,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.7621,
|
|
"mean_token_accuracy": 0.11357782185077667,
|
|
"num_tokens": 1781176.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.854786062240601,
|
|
"epoch": 0.024671767672376315,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.7348,
|
|
"mean_token_accuracy": 0.11998602896928787,
|
|
"num_tokens": 1788118.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.704822015762329,
|
|
"epoch": 0.024780453873576212,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.7596,
|
|
"mean_token_accuracy": 0.12081285044550896,
|
|
"num_tokens": 1797245.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.7488603591918945,
|
|
"epoch": 0.024889140074776106,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.7508,
|
|
"mean_token_accuracy": 0.10823504626750946,
|
|
"num_tokens": 1806358.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.78481707572937,
|
|
"epoch": 0.024997826275976003,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.7131,
|
|
"mean_token_accuracy": 0.12163210511207581,
|
|
"num_tokens": 1814500.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.817404508590698,
|
|
"epoch": 0.025106512477175897,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.7377,
|
|
"mean_token_accuracy": 0.11625442504882813,
|
|
"num_tokens": 1821983.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.750685548782348,
|
|
"epoch": 0.025215198678375794,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.7311,
|
|
"mean_token_accuracy": 0.11820982620120049,
|
|
"num_tokens": 1829416.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.63856954574585,
|
|
"epoch": 0.025323884879575688,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.6535,
|
|
"mean_token_accuracy": 0.12064194157719613,
|
|
"num_tokens": 1836888.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.7434063911437985,
|
|
"epoch": 0.025432571080775585,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.6959,
|
|
"mean_token_accuracy": 0.11727289259433746,
|
|
"num_tokens": 1845318.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.721282291412353,
|
|
"epoch": 0.02554125728197548,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.7254,
|
|
"mean_token_accuracy": 0.11171131953597069,
|
|
"num_tokens": 1855005.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.7714166164398195,
|
|
"epoch": 0.025649943483175376,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.6286,
|
|
"mean_token_accuracy": 0.11785526797175408,
|
|
"num_tokens": 1863104.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.647221803665161,
|
|
"epoch": 0.025758629684375273,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.7371,
|
|
"mean_token_accuracy": 0.12050389349460602,
|
|
"num_tokens": 1870833.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.822182941436767,
|
|
"epoch": 0.025867315885575167,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.6016,
|
|
"mean_token_accuracy": 0.12365994974970818,
|
|
"num_tokens": 1877812.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.610793733596802,
|
|
"epoch": 0.025976002086775064,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.6732,
|
|
"mean_token_accuracy": 0.12029042765498162,
|
|
"num_tokens": 1886265.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.711217069625855,
|
|
"epoch": 0.026084688287974958,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.6321,
|
|
"mean_token_accuracy": 0.12245473191142082,
|
|
"num_tokens": 1894378.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.634425687789917,
|
|
"epoch": 0.026193374489174855,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.5894,
|
|
"mean_token_accuracy": 0.12096654176712036,
|
|
"num_tokens": 1902298.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.674074411392212,
|
|
"epoch": 0.02630206069037475,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.7116,
|
|
"mean_token_accuracy": 0.1144709900021553,
|
|
"num_tokens": 1909320.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.827318143844605,
|
|
"epoch": 0.026410746891574646,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.6938,
|
|
"mean_token_accuracy": 0.119134671241045,
|
|
"num_tokens": 1917053.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.668541002273559,
|
|
"epoch": 0.02651943309277454,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.6332,
|
|
"mean_token_accuracy": 0.12302515879273415,
|
|
"num_tokens": 1925725.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.772843980789185,
|
|
"epoch": 0.026628119293974437,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.666,
|
|
"mean_token_accuracy": 0.1218823440372944,
|
|
"num_tokens": 1933317.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.60233359336853,
|
|
"epoch": 0.02673680549517433,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.7042,
|
|
"mean_token_accuracy": 0.11369993463158608,
|
|
"num_tokens": 1941273.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.729761362075806,
|
|
"epoch": 0.026845491696374228,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.671,
|
|
"mean_token_accuracy": 0.1214449904859066,
|
|
"num_tokens": 1950474.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.670256090164185,
|
|
"epoch": 0.026954177897574125,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.7009,
|
|
"mean_token_accuracy": 0.11673137396574021,
|
|
"num_tokens": 1959019.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.690532398223877,
|
|
"epoch": 0.02706286409877402,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.568,
|
|
"mean_token_accuracy": 0.12435560747981071,
|
|
"num_tokens": 1967265.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.617759323120117,
|
|
"epoch": 0.027171550299973916,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.5434,
|
|
"mean_token_accuracy": 0.12447248473763466,
|
|
"num_tokens": 1974239.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.75258617401123,
|
|
"epoch": 0.02728023650117381,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.6743,
|
|
"mean_token_accuracy": 0.1195609726011753,
|
|
"num_tokens": 1981982.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.5735856056213375,
|
|
"epoch": 0.027388922702373707,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.5964,
|
|
"mean_token_accuracy": 0.12709427028894424,
|
|
"num_tokens": 1989488.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.660828065872193,
|
|
"epoch": 0.0274976089035736,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.6906,
|
|
"mean_token_accuracy": 0.11730890870094299,
|
|
"num_tokens": 1997447.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.7316429138183596,
|
|
"epoch": 0.0276062951047735,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.7033,
|
|
"mean_token_accuracy": 0.1155233234167099,
|
|
"num_tokens": 2006443.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.715520000457763,
|
|
"epoch": 0.027714981305973392,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.6559,
|
|
"mean_token_accuracy": 0.11586152538657188,
|
|
"num_tokens": 2015339.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.641583824157715,
|
|
"epoch": 0.02782366750717329,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.5521,
|
|
"mean_token_accuracy": 0.12834264934062958,
|
|
"num_tokens": 2022976.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.611687469482422,
|
|
"epoch": 0.027932353708373187,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.6215,
|
|
"mean_token_accuracy": 0.12682543843984603,
|
|
"num_tokens": 2031252.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.703845119476318,
|
|
"epoch": 0.02804103990957308,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.611,
|
|
"mean_token_accuracy": 0.1212866634130478,
|
|
"num_tokens": 2038465.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.586824417114258,
|
|
"epoch": 0.028149726110772978,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.5856,
|
|
"mean_token_accuracy": 0.12294368520379066,
|
|
"num_tokens": 2045921.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.655889320373535,
|
|
"epoch": 0.02825841231197287,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.5343,
|
|
"mean_token_accuracy": 0.1270897440612316,
|
|
"num_tokens": 2054127.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.691068410873413,
|
|
"epoch": 0.02836709851317277,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.6147,
|
|
"mean_token_accuracy": 0.12068610191345215,
|
|
"num_tokens": 2061889.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.5537497997283936,
|
|
"epoch": 0.028475784714372662,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.5846,
|
|
"mean_token_accuracy": 0.12348945364356041,
|
|
"num_tokens": 2069620.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.72159972190857,
|
|
"epoch": 0.02858447091557256,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.5326,
|
|
"mean_token_accuracy": 0.13243458196520805,
|
|
"num_tokens": 2077371.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.541279745101929,
|
|
"epoch": 0.028693157116772453,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.6154,
|
|
"mean_token_accuracy": 0.12915401235222818,
|
|
"num_tokens": 2085727.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.756937503814697,
|
|
"epoch": 0.02880184331797235,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.6528,
|
|
"mean_token_accuracy": 0.12067537754774094,
|
|
"num_tokens": 2093858.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.603250503540039,
|
|
"epoch": 0.028910529519172244,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.6029,
|
|
"mean_token_accuracy": 0.12590081617236137,
|
|
"num_tokens": 2101776.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.552526473999023,
|
|
"epoch": 0.02901921572037214,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.5071,
|
|
"mean_token_accuracy": 0.12628837302327156,
|
|
"num_tokens": 2108932.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.61759204864502,
|
|
"epoch": 0.02912790192157204,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.5878,
|
|
"mean_token_accuracy": 0.12770043164491654,
|
|
"num_tokens": 2116786.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.6530732154846195,
|
|
"epoch": 0.029236588122771932,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.6136,
|
|
"mean_token_accuracy": 0.11667786464095116,
|
|
"num_tokens": 2124884.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.576787042617798,
|
|
"epoch": 0.02934527432397183,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.6245,
|
|
"mean_token_accuracy": 0.12192277386784553,
|
|
"num_tokens": 2132173.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.7595985412597654,
|
|
"epoch": 0.029453960525171723,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.5765,
|
|
"mean_token_accuracy": 0.12141433283686638,
|
|
"num_tokens": 2140554.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.474835062026978,
|
|
"epoch": 0.02956264672637162,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.5187,
|
|
"mean_token_accuracy": 0.12662884443998337,
|
|
"num_tokens": 2148350.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.634714126586914,
|
|
"epoch": 0.029671332927571514,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.5471,
|
|
"mean_token_accuracy": 0.1328614644706249,
|
|
"num_tokens": 2156652.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.579633331298828,
|
|
"epoch": 0.02978001912877141,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.6294,
|
|
"mean_token_accuracy": 0.11975409835577011,
|
|
"num_tokens": 2165406.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.649616336822509,
|
|
"epoch": 0.029888705329971305,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.5915,
|
|
"mean_token_accuracy": 0.1256055273115635,
|
|
"num_tokens": 2173663.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.660815572738647,
|
|
"epoch": 0.029997391531171202,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.5002,
|
|
"mean_token_accuracy": 0.12725673615932465,
|
|
"num_tokens": 2181042.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.612697267532349,
|
|
"epoch": 0.0301060777323711,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.6187,
|
|
"mean_token_accuracy": 0.11868590340018273,
|
|
"num_tokens": 2189686.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.637413787841797,
|
|
"epoch": 0.030214763933570993,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.6813,
|
|
"mean_token_accuracy": 0.12129974663257599,
|
|
"num_tokens": 2198924.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.662744569778442,
|
|
"epoch": 0.03032345013477089,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.6561,
|
|
"mean_token_accuracy": 0.11987720131874084,
|
|
"num_tokens": 2206770.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.626357221603394,
|
|
"epoch": 0.030432136335970784,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.5734,
|
|
"mean_token_accuracy": 0.12084893509745598,
|
|
"num_tokens": 2214398.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.651124620437622,
|
|
"epoch": 0.03054082253717068,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.5953,
|
|
"mean_token_accuracy": 0.12142176851630211,
|
|
"num_tokens": 2222338.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.602102756500244,
|
|
"epoch": 0.030649508738370575,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.5087,
|
|
"mean_token_accuracy": 0.13247907012701035,
|
|
"num_tokens": 2229815.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.539565515518189,
|
|
"epoch": 0.030758194939570473,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.4888,
|
|
"mean_token_accuracy": 0.1223328985273838,
|
|
"num_tokens": 2237663.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.608070755004883,
|
|
"epoch": 0.030866881140770366,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.5472,
|
|
"mean_token_accuracy": 0.13164723366498948,
|
|
"num_tokens": 2245703.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.681449699401855,
|
|
"epoch": 0.030975567341970264,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.5276,
|
|
"mean_token_accuracy": 0.12454903274774551,
|
|
"num_tokens": 2253178.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.496039152145386,
|
|
"epoch": 0.03108425354317016,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.5428,
|
|
"mean_token_accuracy": 0.12364711612462997,
|
|
"num_tokens": 2260564.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.588859939575196,
|
|
"epoch": 0.031192939744370055,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.5566,
|
|
"mean_token_accuracy": 0.11912669464945794,
|
|
"num_tokens": 2268988.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.649993419647217,
|
|
"epoch": 0.03130162594556995,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.508,
|
|
"mean_token_accuracy": 0.12953050956130027,
|
|
"num_tokens": 2276400.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.583322143554687,
|
|
"epoch": 0.03141031214676985,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.5249,
|
|
"mean_token_accuracy": 0.1304257981479168,
|
|
"num_tokens": 2284540.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.596342277526856,
|
|
"epoch": 0.03151899834796974,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.5765,
|
|
"mean_token_accuracy": 0.12414052039384842,
|
|
"num_tokens": 2293655.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.667462110519409,
|
|
"epoch": 0.031627684549169637,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.5642,
|
|
"mean_token_accuracy": 0.1251768171787262,
|
|
"num_tokens": 2302647.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.515790939331055,
|
|
"epoch": 0.03173637075036953,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.5314,
|
|
"mean_token_accuracy": 0.12692836448550224,
|
|
"num_tokens": 2311064.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.6385047912597654,
|
|
"epoch": 0.03184505695156943,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.5232,
|
|
"mean_token_accuracy": 0.13005399256944655,
|
|
"num_tokens": 2318608.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.426021480560303,
|
|
"epoch": 0.031953743152769325,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.4206,
|
|
"mean_token_accuracy": 0.1302155315876007,
|
|
"num_tokens": 2326806.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.695640182495117,
|
|
"epoch": 0.03206242935396922,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.5783,
|
|
"mean_token_accuracy": 0.12959000319242478,
|
|
"num_tokens": 2334568.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.571360015869141,
|
|
"epoch": 0.03217111555516912,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.5223,
|
|
"mean_token_accuracy": 0.12900073900818826,
|
|
"num_tokens": 2342361.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.571485424041748,
|
|
"epoch": 0.03227980175636901,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.4626,
|
|
"mean_token_accuracy": 0.13145333901047707,
|
|
"num_tokens": 2349593.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.519814300537109,
|
|
"epoch": 0.03238848795756891,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.5831,
|
|
"mean_token_accuracy": 0.12430679574608802,
|
|
"num_tokens": 2357192.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.629279899597168,
|
|
"epoch": 0.0324971741587688,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.4638,
|
|
"mean_token_accuracy": 0.12320095226168633,
|
|
"num_tokens": 2365248.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.538986921310425,
|
|
"epoch": 0.0326058603599687,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.4738,
|
|
"mean_token_accuracy": 0.12518679574131966,
|
|
"num_tokens": 2373505.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.536040925979615,
|
|
"epoch": 0.032714546561168595,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 6.5423,
|
|
"mean_token_accuracy": 0.1335771270096302,
|
|
"num_tokens": 2382949.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.558509492874146,
|
|
"epoch": 0.03282323276236849,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.4999,
|
|
"mean_token_accuracy": 0.1334672413766384,
|
|
"num_tokens": 2390134.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.495549535751342,
|
|
"epoch": 0.03293191896356838,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.4612,
|
|
"mean_token_accuracy": 0.1316902793943882,
|
|
"num_tokens": 2397742.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.587743139266967,
|
|
"epoch": 0.03304060516476828,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.5267,
|
|
"mean_token_accuracy": 0.12813749685883521,
|
|
"num_tokens": 2405223.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.575222682952881,
|
|
"epoch": 0.03314929136596818,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.533,
|
|
"mean_token_accuracy": 0.13205342888832092,
|
|
"num_tokens": 2413308.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.633901071548462,
|
|
"epoch": 0.03325797756716807,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.567,
|
|
"mean_token_accuracy": 0.11793626993894576,
|
|
"num_tokens": 2420808.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.5440003871917725,
|
|
"epoch": 0.03336666376836797,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.5334,
|
|
"mean_token_accuracy": 0.12041784524917602,
|
|
"num_tokens": 2429640.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.689853143692017,
|
|
"epoch": 0.033475349969567865,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.5633,
|
|
"mean_token_accuracy": 0.1279229164123535,
|
|
"num_tokens": 2437328.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.499489879608154,
|
|
"epoch": 0.03358403617076776,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.5067,
|
|
"mean_token_accuracy": 0.1295216828584671,
|
|
"num_tokens": 2445244.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.620385551452637,
|
|
"epoch": 0.03369272237196765,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.4369,
|
|
"mean_token_accuracy": 0.1301434338092804,
|
|
"num_tokens": 2453572.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.366896629333496,
|
|
"epoch": 0.03380140857316755,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.4015,
|
|
"mean_token_accuracy": 0.13501286879181862,
|
|
"num_tokens": 2460485.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.632570362091064,
|
|
"epoch": 0.03391009477436745,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.405,
|
|
"mean_token_accuracy": 0.12899469584226608,
|
|
"num_tokens": 2467370.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.527682161331176,
|
|
"epoch": 0.03401878097556734,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 6.5547,
|
|
"mean_token_accuracy": 0.1285148099064827,
|
|
"num_tokens": 2476221.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.6062336444854735,
|
|
"epoch": 0.034127467176767234,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 6.5629,
|
|
"mean_token_accuracy": 0.12118729501962662,
|
|
"num_tokens": 2483971.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.520406293869018,
|
|
"epoch": 0.034236153377967135,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.4336,
|
|
"mean_token_accuracy": 0.13079387471079826,
|
|
"num_tokens": 2491462.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.47422046661377,
|
|
"epoch": 0.03434483957916703,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 6.4056,
|
|
"mean_token_accuracy": 0.1340803399682045,
|
|
"num_tokens": 2499577.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.574036169052124,
|
|
"epoch": 0.03445352578036692,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 6.4668,
|
|
"mean_token_accuracy": 0.13153123632073402,
|
|
"num_tokens": 2508442.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.568566417694091,
|
|
"epoch": 0.03456221198156682,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.649,
|
|
"mean_token_accuracy": 0.1157750979065895,
|
|
"num_tokens": 2517291.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.581708860397339,
|
|
"epoch": 0.03467089818276672,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.4364,
|
|
"mean_token_accuracy": 0.13571289256215097,
|
|
"num_tokens": 2523844.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.505489349365234,
|
|
"epoch": 0.03477958438396661,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 6.5653,
|
|
"mean_token_accuracy": 0.12564576417207718,
|
|
"num_tokens": 2532061.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.633800792694092,
|
|
"epoch": 0.034888270585166505,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 6.4121,
|
|
"mean_token_accuracy": 0.13259042352437972,
|
|
"num_tokens": 2539795.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.473930311203003,
|
|
"epoch": 0.034996956786366405,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 6.4229,
|
|
"mean_token_accuracy": 0.12699868008494378,
|
|
"num_tokens": 2547627.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.554611539840698,
|
|
"epoch": 0.0351056429875663,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.379,
|
|
"mean_token_accuracy": 0.13499154895544052,
|
|
"num_tokens": 2555532.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.449095153808594,
|
|
"epoch": 0.03521432918876619,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.3399,
|
|
"mean_token_accuracy": 0.1383037745952606,
|
|
"num_tokens": 2563568.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.513947772979736,
|
|
"epoch": 0.035323015389966086,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 6.4384,
|
|
"mean_token_accuracy": 0.13582201898097992,
|
|
"num_tokens": 2571599.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.5380042552947994,
|
|
"epoch": 0.03543170159116599,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.4308,
|
|
"mean_token_accuracy": 0.13048155084252358,
|
|
"num_tokens": 2579417.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.477106666564941,
|
|
"epoch": 0.03554038779236588,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 6.4597,
|
|
"mean_token_accuracy": 0.13081906735897064,
|
|
"num_tokens": 2587935.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.538247537612915,
|
|
"epoch": 0.035649073993565775,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 6.3801,
|
|
"mean_token_accuracy": 0.12998153120279313,
|
|
"num_tokens": 2595076.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.523086023330689,
|
|
"epoch": 0.035757760194765675,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 6.421,
|
|
"mean_token_accuracy": 0.13086821809411048,
|
|
"num_tokens": 2602716.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.460543966293335,
|
|
"epoch": 0.03586644639596557,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.4638,
|
|
"mean_token_accuracy": 0.1328990265727043,
|
|
"num_tokens": 2610457.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.583772754669189,
|
|
"epoch": 0.03597513259716546,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 6.4479,
|
|
"mean_token_accuracy": 0.13444233164191247,
|
|
"num_tokens": 2618399.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.464978981018066,
|
|
"epoch": 0.03608381879836536,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.4389,
|
|
"mean_token_accuracy": 0.13300132974982262,
|
|
"num_tokens": 2625975.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.520250415802002,
|
|
"epoch": 0.03619250499956526,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 6.3903,
|
|
"mean_token_accuracy": 0.1342037595808506,
|
|
"num_tokens": 2633861.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.428459119796753,
|
|
"epoch": 0.03630119120076515,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.351,
|
|
"mean_token_accuracy": 0.135555599629879,
|
|
"num_tokens": 2641478.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.458524894714356,
|
|
"epoch": 0.036409877401965045,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.3658,
|
|
"mean_token_accuracy": 0.1361924558877945,
|
|
"num_tokens": 2649282.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.466368627548218,
|
|
"epoch": 0.036518563603164945,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 6.2352,
|
|
"mean_token_accuracy": 0.14827430844306946,
|
|
"num_tokens": 2656129.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.3706169605255125,
|
|
"epoch": 0.03662724980436484,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 6.4333,
|
|
"mean_token_accuracy": 0.1218526192009449,
|
|
"num_tokens": 2663070.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.563682889938354,
|
|
"epoch": 0.03673593600556473,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.354,
|
|
"mean_token_accuracy": 0.13783733993768693,
|
|
"num_tokens": 2670167.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.5183515548706055,
|
|
"epoch": 0.03684462220676463,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.484,
|
|
"mean_token_accuracy": 0.12654226571321486,
|
|
"num_tokens": 2677920.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.550176572799683,
|
|
"epoch": 0.03695330840796453,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 6.3382,
|
|
"mean_token_accuracy": 0.13361814990639687,
|
|
"num_tokens": 2684388.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.384865379333496,
|
|
"epoch": 0.03706199460916442,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 6.3741,
|
|
"mean_token_accuracy": 0.1353348322212696,
|
|
"num_tokens": 2692231.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.443640995025635,
|
|
"epoch": 0.037170680810364315,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 6.4508,
|
|
"mean_token_accuracy": 0.13217496126890182,
|
|
"num_tokens": 2700075.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.5465617179870605,
|
|
"epoch": 0.03727936701156421,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.4546,
|
|
"mean_token_accuracy": 0.1277106761932373,
|
|
"num_tokens": 2708951.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.462393426895142,
|
|
"epoch": 0.03738805321276411,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 6.4102,
|
|
"mean_token_accuracy": 0.13752326667308806,
|
|
"num_tokens": 2716294.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.569583129882813,
|
|
"epoch": 0.037496739413964,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 6.4967,
|
|
"mean_token_accuracy": 0.12161010652780532,
|
|
"num_tokens": 2724093.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.513576364517212,
|
|
"epoch": 0.0376054256151639,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 6.3602,
|
|
"mean_token_accuracy": 0.13403671607375145,
|
|
"num_tokens": 2731710.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.401008653640747,
|
|
"epoch": 0.0377141118163638,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.4097,
|
|
"mean_token_accuracy": 0.13267221450805664,
|
|
"num_tokens": 2740007.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.472448682785034,
|
|
"epoch": 0.03782279801756369,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.4079,
|
|
"mean_token_accuracy": 0.12942041605710983,
|
|
"num_tokens": 2747257.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.5088489055633545,
|
|
"epoch": 0.037931484218763585,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 6.48,
|
|
"mean_token_accuracy": 0.12736652791500092,
|
|
"num_tokens": 2755932.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.545922040939331,
|
|
"epoch": 0.03804017041996348,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.4148,
|
|
"mean_token_accuracy": 0.132827740162611,
|
|
"num_tokens": 2764215.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.4098350524902346,
|
|
"epoch": 0.03814885662116338,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.3495,
|
|
"mean_token_accuracy": 0.13987976163625718,
|
|
"num_tokens": 2771756.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.4801195621490475,
|
|
"epoch": 0.03825754282236327,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.4278,
|
|
"mean_token_accuracy": 0.1352240838110447,
|
|
"num_tokens": 2779545.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.543670701980591,
|
|
"epoch": 0.03836622902356317,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.4494,
|
|
"mean_token_accuracy": 0.13165312260389328,
|
|
"num_tokens": 2787709.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.437568712234497,
|
|
"epoch": 0.03847491522476306,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 6.3888,
|
|
"mean_token_accuracy": 0.1326165869832039,
|
|
"num_tokens": 2795822.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.432380104064942,
|
|
"epoch": 0.03858360142596296,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.3875,
|
|
"mean_token_accuracy": 0.13581312224268913,
|
|
"num_tokens": 2804438.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.4600687503814695,
|
|
"epoch": 0.038692287627162855,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 6.3309,
|
|
"mean_token_accuracy": 0.13892574086785317,
|
|
"num_tokens": 2812051.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.426763820648193,
|
|
"epoch": 0.03880097382836275,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 6.3632,
|
|
"mean_token_accuracy": 0.13622673973441124,
|
|
"num_tokens": 2820478.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.482985639572144,
|
|
"epoch": 0.03890966002956265,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.3994,
|
|
"mean_token_accuracy": 0.13121038153767586,
|
|
"num_tokens": 2828756.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.399754571914673,
|
|
"epoch": 0.03901834623076254,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 6.3088,
|
|
"mean_token_accuracy": 0.14511832892894744,
|
|
"num_tokens": 2836163.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.380357122421264,
|
|
"epoch": 0.03912703243196244,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.381,
|
|
"mean_token_accuracy": 0.13844038397073746,
|
|
"num_tokens": 2844323.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.459079074859619,
|
|
"epoch": 0.03923571863316233,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.3475,
|
|
"mean_token_accuracy": 0.13634610474109649,
|
|
"num_tokens": 2852158.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.420756530761719,
|
|
"epoch": 0.03934440483436223,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 6.3166,
|
|
"mean_token_accuracy": 0.13739913552999497,
|
|
"num_tokens": 2859418.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.426111793518066,
|
|
"epoch": 0.039453091035562125,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 6.4268,
|
|
"mean_token_accuracy": 0.12757971361279488,
|
|
"num_tokens": 2866993.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.48082594871521,
|
|
"epoch": 0.03956177723676202,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 6.2442,
|
|
"mean_token_accuracy": 0.14385330006480218,
|
|
"num_tokens": 2874341.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.469588851928711,
|
|
"epoch": 0.03967046343796192,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 6.4488,
|
|
"mean_token_accuracy": 0.1317708507180214,
|
|
"num_tokens": 2882332.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.4172779560089115,
|
|
"epoch": 0.039779149639161814,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 6.3017,
|
|
"mean_token_accuracy": 0.14140139743685723,
|
|
"num_tokens": 2889901.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.5247608661651615,
|
|
"epoch": 0.03988783584036171,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.429,
|
|
"mean_token_accuracy": 0.12834445238113404,
|
|
"num_tokens": 2897732.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.410207319259643,
|
|
"epoch": 0.0399965220415616,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 6.3051,
|
|
"mean_token_accuracy": 0.13572658076882363,
|
|
"num_tokens": 2906036.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.344397115707397,
|
|
"epoch": 0.0401052082427615,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.336,
|
|
"mean_token_accuracy": 0.13653432950377464,
|
|
"num_tokens": 2913406.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.4600622177124025,
|
|
"epoch": 0.040213894443961395,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 6.3678,
|
|
"mean_token_accuracy": 0.13492351323366164,
|
|
"num_tokens": 2921468.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.381730365753174,
|
|
"epoch": 0.04032258064516129,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 6.2972,
|
|
"mean_token_accuracy": 0.1388672910630703,
|
|
"num_tokens": 2929220.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.502804088592529,
|
|
"epoch": 0.04043126684636118,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 6.4621,
|
|
"mean_token_accuracy": 0.13003402724862098,
|
|
"num_tokens": 2937452.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.5060210704803465,
|
|
"epoch": 0.040539953047561084,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.3967,
|
|
"mean_token_accuracy": 0.13554264083504677,
|
|
"num_tokens": 2946157.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.35983681678772,
|
|
"epoch": 0.04064863924876098,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 6.2866,
|
|
"mean_token_accuracy": 0.13986295610666274,
|
|
"num_tokens": 2953445.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.422707271575928,
|
|
"epoch": 0.04075732544996087,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 6.3382,
|
|
"mean_token_accuracy": 0.13599542155861855,
|
|
"num_tokens": 2961818.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.40128526687622,
|
|
"epoch": 0.04086601165116077,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.335,
|
|
"mean_token_accuracy": 0.13727616444230079,
|
|
"num_tokens": 2969811.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.453525066375732,
|
|
"epoch": 0.040974697852360666,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 6.3521,
|
|
"mean_token_accuracy": 0.13766252547502517,
|
|
"num_tokens": 2978291.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.41055645942688,
|
|
"epoch": 0.04108338405356056,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.317,
|
|
"mean_token_accuracy": 0.13951440081000327,
|
|
"num_tokens": 2985624.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.40109486579895,
|
|
"epoch": 0.04119207025476045,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.3304,
|
|
"mean_token_accuracy": 0.13785064667463304,
|
|
"num_tokens": 2993078.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.453061771392822,
|
|
"epoch": 0.041300756455960354,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 6.3396,
|
|
"mean_token_accuracy": 0.14081092923879623,
|
|
"num_tokens": 3001208.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.355777645111084,
|
|
"epoch": 0.04140944265716025,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 6.2945,
|
|
"mean_token_accuracy": 0.13765922859311103,
|
|
"num_tokens": 3008721.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.405518579483032,
|
|
"epoch": 0.04151812885836014,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.3345,
|
|
"mean_token_accuracy": 0.13959560170769691,
|
|
"num_tokens": 3016895.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.4162850856781,
|
|
"epoch": 0.041626815059560035,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 6.2956,
|
|
"mean_token_accuracy": 0.14023387357592582,
|
|
"num_tokens": 3024525.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.41334981918335,
|
|
"epoch": 0.041735501260759936,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.3748,
|
|
"mean_token_accuracy": 0.1350133463740349,
|
|
"num_tokens": 3032628.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.421497774124146,
|
|
"epoch": 0.04184418746195983,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.2696,
|
|
"mean_token_accuracy": 0.13667341843247413,
|
|
"num_tokens": 3041123.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.2926805973052975,
|
|
"epoch": 0.04195287366315972,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.2008,
|
|
"mean_token_accuracy": 0.14734140411019325,
|
|
"num_tokens": 3048367.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.407898807525635,
|
|
"epoch": 0.042061559864359624,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 6.3414,
|
|
"mean_token_accuracy": 0.13200511485338212,
|
|
"num_tokens": 3057257.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.4973756790161135,
|
|
"epoch": 0.04217024606555952,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.3839,
|
|
"mean_token_accuracy": 0.13520105555653572,
|
|
"num_tokens": 3065538.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.375053262710571,
|
|
"epoch": 0.04227893226675941,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 6.3181,
|
|
"mean_token_accuracy": 0.14364981800317764,
|
|
"num_tokens": 3073215.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.431594753265381,
|
|
"epoch": 0.042387618467959305,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 6.3236,
|
|
"mean_token_accuracy": 0.14194342046976088,
|
|
"num_tokens": 3080713.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.475106859207154,
|
|
"epoch": 0.042496304669159206,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 6.4159,
|
|
"mean_token_accuracy": 0.1319071814417839,
|
|
"num_tokens": 3089339.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.373333311080932,
|
|
"epoch": 0.0426049908703591,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.3513,
|
|
"mean_token_accuracy": 0.14044444710016252,
|
|
"num_tokens": 3097315.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.446169137954712,
|
|
"epoch": 0.04271367707155899,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 6.3213,
|
|
"mean_token_accuracy": 0.13605728298425673,
|
|
"num_tokens": 3104924.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.359133243560791,
|
|
"epoch": 0.04282236327275889,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 6.2606,
|
|
"mean_token_accuracy": 0.14185562431812287,
|
|
"num_tokens": 3113107.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.3946362972259525,
|
|
"epoch": 0.04293104947395879,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 6.3506,
|
|
"mean_token_accuracy": 0.13473108187317848,
|
|
"num_tokens": 3121410.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.389178657531739,
|
|
"epoch": 0.04303973567515868,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 6.3093,
|
|
"mean_token_accuracy": 0.1421495869755745,
|
|
"num_tokens": 3129020.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.429538202285767,
|
|
"epoch": 0.043148421876358575,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.2758,
|
|
"mean_token_accuracy": 0.13987535163760184,
|
|
"num_tokens": 3136934.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.408135843276978,
|
|
"epoch": 0.043257108077558476,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 6.3525,
|
|
"mean_token_accuracy": 0.13694941774010658,
|
|
"num_tokens": 3144742.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.473373699188232,
|
|
"epoch": 0.04336579427875837,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 6.319,
|
|
"mean_token_accuracy": 0.1362934321165085,
|
|
"num_tokens": 3151731.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.4309955596923825,
|
|
"epoch": 0.043474480479958263,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 6.3891,
|
|
"mean_token_accuracy": 0.13514773100614547,
|
|
"num_tokens": 3159781.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.431392335891724,
|
|
"epoch": 0.04358316668115816,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 6.3403,
|
|
"mean_token_accuracy": 0.13589958250522613,
|
|
"num_tokens": 3167896.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.383094835281372,
|
|
"epoch": 0.04369185288235806,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 6.3414,
|
|
"mean_token_accuracy": 0.1376558393239975,
|
|
"num_tokens": 3176573.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.377293491363526,
|
|
"epoch": 0.04380053908355795,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 6.1985,
|
|
"mean_token_accuracy": 0.1522999309003353,
|
|
"num_tokens": 3184935.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.331845140457153,
|
|
"epoch": 0.043909225284757845,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 6.264,
|
|
"mean_token_accuracy": 0.1453753650188446,
|
|
"num_tokens": 3192220.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.460266923904419,
|
|
"epoch": 0.044017911485957746,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 6.2932,
|
|
"mean_token_accuracy": 0.14447869211435319,
|
|
"num_tokens": 3200330.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.3918225288391115,
|
|
"epoch": 0.04412659768715764,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 6.328,
|
|
"mean_token_accuracy": 0.1438807040452957,
|
|
"num_tokens": 3208076.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.434448146820069,
|
|
"epoch": 0.044235283888357534,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 6.2808,
|
|
"mean_token_accuracy": 0.1417445071041584,
|
|
"num_tokens": 3216326.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.358164215087891,
|
|
"epoch": 0.04434397008955743,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 6.3514,
|
|
"mean_token_accuracy": 0.13139131888747216,
|
|
"num_tokens": 3224322.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.407972049713135,
|
|
"epoch": 0.04445265629075733,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 6.2797,
|
|
"mean_token_accuracy": 0.1365787222981453,
|
|
"num_tokens": 3231766.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.426673221588135,
|
|
"epoch": 0.04456134249195722,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 6.3085,
|
|
"mean_token_accuracy": 0.136894803494215,
|
|
"num_tokens": 3239936.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.341384601593018,
|
|
"epoch": 0.044670028693157116,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 6.2338,
|
|
"mean_token_accuracy": 0.13770833760499954,
|
|
"num_tokens": 3248185.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.430909919738769,
|
|
"epoch": 0.04477871489435701,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 6.2252,
|
|
"mean_token_accuracy": 0.14265335500240325,
|
|
"num_tokens": 3256051.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.481796169281006,
|
|
"epoch": 0.04488740109555691,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 6.3142,
|
|
"mean_token_accuracy": 0.13447861969470978,
|
|
"num_tokens": 3264128.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.342309761047363,
|
|
"epoch": 0.044996087296756804,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 6.383,
|
|
"mean_token_accuracy": 0.13557247668504716,
|
|
"num_tokens": 3272633.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.4529866695404055,
|
|
"epoch": 0.0451047734979567,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 6.386,
|
|
"mean_token_accuracy": 0.142952311784029,
|
|
"num_tokens": 3281139.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.303681421279907,
|
|
"epoch": 0.0452134596991566,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 6.1852,
|
|
"mean_token_accuracy": 0.1471434824168682,
|
|
"num_tokens": 3289384.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.332111644744873,
|
|
"epoch": 0.04532214590035649,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 6.272,
|
|
"mean_token_accuracy": 0.14111893102526665,
|
|
"num_tokens": 3297103.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.413033151626587,
|
|
"epoch": 0.045430832101556386,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 6.2727,
|
|
"mean_token_accuracy": 0.14211916625499726,
|
|
"num_tokens": 3304497.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.440083169937134,
|
|
"epoch": 0.04553951830275628,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 6.3613,
|
|
"mean_token_accuracy": 0.13257155790925026,
|
|
"num_tokens": 3311691.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.319086742401123,
|
|
"epoch": 0.04564820450395618,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 6.2465,
|
|
"mean_token_accuracy": 0.14056456089019775,
|
|
"num_tokens": 3320547.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.4411708354949955,
|
|
"epoch": 0.045756890705156074,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 6.2822,
|
|
"mean_token_accuracy": 0.13730876967310907,
|
|
"num_tokens": 3328629.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.42737603187561,
|
|
"epoch": 0.04586557690635597,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 6.3231,
|
|
"mean_token_accuracy": 0.14296908006072045,
|
|
"num_tokens": 3336050.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.277024555206299,
|
|
"epoch": 0.04597426310755586,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 6.1642,
|
|
"mean_token_accuracy": 0.14555428475141524,
|
|
"num_tokens": 3343736.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.362267398834229,
|
|
"epoch": 0.04608294930875576,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 6.2673,
|
|
"mean_token_accuracy": 0.13701179027557372,
|
|
"num_tokens": 3350841.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.399694347381592,
|
|
"epoch": 0.046191635509955656,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 6.4041,
|
|
"mean_token_accuracy": 0.1311640538275242,
|
|
"num_tokens": 3359792.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.386282014846802,
|
|
"epoch": 0.04630032171115555,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 6.2087,
|
|
"mean_token_accuracy": 0.14746672958135604,
|
|
"num_tokens": 3366959.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.345727109909058,
|
|
"epoch": 0.04640900791235545,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 6.3342,
|
|
"mean_token_accuracy": 0.13067435547709466,
|
|
"num_tokens": 3375243.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.332585048675537,
|
|
"epoch": 0.046517694113555344,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 6.1901,
|
|
"mean_token_accuracy": 0.14754107296466829,
|
|
"num_tokens": 3382735.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.329954290390015,
|
|
"epoch": 0.04662638031475524,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 6.2346,
|
|
"mean_token_accuracy": 0.14045259580016137,
|
|
"num_tokens": 3390041.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.354192924499512,
|
|
"epoch": 0.04673506651595513,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 6.2207,
|
|
"mean_token_accuracy": 0.14744865596294404,
|
|
"num_tokens": 3397385.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.311786699295044,
|
|
"epoch": 0.04684375271715503,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 6.2298,
|
|
"mean_token_accuracy": 0.1454736463725567,
|
|
"num_tokens": 3404499.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.40657868385315,
|
|
"epoch": 0.046952438918354926,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 6.2401,
|
|
"mean_token_accuracy": 0.14431775584816933,
|
|
"num_tokens": 3412650.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.263350868225098,
|
|
"epoch": 0.04706112511955482,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 6.2667,
|
|
"mean_token_accuracy": 0.1401475727558136,
|
|
"num_tokens": 3420282.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.392113447189331,
|
|
"epoch": 0.04716981132075472,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 6.2028,
|
|
"mean_token_accuracy": 0.14057230800390244,
|
|
"num_tokens": 3427706.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.307991933822632,
|
|
"epoch": 0.047278497521954614,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 6.3039,
|
|
"mean_token_accuracy": 0.1375028222799301,
|
|
"num_tokens": 3436088.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.364810371398926,
|
|
"epoch": 0.04738718372315451,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 6.1777,
|
|
"mean_token_accuracy": 0.15691623613238334,
|
|
"num_tokens": 3444200.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.327755641937256,
|
|
"epoch": 0.0474958699243544,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 6.2187,
|
|
"mean_token_accuracy": 0.143999794870615,
|
|
"num_tokens": 3452684.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.2792760848999025,
|
|
"epoch": 0.0476045561255543,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 6.1575,
|
|
"mean_token_accuracy": 0.14571396857500077,
|
|
"num_tokens": 3460347.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.320020341873169,
|
|
"epoch": 0.047713242326754196,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 6.2914,
|
|
"mean_token_accuracy": 0.13896366208791733,
|
|
"num_tokens": 3467956.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.35538878440857,
|
|
"epoch": 0.04782192852795409,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 6.175,
|
|
"mean_token_accuracy": 0.1493356004357338,
|
|
"num_tokens": 3474803.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.3529205322265625,
|
|
"epoch": 0.047930614729153984,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 6.2878,
|
|
"mean_token_accuracy": 0.13683823347091675,
|
|
"num_tokens": 3482390.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.39755654335022,
|
|
"epoch": 0.048039300930353884,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 6.2117,
|
|
"mean_token_accuracy": 0.14897590279579162,
|
|
"num_tokens": 3489742.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.369425535202026,
|
|
"epoch": 0.04814798713155378,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 6.2722,
|
|
"mean_token_accuracy": 0.14080924764275551,
|
|
"num_tokens": 3497257.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.442922210693359,
|
|
"epoch": 0.04825667333275367,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 6.3458,
|
|
"mean_token_accuracy": 0.130158282071352,
|
|
"num_tokens": 3506258.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.398208045959473,
|
|
"epoch": 0.04836535953395357,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 6.2848,
|
|
"mean_token_accuracy": 0.14103573709726333,
|
|
"num_tokens": 3514109.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.347044658660889,
|
|
"epoch": 0.048474045735153466,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 6.1916,
|
|
"mean_token_accuracy": 0.14575656726956368,
|
|
"num_tokens": 3521654.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.344361257553101,
|
|
"epoch": 0.04858273193635336,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 6.1765,
|
|
"mean_token_accuracy": 0.14519614577293397,
|
|
"num_tokens": 3528943.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.325352907180786,
|
|
"epoch": 0.048691418137553254,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 6.1874,
|
|
"mean_token_accuracy": 0.1502646818757057,
|
|
"num_tokens": 3536182.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.360516691207886,
|
|
"epoch": 0.048800104338753154,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 6.1989,
|
|
"mean_token_accuracy": 0.1458291694521904,
|
|
"num_tokens": 3543446.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.3320952415466305,
|
|
"epoch": 0.04890879053995305,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 6.3055,
|
|
"mean_token_accuracy": 0.1307961642742157,
|
|
"num_tokens": 3550418.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.3693479061126705,
|
|
"epoch": 0.04901747674115294,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 6.2303,
|
|
"mean_token_accuracy": 0.14581648483872414,
|
|
"num_tokens": 3558403.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.360597085952759,
|
|
"epoch": 0.049126162942352836,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 6.2327,
|
|
"mean_token_accuracy": 0.14444623440504073,
|
|
"num_tokens": 3566365.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.362605428695678,
|
|
"epoch": 0.049234849143552736,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 6.254,
|
|
"mean_token_accuracy": 0.14103433042764663,
|
|
"num_tokens": 3574648.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.3952125072479244,
|
|
"epoch": 0.04934353534475263,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 6.2518,
|
|
"mean_token_accuracy": 0.1349252924323082,
|
|
"num_tokens": 3581993.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.285873746871948,
|
|
"epoch": 0.049452221545952524,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 6.0713,
|
|
"mean_token_accuracy": 0.14827482253313065,
|
|
"num_tokens": 3589229.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.245410346984864,
|
|
"epoch": 0.049560907747152425,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 6.1994,
|
|
"mean_token_accuracy": 0.14884500652551652,
|
|
"num_tokens": 3597560.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.373068618774414,
|
|
"epoch": 0.04966959394835232,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 6.1644,
|
|
"mean_token_accuracy": 0.14538974687457085,
|
|
"num_tokens": 3605558.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.245131349563598,
|
|
"epoch": 0.04977828014955221,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 6.1659,
|
|
"mean_token_accuracy": 0.15103522837162017,
|
|
"num_tokens": 3613080.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.331179475784301,
|
|
"epoch": 0.049886966350752106,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 6.3188,
|
|
"mean_token_accuracy": 0.14189708456397057,
|
|
"num_tokens": 3620907.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.475124979019165,
|
|
"epoch": 0.049995652551952006,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 6.2626,
|
|
"mean_token_accuracy": 0.13620245233178138,
|
|
"num_tokens": 3630046.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.352400922775269,
|
|
"epoch": 0.0501043387531519,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 6.1488,
|
|
"mean_token_accuracy": 0.145827666670084,
|
|
"num_tokens": 3637492.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.361015510559082,
|
|
"epoch": 0.050213024954351794,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 6.2743,
|
|
"mean_token_accuracy": 0.14036672934889793,
|
|
"num_tokens": 3645770.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.382439994812012,
|
|
"epoch": 0.05032171115555169,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 6.2207,
|
|
"mean_token_accuracy": 0.13502092361450196,
|
|
"num_tokens": 3654171.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.313130712509155,
|
|
"epoch": 0.05043039735675159,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 6.1364,
|
|
"mean_token_accuracy": 0.15124589800834656,
|
|
"num_tokens": 3661158.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.174329662322998,
|
|
"epoch": 0.05053908355795148,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 6.1544,
|
|
"mean_token_accuracy": 0.14774754047393798,
|
|
"num_tokens": 3669247.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.351832914352417,
|
|
"epoch": 0.050647769759151376,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 6.2254,
|
|
"mean_token_accuracy": 0.14423187524080278,
|
|
"num_tokens": 3676917.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.440951728820801,
|
|
"epoch": 0.05075645596035128,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 6.2574,
|
|
"mean_token_accuracy": 0.14545449316501619,
|
|
"num_tokens": 3684664.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.275143241882324,
|
|
"epoch": 0.05086514216155117,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 6.1038,
|
|
"mean_token_accuracy": 0.14968376085162163,
|
|
"num_tokens": 3691938.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.226733779907226,
|
|
"epoch": 0.050973828362751064,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 6.2031,
|
|
"mean_token_accuracy": 0.13864690586924552,
|
|
"num_tokens": 3699489.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.331015968322754,
|
|
"epoch": 0.05108251456395096,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 6.1655,
|
|
"mean_token_accuracy": 0.15013981610536575,
|
|
"num_tokens": 3707804.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.322420215606689,
|
|
"epoch": 0.05119120076515086,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 6.2085,
|
|
"mean_token_accuracy": 0.13792425468564035,
|
|
"num_tokens": 3715175.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.352394914627075,
|
|
"epoch": 0.05129988696635075,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 6.1127,
|
|
"mean_token_accuracy": 0.14395364746451378,
|
|
"num_tokens": 3722468.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.23946738243103,
|
|
"epoch": 0.051408573167550646,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 6.0647,
|
|
"mean_token_accuracy": 0.16229498982429505,
|
|
"num_tokens": 3729029.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.157553577423096,
|
|
"epoch": 0.05151725936875055,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 6.1303,
|
|
"mean_token_accuracy": 0.14998419880867003,
|
|
"num_tokens": 3736292.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.257620859146118,
|
|
"epoch": 0.05162594556995044,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 6.1493,
|
|
"mean_token_accuracy": 0.1474681794643402,
|
|
"num_tokens": 3744264.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.428153324127197,
|
|
"epoch": 0.051734631771150334,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 6.189,
|
|
"mean_token_accuracy": 0.14731017872691154,
|
|
"num_tokens": 3751732.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.319741821289062,
|
|
"epoch": 0.05184331797235023,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 6.1513,
|
|
"mean_token_accuracy": 0.1528203696012497,
|
|
"num_tokens": 3759273.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.320394086837768,
|
|
"epoch": 0.05195200417355013,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 6.1181,
|
|
"mean_token_accuracy": 0.14740897715091705,
|
|
"num_tokens": 3766929.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.296305561065674,
|
|
"epoch": 0.05206069037475002,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 6.1932,
|
|
"mean_token_accuracy": 0.14625833109021186,
|
|
"num_tokens": 3774453.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.293846559524536,
|
|
"epoch": 0.052169376575949916,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 6.2742,
|
|
"mean_token_accuracy": 0.14566972702741623,
|
|
"num_tokens": 3783329.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.346486234664917,
|
|
"epoch": 0.05227806277714981,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 6.1299,
|
|
"mean_token_accuracy": 0.1495580941438675,
|
|
"num_tokens": 3791201.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.227714109420776,
|
|
"epoch": 0.05238674897834971,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 6.121,
|
|
"mean_token_accuracy": 0.14442274942994118,
|
|
"num_tokens": 3799088.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.392905378341675,
|
|
"epoch": 0.052495435179549604,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 6.2871,
|
|
"mean_token_accuracy": 0.13784987702965737,
|
|
"num_tokens": 3807156.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.408941173553467,
|
|
"epoch": 0.0526041213807495,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 6.1201,
|
|
"mean_token_accuracy": 0.15032362788915635,
|
|
"num_tokens": 3814121.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.283007287979126,
|
|
"epoch": 0.0527128075819494,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 6.1984,
|
|
"mean_token_accuracy": 0.1452282190322876,
|
|
"num_tokens": 3821502.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.33258056640625,
|
|
"epoch": 0.05282149378314929,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 6.2369,
|
|
"mean_token_accuracy": 0.14194708019495011,
|
|
"num_tokens": 3830907.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.310719108581543,
|
|
"epoch": 0.052930179984349186,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 6.2992,
|
|
"mean_token_accuracy": 0.13263836950063707,
|
|
"num_tokens": 3838643.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.362523603439331,
|
|
"epoch": 0.05303886618554908,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 6.2052,
|
|
"mean_token_accuracy": 0.14495603367686272,
|
|
"num_tokens": 3846950.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.364418840408325,
|
|
"epoch": 0.05314755238674898,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 6.234,
|
|
"mean_token_accuracy": 0.15189555883407593,
|
|
"num_tokens": 3855373.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.281556272506714,
|
|
"epoch": 0.053256238587948874,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 6.1334,
|
|
"mean_token_accuracy": 0.1457263521850109,
|
|
"num_tokens": 3862973.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.234706211090088,
|
|
"epoch": 0.05336492478914877,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 6.0761,
|
|
"mean_token_accuracy": 0.15153853297233583,
|
|
"num_tokens": 3870713.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.313697624206543,
|
|
"epoch": 0.05347361099034866,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 6.1607,
|
|
"mean_token_accuracy": 0.1522718347609043,
|
|
"num_tokens": 3878696.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.309638118743896,
|
|
"epoch": 0.05358229719154856,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 6.1553,
|
|
"mean_token_accuracy": 0.1505454257130623,
|
|
"num_tokens": 3886072.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.325492429733276,
|
|
"epoch": 0.053690983392748456,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 6.1877,
|
|
"mean_token_accuracy": 0.1469157561659813,
|
|
"num_tokens": 3893970.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.351982879638672,
|
|
"epoch": 0.05379966959394835,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 6.1673,
|
|
"mean_token_accuracy": 0.1419338807463646,
|
|
"num_tokens": 3902017.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.343962669372559,
|
|
"epoch": 0.05390835579514825,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 6.2655,
|
|
"mean_token_accuracy": 0.1384866639971733,
|
|
"num_tokens": 3910665.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.423051500320435,
|
|
"epoch": 0.054017041996348145,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 6.2664,
|
|
"mean_token_accuracy": 0.14044074714183807,
|
|
"num_tokens": 3918593.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.360489845275879,
|
|
"epoch": 0.05412572819754804,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 6.0691,
|
|
"mean_token_accuracy": 0.14841551929712296,
|
|
"num_tokens": 3925687.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.26637830734253,
|
|
"epoch": 0.05423441439874793,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 6.2016,
|
|
"mean_token_accuracy": 0.15307127833366393,
|
|
"num_tokens": 3932641.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.261042737960816,
|
|
"epoch": 0.05434310059994783,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 6.2238,
|
|
"mean_token_accuracy": 0.1425244577229023,
|
|
"num_tokens": 3941158.0,
|
|
"step": 2500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"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": 6447152259072000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|