2035 lines
54 KiB
JSON
2035 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.021737240239979132,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 2577939849216000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|