af430a657a
Model: fpadovani/ita-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
10145 lines
276 KiB
JSON
10145 lines
276 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 5.040342914775593,
|
|
"eval_steps": 500,
|
|
"global_step": 5000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.798974180221558,
|
|
"epoch": 0.005042864346949067,
|
|
"grad_norm": 13.8125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.3629,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 11445.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.794778347015381,
|
|
"epoch": 0.010085728693898134,
|
|
"grad_norm": 14.6875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.26,
|
|
"mean_token_accuracy": 6.73400645609945e-05,
|
|
"num_tokens": 22095.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.854873609542847,
|
|
"epoch": 0.015128593040847202,
|
|
"grad_norm": 19.25,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.0607,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 33847.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.200687265396118,
|
|
"epoch": 0.020171457387796268,
|
|
"grad_norm": 30.25,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.3296,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 44495.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.847039031982422,
|
|
"epoch": 0.025214321734745335,
|
|
"grad_norm": 9.6875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.6749,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 55298.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.538958072662354,
|
|
"epoch": 0.030257186081694403,
|
|
"grad_norm": 3.78125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.8508,
|
|
"mean_token_accuracy": 0.0004506777215283364,
|
|
"num_tokens": 66391.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.692523670196532,
|
|
"epoch": 0.03530005042864347,
|
|
"grad_norm": 3.3125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.6137,
|
|
"mean_token_accuracy": 0.009274485579226166,
|
|
"num_tokens": 77624.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.688490772247315,
|
|
"epoch": 0.040342914775592535,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2443,
|
|
"mean_token_accuracy": 0.029242578893899918,
|
|
"num_tokens": 89426.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.590974426269531,
|
|
"epoch": 0.0453857791225416,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.9628,
|
|
"mean_token_accuracy": 0.03920505605638027,
|
|
"num_tokens": 100252.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.583111190795899,
|
|
"epoch": 0.05042864346949067,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.7767,
|
|
"mean_token_accuracy": 0.035502530820667746,
|
|
"num_tokens": 110578.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.582507705688476,
|
|
"epoch": 0.05547150781643974,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.6849,
|
|
"mean_token_accuracy": 0.03719575461000204,
|
|
"num_tokens": 121631.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.566785621643067,
|
|
"epoch": 0.060514372163388806,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5638,
|
|
"mean_token_accuracy": 0.04203882701694965,
|
|
"num_tokens": 134254.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.553144550323486,
|
|
"epoch": 0.06555723651033787,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.5596,
|
|
"mean_token_accuracy": 0.03803328052163124,
|
|
"num_tokens": 144928.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.5373703956604,
|
|
"epoch": 0.07060010085728693,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.409,
|
|
"mean_token_accuracy": 0.039696192182600495,
|
|
"num_tokens": 155518.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.525967884063721,
|
|
"epoch": 0.07564296520423601,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.3081,
|
|
"mean_token_accuracy": 0.04226764794439077,
|
|
"num_tokens": 165850.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.500716590881348,
|
|
"epoch": 0.08068582955118507,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.2903,
|
|
"mean_token_accuracy": 0.04092370942234993,
|
|
"num_tokens": 176787.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.483646583557128,
|
|
"epoch": 0.08572869389813415,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1961,
|
|
"mean_token_accuracy": 0.03919961098581552,
|
|
"num_tokens": 188345.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.474784088134765,
|
|
"epoch": 0.0907715582450832,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.1426,
|
|
"mean_token_accuracy": 0.032541034184396266,
|
|
"num_tokens": 199315.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.458439445495605,
|
|
"epoch": 0.09581442259203228,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.9644,
|
|
"mean_token_accuracy": 0.04057908412069082,
|
|
"num_tokens": 209627.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.382002353668213,
|
|
"epoch": 0.10085728693898134,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8705,
|
|
"mean_token_accuracy": 0.043775437772274016,
|
|
"num_tokens": 220872.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.296143341064454,
|
|
"epoch": 0.1059001512859304,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.7393,
|
|
"mean_token_accuracy": 0.04108128827065229,
|
|
"num_tokens": 233272.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.234007835388184,
|
|
"epoch": 0.11094301563287948,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.6429,
|
|
"mean_token_accuracy": 0.043316423520445825,
|
|
"num_tokens": 243529.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.142849445343018,
|
|
"epoch": 0.11598587997982854,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.529,
|
|
"mean_token_accuracy": 0.04300804398953915,
|
|
"num_tokens": 254942.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.024840545654296,
|
|
"epoch": 0.12102874432677761,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.4186,
|
|
"mean_token_accuracy": 0.04885575398802757,
|
|
"num_tokens": 266552.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.812107372283936,
|
|
"epoch": 0.1260716086737267,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.2492,
|
|
"mean_token_accuracy": 0.04936607554554939,
|
|
"num_tokens": 277906.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.587112522125244,
|
|
"epoch": 0.13111447302067575,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.1241,
|
|
"mean_token_accuracy": 0.05138890668749809,
|
|
"num_tokens": 289079.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.357858276367187,
|
|
"epoch": 0.1361573373676248,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.0736,
|
|
"mean_token_accuracy": 0.04534509554505348,
|
|
"num_tokens": 300107.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.149685192108155,
|
|
"epoch": 0.14120020171457387,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.9221,
|
|
"mean_token_accuracy": 0.04426998291164637,
|
|
"num_tokens": 310779.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.934042835235596,
|
|
"epoch": 0.14624306606152296,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.9263,
|
|
"mean_token_accuracy": 0.04932713992893696,
|
|
"num_tokens": 323478.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.715434741973876,
|
|
"epoch": 0.15128593040847202,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.8083,
|
|
"mean_token_accuracy": 0.05145653560757637,
|
|
"num_tokens": 335255.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.460878372192383,
|
|
"epoch": 0.15632879475542108,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.7481,
|
|
"mean_token_accuracy": 0.050404298678040506,
|
|
"num_tokens": 344824.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.369724655151368,
|
|
"epoch": 0.16137165910237014,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.7789,
|
|
"mean_token_accuracy": 0.046216747164726256,
|
|
"num_tokens": 356283.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.261080932617187,
|
|
"epoch": 0.1664145234493192,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7572,
|
|
"mean_token_accuracy": 0.048884724453091624,
|
|
"num_tokens": 368098.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.193770790100098,
|
|
"epoch": 0.1714573877962683,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7121,
|
|
"mean_token_accuracy": 0.04776199422776699,
|
|
"num_tokens": 380948.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.10466547012329,
|
|
"epoch": 0.17650025214321735,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.655,
|
|
"mean_token_accuracy": 0.05172229073941707,
|
|
"num_tokens": 390392.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.129018688201905,
|
|
"epoch": 0.1815431164901664,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.6849,
|
|
"mean_token_accuracy": 0.05508568845689297,
|
|
"num_tokens": 400328.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.094326400756836,
|
|
"epoch": 0.18658598083711547,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.6912,
|
|
"mean_token_accuracy": 0.05726216584444046,
|
|
"num_tokens": 410909.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.090226984024047,
|
|
"epoch": 0.19162884518406456,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.6585,
|
|
"mean_token_accuracy": 0.057058770582079886,
|
|
"num_tokens": 422726.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.073493576049804,
|
|
"epoch": 0.19667170953101362,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.5669,
|
|
"mean_token_accuracy": 0.06086595430970192,
|
|
"num_tokens": 433376.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.003545379638672,
|
|
"epoch": 0.20171457387796268,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.61,
|
|
"mean_token_accuracy": 0.05815195962786675,
|
|
"num_tokens": 443940.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.07164282798767,
|
|
"epoch": 0.20675743822491174,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5019,
|
|
"mean_token_accuracy": 0.060009703412652014,
|
|
"num_tokens": 455016.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.987448406219483,
|
|
"epoch": 0.2118003025718608,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.5273,
|
|
"mean_token_accuracy": 0.06523391492664814,
|
|
"num_tokens": 466380.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.967493391036987,
|
|
"epoch": 0.2168431669188099,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3896,
|
|
"mean_token_accuracy": 0.0666438702493906,
|
|
"num_tokens": 476915.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.933144903182983,
|
|
"epoch": 0.22188603126575895,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.4639,
|
|
"mean_token_accuracy": 0.07083339318633079,
|
|
"num_tokens": 488924.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.8961576461792,
|
|
"epoch": 0.22692889561270801,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.4065,
|
|
"mean_token_accuracy": 0.06823427751660346,
|
|
"num_tokens": 498965.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.938161325454712,
|
|
"epoch": 0.23197175995965708,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.4277,
|
|
"mean_token_accuracy": 0.0643129862844944,
|
|
"num_tokens": 509755.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.8996302604675295,
|
|
"epoch": 0.23701462430660616,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.3917,
|
|
"mean_token_accuracy": 0.06402169205248356,
|
|
"num_tokens": 521221.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.879079818725586,
|
|
"epoch": 0.24205748865355523,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.4935,
|
|
"mean_token_accuracy": 0.06531766764819621,
|
|
"num_tokens": 533595.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.888955450057983,
|
|
"epoch": 0.24710035300050429,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.4383,
|
|
"mean_token_accuracy": 0.06679745763540268,
|
|
"num_tokens": 543672.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.8185327529907225,
|
|
"epoch": 0.2521432173474534,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.3155,
|
|
"mean_token_accuracy": 0.0741234589368105,
|
|
"num_tokens": 554906.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.80477933883667,
|
|
"epoch": 0.25718608169440244,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.3978,
|
|
"mean_token_accuracy": 0.07233219593763351,
|
|
"num_tokens": 566419.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.773422384262085,
|
|
"epoch": 0.2622289460413515,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3585,
|
|
"mean_token_accuracy": 0.06727737858891487,
|
|
"num_tokens": 578848.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.732848262786865,
|
|
"epoch": 0.26727181038830056,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.3033,
|
|
"mean_token_accuracy": 0.06985007859766483,
|
|
"num_tokens": 590601.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.773223400115967,
|
|
"epoch": 0.2723146747352496,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3549,
|
|
"mean_token_accuracy": 0.0672258909791708,
|
|
"num_tokens": 602021.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.746130037307739,
|
|
"epoch": 0.2773575390821987,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.436,
|
|
"mean_token_accuracy": 0.06151515953242779,
|
|
"num_tokens": 613966.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.81422176361084,
|
|
"epoch": 0.28240040342914774,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2641,
|
|
"mean_token_accuracy": 0.07085276804864407,
|
|
"num_tokens": 624851.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.6415393352508545,
|
|
"epoch": 0.2874432677760968,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.3566,
|
|
"mean_token_accuracy": 0.06834671199321747,
|
|
"num_tokens": 635580.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.7174467086792,
|
|
"epoch": 0.2924861321230459,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.4288,
|
|
"mean_token_accuracy": 0.06711922027170658,
|
|
"num_tokens": 646927.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.766084957122803,
|
|
"epoch": 0.297528996469995,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.2366,
|
|
"mean_token_accuracy": 0.07219495773315429,
|
|
"num_tokens": 658116.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.702093029022217,
|
|
"epoch": 0.30257186081694404,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2469,
|
|
"mean_token_accuracy": 0.06866099908947945,
|
|
"num_tokens": 669054.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.672720432281494,
|
|
"epoch": 0.3076147251638931,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2934,
|
|
"mean_token_accuracy": 0.07351624183356761,
|
|
"num_tokens": 680439.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.5994908809661865,
|
|
"epoch": 0.31265758951084216,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2394,
|
|
"mean_token_accuracy": 0.07195786684751511,
|
|
"num_tokens": 690909.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.61633038520813,
|
|
"epoch": 0.3177004538577912,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.2214,
|
|
"mean_token_accuracy": 0.07065910324454308,
|
|
"num_tokens": 701744.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.582859230041504,
|
|
"epoch": 0.3227433182047403,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.2561,
|
|
"mean_token_accuracy": 0.07319894470274449,
|
|
"num_tokens": 712268.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.633519649505615,
|
|
"epoch": 0.32778618255168934,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.2916,
|
|
"mean_token_accuracy": 0.06789168566465378,
|
|
"num_tokens": 722978.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.636728763580322,
|
|
"epoch": 0.3328290468986384,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.2761,
|
|
"mean_token_accuracy": 0.06828575171530246,
|
|
"num_tokens": 734278.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.655182218551635,
|
|
"epoch": 0.3378719112455875,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1973,
|
|
"mean_token_accuracy": 0.06927417740225791,
|
|
"num_tokens": 746109.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.508404159545899,
|
|
"epoch": 0.3429147755925366,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.1252,
|
|
"mean_token_accuracy": 0.07408354654908181,
|
|
"num_tokens": 757519.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.551991367340088,
|
|
"epoch": 0.34795763993948564,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1015,
|
|
"mean_token_accuracy": 0.07286596298217773,
|
|
"num_tokens": 767665.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.528620338439941,
|
|
"epoch": 0.3530005042864347,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1368,
|
|
"mean_token_accuracy": 0.07722761034965515,
|
|
"num_tokens": 778880.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.41562442779541,
|
|
"epoch": 0.35804336863338376,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.1417,
|
|
"mean_token_accuracy": 0.08068987801671028,
|
|
"num_tokens": 790360.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.50475926399231,
|
|
"epoch": 0.3630862329803328,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.1075,
|
|
"mean_token_accuracy": 0.07803598418831825,
|
|
"num_tokens": 800889.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.5319267272949215,
|
|
"epoch": 0.3681290973272819,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.1137,
|
|
"mean_token_accuracy": 0.0743459228426218,
|
|
"num_tokens": 811414.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.38033766746521,
|
|
"epoch": 0.37317196167423095,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.1045,
|
|
"mean_token_accuracy": 0.07797230668365955,
|
|
"num_tokens": 822822.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.461845874786377,
|
|
"epoch": 0.37821482602118,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0241,
|
|
"mean_token_accuracy": 0.08001302257180214,
|
|
"num_tokens": 833145.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.3793505191802975,
|
|
"epoch": 0.3832576903681291,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1239,
|
|
"mean_token_accuracy": 0.07612378522753716,
|
|
"num_tokens": 844015.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.428417205810547,
|
|
"epoch": 0.3883005547150782,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.0585,
|
|
"mean_token_accuracy": 0.0799956239759922,
|
|
"num_tokens": 855806.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.3365089893341064,
|
|
"epoch": 0.39334341906202724,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.9714,
|
|
"mean_token_accuracy": 0.08045700266957283,
|
|
"num_tokens": 865750.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.428045749664307,
|
|
"epoch": 0.3983862834089763,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.1142,
|
|
"mean_token_accuracy": 0.07444334477186203,
|
|
"num_tokens": 877630.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.323162603378296,
|
|
"epoch": 0.40342914775592537,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.0023,
|
|
"mean_token_accuracy": 0.08490228429436683,
|
|
"num_tokens": 888667.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.440687036514282,
|
|
"epoch": 0.4084720121028744,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0748,
|
|
"mean_token_accuracy": 0.07444942444562912,
|
|
"num_tokens": 900989.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.371865701675415,
|
|
"epoch": 0.4135148764498235,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.9561,
|
|
"mean_token_accuracy": 0.07560547441244125,
|
|
"num_tokens": 911479.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.36584963798523,
|
|
"epoch": 0.41855774079677255,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.9798,
|
|
"mean_token_accuracy": 0.07795524820685387,
|
|
"num_tokens": 922434.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.303012132644653,
|
|
"epoch": 0.4236006051437216,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9825,
|
|
"mean_token_accuracy": 0.08246111869812012,
|
|
"num_tokens": 933405.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.286214828491211,
|
|
"epoch": 0.4286434694906707,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.983,
|
|
"mean_token_accuracy": 0.07633188813924789,
|
|
"num_tokens": 943567.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.275108814239502,
|
|
"epoch": 0.4336863338376198,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0033,
|
|
"mean_token_accuracy": 0.07348216734826565,
|
|
"num_tokens": 955123.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.364081239700317,
|
|
"epoch": 0.43872919818456885,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.0296,
|
|
"mean_token_accuracy": 0.08127171024680138,
|
|
"num_tokens": 965565.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.299008512496949,
|
|
"epoch": 0.4437720625315179,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.9695,
|
|
"mean_token_accuracy": 0.07634622417390347,
|
|
"num_tokens": 977101.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.371659612655639,
|
|
"epoch": 0.44881492687846697,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0095,
|
|
"mean_token_accuracy": 0.08116184324026107,
|
|
"num_tokens": 987445.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.221897411346435,
|
|
"epoch": 0.45385779122541603,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.8981,
|
|
"mean_token_accuracy": 0.08461304530501365,
|
|
"num_tokens": 999379.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.212428951263428,
|
|
"epoch": 0.4589006555723651,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.8401,
|
|
"mean_token_accuracy": 0.08424894288182258,
|
|
"num_tokens": 1010744.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.2306849479675295,
|
|
"epoch": 0.46394351991931415,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9091,
|
|
"mean_token_accuracy": 0.07732984013855457,
|
|
"num_tokens": 1021808.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.284563875198364,
|
|
"epoch": 0.4689863842662632,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.9519,
|
|
"mean_token_accuracy": 0.08044612407684326,
|
|
"num_tokens": 1033694.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.271745586395264,
|
|
"epoch": 0.47402924861321233,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.9258,
|
|
"mean_token_accuracy": 0.08233876228332519,
|
|
"num_tokens": 1044690.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.282025527954102,
|
|
"epoch": 0.4790721129601614,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.9288,
|
|
"mean_token_accuracy": 0.07964781187474727,
|
|
"num_tokens": 1055557.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.1536393642425535,
|
|
"epoch": 0.48411497730711045,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8648,
|
|
"mean_token_accuracy": 0.08270099312067032,
|
|
"num_tokens": 1067304.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.216547870635987,
|
|
"epoch": 0.4891578416540595,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8894,
|
|
"mean_token_accuracy": 0.08301360458135605,
|
|
"num_tokens": 1078955.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.128256750106812,
|
|
"epoch": 0.49420070600100857,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.924,
|
|
"mean_token_accuracy": 0.0845364436507225,
|
|
"num_tokens": 1091405.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.222031497955323,
|
|
"epoch": 0.49924357034795763,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8924,
|
|
"mean_token_accuracy": 0.0861407183110714,
|
|
"num_tokens": 1103227.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.179231071472168,
|
|
"epoch": 0.5042864346949067,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8366,
|
|
"mean_token_accuracy": 0.08988775014877319,
|
|
"num_tokens": 1114485.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5042864346949067,
|
|
"eval_entropy": 7.116292174535853,
|
|
"eval_loss": 6.949538707733154,
|
|
"eval_mean_token_accuracy": 0.0879388821640694,
|
|
"eval_num_tokens": 1114485.0,
|
|
"eval_runtime": 2.7822,
|
|
"eval_samples_per_second": 1127.884,
|
|
"eval_steps_per_second": 141.255,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.127004718780517,
|
|
"epoch": 0.5093292990418558,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.863,
|
|
"mean_token_accuracy": 0.08901753202080727,
|
|
"num_tokens": 1126789.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.197589540481568,
|
|
"epoch": 0.5143721633888049,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.8414,
|
|
"mean_token_accuracy": 0.08390456512570381,
|
|
"num_tokens": 1138192.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.102254104614258,
|
|
"epoch": 0.5194150277357539,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7849,
|
|
"mean_token_accuracy": 0.08334773629903794,
|
|
"num_tokens": 1148526.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.1709747314453125,
|
|
"epoch": 0.524457892082703,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.8869,
|
|
"mean_token_accuracy": 0.0790933720767498,
|
|
"num_tokens": 1159194.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.061290121078491,
|
|
"epoch": 0.529500756429652,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.7879,
|
|
"mean_token_accuracy": 0.09050186872482299,
|
|
"num_tokens": 1170601.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.138358449935913,
|
|
"epoch": 0.5345436207766011,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7923,
|
|
"mean_token_accuracy": 0.08816501796245575,
|
|
"num_tokens": 1182441.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.954041624069214,
|
|
"epoch": 0.5395864851235502,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6839,
|
|
"mean_token_accuracy": 0.09594669714570045,
|
|
"num_tokens": 1192530.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.019408082962036,
|
|
"epoch": 0.5446293494704992,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8118,
|
|
"mean_token_accuracy": 0.08356897532939911,
|
|
"num_tokens": 1203831.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.082006740570068,
|
|
"epoch": 0.5496722138174484,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.7515,
|
|
"mean_token_accuracy": 0.08889398127794265,
|
|
"num_tokens": 1214164.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.96448278427124,
|
|
"epoch": 0.5547150781643974,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.757,
|
|
"mean_token_accuracy": 0.08531036004424095,
|
|
"num_tokens": 1226932.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.108938455581665,
|
|
"epoch": 0.5597579425113465,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.8221,
|
|
"mean_token_accuracy": 0.0844253160059452,
|
|
"num_tokens": 1237212.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.067666864395141,
|
|
"epoch": 0.5648008068582955,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.7244,
|
|
"mean_token_accuracy": 0.08955233469605446,
|
|
"num_tokens": 1249340.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.999148416519165,
|
|
"epoch": 0.5698436712052446,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7293,
|
|
"mean_token_accuracy": 0.09257269129157067,
|
|
"num_tokens": 1260778.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.01264066696167,
|
|
"epoch": 0.5748865355521936,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7454,
|
|
"mean_token_accuracy": 0.0893318273127079,
|
|
"num_tokens": 1272216.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.973303890228271,
|
|
"epoch": 0.5799293998991427,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.822,
|
|
"mean_token_accuracy": 0.0911563828587532,
|
|
"num_tokens": 1283558.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.136276054382324,
|
|
"epoch": 0.5849722642460918,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.8604,
|
|
"mean_token_accuracy": 0.08416322097182274,
|
|
"num_tokens": 1294638.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.052536869049073,
|
|
"epoch": 0.5900151285930408,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.7796,
|
|
"mean_token_accuracy": 0.08722122386097908,
|
|
"num_tokens": 1304697.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.01437816619873,
|
|
"epoch": 0.59505799293999,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.7564,
|
|
"mean_token_accuracy": 0.08593685999512672,
|
|
"num_tokens": 1316618.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.061299085617065,
|
|
"epoch": 0.600100857286939,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.8535,
|
|
"mean_token_accuracy": 0.08204089105129242,
|
|
"num_tokens": 1328134.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.055108070373535,
|
|
"epoch": 0.6051437216338881,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.7722,
|
|
"mean_token_accuracy": 0.08621780574321747,
|
|
"num_tokens": 1339101.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.979379367828369,
|
|
"epoch": 0.6101865859808371,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.638,
|
|
"mean_token_accuracy": 0.08904936537146568,
|
|
"num_tokens": 1349181.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.934052801132202,
|
|
"epoch": 0.6152294503277862,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.7762,
|
|
"mean_token_accuracy": 0.09283258244395257,
|
|
"num_tokens": 1361443.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.0075397968292235,
|
|
"epoch": 0.6202723146747352,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7061,
|
|
"mean_token_accuracy": 0.0944965936243534,
|
|
"num_tokens": 1372807.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.948248767852784,
|
|
"epoch": 0.6253151790216843,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7824,
|
|
"mean_token_accuracy": 0.090505401045084,
|
|
"num_tokens": 1385509.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.898936414718628,
|
|
"epoch": 0.6303580433686334,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6819,
|
|
"mean_token_accuracy": 0.09380084425210952,
|
|
"num_tokens": 1396035.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.905163812637329,
|
|
"epoch": 0.6354009077155824,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.6321,
|
|
"mean_token_accuracy": 0.09524858817458152,
|
|
"num_tokens": 1406839.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.869116592407226,
|
|
"epoch": 0.6404437720625316,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6022,
|
|
"mean_token_accuracy": 0.09236029610037803,
|
|
"num_tokens": 1417708.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.954378890991211,
|
|
"epoch": 0.6454866364094806,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7294,
|
|
"mean_token_accuracy": 0.0865604043006897,
|
|
"num_tokens": 1428985.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.925297403335572,
|
|
"epoch": 0.6505295007564297,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.7463,
|
|
"mean_token_accuracy": 0.09332452937960625,
|
|
"num_tokens": 1440232.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.7807807445526125,
|
|
"epoch": 0.6555723651033787,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5228,
|
|
"mean_token_accuracy": 0.0976889856159687,
|
|
"num_tokens": 1451735.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.86837100982666,
|
|
"epoch": 0.6606152294503278,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6586,
|
|
"mean_token_accuracy": 0.09053487479686737,
|
|
"num_tokens": 1463477.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.816041040420532,
|
|
"epoch": 0.6656580937972768,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.6748,
|
|
"mean_token_accuracy": 0.08478344976902008,
|
|
"num_tokens": 1473802.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.890794849395752,
|
|
"epoch": 0.6707009581442259,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.681,
|
|
"mean_token_accuracy": 0.08961676359176636,
|
|
"num_tokens": 1485519.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.87843132019043,
|
|
"epoch": 0.675743822491175,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6714,
|
|
"mean_token_accuracy": 0.0913410633802414,
|
|
"num_tokens": 1496957.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.819877815246582,
|
|
"epoch": 0.680786686838124,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6704,
|
|
"mean_token_accuracy": 0.09737999215722085,
|
|
"num_tokens": 1508336.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.818086719512939,
|
|
"epoch": 0.6858295511850732,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.6488,
|
|
"mean_token_accuracy": 0.09069961532950402,
|
|
"num_tokens": 1520016.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.743166494369507,
|
|
"epoch": 0.6908724155320222,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.5036,
|
|
"mean_token_accuracy": 0.1011080227792263,
|
|
"num_tokens": 1531877.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.714833211898804,
|
|
"epoch": 0.6959152798789713,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4948,
|
|
"mean_token_accuracy": 0.09976917505264282,
|
|
"num_tokens": 1541553.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.845389795303345,
|
|
"epoch": 0.7009581442259203,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6568,
|
|
"mean_token_accuracy": 0.0985904410481453,
|
|
"num_tokens": 1552491.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.696604156494141,
|
|
"epoch": 0.7060010085728694,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5605,
|
|
"mean_token_accuracy": 0.09608995988965034,
|
|
"num_tokens": 1563695.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.790359306335449,
|
|
"epoch": 0.7110438729198184,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.587,
|
|
"mean_token_accuracy": 0.09952320456504822,
|
|
"num_tokens": 1573467.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.7038726806640625,
|
|
"epoch": 0.7160867372667675,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5779,
|
|
"mean_token_accuracy": 0.09964438900351524,
|
|
"num_tokens": 1584367.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.759761142730713,
|
|
"epoch": 0.7211296016137166,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5134,
|
|
"mean_token_accuracy": 0.10387681797146797,
|
|
"num_tokens": 1595301.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.706775236129761,
|
|
"epoch": 0.7261724659606656,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5675,
|
|
"mean_token_accuracy": 0.09774501919746399,
|
|
"num_tokens": 1605916.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.756863021850586,
|
|
"epoch": 0.7312153303076148,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5649,
|
|
"mean_token_accuracy": 0.09686506614089012,
|
|
"num_tokens": 1616455.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.688662767410278,
|
|
"epoch": 0.7362581946545638,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.5471,
|
|
"mean_token_accuracy": 0.10123999193310737,
|
|
"num_tokens": 1627396.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.9027222156524655,
|
|
"epoch": 0.7413010590015129,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6822,
|
|
"mean_token_accuracy": 0.08975760117173195,
|
|
"num_tokens": 1639073.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.6817467212677,
|
|
"epoch": 0.7463439233484619,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5416,
|
|
"mean_token_accuracy": 0.0988863579928875,
|
|
"num_tokens": 1649787.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.7015870094299315,
|
|
"epoch": 0.751386787695411,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4891,
|
|
"mean_token_accuracy": 0.09893131330609321,
|
|
"num_tokens": 1658883.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.636654996871949,
|
|
"epoch": 0.75642965204236,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4617,
|
|
"mean_token_accuracy": 0.1026472918689251,
|
|
"num_tokens": 1669099.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.614272832870483,
|
|
"epoch": 0.7614725163893091,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4674,
|
|
"mean_token_accuracy": 0.09820608049631119,
|
|
"num_tokens": 1680227.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.69797043800354,
|
|
"epoch": 0.7665153807362582,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4523,
|
|
"mean_token_accuracy": 0.1019385151565075,
|
|
"num_tokens": 1689901.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.533325576782227,
|
|
"epoch": 0.7715582450832073,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5359,
|
|
"mean_token_accuracy": 0.10542306154966355,
|
|
"num_tokens": 1700498.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.658027696609497,
|
|
"epoch": 0.7766011094301564,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4573,
|
|
"mean_token_accuracy": 0.10173970907926559,
|
|
"num_tokens": 1710715.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.7453186988830565,
|
|
"epoch": 0.7816439737771054,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5749,
|
|
"mean_token_accuracy": 0.09943247437477112,
|
|
"num_tokens": 1723858.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.620595216751099,
|
|
"epoch": 0.7866868381240545,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5399,
|
|
"mean_token_accuracy": 0.09783893302083016,
|
|
"num_tokens": 1734678.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.601981925964355,
|
|
"epoch": 0.7917297024710035,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4499,
|
|
"mean_token_accuracy": 0.10742662996053695,
|
|
"num_tokens": 1743869.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.722983407974243,
|
|
"epoch": 0.7967725668179526,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5238,
|
|
"mean_token_accuracy": 0.09807569980621338,
|
|
"num_tokens": 1755266.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.491190385818482,
|
|
"epoch": 0.8018154311649016,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4829,
|
|
"mean_token_accuracy": 0.10032272264361382,
|
|
"num_tokens": 1766089.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.627046537399292,
|
|
"epoch": 0.8068582955118507,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.467,
|
|
"mean_token_accuracy": 0.10166476890444756,
|
|
"num_tokens": 1776057.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.656621074676513,
|
|
"epoch": 0.8119011598587998,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5303,
|
|
"mean_token_accuracy": 0.0997236855328083,
|
|
"num_tokens": 1786729.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.6179163455963135,
|
|
"epoch": 0.8169440242057489,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4616,
|
|
"mean_token_accuracy": 0.09969348832964897,
|
|
"num_tokens": 1797708.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.630392980575562,
|
|
"epoch": 0.821986888552698,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4925,
|
|
"mean_token_accuracy": 0.10339146554470062,
|
|
"num_tokens": 1807326.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.530580949783325,
|
|
"epoch": 0.827029752899647,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4036,
|
|
"mean_token_accuracy": 0.10418965294957161,
|
|
"num_tokens": 1817266.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.583774137496948,
|
|
"epoch": 0.8320726172465961,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4454,
|
|
"mean_token_accuracy": 0.10896264910697936,
|
|
"num_tokens": 1827840.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.4405241966247555,
|
|
"epoch": 0.8371154815935451,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4218,
|
|
"mean_token_accuracy": 0.10164540708065033,
|
|
"num_tokens": 1838158.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.641921854019165,
|
|
"epoch": 0.8421583459404942,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.5088,
|
|
"mean_token_accuracy": 0.09839870259165764,
|
|
"num_tokens": 1848317.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.549338483810425,
|
|
"epoch": 0.8472012102874432,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.4267,
|
|
"mean_token_accuracy": 0.10634519904851913,
|
|
"num_tokens": 1858751.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.638240051269531,
|
|
"epoch": 0.8522440746343923,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4398,
|
|
"mean_token_accuracy": 0.10339820608496667,
|
|
"num_tokens": 1870739.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.620791673660278,
|
|
"epoch": 0.8572869389813415,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5589,
|
|
"mean_token_accuracy": 0.10414598137140274,
|
|
"num_tokens": 1880995.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.586583852767944,
|
|
"epoch": 0.8623298033282905,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5018,
|
|
"mean_token_accuracy": 0.10163085609674453,
|
|
"num_tokens": 1892633.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.634851551055908,
|
|
"epoch": 0.8673726676752396,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5586,
|
|
"mean_token_accuracy": 0.0967508465051651,
|
|
"num_tokens": 1904802.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.60012035369873,
|
|
"epoch": 0.8724155320221886,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.5619,
|
|
"mean_token_accuracy": 0.10105684995651246,
|
|
"num_tokens": 1915755.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.494285821914673,
|
|
"epoch": 0.8774583963691377,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3066,
|
|
"mean_token_accuracy": 0.10292746499180794,
|
|
"num_tokens": 1926098.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.5870774269104,
|
|
"epoch": 0.8825012607160867,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.5364,
|
|
"mean_token_accuracy": 0.10066070631146432,
|
|
"num_tokens": 1936607.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.500468444824219,
|
|
"epoch": 0.8875441250630358,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4249,
|
|
"mean_token_accuracy": 0.10577242746949196,
|
|
"num_tokens": 1948596.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.511220216751099,
|
|
"epoch": 0.8925869894099848,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4901,
|
|
"mean_token_accuracy": 0.10860564857721329,
|
|
"num_tokens": 1960322.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.646132707595825,
|
|
"epoch": 0.8976298537569339,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4182,
|
|
"mean_token_accuracy": 0.10837725177407265,
|
|
"num_tokens": 1971052.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.583981227874756,
|
|
"epoch": 0.902672718103883,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4261,
|
|
"mean_token_accuracy": 0.09693335890769958,
|
|
"num_tokens": 1980917.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.396654939651489,
|
|
"epoch": 0.9077155824508321,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3978,
|
|
"mean_token_accuracy": 0.10672854483127595,
|
|
"num_tokens": 1991035.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.52168550491333,
|
|
"epoch": 0.9127584467977812,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.41,
|
|
"mean_token_accuracy": 0.10608047693967819,
|
|
"num_tokens": 2000587.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.509775876998901,
|
|
"epoch": 0.9178013111447302,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3702,
|
|
"mean_token_accuracy": 0.10528609082102776,
|
|
"num_tokens": 2011087.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.52767915725708,
|
|
"epoch": 0.9228441754916793,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4095,
|
|
"mean_token_accuracy": 0.105078125,
|
|
"num_tokens": 2023472.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.501510429382324,
|
|
"epoch": 0.9278870398386283,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4041,
|
|
"mean_token_accuracy": 0.10640478283166885,
|
|
"num_tokens": 2033783.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.399911546707154,
|
|
"epoch": 0.9329299041855774,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4192,
|
|
"mean_token_accuracy": 0.10806669071316719,
|
|
"num_tokens": 2045882.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.524492359161377,
|
|
"epoch": 0.9379727685325264,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4363,
|
|
"mean_token_accuracy": 0.09985145330429077,
|
|
"num_tokens": 2056549.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.550910091400146,
|
|
"epoch": 0.9430156328794755,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4306,
|
|
"mean_token_accuracy": 0.11042450070381164,
|
|
"num_tokens": 2067540.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.420332193374634,
|
|
"epoch": 0.9480584972264247,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3328,
|
|
"mean_token_accuracy": 0.10801910012960433,
|
|
"num_tokens": 2077386.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.449780035018921,
|
|
"epoch": 0.9531013615733737,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.3002,
|
|
"mean_token_accuracy": 0.11137946620583535,
|
|
"num_tokens": 2087777.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.31224308013916,
|
|
"epoch": 0.9581442259203228,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.2923,
|
|
"mean_token_accuracy": 0.10804227143526077,
|
|
"num_tokens": 2099071.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.359074258804322,
|
|
"epoch": 0.9631870902672718,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3637,
|
|
"mean_token_accuracy": 0.10778041705489158,
|
|
"num_tokens": 2109168.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.582278728485107,
|
|
"epoch": 0.9682299546142209,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4789,
|
|
"mean_token_accuracy": 0.10643763318657876,
|
|
"num_tokens": 2120696.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.454548788070679,
|
|
"epoch": 0.9732728189611699,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4063,
|
|
"mean_token_accuracy": 0.10812746360898018,
|
|
"num_tokens": 2132427.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.398885440826416,
|
|
"epoch": 0.978315683308119,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3979,
|
|
"mean_token_accuracy": 0.10866176411509514,
|
|
"num_tokens": 2143120.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.551682662963867,
|
|
"epoch": 0.983358547655068,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3837,
|
|
"mean_token_accuracy": 0.10492091253399849,
|
|
"num_tokens": 2154154.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.396205472946167,
|
|
"epoch": 0.9884014120020171,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5125,
|
|
"mean_token_accuracy": 0.10000978037714958,
|
|
"num_tokens": 2167000.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.504131698608399,
|
|
"epoch": 0.9934442763489663,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.4428,
|
|
"mean_token_accuracy": 0.10678203850984573,
|
|
"num_tokens": 2178100.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.417101860046387,
|
|
"epoch": 0.9984871406959153,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.371,
|
|
"mean_token_accuracy": 0.10810489282011986,
|
|
"num_tokens": 2189063.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.555997848510742,
|
|
"epoch": 1.0030257186081695,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3104,
|
|
"mean_token_accuracy": 0.11417724440495174,
|
|
"num_tokens": 2198302.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.342493772506714,
|
|
"epoch": 1.0080685829551186,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2591,
|
|
"mean_token_accuracy": 0.11025140136480331,
|
|
"num_tokens": 2209648.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0080685829551186,
|
|
"eval_entropy": 6.458200654910721,
|
|
"eval_loss": 6.430931091308594,
|
|
"eval_mean_token_accuracy": 0.1114898744790457,
|
|
"eval_num_tokens": 2209648.0,
|
|
"eval_runtime": 2.6529,
|
|
"eval_samples_per_second": 1182.846,
|
|
"eval_steps_per_second": 148.138,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.405465030670166,
|
|
"epoch": 1.0131114473020675,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999997762223182,
|
|
"loss": 6.1818,
|
|
"mean_token_accuracy": 0.11324734687805176,
|
|
"num_tokens": 2220806.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.294467878341675,
|
|
"epoch": 1.0181543116490166,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999988671262489,
|
|
"loss": 6.1893,
|
|
"mean_token_accuracy": 0.1121673695743084,
|
|
"num_tokens": 2231806.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.352563190460205,
|
|
"epoch": 1.0231971759959657,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999972587285105,
|
|
"loss": 6.2336,
|
|
"mean_token_accuracy": 0.1214149460196495,
|
|
"num_tokens": 2243835.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.197252893447876,
|
|
"epoch": 1.0282400403429148,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999949510341015,
|
|
"loss": 6.151,
|
|
"mean_token_accuracy": 0.11364971101284027,
|
|
"num_tokens": 2254722.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.337039661407471,
|
|
"epoch": 1.033282904689864,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999919440501947,
|
|
"loss": 6.0904,
|
|
"mean_token_accuracy": 0.12333657965064049,
|
|
"num_tokens": 2264806.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.413045501708984,
|
|
"epoch": 1.0383257690368128,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999882377861355,
|
|
"loss": 6.3107,
|
|
"mean_token_accuracy": 0.1120715782046318,
|
|
"num_tokens": 2277037.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.356806755065918,
|
|
"epoch": 1.043368633383762,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999838322534433,
|
|
"loss": 6.2392,
|
|
"mean_token_accuracy": 0.10485389232635497,
|
|
"num_tokens": 2289829.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.31101713180542,
|
|
"epoch": 1.048411497730711,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999787274658107,
|
|
"loss": 6.1344,
|
|
"mean_token_accuracy": 0.1157944954931736,
|
|
"num_tokens": 2300340.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.289593505859375,
|
|
"epoch": 1.0534543620776602,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999729234391032,
|
|
"loss": 6.1735,
|
|
"mean_token_accuracy": 0.11508876532316208,
|
|
"num_tokens": 2310009.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.291887617111206,
|
|
"epoch": 1.058497226424609,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00049996642019136,
|
|
"loss": 6.1965,
|
|
"mean_token_accuracy": 0.11197640299797058,
|
|
"num_tokens": 2319290.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.3000836849212645,
|
|
"epoch": 1.0635400907715582,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999592177427933,
|
|
"loss": 6.1402,
|
|
"mean_token_accuracy": 0.11916298121213913,
|
|
"num_tokens": 2329170.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.299685478210449,
|
|
"epoch": 1.0685829551185073,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999513161157885,
|
|
"loss": 6.1672,
|
|
"mean_token_accuracy": 0.10879501402378082,
|
|
"num_tokens": 2339617.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.300838565826416,
|
|
"epoch": 1.0736258194654564,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999427153349041,
|
|
"loss": 6.1473,
|
|
"mean_token_accuracy": 0.12195768058300019,
|
|
"num_tokens": 2349548.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.1825345993042,
|
|
"epoch": 1.0786686838124055,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999334154268714,
|
|
"loss": 6.1228,
|
|
"mean_token_accuracy": 0.11572569757699966,
|
|
"num_tokens": 2359116.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.32333345413208,
|
|
"epoch": 1.0837115481593544,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999234164205949,
|
|
"loss": 6.1872,
|
|
"mean_token_accuracy": 0.1209019511938095,
|
|
"num_tokens": 2369412.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.269948101043701,
|
|
"epoch": 1.0887544125063036,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999127183471517,
|
|
"loss": 6.2578,
|
|
"mean_token_accuracy": 0.1133066512644291,
|
|
"num_tokens": 2381276.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.315757846832275,
|
|
"epoch": 1.0937972768532527,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004999013212397915,
|
|
"loss": 6.16,
|
|
"mean_token_accuracy": 0.12085302025079728,
|
|
"num_tokens": 2392135.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.140313339233399,
|
|
"epoch": 1.0988401412002018,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004998892251339368,
|
|
"loss": 6.0586,
|
|
"mean_token_accuracy": 0.12610973864793779,
|
|
"num_tokens": 2402957.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.341390085220337,
|
|
"epoch": 1.1038830055471507,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998764300671827,
|
|
"loss": 6.2069,
|
|
"mean_token_accuracy": 0.12120629996061325,
|
|
"num_tokens": 2413386.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.2240088939666744,
|
|
"epoch": 1.1089258698940998,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998629360792965,
|
|
"loss": 6.1891,
|
|
"mean_token_accuracy": 0.11651870384812354,
|
|
"num_tokens": 2424441.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.251725578308106,
|
|
"epoch": 1.113968734241049,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004998487432122179,
|
|
"loss": 6.2143,
|
|
"mean_token_accuracy": 0.11560561358928681,
|
|
"num_tokens": 2436144.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.313767528533935,
|
|
"epoch": 1.119011598587998,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998338515100585,
|
|
"loss": 6.1214,
|
|
"mean_token_accuracy": 0.12051198184490204,
|
|
"num_tokens": 2447393.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.141254854202271,
|
|
"epoch": 1.1240544629349472,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998182610191022,
|
|
"loss": 6.0922,
|
|
"mean_token_accuracy": 0.12103575617074966,
|
|
"num_tokens": 2458632.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.278240442276001,
|
|
"epoch": 1.129097327281896,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998019717878047,
|
|
"loss": 6.1761,
|
|
"mean_token_accuracy": 0.11516001969575881,
|
|
"num_tokens": 2469331.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.334226083755493,
|
|
"epoch": 1.1341401916288452,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000499784983866793,
|
|
"loss": 6.2084,
|
|
"mean_token_accuracy": 0.11247783154249191,
|
|
"num_tokens": 2479501.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.137297296524048,
|
|
"epoch": 1.1391830559757943,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997672973088661,
|
|
"loss": 6.0872,
|
|
"mean_token_accuracy": 0.1245164304971695,
|
|
"num_tokens": 2490087.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.2877974033355715,
|
|
"epoch": 1.1442259203227434,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997489121689944,
|
|
"loss": 6.1483,
|
|
"mean_token_accuracy": 0.1135520800948143,
|
|
"num_tokens": 2501130.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.271280956268311,
|
|
"epoch": 1.1492687846696923,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997298285043189,
|
|
"loss": 6.2743,
|
|
"mean_token_accuracy": 0.1111614890396595,
|
|
"num_tokens": 2513999.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.2970366954803465,
|
|
"epoch": 1.1543116490166414,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997100463741527,
|
|
"loss": 6.1762,
|
|
"mean_token_accuracy": 0.11256429478526116,
|
|
"num_tokens": 2525579.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.190529918670654,
|
|
"epoch": 1.1593545133635905,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004996895658399785,
|
|
"loss": 6.153,
|
|
"mean_token_accuracy": 0.11859855502843857,
|
|
"num_tokens": 2536091.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.212759208679199,
|
|
"epoch": 1.1643973777105396,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996683869654507,
|
|
"loss": 6.1914,
|
|
"mean_token_accuracy": 0.12091434523463249,
|
|
"num_tokens": 2546073.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.293171787261963,
|
|
"epoch": 1.1694402420574885,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996465098163934,
|
|
"loss": 6.143,
|
|
"mean_token_accuracy": 0.11481736078858376,
|
|
"num_tokens": 2556626.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.234065008163452,
|
|
"epoch": 1.1744831064044376,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004996239344608015,
|
|
"loss": 6.1563,
|
|
"mean_token_accuracy": 0.12127474620938301,
|
|
"num_tokens": 2567304.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.170280504226684,
|
|
"epoch": 1.1795259707513868,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996006609688398,
|
|
"loss": 6.0268,
|
|
"mean_token_accuracy": 0.11522144079208374,
|
|
"num_tokens": 2577242.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.203977108001709,
|
|
"epoch": 1.1845688350983359,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995766894128426,
|
|
"loss": 6.1415,
|
|
"mean_token_accuracy": 0.11265210136771202,
|
|
"num_tokens": 2588330.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.207689332962036,
|
|
"epoch": 1.189611699445285,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995520198673143,
|
|
"loss": 6.1477,
|
|
"mean_token_accuracy": 0.11861827671527862,
|
|
"num_tokens": 2598471.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.2418512344360355,
|
|
"epoch": 1.194654563792234,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995266524089283,
|
|
"loss": 6.2229,
|
|
"mean_token_accuracy": 0.11527781486511231,
|
|
"num_tokens": 2610238.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.1659543991088865,
|
|
"epoch": 1.199697428139183,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995005871165271,
|
|
"loss": 6.0876,
|
|
"mean_token_accuracy": 0.12224913090467453,
|
|
"num_tokens": 2619870.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.280199241638184,
|
|
"epoch": 1.2047402924861321,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004994738240711225,
|
|
"loss": 6.2382,
|
|
"mean_token_accuracy": 0.11500063091516495,
|
|
"num_tokens": 2631541.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.14766297340393,
|
|
"epoch": 1.2097831568330812,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004994463633558945,
|
|
"loss": 6.0612,
|
|
"mean_token_accuracy": 0.12328560650348663,
|
|
"num_tokens": 2641896.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.187077808380127,
|
|
"epoch": 1.2148260211800301,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004994182050561917,
|
|
"loss": 6.1498,
|
|
"mean_token_accuracy": 0.11411400809884072,
|
|
"num_tokens": 2654208.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.28818678855896,
|
|
"epoch": 1.2198688855269793,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004993893492595306,
|
|
"loss": 6.1362,
|
|
"mean_token_accuracy": 0.12012557908892632,
|
|
"num_tokens": 2665115.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.16124267578125,
|
|
"epoch": 1.2249117498739284,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004993597960555959,
|
|
"loss": 6.2187,
|
|
"mean_token_accuracy": 0.11932125166058541,
|
|
"num_tokens": 2676093.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.24350380897522,
|
|
"epoch": 1.2299546142208775,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004993295455362395,
|
|
"loss": 6.1806,
|
|
"mean_token_accuracy": 0.11729151234030724,
|
|
"num_tokens": 2687866.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.228955221176148,
|
|
"epoch": 1.2349974785678266,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004992985977954808,
|
|
"loss": 6.1755,
|
|
"mean_token_accuracy": 0.11626276522874832,
|
|
"num_tokens": 2699814.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.2397219181060795,
|
|
"epoch": 1.2400403429147757,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000499266952929506,
|
|
"loss": 6.1405,
|
|
"mean_token_accuracy": 0.12563820108771323,
|
|
"num_tokens": 2710384.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.220791053771973,
|
|
"epoch": 1.2450832072617246,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004992346110366682,
|
|
"loss": 6.1094,
|
|
"mean_token_accuracy": 0.1186500407755375,
|
|
"num_tokens": 2721442.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.196668577194214,
|
|
"epoch": 1.2501260716086737,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004992015722174865,
|
|
"loss": 6.1401,
|
|
"mean_token_accuracy": 0.11508488729596138,
|
|
"num_tokens": 2733605.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.185384273529053,
|
|
"epoch": 1.2551689359556228,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991678365746466,
|
|
"loss": 6.1695,
|
|
"mean_token_accuracy": 0.12907450795173644,
|
|
"num_tokens": 2744853.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.098152780532837,
|
|
"epoch": 1.2602118003025717,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004991334042129996,
|
|
"loss": 6.071,
|
|
"mean_token_accuracy": 0.13040545731782913,
|
|
"num_tokens": 2754477.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.2803586483001705,
|
|
"epoch": 1.2652546646495209,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004990982752395618,
|
|
"loss": 6.0709,
|
|
"mean_token_accuracy": 0.11997993364930153,
|
|
"num_tokens": 2765895.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.113924503326416,
|
|
"epoch": 1.27029752899647,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004990624497635152,
|
|
"loss": 6.1165,
|
|
"mean_token_accuracy": 0.11766326203942298,
|
|
"num_tokens": 2776897.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.239742136001587,
|
|
"epoch": 1.275340393343419,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000499025927896206,
|
|
"loss": 6.1527,
|
|
"mean_token_accuracy": 0.11527862325310707,
|
|
"num_tokens": 2788666.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.084677743911743,
|
|
"epoch": 1.2803832576903682,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004989887097511451,
|
|
"loss": 6.0415,
|
|
"mean_token_accuracy": 0.11637270450592041,
|
|
"num_tokens": 2799167.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.1817937850952145,
|
|
"epoch": 1.2854261220373173,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004989507954440074,
|
|
"loss": 6.0501,
|
|
"mean_token_accuracy": 0.11778480559587479,
|
|
"num_tokens": 2809774.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.182520818710327,
|
|
"epoch": 1.2904689863842662,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004989121850926313,
|
|
"loss": 6.1539,
|
|
"mean_token_accuracy": 0.11690962538123131,
|
|
"num_tokens": 2820515.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.1439462184906,
|
|
"epoch": 1.2955118507312153,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004988728788170188,
|
|
"loss": 6.0935,
|
|
"mean_token_accuracy": 0.12135230079293251,
|
|
"num_tokens": 2832163.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.165696620941162,
|
|
"epoch": 1.3005547150781644,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004988328767393346,
|
|
"loss": 6.1012,
|
|
"mean_token_accuracy": 0.12144723758101464,
|
|
"num_tokens": 2843717.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.036713552474976,
|
|
"epoch": 1.3055975794251133,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004987921789839061,
|
|
"loss": 6.056,
|
|
"mean_token_accuracy": 0.11873352527618408,
|
|
"num_tokens": 2854489.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.1999259948730465,
|
|
"epoch": 1.3106404437720625,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004987507856772229,
|
|
"loss": 6.0996,
|
|
"mean_token_accuracy": 0.11998600885272026,
|
|
"num_tokens": 2865317.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.1370643138885494,
|
|
"epoch": 1.3156833081190116,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004987086969479364,
|
|
"loss": 6.009,
|
|
"mean_token_accuracy": 0.12642660290002822,
|
|
"num_tokens": 2877466.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.0906274795532225,
|
|
"epoch": 1.3207261724659607,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004986659129268592,
|
|
"loss": 6.0479,
|
|
"mean_token_accuracy": 0.12097594514489174,
|
|
"num_tokens": 2887698.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.014158010482788,
|
|
"epoch": 1.3257690368129098,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004986224337469652,
|
|
"loss": 6.0012,
|
|
"mean_token_accuracy": 0.12554697692394257,
|
|
"num_tokens": 2899339.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.151623344421386,
|
|
"epoch": 1.330811901159859,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004985782595433886,
|
|
"loss": 6.0375,
|
|
"mean_token_accuracy": 0.12203648909926415,
|
|
"num_tokens": 2908340.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.130350542068482,
|
|
"epoch": 1.3358547655068078,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004985333904534237,
|
|
"loss": 6.1435,
|
|
"mean_token_accuracy": 0.12421361953020096,
|
|
"num_tokens": 2920432.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.201553821563721,
|
|
"epoch": 1.340897629853757,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000498487826616525,
|
|
"loss": 6.1003,
|
|
"mean_token_accuracy": 0.11795696169137955,
|
|
"num_tokens": 2930385.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.142268180847168,
|
|
"epoch": 1.345940494200706,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004984415681743056,
|
|
"loss": 6.0376,
|
|
"mean_token_accuracy": 0.12345123142004014,
|
|
"num_tokens": 2940760.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.155660104751587,
|
|
"epoch": 1.350983358547655,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004983946152705382,
|
|
"loss": 6.0635,
|
|
"mean_token_accuracy": 0.12443178966641426,
|
|
"num_tokens": 2952441.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.119363737106323,
|
|
"epoch": 1.356026222894604,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004983469680511531,
|
|
"loss": 6.081,
|
|
"mean_token_accuracy": 0.12072315961122512,
|
|
"num_tokens": 2962766.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.042787265777588,
|
|
"epoch": 1.3610690872415532,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004982986266642393,
|
|
"loss": 5.9704,
|
|
"mean_token_accuracy": 0.12917595431208612,
|
|
"num_tokens": 2974239.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.080689191818237,
|
|
"epoch": 1.3661119515885023,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004982495912600426,
|
|
"loss": 6.0177,
|
|
"mean_token_accuracy": 0.12488547414541244,
|
|
"num_tokens": 2984497.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.136371040344239,
|
|
"epoch": 1.3711548159354514,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004981998619909664,
|
|
"loss": 6.0001,
|
|
"mean_token_accuracy": 0.12997711077332497,
|
|
"num_tokens": 2995367.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.085458517074585,
|
|
"epoch": 1.3761976802824005,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004981494390115704,
|
|
"loss": 6.028,
|
|
"mean_token_accuracy": 0.12649903818964958,
|
|
"num_tokens": 3007687.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.0920158386230465,
|
|
"epoch": 1.3812405446293494,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004980983224785704,
|
|
"loss": 6.0108,
|
|
"mean_token_accuracy": 0.12236928418278695,
|
|
"num_tokens": 3017505.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.0635899066925045,
|
|
"epoch": 1.3862834089762985,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004980465125508376,
|
|
"loss": 5.9586,
|
|
"mean_token_accuracy": 0.12392374128103256,
|
|
"num_tokens": 3028271.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.042300415039063,
|
|
"epoch": 1.3913262733232477,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004979940093893986,
|
|
"loss": 6.0168,
|
|
"mean_token_accuracy": 0.12344199493527412,
|
|
"num_tokens": 3040440.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.025495195388794,
|
|
"epoch": 1.3963691376701965,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004979408131574345,
|
|
"loss": 5.9876,
|
|
"mean_token_accuracy": 0.12624408453702926,
|
|
"num_tokens": 3050999.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.1228841781616214,
|
|
"epoch": 1.4014120020171457,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004978869240202802,
|
|
"loss": 6.0356,
|
|
"mean_token_accuracy": 0.12834085524082184,
|
|
"num_tokens": 3063374.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.068223333358764,
|
|
"epoch": 1.4064548663640948,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004978323421454246,
|
|
"loss": 6.0697,
|
|
"mean_token_accuracy": 0.12431771457195281,
|
|
"num_tokens": 3075464.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.05127100944519,
|
|
"epoch": 1.411497730711044,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004977770677025092,
|
|
"loss": 5.9442,
|
|
"mean_token_accuracy": 0.12418310642242432,
|
|
"num_tokens": 3086398.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.0082621574401855,
|
|
"epoch": 1.416540595057993,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004977211008633284,
|
|
"loss": 6.0749,
|
|
"mean_token_accuracy": 0.11579525694251061,
|
|
"num_tokens": 3097777.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.13661413192749,
|
|
"epoch": 1.4215834594049421,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004976644418018285,
|
|
"loss": 6.0374,
|
|
"mean_token_accuracy": 0.1305582657456398,
|
|
"num_tokens": 3108711.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.090465688705445,
|
|
"epoch": 1.426626323751891,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000497607090694107,
|
|
"loss": 6.0721,
|
|
"mean_token_accuracy": 0.1194371834397316,
|
|
"num_tokens": 3119589.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.05711932182312,
|
|
"epoch": 1.4316691880988401,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004975490477184126,
|
|
"loss": 5.9957,
|
|
"mean_token_accuracy": 0.13197765797376632,
|
|
"num_tokens": 3131744.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.148442125320434,
|
|
"epoch": 1.4367120524457893,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000497490313055144,
|
|
"loss": 6.0373,
|
|
"mean_token_accuracy": 0.12273103445768356,
|
|
"num_tokens": 3143526.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.999296522140503,
|
|
"epoch": 1.4417549167927382,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004974308868868501,
|
|
"loss": 5.9674,
|
|
"mean_token_accuracy": 0.12395532280206681,
|
|
"num_tokens": 3154561.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.131358575820923,
|
|
"epoch": 1.4467977811396873,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004973707693982287,
|
|
"loss": 6.1497,
|
|
"mean_token_accuracy": 0.11991134583950043,
|
|
"num_tokens": 3166052.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.151666879653931,
|
|
"epoch": 1.4518406454866364,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004973099607761264,
|
|
"loss": 6.1022,
|
|
"mean_token_accuracy": 0.11915978640317917,
|
|
"num_tokens": 3179335.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.043399906158447,
|
|
"epoch": 1.4568835098335855,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004972484612095378,
|
|
"loss": 5.9872,
|
|
"mean_token_accuracy": 0.12339468076825141,
|
|
"num_tokens": 3192129.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.134992694854736,
|
|
"epoch": 1.4619263741805346,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000497186270889605,
|
|
"loss": 6.0985,
|
|
"mean_token_accuracy": 0.126957655698061,
|
|
"num_tokens": 3202792.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.145917558670044,
|
|
"epoch": 1.4669692385274837,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004971233900096169,
|
|
"loss": 6.0516,
|
|
"mean_token_accuracy": 0.12146999910473824,
|
|
"num_tokens": 3213707.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.0715807437896725,
|
|
"epoch": 1.4720121028744326,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004970598187650085,
|
|
"loss": 6.0152,
|
|
"mean_token_accuracy": 0.12506213039159775,
|
|
"num_tokens": 3224531.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.101908493041992,
|
|
"epoch": 1.4770549672213817,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000496995557353361,
|
|
"loss": 6.0029,
|
|
"mean_token_accuracy": 0.1283414624631405,
|
|
"num_tokens": 3234656.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.011713361740112,
|
|
"epoch": 1.4820978315683309,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004969306059744004,
|
|
"loss": 6.0374,
|
|
"mean_token_accuracy": 0.12599452063441277,
|
|
"num_tokens": 3246745.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.042939615249634,
|
|
"epoch": 1.4871406959152798,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004968649648299968,
|
|
"loss": 6.008,
|
|
"mean_token_accuracy": 0.12671835720539093,
|
|
"num_tokens": 3257277.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.111767959594727,
|
|
"epoch": 1.4921835602622289,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004967986341241646,
|
|
"loss": 5.9964,
|
|
"mean_token_accuracy": 0.1276458077132702,
|
|
"num_tokens": 3267704.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.070793342590332,
|
|
"epoch": 1.497226424609178,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004967316140630611,
|
|
"loss": 6.049,
|
|
"mean_token_accuracy": 0.12442609444260597,
|
|
"num_tokens": 3279772.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.053121614456177,
|
|
"epoch": 1.502269288956127,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004966639048549862,
|
|
"loss": 5.9485,
|
|
"mean_token_accuracy": 0.13137247040867805,
|
|
"num_tokens": 3290536.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.973325300216675,
|
|
"epoch": 1.5073121533030762,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004965955067103816,
|
|
"loss": 6.0261,
|
|
"mean_token_accuracy": 0.12764430716633796,
|
|
"num_tokens": 3302051.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.1504730701446535,
|
|
"epoch": 1.5123550176500253,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004965264198418305,
|
|
"loss": 5.9992,
|
|
"mean_token_accuracy": 0.12112134248018265,
|
|
"num_tokens": 3314324.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.5123550176500253,
|
|
"eval_entropy": 5.864291856003779,
|
|
"eval_loss": 6.149571418762207,
|
|
"eval_mean_token_accuracy": 0.12647812586021787,
|
|
"eval_num_tokens": 3314324.0,
|
|
"eval_runtime": 2.6524,
|
|
"eval_samples_per_second": 1183.094,
|
|
"eval_steps_per_second": 148.17,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.9334522724151615,
|
|
"epoch": 1.5173978819969742,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004964566444640564,
|
|
"loss": 5.9942,
|
|
"mean_token_accuracy": 0.12911396101117134,
|
|
"num_tokens": 3326246.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.03910117149353,
|
|
"epoch": 1.5224407463439233,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004963861807939227,
|
|
"loss": 5.9882,
|
|
"mean_token_accuracy": 0.12805913537740707,
|
|
"num_tokens": 3336308.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.965415620803833,
|
|
"epoch": 1.5274836106908725,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004963150290504321,
|
|
"loss": 5.9282,
|
|
"mean_token_accuracy": 0.1313933663070202,
|
|
"num_tokens": 3347596.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.956177520751953,
|
|
"epoch": 1.5325264750378214,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004962431894547259,
|
|
"loss": 5.9147,
|
|
"mean_token_accuracy": 0.12645500227808953,
|
|
"num_tokens": 3358370.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.085400199890136,
|
|
"epoch": 1.5375693393847705,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004961706622300834,
|
|
"loss": 5.9399,
|
|
"mean_token_accuracy": 0.12542775198817252,
|
|
"num_tokens": 3368630.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.011035633087158,
|
|
"epoch": 1.5426122037317196,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004960974476019206,
|
|
"loss": 5.9562,
|
|
"mean_token_accuracy": 0.12478697001934051,
|
|
"num_tokens": 3379952.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.001274585723877,
|
|
"epoch": 1.5476550680786687,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004960235457977905,
|
|
"loss": 5.8761,
|
|
"mean_token_accuracy": 0.13284963369369507,
|
|
"num_tokens": 3390097.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.901377439498901,
|
|
"epoch": 1.5526979324256178,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004959489570473814,
|
|
"loss": 5.8835,
|
|
"mean_token_accuracy": 0.13172722682356836,
|
|
"num_tokens": 3399778.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.054537296295166,
|
|
"epoch": 1.557740796772567,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000495873681582517,
|
|
"loss": 6.0355,
|
|
"mean_token_accuracy": 0.12638389319181442,
|
|
"num_tokens": 3410213.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.0968907356262205,
|
|
"epoch": 1.5627836611195158,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004957977196371552,
|
|
"loss": 6.0347,
|
|
"mean_token_accuracy": 0.12559032961726188,
|
|
"num_tokens": 3422071.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.062275838851929,
|
|
"epoch": 1.567826525466465,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004957210714473873,
|
|
"loss": 6.0062,
|
|
"mean_token_accuracy": 0.12689514607191085,
|
|
"num_tokens": 3432102.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.974687957763672,
|
|
"epoch": 1.572869389813414,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004956437372514378,
|
|
"loss": 5.9851,
|
|
"mean_token_accuracy": 0.1266164645552635,
|
|
"num_tokens": 3442821.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.000227212905884,
|
|
"epoch": 1.577912254160363,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004955657172896631,
|
|
"loss": 5.9408,
|
|
"mean_token_accuracy": 0.12739403322339057,
|
|
"num_tokens": 3454378.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.965472650527954,
|
|
"epoch": 1.582955118507312,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004954870118045511,
|
|
"loss": 5.9621,
|
|
"mean_token_accuracy": 0.13084643110632896,
|
|
"num_tokens": 3463933.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.005740642547607,
|
|
"epoch": 1.5879979828542612,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004954076210407203,
|
|
"loss": 5.9067,
|
|
"mean_token_accuracy": 0.12802384719252585,
|
|
"num_tokens": 3473993.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.995165014266968,
|
|
"epoch": 1.5930408472012103,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004953275452449188,
|
|
"loss": 6.0559,
|
|
"mean_token_accuracy": 0.12162444740533829,
|
|
"num_tokens": 3484882.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.034287214279175,
|
|
"epoch": 1.5980837115481594,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004952467846660243,
|
|
"loss": 5.9718,
|
|
"mean_token_accuracy": 0.1258918412029743,
|
|
"num_tokens": 3496503.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.073749017715454,
|
|
"epoch": 1.6031265758951085,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004951653395550423,
|
|
"loss": 6.0583,
|
|
"mean_token_accuracy": 0.12465130165219307,
|
|
"num_tokens": 3508696.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.072420835494995,
|
|
"epoch": 1.6081694402420574,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004950832101651062,
|
|
"loss": 5.9674,
|
|
"mean_token_accuracy": 0.12676426842808725,
|
|
"num_tokens": 3519542.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.970708322525025,
|
|
"epoch": 1.6132123045890066,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004950003967514762,
|
|
"loss": 5.9232,
|
|
"mean_token_accuracy": 0.13130599558353423,
|
|
"num_tokens": 3531576.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.022836399078369,
|
|
"epoch": 1.6182551689359557,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000494916899571538,
|
|
"loss": 5.8962,
|
|
"mean_token_accuracy": 0.13517886400222778,
|
|
"num_tokens": 3542816.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.012871313095093,
|
|
"epoch": 1.6232980332829046,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004948327188848029,
|
|
"loss": 6.0192,
|
|
"mean_token_accuracy": 0.12498244643211365,
|
|
"num_tokens": 3554100.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.0960804462432865,
|
|
"epoch": 1.6283408976298537,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004947478549529065,
|
|
"loss": 5.8722,
|
|
"mean_token_accuracy": 0.1253907360136509,
|
|
"num_tokens": 3565048.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.964908075332642,
|
|
"epoch": 1.6333837619768028,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004946623080396077,
|
|
"loss": 5.9671,
|
|
"mean_token_accuracy": 0.13184850215911864,
|
|
"num_tokens": 3575587.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.005288934707641,
|
|
"epoch": 1.638426626323752,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004945760784107884,
|
|
"loss": 6.0434,
|
|
"mean_token_accuracy": 0.12770640552043916,
|
|
"num_tokens": 3587059.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.953710603713989,
|
|
"epoch": 1.643469490670701,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004944891663344522,
|
|
"loss": 5.8552,
|
|
"mean_token_accuracy": 0.13130738958716393,
|
|
"num_tokens": 3598492.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.962081098556519,
|
|
"epoch": 1.6485123550176501,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004944015720807239,
|
|
"loss": 5.9245,
|
|
"mean_token_accuracy": 0.12875783294439316,
|
|
"num_tokens": 3609513.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.011997222900391,
|
|
"epoch": 1.653555219364599,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004943132959218484,
|
|
"loss": 5.9686,
|
|
"mean_token_accuracy": 0.12951236963272095,
|
|
"num_tokens": 3621327.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.94477105140686,
|
|
"epoch": 1.6585980837115482,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00049422433813219,
|
|
"loss": 5.8743,
|
|
"mean_token_accuracy": 0.13340896219015122,
|
|
"num_tokens": 3632136.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.9721160411834715,
|
|
"epoch": 1.6636409480584973,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004941346989882318,
|
|
"loss": 5.8301,
|
|
"mean_token_accuracy": 0.13147396594285965,
|
|
"num_tokens": 3642271.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.862770223617554,
|
|
"epoch": 1.6686838124054462,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004940443787685739,
|
|
"loss": 5.9805,
|
|
"mean_token_accuracy": 0.12712769135832785,
|
|
"num_tokens": 3652355.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.093797254562378,
|
|
"epoch": 1.6737266767523953,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004939533777539341,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.12801686748862268,
|
|
"num_tokens": 3663714.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.878468751907349,
|
|
"epoch": 1.6787695410993444,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004938616962271453,
|
|
"loss": 5.8946,
|
|
"mean_token_accuracy": 0.1276872582733631,
|
|
"num_tokens": 3674675.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.971212911605835,
|
|
"epoch": 1.6838124054462935,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004937693344731561,
|
|
"loss": 5.9785,
|
|
"mean_token_accuracy": 0.13142225295305252,
|
|
"num_tokens": 3686403.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.977137470245362,
|
|
"epoch": 1.6888552697932426,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004936762927790286,
|
|
"loss": 5.868,
|
|
"mean_token_accuracy": 0.1354008622467518,
|
|
"num_tokens": 3697259.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.918380641937256,
|
|
"epoch": 1.6938981341401917,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004935825714339389,
|
|
"loss": 5.9338,
|
|
"mean_token_accuracy": 0.13212197870016099,
|
|
"num_tokens": 3708536.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.113958168029785,
|
|
"epoch": 1.6989409984871406,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004934881707291748,
|
|
"loss": 5.9316,
|
|
"mean_token_accuracy": 0.12909703329205513,
|
|
"num_tokens": 3719181.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.859345769882202,
|
|
"epoch": 1.7039838628340898,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004933930909581362,
|
|
"loss": 5.7748,
|
|
"mean_token_accuracy": 0.13730273768305779,
|
|
"num_tokens": 3729325.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.905084085464478,
|
|
"epoch": 1.7090267271810389,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004932973324163328,
|
|
"loss": 5.9148,
|
|
"mean_token_accuracy": 0.12552084773778915,
|
|
"num_tokens": 3740462.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.018700504302979,
|
|
"epoch": 1.7140695915279878,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004932008954013846,
|
|
"loss": 6.0062,
|
|
"mean_token_accuracy": 0.1278520070016384,
|
|
"num_tokens": 3752523.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.909162616729736,
|
|
"epoch": 1.7191124558749369,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004931037802130201,
|
|
"loss": 5.8645,
|
|
"mean_token_accuracy": 0.12955691739916803,
|
|
"num_tokens": 3764120.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.9912786960601805,
|
|
"epoch": 1.724155320221886,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004930059871530755,
|
|
"loss": 5.9983,
|
|
"mean_token_accuracy": 0.12769825533032417,
|
|
"num_tokens": 3776733.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.006366109848022,
|
|
"epoch": 1.7291981845688351,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004929075165254938,
|
|
"loss": 5.9695,
|
|
"mean_token_accuracy": 0.1266148678958416,
|
|
"num_tokens": 3787799.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.869033432006836,
|
|
"epoch": 1.7342410489157842,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004928083686363241,
|
|
"loss": 5.8866,
|
|
"mean_token_accuracy": 0.12758740782737732,
|
|
"num_tokens": 3797995.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.028029441833496,
|
|
"epoch": 1.7392839132627333,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004927085437937204,
|
|
"loss": 5.9694,
|
|
"mean_token_accuracy": 0.12862191274762153,
|
|
"num_tokens": 3809580.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.001883459091187,
|
|
"epoch": 1.7443267776096822,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004926080423079404,
|
|
"loss": 5.9956,
|
|
"mean_token_accuracy": 0.12958402037620545,
|
|
"num_tokens": 3820150.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.948792552947998,
|
|
"epoch": 1.7493696419566314,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492506864491345,
|
|
"loss": 5.8413,
|
|
"mean_token_accuracy": 0.1305758535861969,
|
|
"num_tokens": 3830391.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.850123071670533,
|
|
"epoch": 1.7544125063035805,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004924050106583976,
|
|
"loss": 5.8606,
|
|
"mean_token_accuracy": 0.1337277926504612,
|
|
"num_tokens": 3842773.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.9990283966064455,
|
|
"epoch": 1.7594553706505294,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004923024811256618,
|
|
"loss": 5.8985,
|
|
"mean_token_accuracy": 0.1337128020823002,
|
|
"num_tokens": 3853646.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.929308748245239,
|
|
"epoch": 1.7644982349974785,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000492199276211802,
|
|
"loss": 5.8556,
|
|
"mean_token_accuracy": 0.13106711879372596,
|
|
"num_tokens": 3865319.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.936299753189087,
|
|
"epoch": 1.7695410993444276,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004920953962375813,
|
|
"loss": 5.9191,
|
|
"mean_token_accuracy": 0.1285109229385853,
|
|
"num_tokens": 3877107.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.973802900314331,
|
|
"epoch": 1.7745839636913767,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004919908415258612,
|
|
"loss": 5.96,
|
|
"mean_token_accuracy": 0.1286729894578457,
|
|
"num_tokens": 3889334.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.960503768920899,
|
|
"epoch": 1.7796268280383258,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004918856124016002,
|
|
"loss": 5.8451,
|
|
"mean_token_accuracy": 0.13625560626387595,
|
|
"num_tokens": 3899482.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.865010595321655,
|
|
"epoch": 1.784669692385275,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004917797091918526,
|
|
"loss": 5.786,
|
|
"mean_token_accuracy": 0.13576927557587623,
|
|
"num_tokens": 3910216.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.978150987625122,
|
|
"epoch": 1.789712556732224,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004916731322257683,
|
|
"loss": 5.9347,
|
|
"mean_token_accuracy": 0.13337727561593055,
|
|
"num_tokens": 3922246.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.903500509262085,
|
|
"epoch": 1.794755421079173,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004915658818345908,
|
|
"loss": 5.8299,
|
|
"mean_token_accuracy": 0.1367899715900421,
|
|
"num_tokens": 3933688.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.987347650527954,
|
|
"epoch": 1.799798285426122,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004914579583516568,
|
|
"loss": 5.9358,
|
|
"mean_token_accuracy": 0.12906345203518868,
|
|
"num_tokens": 3944168.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.907552337646484,
|
|
"epoch": 1.804841149773071,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491349362112395,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.13501922264695168,
|
|
"num_tokens": 3954346.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.966559553146363,
|
|
"epoch": 1.80988401412002,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000491240093454325,
|
|
"loss": 5.9741,
|
|
"mean_token_accuracy": 0.1284855455160141,
|
|
"num_tokens": 3966375.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.9653175354003904,
|
|
"epoch": 1.8149268784669692,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004911301527170563,
|
|
"loss": 5.9301,
|
|
"mean_token_accuracy": 0.13323267549276352,
|
|
"num_tokens": 3977030.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.888005161285401,
|
|
"epoch": 1.8199697428139183,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004910195402422873,
|
|
"loss": 5.8496,
|
|
"mean_token_accuracy": 0.13025863468647003,
|
|
"num_tokens": 3988064.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.95345287322998,
|
|
"epoch": 1.8250126071608674,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004909082563738042,
|
|
"loss": 5.8862,
|
|
"mean_token_accuracy": 0.12944839894771576,
|
|
"num_tokens": 3997994.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.986985683441162,
|
|
"epoch": 1.8300554715078166,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004907963014574797,
|
|
"loss": 5.954,
|
|
"mean_token_accuracy": 0.1267230324447155,
|
|
"num_tokens": 4008645.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.92745943069458,
|
|
"epoch": 1.8350983358547657,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004906836758412722,
|
|
"loss": 5.8873,
|
|
"mean_token_accuracy": 0.13214431703090668,
|
|
"num_tokens": 4020387.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.783616828918457,
|
|
"epoch": 1.8401412002017146,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004905703798752248,
|
|
"loss": 5.7658,
|
|
"mean_token_accuracy": 0.13267371356487273,
|
|
"num_tokens": 4032720.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.915006446838379,
|
|
"epoch": 1.8451840645486637,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000490456413911464,
|
|
"loss": 5.8502,
|
|
"mean_token_accuracy": 0.1391013965010643,
|
|
"num_tokens": 4043438.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.847107315063477,
|
|
"epoch": 1.8502269288956126,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004903417783041987,
|
|
"loss": 5.8221,
|
|
"mean_token_accuracy": 0.13950808122754096,
|
|
"num_tokens": 4054467.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.879653882980347,
|
|
"epoch": 1.8552697932425617,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004902264734097188,
|
|
"loss": 5.8135,
|
|
"mean_token_accuracy": 0.13670793995261193,
|
|
"num_tokens": 4066418.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.988006734848023,
|
|
"epoch": 1.8603126575895108,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004901104995863948,
|
|
"loss": 5.9167,
|
|
"mean_token_accuracy": 0.12878741100430488,
|
|
"num_tokens": 4078770.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.905097246170044,
|
|
"epoch": 1.86535552193646,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004899938571946759,
|
|
"loss": 5.9471,
|
|
"mean_token_accuracy": 0.12999097257852554,
|
|
"num_tokens": 4089998.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.934671878814697,
|
|
"epoch": 1.870398386283409,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004898765465970894,
|
|
"loss": 5.8473,
|
|
"mean_token_accuracy": 0.13673469424247742,
|
|
"num_tokens": 4101409.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.9459198951721195,
|
|
"epoch": 1.8754412506303582,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004897585681582393,
|
|
"loss": 5.9822,
|
|
"mean_token_accuracy": 0.12754397690296174,
|
|
"num_tokens": 4113598.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.887400436401367,
|
|
"epoch": 1.8804841149773073,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004896399222448053,
|
|
"loss": 5.9212,
|
|
"mean_token_accuracy": 0.13141812086105348,
|
|
"num_tokens": 4124639.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.034184169769287,
|
|
"epoch": 1.8855269793242562,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004895206092255417,
|
|
"loss": 5.9825,
|
|
"mean_token_accuracy": 0.12615976333618165,
|
|
"num_tokens": 4136280.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.897401332855225,
|
|
"epoch": 1.8905698436712053,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004894006294712761,
|
|
"loss": 5.8223,
|
|
"mean_token_accuracy": 0.12734010741114615,
|
|
"num_tokens": 4148251.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.965174198150635,
|
|
"epoch": 1.8956127080181542,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004892799833549083,
|
|
"loss": 5.9036,
|
|
"mean_token_accuracy": 0.13231924250721933,
|
|
"num_tokens": 4159298.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.938666772842407,
|
|
"epoch": 1.9006555723651033,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004891586712514091,
|
|
"loss": 5.8149,
|
|
"mean_token_accuracy": 0.1375449776649475,
|
|
"num_tokens": 4169974.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.869181060791016,
|
|
"epoch": 1.9056984367120524,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004890366935378196,
|
|
"loss": 5.8375,
|
|
"mean_token_accuracy": 0.13779282346367835,
|
|
"num_tokens": 4181442.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.881608295440674,
|
|
"epoch": 1.9107413010590015,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004889140505932492,
|
|
"loss": 5.8617,
|
|
"mean_token_accuracy": 0.1402617871761322,
|
|
"num_tokens": 4191535.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.936998558044434,
|
|
"epoch": 1.9157841654059506,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004887907427988749,
|
|
"loss": 5.8886,
|
|
"mean_token_accuracy": 0.13966521695256234,
|
|
"num_tokens": 4201741.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.869378900527954,
|
|
"epoch": 1.9208270297528998,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004886667705379403,
|
|
"loss": 5.781,
|
|
"mean_token_accuracy": 0.1395824022591114,
|
|
"num_tokens": 4211762.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.957179403305053,
|
|
"epoch": 1.9258698940998489,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004885421341957541,
|
|
"loss": 5.8717,
|
|
"mean_token_accuracy": 0.12964706718921662,
|
|
"num_tokens": 4222322.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.93777813911438,
|
|
"epoch": 1.9309127584467978,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004884168341596888,
|
|
"loss": 5.9455,
|
|
"mean_token_accuracy": 0.13559314534068106,
|
|
"num_tokens": 4232166.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.911707353591919,
|
|
"epoch": 1.9359556227937469,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004882908708191797,
|
|
"loss": 5.896,
|
|
"mean_token_accuracy": 0.1303203970193863,
|
|
"num_tokens": 4243651.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.9245704174041744,
|
|
"epoch": 1.9409984871406958,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00048816424456572403,
|
|
"loss": 5.8105,
|
|
"mean_token_accuracy": 0.14012144654989242,
|
|
"num_tokens": 4255137.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.810333347320556,
|
|
"epoch": 1.946041351487645,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048803695579287905,
|
|
"loss": 5.8252,
|
|
"mean_token_accuracy": 0.13574693575501443,
|
|
"num_tokens": 4264904.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.8669086456298825,
|
|
"epoch": 1.951084215834594,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048790900489626104,
|
|
"loss": 5.7953,
|
|
"mean_token_accuracy": 0.13575853258371354,
|
|
"num_tokens": 4275305.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.900700807571411,
|
|
"epoch": 1.9561270801815431,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004877803922735445,
|
|
"loss": 5.8906,
|
|
"mean_token_accuracy": 0.12943072393536567,
|
|
"num_tokens": 4287038.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.007995367050171,
|
|
"epoch": 1.9611699445284922,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004876511183244603,
|
|
"loss": 6.0065,
|
|
"mean_token_accuracy": 0.12546259090304374,
|
|
"num_tokens": 4298668.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.829781103134155,
|
|
"epoch": 1.9662128088754414,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00048752118345079496,
|
|
"loss": 5.7539,
|
|
"mean_token_accuracy": 0.13618924468755722,
|
|
"num_tokens": 4309429.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.923287296295166,
|
|
"epoch": 1.9712556732223905,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048739058805638905,
|
|
"loss": 5.8425,
|
|
"mean_token_accuracy": 0.13548057079315184,
|
|
"num_tokens": 4321137.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.871845054626465,
|
|
"epoch": 1.9762985375693394,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000487259332547136,
|
|
"loss": 5.8945,
|
|
"mean_token_accuracy": 0.12442709431052208,
|
|
"num_tokens": 4331249.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.9540855407714846,
|
|
"epoch": 1.9813414019162885,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000487127417330981,
|
|
"loss": 5.8644,
|
|
"mean_token_accuracy": 0.12946647256612778,
|
|
"num_tokens": 4342154.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.956533145904541,
|
|
"epoch": 1.9863842662632374,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004869948428179196,
|
|
"loss": 5.9176,
|
|
"mean_token_accuracy": 0.12653955966234207,
|
|
"num_tokens": 4353906.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.914293241500855,
|
|
"epoch": 1.9914271306101865,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004868616094199964,
|
|
"loss": 5.9594,
|
|
"mean_token_accuracy": 0.12459009885787964,
|
|
"num_tokens": 4366416.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.935325622558594,
|
|
"epoch": 1.9964699949571356,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000486727717551304,
|
|
"loss": 5.8479,
|
|
"mean_token_accuracy": 0.13273880258202553,
|
|
"num_tokens": 4376743.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.808435704973009,
|
|
"epoch": 2.00100857286939,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004865931676279813,
|
|
"loss": 5.6119,
|
|
"mean_token_accuracy": 0.15014778905444676,
|
|
"num_tokens": 4385379.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.88061752319336,
|
|
"epoch": 2.006051437216339,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004864579600682125,
|
|
"loss": 5.5627,
|
|
"mean_token_accuracy": 0.14014450311660767,
|
|
"num_tokens": 4396872.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.83287410736084,
|
|
"epoch": 2.011094301563288,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048632209529222607,
|
|
"loss": 5.4937,
|
|
"mean_token_accuracy": 0.14493266195058824,
|
|
"num_tokens": 4407075.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.780134487152099,
|
|
"epoch": 2.016137165910237,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048618557372229253,
|
|
"loss": 5.5274,
|
|
"mean_token_accuracy": 0.15092427432537078,
|
|
"num_tokens": 4418020.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.016137165910237,
|
|
"eval_entropy": 5.737285053457013,
|
|
"eval_loss": 5.987236976623535,
|
|
"eval_mean_token_accuracy": 0.1333525952037054,
|
|
"eval_num_tokens": 4418020.0,
|
|
"eval_runtime": 2.647,
|
|
"eval_samples_per_second": 1185.48,
|
|
"eval_steps_per_second": 148.468,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.809401512145996,
|
|
"epoch": 2.0211800302571863,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004860483957827243,
|
|
"loss": 5.5503,
|
|
"mean_token_accuracy": 0.14347360730171205,
|
|
"num_tokens": 4429740.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.803671216964721,
|
|
"epoch": 2.026222894604135,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004859105618998735,
|
|
"loss": 5.5234,
|
|
"mean_token_accuracy": 0.14536709189414979,
|
|
"num_tokens": 4440183.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.811364841461182,
|
|
"epoch": 2.031265758951084,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004857720725021311,
|
|
"loss": 5.5086,
|
|
"mean_token_accuracy": 0.14281092286109925,
|
|
"num_tokens": 4451750.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.654035997390747,
|
|
"epoch": 2.036308623298033,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048563292801992534,
|
|
"loss": 5.4667,
|
|
"mean_token_accuracy": 0.1473421186208725,
|
|
"num_tokens": 4462071.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.911445903778076,
|
|
"epoch": 2.0413514876449823,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004854931288857203,
|
|
"loss": 5.6541,
|
|
"mean_token_accuracy": 0.13716498389840126,
|
|
"num_tokens": 4474241.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.776958703994751,
|
|
"epoch": 2.0463943519919314,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00048535267553401505,
|
|
"loss": 5.491,
|
|
"mean_token_accuracy": 0.14453579783439635,
|
|
"num_tokens": 4487083.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.71920690536499,
|
|
"epoch": 2.0514372163388805,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048521156840134193,
|
|
"loss": 5.5464,
|
|
"mean_token_accuracy": 0.152300226688385,
|
|
"num_tokens": 4497181.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.802086544036865,
|
|
"epoch": 2.0564800806858297,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000485069807926265,
|
|
"loss": 5.5087,
|
|
"mean_token_accuracy": 0.1469719037413597,
|
|
"num_tokens": 4508250.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.782970714569092,
|
|
"epoch": 2.0615229450327788,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004849273945493792,
|
|
"loss": 5.5308,
|
|
"mean_token_accuracy": 0.15052359253168107,
|
|
"num_tokens": 4519778.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.725796842575074,
|
|
"epoch": 2.066565809379728,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004847843287133084,
|
|
"loss": 5.5111,
|
|
"mean_token_accuracy": 0.14419954493641854,
|
|
"num_tokens": 4530879.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.771544408798218,
|
|
"epoch": 2.0716086737266766,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004846406108627048,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.14897709637880324,
|
|
"num_tokens": 4542560.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.699616146087647,
|
|
"epoch": 2.0766515380736257,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048449624144424665,
|
|
"loss": 5.4682,
|
|
"mean_token_accuracy": 0.15372542589902877,
|
|
"num_tokens": 4552213.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.757037782669068,
|
|
"epoch": 2.081694402420575,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048435122090663744,
|
|
"loss": 5.4892,
|
|
"mean_token_accuracy": 0.15251175612211226,
|
|
"num_tokens": 4562798.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.709850597381592,
|
|
"epoch": 2.086737266767524,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048420554970060424,
|
|
"loss": 5.5303,
|
|
"mean_token_accuracy": 0.14272531345486641,
|
|
"num_tokens": 4574534.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.754495573043823,
|
|
"epoch": 2.091780131114473,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048405922827889677,
|
|
"loss": 5.5165,
|
|
"mean_token_accuracy": 0.14036599695682525,
|
|
"num_tokens": 4586239.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.677375364303589,
|
|
"epoch": 2.096822995461422,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048391225709628515,
|
|
"loss": 5.5462,
|
|
"mean_token_accuracy": 0.14933456629514694,
|
|
"num_tokens": 4598018.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.75076642036438,
|
|
"epoch": 2.1018658598083713,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004837646366095593,
|
|
"loss": 5.523,
|
|
"mean_token_accuracy": 0.14465831518173217,
|
|
"num_tokens": 4608985.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.79936032295227,
|
|
"epoch": 2.1069087241553204,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048361636727752716,
|
|
"loss": 5.586,
|
|
"mean_token_accuracy": 0.14790917485952376,
|
|
"num_tokens": 4621103.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.725647449493408,
|
|
"epoch": 2.1119515885022695,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004834674495610131,
|
|
"loss": 5.5624,
|
|
"mean_token_accuracy": 0.14707302153110505,
|
|
"num_tokens": 4632244.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.728663396835327,
|
|
"epoch": 2.116994452849218,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048331788392285693,
|
|
"loss": 5.4739,
|
|
"mean_token_accuracy": 0.14647374972701072,
|
|
"num_tokens": 4642387.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.753600311279297,
|
|
"epoch": 2.1220373171961673,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048316767082791205,
|
|
"loss": 5.6162,
|
|
"mean_token_accuracy": 0.14376127645373343,
|
|
"num_tokens": 4655181.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.67578501701355,
|
|
"epoch": 2.1270801815431164,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048301681074304427,
|
|
"loss": 5.5267,
|
|
"mean_token_accuracy": 0.14497065246105195,
|
|
"num_tokens": 4665664.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.686625146865845,
|
|
"epoch": 2.1321230458900655,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00048286530413713016,
|
|
"loss": 5.4954,
|
|
"mean_token_accuracy": 0.15223089456558228,
|
|
"num_tokens": 4677683.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.751373338699341,
|
|
"epoch": 2.1371659102370146,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048271315148105574,
|
|
"loss": 5.5368,
|
|
"mean_token_accuracy": 0.13940742313861848,
|
|
"num_tokens": 4688891.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.719531106948852,
|
|
"epoch": 2.1422087745839637,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000482560353247715,
|
|
"loss": 5.5641,
|
|
"mean_token_accuracy": 0.14690410792827607,
|
|
"num_tokens": 4700765.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.729419279098511,
|
|
"epoch": 2.147251638930913,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004824069099120084,
|
|
"loss": 5.5805,
|
|
"mean_token_accuracy": 0.14375081583857535,
|
|
"num_tokens": 4712004.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.696139192581176,
|
|
"epoch": 2.152294503277862,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004822528219508413,
|
|
"loss": 5.4749,
|
|
"mean_token_accuracy": 0.1454821676015854,
|
|
"num_tokens": 4722427.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.791658639907837,
|
|
"epoch": 2.157337367624811,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004820980898431227,
|
|
"loss": 5.5753,
|
|
"mean_token_accuracy": 0.14931401386857032,
|
|
"num_tokens": 4733815.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.6779876232147215,
|
|
"epoch": 2.1623802319717598,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004819427140697635,
|
|
"loss": 5.5312,
|
|
"mean_token_accuracy": 0.14588711857795716,
|
|
"num_tokens": 4745070.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.778981351852417,
|
|
"epoch": 2.167423096318709,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004817866951136753,
|
|
"loss": 5.6063,
|
|
"mean_token_accuracy": 0.13915701657533647,
|
|
"num_tokens": 4756066.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.772104787826538,
|
|
"epoch": 2.172465960665658,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00048163003345976847,
|
|
"loss": 5.6164,
|
|
"mean_token_accuracy": 0.1358711190521717,
|
|
"num_tokens": 4767153.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.731611204147339,
|
|
"epoch": 2.177508825012607,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004814727295949512,
|
|
"loss": 5.5123,
|
|
"mean_token_accuracy": 0.14932651072740555,
|
|
"num_tokens": 4777554.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.670349216461181,
|
|
"epoch": 2.1825516893595562,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00048131478400812735,
|
|
"loss": 5.563,
|
|
"mean_token_accuracy": 0.1438249096274376,
|
|
"num_tokens": 4788881.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.754054307937622,
|
|
"epoch": 2.1875945537065054,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004811561971901956,
|
|
"loss": 5.5311,
|
|
"mean_token_accuracy": 0.14462420493364334,
|
|
"num_tokens": 4798680.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.766398811340332,
|
|
"epoch": 2.1926374180534545,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004809969696340474,
|
|
"loss": 5.5354,
|
|
"mean_token_accuracy": 0.1441192775964737,
|
|
"num_tokens": 4808096.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.6534160614013675,
|
|
"epoch": 2.1976802824004036,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048083710183456554,
|
|
"loss": 5.4856,
|
|
"mean_token_accuracy": 0.15018677413463594,
|
|
"num_tokens": 4818778.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.682404947280884,
|
|
"epoch": 2.2027231467473527,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048067659428862306,
|
|
"loss": 5.4984,
|
|
"mean_token_accuracy": 0.14967529624700546,
|
|
"num_tokens": 4828586.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.688867807388306,
|
|
"epoch": 2.2077660110943014,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048051544749508096,
|
|
"loss": 5.6025,
|
|
"mean_token_accuracy": 0.14644743204116822,
|
|
"num_tokens": 4839419.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.698174953460693,
|
|
"epoch": 2.2128088754412505,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004803536619547872,
|
|
"loss": 5.5149,
|
|
"mean_token_accuracy": 0.14952964037656785,
|
|
"num_tokens": 4849700.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.713203430175781,
|
|
"epoch": 2.2178517397881996,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000480191238170575,
|
|
"loss": 5.6779,
|
|
"mean_token_accuracy": 0.13699523061513902,
|
|
"num_tokens": 4860261.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.816243028640747,
|
|
"epoch": 2.2228946041351487,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004800281766472612,
|
|
"loss": 5.5874,
|
|
"mean_token_accuracy": 0.1394091561436653,
|
|
"num_tokens": 4872277.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.741755247116089,
|
|
"epoch": 2.227937468482098,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047986447789164486,
|
|
"loss": 5.6447,
|
|
"mean_token_accuracy": 0.14800818413496017,
|
|
"num_tokens": 4883512.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.70157265663147,
|
|
"epoch": 2.232980332829047,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004797001424125054,
|
|
"loss": 5.5662,
|
|
"mean_token_accuracy": 0.14716027528047562,
|
|
"num_tokens": 4896197.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.742011785507202,
|
|
"epoch": 2.238023197175996,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047953517072060124,
|
|
"loss": 5.5198,
|
|
"mean_token_accuracy": 0.14899527728557588,
|
|
"num_tokens": 4907397.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.719397354125976,
|
|
"epoch": 2.243066061522945,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004793695633286684,
|
|
"loss": 5.5959,
|
|
"mean_token_accuracy": 0.14191670343279839,
|
|
"num_tokens": 4919008.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.669133090972901,
|
|
"epoch": 2.2481089258698943,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047920332075141825,
|
|
"loss": 5.5247,
|
|
"mean_token_accuracy": 0.14321326985955238,
|
|
"num_tokens": 4930877.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.711592388153076,
|
|
"epoch": 2.253151790216843,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004790364435055367,
|
|
"loss": 5.4284,
|
|
"mean_token_accuracy": 0.15727580785751344,
|
|
"num_tokens": 4942274.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.637373447418213,
|
|
"epoch": 2.258194654563792,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000478868932109682,
|
|
"loss": 5.5271,
|
|
"mean_token_accuracy": 0.14538483768701554,
|
|
"num_tokens": 4953740.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.768154001235962,
|
|
"epoch": 2.263237518910741,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004787007870844835,
|
|
"loss": 5.5639,
|
|
"mean_token_accuracy": 0.1475571945309639,
|
|
"num_tokens": 4964760.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.682580375671387,
|
|
"epoch": 2.2682803832576903,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047853200895253976,
|
|
"loss": 5.6204,
|
|
"mean_token_accuracy": 0.14218731597065926,
|
|
"num_tokens": 4976404.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.706311130523682,
|
|
"epoch": 2.2733232476046394,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047836259823841716,
|
|
"loss": 5.5189,
|
|
"mean_token_accuracy": 0.1423356331884861,
|
|
"num_tokens": 4987552.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.664641046524048,
|
|
"epoch": 2.2783661119515886,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047819255546864797,
|
|
"loss": 5.51,
|
|
"mean_token_accuracy": 0.1469644159078598,
|
|
"num_tokens": 4998912.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.729035329818726,
|
|
"epoch": 2.2834089762985377,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004780218811717293,
|
|
"loss": 5.4879,
|
|
"mean_token_accuracy": 0.15003155618906022,
|
|
"num_tokens": 5009556.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.636032962799073,
|
|
"epoch": 2.288451840645487,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047785057587812057,
|
|
"loss": 5.5309,
|
|
"mean_token_accuracy": 0.15048451870679855,
|
|
"num_tokens": 5021569.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.747111511230469,
|
|
"epoch": 2.293494704992436,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047767864012024266,
|
|
"loss": 5.6083,
|
|
"mean_token_accuracy": 0.14447000846266747,
|
|
"num_tokens": 5032994.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.724060678482056,
|
|
"epoch": 2.2985375693393846,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047750607443247585,
|
|
"loss": 5.5622,
|
|
"mean_token_accuracy": 0.14192608892917633,
|
|
"num_tokens": 5044681.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.654747867584229,
|
|
"epoch": 2.3035804336863337,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047733287935115837,
|
|
"loss": 5.4719,
|
|
"mean_token_accuracy": 0.15386678725481034,
|
|
"num_tokens": 5055509.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.563857364654541,
|
|
"epoch": 2.308623298033283,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004771590554145844,
|
|
"loss": 5.4905,
|
|
"mean_token_accuracy": 0.15451828390359879,
|
|
"num_tokens": 5066903.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.746557378768921,
|
|
"epoch": 2.313666162380232,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047698460316300285,
|
|
"loss": 5.5525,
|
|
"mean_token_accuracy": 0.14106202200055123,
|
|
"num_tokens": 5078411.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.687992429733276,
|
|
"epoch": 2.318709026727181,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004768095231386153,
|
|
"loss": 5.5435,
|
|
"mean_token_accuracy": 0.14831157326698302,
|
|
"num_tokens": 5089811.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.693154764175415,
|
|
"epoch": 2.32375189107413,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004766338158855744,
|
|
"loss": 5.5542,
|
|
"mean_token_accuracy": 0.15550451576709748,
|
|
"num_tokens": 5101196.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.620962238311767,
|
|
"epoch": 2.3287947554210793,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004764574819499826,
|
|
"loss": 5.5324,
|
|
"mean_token_accuracy": 0.14903795272111892,
|
|
"num_tokens": 5112379.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.770616626739502,
|
|
"epoch": 2.3338376197680284,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047628052187988955,
|
|
"loss": 5.6501,
|
|
"mean_token_accuracy": 0.1362842358648777,
|
|
"num_tokens": 5125065.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.766256952285767,
|
|
"epoch": 2.338880484114977,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004761029362252914,
|
|
"loss": 5.6289,
|
|
"mean_token_accuracy": 0.14608695954084397,
|
|
"num_tokens": 5137471.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.759060907363891,
|
|
"epoch": 2.343923348461926,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047592472553812856,
|
|
"loss": 5.5592,
|
|
"mean_token_accuracy": 0.14957329034805297,
|
|
"num_tokens": 5147682.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.697702646255493,
|
|
"epoch": 2.3489662128088753,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004757458903722837,
|
|
"loss": 5.5497,
|
|
"mean_token_accuracy": 0.1418791577219963,
|
|
"num_tokens": 5156143.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.708810091018677,
|
|
"epoch": 2.3540090771558244,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00047556643128358075,
|
|
"loss": 5.5827,
|
|
"mean_token_accuracy": 0.1440706342458725,
|
|
"num_tokens": 5167462.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.645598363876343,
|
|
"epoch": 2.3590519415027735,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047538634882978277,
|
|
"loss": 5.5152,
|
|
"mean_token_accuracy": 0.14537022709846498,
|
|
"num_tokens": 5178030.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.65037088394165,
|
|
"epoch": 2.3640948058497226,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047520564357059013,
|
|
"loss": 5.5418,
|
|
"mean_token_accuracy": 0.14569444060325623,
|
|
"num_tokens": 5188296.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.610379076004028,
|
|
"epoch": 2.3691376701966718,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004750243160676389,
|
|
"loss": 5.5213,
|
|
"mean_token_accuracy": 0.1517985060811043,
|
|
"num_tokens": 5199058.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.684592199325562,
|
|
"epoch": 2.374180534543621,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004748423668844992,
|
|
"loss": 5.4952,
|
|
"mean_token_accuracy": 0.15093099176883698,
|
|
"num_tokens": 5209314.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.748591423034668,
|
|
"epoch": 2.37922339889057,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004746597965866733,
|
|
"loss": 5.6261,
|
|
"mean_token_accuracy": 0.13885305523872377,
|
|
"num_tokens": 5220248.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.738989448547363,
|
|
"epoch": 2.384266263237519,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004744766057415938,
|
|
"loss": 5.5906,
|
|
"mean_token_accuracy": 0.13861018046736717,
|
|
"num_tokens": 5230874.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.672354984283447,
|
|
"epoch": 2.389309127584468,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004742927949186222,
|
|
"loss": 5.5164,
|
|
"mean_token_accuracy": 0.15369487255811692,
|
|
"num_tokens": 5243328.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.606172132492065,
|
|
"epoch": 2.394351991931417,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047410836468904664,
|
|
"loss": 5.4689,
|
|
"mean_token_accuracy": 0.15051238089799882,
|
|
"num_tokens": 5254357.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.610012006759644,
|
|
"epoch": 2.399394856278366,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047392331562608073,
|
|
"loss": 5.4796,
|
|
"mean_token_accuracy": 0.1530187249183655,
|
|
"num_tokens": 5264492.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.563921403884888,
|
|
"epoch": 2.404437720625315,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004737376483048609,
|
|
"loss": 5.431,
|
|
"mean_token_accuracy": 0.15398082733154297,
|
|
"num_tokens": 5274816.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.650123357772827,
|
|
"epoch": 2.4094805849722642,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004735513633024458,
|
|
"loss": 5.542,
|
|
"mean_token_accuracy": 0.14859297275543212,
|
|
"num_tokens": 5286855.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.764153528213501,
|
|
"epoch": 2.4145234493192134,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004733644611978134,
|
|
"loss": 5.6489,
|
|
"mean_token_accuracy": 0.1342922605574131,
|
|
"num_tokens": 5299594.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.604626798629761,
|
|
"epoch": 2.4195663136661625,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004731769425718597,
|
|
"loss": 5.4761,
|
|
"mean_token_accuracy": 0.14971282929182053,
|
|
"num_tokens": 5310078.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.61399245262146,
|
|
"epoch": 2.4246091780131116,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000472988808007397,
|
|
"loss": 5.5083,
|
|
"mean_token_accuracy": 0.15255012810230256,
|
|
"num_tokens": 5319544.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.6471294403076175,
|
|
"epoch": 2.4296520423600603,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047280005808915176,
|
|
"loss": 5.5765,
|
|
"mean_token_accuracy": 0.1412188343703747,
|
|
"num_tokens": 5331025.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.57730622291565,
|
|
"epoch": 2.4346949067070094,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004726106934037631,
|
|
"loss": 5.541,
|
|
"mean_token_accuracy": 0.14827449172735213,
|
|
"num_tokens": 5343086.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.7858335971832275,
|
|
"epoch": 2.4397377710539585,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004724207145397809,
|
|
"loss": 5.5865,
|
|
"mean_token_accuracy": 0.1527317628264427,
|
|
"num_tokens": 5353598.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.646032762527466,
|
|
"epoch": 2.4447806354009076,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047223012208766384,
|
|
"loss": 5.4765,
|
|
"mean_token_accuracy": 0.14786297380924224,
|
|
"num_tokens": 5363607.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.625754690170288,
|
|
"epoch": 2.4498234997478567,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004720389166397777,
|
|
"loss": 5.5408,
|
|
"mean_token_accuracy": 0.14737689197063447,
|
|
"num_tokens": 5375198.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.602023267745972,
|
|
"epoch": 2.454866364094806,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047184709879039325,
|
|
"loss": 5.5168,
|
|
"mean_token_accuracy": 0.1479905441403389,
|
|
"num_tokens": 5386058.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.733691644668579,
|
|
"epoch": 2.459909228441755,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047165466913568497,
|
|
"loss": 5.5586,
|
|
"mean_token_accuracy": 0.14219130128622054,
|
|
"num_tokens": 5395720.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.632841348648071,
|
|
"epoch": 2.464952092788704,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004714616282737285,
|
|
"loss": 5.6011,
|
|
"mean_token_accuracy": 0.14688002765178682,
|
|
"num_tokens": 5406252.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.6775250911712645,
|
|
"epoch": 2.469994957135653,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004712679768044995,
|
|
"loss": 5.5973,
|
|
"mean_token_accuracy": 0.14558053016662598,
|
|
"num_tokens": 5418125.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.705634069442749,
|
|
"epoch": 2.4750378214826023,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000471073715329871,
|
|
"loss": 5.5552,
|
|
"mean_token_accuracy": 0.15152572393417357,
|
|
"num_tokens": 5428095.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.694837427139282,
|
|
"epoch": 2.4800806858295514,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004708788444536123,
|
|
"loss": 5.5744,
|
|
"mean_token_accuracy": 0.1408723622560501,
|
|
"num_tokens": 5438359.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.698658752441406,
|
|
"epoch": 2.4851235501765,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004706833647813865,
|
|
"loss": 5.512,
|
|
"mean_token_accuracy": 0.15103893131017684,
|
|
"num_tokens": 5448559.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.585665941238403,
|
|
"epoch": 2.4901664145234492,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047048727692074893,
|
|
"loss": 5.5147,
|
|
"mean_token_accuracy": 0.14688608497381211,
|
|
"num_tokens": 5459739.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.614408016204834,
|
|
"epoch": 2.4952092788703983,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004702905814811452,
|
|
"loss": 5.4881,
|
|
"mean_token_accuracy": 0.14290664494037628,
|
|
"num_tokens": 5471160.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.615060520172119,
|
|
"epoch": 2.5002521432173475,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00047009327907390926,
|
|
"loss": 5.5536,
|
|
"mean_token_accuracy": 0.14889164119958878,
|
|
"num_tokens": 5482344.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.708916330337525,
|
|
"epoch": 2.5052950075642966,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004698953703122616,
|
|
"loss": 5.5544,
|
|
"mean_token_accuracy": 0.14673320651054383,
|
|
"num_tokens": 5493111.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.499578380584717,
|
|
"epoch": 2.5103378719112457,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046969685581130715,
|
|
"loss": 5.4974,
|
|
"mean_token_accuracy": 0.15034912675619125,
|
|
"num_tokens": 5503877.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.670448541641235,
|
|
"epoch": 2.515380736258195,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046949773618803353,
|
|
"loss": 5.514,
|
|
"mean_token_accuracy": 0.1452426664531231,
|
|
"num_tokens": 5514067.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.724104166030884,
|
|
"epoch": 2.5204236006051435,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00046929801206130925,
|
|
"loss": 5.5581,
|
|
"mean_token_accuracy": 0.14580155909061432,
|
|
"num_tokens": 5525534.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.5204236006051435,
|
|
"eval_entropy": 5.600534638072395,
|
|
"eval_loss": 5.905900001525879,
|
|
"eval_mean_token_accuracy": 0.13783107661943095,
|
|
"eval_num_tokens": 5525534.0,
|
|
"eval_runtime": 2.8541,
|
|
"eval_samples_per_second": 1099.485,
|
|
"eval_steps_per_second": 137.698,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.666773843765259,
|
|
"epoch": 2.5254664649520926,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046909768405188135,
|
|
"loss": 5.5284,
|
|
"mean_token_accuracy": 0.15276879966259002,
|
|
"num_tokens": 5535525.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.569863271713257,
|
|
"epoch": 2.5305093292990417,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046889675278237404,
|
|
"loss": 5.5314,
|
|
"mean_token_accuracy": 0.15008983239531518,
|
|
"num_tokens": 5546129.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.778241443634033,
|
|
"epoch": 2.535552193645991,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004686952188772863,
|
|
"loss": 5.6515,
|
|
"mean_token_accuracy": 0.14541684687137604,
|
|
"num_tokens": 5557548.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.710212659835816,
|
|
"epoch": 2.54059505799294,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00046849308296299005,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.15283845961093903,
|
|
"num_tokens": 5567367.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.665985774993897,
|
|
"epoch": 2.545637922339889,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004682903456677285,
|
|
"loss": 5.4882,
|
|
"mean_token_accuracy": 0.1479597195982933,
|
|
"num_tokens": 5577489.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.635674524307251,
|
|
"epoch": 2.550680786686838,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046808700762161385,
|
|
"loss": 5.5392,
|
|
"mean_token_accuracy": 0.13918927758932115,
|
|
"num_tokens": 5587877.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.553708171844482,
|
|
"epoch": 2.5557236510337873,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004678830694566253,
|
|
"loss": 5.5413,
|
|
"mean_token_accuracy": 0.1495114952325821,
|
|
"num_tokens": 5597863.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.633197546005249,
|
|
"epoch": 2.5607665153807364,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004676785318066075,
|
|
"loss": 5.5456,
|
|
"mean_token_accuracy": 0.1501670554280281,
|
|
"num_tokens": 5608141.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.622183847427368,
|
|
"epoch": 2.5658093797276855,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004674733953072682,
|
|
"loss": 5.6176,
|
|
"mean_token_accuracy": 0.14584484994411467,
|
|
"num_tokens": 5619660.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.796398591995239,
|
|
"epoch": 2.5708522440746346,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004672676605961763,
|
|
"loss": 5.599,
|
|
"mean_token_accuracy": 0.14210569560527803,
|
|
"num_tokens": 5630271.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.660651731491089,
|
|
"epoch": 2.5758951084215833,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046706132831276007,
|
|
"loss": 5.5872,
|
|
"mean_token_accuracy": 0.1474399074912071,
|
|
"num_tokens": 5642594.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.644368505477905,
|
|
"epoch": 2.5809379727685324,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046685439909830486,
|
|
"loss": 5.535,
|
|
"mean_token_accuracy": 0.14901075512170792,
|
|
"num_tokens": 5653361.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.585793113708496,
|
|
"epoch": 2.5859808371154815,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004666468735959516,
|
|
"loss": 5.4526,
|
|
"mean_token_accuracy": 0.14740752205252647,
|
|
"num_tokens": 5664167.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.631259918212891,
|
|
"epoch": 2.5910237014624307,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004664387524506943,
|
|
"loss": 5.5454,
|
|
"mean_token_accuracy": 0.1464836224913597,
|
|
"num_tokens": 5675903.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.665949106216431,
|
|
"epoch": 2.59606656580938,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004662300363093784,
|
|
"loss": 5.5227,
|
|
"mean_token_accuracy": 0.14747685864567756,
|
|
"num_tokens": 5686319.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.737546682357788,
|
|
"epoch": 2.601109430156329,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004660207258206983,
|
|
"loss": 5.6216,
|
|
"mean_token_accuracy": 0.14249770417809487,
|
|
"num_tokens": 5696000.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.6564465999603275,
|
|
"epoch": 2.606152294503278,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046581082163519585,
|
|
"loss": 5.5242,
|
|
"mean_token_accuracy": 0.14804662466049195,
|
|
"num_tokens": 5708145.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.575733470916748,
|
|
"epoch": 2.6111951588502267,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004656003244052582,
|
|
"loss": 5.4823,
|
|
"mean_token_accuracy": 0.1528120219707489,
|
|
"num_tokens": 5718053.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.65503158569336,
|
|
"epoch": 2.616238023197176,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004653892347851155,
|
|
"loss": 5.5574,
|
|
"mean_token_accuracy": 0.1495755970478058,
|
|
"num_tokens": 5728723.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.554058837890625,
|
|
"epoch": 2.621280887544125,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004651775534308392,
|
|
"loss": 5.4818,
|
|
"mean_token_accuracy": 0.1565033659338951,
|
|
"num_tokens": 5739414.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.708927392959595,
|
|
"epoch": 2.626323751891074,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004649652810003398,
|
|
"loss": 5.6331,
|
|
"mean_token_accuracy": 0.1417702041566372,
|
|
"num_tokens": 5751247.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.624011087417602,
|
|
"epoch": 2.631366616238023,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046475241815336504,
|
|
"loss": 5.4838,
|
|
"mean_token_accuracy": 0.15057189613580704,
|
|
"num_tokens": 5761791.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.5540141582489015,
|
|
"epoch": 2.6364094805849723,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004645389655514974,
|
|
"loss": 5.4792,
|
|
"mean_token_accuracy": 0.14652240872383118,
|
|
"num_tokens": 5773597.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.608112382888794,
|
|
"epoch": 2.6414523449319214,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004643249238581525,
|
|
"loss": 5.4888,
|
|
"mean_token_accuracy": 0.14806195273995398,
|
|
"num_tokens": 5782963.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.536990547180176,
|
|
"epoch": 2.6464952092788705,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004641102937385769,
|
|
"loss": 5.5192,
|
|
"mean_token_accuracy": 0.14364688619971275,
|
|
"num_tokens": 5793474.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.685369443893433,
|
|
"epoch": 2.6515380736258196,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004638950758598461,
|
|
"loss": 5.5157,
|
|
"mean_token_accuracy": 0.14668485075235366,
|
|
"num_tokens": 5804018.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.647674226760865,
|
|
"epoch": 2.6565809379727687,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046367927089086203,
|
|
"loss": 5.5466,
|
|
"mean_token_accuracy": 0.1479173108935356,
|
|
"num_tokens": 5815897.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.636591053009033,
|
|
"epoch": 2.661623802319718,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046346287950235163,
|
|
"loss": 5.5778,
|
|
"mean_token_accuracy": 0.1426468722522259,
|
|
"num_tokens": 5825883.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.67188081741333,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004632459023668642,
|
|
"loss": 5.5473,
|
|
"mean_token_accuracy": 0.1464351862668991,
|
|
"num_tokens": 5836668.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.678479337692261,
|
|
"epoch": 2.6717095310136156,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004630283401587699,
|
|
"loss": 5.6682,
|
|
"mean_token_accuracy": 0.14157723709940911,
|
|
"num_tokens": 5848497.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.614854860305786,
|
|
"epoch": 2.6767523953605648,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004628101935542568,
|
|
"loss": 5.481,
|
|
"mean_token_accuracy": 0.15395515114068986,
|
|
"num_tokens": 5859231.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.635919427871704,
|
|
"epoch": 2.681795259707514,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046259146323132986,
|
|
"loss": 5.6279,
|
|
"mean_token_accuracy": 0.1475918710231781,
|
|
"num_tokens": 5871190.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.672294855117798,
|
|
"epoch": 2.686838124054463,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046237214986980776,
|
|
"loss": 5.4987,
|
|
"mean_token_accuracy": 0.15098831057548523,
|
|
"num_tokens": 5881788.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.614300060272217,
|
|
"epoch": 2.691880988401412,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004621522541513216,
|
|
"loss": 5.484,
|
|
"mean_token_accuracy": 0.15050791203975677,
|
|
"num_tokens": 5891289.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.5231529712677006,
|
|
"epoch": 2.696923852748361,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004619317767593123,
|
|
"loss": 5.5186,
|
|
"mean_token_accuracy": 0.14828689843416215,
|
|
"num_tokens": 5903594.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.75124225616455,
|
|
"epoch": 2.70196671709531,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046171071837902873,
|
|
"loss": 5.6426,
|
|
"mean_token_accuracy": 0.1478206306695938,
|
|
"num_tokens": 5914592.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.60038194656372,
|
|
"epoch": 2.707009581442259,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004614890796975254,
|
|
"loss": 5.453,
|
|
"mean_token_accuracy": 0.15100838541984557,
|
|
"num_tokens": 5927137.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.56253399848938,
|
|
"epoch": 2.712052445789208,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004612668614036605,
|
|
"loss": 5.5716,
|
|
"mean_token_accuracy": 0.14723662585020064,
|
|
"num_tokens": 5938730.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.712194919586182,
|
|
"epoch": 2.7170953101361572,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004610440641880934,
|
|
"loss": 5.5717,
|
|
"mean_token_accuracy": 0.14497406929731368,
|
|
"num_tokens": 5948912.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.5859886646270756,
|
|
"epoch": 2.7221381744831064,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046082068874328324,
|
|
"loss": 5.5159,
|
|
"mean_token_accuracy": 0.15253227651119233,
|
|
"num_tokens": 5960253.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.707232904434204,
|
|
"epoch": 2.7271810388300555,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046059673576348596,
|
|
"loss": 5.6009,
|
|
"mean_token_accuracy": 0.14174754545092583,
|
|
"num_tokens": 5972615.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.689057874679565,
|
|
"epoch": 2.7322239031770046,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046037220594475264,
|
|
"loss": 5.5742,
|
|
"mean_token_accuracy": 0.15167654901742936,
|
|
"num_tokens": 5985023.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.56736273765564,
|
|
"epoch": 2.7372667675239537,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000460147099984927,
|
|
"loss": 5.4731,
|
|
"mean_token_accuracy": 0.15524855107069016,
|
|
"num_tokens": 5996445.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.6111010074615475,
|
|
"epoch": 2.742309631870903,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004599214185836438,
|
|
"loss": 5.5326,
|
|
"mean_token_accuracy": 0.1473022297024727,
|
|
"num_tokens": 6007385.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.612886762619018,
|
|
"epoch": 2.747352496217852,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045969516244232574,
|
|
"loss": 5.5623,
|
|
"mean_token_accuracy": 0.15052144825458527,
|
|
"num_tokens": 6018174.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.6352783203125,
|
|
"epoch": 2.752395360564801,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00045946833226418235,
|
|
"loss": 5.4906,
|
|
"mean_token_accuracy": 0.1489933043718338,
|
|
"num_tokens": 6030049.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.56928391456604,
|
|
"epoch": 2.7574382249117497,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004592409287542069,
|
|
"loss": 5.5283,
|
|
"mean_token_accuracy": 0.14642690122127533,
|
|
"num_tokens": 6040155.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.578905773162842,
|
|
"epoch": 2.762481089258699,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004590129526191748,
|
|
"loss": 5.5237,
|
|
"mean_token_accuracy": 0.15142621397972106,
|
|
"num_tokens": 6050569.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.652458143234253,
|
|
"epoch": 2.767523953605648,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004587844045676413,
|
|
"loss": 5.4728,
|
|
"mean_token_accuracy": 0.15524515211582185,
|
|
"num_tokens": 6060871.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.588338565826416,
|
|
"epoch": 2.772566817952597,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045855528530993874,
|
|
"loss": 5.4885,
|
|
"mean_token_accuracy": 0.1549958810210228,
|
|
"num_tokens": 6073045.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.602509307861328,
|
|
"epoch": 2.777609682299546,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045832559555817525,
|
|
"loss": 5.5415,
|
|
"mean_token_accuracy": 0.14560777842998504,
|
|
"num_tokens": 6084798.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.591440486907959,
|
|
"epoch": 2.7826525466464953,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045809533602623185,
|
|
"loss": 5.4789,
|
|
"mean_token_accuracy": 0.14812592715024947,
|
|
"num_tokens": 6095831.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.624585247039795,
|
|
"epoch": 2.7876954109934444,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004578645074297603,
|
|
"loss": 5.4447,
|
|
"mean_token_accuracy": 0.15388002395629882,
|
|
"num_tokens": 6106958.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.54981746673584,
|
|
"epoch": 2.792738275340393,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045763311048618144,
|
|
"loss": 5.4715,
|
|
"mean_token_accuracy": 0.15723398625850676,
|
|
"num_tokens": 6118165.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.565126419067383,
|
|
"epoch": 2.797781139687342,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004574011459146822,
|
|
"loss": 5.4768,
|
|
"mean_token_accuracy": 0.155544276535511,
|
|
"num_tokens": 6129297.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.627363061904907,
|
|
"epoch": 2.8028240040342913,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00045716861443621386,
|
|
"loss": 5.4924,
|
|
"mean_token_accuracy": 0.14900696873664857,
|
|
"num_tokens": 6140071.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.578464984893799,
|
|
"epoch": 2.8078668683812404,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045693551677348967,
|
|
"loss": 5.5061,
|
|
"mean_token_accuracy": 0.14534874707460405,
|
|
"num_tokens": 6150940.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.612773418426514,
|
|
"epoch": 2.8129097327281896,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004567018536509823,
|
|
"loss": 5.5071,
|
|
"mean_token_accuracy": 0.14299963265657425,
|
|
"num_tokens": 6161966.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.690543937683105,
|
|
"epoch": 2.8179525970751387,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045646762579492245,
|
|
"loss": 5.5642,
|
|
"mean_token_accuracy": 0.1474120259284973,
|
|
"num_tokens": 6174455.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.583759737014771,
|
|
"epoch": 2.822995461422088,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045623283393329546,
|
|
"loss": 5.4819,
|
|
"mean_token_accuracy": 0.14665819257497786,
|
|
"num_tokens": 6184789.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.61003623008728,
|
|
"epoch": 2.828038325769037,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045599747879584,
|
|
"loss": 5.5761,
|
|
"mean_token_accuracy": 0.14073846191167833,
|
|
"num_tokens": 6195704.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.677083444595337,
|
|
"epoch": 2.833081190115986,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004557615611140451,
|
|
"loss": 5.4681,
|
|
"mean_token_accuracy": 0.15004574358463288,
|
|
"num_tokens": 6206290.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.643630456924439,
|
|
"epoch": 2.838124054462935,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004555250816211485,
|
|
"loss": 5.5673,
|
|
"mean_token_accuracy": 0.14479778259992598,
|
|
"num_tokens": 6216660.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.570545816421509,
|
|
"epoch": 2.8431669188098843,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00045528804105213395,
|
|
"loss": 5.5061,
|
|
"mean_token_accuracy": 0.15498915314674377,
|
|
"num_tokens": 6228005.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.557367277145386,
|
|
"epoch": 2.848209783156833,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004550504401437289,
|
|
"loss": 5.4691,
|
|
"mean_token_accuracy": 0.15004847794771195,
|
|
"num_tokens": 6239447.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.654346990585327,
|
|
"epoch": 2.853252647503782,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00045481227963440235,
|
|
"loss": 5.5754,
|
|
"mean_token_accuracy": 0.13897361159324645,
|
|
"num_tokens": 6252312.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.616372871398926,
|
|
"epoch": 2.858295511850731,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045457356026436295,
|
|
"loss": 5.5136,
|
|
"mean_token_accuracy": 0.1475168228149414,
|
|
"num_tokens": 6264131.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.482991456985474,
|
|
"epoch": 2.8633383761976803,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00045433428277555577,
|
|
"loss": 5.4617,
|
|
"mean_token_accuracy": 0.1529868423938751,
|
|
"num_tokens": 6275161.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.57648344039917,
|
|
"epoch": 2.8683812405446294,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004540944479116609,
|
|
"loss": 5.5013,
|
|
"mean_token_accuracy": 0.14636597633361817,
|
|
"num_tokens": 6285655.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.657945346832276,
|
|
"epoch": 2.8734241048915785,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004538540564180906,
|
|
"loss": 5.6137,
|
|
"mean_token_accuracy": 0.14963025376200675,
|
|
"num_tokens": 6296532.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.7276524066925045,
|
|
"epoch": 2.8784669692385276,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004536131090419871,
|
|
"loss": 5.6315,
|
|
"mean_token_accuracy": 0.14828269481658934,
|
|
"num_tokens": 6308229.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.607042980194092,
|
|
"epoch": 2.8835098335854763,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004533716065322205,
|
|
"loss": 5.5516,
|
|
"mean_token_accuracy": 0.15048472285270692,
|
|
"num_tokens": 6321257.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.697853660583496,
|
|
"epoch": 2.8885526979324254,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004531295496393861,
|
|
"loss": 5.4976,
|
|
"mean_token_accuracy": 0.14919596016407013,
|
|
"num_tokens": 6332040.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.541913366317749,
|
|
"epoch": 2.8935955622793745,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004528869391158023,
|
|
"loss": 5.3792,
|
|
"mean_token_accuracy": 0.15894148051738738,
|
|
"num_tokens": 6342925.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.605545425415039,
|
|
"epoch": 2.8986384266263236,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00045264377571550834,
|
|
"loss": 5.5548,
|
|
"mean_token_accuracy": 0.14902669340372085,
|
|
"num_tokens": 6354263.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.712835454940796,
|
|
"epoch": 2.9036812909732728,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004524000601942615,
|
|
"loss": 5.5565,
|
|
"mean_token_accuracy": 0.14853587597608567,
|
|
"num_tokens": 6365497.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.66733980178833,
|
|
"epoch": 2.908724155320222,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004521557933095354,
|
|
"loss": 5.602,
|
|
"mean_token_accuracy": 0.1435862325131893,
|
|
"num_tokens": 6376501.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.625383424758911,
|
|
"epoch": 2.913767019667171,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000451910975820517,
|
|
"loss": 5.5404,
|
|
"mean_token_accuracy": 0.14271508753299714,
|
|
"num_tokens": 6387632.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.552101802825928,
|
|
"epoch": 2.91880988401412,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045166560848810476,
|
|
"loss": 5.5446,
|
|
"mean_token_accuracy": 0.15111299604177475,
|
|
"num_tokens": 6398813.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.683384370803833,
|
|
"epoch": 2.9238527483610692,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045141969207490606,
|
|
"loss": 5.4446,
|
|
"mean_token_accuracy": 0.151677268743515,
|
|
"num_tokens": 6409532.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.61807107925415,
|
|
"epoch": 2.9288956127080183,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004511732273452347,
|
|
"loss": 5.5862,
|
|
"mean_token_accuracy": 0.14124868363142012,
|
|
"num_tokens": 6420755.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.69642915725708,
|
|
"epoch": 2.9339384770549675,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004509262150651089,
|
|
"loss": 5.6112,
|
|
"mean_token_accuracy": 0.15009240955114364,
|
|
"num_tokens": 6430662.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.594843959808349,
|
|
"epoch": 2.938981341401916,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045067865600224833,
|
|
"loss": 5.4879,
|
|
"mean_token_accuracy": 0.1497061714529991,
|
|
"num_tokens": 6441230.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.636964797973633,
|
|
"epoch": 2.9440242057488653,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004504305509260724,
|
|
"loss": 5.619,
|
|
"mean_token_accuracy": 0.13896541818976402,
|
|
"num_tokens": 6452574.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.638625764846802,
|
|
"epoch": 2.9490670700958144,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004501819006076974,
|
|
"loss": 5.4786,
|
|
"mean_token_accuracy": 0.15589796304702758,
|
|
"num_tokens": 6462562.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.529220104217529,
|
|
"epoch": 2.9541099344427635,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044993270581993413,
|
|
"loss": 5.5053,
|
|
"mean_token_accuracy": 0.14860141575336455,
|
|
"num_tokens": 6473594.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.643556213378906,
|
|
"epoch": 2.9591527987897126,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00044968296733728587,
|
|
"loss": 5.5162,
|
|
"mean_token_accuracy": 0.15147386044263839,
|
|
"num_tokens": 6484129.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.679654693603515,
|
|
"epoch": 2.9641956631366617,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004494326859359456,
|
|
"loss": 5.4922,
|
|
"mean_token_accuracy": 0.1495089814066887,
|
|
"num_tokens": 6495584.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.571310520172119,
|
|
"epoch": 2.969238527483611,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044918186239379363,
|
|
"loss": 5.581,
|
|
"mean_token_accuracy": 0.1475920483469963,
|
|
"num_tokens": 6507872.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.576633024215698,
|
|
"epoch": 2.9742813918305595,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004489304974903953,
|
|
"loss": 5.4523,
|
|
"mean_token_accuracy": 0.1542717307806015,
|
|
"num_tokens": 6519172.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.614698839187622,
|
|
"epoch": 2.9793242561775086,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004486785920069985,
|
|
"loss": 5.5553,
|
|
"mean_token_accuracy": 0.15236086547374725,
|
|
"num_tokens": 6530151.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.648719215393067,
|
|
"epoch": 2.9843671205244577,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004484261467265313,
|
|
"loss": 5.5313,
|
|
"mean_token_accuracy": 0.14916694536805153,
|
|
"num_tokens": 6541881.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.620734024047851,
|
|
"epoch": 2.989409984871407,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00044817316243359956,
|
|
"loss": 5.5525,
|
|
"mean_token_accuracy": 0.1492960438132286,
|
|
"num_tokens": 6553390.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.61136474609375,
|
|
"epoch": 2.994452849218356,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044791963991448427,
|
|
"loss": 5.3773,
|
|
"mean_token_accuracy": 0.16189919263124466,
|
|
"num_tokens": 6563754.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.53472146987915,
|
|
"epoch": 2.999495713565305,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044766557995713915,
|
|
"loss": 5.4588,
|
|
"mean_token_accuracy": 0.14736676812171937,
|
|
"num_tokens": 6574124.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.634968545701769,
|
|
"epoch": 3.0040342914775593,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004474109833511885,
|
|
"loss": 5.2539,
|
|
"mean_token_accuracy": 0.15890773137410483,
|
|
"num_tokens": 6582795.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.53092737197876,
|
|
"epoch": 3.0090771558245084,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004471558508879244,
|
|
"loss": 5.1777,
|
|
"mean_token_accuracy": 0.167185677587986,
|
|
"num_tokens": 6593649.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.489673662185669,
|
|
"epoch": 3.0141200201714575,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004469001833603045,
|
|
"loss": 5.1972,
|
|
"mean_token_accuracy": 0.16715353727340698,
|
|
"num_tokens": 6606370.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.518502235412598,
|
|
"epoch": 3.0191628845184066,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004466439815629493,
|
|
"loss": 5.1302,
|
|
"mean_token_accuracy": 0.1652125358581543,
|
|
"num_tokens": 6616599.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.5980949878692625,
|
|
"epoch": 3.0242057488653553,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044638724629214,
|
|
"loss": 5.2149,
|
|
"mean_token_accuracy": 0.15740568339824676,
|
|
"num_tokens": 6627707.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.0242057488653553,
|
|
"eval_entropy": 5.453453542015328,
|
|
"eval_loss": 5.838198661804199,
|
|
"eval_mean_token_accuracy": 0.14045951890808936,
|
|
"eval_num_tokens": 6627707.0,
|
|
"eval_runtime": 2.6348,
|
|
"eval_samples_per_second": 1190.983,
|
|
"eval_steps_per_second": 149.158,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.545943260192871,
|
|
"epoch": 3.0292486132123044,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044612997834581557,
|
|
"loss": 5.1257,
|
|
"mean_token_accuracy": 0.1654035523533821,
|
|
"num_tokens": 6639001.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.500567817687989,
|
|
"epoch": 3.0342914775592535,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044587217852357086,
|
|
"loss": 5.1983,
|
|
"mean_token_accuracy": 0.16164908409118653,
|
|
"num_tokens": 6649318.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.5300596237182615,
|
|
"epoch": 3.0393343419062027,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044561384762665363,
|
|
"loss": 5.2425,
|
|
"mean_token_accuracy": 0.15802418440580368,
|
|
"num_tokens": 6660584.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.496360540390015,
|
|
"epoch": 3.044377206253152,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004453549864579623,
|
|
"loss": 5.079,
|
|
"mean_token_accuracy": 0.17203515022993088,
|
|
"num_tokens": 6671361.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.531878709793091,
|
|
"epoch": 3.049420070600101,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004450955958220433,
|
|
"loss": 5.3063,
|
|
"mean_token_accuracy": 0.15716103464365005,
|
|
"num_tokens": 6681896.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.5471045017242435,
|
|
"epoch": 3.05446293494705,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00044483567652508864,
|
|
"loss": 5.1726,
|
|
"mean_token_accuracy": 0.1614293023943901,
|
|
"num_tokens": 6693360.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.481127166748047,
|
|
"epoch": 3.059505799293999,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004445752293749337,
|
|
"loss": 5.2316,
|
|
"mean_token_accuracy": 0.1649590477347374,
|
|
"num_tokens": 6704921.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.5438759326934814,
|
|
"epoch": 3.0645486636409482,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004443142551810541,
|
|
"loss": 5.1877,
|
|
"mean_token_accuracy": 0.1661015808582306,
|
|
"num_tokens": 6715659.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.445790338516235,
|
|
"epoch": 3.069591527987897,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044405275475456365,
|
|
"loss": 5.1171,
|
|
"mean_token_accuracy": 0.17525551319122315,
|
|
"num_tokens": 6726431.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.458484840393067,
|
|
"epoch": 3.074634392334846,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004437907289082116,
|
|
"loss": 5.1968,
|
|
"mean_token_accuracy": 0.16822191774845124,
|
|
"num_tokens": 6738002.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.507405233383179,
|
|
"epoch": 3.079677256681795,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044352817845638053,
|
|
"loss": 5.1644,
|
|
"mean_token_accuracy": 0.16749805063009263,
|
|
"num_tokens": 6748509.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.530843257904053,
|
|
"epoch": 3.0847201210287443,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004432651042150833,
|
|
"loss": 5.2062,
|
|
"mean_token_accuracy": 0.1669891133904457,
|
|
"num_tokens": 6759258.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.473256063461304,
|
|
"epoch": 3.0897629853756934,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044300150700196054,
|
|
"loss": 5.2222,
|
|
"mean_token_accuracy": 0.1665765330195427,
|
|
"num_tokens": 6771263.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.5137354850769045,
|
|
"epoch": 3.0948058497226425,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004427373876362786,
|
|
"loss": 5.1756,
|
|
"mean_token_accuracy": 0.16439436823129655,
|
|
"num_tokens": 6779965.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.486676788330078,
|
|
"epoch": 3.0998487140695916,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004424727469389265,
|
|
"loss": 5.1614,
|
|
"mean_token_accuracy": 0.16554953157901764,
|
|
"num_tokens": 6791521.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.5311308860778805,
|
|
"epoch": 3.1048915784165407,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004422075857324137,
|
|
"loss": 5.2315,
|
|
"mean_token_accuracy": 0.16569456905126573,
|
|
"num_tokens": 6803537.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.526635265350341,
|
|
"epoch": 3.10993444276349,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004419419048408673,
|
|
"loss": 5.2213,
|
|
"mean_token_accuracy": 0.16253284215927125,
|
|
"num_tokens": 6813811.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.483383989334106,
|
|
"epoch": 3.1149773071104385,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004416757050900297,
|
|
"loss": 5.1621,
|
|
"mean_token_accuracy": 0.16567203551530837,
|
|
"num_tokens": 6825422.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.612085771560669,
|
|
"epoch": 3.1200201714573876,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004414089873072557,
|
|
"loss": 5.2946,
|
|
"mean_token_accuracy": 0.15961285680532455,
|
|
"num_tokens": 6837086.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.466375160217285,
|
|
"epoch": 3.1250630358043368,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004411417523215105,
|
|
"loss": 5.1686,
|
|
"mean_token_accuracy": 0.16706090718507766,
|
|
"num_tokens": 6847195.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.425228595733643,
|
|
"epoch": 3.130105900151286,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004408740009633665,
|
|
"loss": 5.1579,
|
|
"mean_token_accuracy": 0.16677275151014329,
|
|
"num_tokens": 6857492.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.548114442825318,
|
|
"epoch": 3.135148764498235,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00044060573406500117,
|
|
"loss": 5.3486,
|
|
"mean_token_accuracy": 0.15285347253084183,
|
|
"num_tokens": 6869190.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.5041419506073,
|
|
"epoch": 3.140191628845184,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004403369524601941,
|
|
"loss": 5.1479,
|
|
"mean_token_accuracy": 0.1656806841492653,
|
|
"num_tokens": 6880728.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.425574016571045,
|
|
"epoch": 3.145234493192133,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004400676569843249,
|
|
"loss": 5.1779,
|
|
"mean_token_accuracy": 0.1623378574848175,
|
|
"num_tokens": 6892043.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.507378196716308,
|
|
"epoch": 3.1502773575390823,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043979784847436994,
|
|
"loss": 5.1989,
|
|
"mean_token_accuracy": 0.1686200961470604,
|
|
"num_tokens": 6903913.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.36406626701355,
|
|
"epoch": 3.1553202218860315,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004395275277689004,
|
|
"loss": 5.1814,
|
|
"mean_token_accuracy": 0.16108253002166747,
|
|
"num_tokens": 6913951.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.423825979232788,
|
|
"epoch": 3.16036308623298,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004392566957080793,
|
|
"loss": 5.1705,
|
|
"mean_token_accuracy": 0.16641268283128738,
|
|
"num_tokens": 6924762.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.490802478790283,
|
|
"epoch": 3.1654059505799292,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00043898535313365884,
|
|
"loss": 5.1439,
|
|
"mean_token_accuracy": 0.16724391728639604,
|
|
"num_tokens": 6935845.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.4740684032440186,
|
|
"epoch": 3.1704488149268784,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000438713500888978,
|
|
"loss": 5.1981,
|
|
"mean_token_accuracy": 0.1667625367641449,
|
|
"num_tokens": 6947338.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.480612707138062,
|
|
"epoch": 3.1754916792738275,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004384411398189599,
|
|
"loss": 5.2206,
|
|
"mean_token_accuracy": 0.16403878182172776,
|
|
"num_tokens": 6958084.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.461968803405762,
|
|
"epoch": 3.1805345436207766,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000438168270770109,
|
|
"loss": 5.1728,
|
|
"mean_token_accuracy": 0.1674924984574318,
|
|
"num_tokens": 6968063.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.491331672668457,
|
|
"epoch": 3.1855774079677257,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004378948945905085,
|
|
"loss": 5.1654,
|
|
"mean_token_accuracy": 0.16830475777387618,
|
|
"num_tokens": 6978424.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.355680131912232,
|
|
"epoch": 3.190620272314675,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00043762101212981805,
|
|
"loss": 5.1507,
|
|
"mean_token_accuracy": 0.16448075771331788,
|
|
"num_tokens": 6991067.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.439352703094483,
|
|
"epoch": 3.195663136661624,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043734662423927054,
|
|
"loss": 5.2403,
|
|
"mean_token_accuracy": 0.16961376667022704,
|
|
"num_tokens": 7000642.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.58800859451294,
|
|
"epoch": 3.200706001008573,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004370717317716699,
|
|
"loss": 5.2629,
|
|
"mean_token_accuracy": 0.16388253569602967,
|
|
"num_tokens": 7011305.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.457569980621338,
|
|
"epoch": 3.205748865355522,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004367963355813882,
|
|
"loss": 5.2117,
|
|
"mean_token_accuracy": 0.1585549458861351,
|
|
"num_tokens": 7021452.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.451887798309326,
|
|
"epoch": 3.210791729702471,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043652043652436323,
|
|
"loss": 5.2788,
|
|
"mean_token_accuracy": 0.1594686910510063,
|
|
"num_tokens": 7033189.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.539680480957031,
|
|
"epoch": 3.21583459404942,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004362440354580956,
|
|
"loss": 5.2273,
|
|
"mean_token_accuracy": 0.16164455115795134,
|
|
"num_tokens": 7044153.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.46336259841919,
|
|
"epoch": 3.220877458396369,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004359671332416462,
|
|
"loss": 5.267,
|
|
"mean_token_accuracy": 0.1549703910946846,
|
|
"num_tokens": 7056679.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.513670921325684,
|
|
"epoch": 3.225920322743318,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004356897307356335,
|
|
"loss": 5.1625,
|
|
"mean_token_accuracy": 0.1659749686717987,
|
|
"num_tokens": 7067428.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.432787227630615,
|
|
"epoch": 3.2309631870902673,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004354118288022309,
|
|
"loss": 5.2629,
|
|
"mean_token_accuracy": 0.15948830991983415,
|
|
"num_tokens": 7079196.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.481943511962891,
|
|
"epoch": 3.2360060514372164,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043513342830516405,
|
|
"loss": 5.1599,
|
|
"mean_token_accuracy": 0.16453344076871873,
|
|
"num_tokens": 7088936.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.449812030792236,
|
|
"epoch": 3.2410489157841655,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000434854530109708,
|
|
"loss": 5.254,
|
|
"mean_token_accuracy": 0.16199739575386046,
|
|
"num_tokens": 7101504.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.5198564529418945,
|
|
"epoch": 3.2460917801311147,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00043457513508268487,
|
|
"loss": 5.2324,
|
|
"mean_token_accuracy": 0.16156151592731477,
|
|
"num_tokens": 7114121.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.405584192276001,
|
|
"epoch": 3.2511346444780633,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043429524409246083,
|
|
"loss": 5.1348,
|
|
"mean_token_accuracy": 0.16553013175725936,
|
|
"num_tokens": 7123321.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.447930097579956,
|
|
"epoch": 3.2561775088250124,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004340148580089436,
|
|
"loss": 5.2105,
|
|
"mean_token_accuracy": 0.16655311584472657,
|
|
"num_tokens": 7134145.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.565892601013184,
|
|
"epoch": 3.2612203731719616,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004337339777035794,
|
|
"loss": 5.2691,
|
|
"mean_token_accuracy": 0.15880045890808106,
|
|
"num_tokens": 7144727.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.381158876419067,
|
|
"epoch": 3.2662632375189107,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000433452604049351,
|
|
"loss": 5.1666,
|
|
"mean_token_accuracy": 0.17045472413301468,
|
|
"num_tokens": 7155531.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.40645923614502,
|
|
"epoch": 3.27130610186586,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00043317073792077394,
|
|
"loss": 5.2005,
|
|
"mean_token_accuracy": 0.16811218708753586,
|
|
"num_tokens": 7167234.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.487696981430053,
|
|
"epoch": 3.276348966212809,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004328883801938947,
|
|
"loss": 5.2869,
|
|
"mean_token_accuracy": 0.15416017770767212,
|
|
"num_tokens": 7178556.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.398592710494995,
|
|
"epoch": 3.281391830559758,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043260553174628776,
|
|
"loss": 5.1238,
|
|
"mean_token_accuracy": 0.17282683700323104,
|
|
"num_tokens": 7189074.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.4875829219818115,
|
|
"epoch": 3.286434694906707,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004323221934570524,
|
|
"loss": 5.2749,
|
|
"mean_token_accuracy": 0.1618475630879402,
|
|
"num_tokens": 7200567.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.488211441040039,
|
|
"epoch": 3.2914775592536563,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043203836620681074,
|
|
"loss": 5.3302,
|
|
"mean_token_accuracy": 0.15507401376962662,
|
|
"num_tokens": 7211012.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.4462278366088865,
|
|
"epoch": 3.2965204236006054,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043175405087770435,
|
|
"loss": 5.2034,
|
|
"mean_token_accuracy": 0.1673835724592209,
|
|
"num_tokens": 7222271.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.44311637878418,
|
|
"epoch": 3.301563287947554,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00043146924835339186,
|
|
"loss": 5.1591,
|
|
"mean_token_accuracy": 0.16978632360696794,
|
|
"num_tokens": 7235067.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.423116779327392,
|
|
"epoch": 3.306606152294503,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043118395951904605,
|
|
"loss": 5.2365,
|
|
"mean_token_accuracy": 0.1662725940346718,
|
|
"num_tokens": 7246159.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.557249546051025,
|
|
"epoch": 3.3116490166414523,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004308981852613512,
|
|
"loss": 5.277,
|
|
"mean_token_accuracy": 0.158920818567276,
|
|
"num_tokens": 7255944.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.45926194190979,
|
|
"epoch": 3.3166918809884014,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004306119264685003,
|
|
"loss": 5.1895,
|
|
"mean_token_accuracy": 0.16631488502025604,
|
|
"num_tokens": 7267273.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.400034379959107,
|
|
"epoch": 3.3217347453353505,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004303251840301923,
|
|
"loss": 5.2061,
|
|
"mean_token_accuracy": 0.15908047705888748,
|
|
"num_tokens": 7277622.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.408486938476562,
|
|
"epoch": 3.3267776096822996,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00043003795883762933,
|
|
"loss": 5.2413,
|
|
"mean_token_accuracy": 0.1579745665192604,
|
|
"num_tokens": 7288851.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.542595911026001,
|
|
"epoch": 3.3318204740292487,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00042975025178351386,
|
|
"loss": 5.2947,
|
|
"mean_token_accuracy": 0.1575707256793976,
|
|
"num_tokens": 7302175.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.452737283706665,
|
|
"epoch": 3.336863338376198,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000429462063762046,
|
|
"loss": 5.1604,
|
|
"mean_token_accuracy": 0.17357034385204315,
|
|
"num_tokens": 7312693.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.438597202301025,
|
|
"epoch": 3.3419062027231465,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042917339566892095,
|
|
"loss": 5.2133,
|
|
"mean_token_accuracy": 0.15934556275606154,
|
|
"num_tokens": 7323790.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.397851085662841,
|
|
"epoch": 3.3469490670700957,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004288842484013257,
|
|
"loss": 5.1943,
|
|
"mean_token_accuracy": 0.16560234874486923,
|
|
"num_tokens": 7333797.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.4009946346282955,
|
|
"epoch": 3.3519919314170448,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004285946228579365,
|
|
"loss": 5.2165,
|
|
"mean_token_accuracy": 0.1717174917459488,
|
|
"num_tokens": 7344674.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.473981142044067,
|
|
"epoch": 3.357034795763994,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004283045199389164,
|
|
"loss": 5.228,
|
|
"mean_token_accuracy": 0.16603025048971176,
|
|
"num_tokens": 7355978.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.555987453460693,
|
|
"epoch": 3.362077660110943,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042801394054591197,
|
|
"loss": 5.3154,
|
|
"mean_token_accuracy": 0.16348308175802231,
|
|
"num_tokens": 7367054.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.4604799270629885,
|
|
"epoch": 3.367120524457892,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042772288558205057,
|
|
"loss": 5.2873,
|
|
"mean_token_accuracy": 0.15880271643400193,
|
|
"num_tokens": 7379588.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.44103889465332,
|
|
"epoch": 3.3721633888048412,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004274313559519377,
|
|
"loss": 5.2549,
|
|
"mean_token_accuracy": 0.16040123403072357,
|
|
"num_tokens": 7392411.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.42080078125,
|
|
"epoch": 3.3772062531517903,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004271393525616544,
|
|
"loss": 5.1517,
|
|
"mean_token_accuracy": 0.16182904094457626,
|
|
"num_tokens": 7402801.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.419835615158081,
|
|
"epoch": 3.3822491174987395,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042684687631875377,
|
|
"loss": 5.2229,
|
|
"mean_token_accuracy": 0.1577916204929352,
|
|
"num_tokens": 7413084.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.410536527633667,
|
|
"epoch": 3.3872919818456886,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00042655392813225864,
|
|
"loss": 5.1571,
|
|
"mean_token_accuracy": 0.16673170924186706,
|
|
"num_tokens": 7424004.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.383282136917114,
|
|
"epoch": 3.3923348461926373,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042626050891265884,
|
|
"loss": 5.1879,
|
|
"mean_token_accuracy": 0.16679954826831817,
|
|
"num_tokens": 7434461.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.496283626556396,
|
|
"epoch": 3.3973777105395864,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004259666195719081,
|
|
"loss": 5.2307,
|
|
"mean_token_accuracy": 0.16463324725627898,
|
|
"num_tokens": 7445552.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.459835338592529,
|
|
"epoch": 3.4024205748865355,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042567226102342134,
|
|
"loss": 5.2424,
|
|
"mean_token_accuracy": 0.16179776787757874,
|
|
"num_tokens": 7457073.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.436721515655518,
|
|
"epoch": 3.4074634392334846,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004253774341820715,
|
|
"loss": 5.2247,
|
|
"mean_token_accuracy": 0.16673760414123534,
|
|
"num_tokens": 7468197.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.450838327407837,
|
|
"epoch": 3.4125063035804337,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004250821399641874,
|
|
"loss": 5.1868,
|
|
"mean_token_accuracy": 0.1643204480409622,
|
|
"num_tokens": 7480751.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.439301919937134,
|
|
"epoch": 3.417549167927383,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004247863792875502,
|
|
"loss": 5.1983,
|
|
"mean_token_accuracy": 0.16241565942764283,
|
|
"num_tokens": 7492259.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.426845550537109,
|
|
"epoch": 3.422592032274332,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042449015307139076,
|
|
"loss": 5.294,
|
|
"mean_token_accuracy": 0.15822483003139495,
|
|
"num_tokens": 7504317.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.457805871963501,
|
|
"epoch": 3.427634896621281,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042419346223638725,
|
|
"loss": 5.199,
|
|
"mean_token_accuracy": 0.15872539579868317,
|
|
"num_tokens": 7516039.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.459451484680176,
|
|
"epoch": 3.4326777609682297,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004238963077046616,
|
|
"loss": 5.2415,
|
|
"mean_token_accuracy": 0.1618718534708023,
|
|
"num_tokens": 7526095.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.364714002609253,
|
|
"epoch": 3.437720625315179,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004235986903997767,
|
|
"loss": 5.1987,
|
|
"mean_token_accuracy": 0.1616871327161789,
|
|
"num_tokens": 7537320.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.454429388046265,
|
|
"epoch": 3.442763489662128,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042330061124673437,
|
|
"loss": 5.2672,
|
|
"mean_token_accuracy": 0.16626658141613007,
|
|
"num_tokens": 7548100.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.504119157791138,
|
|
"epoch": 3.447806354009077,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042300207117197116,
|
|
"loss": 5.2857,
|
|
"mean_token_accuracy": 0.15909409075975417,
|
|
"num_tokens": 7560483.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.4595019817352295,
|
|
"epoch": 3.452849218356026,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042270307110335657,
|
|
"loss": 5.1799,
|
|
"mean_token_accuracy": 0.16843537092208863,
|
|
"num_tokens": 7571256.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.465969181060791,
|
|
"epoch": 3.4578920827029753,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004224036119701899,
|
|
"loss": 5.1671,
|
|
"mean_token_accuracy": 0.16369685381650925,
|
|
"num_tokens": 7581716.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.397514152526855,
|
|
"epoch": 3.4629349470499244,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042210369470319694,
|
|
"loss": 5.211,
|
|
"mean_token_accuracy": 0.16916721910238267,
|
|
"num_tokens": 7591932.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.454275703430175,
|
|
"epoch": 3.4679778113968736,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004218033202345274,
|
|
"loss": 5.2047,
|
|
"mean_token_accuracy": 0.16532821655273439,
|
|
"num_tokens": 7602842.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.4039167881011965,
|
|
"epoch": 3.4730206757438227,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004215024894977522,
|
|
"loss": 5.1713,
|
|
"mean_token_accuracy": 0.16072049587965012,
|
|
"num_tokens": 7614075.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.497942018508911,
|
|
"epoch": 3.478063540090772,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004212012034278601,
|
|
"loss": 5.3453,
|
|
"mean_token_accuracy": 0.15834219455718995,
|
|
"num_tokens": 7625642.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.478870058059693,
|
|
"epoch": 3.4831064044377205,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042089946296125513,
|
|
"loss": 5.218,
|
|
"mean_token_accuracy": 0.16457697749137878,
|
|
"num_tokens": 7636447.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.343750286102295,
|
|
"epoch": 3.4881492687846696,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00042059726903575373,
|
|
"loss": 5.1297,
|
|
"mean_token_accuracy": 0.16676267087459565,
|
|
"num_tokens": 7648684.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.376708364486694,
|
|
"epoch": 3.4931921331316187,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004202946225905815,
|
|
"loss": 5.2019,
|
|
"mean_token_accuracy": 0.16291389167308806,
|
|
"num_tokens": 7660864.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.406888151168824,
|
|
"epoch": 3.498234997478568,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004199915245663706,
|
|
"loss": 5.2202,
|
|
"mean_token_accuracy": 0.16525095999240874,
|
|
"num_tokens": 7671939.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.5601184368133545,
|
|
"epoch": 3.503277861825517,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041968797590515673,
|
|
"loss": 5.2579,
|
|
"mean_token_accuracy": 0.16321595162153243,
|
|
"num_tokens": 7681536.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.481729984283447,
|
|
"epoch": 3.508320726172466,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004193839775503761,
|
|
"loss": 5.1939,
|
|
"mean_token_accuracy": 0.1660198450088501,
|
|
"num_tokens": 7693674.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.366575145721436,
|
|
"epoch": 3.513363590519415,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00041907953044686253,
|
|
"loss": 5.2268,
|
|
"mean_token_accuracy": 0.17419916093349458,
|
|
"num_tokens": 7704533.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.393583631515503,
|
|
"epoch": 3.518406454866364,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004187746355408448,
|
|
"loss": 5.2711,
|
|
"mean_token_accuracy": 0.16050518453121185,
|
|
"num_tokens": 7715448.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.500758409500122,
|
|
"epoch": 3.523449319213313,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004184692937799431,
|
|
"loss": 5.2398,
|
|
"mean_token_accuracy": 0.16774088889360428,
|
|
"num_tokens": 7726808.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.565917539596557,
|
|
"epoch": 3.528492183560262,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041816350611316693,
|
|
"loss": 5.224,
|
|
"mean_token_accuracy": 0.16495896130800247,
|
|
"num_tokens": 7737899.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.528492183560262,
|
|
"eval_entropy": 5.334265962508495,
|
|
"eval_loss": 5.817132472991943,
|
|
"eval_mean_token_accuracy": 0.1448191880952311,
|
|
"eval_num_tokens": 7737899.0,
|
|
"eval_runtime": 2.6576,
|
|
"eval_samples_per_second": 1180.751,
|
|
"eval_steps_per_second": 147.876,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.39196662902832,
|
|
"epoch": 3.533535047907211,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00041785727349091117,
|
|
"loss": 5.1847,
|
|
"mean_token_accuracy": 0.1667127326130867,
|
|
"num_tokens": 7749268.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.460740423202514,
|
|
"epoch": 3.5385779122541603,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000417550596864954,
|
|
"loss": 5.2667,
|
|
"mean_token_accuracy": 0.1632498636841774,
|
|
"num_tokens": 7760312.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.434720420837403,
|
|
"epoch": 3.5436207766011094,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041724347718845344,
|
|
"loss": 5.2506,
|
|
"mean_token_accuracy": 0.16706440597772598,
|
|
"num_tokens": 7771130.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.429781150817871,
|
|
"epoch": 3.5486636409480585,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041693591541594455,
|
|
"loss": 5.2214,
|
|
"mean_token_accuracy": 0.16056369692087175,
|
|
"num_tokens": 7783068.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.434239673614502,
|
|
"epoch": 3.5537065052950076,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004166279125033363,
|
|
"loss": 5.3265,
|
|
"mean_token_accuracy": 0.16134314090013505,
|
|
"num_tokens": 7794314.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.52089490890503,
|
|
"epoch": 3.5587493696419568,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041631946940790893,
|
|
"loss": 5.2448,
|
|
"mean_token_accuracy": 0.1675504580140114,
|
|
"num_tokens": 7806888.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.380189275741577,
|
|
"epoch": 3.563792233988906,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041601058708831077,
|
|
"loss": 5.2745,
|
|
"mean_token_accuracy": 0.16096184253692628,
|
|
"num_tokens": 7818690.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.448488569259643,
|
|
"epoch": 3.568835098335855,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000415701266504555,
|
|
"loss": 5.1787,
|
|
"mean_token_accuracy": 0.1556127056479454,
|
|
"num_tokens": 7829321.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.41423192024231,
|
|
"epoch": 3.5738779626828037,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004153915086180173,
|
|
"loss": 5.2407,
|
|
"mean_token_accuracy": 0.1691288858652115,
|
|
"num_tokens": 7840223.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.418885135650635,
|
|
"epoch": 3.578920827029753,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004150813143914323,
|
|
"loss": 5.2261,
|
|
"mean_token_accuracy": 0.15754330903291702,
|
|
"num_tokens": 7850999.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.523201036453247,
|
|
"epoch": 3.583963691376702,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004147706847888908,
|
|
"loss": 5.2831,
|
|
"mean_token_accuracy": 0.16125359684228896,
|
|
"num_tokens": 7861935.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.390472364425659,
|
|
"epoch": 3.589006555723651,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004144596207758367,
|
|
"loss": 5.2789,
|
|
"mean_token_accuracy": 0.1672527238726616,
|
|
"num_tokens": 7872960.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.481933689117431,
|
|
"epoch": 3.5940494200706,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004141481233190642,
|
|
"loss": 5.3011,
|
|
"mean_token_accuracy": 0.1603357806801796,
|
|
"num_tokens": 7883881.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.502127313613892,
|
|
"epoch": 3.5990922844175492,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004138361933867145,
|
|
"loss": 5.2989,
|
|
"mean_token_accuracy": 0.16143314242362977,
|
|
"num_tokens": 7895106.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.423171854019165,
|
|
"epoch": 3.6041351487644984,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004135238319482731,
|
|
"loss": 5.2006,
|
|
"mean_token_accuracy": 0.16672123223543167,
|
|
"num_tokens": 7906020.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.386684799194336,
|
|
"epoch": 3.609178013111447,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004132110399745666,
|
|
"loss": 5.2734,
|
|
"mean_token_accuracy": 0.15903358161449432,
|
|
"num_tokens": 7918959.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.514396715164184,
|
|
"epoch": 3.614220877458396,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004128978184377596,
|
|
"loss": 5.2324,
|
|
"mean_token_accuracy": 0.1624571457505226,
|
|
"num_tokens": 7929804.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.413375949859619,
|
|
"epoch": 3.6192637418053453,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004125841683113518,
|
|
"loss": 5.2356,
|
|
"mean_token_accuracy": 0.16197211891412736,
|
|
"num_tokens": 7941558.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.446769952774048,
|
|
"epoch": 3.6243066061522944,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041227009057017523,
|
|
"loss": 5.2659,
|
|
"mean_token_accuracy": 0.16411940455436708,
|
|
"num_tokens": 7951737.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.381877040863037,
|
|
"epoch": 3.6293494704992435,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004119555861903906,
|
|
"loss": 5.2067,
|
|
"mean_token_accuracy": 0.17239036560058593,
|
|
"num_tokens": 7963164.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.426313734054565,
|
|
"epoch": 3.6343923348461926,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004116406561494847,
|
|
"loss": 5.2343,
|
|
"mean_token_accuracy": 0.16230566799640656,
|
|
"num_tokens": 7973275.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.415792417526245,
|
|
"epoch": 3.6394351991931417,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041132530142626756,
|
|
"loss": 5.2023,
|
|
"mean_token_accuracy": 0.17363037317991256,
|
|
"num_tokens": 7984568.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.431008672714233,
|
|
"epoch": 3.644478063540091,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041100952300086884,
|
|
"loss": 5.2418,
|
|
"mean_token_accuracy": 0.15813132226467133,
|
|
"num_tokens": 7994263.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.552149391174316,
|
|
"epoch": 3.64952092788704,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00041069332185473514,
|
|
"loss": 5.2165,
|
|
"mean_token_accuracy": 0.16444766223430635,
|
|
"num_tokens": 8004881.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.390504550933838,
|
|
"epoch": 3.654563792233989,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000410376698970627,
|
|
"loss": 5.2059,
|
|
"mean_token_accuracy": 0.16623317301273347,
|
|
"num_tokens": 8015955.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.345073175430298,
|
|
"epoch": 3.659606656580938,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041005965533261554,
|
|
"loss": 5.2259,
|
|
"mean_token_accuracy": 0.16372257769107817,
|
|
"num_tokens": 8027010.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.4124006748199465,
|
|
"epoch": 3.6646495209278873,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040974219192607973,
|
|
"loss": 5.1716,
|
|
"mean_token_accuracy": 0.16332909166812898,
|
|
"num_tokens": 8038564.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.446395444869995,
|
|
"epoch": 3.669692385274836,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00040942430973770323,
|
|
"loss": 5.2473,
|
|
"mean_token_accuracy": 0.1610390529036522,
|
|
"num_tokens": 8049975.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.4358278751373295,
|
|
"epoch": 3.674735249621785,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004091060097554712,
|
|
"loss": 5.2424,
|
|
"mean_token_accuracy": 0.16687394976615905,
|
|
"num_tokens": 8061042.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.4241186618804935,
|
|
"epoch": 3.679778113968734,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004087872929686673,
|
|
"loss": 5.1779,
|
|
"mean_token_accuracy": 0.16943740099668503,
|
|
"num_tokens": 8071777.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.3719922542572025,
|
|
"epoch": 3.6848209783156833,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00040846816036787056,
|
|
"loss": 5.2673,
|
|
"mean_token_accuracy": 0.1617873266339302,
|
|
"num_tokens": 8082387.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.525816440582275,
|
|
"epoch": 3.6898638426626325,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004081486129449527,
|
|
"loss": 5.1915,
|
|
"mean_token_accuracy": 0.16953505277633668,
|
|
"num_tokens": 8091818.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.4837347030639645,
|
|
"epoch": 3.6949067070095816,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00040782865169307423,
|
|
"loss": 5.2453,
|
|
"mean_token_accuracy": 0.15914358496665953,
|
|
"num_tokens": 8103377.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.435573720932007,
|
|
"epoch": 3.6999495713565302,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004075082776066822,
|
|
"loss": 5.2858,
|
|
"mean_token_accuracy": 0.15753725916147232,
|
|
"num_tokens": 8114723.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.3904155731201175,
|
|
"epoch": 3.7049924357034794,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004071874916815068,
|
|
"loss": 5.1279,
|
|
"mean_token_accuracy": 0.1690109208226204,
|
|
"num_tokens": 8124530.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.32537522315979,
|
|
"epoch": 3.7100353000504285,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040686629491455787,
|
|
"loss": 5.2159,
|
|
"mean_token_accuracy": 0.16981774270534516,
|
|
"num_tokens": 8134894.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.422755002975464,
|
|
"epoch": 3.7150781643973776,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040654468830412247,
|
|
"loss": 5.1803,
|
|
"mean_token_accuracy": 0.1712821200489998,
|
|
"num_tokens": 8143499.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.4395421028137205,
|
|
"epoch": 3.7201210287443267,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040622267284976134,
|
|
"loss": 5.2202,
|
|
"mean_token_accuracy": 0.1582912474870682,
|
|
"num_tokens": 8154556.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.382652854919433,
|
|
"epoch": 3.725163893091276,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040590024955230596,
|
|
"loss": 5.3392,
|
|
"mean_token_accuracy": 0.15436975955963134,
|
|
"num_tokens": 8166047.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.469343185424805,
|
|
"epoch": 3.730206757438225,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004055774194138554,
|
|
"loss": 5.2293,
|
|
"mean_token_accuracy": 0.1651908576488495,
|
|
"num_tokens": 8177068.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.4093903541564945,
|
|
"epoch": 3.735249621785174,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004052541834377731,
|
|
"loss": 5.2214,
|
|
"mean_token_accuracy": 0.16594937741756438,
|
|
"num_tokens": 8187316.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.44281702041626,
|
|
"epoch": 3.740292486132123,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040493054262868397,
|
|
"loss": 5.3312,
|
|
"mean_token_accuracy": 0.16203437596559525,
|
|
"num_tokens": 8198642.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.414805889129639,
|
|
"epoch": 3.7453353504790723,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040460649799247116,
|
|
"loss": 5.2071,
|
|
"mean_token_accuracy": 0.16936664283275604,
|
|
"num_tokens": 8208476.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.392314434051514,
|
|
"epoch": 3.7503782148260214,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004042820505362728,
|
|
"loss": 5.2677,
|
|
"mean_token_accuracy": 0.16083402782678605,
|
|
"num_tokens": 8219590.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.402860355377197,
|
|
"epoch": 3.7554210791729705,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004039572012684792,
|
|
"loss": 5.1757,
|
|
"mean_token_accuracy": 0.1690693035721779,
|
|
"num_tokens": 8230707.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.460808849334716,
|
|
"epoch": 3.760463943519919,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00040363195119872916,
|
|
"loss": 5.2472,
|
|
"mean_token_accuracy": 0.15368150621652604,
|
|
"num_tokens": 8241774.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.502851819992065,
|
|
"epoch": 3.7655068078668683,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040330630133790757,
|
|
"loss": 5.3676,
|
|
"mean_token_accuracy": 0.15392172634601592,
|
|
"num_tokens": 8254314.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.434267234802246,
|
|
"epoch": 3.7705496722138174,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040298025269814165,
|
|
"loss": 5.2139,
|
|
"mean_token_accuracy": 0.16667702049016953,
|
|
"num_tokens": 8265650.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.398479080200195,
|
|
"epoch": 3.7755925365607665,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004026538062927981,
|
|
"loss": 5.1913,
|
|
"mean_token_accuracy": 0.16175859570503234,
|
|
"num_tokens": 8276808.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.397066259384156,
|
|
"epoch": 3.7806354009077157,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040232696313647985,
|
|
"loss": 5.1226,
|
|
"mean_token_accuracy": 0.16555678248405456,
|
|
"num_tokens": 8287867.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.378733825683594,
|
|
"epoch": 3.7856782652546648,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004019997242450231,
|
|
"loss": 5.1919,
|
|
"mean_token_accuracy": 0.16552965790033342,
|
|
"num_tokens": 8298707.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.359284353256226,
|
|
"epoch": 3.7907211296016134,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004016720906354937,
|
|
"loss": 5.1458,
|
|
"mean_token_accuracy": 0.16907446831464767,
|
|
"num_tokens": 8309363.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.473581409454345,
|
|
"epoch": 3.7957639939485626,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004013440633261845,
|
|
"loss": 5.2897,
|
|
"mean_token_accuracy": 0.16341162025928496,
|
|
"num_tokens": 8319376.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.403111839294434,
|
|
"epoch": 3.8008068582955117,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00040101564333661213,
|
|
"loss": 5.2022,
|
|
"mean_token_accuracy": 0.16800960004329682,
|
|
"num_tokens": 8330842.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.35449914932251,
|
|
"epoch": 3.805849722642461,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040068683168751316,
|
|
"loss": 5.1789,
|
|
"mean_token_accuracy": 0.17052550315856935,
|
|
"num_tokens": 8341588.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.404164505004883,
|
|
"epoch": 3.81089258698941,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00040035762940084205,
|
|
"loss": 5.2126,
|
|
"mean_token_accuracy": 0.16426900178194045,
|
|
"num_tokens": 8351664.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.490190029144287,
|
|
"epoch": 3.815935451336359,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000400028037499767,
|
|
"loss": 5.2968,
|
|
"mean_token_accuracy": 0.15747719556093215,
|
|
"num_tokens": 8362803.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.35491418838501,
|
|
"epoch": 3.820978315683308,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003996980570086673,
|
|
"loss": 5.2281,
|
|
"mean_token_accuracy": 0.16248423606157303,
|
|
"num_tokens": 8373791.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.423542547225952,
|
|
"epoch": 3.8260211800302573,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039936768895312987,
|
|
"loss": 5.2286,
|
|
"mean_token_accuracy": 0.16731416434049606,
|
|
"num_tokens": 8385738.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.403216075897217,
|
|
"epoch": 3.8310640443772064,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039903693435994626,
|
|
"loss": 5.2206,
|
|
"mean_token_accuracy": 0.16358872056007384,
|
|
"num_tokens": 8396924.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.446047639846801,
|
|
"epoch": 3.8361069087241555,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039870579425710937,
|
|
"loss": 5.3056,
|
|
"mean_token_accuracy": 0.15711609199643134,
|
|
"num_tokens": 8409040.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.344608306884766,
|
|
"epoch": 3.8411497730711046,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003983742696738103,
|
|
"loss": 5.1436,
|
|
"mean_token_accuracy": 0.16874172240495683,
|
|
"num_tokens": 8419630.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.4774617671966555,
|
|
"epoch": 3.8461926374180537,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003980423616404351,
|
|
"loss": 5.2999,
|
|
"mean_token_accuracy": 0.16610458195209504,
|
|
"num_tokens": 8430036.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.420436716079712,
|
|
"epoch": 3.8512355017650024,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003977100711885616,
|
|
"loss": 5.1721,
|
|
"mean_token_accuracy": 0.16778469681739808,
|
|
"num_tokens": 8439829.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.367993593215942,
|
|
"epoch": 3.8562783661119515,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039737739935095614,
|
|
"loss": 5.1933,
|
|
"mean_token_accuracy": 0.16249583214521407,
|
|
"num_tokens": 8450492.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.3878308773040775,
|
|
"epoch": 3.8613212304589006,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003970443471615705,
|
|
"loss": 5.2594,
|
|
"mean_token_accuracy": 0.16098808199167253,
|
|
"num_tokens": 8462200.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.378919792175293,
|
|
"epoch": 3.8663640948058497,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003967109156555386,
|
|
"loss": 5.2213,
|
|
"mean_token_accuracy": 0.1622820243239403,
|
|
"num_tokens": 8473393.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.41911997795105,
|
|
"epoch": 3.871406959152799,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039637710586917324,
|
|
"loss": 5.1976,
|
|
"mean_token_accuracy": 0.16271570175886155,
|
|
"num_tokens": 8484646.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.445470809936523,
|
|
"epoch": 3.876449823499748,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039604291883996287,
|
|
"loss": 5.241,
|
|
"mean_token_accuracy": 0.16419944614171983,
|
|
"num_tokens": 8495371.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.354435062408447,
|
|
"epoch": 3.8814926878466967,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039570835560656854,
|
|
"loss": 5.2341,
|
|
"mean_token_accuracy": 0.16806088089942933,
|
|
"num_tokens": 8507976.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.4419700622558596,
|
|
"epoch": 3.8865355521936458,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003953734172088205,
|
|
"loss": 5.2476,
|
|
"mean_token_accuracy": 0.17115501165390015,
|
|
"num_tokens": 8519120.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.39558367729187,
|
|
"epoch": 3.891578416540595,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039503810468771494,
|
|
"loss": 5.2218,
|
|
"mean_token_accuracy": 0.16799409985542296,
|
|
"num_tokens": 8529727.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.383208894729615,
|
|
"epoch": 3.896621280887544,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003947024190854111,
|
|
"loss": 5.1949,
|
|
"mean_token_accuracy": 0.16123012602329254,
|
|
"num_tokens": 8540202.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.457066535949707,
|
|
"epoch": 3.901664145234493,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039436636144522737,
|
|
"loss": 5.1837,
|
|
"mean_token_accuracy": 0.1703233912587166,
|
|
"num_tokens": 8550810.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.368924331665039,
|
|
"epoch": 3.9067070095814422,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039402993281163875,
|
|
"loss": 5.1907,
|
|
"mean_token_accuracy": 0.1700340211391449,
|
|
"num_tokens": 8561120.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.440017032623291,
|
|
"epoch": 3.9117498739283914,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003936931342302732,
|
|
"loss": 5.2343,
|
|
"mean_token_accuracy": 0.1647605061531067,
|
|
"num_tokens": 8572153.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.399631977081299,
|
|
"epoch": 3.9167927382753405,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003933559667479084,
|
|
"loss": 5.2081,
|
|
"mean_token_accuracy": 0.16367449164390563,
|
|
"num_tokens": 8583740.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.418714427947998,
|
|
"epoch": 3.9218356026222896,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039301843141246876,
|
|
"loss": 5.2334,
|
|
"mean_token_accuracy": 0.15926068127155305,
|
|
"num_tokens": 8596017.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.443010854721069,
|
|
"epoch": 3.9268784669692387,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039268052927302185,
|
|
"loss": 5.2258,
|
|
"mean_token_accuracy": 0.1671770140528679,
|
|
"num_tokens": 8608121.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.337048530578613,
|
|
"epoch": 3.931921331316188,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003923422613797753,
|
|
"loss": 5.254,
|
|
"mean_token_accuracy": 0.16486521512269975,
|
|
"num_tokens": 8621446.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.471338605880737,
|
|
"epoch": 3.936964195663137,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003920036287840734,
|
|
"loss": 5.2594,
|
|
"mean_token_accuracy": 0.16825543493032455,
|
|
"num_tokens": 8632088.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.413474273681641,
|
|
"epoch": 3.9420070600100856,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039166463253839434,
|
|
"loss": 5.1032,
|
|
"mean_token_accuracy": 0.17967729419469833,
|
|
"num_tokens": 8642546.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.360627174377441,
|
|
"epoch": 3.9470499243570347,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003913252736963462,
|
|
"loss": 5.2213,
|
|
"mean_token_accuracy": 0.16438516080379487,
|
|
"num_tokens": 8652869.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.449562978744507,
|
|
"epoch": 3.952092788703984,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000390985553312664,
|
|
"loss": 5.1677,
|
|
"mean_token_accuracy": 0.16542577743530273,
|
|
"num_tokens": 8662834.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.413205575942993,
|
|
"epoch": 3.957135653050933,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003906454724432065,
|
|
"loss": 5.3053,
|
|
"mean_token_accuracy": 0.16047335118055345,
|
|
"num_tokens": 8673372.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.418970155715942,
|
|
"epoch": 3.962178517397882,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039030503214495296,
|
|
"loss": 5.2151,
|
|
"mean_token_accuracy": 0.1663113847374916,
|
|
"num_tokens": 8683660.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.422430324554443,
|
|
"epoch": 3.967221381744831,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003899642334759996,
|
|
"loss": 5.2461,
|
|
"mean_token_accuracy": 0.16416452527046205,
|
|
"num_tokens": 8694044.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.412546014785766,
|
|
"epoch": 3.97226424609178,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003896230774955567,
|
|
"loss": 5.2695,
|
|
"mean_token_accuracy": 0.16129498183727264,
|
|
"num_tokens": 8705014.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.411037349700928,
|
|
"epoch": 3.977307110438729,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038928156526394476,
|
|
"loss": 5.204,
|
|
"mean_token_accuracy": 0.16352577060461043,
|
|
"num_tokens": 8716007.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.445450735092163,
|
|
"epoch": 3.982349974785678,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003889396978425918,
|
|
"loss": 5.2236,
|
|
"mean_token_accuracy": 0.16732474863529206,
|
|
"num_tokens": 8728390.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.390902042388916,
|
|
"epoch": 3.987392839132627,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003885974762940295,
|
|
"loss": 5.1634,
|
|
"mean_token_accuracy": 0.16677574962377548,
|
|
"num_tokens": 8738806.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.345384407043457,
|
|
"epoch": 3.9924357034795763,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003882549016818905,
|
|
"loss": 5.2117,
|
|
"mean_token_accuracy": 0.16485427170991898,
|
|
"num_tokens": 8749667.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.393984365463257,
|
|
"epoch": 3.9974785678265254,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038791197507090456,
|
|
"loss": 5.2328,
|
|
"mean_token_accuracy": 0.16357189267873765,
|
|
"num_tokens": 8761380.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.37272310256958,
|
|
"epoch": 4.00201714573878,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038756869752689527,
|
|
"loss": 5.0532,
|
|
"mean_token_accuracy": 0.1723750051524904,
|
|
"num_tokens": 8770921.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.317253494262696,
|
|
"epoch": 4.007060010085729,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003872250701167775,
|
|
"loss": 4.8307,
|
|
"mean_token_accuracy": 0.1818503513932228,
|
|
"num_tokens": 8782706.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.330973768234253,
|
|
"epoch": 4.012102874432678,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00038688109390855297,
|
|
"loss": 4.9574,
|
|
"mean_token_accuracy": 0.17943090200424194,
|
|
"num_tokens": 8794600.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.368309211730957,
|
|
"epoch": 4.017145738779627,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003865367699713077,
|
|
"loss": 4.8508,
|
|
"mean_token_accuracy": 0.18577467948198317,
|
|
"num_tokens": 8804149.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.357670831680298,
|
|
"epoch": 4.022188603126576,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00038619209937520836,
|
|
"loss": 4.8582,
|
|
"mean_token_accuracy": 0.1859374850988388,
|
|
"num_tokens": 8814723.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.386159420013428,
|
|
"epoch": 4.027231467473525,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00038584708319149924,
|
|
"loss": 4.9534,
|
|
"mean_token_accuracy": 0.18180705159902572,
|
|
"num_tokens": 8827258.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.339230346679687,
|
|
"epoch": 4.032274331820474,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038550172249249856,
|
|
"loss": 4.9313,
|
|
"mean_token_accuracy": 0.17747099548578263,
|
|
"num_tokens": 8837528.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.032274331820474,
|
|
"eval_entropy": 5.217217032842661,
|
|
"eval_loss": 5.773364067077637,
|
|
"eval_mean_token_accuracy": 0.14623777820484632,
|
|
"eval_num_tokens": 8837528.0,
|
|
"eval_runtime": 2.6324,
|
|
"eval_samples_per_second": 1192.07,
|
|
"eval_steps_per_second": 149.294,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.302639150619507,
|
|
"epoch": 4.0373171961674235,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003851560183515953,
|
|
"loss": 4.9102,
|
|
"mean_token_accuracy": 0.17870035618543625,
|
|
"num_tokens": 8848154.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.325091743469239,
|
|
"epoch": 4.042360060514373,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000384809971843246,
|
|
"loss": 4.8772,
|
|
"mean_token_accuracy": 0.1887033015489578,
|
|
"num_tokens": 8858396.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.397021961212158,
|
|
"epoch": 4.047402924861321,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000384463584042971,
|
|
"loss": 4.8814,
|
|
"mean_token_accuracy": 0.19303392469882966,
|
|
"num_tokens": 8868590.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.295708847045899,
|
|
"epoch": 4.05244578920827,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038411685602735175,
|
|
"loss": 4.9202,
|
|
"mean_token_accuracy": 0.1841244399547577,
|
|
"num_tokens": 8878955.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.294881296157837,
|
|
"epoch": 4.057488653555219,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038376978887402687,
|
|
"loss": 4.8729,
|
|
"mean_token_accuracy": 0.1906580850481987,
|
|
"num_tokens": 8889402.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.316621112823486,
|
|
"epoch": 4.062531517902168,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003834223836616891,
|
|
"loss": 4.8765,
|
|
"mean_token_accuracy": 0.18027233183383942,
|
|
"num_tokens": 8901433.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.257454490661621,
|
|
"epoch": 4.067574382249117,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038307464147008176,
|
|
"loss": 4.9539,
|
|
"mean_token_accuracy": 0.17985324561595917,
|
|
"num_tokens": 8912895.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.35434889793396,
|
|
"epoch": 4.072617246596066,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003827265633799957,
|
|
"loss": 4.9414,
|
|
"mean_token_accuracy": 0.18041590601205826,
|
|
"num_tokens": 8923529.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.367965745925903,
|
|
"epoch": 4.0776601109430155,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003823781504732657,
|
|
"loss": 4.8524,
|
|
"mean_token_accuracy": 0.17984197288751602,
|
|
"num_tokens": 8933279.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.3334118843078615,
|
|
"epoch": 4.082702975289965,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003820294038327669,
|
|
"loss": 4.9474,
|
|
"mean_token_accuracy": 0.18405907303094865,
|
|
"num_tokens": 8944396.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.256776905059814,
|
|
"epoch": 4.087745839636914,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003816803245424121,
|
|
"loss": 4.7486,
|
|
"mean_token_accuracy": 0.1973017930984497,
|
|
"num_tokens": 8955535.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.32894082069397,
|
|
"epoch": 4.092788703983863,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003813309136871478,
|
|
"loss": 4.9287,
|
|
"mean_token_accuracy": 0.17828773707151413,
|
|
"num_tokens": 8965859.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.347507905960083,
|
|
"epoch": 4.097831568330812,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038098117235295083,
|
|
"loss": 4.9744,
|
|
"mean_token_accuracy": 0.1799134135246277,
|
|
"num_tokens": 8976580.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.280542993545533,
|
|
"epoch": 4.102874432677761,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003806311016268254,
|
|
"loss": 4.9268,
|
|
"mean_token_accuracy": 0.18765482157468796,
|
|
"num_tokens": 8988154.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.350623512268067,
|
|
"epoch": 4.10791729702471,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003802807025967995,
|
|
"loss": 4.9526,
|
|
"mean_token_accuracy": 0.17829257845878602,
|
|
"num_tokens": 8999758.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.303531694412231,
|
|
"epoch": 4.112960161371659,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003799299763519213,
|
|
"loss": 4.8998,
|
|
"mean_token_accuracy": 0.1823983371257782,
|
|
"num_tokens": 9010371.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.277239656448364,
|
|
"epoch": 4.118003025718608,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003795789239822561,
|
|
"loss": 4.8944,
|
|
"mean_token_accuracy": 0.1888773575425148,
|
|
"num_tokens": 9021224.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.30113377571106,
|
|
"epoch": 4.1230458900655576,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003792275465788826,
|
|
"loss": 4.9526,
|
|
"mean_token_accuracy": 0.1857559561729431,
|
|
"num_tokens": 9033569.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.352548122406006,
|
|
"epoch": 4.128088754412507,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00037887584523389013,
|
|
"loss": 4.9748,
|
|
"mean_token_accuracy": 0.1786218151450157,
|
|
"num_tokens": 9046166.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.328422498703003,
|
|
"epoch": 4.133131618759456,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00037852382104037454,
|
|
"loss": 4.9512,
|
|
"mean_token_accuracy": 0.1840299740433693,
|
|
"num_tokens": 9056210.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.3232685089111325,
|
|
"epoch": 4.138174483106404,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00037817147509243497,
|
|
"loss": 4.9097,
|
|
"mean_token_accuracy": 0.18457420766353608,
|
|
"num_tokens": 9067473.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.348079013824463,
|
|
"epoch": 4.143217347453353,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003778188084851709,
|
|
"loss": 4.9678,
|
|
"mean_token_accuracy": 0.17711228132247925,
|
|
"num_tokens": 9077744.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.313373136520386,
|
|
"epoch": 4.148260211800302,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00037746582231467816,
|
|
"loss": 4.9438,
|
|
"mean_token_accuracy": 0.17991970032453536,
|
|
"num_tokens": 9088988.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.359608554840088,
|
|
"epoch": 4.153303076147251,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003771125176780461,
|
|
"loss": 4.9832,
|
|
"mean_token_accuracy": 0.17172636240720748,
|
|
"num_tokens": 9100584.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.293414354324341,
|
|
"epoch": 4.1583459404942005,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00037675889567335345,
|
|
"loss": 4.9584,
|
|
"mean_token_accuracy": 0.17977921068668365,
|
|
"num_tokens": 9112442.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.3064381122589115,
|
|
"epoch": 4.16338880484115,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037640495739966573,
|
|
"loss": 5.0095,
|
|
"mean_token_accuracy": 0.17658485025167464,
|
|
"num_tokens": 9122903.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.440914297103882,
|
|
"epoch": 4.168431669188099,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037605070395703124,
|
|
"loss": 5.0105,
|
|
"mean_token_accuracy": 0.17761027663946152,
|
|
"num_tokens": 9134301.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.450673341751099,
|
|
"epoch": 4.173474533535048,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037569613644647783,
|
|
"loss": 5.0126,
|
|
"mean_token_accuracy": 0.1710635393857956,
|
|
"num_tokens": 9144503.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.24276442527771,
|
|
"epoch": 4.178517397881997,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003753412559700094,
|
|
"loss": 4.8448,
|
|
"mean_token_accuracy": 0.18016259223222733,
|
|
"num_tokens": 9154173.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.349482870101928,
|
|
"epoch": 4.183560262228946,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037498606363060285,
|
|
"loss": 4.9787,
|
|
"mean_token_accuracy": 0.1813860848546028,
|
|
"num_tokens": 9163772.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.388900852203369,
|
|
"epoch": 4.188603126575895,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003746305605322041,
|
|
"loss": 4.9863,
|
|
"mean_token_accuracy": 0.1794368267059326,
|
|
"num_tokens": 9174618.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.402277565002441,
|
|
"epoch": 4.193645990922844,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037427474777972477,
|
|
"loss": 5.048,
|
|
"mean_token_accuracy": 0.17857434451580048,
|
|
"num_tokens": 9185199.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.332354402542114,
|
|
"epoch": 4.198688855269793,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00037391862647903943,
|
|
"loss": 4.9478,
|
|
"mean_token_accuracy": 0.17822057902812957,
|
|
"num_tokens": 9197329.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.310865211486816,
|
|
"epoch": 4.2037317196167425,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037356219773698106,
|
|
"loss": 4.9259,
|
|
"mean_token_accuracy": 0.1809712678194046,
|
|
"num_tokens": 9207516.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.304825162887573,
|
|
"epoch": 4.208774583963692,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037320546266133854,
|
|
"loss": 5.0068,
|
|
"mean_token_accuracy": 0.16734636276960374,
|
|
"num_tokens": 9219598.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.269561529159546,
|
|
"epoch": 4.213817448310641,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003728484223608526,
|
|
"loss": 4.8822,
|
|
"mean_token_accuracy": 0.18481125682592392,
|
|
"num_tokens": 9229675.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.275122451782226,
|
|
"epoch": 4.21886031265759,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00037249107794521286,
|
|
"loss": 4.9289,
|
|
"mean_token_accuracy": 0.1843302696943283,
|
|
"num_tokens": 9240308.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.261275959014893,
|
|
"epoch": 4.223903177004539,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037213343052505394,
|
|
"loss": 4.9374,
|
|
"mean_token_accuracy": 0.17706461250782013,
|
|
"num_tokens": 9250299.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.343583965301514,
|
|
"epoch": 4.228946041351487,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003717754812119523,
|
|
"loss": 4.9709,
|
|
"mean_token_accuracy": 0.17234741449356078,
|
|
"num_tokens": 9262659.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.360939073562622,
|
|
"epoch": 4.233988905698436,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003714172311184226,
|
|
"loss": 4.9197,
|
|
"mean_token_accuracy": 0.18292128443717956,
|
|
"num_tokens": 9274430.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.269550418853759,
|
|
"epoch": 4.2390317700453854,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003710586813579144,
|
|
"loss": 5.027,
|
|
"mean_token_accuracy": 0.17529074251651763,
|
|
"num_tokens": 9285484.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.323139095306397,
|
|
"epoch": 4.244074634392335,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037069983304480875,
|
|
"loss": 4.925,
|
|
"mean_token_accuracy": 0.18182601928710937,
|
|
"num_tokens": 9296253.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.331530332565308,
|
|
"epoch": 4.249117498739284,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003703406872944144,
|
|
"loss": 4.9525,
|
|
"mean_token_accuracy": 0.1798201635479927,
|
|
"num_tokens": 9308491.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.349999618530274,
|
|
"epoch": 4.254160363086233,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003699812452229645,
|
|
"loss": 4.9379,
|
|
"mean_token_accuracy": 0.17934981286525725,
|
|
"num_tokens": 9319722.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.26986517906189,
|
|
"epoch": 4.259203227433182,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036962150794761354,
|
|
"loss": 4.8781,
|
|
"mean_token_accuracy": 0.18720467388629913,
|
|
"num_tokens": 9329324.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.307141971588135,
|
|
"epoch": 4.264246091780131,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003692614765864331,
|
|
"loss": 4.961,
|
|
"mean_token_accuracy": 0.17951072305440902,
|
|
"num_tokens": 9340060.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.299160003662109,
|
|
"epoch": 4.26928895612708,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00036890115225840904,
|
|
"loss": 4.9599,
|
|
"mean_token_accuracy": 0.18478352576494217,
|
|
"num_tokens": 9351071.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.359957361221314,
|
|
"epoch": 4.274331820474029,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003685405360834376,
|
|
"loss": 5.0271,
|
|
"mean_token_accuracy": 0.17086062133312224,
|
|
"num_tokens": 9361910.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.3686933517456055,
|
|
"epoch": 4.279374684820978,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036817962918232234,
|
|
"loss": 4.944,
|
|
"mean_token_accuracy": 0.17759687155485154,
|
|
"num_tokens": 9371964.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.32645320892334,
|
|
"epoch": 4.2844175491679275,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036781843267677,
|
|
"loss": 4.9251,
|
|
"mean_token_accuracy": 0.18574418723583222,
|
|
"num_tokens": 9381922.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.304310655593872,
|
|
"epoch": 4.289460413514877,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003674569476893878,
|
|
"loss": 4.8589,
|
|
"mean_token_accuracy": 0.1915180817246437,
|
|
"num_tokens": 9393256.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.258758497238159,
|
|
"epoch": 4.294503277861826,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003670951753436794,
|
|
"loss": 4.9242,
|
|
"mean_token_accuracy": 0.1819936826825142,
|
|
"num_tokens": 9403703.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.319845676422119,
|
|
"epoch": 4.299546142208775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00036673311676404163,
|
|
"loss": 5.0105,
|
|
"mean_token_accuracy": 0.1778859853744507,
|
|
"num_tokens": 9414454.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.234507894515991,
|
|
"epoch": 4.304589006555724,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003663707730757608,
|
|
"loss": 4.907,
|
|
"mean_token_accuracy": 0.1859314978122711,
|
|
"num_tokens": 9425566.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.28688759803772,
|
|
"epoch": 4.309631870902673,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036600814540500956,
|
|
"loss": 5.0426,
|
|
"mean_token_accuracy": 0.17424253821372987,
|
|
"num_tokens": 9435555.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.4263842582702635,
|
|
"epoch": 4.314674735249622,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000365645234878843,
|
|
"loss": 5.0613,
|
|
"mean_token_accuracy": 0.16906643956899642,
|
|
"num_tokens": 9447784.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.374078559875488,
|
|
"epoch": 4.31971759959657,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00036528204262519545,
|
|
"loss": 4.9555,
|
|
"mean_token_accuracy": 0.17729347795248032,
|
|
"num_tokens": 9458243.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.283599996566773,
|
|
"epoch": 4.3247604639435195,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003649185697728768,
|
|
"loss": 4.9361,
|
|
"mean_token_accuracy": 0.17935186624526978,
|
|
"num_tokens": 9470258.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.331258153915405,
|
|
"epoch": 4.329803328290469,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00036455481745156906,
|
|
"loss": 4.9483,
|
|
"mean_token_accuracy": 0.18029385358095168,
|
|
"num_tokens": 9480787.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.3175701141357425,
|
|
"epoch": 4.334846192637418,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003641907867918228,
|
|
"loss": 4.9344,
|
|
"mean_token_accuracy": 0.18013985604047775,
|
|
"num_tokens": 9492459.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.360218048095703,
|
|
"epoch": 4.339889056984367,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003638264789250538,
|
|
"loss": 5.0195,
|
|
"mean_token_accuracy": 0.17920296341180803,
|
|
"num_tokens": 9504600.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.371557807922363,
|
|
"epoch": 4.344931921331316,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036346189498353914,
|
|
"loss": 5.0188,
|
|
"mean_token_accuracy": 0.17976579815149307,
|
|
"num_tokens": 9515853.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.282398319244384,
|
|
"epoch": 4.349974785678265,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003630970361004143,
|
|
"loss": 4.9976,
|
|
"mean_token_accuracy": 0.18170807212591172,
|
|
"num_tokens": 9528109.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.3153290271759035,
|
|
"epoch": 4.355017650025214,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003627319034096691,
|
|
"loss": 4.9563,
|
|
"mean_token_accuracy": 0.17810409069061278,
|
|
"num_tokens": 9539795.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.282315683364868,
|
|
"epoch": 4.360060514372163,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003623664980461443,
|
|
"loss": 4.9874,
|
|
"mean_token_accuracy": 0.17548182904720305,
|
|
"num_tokens": 9551020.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.33734974861145,
|
|
"epoch": 4.3651033787191125,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036200082114552806,
|
|
"loss": 4.9827,
|
|
"mean_token_accuracy": 0.17971308380365372,
|
|
"num_tokens": 9562540.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.358851337432862,
|
|
"epoch": 4.370146243066062,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00036163487384435294,
|
|
"loss": 5.0138,
|
|
"mean_token_accuracy": 0.1736428275704384,
|
|
"num_tokens": 9574122.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.303806972503662,
|
|
"epoch": 4.375189107413011,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003612686572799914,
|
|
"loss": 4.9847,
|
|
"mean_token_accuracy": 0.1820244237780571,
|
|
"num_tokens": 9584781.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.376692438125611,
|
|
"epoch": 4.38023197175996,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003609021725906529,
|
|
"loss": 5.0007,
|
|
"mean_token_accuracy": 0.1797618880867958,
|
|
"num_tokens": 9595230.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.362706279754638,
|
|
"epoch": 4.385274836106909,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036053542091538045,
|
|
"loss": 4.9297,
|
|
"mean_token_accuracy": 0.1835813507437706,
|
|
"num_tokens": 9605814.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.294155502319336,
|
|
"epoch": 4.390317700453858,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00036016840339404664,
|
|
"loss": 4.94,
|
|
"mean_token_accuracy": 0.18150079548358916,
|
|
"num_tokens": 9617900.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.296058654785156,
|
|
"epoch": 4.395360564800807,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003598011211673505,
|
|
"loss": 4.9651,
|
|
"mean_token_accuracy": 0.18516950458288192,
|
|
"num_tokens": 9628987.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.2897748947143555,
|
|
"epoch": 4.400403429147756,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003594335753768135,
|
|
"loss": 4.9338,
|
|
"mean_token_accuracy": 0.1812305122613907,
|
|
"num_tokens": 9638419.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.294053840637207,
|
|
"epoch": 4.405446293494705,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003590657671647766,
|
|
"loss": 4.949,
|
|
"mean_token_accuracy": 0.18147448301315308,
|
|
"num_tokens": 9648975.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.280570459365845,
|
|
"epoch": 4.4104891578416545,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035869769767439606,
|
|
"loss": 4.9113,
|
|
"mean_token_accuracy": 0.1840442806482315,
|
|
"num_tokens": 9659935.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.367076683044433,
|
|
"epoch": 4.415532022188603,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003583293680496404,
|
|
"loss": 5.043,
|
|
"mean_token_accuracy": 0.17699178606271743,
|
|
"num_tokens": 9671269.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.298137378692627,
|
|
"epoch": 4.420574886535552,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003579607794352867,
|
|
"loss": 4.9588,
|
|
"mean_token_accuracy": 0.18067768216133118,
|
|
"num_tokens": 9682627.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.260059213638305,
|
|
"epoch": 4.425617750882501,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035759193297691674,
|
|
"loss": 4.8807,
|
|
"mean_token_accuracy": 0.183808770775795,
|
|
"num_tokens": 9693482.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.305668020248413,
|
|
"epoch": 4.43066061522945,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00035722282982091373,
|
|
"loss": 5.0006,
|
|
"mean_token_accuracy": 0.18011518120765685,
|
|
"num_tokens": 9705279.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.290925550460815,
|
|
"epoch": 4.435703479576399,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003568534711144591,
|
|
"loss": 4.9715,
|
|
"mean_token_accuracy": 0.17928884625434877,
|
|
"num_tokens": 9716695.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.252958965301514,
|
|
"epoch": 4.440746343923348,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000356483858005528,
|
|
"loss": 4.9176,
|
|
"mean_token_accuracy": 0.18401297628879548,
|
|
"num_tokens": 9727960.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.281794738769531,
|
|
"epoch": 4.445789208270297,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00035611399164288654,
|
|
"loss": 4.9974,
|
|
"mean_token_accuracy": 0.17457676231861113,
|
|
"num_tokens": 9738341.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.289261817932129,
|
|
"epoch": 4.450832072617247,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003557438731760879,
|
|
"loss": 4.9885,
|
|
"mean_token_accuracy": 0.18040503263473512,
|
|
"num_tokens": 9748261.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.39917140007019,
|
|
"epoch": 4.455874936964196,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003553735037554689,
|
|
"loss": 4.9885,
|
|
"mean_token_accuracy": 0.1773747369647026,
|
|
"num_tokens": 9759109.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.313023090362549,
|
|
"epoch": 4.460917801311145,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003550028845321461,
|
|
"loss": 4.9532,
|
|
"mean_token_accuracy": 0.18129721879959107,
|
|
"num_tokens": 9770431.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.263807106018066,
|
|
"epoch": 4.465960665658094,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003546320166580127,
|
|
"loss": 5.0161,
|
|
"mean_token_accuracy": 0.1718140348792076,
|
|
"num_tokens": 9783459.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.321491289138794,
|
|
"epoch": 4.471003530005043,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00035426090128573464,
|
|
"loss": 4.9267,
|
|
"mean_token_accuracy": 0.18388036787509918,
|
|
"num_tokens": 9793961.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.37356743812561,
|
|
"epoch": 4.476046394351992,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035388953956874686,
|
|
"loss": 5.0231,
|
|
"mean_token_accuracy": 0.17222304940223693,
|
|
"num_tokens": 9803964.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.278263854980469,
|
|
"epoch": 4.481089258698941,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003535179326612504,
|
|
"loss": 4.9285,
|
|
"mean_token_accuracy": 0.18352929949760438,
|
|
"num_tokens": 9816176.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.199662351608277,
|
|
"epoch": 4.48613212304589,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035314608171820787,
|
|
"loss": 4.899,
|
|
"mean_token_accuracy": 0.19029782712459564,
|
|
"num_tokens": 9826649.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.324962520599366,
|
|
"epoch": 4.4911749873928395,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00035277398789534075,
|
|
"loss": 5.0416,
|
|
"mean_token_accuracy": 0.1732318162918091,
|
|
"num_tokens": 9838396.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.352452087402344,
|
|
"epoch": 4.496217851739789,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003524016523491252,
|
|
"loss": 4.9935,
|
|
"mean_token_accuracy": 0.17702273577451705,
|
|
"num_tokens": 9848863.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.222486400604248,
|
|
"epoch": 4.501260716086737,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00035202907623678865,
|
|
"loss": 4.924,
|
|
"mean_token_accuracy": 0.17421945929527283,
|
|
"num_tokens": 9860673.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.32892074584961,
|
|
"epoch": 4.506303580433686,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003516562607163062,
|
|
"loss": 4.9834,
|
|
"mean_token_accuracy": 0.17627667039632797,
|
|
"num_tokens": 9871246.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.34592719078064,
|
|
"epoch": 4.511346444780635,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003512832069463971,
|
|
"loss": 5.0363,
|
|
"mean_token_accuracy": 0.17620474845170975,
|
|
"num_tokens": 9883352.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.293773651123047,
|
|
"epoch": 4.516389309127584,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035090991608652125,
|
|
"loss": 5.0215,
|
|
"mean_token_accuracy": 0.17724793404340744,
|
|
"num_tokens": 9894797.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.323146486282349,
|
|
"epoch": 4.521432173474533,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00035053638929687504,
|
|
"loss": 4.9641,
|
|
"mean_token_accuracy": 0.1770881101489067,
|
|
"num_tokens": 9905618.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.261654233932495,
|
|
"epoch": 4.526475037821482,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003501626277383884,
|
|
"loss": 4.9761,
|
|
"mean_token_accuracy": 0.17481629103422164,
|
|
"num_tokens": 9916817.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.300576257705688,
|
|
"epoch": 4.5315179021684315,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000349788632572721,
|
|
"loss": 4.9395,
|
|
"mean_token_accuracy": 0.17969692200422288,
|
|
"num_tokens": 9929052.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.273266744613648,
|
|
"epoch": 4.536560766515381,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003494144049622582,
|
|
"loss": 4.8594,
|
|
"mean_token_accuracy": 0.17967883199453355,
|
|
"num_tokens": 9939513.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.536560766515381,
|
|
"eval_entropy": 5.206290680640223,
|
|
"eval_loss": 5.754478931427002,
|
|
"eval_mean_token_accuracy": 0.14864040587012095,
|
|
"eval_num_tokens": 9939513.0,
|
|
"eval_runtime": 2.6511,
|
|
"eval_samples_per_second": 1183.651,
|
|
"eval_steps_per_second": 148.239,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.20522141456604,
|
|
"epoch": 4.54160363086233,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003490399460701083,
|
|
"loss": 4.9277,
|
|
"mean_token_accuracy": 0.18226718306541442,
|
|
"num_tokens": 9950149.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.374950313568116,
|
|
"epoch": 4.546646495209279,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000348665257060098,
|
|
"loss": 4.9608,
|
|
"mean_token_accuracy": 0.18251149952411652,
|
|
"num_tokens": 9961696.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.325359582901001,
|
|
"epoch": 4.551689359556228,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003482903390967695,
|
|
"loss": 4.9811,
|
|
"mean_token_accuracy": 0.17832767069339753,
|
|
"num_tokens": 9971446.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.298169708251953,
|
|
"epoch": 4.556732223903177,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00034791519334537635,
|
|
"loss": 4.9176,
|
|
"mean_token_accuracy": 0.19268935173749924,
|
|
"num_tokens": 9982992.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.215664672851562,
|
|
"epoch": 4.561775088250126,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003475398209718804,
|
|
"loss": 4.87,
|
|
"mean_token_accuracy": 0.18471892774105073,
|
|
"num_tokens": 9993443.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.267688751220703,
|
|
"epoch": 4.566817952597075,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003471642231429475,
|
|
"loss": 5.0112,
|
|
"mean_token_accuracy": 0.18080305755138398,
|
|
"num_tokens": 10004531.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.319534063339233,
|
|
"epoch": 4.5718608169440245,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003467884010259443,
|
|
"loss": 4.9507,
|
|
"mean_token_accuracy": 0.18149352967739105,
|
|
"num_tokens": 10016455.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.323209285736084,
|
|
"epoch": 4.576903681290974,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00034641235578893476,
|
|
"loss": 4.8672,
|
|
"mean_token_accuracy": 0.19021911472082137,
|
|
"num_tokens": 10028067.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.303041315078735,
|
|
"epoch": 4.581946545637923,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003460360886006761,
|
|
"loss": 5.0776,
|
|
"mean_token_accuracy": 0.1727398544549942,
|
|
"num_tokens": 10038615.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.295140409469605,
|
|
"epoch": 4.586989409984872,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034565960063061534,
|
|
"loss": 5.017,
|
|
"mean_token_accuracy": 0.17707218676805497,
|
|
"num_tokens": 10050022.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.272798013687134,
|
|
"epoch": 4.592032274331821,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003452828930488859,
|
|
"loss": 4.9468,
|
|
"mean_token_accuracy": 0.18265050947666167,
|
|
"num_tokens": 10062565.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.363107681274414,
|
|
"epoch": 4.597075138678769,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034490596702630355,
|
|
"loss": 5.0854,
|
|
"mean_token_accuracy": 0.17105167210102082,
|
|
"num_tokens": 10075553.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.341572523117065,
|
|
"epoch": 4.602118003025718,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003445288237343632,
|
|
"loss": 5.0046,
|
|
"mean_token_accuracy": 0.17526729702949523,
|
|
"num_tokens": 10086144.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.248961353302002,
|
|
"epoch": 4.607160867372667,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034415146434523463,
|
|
"loss": 4.9939,
|
|
"mean_token_accuracy": 0.17357587963342666,
|
|
"num_tokens": 10097218.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.2435548305511475,
|
|
"epoch": 4.6122037317196165,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00034377389003175984,
|
|
"loss": 4.9758,
|
|
"mean_token_accuracy": 0.17142492085695266,
|
|
"num_tokens": 10107614.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.404813289642334,
|
|
"epoch": 4.617246596066566,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003433961019674483,
|
|
"loss": 5.0272,
|
|
"mean_token_accuracy": 0.1724199503660202,
|
|
"num_tokens": 10118976.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.330430746078491,
|
|
"epoch": 4.622289460413515,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003430181013264741,
|
|
"loss": 4.9482,
|
|
"mean_token_accuracy": 0.1810013860464096,
|
|
"num_tokens": 10129914.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.343965864181518,
|
|
"epoch": 4.627332324760464,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003426398892836719,
|
|
"loss": 4.9819,
|
|
"mean_token_accuracy": 0.17649532109498978,
|
|
"num_tokens": 10139490.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.3509180545806885,
|
|
"epoch": 4.632375189107413,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003422614670145336,
|
|
"loss": 5.0327,
|
|
"mean_token_accuracy": 0.1800553783774376,
|
|
"num_tokens": 10149874.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.3082324981689455,
|
|
"epoch": 4.637418053454362,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003418828356952041,
|
|
"loss": 5.0389,
|
|
"mean_token_accuracy": 0.1774437442421913,
|
|
"num_tokens": 10161373.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.267008113861084,
|
|
"epoch": 4.642460917801311,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034150399650247846,
|
|
"loss": 4.9427,
|
|
"mean_token_accuracy": 0.18608315885066987,
|
|
"num_tokens": 10172044.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.326416635513306,
|
|
"epoch": 4.64750378214826,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00034112495061379764,
|
|
"loss": 4.9901,
|
|
"mean_token_accuracy": 0.18243445008993148,
|
|
"num_tokens": 10183127.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.319555139541626,
|
|
"epoch": 4.652546646495209,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00034074569920724483,
|
|
"loss": 5.032,
|
|
"mean_token_accuracy": 0.17033783942461014,
|
|
"num_tokens": 10194366.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.306932878494263,
|
|
"epoch": 4.6575895108421586,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003403662434615423,
|
|
"loss": 4.9906,
|
|
"mean_token_accuracy": 0.17832741439342498,
|
|
"num_tokens": 10205655.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.307049179077149,
|
|
"epoch": 4.662632375189108,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033998658455604736,
|
|
"loss": 4.9699,
|
|
"mean_token_accuracy": 0.18948812186717987,
|
|
"num_tokens": 10215596.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.269381475448609,
|
|
"epoch": 4.667675239536057,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033960672367074827,
|
|
"loss": 4.9689,
|
|
"mean_token_accuracy": 0.18027861565351486,
|
|
"num_tokens": 10225298.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.413548707962036,
|
|
"epoch": 4.672718103883006,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003392266619862619,
|
|
"loss": 5.0784,
|
|
"mean_token_accuracy": 0.17301628440618516,
|
|
"num_tokens": 10236979.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.28053503036499,
|
|
"epoch": 4.677760968229954,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003388464006838284,
|
|
"loss": 4.9326,
|
|
"mean_token_accuracy": 0.1808907210826874,
|
|
"num_tokens": 10248109.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.265711498260498,
|
|
"epoch": 4.682803832576903,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033846594094530874,
|
|
"loss": 5.0134,
|
|
"mean_token_accuracy": 0.1820990562438965,
|
|
"num_tokens": 10258931.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.306347942352295,
|
|
"epoch": 4.687846696923852,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003380852839531806,
|
|
"loss": 5.0327,
|
|
"mean_token_accuracy": 0.18318586498498918,
|
|
"num_tokens": 10272521.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.285265970230102,
|
|
"epoch": 4.6928895612708015,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033770443089053476,
|
|
"loss": 4.9473,
|
|
"mean_token_accuracy": 0.18003821223974228,
|
|
"num_tokens": 10283481.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.279108762741089,
|
|
"epoch": 4.697932425617751,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033732338294107116,
|
|
"loss": 5.0107,
|
|
"mean_token_accuracy": 0.1746044009923935,
|
|
"num_tokens": 10294258.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.372565841674804,
|
|
"epoch": 4.7029752899647,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033694214128909567,
|
|
"loss": 4.9664,
|
|
"mean_token_accuracy": 0.183512319624424,
|
|
"num_tokens": 10304464.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.317605018615723,
|
|
"epoch": 4.708018154311649,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033656070711951633,
|
|
"loss": 4.9878,
|
|
"mean_token_accuracy": 0.17851071506738664,
|
|
"num_tokens": 10316343.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.313731288909912,
|
|
"epoch": 4.713061018658598,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003361790816178391,
|
|
"loss": 5.0319,
|
|
"mean_token_accuracy": 0.17628833204507827,
|
|
"num_tokens": 10327057.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.370765495300293,
|
|
"epoch": 4.718103883005547,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003357972659701649,
|
|
"loss": 5.0287,
|
|
"mean_token_accuracy": 0.18172771632671356,
|
|
"num_tokens": 10338483.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.300267887115479,
|
|
"epoch": 4.723146747352496,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00033541526136318564,
|
|
"loss": 4.9546,
|
|
"mean_token_accuracy": 0.18281051963567735,
|
|
"num_tokens": 10349522.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.263338661193847,
|
|
"epoch": 4.728189611699445,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003350330689841804,
|
|
"loss": 4.9431,
|
|
"mean_token_accuracy": 0.17673654705286027,
|
|
"num_tokens": 10361036.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.336398124694824,
|
|
"epoch": 4.733232476046394,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003346506900210118,
|
|
"loss": 5.0677,
|
|
"mean_token_accuracy": 0.17829442769289017,
|
|
"num_tokens": 10372673.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.335520887374878,
|
|
"epoch": 4.7382753403933435,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003342681256621227,
|
|
"loss": 5.0906,
|
|
"mean_token_accuracy": 0.17316668182611467,
|
|
"num_tokens": 10384100.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.366275787353516,
|
|
"epoch": 4.743318204740293,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003338853770965319,
|
|
"loss": 4.9945,
|
|
"mean_token_accuracy": 0.17589996606111527,
|
|
"num_tokens": 10395503.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.30719027519226,
|
|
"epoch": 4.748361069087242,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033350244551383064,
|
|
"loss": 4.948,
|
|
"mean_token_accuracy": 0.18412147611379623,
|
|
"num_tokens": 10406147.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.256329154968261,
|
|
"epoch": 4.753403933434191,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003331193321041793,
|
|
"loss": 4.8812,
|
|
"mean_token_accuracy": 0.18300259709358216,
|
|
"num_tokens": 10416659.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.306020593643188,
|
|
"epoch": 4.75844679778114,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00033273603805830296,
|
|
"loss": 4.9937,
|
|
"mean_token_accuracy": 0.17472372949123383,
|
|
"num_tokens": 10427765.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.3129644870758055,
|
|
"epoch": 4.763489662128089,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033235256456748847,
|
|
"loss": 4.9401,
|
|
"mean_token_accuracy": 0.18560262471437455,
|
|
"num_tokens": 10439376.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.240534543991089,
|
|
"epoch": 4.768532526475038,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003319689128235804,
|
|
"loss": 5.0057,
|
|
"mean_token_accuracy": 0.17537233233451843,
|
|
"num_tokens": 10449232.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.240880107879638,
|
|
"epoch": 4.773575390821987,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033158508401897707,
|
|
"loss": 4.9413,
|
|
"mean_token_accuracy": 0.17986069321632386,
|
|
"num_tokens": 10459258.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.329692363739014,
|
|
"epoch": 4.778618255168936,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003312010793466273,
|
|
"loss": 5.0051,
|
|
"mean_token_accuracy": 0.1768033042550087,
|
|
"num_tokens": 10469909.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.3311395168304445,
|
|
"epoch": 4.783661119515885,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003308169000000265,
|
|
"loss": 4.9545,
|
|
"mean_token_accuracy": 0.18208848387002946,
|
|
"num_tokens": 10479848.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.31560263633728,
|
|
"epoch": 4.788703983862834,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003304325471732129,
|
|
"loss": 4.9974,
|
|
"mean_token_accuracy": 0.17569656074047088,
|
|
"num_tokens": 10490352.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.25335464477539,
|
|
"epoch": 4.793746848209783,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033004802206076396,
|
|
"loss": 4.9158,
|
|
"mean_token_accuracy": 0.18476161509752273,
|
|
"num_tokens": 10501905.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.299323225021363,
|
|
"epoch": 4.798789712556732,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00032966332585779263,
|
|
"loss": 5.0504,
|
|
"mean_token_accuracy": 0.17709101885557174,
|
|
"num_tokens": 10512491.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.266164302825928,
|
|
"epoch": 4.803832576903681,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00032927845975994356,
|
|
"loss": 4.859,
|
|
"mean_token_accuracy": 0.19434284269809723,
|
|
"num_tokens": 10522686.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.245361042022705,
|
|
"epoch": 4.80887544125063,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00032889342496338953,
|
|
"loss": 4.9296,
|
|
"mean_token_accuracy": 0.17708287984132767,
|
|
"num_tokens": 10534240.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.321494150161743,
|
|
"epoch": 4.813918305597579,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003285082226648273,
|
|
"loss": 5.0546,
|
|
"mean_token_accuracy": 0.17531771659851075,
|
|
"num_tokens": 10545096.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.37279782295227,
|
|
"epoch": 4.8189611699445285,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003281228540614749,
|
|
"loss": 5.0453,
|
|
"mean_token_accuracy": 0.17625311613082886,
|
|
"num_tokens": 10557076.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.210517740249633,
|
|
"epoch": 4.824004034291478,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032773732035106657,
|
|
"loss": 4.8535,
|
|
"mean_token_accuracy": 0.18497144132852555,
|
|
"num_tokens": 10567635.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.34490385055542,
|
|
"epoch": 4.829046898638427,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00032735162273185006,
|
|
"loss": 5.0762,
|
|
"mean_token_accuracy": 0.1669636458158493,
|
|
"num_tokens": 10579488.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.280808830261231,
|
|
"epoch": 4.834089762985376,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003269657624025824,
|
|
"loss": 4.9335,
|
|
"mean_token_accuracy": 0.17956365048885345,
|
|
"num_tokens": 10591092.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.253216791152954,
|
|
"epoch": 4.839132627332325,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003265797405625265,
|
|
"loss": 5.0147,
|
|
"mean_token_accuracy": 0.17421629577875136,
|
|
"num_tokens": 10603165.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.2426494598388675,
|
|
"epoch": 4.844175491679274,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003261935584114471,
|
|
"loss": 4.8896,
|
|
"mean_token_accuracy": 0.19238029420375824,
|
|
"num_tokens": 10615149.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.3818216800689695,
|
|
"epoch": 4.849218356026223,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003258072171496072,
|
|
"loss": 5.0542,
|
|
"mean_token_accuracy": 0.17845158725976945,
|
|
"num_tokens": 10626786.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.326459217071533,
|
|
"epoch": 4.854261220373172,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003254207179777644,
|
|
"loss": 4.9514,
|
|
"mean_token_accuracy": 0.18064935356378556,
|
|
"num_tokens": 10637107.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.245392990112305,
|
|
"epoch": 4.8593040847201205,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000325034062097167,
|
|
"loss": 4.9546,
|
|
"mean_token_accuracy": 0.18254856169223785,
|
|
"num_tokens": 10648198.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.376431655883789,
|
|
"epoch": 4.86434694906707,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00032464725070955046,
|
|
"loss": 5.1005,
|
|
"mean_token_accuracy": 0.17046169936656952,
|
|
"num_tokens": 10659867.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.357150506973267,
|
|
"epoch": 4.869389813414019,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00032426028501713347,
|
|
"loss": 5.0168,
|
|
"mean_token_accuracy": 0.17657135277986527,
|
|
"num_tokens": 10672117.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.269250965118408,
|
|
"epoch": 4.874432677760968,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032387316622261425,
|
|
"loss": 4.9233,
|
|
"mean_token_accuracy": 0.1793333947658539,
|
|
"num_tokens": 10682014.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.320382833480835,
|
|
"epoch": 4.879475542107917,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032348589552916703,
|
|
"loss": 5.0729,
|
|
"mean_token_accuracy": 0.16703013181686402,
|
|
"num_tokens": 10693330.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.3258302211761475,
|
|
"epoch": 4.884518406454866,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003230984741404381,
|
|
"loss": 4.9345,
|
|
"mean_token_accuracy": 0.18443776965141295,
|
|
"num_tokens": 10703772.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.317282438278198,
|
|
"epoch": 4.889561270801815,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003227109032605419,
|
|
"loss": 5.0072,
|
|
"mean_token_accuracy": 0.1728968784213066,
|
|
"num_tokens": 10714088.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.178986072540283,
|
|
"epoch": 4.894604135148764,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003223231840940578,
|
|
"loss": 4.8194,
|
|
"mean_token_accuracy": 0.18739395141601561,
|
|
"num_tokens": 10724701.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.27409930229187,
|
|
"epoch": 4.8996469994957135,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032193531784602595,
|
|
"loss": 4.9416,
|
|
"mean_token_accuracy": 0.18390117585659027,
|
|
"num_tokens": 10735361.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.33750433921814,
|
|
"epoch": 4.904689863842663,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032154730572194346,
|
|
"loss": 4.9629,
|
|
"mean_token_accuracy": 0.17565561532974244,
|
|
"num_tokens": 10745909.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.261441802978515,
|
|
"epoch": 4.909732728189612,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000321159148927761,
|
|
"loss": 5.0851,
|
|
"mean_token_accuracy": 0.17518814653158188,
|
|
"num_tokens": 10759305.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.280098009109497,
|
|
"epoch": 4.914775592536561,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000320770848669879,
|
|
"loss": 5.0249,
|
|
"mean_token_accuracy": 0.1832031324505806,
|
|
"num_tokens": 10770451.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.275878000259399,
|
|
"epoch": 4.91981845688351,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00032038240615514345,
|
|
"loss": 4.9485,
|
|
"mean_token_accuracy": 0.1822526454925537,
|
|
"num_tokens": 10782376.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.270547008514404,
|
|
"epoch": 4.924861321230459,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003199938225908426,
|
|
"loss": 5.0695,
|
|
"mean_token_accuracy": 0.1779208466410637,
|
|
"num_tokens": 10794252.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.356183004379273,
|
|
"epoch": 4.929904185577408,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003196050991847033,
|
|
"loss": 5.019,
|
|
"mean_token_accuracy": 0.17516678869724273,
|
|
"num_tokens": 10804117.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.299526643753052,
|
|
"epoch": 4.934947049924357,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003192162371448868,
|
|
"loss": 4.9493,
|
|
"mean_token_accuracy": 0.18124551326036453,
|
|
"num_tokens": 10815798.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.229008340835572,
|
|
"epoch": 4.939989914271306,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00031882723767998517,
|
|
"loss": 4.9736,
|
|
"mean_token_accuracy": 0.18518155068159103,
|
|
"num_tokens": 10827135.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.322872686386108,
|
|
"epoch": 4.9450327786182555,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00031843810199901795,
|
|
"loss": 5.0654,
|
|
"mean_token_accuracy": 0.1708437204360962,
|
|
"num_tokens": 10838988.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.283534145355224,
|
|
"epoch": 4.950075642965205,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003180488313114277,
|
|
"loss": 4.8944,
|
|
"mean_token_accuracy": 0.1870134472846985,
|
|
"num_tokens": 10848223.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.2628453254699705,
|
|
"epoch": 4.955118507312154,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003176594268270766,
|
|
"loss": 4.9292,
|
|
"mean_token_accuracy": 0.18034395277500154,
|
|
"num_tokens": 10859052.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.289291715621948,
|
|
"epoch": 4.960161371659103,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00031726988975624293,
|
|
"loss": 4.9833,
|
|
"mean_token_accuracy": 0.18065698444843292,
|
|
"num_tokens": 10869108.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.334167766571045,
|
|
"epoch": 4.965204236006051,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00031688022130961675,
|
|
"loss": 4.9871,
|
|
"mean_token_accuracy": 0.17840698063373567,
|
|
"num_tokens": 10880691.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.2007341384887695,
|
|
"epoch": 4.970247100353,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003164904226982967,
|
|
"loss": 4.8959,
|
|
"mean_token_accuracy": 0.18399753719568251,
|
|
"num_tokens": 10892206.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.275009441375732,
|
|
"epoch": 4.975289964699949,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031610049513378574,
|
|
"loss": 4.868,
|
|
"mean_token_accuracy": 0.19645134806632997,
|
|
"num_tokens": 10902130.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.324296379089356,
|
|
"epoch": 4.9803328290468984,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00031571043982798786,
|
|
"loss": 4.9984,
|
|
"mean_token_accuracy": 0.17903662323951722,
|
|
"num_tokens": 10913424.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.271382093429565,
|
|
"epoch": 4.985375693393848,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003153202579932039,
|
|
"loss": 5.0679,
|
|
"mean_token_accuracy": 0.17533862590789795,
|
|
"num_tokens": 10925718.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.272064256668091,
|
|
"epoch": 4.990418557740797,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031492995084212794,
|
|
"loss": 4.9363,
|
|
"mean_token_accuracy": 0.18015358597040176,
|
|
"num_tokens": 10937102.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.267334270477295,
|
|
"epoch": 4.995461422087746,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003145395195878437,
|
|
"loss": 4.9358,
|
|
"mean_token_accuracy": 0.18337683081626893,
|
|
"num_tokens": 10948679.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.2274753782484265,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00031414896544382047,
|
|
"loss": 4.8808,
|
|
"mean_token_accuracy": 0.1853246142466863,
|
|
"num_tokens": 10958200.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.272035741806031,
|
|
"epoch": 5.005042864346949,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003137582896239095,
|
|
"loss": 4.6656,
|
|
"mean_token_accuracy": 0.20294922441244126,
|
|
"num_tokens": 10969775.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.2435506820678714,
|
|
"epoch": 5.010085728693898,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003133674933423404,
|
|
"loss": 4.6665,
|
|
"mean_token_accuracy": 0.19544582217931747,
|
|
"num_tokens": 10980423.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.215020036697387,
|
|
"epoch": 5.015128593040847,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00031297657781371687,
|
|
"loss": 4.6913,
|
|
"mean_token_accuracy": 0.19675736278295516,
|
|
"num_tokens": 10992716.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.279621171951294,
|
|
"epoch": 5.0201714573877965,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003125855442530135,
|
|
"loss": 4.7678,
|
|
"mean_token_accuracy": 0.1925695925951004,
|
|
"num_tokens": 11004597.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.287301778793335,
|
|
"epoch": 5.025214321734746,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00031219439387557155,
|
|
"loss": 4.7384,
|
|
"mean_token_accuracy": 0.19644169509410858,
|
|
"num_tokens": 11015423.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.305578088760376,
|
|
"epoch": 5.030257186081695,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00031180312789709535,
|
|
"loss": 4.7685,
|
|
"mean_token_accuracy": 0.20085109770298004,
|
|
"num_tokens": 11028191.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.2362977981567385,
|
|
"epoch": 5.035300050428644,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003114117475336487,
|
|
"loss": 4.7778,
|
|
"mean_token_accuracy": 0.19998842477798462,
|
|
"num_tokens": 11040690.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.2070547580719,
|
|
"epoch": 5.040342914775593,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003110202540016509,
|
|
"loss": 4.6353,
|
|
"mean_token_accuracy": 0.1992221862077713,
|
|
"num_tokens": 11052580.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.040342914775593,
|
|
"eval_entropy": 5.200771926316899,
|
|
"eval_loss": 5.732107639312744,
|
|
"eval_mean_token_accuracy": 0.15048249594810356,
|
|
"eval_num_tokens": 11052580.0,
|
|
"eval_runtime": 2.6318,
|
|
"eval_samples_per_second": 1192.351,
|
|
"eval_steps_per_second": 149.329,
|
|
"step": 5000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9910,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.8487332930816e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|