af430a657a
Model: fpadovani/ita-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.5123550176500253,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 5570127084672000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|