af430a657a
Model: fpadovani/ita-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.5042864346949067,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 1868172705792000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|