741ef3be25
Model: fpadovani/dan-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
8123 lines
221 KiB
JSON
8123 lines
221 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.3472803347280333,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 6.802199983596802,
|
|
"epoch": 0.0041841004184100415,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.4445,
|
|
"mean_token_accuracy": 0.10309398248791694,
|
|
"num_tokens": 9848.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 6.861799955368042,
|
|
"epoch": 0.008368200836820083,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.4842,
|
|
"mean_token_accuracy": 0.10960042849183083,
|
|
"num_tokens": 18196.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 6.8262049674987795,
|
|
"epoch": 0.012552301255230125,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.4037,
|
|
"mean_token_accuracy": 0.11181848868727684,
|
|
"num_tokens": 26927.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 6.852504396438599,
|
|
"epoch": 0.016736401673640166,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.4858,
|
|
"mean_token_accuracy": 0.10887380167841912,
|
|
"num_tokens": 34987.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 6.775377893447876,
|
|
"epoch": 0.02092050209205021,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.4787,
|
|
"mean_token_accuracy": 0.10919283404946327,
|
|
"num_tokens": 44936.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 6.746171236038208,
|
|
"epoch": 0.02510460251046025,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.5022,
|
|
"mean_token_accuracy": 0.10517153143882751,
|
|
"num_tokens": 54562.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 6.737930631637573,
|
|
"epoch": 0.029288702928870293,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.5538,
|
|
"mean_token_accuracy": 0.1072025939822197,
|
|
"num_tokens": 63844.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 6.756887722015381,
|
|
"epoch": 0.03347280334728033,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.543,
|
|
"mean_token_accuracy": 0.11064185202121735,
|
|
"num_tokens": 74331.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 6.818012380599976,
|
|
"epoch": 0.03765690376569038,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.5329,
|
|
"mean_token_accuracy": 0.1078970804810524,
|
|
"num_tokens": 83028.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 6.826114511489868,
|
|
"epoch": 0.04184100418410042,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.4529,
|
|
"mean_token_accuracy": 0.11463268399238587,
|
|
"num_tokens": 91395.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 6.710237836837768,
|
|
"epoch": 0.04602510460251046,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.3486,
|
|
"mean_token_accuracy": 0.11258858814835548,
|
|
"num_tokens": 101868.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 6.751042079925537,
|
|
"epoch": 0.0502092050209205,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.4061,
|
|
"mean_token_accuracy": 0.11649479269981385,
|
|
"num_tokens": 110403.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 6.72459921836853,
|
|
"epoch": 0.05439330543933055,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.4011,
|
|
"mean_token_accuracy": 0.11154339239001274,
|
|
"num_tokens": 119557.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 6.770873069763184,
|
|
"epoch": 0.058577405857740586,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.4342,
|
|
"mean_token_accuracy": 0.11136164590716362,
|
|
"num_tokens": 128903.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 6.825613069534302,
|
|
"epoch": 0.06276150627615062,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.4633,
|
|
"mean_token_accuracy": 0.11030099019408227,
|
|
"num_tokens": 137630.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 6.702863502502441,
|
|
"epoch": 0.06694560669456066,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.2417,
|
|
"mean_token_accuracy": 0.11854962259531021,
|
|
"num_tokens": 146335.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 6.701785278320313,
|
|
"epoch": 0.07112970711297072,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.3738,
|
|
"mean_token_accuracy": 0.1200543224811554,
|
|
"num_tokens": 155766.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 6.6252336502075195,
|
|
"epoch": 0.07531380753138076,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.42,
|
|
"mean_token_accuracy": 0.11267679929733276,
|
|
"num_tokens": 165844.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 6.688562870025635,
|
|
"epoch": 0.0794979079497908,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.4734,
|
|
"mean_token_accuracy": 0.10707445964217185,
|
|
"num_tokens": 174858.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 6.8517701625823975,
|
|
"epoch": 0.08368200836820083,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.4229,
|
|
"mean_token_accuracy": 0.11569512039422988,
|
|
"num_tokens": 184020.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 6.7154582977294925,
|
|
"epoch": 0.08786610878661087,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.3312,
|
|
"mean_token_accuracy": 0.11155860349535943,
|
|
"num_tokens": 193714.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 6.720305871963501,
|
|
"epoch": 0.09205020920502092,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.5139,
|
|
"mean_token_accuracy": 0.1125582680106163,
|
|
"num_tokens": 203139.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 6.8431463718414305,
|
|
"epoch": 0.09623430962343096,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.4414,
|
|
"mean_token_accuracy": 0.11265396922826768,
|
|
"num_tokens": 212446.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 6.769783306121826,
|
|
"epoch": 0.100418410041841,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.3793,
|
|
"mean_token_accuracy": 0.12033891826868057,
|
|
"num_tokens": 220910.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 6.643674516677857,
|
|
"epoch": 0.10460251046025104,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.4122,
|
|
"mean_token_accuracy": 0.11757991164922714,
|
|
"num_tokens": 230522.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 6.830571556091309,
|
|
"epoch": 0.1087866108786611,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.4109,
|
|
"mean_token_accuracy": 0.11237624883651734,
|
|
"num_tokens": 240092.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 6.689486694335938,
|
|
"epoch": 0.11297071129707113,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.4836,
|
|
"mean_token_accuracy": 0.11546132564544678,
|
|
"num_tokens": 249478.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 6.702459669113159,
|
|
"epoch": 0.11715481171548117,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.3239,
|
|
"mean_token_accuracy": 0.1283320315182209,
|
|
"num_tokens": 257714.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 6.72728910446167,
|
|
"epoch": 0.12133891213389121,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.3811,
|
|
"mean_token_accuracy": 0.12412762865424157,
|
|
"num_tokens": 266569.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 6.742079639434815,
|
|
"epoch": 0.12552301255230125,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.3738,
|
|
"mean_token_accuracy": 0.11696210950613022,
|
|
"num_tokens": 276688.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 6.649025583267212,
|
|
"epoch": 0.1297071129707113,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.256,
|
|
"mean_token_accuracy": 0.1157757207751274,
|
|
"num_tokens": 285430.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 6.6111983299255375,
|
|
"epoch": 0.13389121338912133,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.4268,
|
|
"mean_token_accuracy": 0.11469644829630851,
|
|
"num_tokens": 295727.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 6.73095850944519,
|
|
"epoch": 0.13807531380753138,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.3704,
|
|
"mean_token_accuracy": 0.12031615525484085,
|
|
"num_tokens": 304854.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 6.7522295951843265,
|
|
"epoch": 0.14225941422594143,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.4069,
|
|
"mean_token_accuracy": 0.12297785431146621,
|
|
"num_tokens": 313579.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 6.770452976226807,
|
|
"epoch": 0.14644351464435146,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.5024,
|
|
"mean_token_accuracy": 0.11297214776277542,
|
|
"num_tokens": 323310.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 6.7838945388793945,
|
|
"epoch": 0.1506276150627615,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.368,
|
|
"mean_token_accuracy": 0.1218890331685543,
|
|
"num_tokens": 331720.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 6.694746971130371,
|
|
"epoch": 0.15481171548117154,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.3376,
|
|
"mean_token_accuracy": 0.12067973613739014,
|
|
"num_tokens": 340331.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 6.690021705627442,
|
|
"epoch": 0.1589958158995816,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.3738,
|
|
"mean_token_accuracy": 0.11946730315685272,
|
|
"num_tokens": 349677.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 6.797479248046875,
|
|
"epoch": 0.16317991631799164,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.3639,
|
|
"mean_token_accuracy": 0.11213221102952957,
|
|
"num_tokens": 358674.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 6.5850146293640135,
|
|
"epoch": 0.16736401673640167,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.4279,
|
|
"mean_token_accuracy": 0.11513617783784866,
|
|
"num_tokens": 368258.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 6.760379457473755,
|
|
"epoch": 0.17154811715481172,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.3138,
|
|
"mean_token_accuracy": 0.12832038179039956,
|
|
"num_tokens": 377083.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 6.707645225524902,
|
|
"epoch": 0.17573221757322174,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.3858,
|
|
"mean_token_accuracy": 0.11404623836278915,
|
|
"num_tokens": 385456.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 6.636844539642334,
|
|
"epoch": 0.1799163179916318,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.3039,
|
|
"mean_token_accuracy": 0.12129815816879272,
|
|
"num_tokens": 395295.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 6.709037256240845,
|
|
"epoch": 0.18410041841004185,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.3763,
|
|
"mean_token_accuracy": 0.11955561861395836,
|
|
"num_tokens": 405023.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 6.674328136444092,
|
|
"epoch": 0.18828451882845187,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.4302,
|
|
"mean_token_accuracy": 0.11591066569089889,
|
|
"num_tokens": 414224.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 6.727414226531982,
|
|
"epoch": 0.19246861924686193,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.2902,
|
|
"mean_token_accuracy": 0.11443310081958771,
|
|
"num_tokens": 423951.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 6.675467681884766,
|
|
"epoch": 0.19665271966527198,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.3461,
|
|
"mean_token_accuracy": 0.1151287503540516,
|
|
"num_tokens": 432427.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 6.5947874069213865,
|
|
"epoch": 0.200836820083682,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.307,
|
|
"mean_token_accuracy": 0.12089343965053559,
|
|
"num_tokens": 441560.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 6.724042892456055,
|
|
"epoch": 0.20502092050209206,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.387,
|
|
"mean_token_accuracy": 0.11083290576934815,
|
|
"num_tokens": 450364.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 6.667903852462769,
|
|
"epoch": 0.20920502092050208,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.3592,
|
|
"mean_token_accuracy": 0.11416003108024597,
|
|
"num_tokens": 460076.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 6.653390169143677,
|
|
"epoch": 0.21338912133891214,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.2896,
|
|
"mean_token_accuracy": 0.12183637991547584,
|
|
"num_tokens": 468718.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 6.727795171737671,
|
|
"epoch": 0.2175732217573222,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.39,
|
|
"mean_token_accuracy": 0.11337619721889496,
|
|
"num_tokens": 478436.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 6.739923620223999,
|
|
"epoch": 0.2217573221757322,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.3767,
|
|
"mean_token_accuracy": 0.11208422631025314,
|
|
"num_tokens": 486833.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 6.713146734237671,
|
|
"epoch": 0.22594142259414227,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.2926,
|
|
"mean_token_accuracy": 0.12398196682333947,
|
|
"num_tokens": 495119.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 6.724122428894043,
|
|
"epoch": 0.2301255230125523,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.475,
|
|
"mean_token_accuracy": 0.11590266525745392,
|
|
"num_tokens": 505396.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 6.622205924987793,
|
|
"epoch": 0.23430962343096234,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.3366,
|
|
"mean_token_accuracy": 0.11645356938242912,
|
|
"num_tokens": 515259.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 6.795593929290772,
|
|
"epoch": 0.2384937238493724,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.3565,
|
|
"mean_token_accuracy": 0.1146524854004383,
|
|
"num_tokens": 523608.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 6.573353290557861,
|
|
"epoch": 0.24267782426778242,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.3411,
|
|
"mean_token_accuracy": 0.11209621205925942,
|
|
"num_tokens": 534613.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 6.745756149291992,
|
|
"epoch": 0.24686192468619247,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.4111,
|
|
"mean_token_accuracy": 0.119757891446352,
|
|
"num_tokens": 543347.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 6.671696519851684,
|
|
"epoch": 0.2510460251046025,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.3369,
|
|
"mean_token_accuracy": 0.11111167520284652,
|
|
"num_tokens": 552371.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 6.657607221603394,
|
|
"epoch": 0.25523012552301255,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.3096,
|
|
"mean_token_accuracy": 0.11953857839107514,
|
|
"num_tokens": 560742.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 6.732522964477539,
|
|
"epoch": 0.2594142259414226,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.4478,
|
|
"mean_token_accuracy": 0.11639050245285035,
|
|
"num_tokens": 569960.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 6.726744174957275,
|
|
"epoch": 0.26359832635983266,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.3837,
|
|
"mean_token_accuracy": 0.11821903139352799,
|
|
"num_tokens": 578859.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 6.730755567550659,
|
|
"epoch": 0.26778242677824265,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.3894,
|
|
"mean_token_accuracy": 0.1184967190027237,
|
|
"num_tokens": 588146.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 6.5525407791137695,
|
|
"epoch": 0.2719665271966527,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.3298,
|
|
"mean_token_accuracy": 0.11562250927090645,
|
|
"num_tokens": 597268.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 6.6291014671325685,
|
|
"epoch": 0.27615062761506276,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.288,
|
|
"mean_token_accuracy": 0.12023762464523316,
|
|
"num_tokens": 606199.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 6.786021327972412,
|
|
"epoch": 0.2803347280334728,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.4049,
|
|
"mean_token_accuracy": 0.10941854566335678,
|
|
"num_tokens": 615966.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 6.667136526107788,
|
|
"epoch": 0.28451882845188287,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.3727,
|
|
"mean_token_accuracy": 0.11674070581793786,
|
|
"num_tokens": 625689.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 6.698928308486939,
|
|
"epoch": 0.28870292887029286,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.4012,
|
|
"mean_token_accuracy": 0.11340371370315552,
|
|
"num_tokens": 634877.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 6.716418790817261,
|
|
"epoch": 0.2928870292887029,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.3268,
|
|
"mean_token_accuracy": 0.12364484667778015,
|
|
"num_tokens": 644042.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 6.608269596099854,
|
|
"epoch": 0.29707112970711297,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.3555,
|
|
"mean_token_accuracy": 0.11423439830541611,
|
|
"num_tokens": 653072.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 6.664682340621948,
|
|
"epoch": 0.301255230125523,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.3614,
|
|
"mean_token_accuracy": 0.11689013764262199,
|
|
"num_tokens": 663686.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 6.632741975784302,
|
|
"epoch": 0.3054393305439331,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.2951,
|
|
"mean_token_accuracy": 0.11744664832949639,
|
|
"num_tokens": 672293.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 6.489195203781128,
|
|
"epoch": 0.30962343096234307,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.266,
|
|
"mean_token_accuracy": 0.11952990591526032,
|
|
"num_tokens": 681758.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 6.6848828315734865,
|
|
"epoch": 0.3138075313807531,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.3038,
|
|
"mean_token_accuracy": 0.11904869601130486,
|
|
"num_tokens": 690727.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.636202001571656,
|
|
"epoch": 0.3179916317991632,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.2574,
|
|
"mean_token_accuracy": 0.11928709298372268,
|
|
"num_tokens": 701210.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 6.494536590576172,
|
|
"epoch": 0.32217573221757323,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.2645,
|
|
"mean_token_accuracy": 0.12418703511357307,
|
|
"num_tokens": 710869.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.6857581615448,
|
|
"epoch": 0.3263598326359833,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.405,
|
|
"mean_token_accuracy": 0.11158787459135056,
|
|
"num_tokens": 721455.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.52554292678833,
|
|
"epoch": 0.3305439330543933,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.2772,
|
|
"mean_token_accuracy": 0.1241541288793087,
|
|
"num_tokens": 730917.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.651937246322632,
|
|
"epoch": 0.33472803347280333,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.3144,
|
|
"mean_token_accuracy": 0.11836001500487328,
|
|
"num_tokens": 739302.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.575543546676636,
|
|
"epoch": 0.3389121338912134,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.2823,
|
|
"mean_token_accuracy": 0.1181989461183548,
|
|
"num_tokens": 748785.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.55746021270752,
|
|
"epoch": 0.34309623430962344,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.2125,
|
|
"mean_token_accuracy": 0.12301687523722649,
|
|
"num_tokens": 757215.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.495482349395752,
|
|
"epoch": 0.3472803347280335,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.158,
|
|
"mean_token_accuracy": 0.12710630893707275,
|
|
"num_tokens": 765682.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.671501731872558,
|
|
"epoch": 0.3514644351464435,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.3598,
|
|
"mean_token_accuracy": 0.1148804098367691,
|
|
"num_tokens": 774814.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.54171838760376,
|
|
"epoch": 0.35564853556485354,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.2819,
|
|
"mean_token_accuracy": 0.1194837212562561,
|
|
"num_tokens": 784019.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.477924489974976,
|
|
"epoch": 0.3598326359832636,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.1587,
|
|
"mean_token_accuracy": 0.12297921627759933,
|
|
"num_tokens": 793237.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.627668190002441,
|
|
"epoch": 0.36401673640167365,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.2292,
|
|
"mean_token_accuracy": 0.11881406381726264,
|
|
"num_tokens": 801548.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.450384140014648,
|
|
"epoch": 0.3682008368200837,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.2893,
|
|
"mean_token_accuracy": 0.12606984302401542,
|
|
"num_tokens": 810974.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.513573360443115,
|
|
"epoch": 0.3723849372384937,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.1682,
|
|
"mean_token_accuracy": 0.12752986550331116,
|
|
"num_tokens": 820023.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.611561870574951,
|
|
"epoch": 0.37656903765690375,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.2589,
|
|
"mean_token_accuracy": 0.12134803086519241,
|
|
"num_tokens": 828297.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.553285312652588,
|
|
"epoch": 0.3807531380753138,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.3043,
|
|
"mean_token_accuracy": 0.1155056394636631,
|
|
"num_tokens": 838268.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.6202904224395756,
|
|
"epoch": 0.38493723849372385,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.2264,
|
|
"mean_token_accuracy": 0.1235147014260292,
|
|
"num_tokens": 846491.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.374619865417481,
|
|
"epoch": 0.3891213389121339,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.1882,
|
|
"mean_token_accuracy": 0.12715737968683244,
|
|
"num_tokens": 856501.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.529414176940918,
|
|
"epoch": 0.39330543933054396,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.2449,
|
|
"mean_token_accuracy": 0.12165890857577324,
|
|
"num_tokens": 867054.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.44644775390625,
|
|
"epoch": 0.39748953974895396,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.271,
|
|
"mean_token_accuracy": 0.12292861416935921,
|
|
"num_tokens": 875956.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.492046117782593,
|
|
"epoch": 0.401673640167364,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.1897,
|
|
"mean_token_accuracy": 0.12475509718060493,
|
|
"num_tokens": 884427.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.615246963500977,
|
|
"epoch": 0.40585774058577406,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.2791,
|
|
"mean_token_accuracy": 0.12459831088781356,
|
|
"num_tokens": 894181.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.427358102798462,
|
|
"epoch": 0.4100418410041841,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.1205,
|
|
"mean_token_accuracy": 0.12992214635014535,
|
|
"num_tokens": 904076.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.470092678070069,
|
|
"epoch": 0.41422594142259417,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.2517,
|
|
"mean_token_accuracy": 0.12282974123954774,
|
|
"num_tokens": 913915.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.594865226745606,
|
|
"epoch": 0.41841004184100417,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.3694,
|
|
"mean_token_accuracy": 0.11399232223629951,
|
|
"num_tokens": 922876.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.41841004184100417,
|
|
"eval_entropy": 6.477647836954316,
|
|
"eval_loss": 6.225774765014648,
|
|
"eval_mean_token_accuracy": 0.1297860539809692,
|
|
"eval_num_tokens": 922876.0,
|
|
"eval_runtime": 2.8824,
|
|
"eval_samples_per_second": 1355.489,
|
|
"eval_steps_per_second": 169.653,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.409206914901733,
|
|
"epoch": 0.4225941422594142,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.1393,
|
|
"mean_token_accuracy": 0.1315084345638752,
|
|
"num_tokens": 931998.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.55022029876709,
|
|
"epoch": 0.42677824267782427,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.214,
|
|
"mean_token_accuracy": 0.12237880975008011,
|
|
"num_tokens": 941387.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.381158638000488,
|
|
"epoch": 0.4309623430962343,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.2484,
|
|
"mean_token_accuracy": 0.11762547865509987,
|
|
"num_tokens": 951866.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.425309085845948,
|
|
"epoch": 0.4351464435146444,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.2003,
|
|
"mean_token_accuracy": 0.12336372956633568,
|
|
"num_tokens": 961252.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.572003746032715,
|
|
"epoch": 0.4393305439330544,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.2033,
|
|
"mean_token_accuracy": 0.1296585388481617,
|
|
"num_tokens": 969857.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.4825043201446535,
|
|
"epoch": 0.4435146443514644,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.1998,
|
|
"mean_token_accuracy": 0.12262558117508889,
|
|
"num_tokens": 978754.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.42857813835144,
|
|
"epoch": 0.4476987447698745,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.1703,
|
|
"mean_token_accuracy": 0.11900547742843628,
|
|
"num_tokens": 987935.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.492337369918824,
|
|
"epoch": 0.45188284518828453,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.1494,
|
|
"mean_token_accuracy": 0.12749148234725,
|
|
"num_tokens": 996873.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.484167146682739,
|
|
"epoch": 0.4560669456066946,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.2565,
|
|
"mean_token_accuracy": 0.12450389862060547,
|
|
"num_tokens": 1006487.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.306475353240967,
|
|
"epoch": 0.4602510460251046,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.0848,
|
|
"mean_token_accuracy": 0.1302729830145836,
|
|
"num_tokens": 1015715.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.4527686595916744,
|
|
"epoch": 0.46443514644351463,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.2025,
|
|
"mean_token_accuracy": 0.12450708001852036,
|
|
"num_tokens": 1025091.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.397480440139771,
|
|
"epoch": 0.4686192468619247,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.1394,
|
|
"mean_token_accuracy": 0.12280101329088211,
|
|
"num_tokens": 1034174.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.512916564941406,
|
|
"epoch": 0.47280334728033474,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.3333,
|
|
"mean_token_accuracy": 0.11398612335324287,
|
|
"num_tokens": 1043777.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.476423072814941,
|
|
"epoch": 0.4769874476987448,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.1838,
|
|
"mean_token_accuracy": 0.12991034761071205,
|
|
"num_tokens": 1052749.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.485992193222046,
|
|
"epoch": 0.4811715481171548,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.207,
|
|
"mean_token_accuracy": 0.12597287371754645,
|
|
"num_tokens": 1061564.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.389427852630615,
|
|
"epoch": 0.48535564853556484,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.1204,
|
|
"mean_token_accuracy": 0.13354995474219322,
|
|
"num_tokens": 1070006.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.363674736022949,
|
|
"epoch": 0.4895397489539749,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.1847,
|
|
"mean_token_accuracy": 0.12432906851172447,
|
|
"num_tokens": 1079799.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.432363891601563,
|
|
"epoch": 0.49372384937238495,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.0386,
|
|
"mean_token_accuracy": 0.13422225415706635,
|
|
"num_tokens": 1088842.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.306155967712402,
|
|
"epoch": 0.497907949790795,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.1119,
|
|
"mean_token_accuracy": 0.13670141398906707,
|
|
"num_tokens": 1097960.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.31667046546936,
|
|
"epoch": 0.502092050209205,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.0895,
|
|
"mean_token_accuracy": 0.12390791401267051,
|
|
"num_tokens": 1106797.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.38304967880249,
|
|
"epoch": 0.5062761506276151,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.2594,
|
|
"mean_token_accuracy": 0.13380709663033485,
|
|
"num_tokens": 1116384.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.254551887512207,
|
|
"epoch": 0.5104602510460251,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.012,
|
|
"mean_token_accuracy": 0.13452231362462044,
|
|
"num_tokens": 1125206.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.527569246292114,
|
|
"epoch": 0.5146443514644351,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.2528,
|
|
"mean_token_accuracy": 0.12866818457841872,
|
|
"num_tokens": 1134110.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.497180366516114,
|
|
"epoch": 0.5188284518828452,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.2468,
|
|
"mean_token_accuracy": 0.12335871756076813,
|
|
"num_tokens": 1142833.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.333167219161988,
|
|
"epoch": 0.5230125523012552,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.1645,
|
|
"mean_token_accuracy": 0.13038308247923852,
|
|
"num_tokens": 1152006.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.3656152248382565,
|
|
"epoch": 0.5271966527196653,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.0956,
|
|
"mean_token_accuracy": 0.13017264977097512,
|
|
"num_tokens": 1161061.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.417209196090698,
|
|
"epoch": 0.5313807531380753,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.1234,
|
|
"mean_token_accuracy": 0.1287168890237808,
|
|
"num_tokens": 1170754.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.204028367996216,
|
|
"epoch": 0.5355648535564853,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.0824,
|
|
"mean_token_accuracy": 0.1296873278915882,
|
|
"num_tokens": 1180854.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.387170028686524,
|
|
"epoch": 0.5397489539748954,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.1625,
|
|
"mean_token_accuracy": 0.12682752832770347,
|
|
"num_tokens": 1190567.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.313226795196533,
|
|
"epoch": 0.5439330543933054,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.0333,
|
|
"mean_token_accuracy": 0.13330961391329765,
|
|
"num_tokens": 1200247.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.270791339874267,
|
|
"epoch": 0.5481171548117155,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.1335,
|
|
"mean_token_accuracy": 0.12965440675616263,
|
|
"num_tokens": 1209358.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.285220718383789,
|
|
"epoch": 0.5523012552301255,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.1296,
|
|
"mean_token_accuracy": 0.13011690378189086,
|
|
"num_tokens": 1217746.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.305118894577026,
|
|
"epoch": 0.5564853556485355,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.1036,
|
|
"mean_token_accuracy": 0.13556399419903756,
|
|
"num_tokens": 1226829.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.277094221115112,
|
|
"epoch": 0.5606694560669456,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.0266,
|
|
"mean_token_accuracy": 0.1367674969136715,
|
|
"num_tokens": 1236642.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.425677394866943,
|
|
"epoch": 0.5648535564853556,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.1464,
|
|
"mean_token_accuracy": 0.13671540543437005,
|
|
"num_tokens": 1244934.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.204587936401367,
|
|
"epoch": 0.5690376569037657,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.0569,
|
|
"mean_token_accuracy": 0.1287803091108799,
|
|
"num_tokens": 1254786.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.193703556060791,
|
|
"epoch": 0.5732217573221757,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.0179,
|
|
"mean_token_accuracy": 0.13714393973350525,
|
|
"num_tokens": 1263745.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.277493715286255,
|
|
"epoch": 0.5774058577405857,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.1458,
|
|
"mean_token_accuracy": 0.13557302355766296,
|
|
"num_tokens": 1273101.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.420682430267334,
|
|
"epoch": 0.5815899581589958,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.1451,
|
|
"mean_token_accuracy": 0.12692318111658096,
|
|
"num_tokens": 1282547.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.242686367034912,
|
|
"epoch": 0.5857740585774058,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.1568,
|
|
"mean_token_accuracy": 0.1301800861954689,
|
|
"num_tokens": 1291594.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.311301374435425,
|
|
"epoch": 0.5899581589958159,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.0677,
|
|
"mean_token_accuracy": 0.13865734487771988,
|
|
"num_tokens": 1300282.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.283356428146362,
|
|
"epoch": 0.5941422594142259,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.1457,
|
|
"mean_token_accuracy": 0.13872819766402245,
|
|
"num_tokens": 1309304.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.22599310874939,
|
|
"epoch": 0.5983263598326359,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 5.9905,
|
|
"mean_token_accuracy": 0.13430829271674155,
|
|
"num_tokens": 1317638.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.374803638458252,
|
|
"epoch": 0.602510460251046,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.1976,
|
|
"mean_token_accuracy": 0.12840964049100875,
|
|
"num_tokens": 1327242.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.2666154384613035,
|
|
"epoch": 0.606694560669456,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.014,
|
|
"mean_token_accuracy": 0.13483576700091363,
|
|
"num_tokens": 1335880.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.198787117004395,
|
|
"epoch": 0.6108786610878661,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.0882,
|
|
"mean_token_accuracy": 0.13066002652049064,
|
|
"num_tokens": 1344909.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.220889329910278,
|
|
"epoch": 0.6150627615062761,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000367,
|
|
"loss": 5.9637,
|
|
"mean_token_accuracy": 0.1385305680334568,
|
|
"num_tokens": 1353997.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.250986480712891,
|
|
"epoch": 0.6192468619246861,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.1325,
|
|
"mean_token_accuracy": 0.13334286585450172,
|
|
"num_tokens": 1363293.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.397185230255127,
|
|
"epoch": 0.6234309623430963,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.1387,
|
|
"mean_token_accuracy": 0.12498829737305642,
|
|
"num_tokens": 1372174.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.285424041748047,
|
|
"epoch": 0.6276150627615062,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.0244,
|
|
"mean_token_accuracy": 0.13294555693864823,
|
|
"num_tokens": 1381067.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.165482044219971,
|
|
"epoch": 0.6317991631799164,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.0349,
|
|
"mean_token_accuracy": 0.13717807978391647,
|
|
"num_tokens": 1390072.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.254514551162719,
|
|
"epoch": 0.6359832635983264,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.017,
|
|
"mean_token_accuracy": 0.1379196561872959,
|
|
"num_tokens": 1399391.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.182499074935913,
|
|
"epoch": 0.6401673640167364,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.0082,
|
|
"mean_token_accuracy": 0.1376515731215477,
|
|
"num_tokens": 1408054.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.243570137023926,
|
|
"epoch": 0.6443514644351465,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.0371,
|
|
"mean_token_accuracy": 0.12797121033072473,
|
|
"num_tokens": 1417406.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.195934247970581,
|
|
"epoch": 0.6485355648535565,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.1035,
|
|
"mean_token_accuracy": 0.12684772536158562,
|
|
"num_tokens": 1426907.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.2838667869567875,
|
|
"epoch": 0.6527196652719666,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.1754,
|
|
"mean_token_accuracy": 0.12340578138828277,
|
|
"num_tokens": 1436085.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.26796145439148,
|
|
"epoch": 0.6569037656903766,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.0634,
|
|
"mean_token_accuracy": 0.1289584718644619,
|
|
"num_tokens": 1444823.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.289790201187134,
|
|
"epoch": 0.6610878661087866,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.0986,
|
|
"mean_token_accuracy": 0.13113027885556222,
|
|
"num_tokens": 1454621.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.109712553024292,
|
|
"epoch": 0.6652719665271967,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 5.9769,
|
|
"mean_token_accuracy": 0.1314921833574772,
|
|
"num_tokens": 1463784.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.236892223358154,
|
|
"epoch": 0.6694560669456067,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.0932,
|
|
"mean_token_accuracy": 0.13157428726553916,
|
|
"num_tokens": 1473257.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.230193567276001,
|
|
"epoch": 0.6736401673640168,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.0429,
|
|
"mean_token_accuracy": 0.13767861425876618,
|
|
"num_tokens": 1482195.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.138327884674072,
|
|
"epoch": 0.6778242677824268,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.0855,
|
|
"mean_token_accuracy": 0.1348243311047554,
|
|
"num_tokens": 1491590.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.219961166381836,
|
|
"epoch": 0.6820083682008368,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 5.9732,
|
|
"mean_token_accuracy": 0.14189569428563117,
|
|
"num_tokens": 1500208.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.153802299499512,
|
|
"epoch": 0.6861924686192469,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.1078,
|
|
"mean_token_accuracy": 0.13705730587244033,
|
|
"num_tokens": 1509412.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.214185428619385,
|
|
"epoch": 0.6903765690376569,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.1048,
|
|
"mean_token_accuracy": 0.13078184500336648,
|
|
"num_tokens": 1518426.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.275689888000488,
|
|
"epoch": 0.694560669456067,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.0391,
|
|
"mean_token_accuracy": 0.1352381370961666,
|
|
"num_tokens": 1528605.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.123213243484497,
|
|
"epoch": 0.698744769874477,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.0004,
|
|
"mean_token_accuracy": 0.14104130268096923,
|
|
"num_tokens": 1537449.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.1678873062133786,
|
|
"epoch": 0.702928870292887,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.0386,
|
|
"mean_token_accuracy": 0.13681078255176543,
|
|
"num_tokens": 1546743.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.144597291946411,
|
|
"epoch": 0.7071129707112971,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.0213,
|
|
"mean_token_accuracy": 0.13631223142147064,
|
|
"num_tokens": 1555691.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.11121015548706,
|
|
"epoch": 0.7112970711297071,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 5.9676,
|
|
"mean_token_accuracy": 0.14060806781053542,
|
|
"num_tokens": 1564957.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.265472888946533,
|
|
"epoch": 0.7154811715481172,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.0478,
|
|
"mean_token_accuracy": 0.13608204945921898,
|
|
"num_tokens": 1573901.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.0549863338470455,
|
|
"epoch": 0.7196652719665272,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.13866082727909088,
|
|
"num_tokens": 1582644.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.237826824188232,
|
|
"epoch": 0.7238493723849372,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.1671,
|
|
"mean_token_accuracy": 0.12615143060684203,
|
|
"num_tokens": 1592363.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.219017791748047,
|
|
"epoch": 0.7280334728033473,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.1152,
|
|
"mean_token_accuracy": 0.131891880184412,
|
|
"num_tokens": 1602012.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.147130346298217,
|
|
"epoch": 0.7322175732217573,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.0066,
|
|
"mean_token_accuracy": 0.1359176442027092,
|
|
"num_tokens": 1611625.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.027130603790283,
|
|
"epoch": 0.7364016736401674,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 5.96,
|
|
"mean_token_accuracy": 0.13448768407106398,
|
|
"num_tokens": 1620941.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.1633179664611815,
|
|
"epoch": 0.7405857740585774,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.0589,
|
|
"mean_token_accuracy": 0.13061976209282875,
|
|
"num_tokens": 1629225.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.152882194519043,
|
|
"epoch": 0.7447698744769874,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.0312,
|
|
"mean_token_accuracy": 0.12992663234472274,
|
|
"num_tokens": 1639551.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.138622665405274,
|
|
"epoch": 0.7489539748953975,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.02,
|
|
"mean_token_accuracy": 0.12998297885060311,
|
|
"num_tokens": 1648963.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.0763167381286625,
|
|
"epoch": 0.7531380753138075,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 5.9294,
|
|
"mean_token_accuracy": 0.14245674312114714,
|
|
"num_tokens": 1657735.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.254350471496582,
|
|
"epoch": 0.7573221757322176,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 5.9799,
|
|
"mean_token_accuracy": 0.13656511679291725,
|
|
"num_tokens": 1666349.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 5.909399604797363,
|
|
"epoch": 0.7615062761506276,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 5.9225,
|
|
"mean_token_accuracy": 0.13355625793337822,
|
|
"num_tokens": 1675920.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.054324007034301,
|
|
"epoch": 0.7656903765690377,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 5.9577,
|
|
"mean_token_accuracy": 0.13852019906044005,
|
|
"num_tokens": 1685626.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.240289211273193,
|
|
"epoch": 0.7698744769874477,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.0816,
|
|
"mean_token_accuracy": 0.14210813418030738,
|
|
"num_tokens": 1694972.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.101596355438232,
|
|
"epoch": 0.7740585774058577,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 5.9615,
|
|
"mean_token_accuracy": 0.14294997751712799,
|
|
"num_tokens": 1703806.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.082038736343383,
|
|
"epoch": 0.7782426778242678,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.0631,
|
|
"mean_token_accuracy": 0.1384358860552311,
|
|
"num_tokens": 1712949.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.252063655853272,
|
|
"epoch": 0.7824267782426778,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.1782,
|
|
"mean_token_accuracy": 0.134084989130497,
|
|
"num_tokens": 1722328.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.226525402069091,
|
|
"epoch": 0.7866108786610879,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1574,
|
|
"mean_token_accuracy": 0.1339714154601097,
|
|
"num_tokens": 1731880.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.010696268081665,
|
|
"epoch": 0.7907949790794979,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 5.9706,
|
|
"mean_token_accuracy": 0.140121129155159,
|
|
"num_tokens": 1740719.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.144752645492554,
|
|
"epoch": 0.7949790794979079,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.0063,
|
|
"mean_token_accuracy": 0.13252055421471595,
|
|
"num_tokens": 1749753.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.193329572677612,
|
|
"epoch": 0.799163179916318,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.0296,
|
|
"mean_token_accuracy": 0.12951601520180703,
|
|
"num_tokens": 1758913.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.127861022949219,
|
|
"epoch": 0.803347280334728,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.047,
|
|
"mean_token_accuracy": 0.13796046003699303,
|
|
"num_tokens": 1768536.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.048326778411865,
|
|
"epoch": 0.8075313807531381,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000482,
|
|
"loss": 5.9733,
|
|
"mean_token_accuracy": 0.13804301396012306,
|
|
"num_tokens": 1777291.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.0986556053161625,
|
|
"epoch": 0.8117154811715481,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.0696,
|
|
"mean_token_accuracy": 0.13444095477461815,
|
|
"num_tokens": 1787186.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.042343235015869,
|
|
"epoch": 0.8158995815899581,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 5.8767,
|
|
"mean_token_accuracy": 0.14351157248020172,
|
|
"num_tokens": 1795581.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 5.956916570663452,
|
|
"epoch": 0.8200836820083682,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 5.8884,
|
|
"mean_token_accuracy": 0.14488404095172883,
|
|
"num_tokens": 1804627.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.110140657424926,
|
|
"epoch": 0.8242677824267782,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0162,
|
|
"mean_token_accuracy": 0.13789800107479094,
|
|
"num_tokens": 1813808.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.012312602996826,
|
|
"epoch": 0.8284518828451883,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.008,
|
|
"mean_token_accuracy": 0.13816680759191513,
|
|
"num_tokens": 1823422.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.114294052124023,
|
|
"epoch": 0.8326359832635983,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.0489,
|
|
"mean_token_accuracy": 0.13757719099521637,
|
|
"num_tokens": 1832437.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.102473211288452,
|
|
"epoch": 0.8368200836820083,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 5.97,
|
|
"mean_token_accuracy": 0.13490361198782921,
|
|
"num_tokens": 1842063.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8368200836820083,
|
|
"eval_entropy": 5.8732611679592015,
|
|
"eval_loss": 6.004635334014893,
|
|
"eval_mean_token_accuracy": 0.1470446902458653,
|
|
"eval_num_tokens": 1842063.0,
|
|
"eval_runtime": 2.7509,
|
|
"eval_samples_per_second": 1420.266,
|
|
"eval_steps_per_second": 177.76,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.095950555801392,
|
|
"epoch": 0.8410041841004184,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999998518355666,
|
|
"loss": 6.118,
|
|
"mean_token_accuracy": 0.13328226357698442,
|
|
"num_tokens": 1851043.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.020190048217773,
|
|
"epoch": 0.8451882845188284,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999992499178905,
|
|
"loss": 5.9667,
|
|
"mean_token_accuracy": 0.14448187947273256,
|
|
"num_tokens": 1860268.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.136621141433716,
|
|
"epoch": 0.8493723849372385,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999981849879322,
|
|
"loss": 6.0767,
|
|
"mean_token_accuracy": 0.13768067285418512,
|
|
"num_tokens": 1870096.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.0227445602417,
|
|
"epoch": 0.8535564853556485,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999966570478833,
|
|
"loss": 5.9369,
|
|
"mean_token_accuracy": 0.14211413413286209,
|
|
"num_tokens": 1879164.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 5.965367984771729,
|
|
"epoch": 0.8577405857740585,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004999946661008878,
|
|
"loss": 5.9174,
|
|
"mean_token_accuracy": 0.14115409776568413,
|
|
"num_tokens": 1888135.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.037340688705444,
|
|
"epoch": 0.8619246861924686,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000499992212151043,
|
|
"loss": 5.979,
|
|
"mean_token_accuracy": 0.13762755542993546,
|
|
"num_tokens": 1898302.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 5.967270994186402,
|
|
"epoch": 0.8661087866108786,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999892952033987,
|
|
"loss": 5.9534,
|
|
"mean_token_accuracy": 0.13841190561652184,
|
|
"num_tokens": 1907753.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.034883499145508,
|
|
"epoch": 0.8702928870292888,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004999859152639573,
|
|
"loss": 6.0102,
|
|
"mean_token_accuracy": 0.13664262220263482,
|
|
"num_tokens": 1917543.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.117797231674194,
|
|
"epoch": 0.8744769874476988,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004999820723396745,
|
|
"loss": 6.0532,
|
|
"mean_token_accuracy": 0.1308543339371681,
|
|
"num_tokens": 1927335.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.065930318832398,
|
|
"epoch": 0.8786610878661087,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999777664384581,
|
|
"loss": 5.9696,
|
|
"mean_token_accuracy": 0.13757520616054536,
|
|
"num_tokens": 1936381.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.017509794235229,
|
|
"epoch": 0.8828451882845189,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999729975691692,
|
|
"loss": 6.0077,
|
|
"mean_token_accuracy": 0.13833484947681426,
|
|
"num_tokens": 1945559.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 5.9340832233428955,
|
|
"epoch": 0.8870292887029289,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999677657416212,
|
|
"loss": 5.8732,
|
|
"mean_token_accuracy": 0.13803015872836114,
|
|
"num_tokens": 1954709.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.036171865463257,
|
|
"epoch": 0.891213389121339,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999620709665804,
|
|
"loss": 5.9938,
|
|
"mean_token_accuracy": 0.13700244426727295,
|
|
"num_tokens": 1963664.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.11403112411499,
|
|
"epoch": 0.895397489539749,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999559132557659,
|
|
"loss": 6.0379,
|
|
"mean_token_accuracy": 0.1384876064956188,
|
|
"num_tokens": 1974103.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.0869780540466305,
|
|
"epoch": 0.899581589958159,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004999492926218489,
|
|
"loss": 5.9107,
|
|
"mean_token_accuracy": 0.14324470087885857,
|
|
"num_tokens": 1982924.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 5.880443286895752,
|
|
"epoch": 0.9037656903765691,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000499942209078454,
|
|
"loss": 5.8567,
|
|
"mean_token_accuracy": 0.14962902218103408,
|
|
"num_tokens": 1992033.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 5.9380744934082035,
|
|
"epoch": 0.9079497907949791,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999346626401578,
|
|
"loss": 5.8593,
|
|
"mean_token_accuracy": 0.15124175548553467,
|
|
"num_tokens": 2001218.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 5.97272481918335,
|
|
"epoch": 0.9121338912133892,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999266533224897,
|
|
"loss": 6.0209,
|
|
"mean_token_accuracy": 0.13835917562246322,
|
|
"num_tokens": 2010463.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.04907603263855,
|
|
"epoch": 0.9163179916317992,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004999181811419314,
|
|
"loss": 5.8831,
|
|
"mean_token_accuracy": 0.14357504546642302,
|
|
"num_tokens": 2019246.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 5.999930572509766,
|
|
"epoch": 0.9205020920502092,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999092461159177,
|
|
"loss": 5.9529,
|
|
"mean_token_accuracy": 0.1417504332959652,
|
|
"num_tokens": 2028083.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 5.937072658538819,
|
|
"epoch": 0.9246861924686193,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004998998482628351,
|
|
"loss": 5.7845,
|
|
"mean_token_accuracy": 0.14775603413581848,
|
|
"num_tokens": 2036631.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 5.905835151672363,
|
|
"epoch": 0.9288702928870293,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004998899876020229,
|
|
"loss": 5.9174,
|
|
"mean_token_accuracy": 0.14066033884882928,
|
|
"num_tokens": 2045910.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 5.955777645111084,
|
|
"epoch": 0.9330543933054394,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000499879664153773,
|
|
"loss": 5.9551,
|
|
"mean_token_accuracy": 0.15176769345998764,
|
|
"num_tokens": 2055959.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.181238317489624,
|
|
"epoch": 0.9372384937238494,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998688779393291,
|
|
"loss": 6.0553,
|
|
"mean_token_accuracy": 0.13956528753042222,
|
|
"num_tokens": 2065475.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 5.694952154159546,
|
|
"epoch": 0.9414225941422594,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998576289808877,
|
|
"loss": 5.8127,
|
|
"mean_token_accuracy": 0.14700343757867812,
|
|
"num_tokens": 2074469.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 5.991063165664673,
|
|
"epoch": 0.9456066945606695,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998459173015972,
|
|
"loss": 5.9968,
|
|
"mean_token_accuracy": 0.14213211834430695,
|
|
"num_tokens": 2084604.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.020702934265136,
|
|
"epoch": 0.9497907949790795,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004998337429255584,
|
|
"loss": 5.9311,
|
|
"mean_token_accuracy": 0.14435766488313675,
|
|
"num_tokens": 2093377.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.022408723831177,
|
|
"epoch": 0.9539748953974896,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998211058778243,
|
|
"loss": 5.9763,
|
|
"mean_token_accuracy": 0.1393851213157177,
|
|
"num_tokens": 2102798.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.016721439361572,
|
|
"epoch": 0.9581589958158996,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998080061843999,
|
|
"loss": 5.9454,
|
|
"mean_token_accuracy": 0.1397051155567169,
|
|
"num_tokens": 2112516.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.075040674209594,
|
|
"epoch": 0.9623430962343096,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499794443872242,
|
|
"loss": 6.0249,
|
|
"mean_token_accuracy": 0.14221663549542427,
|
|
"num_tokens": 2121677.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 5.8664342880249025,
|
|
"epoch": 0.9665271966527197,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997804189692601,
|
|
"loss": 5.873,
|
|
"mean_token_accuracy": 0.14960964918136596,
|
|
"num_tokens": 2131220.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.041165208816528,
|
|
"epoch": 0.9707112970711297,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997659315043149,
|
|
"loss": 6.0275,
|
|
"mean_token_accuracy": 0.14035088643431665,
|
|
"num_tokens": 2142104.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 5.950460052490234,
|
|
"epoch": 0.9748953974895398,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997509815072191,
|
|
"loss": 5.9345,
|
|
"mean_token_accuracy": 0.14209006875753402,
|
|
"num_tokens": 2151970.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 5.905142402648925,
|
|
"epoch": 0.9790794979079498,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997355690087379,
|
|
"loss": 5.8866,
|
|
"mean_token_accuracy": 0.14416528195142747,
|
|
"num_tokens": 2161371.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 5.8869006633758545,
|
|
"epoch": 0.9832635983263598,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997196940405874,
|
|
"loss": 5.8535,
|
|
"mean_token_accuracy": 0.14034167677164078,
|
|
"num_tokens": 2170650.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 5.926012420654297,
|
|
"epoch": 0.9874476987447699,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997033566354358,
|
|
"loss": 5.9338,
|
|
"mean_token_accuracy": 0.14502293542027472,
|
|
"num_tokens": 2180934.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 5.964864015579224,
|
|
"epoch": 0.9916317991631799,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996865568269028,
|
|
"loss": 5.8545,
|
|
"mean_token_accuracy": 0.15056095868349076,
|
|
"num_tokens": 2189894.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 5.927501249313354,
|
|
"epoch": 0.99581589958159,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996692946495597,
|
|
"loss": 5.8354,
|
|
"mean_token_accuracy": 0.15513431578874587,
|
|
"num_tokens": 2200208.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 5.898432397842408,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004996515701389295,
|
|
"loss": 5.88,
|
|
"mean_token_accuracy": 0.14509762749075888,
|
|
"num_tokens": 2209322.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 5.890221166610718,
|
|
"epoch": 1.00418410041841,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004996333833314862,
|
|
"loss": 5.6625,
|
|
"mean_token_accuracy": 0.15636170506477357,
|
|
"num_tokens": 2217663.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 5.808015203475952,
|
|
"epoch": 1.00836820083682,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996147342646554,
|
|
"loss": 5.7086,
|
|
"mean_token_accuracy": 0.15036543905735017,
|
|
"num_tokens": 2226559.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 5.971857309341431,
|
|
"epoch": 1.0125523012552302,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995956229768137,
|
|
"loss": 5.7512,
|
|
"mean_token_accuracy": 0.1576405018568039,
|
|
"num_tokens": 2235468.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 5.761348056793213,
|
|
"epoch": 1.0167364016736402,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995760495072893,
|
|
"loss": 5.5545,
|
|
"mean_token_accuracy": 0.16244466453790665,
|
|
"num_tokens": 2245176.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 5.851337480545044,
|
|
"epoch": 1.0209205020920502,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995560138963611,
|
|
"loss": 5.7064,
|
|
"mean_token_accuracy": 0.15552184730768204,
|
|
"num_tokens": 2255123.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 5.866237306594849,
|
|
"epoch": 1.0251046025104602,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995355161852593,
|
|
"loss": 5.6223,
|
|
"mean_token_accuracy": 0.15251805186271666,
|
|
"num_tokens": 2263596.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 5.857808399200439,
|
|
"epoch": 1.0292887029288702,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004995145564161648,
|
|
"loss": 5.6833,
|
|
"mean_token_accuracy": 0.14903822392225266,
|
|
"num_tokens": 2272228.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.834920883178711,
|
|
"epoch": 1.0334728033472804,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004994931346322094,
|
|
"loss": 5.7218,
|
|
"mean_token_accuracy": 0.15380252599716188,
|
|
"num_tokens": 2281248.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 5.865534734725952,
|
|
"epoch": 1.0376569037656904,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004994712508774757,
|
|
"loss": 5.6887,
|
|
"mean_token_accuracy": 0.15047468543052672,
|
|
"num_tokens": 2290958.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.7493171215057375,
|
|
"epoch": 1.0418410041841004,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499448905196997,
|
|
"loss": 5.6243,
|
|
"mean_token_accuracy": 0.16273448020219802,
|
|
"num_tokens": 2299763.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.8307215690612795,
|
|
"epoch": 1.0460251046025104,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004994260976367573,
|
|
"loss": 5.6484,
|
|
"mean_token_accuracy": 0.15789262801408768,
|
|
"num_tokens": 2309847.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 5.833254051208496,
|
|
"epoch": 1.0502092050209204,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004994028282436905,
|
|
"loss": 5.6662,
|
|
"mean_token_accuracy": 0.15037693381309508,
|
|
"num_tokens": 2317556.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 5.7204986095428465,
|
|
"epoch": 1.0543933054393306,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004993790970656814,
|
|
"loss": 5.6049,
|
|
"mean_token_accuracy": 0.15735123306512833,
|
|
"num_tokens": 2327151.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.79366283416748,
|
|
"epoch": 1.0585774058577406,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004993549041515652,
|
|
"loss": 5.6041,
|
|
"mean_token_accuracy": 0.15938671827316284,
|
|
"num_tokens": 2336119.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 5.739954948425293,
|
|
"epoch": 1.0627615062761506,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993302495511268,
|
|
"loss": 5.6338,
|
|
"mean_token_accuracy": 0.15804623067378998,
|
|
"num_tokens": 2345262.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.959032106399536,
|
|
"epoch": 1.0669456066945606,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004993051333151014,
|
|
"loss": 5.7792,
|
|
"mean_token_accuracy": 0.14950344413518907,
|
|
"num_tokens": 2355210.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.7098313331604,
|
|
"epoch": 1.0711297071129706,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004992795554951742,
|
|
"loss": 5.5224,
|
|
"mean_token_accuracy": 0.16097740083932877,
|
|
"num_tokens": 2364599.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.690215682983398,
|
|
"epoch": 1.0753138075313808,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004992535161439803,
|
|
"loss": 5.6963,
|
|
"mean_token_accuracy": 0.1504080779850483,
|
|
"num_tokens": 2374143.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 5.8944415092468265,
|
|
"epoch": 1.0794979079497908,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004992270153151045,
|
|
"loss": 5.6979,
|
|
"mean_token_accuracy": 0.15302433222532272,
|
|
"num_tokens": 2383155.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.880524778366089,
|
|
"epoch": 1.0836820083682008,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004992000530630812,
|
|
"loss": 5.7269,
|
|
"mean_token_accuracy": 0.1545798197388649,
|
|
"num_tokens": 2392542.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.726328802108765,
|
|
"epoch": 1.0878661087866108,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004991726294433943,
|
|
"loss": 5.7037,
|
|
"mean_token_accuracy": 0.1482452318072319,
|
|
"num_tokens": 2401851.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.814077091217041,
|
|
"epoch": 1.0920502092050208,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991447445124774,
|
|
"loss": 5.6339,
|
|
"mean_token_accuracy": 0.15687052309513091,
|
|
"num_tokens": 2410726.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.678286838531494,
|
|
"epoch": 1.096234309623431,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499116398327713,
|
|
"loss": 5.7169,
|
|
"mean_token_accuracy": 0.15115260183811188,
|
|
"num_tokens": 2420702.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.710619592666626,
|
|
"epoch": 1.100418410041841,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004990875909474331,
|
|
"loss": 5.6326,
|
|
"mean_token_accuracy": 0.15647043734788896,
|
|
"num_tokens": 2431335.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 5.850708532333374,
|
|
"epoch": 1.104602510460251,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004990583224309187,
|
|
"loss": 5.6258,
|
|
"mean_token_accuracy": 0.15880463123321534,
|
|
"num_tokens": 2440722.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.768325519561768,
|
|
"epoch": 1.108786610878661,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004990285928383997,
|
|
"loss": 5.6871,
|
|
"mean_token_accuracy": 0.1519317150115967,
|
|
"num_tokens": 2451230.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.805948209762573,
|
|
"epoch": 1.112970711297071,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989984022310545,
|
|
"loss": 5.7321,
|
|
"mean_token_accuracy": 0.1496009983122349,
|
|
"num_tokens": 2461442.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.755316114425659,
|
|
"epoch": 1.1171548117154813,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004989677506710109,
|
|
"loss": 5.5444,
|
|
"mean_token_accuracy": 0.16538602709770203,
|
|
"num_tokens": 2471381.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.739071464538574,
|
|
"epoch": 1.1213389121338913,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004989366382213449,
|
|
"loss": 5.5941,
|
|
"mean_token_accuracy": 0.161516273021698,
|
|
"num_tokens": 2479937.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.6999297618865965,
|
|
"epoch": 1.1255230125523012,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004989050649460806,
|
|
"loss": 5.5596,
|
|
"mean_token_accuracy": 0.16487280726432801,
|
|
"num_tokens": 2488974.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.751833486557007,
|
|
"epoch": 1.1297071129707112,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000498873030910191,
|
|
"loss": 5.716,
|
|
"mean_token_accuracy": 0.1517424389719963,
|
|
"num_tokens": 2498800.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.716541719436646,
|
|
"epoch": 1.1338912133891212,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004988405361795969,
|
|
"loss": 5.6171,
|
|
"mean_token_accuracy": 0.15509085208177567,
|
|
"num_tokens": 2506580.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.736952686309815,
|
|
"epoch": 1.1380753138075315,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004988075808211672,
|
|
"loss": 5.6768,
|
|
"mean_token_accuracy": 0.15938233137130736,
|
|
"num_tokens": 2516342.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.701758766174317,
|
|
"epoch": 1.1422594142259415,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987741649027189,
|
|
"loss": 5.7103,
|
|
"mean_token_accuracy": 0.1576656699180603,
|
|
"num_tokens": 2526054.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.816002082824707,
|
|
"epoch": 1.1464435146443515,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004987402884930163,
|
|
"loss": 5.6636,
|
|
"mean_token_accuracy": 0.15955882221460344,
|
|
"num_tokens": 2535556.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.758607339859009,
|
|
"epoch": 1.1506276150627615,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004987059516617717,
|
|
"loss": 5.6258,
|
|
"mean_token_accuracy": 0.1616874799132347,
|
|
"num_tokens": 2544930.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.615150737762451,
|
|
"epoch": 1.1548117154811715,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004986711544796449,
|
|
"loss": 5.534,
|
|
"mean_token_accuracy": 0.1632443681359291,
|
|
"num_tokens": 2554362.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.809719753265381,
|
|
"epoch": 1.1589958158995817,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000498635897018243,
|
|
"loss": 5.6457,
|
|
"mean_token_accuracy": 0.16341828256845475,
|
|
"num_tokens": 2563499.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.688316726684571,
|
|
"epoch": 1.1631799163179917,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004986001793501198,
|
|
"loss": 5.6671,
|
|
"mean_token_accuracy": 0.16028585135936738,
|
|
"num_tokens": 2573346.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.7574742317199705,
|
|
"epoch": 1.1673640167364017,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004985640015487768,
|
|
"loss": 5.5473,
|
|
"mean_token_accuracy": 0.16826957762241362,
|
|
"num_tokens": 2582541.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.6951111316680905,
|
|
"epoch": 1.1715481171548117,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004985273636886621,
|
|
"loss": 5.6295,
|
|
"mean_token_accuracy": 0.15948453098535537,
|
|
"num_tokens": 2592015.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.706931447982788,
|
|
"epoch": 1.1757322175732217,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004984902658451704,
|
|
"loss": 5.6428,
|
|
"mean_token_accuracy": 0.15934522002935408,
|
|
"num_tokens": 2601998.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.648833608627319,
|
|
"epoch": 1.1799163179916319,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004984527080946432,
|
|
"loss": 5.5765,
|
|
"mean_token_accuracy": 0.16093924939632415,
|
|
"num_tokens": 2611435.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.76911244392395,
|
|
"epoch": 1.1841004184100419,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004984146905143683,
|
|
"loss": 5.6767,
|
|
"mean_token_accuracy": 0.15309462398290635,
|
|
"num_tokens": 2621639.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.739169025421143,
|
|
"epoch": 1.1882845188284519,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004983762131825797,
|
|
"loss": 5.637,
|
|
"mean_token_accuracy": 0.15594995766878128,
|
|
"num_tokens": 2631390.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.711789560317993,
|
|
"epoch": 1.1924686192468619,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004983372761784576,
|
|
"loss": 5.689,
|
|
"mean_token_accuracy": 0.1508960619568825,
|
|
"num_tokens": 2640769.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.787328004837036,
|
|
"epoch": 1.196652719665272,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004982978795821281,
|
|
"loss": 5.6816,
|
|
"mean_token_accuracy": 0.15681966096162797,
|
|
"num_tokens": 2651137.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.771102666854858,
|
|
"epoch": 1.200836820083682,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004982580234746632,
|
|
"loss": 5.6053,
|
|
"mean_token_accuracy": 0.15788344889879227,
|
|
"num_tokens": 2660530.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.711075687408448,
|
|
"epoch": 1.205020920502092,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00049821770793808,
|
|
"loss": 5.6621,
|
|
"mean_token_accuracy": 0.15614335089921952,
|
|
"num_tokens": 2669541.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.7661858081817625,
|
|
"epoch": 1.209205020920502,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981769330553418,
|
|
"loss": 5.653,
|
|
"mean_token_accuracy": 0.159392286837101,
|
|
"num_tokens": 2678542.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.784830617904663,
|
|
"epoch": 1.213389121338912,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004981356989103566,
|
|
"loss": 5.7314,
|
|
"mean_token_accuracy": 0.1579895868897438,
|
|
"num_tokens": 2687720.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.78279218673706,
|
|
"epoch": 1.217573221757322,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004980940055879774,
|
|
"loss": 5.5907,
|
|
"mean_token_accuracy": 0.1579319342970848,
|
|
"num_tokens": 2697641.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.657995986938476,
|
|
"epoch": 1.2217573221757323,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004980518531740027,
|
|
"loss": 5.6364,
|
|
"mean_token_accuracy": 0.15723809450864792,
|
|
"num_tokens": 2706600.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.837217473983765,
|
|
"epoch": 1.2259414225941423,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004980092417551752,
|
|
"loss": 5.7534,
|
|
"mean_token_accuracy": 0.15348030626773834,
|
|
"num_tokens": 2716633.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.768991184234619,
|
|
"epoch": 1.2301255230125523,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004979661714191825,
|
|
"loss": 5.5118,
|
|
"mean_token_accuracy": 0.17512491196393967,
|
|
"num_tokens": 2724829.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.669136238098145,
|
|
"epoch": 1.2343096234309623,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004979226422546563,
|
|
"loss": 5.6315,
|
|
"mean_token_accuracy": 0.1603492885828018,
|
|
"num_tokens": 2734276.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.772823715209961,
|
|
"epoch": 1.2384937238493725,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004978786543511726,
|
|
"loss": 5.7359,
|
|
"mean_token_accuracy": 0.16275661140680314,
|
|
"num_tokens": 2744413.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.700097274780274,
|
|
"epoch": 1.2426778242677825,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004978342077992515,
|
|
"loss": 5.6097,
|
|
"mean_token_accuracy": 0.1594856157898903,
|
|
"num_tokens": 2753242.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.543384027481079,
|
|
"epoch": 1.2468619246861925,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004977893026903568,
|
|
"loss": 5.5702,
|
|
"mean_token_accuracy": 0.1632060170173645,
|
|
"num_tokens": 2761811.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.773969602584839,
|
|
"epoch": 1.2510460251046025,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004977439391168963,
|
|
"loss": 5.6585,
|
|
"mean_token_accuracy": 0.16232835799455642,
|
|
"num_tokens": 2770560.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.763203144073486,
|
|
"epoch": 1.2552301255230125,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004976981171722205,
|
|
"loss": 5.6505,
|
|
"mean_token_accuracy": 0.1572619527578354,
|
|
"num_tokens": 2780131.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2552301255230125,
|
|
"eval_entropy": 5.464953191441261,
|
|
"eval_loss": 5.795840740203857,
|
|
"eval_mean_token_accuracy": 0.1601959781412698,
|
|
"eval_num_tokens": 2780131.0,
|
|
"eval_runtime": 2.9552,
|
|
"eval_samples_per_second": 1322.064,
|
|
"eval_steps_per_second": 165.469,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.767412948608398,
|
|
"epoch": 1.2594142259414225,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976518369506237,
|
|
"loss": 5.7321,
|
|
"mean_token_accuracy": 0.1497885599732399,
|
|
"num_tokens": 2789319.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.6755859375,
|
|
"epoch": 1.2635983263598327,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004976050985473433,
|
|
"loss": 5.6321,
|
|
"mean_token_accuracy": 0.15301928967237471,
|
|
"num_tokens": 2799253.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.726134300231934,
|
|
"epoch": 1.2677824267782427,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004975579020585593,
|
|
"loss": 5.7317,
|
|
"mean_token_accuracy": 0.1520630419254303,
|
|
"num_tokens": 2808968.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.82230772972107,
|
|
"epoch": 1.2719665271966527,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004975102475813947,
|
|
"loss": 5.5998,
|
|
"mean_token_accuracy": 0.16555387377738953,
|
|
"num_tokens": 2817282.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.64022912979126,
|
|
"epoch": 1.2761506276150627,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004974621352139144,
|
|
"loss": 5.6673,
|
|
"mean_token_accuracy": 0.15914613455533982,
|
|
"num_tokens": 2826395.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.74616584777832,
|
|
"epoch": 1.280334728033473,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004974135650551262,
|
|
"loss": 5.745,
|
|
"mean_token_accuracy": 0.15512988716363907,
|
|
"num_tokens": 2836195.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.736653757095337,
|
|
"epoch": 1.284518828451883,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004973645372049796,
|
|
"loss": 5.5781,
|
|
"mean_token_accuracy": 0.1622034192085266,
|
|
"num_tokens": 2846020.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.60059118270874,
|
|
"epoch": 1.288702928870293,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004973150517643659,
|
|
"loss": 5.5332,
|
|
"mean_token_accuracy": 0.17065534740686417,
|
|
"num_tokens": 2854895.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.681558084487915,
|
|
"epoch": 1.292887029288703,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004972651088351185,
|
|
"loss": 5.5545,
|
|
"mean_token_accuracy": 0.16582695692777633,
|
|
"num_tokens": 2864491.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.739282846450806,
|
|
"epoch": 1.297071129707113,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004972147085200116,
|
|
"loss": 5.5909,
|
|
"mean_token_accuracy": 0.1584654465317726,
|
|
"num_tokens": 2873962.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.703140783309936,
|
|
"epoch": 1.301255230125523,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004971638509227611,
|
|
"loss": 5.6258,
|
|
"mean_token_accuracy": 0.16314564943313598,
|
|
"num_tokens": 2883530.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.658680057525634,
|
|
"epoch": 1.3054393305439331,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004971125361480236,
|
|
"loss": 5.611,
|
|
"mean_token_accuracy": 0.15674630105495452,
|
|
"num_tokens": 2892867.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.679305553436279,
|
|
"epoch": 1.3096234309623431,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004970607643013969,
|
|
"loss": 5.5824,
|
|
"mean_token_accuracy": 0.16301058530807494,
|
|
"num_tokens": 2902320.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.562259578704834,
|
|
"epoch": 1.3138075313807531,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000497008535489419,
|
|
"loss": 5.5175,
|
|
"mean_token_accuracy": 0.16897710114717485,
|
|
"num_tokens": 2911390.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.681601572036743,
|
|
"epoch": 1.3179916317991631,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004969558498195684,
|
|
"loss": 5.5498,
|
|
"mean_token_accuracy": 0.1685500845313072,
|
|
"num_tokens": 2921295.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.6406128883361815,
|
|
"epoch": 1.3221757322175733,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004969027074002638,
|
|
"loss": 5.5849,
|
|
"mean_token_accuracy": 0.16301420331001282,
|
|
"num_tokens": 2931328.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.615828609466552,
|
|
"epoch": 1.3263598326359833,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968491083408636,
|
|
"loss": 5.533,
|
|
"mean_token_accuracy": 0.16588781774044037,
|
|
"num_tokens": 2940193.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.562009859085083,
|
|
"epoch": 1.3305439330543933,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004967950527516663,
|
|
"loss": 5.5643,
|
|
"mean_token_accuracy": 0.1703370675444603,
|
|
"num_tokens": 2949759.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.608451700210571,
|
|
"epoch": 1.3347280334728033,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004967405407439092,
|
|
"loss": 5.4901,
|
|
"mean_token_accuracy": 0.16004174947738647,
|
|
"num_tokens": 2959185.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.7124756336212155,
|
|
"epoch": 1.3389121338912133,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004966855724297698,
|
|
"loss": 5.6403,
|
|
"mean_token_accuracy": 0.1580813243985176,
|
|
"num_tokens": 2968945.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.666547632217407,
|
|
"epoch": 1.3430962343096233,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004966301479223637,
|
|
"loss": 5.5802,
|
|
"mean_token_accuracy": 0.1660241439938545,
|
|
"num_tokens": 2978395.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.5671368598937985,
|
|
"epoch": 1.3472803347280335,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004965742673357456,
|
|
"loss": 5.498,
|
|
"mean_token_accuracy": 0.17463006377220153,
|
|
"num_tokens": 2987030.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.7067772388458256,
|
|
"epoch": 1.3514644351464435,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004965179307849088,
|
|
"loss": 5.55,
|
|
"mean_token_accuracy": 0.1639631822705269,
|
|
"num_tokens": 2995736.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.665324544906616,
|
|
"epoch": 1.3556485355648535,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000496461138385785,
|
|
"loss": 5.6643,
|
|
"mean_token_accuracy": 0.1616310253739357,
|
|
"num_tokens": 3005105.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.608656644821167,
|
|
"epoch": 1.3598326359832635,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004964038902552438,
|
|
"loss": 5.5831,
|
|
"mean_token_accuracy": 0.16028797775506973,
|
|
"num_tokens": 3013857.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.674657392501831,
|
|
"epoch": 1.3640167364016738,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004963461865110926,
|
|
"loss": 5.6055,
|
|
"mean_token_accuracy": 0.1662692293524742,
|
|
"num_tokens": 3023972.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.584111547470092,
|
|
"epoch": 1.3682008368200838,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004962880272720764,
|
|
"loss": 5.578,
|
|
"mean_token_accuracy": 0.16646245121955872,
|
|
"num_tokens": 3032705.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.674110078811646,
|
|
"epoch": 1.3723849372384938,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004962294126578775,
|
|
"loss": 5.5043,
|
|
"mean_token_accuracy": 0.1671427831053734,
|
|
"num_tokens": 3040900.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.6051475524902346,
|
|
"epoch": 1.3765690376569037,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004961703427891154,
|
|
"loss": 5.5671,
|
|
"mean_token_accuracy": 0.16518825590610503,
|
|
"num_tokens": 3050477.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.678053855895996,
|
|
"epoch": 1.3807531380753137,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004961108177873466,
|
|
"loss": 5.6266,
|
|
"mean_token_accuracy": 0.15941449105739594,
|
|
"num_tokens": 3059743.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.6124502658844,
|
|
"epoch": 1.3849372384937237,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004960508377750638,
|
|
"loss": 5.414,
|
|
"mean_token_accuracy": 0.17518963068723678,
|
|
"num_tokens": 3067801.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.540145063400269,
|
|
"epoch": 1.389121338912134,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004959904028756961,
|
|
"loss": 5.6153,
|
|
"mean_token_accuracy": 0.1577400654554367,
|
|
"num_tokens": 3077404.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.669542837142944,
|
|
"epoch": 1.393305439330544,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004959295132136091,
|
|
"loss": 5.636,
|
|
"mean_token_accuracy": 0.1611045181751251,
|
|
"num_tokens": 3087130.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.718611812591552,
|
|
"epoch": 1.397489539748954,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004958681689141036,
|
|
"loss": 5.6588,
|
|
"mean_token_accuracy": 0.15776030272245406,
|
|
"num_tokens": 3096038.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.6009135246276855,
|
|
"epoch": 1.401673640167364,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004958063701034167,
|
|
"loss": 5.5032,
|
|
"mean_token_accuracy": 0.16299725770950318,
|
|
"num_tokens": 3105668.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.535477876663208,
|
|
"epoch": 1.4058577405857742,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00049574411690872,
|
|
"loss": 5.4834,
|
|
"mean_token_accuracy": 0.162048602104187,
|
|
"num_tokens": 3114020.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.5072493076324465,
|
|
"epoch": 1.4100418410041842,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004956814094581208,
|
|
"loss": 5.5311,
|
|
"mean_token_accuracy": 0.16476096361875534,
|
|
"num_tokens": 3124096.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.711183834075928,
|
|
"epoch": 1.4142259414225942,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000495618247880661,
|
|
"loss": 5.5978,
|
|
"mean_token_accuracy": 0.16633668392896653,
|
|
"num_tokens": 3132806.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.563244771957398,
|
|
"epoch": 1.4184100418410042,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004955546323063168,
|
|
"loss": 5.5167,
|
|
"mean_token_accuracy": 0.17463642209768296,
|
|
"num_tokens": 3142001.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.551546478271485,
|
|
"epoch": 1.4225941422594142,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004954905628659989,
|
|
"loss": 5.4947,
|
|
"mean_token_accuracy": 0.1648888036608696,
|
|
"num_tokens": 3151658.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.600251007080078,
|
|
"epoch": 1.4267782426778242,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004954260396915521,
|
|
"loss": 5.5203,
|
|
"mean_token_accuracy": 0.16872449070215226,
|
|
"num_tokens": 3159573.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.392547225952148,
|
|
"epoch": 1.4309623430962344,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004953610629157544,
|
|
"loss": 5.4006,
|
|
"mean_token_accuracy": 0.17204391062259675,
|
|
"num_tokens": 3168312.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.642241954803467,
|
|
"epoch": 1.4351464435146444,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004952956326723177,
|
|
"loss": 5.4875,
|
|
"mean_token_accuracy": 0.17641883194446564,
|
|
"num_tokens": 3176811.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.633111524581909,
|
|
"epoch": 1.4393305439330544,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000495229749095887,
|
|
"loss": 5.5898,
|
|
"mean_token_accuracy": 0.16307459473609925,
|
|
"num_tokens": 3186599.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.6868643283844,
|
|
"epoch": 1.4435146443514644,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00049516341232204,
|
|
"loss": 5.683,
|
|
"mean_token_accuracy": 0.16141854375600814,
|
|
"num_tokens": 3195905.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.687935781478882,
|
|
"epoch": 1.4476987447698746,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000495096622487287,
|
|
"loss": 5.6378,
|
|
"mean_token_accuracy": 0.16340560615062713,
|
|
"num_tokens": 3205132.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.67728853225708,
|
|
"epoch": 1.4518828451882846,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000495029379729071,
|
|
"loss": 5.6515,
|
|
"mean_token_accuracy": 0.15281365364789962,
|
|
"num_tokens": 3214362.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.6806377410888675,
|
|
"epoch": 1.4560669456066946,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004949616841857666,
|
|
"loss": 5.6356,
|
|
"mean_token_accuracy": 0.16245218068361283,
|
|
"num_tokens": 3224248.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.592656803131104,
|
|
"epoch": 1.4602510460251046,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004948935359966806,
|
|
"loss": 5.516,
|
|
"mean_token_accuracy": 0.16721477955579758,
|
|
"num_tokens": 3233416.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.494255018234253,
|
|
"epoch": 1.4644351464435146,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004948249353020506,
|
|
"loss": 5.4576,
|
|
"mean_token_accuracy": 0.17374701350927352,
|
|
"num_tokens": 3242354.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.624186038970947,
|
|
"epoch": 1.4686192468619246,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004947558822430461,
|
|
"loss": 5.5378,
|
|
"mean_token_accuracy": 0.16421912908554076,
|
|
"num_tokens": 3251026.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.604061794281006,
|
|
"epoch": 1.4728033472803348,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004946863769617672,
|
|
"loss": 5.4957,
|
|
"mean_token_accuracy": 0.17072140723466872,
|
|
"num_tokens": 3259429.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.567508888244629,
|
|
"epoch": 1.4769874476987448,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004946164196012445,
|
|
"loss": 5.5426,
|
|
"mean_token_accuracy": 0.16655807942152023,
|
|
"num_tokens": 3269305.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.579040145874023,
|
|
"epoch": 1.4811715481171548,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004945460103054389,
|
|
"loss": 5.6372,
|
|
"mean_token_accuracy": 0.15401357561349868,
|
|
"num_tokens": 3278846.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.596373271942139,
|
|
"epoch": 1.4853556485355648,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004944751492192416,
|
|
"loss": 5.5364,
|
|
"mean_token_accuracy": 0.1669298455119133,
|
|
"num_tokens": 3287200.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.698515605926514,
|
|
"epoch": 1.489539748953975,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004944038364884734,
|
|
"loss": 5.5618,
|
|
"mean_token_accuracy": 0.16433424800634383,
|
|
"num_tokens": 3296651.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.533484077453613,
|
|
"epoch": 1.493723849372385,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004943320722598842,
|
|
"loss": 5.419,
|
|
"mean_token_accuracy": 0.17311303615570067,
|
|
"num_tokens": 3304887.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.577448987960816,
|
|
"epoch": 1.497907949790795,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004942598566811532,
|
|
"loss": 5.461,
|
|
"mean_token_accuracy": 0.17003400772809982,
|
|
"num_tokens": 3313925.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.551423788070679,
|
|
"epoch": 1.502092050209205,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004941871899008887,
|
|
"loss": 5.4594,
|
|
"mean_token_accuracy": 0.17227572649717332,
|
|
"num_tokens": 3322943.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.5156848430633545,
|
|
"epoch": 1.506276150627615,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000494114072068627,
|
|
"loss": 5.4842,
|
|
"mean_token_accuracy": 0.1615619406104088,
|
|
"num_tokens": 3332742.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.572016906738281,
|
|
"epoch": 1.510460251046025,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004940405033348332,
|
|
"loss": 5.5707,
|
|
"mean_token_accuracy": 0.17072367221117019,
|
|
"num_tokens": 3342228.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.572167015075683,
|
|
"epoch": 1.514644351464435,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004939664838508996,
|
|
"loss": 5.5404,
|
|
"mean_token_accuracy": 0.16432256549596785,
|
|
"num_tokens": 3351854.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.558309173583984,
|
|
"epoch": 1.5188284518828452,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004938920137691465,
|
|
"loss": 5.5147,
|
|
"mean_token_accuracy": 0.16797013580799103,
|
|
"num_tokens": 3360809.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.593948125839233,
|
|
"epoch": 1.5230125523012552,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004938170932428214,
|
|
"loss": 5.549,
|
|
"mean_token_accuracy": 0.16172343194484712,
|
|
"num_tokens": 3371033.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.599111843109131,
|
|
"epoch": 1.5271966527196654,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004937417224260984,
|
|
"loss": 5.4259,
|
|
"mean_token_accuracy": 0.17052858769893647,
|
|
"num_tokens": 3379943.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.475646877288819,
|
|
"epoch": 1.5313807531380754,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004936659014740791,
|
|
"loss": 5.5655,
|
|
"mean_token_accuracy": 0.17111508399248124,
|
|
"num_tokens": 3389316.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.668122625350952,
|
|
"epoch": 1.5355648535564854,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004935896305427904,
|
|
"loss": 5.5923,
|
|
"mean_token_accuracy": 0.15884766727685928,
|
|
"num_tokens": 3398914.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.7507908821105955,
|
|
"epoch": 1.5397489539748954,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004935129097891857,
|
|
"loss": 5.559,
|
|
"mean_token_accuracy": 0.16218522191047668,
|
|
"num_tokens": 3407967.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.5997473239898685,
|
|
"epoch": 1.5439330543933054,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000493435739371144,
|
|
"loss": 5.6008,
|
|
"mean_token_accuracy": 0.16404648274183273,
|
|
"num_tokens": 3416530.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.561355066299439,
|
|
"epoch": 1.5481171548117154,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004933581194474695,
|
|
"loss": 5.4387,
|
|
"mean_token_accuracy": 0.16938610821962358,
|
|
"num_tokens": 3425762.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.534957981109619,
|
|
"epoch": 1.5523012552301254,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004932800501778918,
|
|
"loss": 5.5416,
|
|
"mean_token_accuracy": 0.16275594830513002,
|
|
"num_tokens": 3434799.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.631038427352905,
|
|
"epoch": 1.5564853556485354,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004932015317230646,
|
|
"loss": 5.5531,
|
|
"mean_token_accuracy": 0.16585518270730973,
|
|
"num_tokens": 3442986.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.599202108383179,
|
|
"epoch": 1.5606694560669456,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004931225642445665,
|
|
"loss": 5.5819,
|
|
"mean_token_accuracy": 0.16394275426864624,
|
|
"num_tokens": 3452914.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.555068254470825,
|
|
"epoch": 1.5648535564853556,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004930431479048998,
|
|
"loss": 5.4174,
|
|
"mean_token_accuracy": 0.17184423357248307,
|
|
"num_tokens": 3462093.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.551028251647949,
|
|
"epoch": 1.5690376569037658,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004929632828674904,
|
|
"loss": 5.6183,
|
|
"mean_token_accuracy": 0.161507810652256,
|
|
"num_tokens": 3471259.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.583326292037964,
|
|
"epoch": 1.5732217573221758,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004928829692966879,
|
|
"loss": 5.5311,
|
|
"mean_token_accuracy": 0.17438520193099977,
|
|
"num_tokens": 3480071.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.7039947509765625,
|
|
"epoch": 1.5774058577405858,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004928022073577646,
|
|
"loss": 5.576,
|
|
"mean_token_accuracy": 0.17246243208646775,
|
|
"num_tokens": 3489075.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.533627223968506,
|
|
"epoch": 1.5815899581589958,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004927209972169157,
|
|
"loss": 5.4171,
|
|
"mean_token_accuracy": 0.1735706001520157,
|
|
"num_tokens": 3498495.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.528741264343262,
|
|
"epoch": 1.5857740585774058,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004926393390412586,
|
|
"loss": 5.5479,
|
|
"mean_token_accuracy": 0.16164298355579376,
|
|
"num_tokens": 3508590.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.65437331199646,
|
|
"epoch": 1.5899581589958158,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004925572329988325,
|
|
"loss": 5.5718,
|
|
"mean_token_accuracy": 0.15928141623735428,
|
|
"num_tokens": 3518397.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.5337958335876465,
|
|
"epoch": 1.5941422594142258,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004924746792585987,
|
|
"loss": 5.5356,
|
|
"mean_token_accuracy": 0.17011555880308152,
|
|
"num_tokens": 3526620.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.680136203765869,
|
|
"epoch": 1.5983263598326358,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004923916779904394,
|
|
"loss": 5.5558,
|
|
"mean_token_accuracy": 0.16798217594623566,
|
|
"num_tokens": 3535534.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.505680274963379,
|
|
"epoch": 1.602510460251046,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004923082293651578,
|
|
"loss": 5.5216,
|
|
"mean_token_accuracy": 0.1653604969382286,
|
|
"num_tokens": 3544771.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.5642822742462155,
|
|
"epoch": 1.606694560669456,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000492224333554478,
|
|
"loss": 5.4819,
|
|
"mean_token_accuracy": 0.16008352041244506,
|
|
"num_tokens": 3553970.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.522343111038208,
|
|
"epoch": 1.6108786610878663,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004921399907310437,
|
|
"loss": 5.39,
|
|
"mean_token_accuracy": 0.1683530405163765,
|
|
"num_tokens": 3563116.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.489598798751831,
|
|
"epoch": 1.6150627615062763,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004920552010684191,
|
|
"loss": 5.5231,
|
|
"mean_token_accuracy": 0.16716862767934798,
|
|
"num_tokens": 3571605.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.526372814178467,
|
|
"epoch": 1.6192468619246863,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004919699647410878,
|
|
"loss": 5.6229,
|
|
"mean_token_accuracy": 0.16075178384780883,
|
|
"num_tokens": 3581837.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.63598747253418,
|
|
"epoch": 1.6234309623430963,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004918842819244523,
|
|
"loss": 5.5645,
|
|
"mean_token_accuracy": 0.15917611569166185,
|
|
"num_tokens": 3590853.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.517759466171265,
|
|
"epoch": 1.6276150627615062,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004917981527948342,
|
|
"loss": 5.4493,
|
|
"mean_token_accuracy": 0.169613578915596,
|
|
"num_tokens": 3599708.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.499824714660645,
|
|
"epoch": 1.6317991631799162,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004917115775294731,
|
|
"loss": 5.4722,
|
|
"mean_token_accuracy": 0.16738361418247222,
|
|
"num_tokens": 3608634.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.594522762298584,
|
|
"epoch": 1.6359832635983262,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004916245563065273,
|
|
"loss": 5.5479,
|
|
"mean_token_accuracy": 0.16360496133565902,
|
|
"num_tokens": 3618776.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.53354115486145,
|
|
"epoch": 1.6401673640167362,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004915370893050722,
|
|
"loss": 5.552,
|
|
"mean_token_accuracy": 0.16077069342136383,
|
|
"num_tokens": 3628595.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.691362142562866,
|
|
"epoch": 1.6443514644351465,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004914491767051012,
|
|
"loss": 5.6676,
|
|
"mean_token_accuracy": 0.15586235821247102,
|
|
"num_tokens": 3638133.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.534716415405273,
|
|
"epoch": 1.6485355648535565,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004913608186875241,
|
|
"loss": 5.452,
|
|
"mean_token_accuracy": 0.17007116377353668,
|
|
"num_tokens": 3646679.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.596976804733276,
|
|
"epoch": 1.6527196652719667,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004912720154341674,
|
|
"loss": 5.5778,
|
|
"mean_token_accuracy": 0.16362767666578293,
|
|
"num_tokens": 3656131.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.611526536941528,
|
|
"epoch": 1.6569037656903767,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004911827671277739,
|
|
"loss": 5.4668,
|
|
"mean_token_accuracy": 0.1692266970872879,
|
|
"num_tokens": 3665187.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.548706722259522,
|
|
"epoch": 1.6610878661087867,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004910930739520025,
|
|
"loss": 5.5307,
|
|
"mean_token_accuracy": 0.16843136101961137,
|
|
"num_tokens": 3674649.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.616735553741455,
|
|
"epoch": 1.6652719665271967,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004910029360914271,
|
|
"loss": 5.5674,
|
|
"mean_token_accuracy": 0.15753058046102525,
|
|
"num_tokens": 3684680.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.50072021484375,
|
|
"epoch": 1.6694560669456067,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000490912353731537,
|
|
"loss": 5.4958,
|
|
"mean_token_accuracy": 0.17296883910894395,
|
|
"num_tokens": 3693560.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.554805278778076,
|
|
"epoch": 1.6736401673640167,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004908213270587363,
|
|
"loss": 5.4829,
|
|
"mean_token_accuracy": 0.16652763932943343,
|
|
"num_tokens": 3702632.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6736401673640167,
|
|
"eval_entropy": 5.440355615869378,
|
|
"eval_loss": 5.629227161407471,
|
|
"eval_mean_token_accuracy": 0.17036509641840414,
|
|
"eval_num_tokens": 3702632.0,
|
|
"eval_runtime": 2.7412,
|
|
"eval_samples_per_second": 1425.293,
|
|
"eval_steps_per_second": 178.39,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.490265941619873,
|
|
"epoch": 1.6778242677824267,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004907298562603431,
|
|
"loss": 5.3693,
|
|
"mean_token_accuracy": 0.1780234083533287,
|
|
"num_tokens": 3711022.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.519587564468384,
|
|
"epoch": 1.6820083682008367,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004906379415245896,
|
|
"loss": 5.4912,
|
|
"mean_token_accuracy": 0.1682968758046627,
|
|
"num_tokens": 3719778.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.481550216674805,
|
|
"epoch": 1.6861924686192469,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004905455830406215,
|
|
"loss": 5.4789,
|
|
"mean_token_accuracy": 0.16773745119571687,
|
|
"num_tokens": 3728773.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.569124746322632,
|
|
"epoch": 1.6903765690376569,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004904527809984978,
|
|
"loss": 5.5196,
|
|
"mean_token_accuracy": 0.17281341999769212,
|
|
"num_tokens": 3737781.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.612848854064941,
|
|
"epoch": 1.694560669456067,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004903595355891902,
|
|
"loss": 5.561,
|
|
"mean_token_accuracy": 0.16299786269664765,
|
|
"num_tokens": 3748010.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.534227752685547,
|
|
"epoch": 1.698744769874477,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004902658470045826,
|
|
"loss": 5.5441,
|
|
"mean_token_accuracy": 0.16640205830335617,
|
|
"num_tokens": 3757894.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.637985038757324,
|
|
"epoch": 1.702928870292887,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004901717154374713,
|
|
"loss": 5.6429,
|
|
"mean_token_accuracy": 0.16354548633098603,
|
|
"num_tokens": 3768760.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.5542093276977536,
|
|
"epoch": 1.707112970711297,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004900771410815638,
|
|
"loss": 5.4831,
|
|
"mean_token_accuracy": 0.17497676610946655,
|
|
"num_tokens": 3778273.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.4441554069519045,
|
|
"epoch": 1.711297071129707,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004899821241314788,
|
|
"loss": 5.4908,
|
|
"mean_token_accuracy": 0.17085491120815277,
|
|
"num_tokens": 3787012.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.535651969909668,
|
|
"epoch": 1.715481171548117,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000489886664782746,
|
|
"loss": 5.434,
|
|
"mean_token_accuracy": 0.1750217691063881,
|
|
"num_tokens": 3795144.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.624910879135132,
|
|
"epoch": 1.719665271966527,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004897907632318055,
|
|
"loss": 5.6586,
|
|
"mean_token_accuracy": 0.15991874784231186,
|
|
"num_tokens": 3804534.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.514365386962891,
|
|
"epoch": 1.723849372384937,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004896944196760071,
|
|
"loss": 5.4353,
|
|
"mean_token_accuracy": 0.1699859157204628,
|
|
"num_tokens": 3813757.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.443351697921753,
|
|
"epoch": 1.7280334728033473,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004895976343136106,
|
|
"loss": 5.4811,
|
|
"mean_token_accuracy": 0.17176128923892975,
|
|
"num_tokens": 3823157.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.447935390472412,
|
|
"epoch": 1.7322175732217573,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004895004073437844,
|
|
"loss": 5.3619,
|
|
"mean_token_accuracy": 0.17392124831676484,
|
|
"num_tokens": 3831661.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.408009672164917,
|
|
"epoch": 1.7364016736401675,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004894027389666061,
|
|
"loss": 5.3646,
|
|
"mean_token_accuracy": 0.1749645084142685,
|
|
"num_tokens": 3840591.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.539536762237549,
|
|
"epoch": 1.7405857740585775,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004893046293830616,
|
|
"loss": 5.4868,
|
|
"mean_token_accuracy": 0.1652049645781517,
|
|
"num_tokens": 3850094.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.521003341674804,
|
|
"epoch": 1.7447698744769875,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004892060787950447,
|
|
"loss": 5.5438,
|
|
"mean_token_accuracy": 0.16726065725088118,
|
|
"num_tokens": 3858973.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.3966960430145265,
|
|
"epoch": 1.7489539748953975,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004891070874053564,
|
|
"loss": 5.3147,
|
|
"mean_token_accuracy": 0.17629740983247758,
|
|
"num_tokens": 3868315.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.500430917739868,
|
|
"epoch": 1.7531380753138075,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004890076554177051,
|
|
"loss": 5.4802,
|
|
"mean_token_accuracy": 0.1695767819881439,
|
|
"num_tokens": 3877605.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.587614965438843,
|
|
"epoch": 1.7573221757322175,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004889077830367061,
|
|
"loss": 5.4894,
|
|
"mean_token_accuracy": 0.17201679646968843,
|
|
"num_tokens": 3886998.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.531196069717407,
|
|
"epoch": 1.7615062761506275,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004888074704678806,
|
|
"loss": 5.4475,
|
|
"mean_token_accuracy": 0.17462632656097413,
|
|
"num_tokens": 3896018.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.514407539367676,
|
|
"epoch": 1.7656903765690377,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004887067179176558,
|
|
"loss": 5.4467,
|
|
"mean_token_accuracy": 0.17259093672037124,
|
|
"num_tokens": 3904957.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.503034543991089,
|
|
"epoch": 1.7698744769874477,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004886055255933639,
|
|
"loss": 5.5571,
|
|
"mean_token_accuracy": 0.17202653884887695,
|
|
"num_tokens": 3913427.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.539950847625732,
|
|
"epoch": 1.7740585774058577,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004885038937032429,
|
|
"loss": 5.4603,
|
|
"mean_token_accuracy": 0.17139968723058702,
|
|
"num_tokens": 3923134.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.52648515701294,
|
|
"epoch": 1.778242677824268,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004884018224564345,
|
|
"loss": 5.3742,
|
|
"mean_token_accuracy": 0.1694469630718231,
|
|
"num_tokens": 3931867.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.444571542739868,
|
|
"epoch": 1.782426778242678,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004882993120629852,
|
|
"loss": 5.4804,
|
|
"mean_token_accuracy": 0.1715375080704689,
|
|
"num_tokens": 3941057.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.659233045578003,
|
|
"epoch": 1.786610878661088,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004881963627338449,
|
|
"loss": 5.5931,
|
|
"mean_token_accuracy": 0.1629413604736328,
|
|
"num_tokens": 3950511.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.51923770904541,
|
|
"epoch": 1.790794979079498,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048809297468086654,
|
|
"loss": 5.5011,
|
|
"mean_token_accuracy": 0.17307366877794267,
|
|
"num_tokens": 3959769.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.521024179458618,
|
|
"epoch": 1.794979079497908,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004879891481168063,
|
|
"loss": 5.4298,
|
|
"mean_token_accuracy": 0.17547144293785094,
|
|
"num_tokens": 3968462.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.403447103500366,
|
|
"epoch": 1.799163179916318,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048788488325532235,
|
|
"loss": 5.4883,
|
|
"mean_token_accuracy": 0.17039523422718048,
|
|
"num_tokens": 3977464.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.532100677490234,
|
|
"epoch": 1.803347280334728,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004877801803109753,
|
|
"loss": 5.3911,
|
|
"mean_token_accuracy": 0.17889024466276168,
|
|
"num_tokens": 3986303.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.581285095214843,
|
|
"epoch": 1.8075313807531381,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048767503949922683,
|
|
"loss": 5.4245,
|
|
"mean_token_accuracy": 0.17356548756361007,
|
|
"num_tokens": 3994970.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.5172028064727785,
|
|
"epoch": 1.8117154811715481,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004875694610364398,
|
|
"loss": 5.5307,
|
|
"mean_token_accuracy": 0.16971690058708191,
|
|
"num_tokens": 4004679.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.642525959014892,
|
|
"epoch": 1.8158995815899581,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048746344513987775,
|
|
"loss": 5.5421,
|
|
"mean_token_accuracy": 0.16614817827939987,
|
|
"num_tokens": 4013516.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.51641149520874,
|
|
"epoch": 1.8200836820083683,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004873569920277043,
|
|
"loss": 5.3966,
|
|
"mean_token_accuracy": 0.17818442732095718,
|
|
"num_tokens": 4022335.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.389952516555786,
|
|
"epoch": 1.8242677824267783,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048725010191898296,
|
|
"loss": 5.3969,
|
|
"mean_token_accuracy": 0.17473374754190446,
|
|
"num_tokens": 4031295.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.573140096664429,
|
|
"epoch": 1.8284518828451883,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048714277503367627,
|
|
"loss": 5.6013,
|
|
"mean_token_accuracy": 0.1672340750694275,
|
|
"num_tokens": 4040378.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.609200382232666,
|
|
"epoch": 1.8326359832635983,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048703501159264576,
|
|
"loss": 5.4727,
|
|
"mean_token_accuracy": 0.16173548102378846,
|
|
"num_tokens": 4050217.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.3808427333831785,
|
|
"epoch": 1.8368200836820083,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004869268118176512,
|
|
"loss": 5.3533,
|
|
"mean_token_accuracy": 0.16994479298591614,
|
|
"num_tokens": 4059088.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.546787691116333,
|
|
"epoch": 1.8410041841004183,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048681817593135046,
|
|
"loss": 5.4393,
|
|
"mean_token_accuracy": 0.18028466552495956,
|
|
"num_tokens": 4067609.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.52881498336792,
|
|
"epoch": 1.8451882845188283,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048670910415729855,
|
|
"loss": 5.4406,
|
|
"mean_token_accuracy": 0.17771714627742768,
|
|
"num_tokens": 4076763.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.521932554244995,
|
|
"epoch": 1.8493723849372385,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048659959671994786,
|
|
"loss": 5.5957,
|
|
"mean_token_accuracy": 0.16645434647798538,
|
|
"num_tokens": 4086552.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.419884538650512,
|
|
"epoch": 1.8535564853556485,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048648965384464693,
|
|
"loss": 5.3411,
|
|
"mean_token_accuracy": 0.17399129271507263,
|
|
"num_tokens": 4096200.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.407555675506591,
|
|
"epoch": 1.8577405857740585,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048637927575764057,
|
|
"loss": 5.4131,
|
|
"mean_token_accuracy": 0.17483898252248764,
|
|
"num_tokens": 4105080.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.535524129867554,
|
|
"epoch": 1.8619246861924688,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048626846268606913,
|
|
"loss": 5.3662,
|
|
"mean_token_accuracy": 0.17205767184495926,
|
|
"num_tokens": 4113972.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.554711198806762,
|
|
"epoch": 1.8661087866108788,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048615721485796803,
|
|
"loss": 5.5942,
|
|
"mean_token_accuracy": 0.16348322480916977,
|
|
"num_tokens": 4123615.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.540109729766845,
|
|
"epoch": 1.8702928870292888,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004860455325022676,
|
|
"loss": 5.3667,
|
|
"mean_token_accuracy": 0.1732967182993889,
|
|
"num_tokens": 4132128.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.565926122665405,
|
|
"epoch": 1.8744769874476988,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000485933415848792,
|
|
"loss": 5.5737,
|
|
"mean_token_accuracy": 0.17058898657560348,
|
|
"num_tokens": 4141575.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.537665557861328,
|
|
"epoch": 1.8786610878661087,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004858208651282593,
|
|
"loss": 5.4133,
|
|
"mean_token_accuracy": 0.17372779101133345,
|
|
"num_tokens": 4150372.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.4735640525817875,
|
|
"epoch": 1.8828451882845187,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000485707880572281,
|
|
"loss": 5.519,
|
|
"mean_token_accuracy": 0.1760459750890732,
|
|
"num_tokens": 4159007.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.5536839962005615,
|
|
"epoch": 1.8870292887029287,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004855944624133608,
|
|
"loss": 5.5853,
|
|
"mean_token_accuracy": 0.16366994976997376,
|
|
"num_tokens": 4168646.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.441170740127563,
|
|
"epoch": 1.891213389121339,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048548061088489523,
|
|
"loss": 5.467,
|
|
"mean_token_accuracy": 0.16810427308082582,
|
|
"num_tokens": 4177979.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.556855058670044,
|
|
"epoch": 1.895397489539749,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048536632622117246,
|
|
"loss": 5.3997,
|
|
"mean_token_accuracy": 0.175626739859581,
|
|
"num_tokens": 4186781.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.451776933670044,
|
|
"epoch": 1.899581589958159,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048525160865737186,
|
|
"loss": 5.4549,
|
|
"mean_token_accuracy": 0.16997883915901185,
|
|
"num_tokens": 4195661.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.4366716861724855,
|
|
"epoch": 1.9037656903765692,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048513645842956384,
|
|
"loss": 5.3577,
|
|
"mean_token_accuracy": 0.17414591908454896,
|
|
"num_tokens": 4203902.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.443609809875488,
|
|
"epoch": 1.9079497907949792,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048502087577470895,
|
|
"loss": 5.4229,
|
|
"mean_token_accuracy": 0.1694953128695488,
|
|
"num_tokens": 4213663.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.557037734985352,
|
|
"epoch": 1.9121338912133892,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048490486093065774,
|
|
"loss": 5.469,
|
|
"mean_token_accuracy": 0.17493365705013275,
|
|
"num_tokens": 4222152.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.497901725769043,
|
|
"epoch": 1.9163179916317992,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048478841413615017,
|
|
"loss": 5.4833,
|
|
"mean_token_accuracy": 0.1730593830347061,
|
|
"num_tokens": 4232347.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.565075778961182,
|
|
"epoch": 1.9205020920502092,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048467153563081487,
|
|
"loss": 5.5514,
|
|
"mean_token_accuracy": 0.16933674961328507,
|
|
"num_tokens": 4240607.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.54401798248291,
|
|
"epoch": 1.9246861924686192,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004845542256551692,
|
|
"loss": 5.56,
|
|
"mean_token_accuracy": 0.16783103942871094,
|
|
"num_tokens": 4251131.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.5538262844085695,
|
|
"epoch": 1.9288702928870292,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004844364844506181,
|
|
"loss": 5.4921,
|
|
"mean_token_accuracy": 0.1689032256603241,
|
|
"num_tokens": 4259955.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.564026737213135,
|
|
"epoch": 1.9330543933054394,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004843183122594541,
|
|
"loss": 5.4619,
|
|
"mean_token_accuracy": 0.17760718017816543,
|
|
"num_tokens": 4268686.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.464136457443237,
|
|
"epoch": 1.9372384937238494,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004841997093248566,
|
|
"loss": 5.4324,
|
|
"mean_token_accuracy": 0.17746834307909012,
|
|
"num_tokens": 4277377.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.430179262161255,
|
|
"epoch": 1.9414225941422594,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004840806758908913,
|
|
"loss": 5.4468,
|
|
"mean_token_accuracy": 0.16720687597990036,
|
|
"num_tokens": 4287037.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.609952306747436,
|
|
"epoch": 1.9456066945606696,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048396121220250983,
|
|
"loss": 5.571,
|
|
"mean_token_accuracy": 0.15952135771512985,
|
|
"num_tokens": 4296923.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.435097360610962,
|
|
"epoch": 1.9497907949790796,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004838413185055494,
|
|
"loss": 5.3389,
|
|
"mean_token_accuracy": 0.18353236168622972,
|
|
"num_tokens": 4305873.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.466088438034058,
|
|
"epoch": 1.9539748953974896,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004837209950467318,
|
|
"loss": 5.4963,
|
|
"mean_token_accuracy": 0.16480235755443573,
|
|
"num_tokens": 4314912.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.527186918258667,
|
|
"epoch": 1.9581589958158996,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004836002420736635,
|
|
"loss": 5.4924,
|
|
"mean_token_accuracy": 0.16667744666337966,
|
|
"num_tokens": 4324606.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.43889594078064,
|
|
"epoch": 1.9623430962343096,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004834790598348347,
|
|
"loss": 5.354,
|
|
"mean_token_accuracy": 0.17873935550451278,
|
|
"num_tokens": 4333076.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.561453485488892,
|
|
"epoch": 1.9665271966527196,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004833574485796188,
|
|
"loss": 5.4915,
|
|
"mean_token_accuracy": 0.16966523081064225,
|
|
"num_tokens": 4341893.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.479845237731934,
|
|
"epoch": 1.9707112970711296,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048323540855827244,
|
|
"loss": 5.4677,
|
|
"mean_token_accuracy": 0.16631237864494325,
|
|
"num_tokens": 4352303.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.553999996185302,
|
|
"epoch": 1.9748953974895398,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004831129400219341,
|
|
"loss": 5.4172,
|
|
"mean_token_accuracy": 0.17104581445455552,
|
|
"num_tokens": 4361166.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.5471100330352785,
|
|
"epoch": 1.9790794979079498,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048299004322262457,
|
|
"loss": 5.4641,
|
|
"mean_token_accuracy": 0.1670989915728569,
|
|
"num_tokens": 4369999.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.468943405151367,
|
|
"epoch": 1.9832635983263598,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048286671841324565,
|
|
"loss": 5.429,
|
|
"mean_token_accuracy": 0.17624498158693314,
|
|
"num_tokens": 4379373.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.469639587402344,
|
|
"epoch": 1.98744769874477,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004827429658475799,
|
|
"loss": 5.4191,
|
|
"mean_token_accuracy": 0.17911377549171448,
|
|
"num_tokens": 4389035.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.521197557449341,
|
|
"epoch": 1.99163179916318,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048261878578029026,
|
|
"loss": 5.6026,
|
|
"mean_token_accuracy": 0.16540570557117462,
|
|
"num_tokens": 4399366.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.548810243606567,
|
|
"epoch": 1.99581589958159,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004824941784669193,
|
|
"loss": 5.4466,
|
|
"mean_token_accuracy": 0.17074829936027527,
|
|
"num_tokens": 4408183.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.53822774887085,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004823691441638889,
|
|
"loss": 5.5375,
|
|
"mean_token_accuracy": 0.16014931201934815,
|
|
"num_tokens": 4418644.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.420675230026245,
|
|
"epoch": 2.00418410041841,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004822436831284995,
|
|
"loss": 5.1231,
|
|
"mean_token_accuracy": 0.1838611513376236,
|
|
"num_tokens": 4427828.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.4005396366119385,
|
|
"epoch": 2.00836820083682,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004821177956189297,
|
|
"loss": 5.0386,
|
|
"mean_token_accuracy": 0.18786489367485046,
|
|
"num_tokens": 4436194.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.374574279785156,
|
|
"epoch": 2.01255230125523,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004819914818942359,
|
|
"loss": 5.1157,
|
|
"mean_token_accuracy": 0.18870676159858704,
|
|
"num_tokens": 4444468.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.475891828536987,
|
|
"epoch": 2.01673640167364,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048186474221435127,
|
|
"loss": 5.0311,
|
|
"mean_token_accuracy": 0.19567948579788208,
|
|
"num_tokens": 4452760.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.39983115196228,
|
|
"epoch": 2.02092050209205,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048173757684008583,
|
|
"loss": 5.0963,
|
|
"mean_token_accuracy": 0.1926517978310585,
|
|
"num_tokens": 4462154.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.410884189605713,
|
|
"epoch": 2.0251046025104604,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004816099860331255,
|
|
"loss": 5.1884,
|
|
"mean_token_accuracy": 0.18110420405864716,
|
|
"num_tokens": 4471987.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.386457538604736,
|
|
"epoch": 2.0292887029288704,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048148197005603145,
|
|
"loss": 5.1244,
|
|
"mean_token_accuracy": 0.18814223259687424,
|
|
"num_tokens": 4481645.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.290286874771118,
|
|
"epoch": 2.0334728033472804,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048135352917224005,
|
|
"loss": 5.0699,
|
|
"mean_token_accuracy": 0.19413749724626542,
|
|
"num_tokens": 4491166.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.484658861160279,
|
|
"epoch": 2.0376569037656904,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048122466364606214,
|
|
"loss": 5.0853,
|
|
"mean_token_accuracy": 0.18849048763513565,
|
|
"num_tokens": 4500159.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.337586879730225,
|
|
"epoch": 2.0418410041841004,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048109537374268216,
|
|
"loss": 5.0376,
|
|
"mean_token_accuracy": 0.19409384578466415,
|
|
"num_tokens": 4509737.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.225504207611084,
|
|
"epoch": 2.0460251046025104,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048096565972815793,
|
|
"loss": 4.9718,
|
|
"mean_token_accuracy": 0.18944948464632033,
|
|
"num_tokens": 4518379.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.396557950973511,
|
|
"epoch": 2.0502092050209204,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004808355218694199,
|
|
"loss": 5.1176,
|
|
"mean_token_accuracy": 0.1876818999648094,
|
|
"num_tokens": 4527885.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.3718568801879885,
|
|
"epoch": 2.0543933054393304,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048070496043427116,
|
|
"loss": 5.1515,
|
|
"mean_token_accuracy": 0.17957401126623154,
|
|
"num_tokens": 4537894.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.399702787399292,
|
|
"epoch": 2.0585774058577404,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048057397569138596,
|
|
"loss": 5.0524,
|
|
"mean_token_accuracy": 0.19588130563497544,
|
|
"num_tokens": 4546462.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.3364333152771,
|
|
"epoch": 2.062761506276151,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004804425679103099,
|
|
"loss": 5.0763,
|
|
"mean_token_accuracy": 0.18364191949367523,
|
|
"num_tokens": 4555970.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.346644544601441,
|
|
"epoch": 2.066945606694561,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004803107373614589,
|
|
"loss": 5.1251,
|
|
"mean_token_accuracy": 0.18158886283636094,
|
|
"num_tokens": 4565799.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.434850835800171,
|
|
"epoch": 2.071129707112971,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004801784843161194,
|
|
"loss": 5.1498,
|
|
"mean_token_accuracy": 0.17816840559244157,
|
|
"num_tokens": 4575571.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.310073566436768,
|
|
"epoch": 2.075313807531381,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004800458090464465,
|
|
"loss": 5.0696,
|
|
"mean_token_accuracy": 0.1884886294603348,
|
|
"num_tokens": 4584263.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.318070316314698,
|
|
"epoch": 2.079497907949791,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047991271182546476,
|
|
"loss": 5.0408,
|
|
"mean_token_accuracy": 0.19437606781721115,
|
|
"num_tokens": 4592979.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.329968070983886,
|
|
"epoch": 2.083682008368201,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047977919292706696,
|
|
"loss": 5.072,
|
|
"mean_token_accuracy": 0.19438588619232178,
|
|
"num_tokens": 4602042.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.3358073234558105,
|
|
"epoch": 2.087866108786611,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004796452526260134,
|
|
"loss": 5.1242,
|
|
"mean_token_accuracy": 0.1904394209384918,
|
|
"num_tokens": 4611197.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.2452216148376465,
|
|
"epoch": 2.092050209205021,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047951089119793164,
|
|
"loss": 5.0836,
|
|
"mean_token_accuracy": 0.1864160418510437,
|
|
"num_tokens": 4621612.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.092050209205021,
|
|
"eval_entropy": 5.17471120108856,
|
|
"eval_loss": 5.554454326629639,
|
|
"eval_mean_token_accuracy": 0.17758111425643075,
|
|
"eval_num_tokens": 4621612.0,
|
|
"eval_runtime": 2.7512,
|
|
"eval_samples_per_second": 1420.104,
|
|
"eval_steps_per_second": 177.74,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.246050310134888,
|
|
"epoch": 2.096234309623431,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000479376108919316,
|
|
"loss": 4.9461,
|
|
"mean_token_accuracy": 0.20234525948762894,
|
|
"num_tokens": 4630198.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.370128011703491,
|
|
"epoch": 2.100418410041841,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004792409060675267,
|
|
"loss": 5.0633,
|
|
"mean_token_accuracy": 0.18582278341054917,
|
|
"num_tokens": 4638729.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.320640897750854,
|
|
"epoch": 2.104602510460251,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004791052829207895,
|
|
"loss": 5.1837,
|
|
"mean_token_accuracy": 0.19010317623615264,
|
|
"num_tokens": 4648759.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.332196378707886,
|
|
"epoch": 2.1087866108786613,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000478969239758195,
|
|
"loss": 5.1232,
|
|
"mean_token_accuracy": 0.18839554190635682,
|
|
"num_tokens": 4657744.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.322515821456909,
|
|
"epoch": 2.1129707112970713,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047883277685969815,
|
|
"loss": 5.0777,
|
|
"mean_token_accuracy": 0.19082514941692352,
|
|
"num_tokens": 4666154.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.334899091720581,
|
|
"epoch": 2.1171548117154813,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047869589450611773,
|
|
"loss": 5.1203,
|
|
"mean_token_accuracy": 0.18512035459280013,
|
|
"num_tokens": 4675677.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.273021650314331,
|
|
"epoch": 2.1213389121338913,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004785585929791355,
|
|
"loss": 5.1003,
|
|
"mean_token_accuracy": 0.18778423964977264,
|
|
"num_tokens": 4684733.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.358072900772095,
|
|
"epoch": 2.1255230125523012,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000478420872561296,
|
|
"loss": 5.0226,
|
|
"mean_token_accuracy": 0.190115986764431,
|
|
"num_tokens": 4694155.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.236152172088623,
|
|
"epoch": 2.1297071129707112,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047828273353600577,
|
|
"loss": 5.0455,
|
|
"mean_token_accuracy": 0.19581301361322404,
|
|
"num_tokens": 4703286.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.3488829135894775,
|
|
"epoch": 2.1338912133891212,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047814417618753256,
|
|
"loss": 5.0764,
|
|
"mean_token_accuracy": 0.18645308762788773,
|
|
"num_tokens": 4711666.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.270642232894898,
|
|
"epoch": 2.1380753138075312,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047800520080100526,
|
|
"loss": 5.109,
|
|
"mean_token_accuracy": 0.18238500952720643,
|
|
"num_tokens": 4721644.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.2763121128082275,
|
|
"epoch": 2.1422594142259412,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047786580766241265,
|
|
"loss": 5.0617,
|
|
"mean_token_accuracy": 0.1938999980688095,
|
|
"num_tokens": 4730885.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.254446601867675,
|
|
"epoch": 2.1464435146443517,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004777259970586036,
|
|
"loss": 5.0515,
|
|
"mean_token_accuracy": 0.1930363342165947,
|
|
"num_tokens": 4740871.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.279245233535766,
|
|
"epoch": 2.1506276150627617,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004775857692772859,
|
|
"loss": 5.0131,
|
|
"mean_token_accuracy": 0.18983722776174544,
|
|
"num_tokens": 4750603.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.320049571990967,
|
|
"epoch": 2.1548117154811717,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004774451246070257,
|
|
"loss": 5.1277,
|
|
"mean_token_accuracy": 0.18448849469423295,
|
|
"num_tokens": 4760681.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.33151159286499,
|
|
"epoch": 2.1589958158995817,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004773040633372471,
|
|
"loss": 5.1069,
|
|
"mean_token_accuracy": 0.19268891215324402,
|
|
"num_tokens": 4770237.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.323383474349976,
|
|
"epoch": 2.1631799163179917,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047716258575823147,
|
|
"loss": 5.1348,
|
|
"mean_token_accuracy": 0.18953780084848404,
|
|
"num_tokens": 4779022.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.329308223724365,
|
|
"epoch": 2.1673640167364017,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000477020692161117,
|
|
"loss": 5.1532,
|
|
"mean_token_accuracy": 0.1888073593378067,
|
|
"num_tokens": 4788644.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.305137872695923,
|
|
"epoch": 2.1715481171548117,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004768783828378978,
|
|
"loss": 5.1583,
|
|
"mean_token_accuracy": 0.18073731809854507,
|
|
"num_tokens": 4798628.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.467740678787232,
|
|
"epoch": 2.1757322175732217,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004767356580814237,
|
|
"loss": 5.2674,
|
|
"mean_token_accuracy": 0.18118003010749817,
|
|
"num_tokens": 4807727.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.3576709747314455,
|
|
"epoch": 2.1799163179916317,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004765925181853993,
|
|
"loss": 5.1466,
|
|
"mean_token_accuracy": 0.18318893760442734,
|
|
"num_tokens": 4817896.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.3308957576751705,
|
|
"epoch": 2.1841004184100417,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004764489634443834,
|
|
"loss": 5.0941,
|
|
"mean_token_accuracy": 0.1882518395781517,
|
|
"num_tokens": 4827135.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.382042074203492,
|
|
"epoch": 2.1882845188284517,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047630499415378863,
|
|
"loss": 5.2044,
|
|
"mean_token_accuracy": 0.17700981199741364,
|
|
"num_tokens": 4837554.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.259732341766357,
|
|
"epoch": 2.192468619246862,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004761606106098806,
|
|
"loss": 5.0558,
|
|
"mean_token_accuracy": 0.18896397799253464,
|
|
"num_tokens": 4846739.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.306728982925415,
|
|
"epoch": 2.196652719665272,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047601581310977747,
|
|
"loss": 5.1122,
|
|
"mean_token_accuracy": 0.19147921949625016,
|
|
"num_tokens": 4855707.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.3428887844085695,
|
|
"epoch": 2.200836820083682,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004758706019514493,
|
|
"loss": 5.1178,
|
|
"mean_token_accuracy": 0.18854134827852248,
|
|
"num_tokens": 4865098.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.3770911693573,
|
|
"epoch": 2.205020920502092,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004757249774337172,
|
|
"loss": 5.1242,
|
|
"mean_token_accuracy": 0.1883791819214821,
|
|
"num_tokens": 4874426.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.271085119247436,
|
|
"epoch": 2.209205020920502,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047557893985625293,
|
|
"loss": 5.0972,
|
|
"mean_token_accuracy": 0.19329321682453154,
|
|
"num_tokens": 4883240.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.3018394947052006,
|
|
"epoch": 2.213389121338912,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004754324895195786,
|
|
"loss": 5.0867,
|
|
"mean_token_accuracy": 0.19223089665174484,
|
|
"num_tokens": 4891989.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.330430412292481,
|
|
"epoch": 2.217573221757322,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004752856267250652,
|
|
"loss": 5.1901,
|
|
"mean_token_accuracy": 0.1845463216304779,
|
|
"num_tokens": 4901546.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.224370718002319,
|
|
"epoch": 2.221757322175732,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047513835177493286,
|
|
"loss": 5.0066,
|
|
"mean_token_accuracy": 0.19758611619472505,
|
|
"num_tokens": 4910222.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.324810361862182,
|
|
"epoch": 2.225941422594142,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004749906649722498,
|
|
"loss": 5.2126,
|
|
"mean_token_accuracy": 0.17787486612796782,
|
|
"num_tokens": 4920184.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.291962575912476,
|
|
"epoch": 2.2301255230125525,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004748425666209316,
|
|
"loss": 5.129,
|
|
"mean_token_accuracy": 0.1867642730474472,
|
|
"num_tokens": 4930014.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.310875511169433,
|
|
"epoch": 2.2343096234309625,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004746940570257409,
|
|
"loss": 5.0502,
|
|
"mean_token_accuracy": 0.18978514075279235,
|
|
"num_tokens": 4939556.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.2545092582702635,
|
|
"epoch": 2.2384937238493725,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004745451364922867,
|
|
"loss": 5.1045,
|
|
"mean_token_accuracy": 0.18112686574459075,
|
|
"num_tokens": 4949066.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.4056755065917965,
|
|
"epoch": 2.2426778242677825,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004743958053270233,
|
|
"loss": 5.1133,
|
|
"mean_token_accuracy": 0.18686354160308838,
|
|
"num_tokens": 4957881.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.313535404205322,
|
|
"epoch": 2.2468619246861925,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047424606383725037,
|
|
"loss": 5.15,
|
|
"mean_token_accuracy": 0.18608384430408478,
|
|
"num_tokens": 4967105.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.221853256225586,
|
|
"epoch": 2.2510460251046025,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004740959123311118,
|
|
"loss": 5.0638,
|
|
"mean_token_accuracy": 0.19056562036275865,
|
|
"num_tokens": 4977676.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.240513038635254,
|
|
"epoch": 2.2552301255230125,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047394535111759526,
|
|
"loss": 5.0737,
|
|
"mean_token_accuracy": 0.19114105999469758,
|
|
"num_tokens": 4986163.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.316378784179688,
|
|
"epoch": 2.2594142259414225,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004737943805065314,
|
|
"loss": 5.0803,
|
|
"mean_token_accuracy": 0.18888150602579118,
|
|
"num_tokens": 4995179.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.378285694122314,
|
|
"epoch": 2.2635983263598325,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047364300080859363,
|
|
"loss": 5.1821,
|
|
"mean_token_accuracy": 0.17384993582963942,
|
|
"num_tokens": 5003856.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.348804092407226,
|
|
"epoch": 2.2677824267782425,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047349121233529695,
|
|
"loss": 5.2124,
|
|
"mean_token_accuracy": 0.1831280842423439,
|
|
"num_tokens": 5013506.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.290798139572144,
|
|
"epoch": 2.2719665271966525,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004733390153989976,
|
|
"loss": 5.1439,
|
|
"mean_token_accuracy": 0.18972165137529373,
|
|
"num_tokens": 5022159.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.307530641555786,
|
|
"epoch": 2.276150627615063,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047318641031289245,
|
|
"loss": 5.0866,
|
|
"mean_token_accuracy": 0.1890856981277466,
|
|
"num_tokens": 5031869.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.17536473274231,
|
|
"epoch": 2.280334728033473,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004730333973910183,
|
|
"loss": 4.959,
|
|
"mean_token_accuracy": 0.20518755465745925,
|
|
"num_tokens": 5039904.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.337870121002197,
|
|
"epoch": 2.284518828451883,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004728799769482511,
|
|
"loss": 5.2498,
|
|
"mean_token_accuracy": 0.18180439472198487,
|
|
"num_tokens": 5049144.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.2850103855133055,
|
|
"epoch": 2.288702928870293,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047272614930030533,
|
|
"loss": 5.0029,
|
|
"mean_token_accuracy": 0.19399643391370774,
|
|
"num_tokens": 5058302.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.21338415145874,
|
|
"epoch": 2.292887029288703,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047257191476373375,
|
|
"loss": 5.0452,
|
|
"mean_token_accuracy": 0.1903329998254776,
|
|
"num_tokens": 5067443.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.250432777404785,
|
|
"epoch": 2.297071129707113,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047241727365592603,
|
|
"loss": 5.1744,
|
|
"mean_token_accuracy": 0.1805453270673752,
|
|
"num_tokens": 5076415.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.3315346240997314,
|
|
"epoch": 2.301255230125523,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004722622262951089,
|
|
"loss": 5.1321,
|
|
"mean_token_accuracy": 0.1840568408370018,
|
|
"num_tokens": 5085882.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.325094938278198,
|
|
"epoch": 2.305439330543933,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004721067730003448,
|
|
"loss": 5.1048,
|
|
"mean_token_accuracy": 0.19234621822834014,
|
|
"num_tokens": 5094685.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.2754950523376465,
|
|
"epoch": 2.309623430962343,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004719509140915317,
|
|
"loss": 5.1123,
|
|
"mean_token_accuracy": 0.18495827168226242,
|
|
"num_tokens": 5103885.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.279915857315063,
|
|
"epoch": 2.3138075313807533,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047179464988940205,
|
|
"loss": 5.0279,
|
|
"mean_token_accuracy": 0.19769937545061111,
|
|
"num_tokens": 5112164.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.218002796173096,
|
|
"epoch": 2.3179916317991633,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047163798071552253,
|
|
"loss": 5.0108,
|
|
"mean_token_accuracy": 0.19311033189296722,
|
|
"num_tokens": 5121913.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.295284032821655,
|
|
"epoch": 2.3221757322175733,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004714809068922931,
|
|
"loss": 5.0179,
|
|
"mean_token_accuracy": 0.19485592395067214,
|
|
"num_tokens": 5130447.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.240268898010254,
|
|
"epoch": 2.3263598326359833,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004713234287429465,
|
|
"loss": 5.1335,
|
|
"mean_token_accuracy": 0.1870643138885498,
|
|
"num_tokens": 5139863.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.221497011184693,
|
|
"epoch": 2.3305439330543933,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004711655465915473,
|
|
"loss": 5.1128,
|
|
"mean_token_accuracy": 0.18968554735183715,
|
|
"num_tokens": 5149945.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.3533525466918945,
|
|
"epoch": 2.3347280334728033,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004710072607629917,
|
|
"loss": 5.0432,
|
|
"mean_token_accuracy": 0.18991013020277023,
|
|
"num_tokens": 5160293.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.2163951873779295,
|
|
"epoch": 2.3389121338912133,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047084857158300646,
|
|
"loss": 5.0816,
|
|
"mean_token_accuracy": 0.18396058976650237,
|
|
"num_tokens": 5169896.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.266668844223022,
|
|
"epoch": 2.3430962343096233,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047068947937814844,
|
|
"loss": 5.1211,
|
|
"mean_token_accuracy": 0.18983114808797835,
|
|
"num_tokens": 5178898.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.339715528488159,
|
|
"epoch": 2.3472803347280333,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047052998447580374,
|
|
"loss": 5.1876,
|
|
"mean_token_accuracy": 0.18320218920707704,
|
|
"num_tokens": 5188733.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.22649621963501,
|
|
"epoch": 2.3514644351464433,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047037008720418724,
|
|
"loss": 5.0787,
|
|
"mean_token_accuracy": 0.18901680409908295,
|
|
"num_tokens": 5198117.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.350342702865601,
|
|
"epoch": 2.3556485355648533,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004702097878923419,
|
|
"loss": 5.1353,
|
|
"mean_token_accuracy": 0.187324658036232,
|
|
"num_tokens": 5206791.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.235406064987183,
|
|
"epoch": 2.3598326359832638,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004700490868701377,
|
|
"loss": 5.0322,
|
|
"mean_token_accuracy": 0.197453673183918,
|
|
"num_tokens": 5216427.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.208360719680786,
|
|
"epoch": 2.3640167364016738,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004698879844682718,
|
|
"loss": 5.0165,
|
|
"mean_token_accuracy": 0.1976787641644478,
|
|
"num_tokens": 5225983.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.224909448623658,
|
|
"epoch": 2.3682008368200838,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046972648101826695,
|
|
"loss": 5.1185,
|
|
"mean_token_accuracy": 0.19658852815628053,
|
|
"num_tokens": 5234547.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.3236894607543945,
|
|
"epoch": 2.3723849372384938,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004695645768524712,
|
|
"loss": 5.1387,
|
|
"mean_token_accuracy": 0.1944551408290863,
|
|
"num_tokens": 5243281.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.269918966293335,
|
|
"epoch": 2.3765690376569037,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004694022723040573,
|
|
"loss": 5.054,
|
|
"mean_token_accuracy": 0.18740010708570481,
|
|
"num_tokens": 5252318.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.306603622436524,
|
|
"epoch": 2.3807531380753137,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046923956770702206,
|
|
"loss": 5.142,
|
|
"mean_token_accuracy": 0.19188392609357835,
|
|
"num_tokens": 5261781.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.362006950378418,
|
|
"epoch": 2.3849372384937237,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004690764633961852,
|
|
"loss": 5.0405,
|
|
"mean_token_accuracy": 0.19037206172943116,
|
|
"num_tokens": 5270409.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.118597793579101,
|
|
"epoch": 2.3891213389121337,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004689129597071893,
|
|
"loss": 5.0296,
|
|
"mean_token_accuracy": 0.1998368188738823,
|
|
"num_tokens": 5278766.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.279699325561523,
|
|
"epoch": 2.393305439330544,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046874905697649857,
|
|
"loss": 5.2329,
|
|
"mean_token_accuracy": 0.18201542645692825,
|
|
"num_tokens": 5287572.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.352121734619141,
|
|
"epoch": 2.397489539748954,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046858475554139857,
|
|
"loss": 5.1137,
|
|
"mean_token_accuracy": 0.19088485091924667,
|
|
"num_tokens": 5297348.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.281993722915649,
|
|
"epoch": 2.401673640167364,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004684200557399952,
|
|
"loss": 5.1248,
|
|
"mean_token_accuracy": 0.18940239101648332,
|
|
"num_tokens": 5305748.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.245026397705078,
|
|
"epoch": 2.405857740585774,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046825495791121414,
|
|
"loss": 5.0718,
|
|
"mean_token_accuracy": 0.2004568710923195,
|
|
"num_tokens": 5314482.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.276341438293457,
|
|
"epoch": 2.410041841004184,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046808946239480017,
|
|
"loss": 5.1697,
|
|
"mean_token_accuracy": 0.1845923036336899,
|
|
"num_tokens": 5325161.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.240137243270874,
|
|
"epoch": 2.414225941422594,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004679235695313165,
|
|
"loss": 5.1561,
|
|
"mean_token_accuracy": 0.18108493238687515,
|
|
"num_tokens": 5335262.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.268524217605591,
|
|
"epoch": 2.418410041841004,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046775727966214393,
|
|
"loss": 5.0684,
|
|
"mean_token_accuracy": 0.19307907223701476,
|
|
"num_tokens": 5345145.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.192647314071655,
|
|
"epoch": 2.422594142259414,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004675905931294803,
|
|
"loss": 5.0428,
|
|
"mean_token_accuracy": 0.18985151499509811,
|
|
"num_tokens": 5354715.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.245454978942871,
|
|
"epoch": 2.426778242677824,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004674235102763398,
|
|
"loss": 5.1013,
|
|
"mean_token_accuracy": 0.19313904792070388,
|
|
"num_tokens": 5364180.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.19718189239502,
|
|
"epoch": 2.430962343096234,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046725603144655195,
|
|
"loss": 5.0619,
|
|
"mean_token_accuracy": 0.1946485936641693,
|
|
"num_tokens": 5372919.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.213984584808349,
|
|
"epoch": 2.435146443514644,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004670881569847612,
|
|
"loss": 5.053,
|
|
"mean_token_accuracy": 0.19749044626951218,
|
|
"num_tokens": 5383002.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.3344258785247805,
|
|
"epoch": 2.439330543933054,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046691988723642636,
|
|
"loss": 5.2464,
|
|
"mean_token_accuracy": 0.1803280457854271,
|
|
"num_tokens": 5393622.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.342332363128662,
|
|
"epoch": 2.4435146443514646,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004667512225478195,
|
|
"loss": 5.1218,
|
|
"mean_token_accuracy": 0.18834031075239183,
|
|
"num_tokens": 5403106.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.32227258682251,
|
|
"epoch": 2.4476987447698746,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046658216326602526,
|
|
"loss": 5.0999,
|
|
"mean_token_accuracy": 0.18862743377685548,
|
|
"num_tokens": 5412596.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.321303844451904,
|
|
"epoch": 2.4518828451882846,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004664127097389406,
|
|
"loss": 5.1521,
|
|
"mean_token_accuracy": 0.18980562090873718,
|
|
"num_tokens": 5422301.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.329876470565796,
|
|
"epoch": 2.4560669456066946,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046624286231527367,
|
|
"loss": 5.0702,
|
|
"mean_token_accuracy": 0.18936679512262344,
|
|
"num_tokens": 5432178.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.186872720718384,
|
|
"epoch": 2.4602510460251046,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004660726213445431,
|
|
"loss": 5.1117,
|
|
"mean_token_accuracy": 0.1925972133874893,
|
|
"num_tokens": 5441162.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.261181735992432,
|
|
"epoch": 2.4644351464435146,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004659019871770774,
|
|
"loss": 5.0402,
|
|
"mean_token_accuracy": 0.19670872539281845,
|
|
"num_tokens": 5450566.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.280089902877807,
|
|
"epoch": 2.4686192468619246,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046573096016401435,
|
|
"loss": 5.0793,
|
|
"mean_token_accuracy": 0.1999111920595169,
|
|
"num_tokens": 5460289.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.1828478336334225,
|
|
"epoch": 2.4728033472803346,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004655595406572999,
|
|
"loss": 5.059,
|
|
"mean_token_accuracy": 0.1950552761554718,
|
|
"num_tokens": 5469581.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.311435317993164,
|
|
"epoch": 2.476987447698745,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000465387729009688,
|
|
"loss": 5.1574,
|
|
"mean_token_accuracy": 0.18863597214221955,
|
|
"num_tokens": 5478005.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.234506416320801,
|
|
"epoch": 2.481171548117155,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046521552557473927,
|
|
"loss": 5.1502,
|
|
"mean_token_accuracy": 0.18971240371465684,
|
|
"num_tokens": 5486963.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.274946498870849,
|
|
"epoch": 2.485355648535565,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046504293070682073,
|
|
"loss": 5.0442,
|
|
"mean_token_accuracy": 0.18735662549734117,
|
|
"num_tokens": 5496368.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.211497068405151,
|
|
"epoch": 2.489539748953975,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046486994476110493,
|
|
"loss": 5.032,
|
|
"mean_token_accuracy": 0.1927930787205696,
|
|
"num_tokens": 5505706.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.178287935256958,
|
|
"epoch": 2.493723849372385,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046469656809356916,
|
|
"loss": 5.0211,
|
|
"mean_token_accuracy": 0.19452526569366455,
|
|
"num_tokens": 5513945.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.212137889862061,
|
|
"epoch": 2.497907949790795,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046452280106099454,
|
|
"loss": 5.0682,
|
|
"mean_token_accuracy": 0.1926400512456894,
|
|
"num_tokens": 5523831.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.306842994689942,
|
|
"epoch": 2.502092050209205,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004643486440209659,
|
|
"loss": 5.206,
|
|
"mean_token_accuracy": 0.1866828441619873,
|
|
"num_tokens": 5532705.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.338761186599731,
|
|
"epoch": 2.506276150627615,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004641740973318702,
|
|
"loss": 5.1171,
|
|
"mean_token_accuracy": 0.19268973767757416,
|
|
"num_tokens": 5542793.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.267599630355835,
|
|
"epoch": 2.510460251046025,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004639991613528967,
|
|
"loss": 5.1003,
|
|
"mean_token_accuracy": 0.19482764303684236,
|
|
"num_tokens": 5551938.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.510460251046025,
|
|
"eval_entropy": 5.085457296703003,
|
|
"eval_loss": 5.482717990875244,
|
|
"eval_mean_token_accuracy": 0.18331411985227178,
|
|
"eval_num_tokens": 5551938.0,
|
|
"eval_runtime": 2.7607,
|
|
"eval_samples_per_second": 1415.213,
|
|
"eval_steps_per_second": 177.128,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.283684539794922,
|
|
"epoch": 2.514644351464435,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046382383644403543,
|
|
"loss": 5.1577,
|
|
"mean_token_accuracy": 0.18795475512742996,
|
|
"num_tokens": 5561374.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.189481163024903,
|
|
"epoch": 2.518828451882845,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046364812296607676,
|
|
"loss": 5.0131,
|
|
"mean_token_accuracy": 0.1919141173362732,
|
|
"num_tokens": 5570459.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.175030374526978,
|
|
"epoch": 2.523012552301255,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046347202128061094,
|
|
"loss": 5.1047,
|
|
"mean_token_accuracy": 0.18761301338672637,
|
|
"num_tokens": 5579692.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.328635215759277,
|
|
"epoch": 2.5271966527196654,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004632955317500268,
|
|
"loss": 5.073,
|
|
"mean_token_accuracy": 0.19245171546936035,
|
|
"num_tokens": 5588338.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.242974042892456,
|
|
"epoch": 2.5313807531380754,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046311865473751154,
|
|
"loss": 5.1385,
|
|
"mean_token_accuracy": 0.1903909757733345,
|
|
"num_tokens": 5598589.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.305221652984619,
|
|
"epoch": 2.5355648535564854,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004629413906070495,
|
|
"loss": 5.1302,
|
|
"mean_token_accuracy": 0.19345313906669617,
|
|
"num_tokens": 5608127.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.313396549224853,
|
|
"epoch": 2.5397489539748954,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046276373972342177,
|
|
"loss": 5.1666,
|
|
"mean_token_accuracy": 0.19249054193496704,
|
|
"num_tokens": 5617728.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.202130746841431,
|
|
"epoch": 2.5439330543933054,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004625857024522054,
|
|
"loss": 5.1283,
|
|
"mean_token_accuracy": 0.19441729336977004,
|
|
"num_tokens": 5626823.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.284762048721314,
|
|
"epoch": 2.5481171548117154,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046240727915977236,
|
|
"loss": 5.0954,
|
|
"mean_token_accuracy": 0.1940153196454048,
|
|
"num_tokens": 5636284.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.344861650466919,
|
|
"epoch": 2.5523012552301254,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004622284702132893,
|
|
"loss": 5.2257,
|
|
"mean_token_accuracy": 0.17923278361558914,
|
|
"num_tokens": 5645721.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.222573614120483,
|
|
"epoch": 2.5564853556485354,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00046204927598071595,
|
|
"loss": 5.1455,
|
|
"mean_token_accuracy": 0.18975146263837814,
|
|
"num_tokens": 5654734.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.318623065948486,
|
|
"epoch": 2.560669456066946,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046186969683080547,
|
|
"loss": 5.1221,
|
|
"mean_token_accuracy": 0.20064057111740113,
|
|
"num_tokens": 5664010.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.247400951385498,
|
|
"epoch": 2.564853556485356,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046168973313310294,
|
|
"loss": 5.1082,
|
|
"mean_token_accuracy": 0.19190422147512437,
|
|
"num_tokens": 5673028.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.207634115219117,
|
|
"epoch": 2.569037656903766,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004615093852579445,
|
|
"loss": 5.0781,
|
|
"mean_token_accuracy": 0.19491716772317885,
|
|
"num_tokens": 5682517.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.188964033126831,
|
|
"epoch": 2.573221757322176,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046132865357645717,
|
|
"loss": 5.0002,
|
|
"mean_token_accuracy": 0.2007356435060501,
|
|
"num_tokens": 5690920.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.189310789108276,
|
|
"epoch": 2.577405857740586,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046114753846055774,
|
|
"loss": 5.0901,
|
|
"mean_token_accuracy": 0.18546762466430664,
|
|
"num_tokens": 5700336.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.3212127685546875,
|
|
"epoch": 2.581589958158996,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000460966040282952,
|
|
"loss": 5.2212,
|
|
"mean_token_accuracy": 0.1865714892745018,
|
|
"num_tokens": 5711085.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.249056959152222,
|
|
"epoch": 2.585774058577406,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00046078415941713384,
|
|
"loss": 5.036,
|
|
"mean_token_accuracy": 0.19373459368944168,
|
|
"num_tokens": 5719560.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.1868115901947025,
|
|
"epoch": 2.589958158995816,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004606018962373852,
|
|
"loss": 4.992,
|
|
"mean_token_accuracy": 0.19921741634607315,
|
|
"num_tokens": 5727865.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.229388856887818,
|
|
"epoch": 2.594142259414226,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004604192511187741,
|
|
"loss": 5.1167,
|
|
"mean_token_accuracy": 0.19096483588218688,
|
|
"num_tokens": 5737705.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.3044239521026615,
|
|
"epoch": 2.598326359832636,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046023622443715483,
|
|
"loss": 5.1361,
|
|
"mean_token_accuracy": 0.17921521067619323,
|
|
"num_tokens": 5747057.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.223257923126221,
|
|
"epoch": 2.602510460251046,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046005281656916707,
|
|
"loss": 5.0855,
|
|
"mean_token_accuracy": 0.1901606187224388,
|
|
"num_tokens": 5755793.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.244540691375732,
|
|
"epoch": 2.606694560669456,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004598690278922346,
|
|
"loss": 5.0939,
|
|
"mean_token_accuracy": 0.18625172227621078,
|
|
"num_tokens": 5765385.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.153616189956665,
|
|
"epoch": 2.6108786610878663,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045968485878456497,
|
|
"loss": 5.0152,
|
|
"mean_token_accuracy": 0.19641041308641433,
|
|
"num_tokens": 5773947.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.227500534057617,
|
|
"epoch": 2.6150627615062763,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004595003096251486,
|
|
"loss": 5.0231,
|
|
"mean_token_accuracy": 0.196515990793705,
|
|
"num_tokens": 5782700.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.287982034683227,
|
|
"epoch": 2.6192468619246863,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045931538079375803,
|
|
"loss": 5.1713,
|
|
"mean_token_accuracy": 0.18707239180803298,
|
|
"num_tokens": 5792452.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.330501556396484,
|
|
"epoch": 2.6234309623430963,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045913007267094704,
|
|
"loss": 5.1115,
|
|
"mean_token_accuracy": 0.1955230265855789,
|
|
"num_tokens": 5801270.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.278586912155151,
|
|
"epoch": 2.6276150627615062,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004589443856380499,
|
|
"loss": 5.1445,
|
|
"mean_token_accuracy": 0.1937210589647293,
|
|
"num_tokens": 5810704.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.251729583740234,
|
|
"epoch": 2.6317991631799162,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004587583200771807,
|
|
"loss": 5.0889,
|
|
"mean_token_accuracy": 0.1871331065893173,
|
|
"num_tokens": 5819690.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.1724451065063475,
|
|
"epoch": 2.6359832635983262,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004585718763712325,
|
|
"loss": 5.026,
|
|
"mean_token_accuracy": 0.18755852431058884,
|
|
"num_tokens": 5828912.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.104170465469361,
|
|
"epoch": 2.6401673640167362,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004583850549038764,
|
|
"loss": 4.9471,
|
|
"mean_token_accuracy": 0.19984448105096816,
|
|
"num_tokens": 5836914.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.199622535705567,
|
|
"epoch": 2.6443514644351467,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045819785605956105,
|
|
"loss": 5.1115,
|
|
"mean_token_accuracy": 0.18994835317134856,
|
|
"num_tokens": 5846268.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.234996032714844,
|
|
"epoch": 2.6485355648535567,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004580102802235115,
|
|
"loss": 5.106,
|
|
"mean_token_accuracy": 0.19342414885759354,
|
|
"num_tokens": 5855029.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.269737625122071,
|
|
"epoch": 2.6527196652719667,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004578223277817286,
|
|
"loss": 5.106,
|
|
"mean_token_accuracy": 0.18429109901189805,
|
|
"num_tokens": 5865049.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.291262149810791,
|
|
"epoch": 2.6569037656903767,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045763399912098835,
|
|
"loss": 5.1255,
|
|
"mean_token_accuracy": 0.1859837993979454,
|
|
"num_tokens": 5873593.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.1342895984649655,
|
|
"epoch": 2.6610878661087867,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045744529462884086,
|
|
"loss": 5.0367,
|
|
"mean_token_accuracy": 0.19363411962985994,
|
|
"num_tokens": 5882562.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.219482898712158,
|
|
"epoch": 2.6652719665271967,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045725621469360964,
|
|
"loss": 5.2162,
|
|
"mean_token_accuracy": 0.1834290146827698,
|
|
"num_tokens": 5892516.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.2356750011444095,
|
|
"epoch": 2.6694560669456067,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004570667597043907,
|
|
"loss": 5.0825,
|
|
"mean_token_accuracy": 0.19192732721567154,
|
|
"num_tokens": 5901494.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.253531551361084,
|
|
"epoch": 2.6736401673640167,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000456876930051052,
|
|
"loss": 5.0866,
|
|
"mean_token_accuracy": 0.18902710378170012,
|
|
"num_tokens": 5910750.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.300333690643311,
|
|
"epoch": 2.6778242677824267,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004566867261242325,
|
|
"loss": 5.1335,
|
|
"mean_token_accuracy": 0.19503549188375474,
|
|
"num_tokens": 5920277.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.239656829833985,
|
|
"epoch": 2.6820083682008367,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045649614831534127,
|
|
"loss": 5.1088,
|
|
"mean_token_accuracy": 0.1912435472011566,
|
|
"num_tokens": 5929880.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.354261350631714,
|
|
"epoch": 2.6861924686192467,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045630519701655677,
|
|
"loss": 5.2567,
|
|
"mean_token_accuracy": 0.1914851501584053,
|
|
"num_tokens": 5939538.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.251696252822876,
|
|
"epoch": 2.6903765690376567,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045611387262082604,
|
|
"loss": 5.0885,
|
|
"mean_token_accuracy": 0.1932705298066139,
|
|
"num_tokens": 5948747.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.1684175491333,
|
|
"epoch": 2.694560669456067,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045592217552186407,
|
|
"loss": 5.0811,
|
|
"mean_token_accuracy": 0.18924431651830673,
|
|
"num_tokens": 5958209.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.273152160644531,
|
|
"epoch": 2.698744769874477,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004557301061141525,
|
|
"loss": 5.1263,
|
|
"mean_token_accuracy": 0.19000065326690674,
|
|
"num_tokens": 5967184.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.252749347686768,
|
|
"epoch": 2.702928870292887,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045553766479293945,
|
|
"loss": 5.1432,
|
|
"mean_token_accuracy": 0.18446308374404907,
|
|
"num_tokens": 5976006.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.146998262405395,
|
|
"epoch": 2.707112970711297,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045534485195423806,
|
|
"loss": 4.9865,
|
|
"mean_token_accuracy": 0.1992344841361046,
|
|
"num_tokens": 5984671.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.324666833877563,
|
|
"epoch": 2.711297071129707,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045515166799482633,
|
|
"loss": 5.108,
|
|
"mean_token_accuracy": 0.19198963791131973,
|
|
"num_tokens": 5993190.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.175671720504761,
|
|
"epoch": 2.715481171548117,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045495811331224564,
|
|
"loss": 5.0912,
|
|
"mean_token_accuracy": 0.19211582392454146,
|
|
"num_tokens": 6004080.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.251320219039917,
|
|
"epoch": 2.719665271966527,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004547641883048005,
|
|
"loss": 5.1142,
|
|
"mean_token_accuracy": 0.1906818524003029,
|
|
"num_tokens": 6013212.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.235155344009399,
|
|
"epoch": 2.723849372384937,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004545698933715573,
|
|
"loss": 5.042,
|
|
"mean_token_accuracy": 0.19610748142004014,
|
|
"num_tokens": 6023444.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.195427417755127,
|
|
"epoch": 2.7280334728033475,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045437522891234386,
|
|
"loss": 5.0471,
|
|
"mean_token_accuracy": 0.19615740925073624,
|
|
"num_tokens": 6032217.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.186441087722779,
|
|
"epoch": 2.7322175732217575,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045418019532774827,
|
|
"loss": 5.0092,
|
|
"mean_token_accuracy": 0.20197249948978424,
|
|
"num_tokens": 6040246.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.218929958343506,
|
|
"epoch": 2.7364016736401675,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045398479301911836,
|
|
"loss": 5.1353,
|
|
"mean_token_accuracy": 0.18078775107860565,
|
|
"num_tokens": 6049447.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.078128147125244,
|
|
"epoch": 2.7405857740585775,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045378902238856064,
|
|
"loss": 5.0044,
|
|
"mean_token_accuracy": 0.1962415635585785,
|
|
"num_tokens": 6058294.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.234472227096558,
|
|
"epoch": 2.7447698744769875,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004535928838389396,
|
|
"loss": 5.1187,
|
|
"mean_token_accuracy": 0.19223692119121552,
|
|
"num_tokens": 6067237.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.2860150814056395,
|
|
"epoch": 2.7489539748953975,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004533963777738768,
|
|
"loss": 5.1314,
|
|
"mean_token_accuracy": 0.18676792085170746,
|
|
"num_tokens": 6075599.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.266327524185181,
|
|
"epoch": 2.7531380753138075,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045319950459775024,
|
|
"loss": 5.1162,
|
|
"mean_token_accuracy": 0.1831006035208702,
|
|
"num_tokens": 6085131.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.222617435455322,
|
|
"epoch": 2.7573221757322175,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004530022647156931,
|
|
"loss": 5.0584,
|
|
"mean_token_accuracy": 0.19816596955060958,
|
|
"num_tokens": 6094745.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.16615138053894,
|
|
"epoch": 2.7615062761506275,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004528046585335935,
|
|
"loss": 5.1265,
|
|
"mean_token_accuracy": 0.1877957969903946,
|
|
"num_tokens": 6104182.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.242715072631836,
|
|
"epoch": 2.7656903765690375,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045260668645809304,
|
|
"loss": 4.9852,
|
|
"mean_token_accuracy": 0.2020021751523018,
|
|
"num_tokens": 6113370.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.1468791484832765,
|
|
"epoch": 2.7698744769874475,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004524083488965866,
|
|
"loss": 5.0724,
|
|
"mean_token_accuracy": 0.19289353340864182,
|
|
"num_tokens": 6122202.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.221616172790528,
|
|
"epoch": 2.7740585774058575,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004522096462572208,
|
|
"loss": 5.1339,
|
|
"mean_token_accuracy": 0.18605839014053344,
|
|
"num_tokens": 6131081.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.339215040206909,
|
|
"epoch": 2.778242677824268,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004520105789488939,
|
|
"loss": 5.2008,
|
|
"mean_token_accuracy": 0.18264513462781906,
|
|
"num_tokens": 6141707.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.167083644866944,
|
|
"epoch": 2.782426778242678,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045181114738125427,
|
|
"loss": 5.0537,
|
|
"mean_token_accuracy": 0.19120820611715317,
|
|
"num_tokens": 6151651.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.224046277999878,
|
|
"epoch": 2.786610878661088,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045161135196470015,
|
|
"loss": 5.012,
|
|
"mean_token_accuracy": 0.1992000862956047,
|
|
"num_tokens": 6159928.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.227979707717895,
|
|
"epoch": 2.790794979079498,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004514111931103783,
|
|
"loss": 5.1054,
|
|
"mean_token_accuracy": 0.186365570127964,
|
|
"num_tokens": 6169115.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.244626808166504,
|
|
"epoch": 2.794979079497908,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045121067123018353,
|
|
"loss": 5.1221,
|
|
"mean_token_accuracy": 0.19543938040733339,
|
|
"num_tokens": 6177995.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.26192798614502,
|
|
"epoch": 2.799163179916318,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045100978673675755,
|
|
"loss": 5.1185,
|
|
"mean_token_accuracy": 0.1874633625149727,
|
|
"num_tokens": 6188081.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.287642002105713,
|
|
"epoch": 2.803347280334728,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045080854004348836,
|
|
"loss": 5.0772,
|
|
"mean_token_accuracy": 0.1929886519908905,
|
|
"num_tokens": 6198010.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.103681421279907,
|
|
"epoch": 2.8075313807531384,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045060693156450933,
|
|
"loss": 5.0486,
|
|
"mean_token_accuracy": 0.20039194375276564,
|
|
"num_tokens": 6207164.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.301557302474976,
|
|
"epoch": 2.8117154811715483,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004504049617146983,
|
|
"loss": 5.19,
|
|
"mean_token_accuracy": 0.18494621217250823,
|
|
"num_tokens": 6216512.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.247885656356812,
|
|
"epoch": 2.8158995815899583,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045020263090967674,
|
|
"loss": 5.1446,
|
|
"mean_token_accuracy": 0.1954988032579422,
|
|
"num_tokens": 6226196.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.21619553565979,
|
|
"epoch": 2.8200836820083683,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004499999395658088,
|
|
"loss": 5.0067,
|
|
"mean_token_accuracy": 0.19041311144828796,
|
|
"num_tokens": 6234820.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.1981120109558105,
|
|
"epoch": 2.8242677824267783,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000449796888100201,
|
|
"loss": 5.0795,
|
|
"mean_token_accuracy": 0.19274015575647355,
|
|
"num_tokens": 6243813.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.340387201309204,
|
|
"epoch": 2.8284518828451883,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044959347693070026,
|
|
"loss": 5.1749,
|
|
"mean_token_accuracy": 0.18512681722640992,
|
|
"num_tokens": 6253376.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.235930252075195,
|
|
"epoch": 2.8326359832635983,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004493897064758943,
|
|
"loss": 5.1914,
|
|
"mean_token_accuracy": 0.1867931976914406,
|
|
"num_tokens": 6262640.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.267892265319825,
|
|
"epoch": 2.8368200836820083,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004491855771551098,
|
|
"loss": 5.1408,
|
|
"mean_token_accuracy": 0.1928987294435501,
|
|
"num_tokens": 6270908.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.23227481842041,
|
|
"epoch": 2.8410041841004183,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004489810893884121,
|
|
"loss": 5.0538,
|
|
"mean_token_accuracy": 0.18868430256843566,
|
|
"num_tokens": 6280361.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.244975852966308,
|
|
"epoch": 2.8451882845188283,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004487762435966042,
|
|
"loss": 5.0759,
|
|
"mean_token_accuracy": 0.19281421452760697,
|
|
"num_tokens": 6288832.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.161826848983765,
|
|
"epoch": 2.8493723849372383,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044857104020122585,
|
|
"loss": 5.0446,
|
|
"mean_token_accuracy": 0.19527423232793809,
|
|
"num_tokens": 6298553.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.1375984191894535,
|
|
"epoch": 2.8535564853556483,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004483654796245526,
|
|
"loss": 5.0108,
|
|
"mean_token_accuracy": 0.19439974874258042,
|
|
"num_tokens": 6307692.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.299069738388061,
|
|
"epoch": 2.8577405857740583,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000448159562289595,
|
|
"loss": 5.1064,
|
|
"mean_token_accuracy": 0.1892927199602127,
|
|
"num_tokens": 6316729.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.289239454269409,
|
|
"epoch": 2.8619246861924688,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044795328862009794,
|
|
"loss": 5.1812,
|
|
"mean_token_accuracy": 0.1823544904589653,
|
|
"num_tokens": 6326210.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.150221061706543,
|
|
"epoch": 2.8661087866108788,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00044774665904053943,
|
|
"loss": 5.0761,
|
|
"mean_token_accuracy": 0.1925682857632637,
|
|
"num_tokens": 6335156.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.103571271896362,
|
|
"epoch": 2.8702928870292888,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004475396739761299,
|
|
"loss": 4.9568,
|
|
"mean_token_accuracy": 0.20360913574695588,
|
|
"num_tokens": 6344264.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.200181913375855,
|
|
"epoch": 2.8744769874476988,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044733233385281117,
|
|
"loss": 5.0305,
|
|
"mean_token_accuracy": 0.19665719866752623,
|
|
"num_tokens": 6352997.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.154657411575317,
|
|
"epoch": 2.8786610878661087,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044712463909725626,
|
|
"loss": 5.0428,
|
|
"mean_token_accuracy": 0.19684528708457946,
|
|
"num_tokens": 6362312.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.26666111946106,
|
|
"epoch": 2.8828451882845187,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044691659013686727,
|
|
"loss": 5.1514,
|
|
"mean_token_accuracy": 0.19192761927843094,
|
|
"num_tokens": 6372107.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.210444450378418,
|
|
"epoch": 2.8870292887029287,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004467081873997756,
|
|
"loss": 5.0077,
|
|
"mean_token_accuracy": 0.20240505188703536,
|
|
"num_tokens": 6381306.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.093677139282226,
|
|
"epoch": 2.891213389121339,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044649943131484056,
|
|
"loss": 4.9493,
|
|
"mean_token_accuracy": 0.19492731243371964,
|
|
"num_tokens": 6390404.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.341904926300049,
|
|
"epoch": 2.895397489539749,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044629032231164874,
|
|
"loss": 5.1768,
|
|
"mean_token_accuracy": 0.18503012806177138,
|
|
"num_tokens": 6400366.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.21979079246521,
|
|
"epoch": 2.899581589958159,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004460808608205127,
|
|
"loss": 4.9842,
|
|
"mean_token_accuracy": 0.19835661649703978,
|
|
"num_tokens": 6408955.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.040308332443237,
|
|
"epoch": 2.903765690376569,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004458710472724706,
|
|
"loss": 5.0342,
|
|
"mean_token_accuracy": 0.1961667135357857,
|
|
"num_tokens": 6417172.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.2249524116516115,
|
|
"epoch": 2.907949790794979,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044566088209928485,
|
|
"loss": 5.0573,
|
|
"mean_token_accuracy": 0.19124833941459657,
|
|
"num_tokens": 6425898.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.174408960342407,
|
|
"epoch": 2.912133891213389,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044545036573344175,
|
|
"loss": 5.0929,
|
|
"mean_token_accuracy": 0.18862212598323821,
|
|
"num_tokens": 6435940.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.155632066726684,
|
|
"epoch": 2.916317991631799,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044523949860815015,
|
|
"loss": 5.0182,
|
|
"mean_token_accuracy": 0.19935661405324936,
|
|
"num_tokens": 6445119.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.1995045185089115,
|
|
"epoch": 2.920502092050209,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004450282811573405,
|
|
"loss": 5.0589,
|
|
"mean_token_accuracy": 0.19250494092702866,
|
|
"num_tokens": 6453986.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.178009700775147,
|
|
"epoch": 2.924686192468619,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044481671381566446,
|
|
"loss": 5.0339,
|
|
"mean_token_accuracy": 0.1887597605586052,
|
|
"num_tokens": 6463658.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.182079601287842,
|
|
"epoch": 2.928870292887029,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004446047970184937,
|
|
"loss": 5.0742,
|
|
"mean_token_accuracy": 0.1969494640827179,
|
|
"num_tokens": 6472370.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.928870292887029,
|
|
"eval_entropy": 5.0158472987527984,
|
|
"eval_loss": 5.41502046585083,
|
|
"eval_mean_token_accuracy": 0.1876692632803644,
|
|
"eval_num_tokens": 6472370.0,
|
|
"eval_runtime": 2.9064,
|
|
"eval_samples_per_second": 1344.252,
|
|
"eval_steps_per_second": 168.247,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.2194946765899655,
|
|
"epoch": 2.933054393305439,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044439253120191877,
|
|
"loss": 5.1977,
|
|
"mean_token_accuracy": 0.18319174647331238,
|
|
"num_tokens": 6482642.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.2442113876342775,
|
|
"epoch": 2.937238493723849,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044417991680274874,
|
|
"loss": 5.125,
|
|
"mean_token_accuracy": 0.19336890429258347,
|
|
"num_tokens": 6491818.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.307046842575073,
|
|
"epoch": 2.941422594142259,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004439669542585097,
|
|
"loss": 5.0962,
|
|
"mean_token_accuracy": 0.1879289209842682,
|
|
"num_tokens": 6500908.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.150317144393921,
|
|
"epoch": 2.9456066945606696,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004437536440074444,
|
|
"loss": 5.0565,
|
|
"mean_token_accuracy": 0.195574814081192,
|
|
"num_tokens": 6510182.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.144090938568115,
|
|
"epoch": 2.9497907949790796,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004435399864885111,
|
|
"loss": 5.0436,
|
|
"mean_token_accuracy": 0.19778359085321426,
|
|
"num_tokens": 6519051.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.1417313575744625,
|
|
"epoch": 2.9539748953974896,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004433259821413825,
|
|
"loss": 4.9705,
|
|
"mean_token_accuracy": 0.20726521462202072,
|
|
"num_tokens": 6527949.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.336123609542847,
|
|
"epoch": 2.9581589958158996,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044311163140644524,
|
|
"loss": 5.2479,
|
|
"mean_token_accuracy": 0.18443945944309234,
|
|
"num_tokens": 6537227.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.170662975311279,
|
|
"epoch": 2.9623430962343096,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044289693472479867,
|
|
"loss": 5.0409,
|
|
"mean_token_accuracy": 0.1877062126994133,
|
|
"num_tokens": 6546938.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.169513940811157,
|
|
"epoch": 2.9665271966527196,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044268189253825397,
|
|
"loss": 5.0656,
|
|
"mean_token_accuracy": 0.1903521493077278,
|
|
"num_tokens": 6555750.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.170086336135864,
|
|
"epoch": 2.9707112970711296,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004424665052893334,
|
|
"loss": 4.9926,
|
|
"mean_token_accuracy": 0.1998180255293846,
|
|
"num_tokens": 6565680.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.09064679145813,
|
|
"epoch": 2.97489539748954,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044225077342126935,
|
|
"loss": 4.9792,
|
|
"mean_token_accuracy": 0.19831879436969757,
|
|
"num_tokens": 6574595.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.220278739929199,
|
|
"epoch": 2.97907949790795,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004420346973780033,
|
|
"loss": 5.0895,
|
|
"mean_token_accuracy": 0.19558854699134826,
|
|
"num_tokens": 6583333.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.249635028839111,
|
|
"epoch": 2.98326359832636,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000441818277604185,
|
|
"loss": 5.108,
|
|
"mean_token_accuracy": 0.19843464195728303,
|
|
"num_tokens": 6592252.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.1930504322052,
|
|
"epoch": 2.98744769874477,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004416015145451715,
|
|
"loss": 4.9685,
|
|
"mean_token_accuracy": 0.1990971624851227,
|
|
"num_tokens": 6600675.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.126534414291382,
|
|
"epoch": 2.99163179916318,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004413844086470264,
|
|
"loss": 5.0472,
|
|
"mean_token_accuracy": 0.19561843425035477,
|
|
"num_tokens": 6609722.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.208112573623657,
|
|
"epoch": 2.99581589958159,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004411669603565187,
|
|
"loss": 5.048,
|
|
"mean_token_accuracy": 0.1993239313364029,
|
|
"num_tokens": 6618415.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.1942760944366455,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000440949170121122,
|
|
"loss": 5.0552,
|
|
"mean_token_accuracy": 0.20129602700471877,
|
|
"num_tokens": 6627966.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.211493825912475,
|
|
"epoch": 3.00418410041841,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044073103838901373,
|
|
"loss": 4.6836,
|
|
"mean_token_accuracy": 0.21819210201501846,
|
|
"num_tokens": 6636820.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.0570331573486325,
|
|
"epoch": 3.00836820083682,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044051256560907374,
|
|
"loss": 4.7162,
|
|
"mean_token_accuracy": 0.20476453006267548,
|
|
"num_tokens": 6645792.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.208727741241455,
|
|
"epoch": 3.01255230125523,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044029375223088394,
|
|
"loss": 4.8173,
|
|
"mean_token_accuracy": 0.2045898750424385,
|
|
"num_tokens": 6654850.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.100937986373902,
|
|
"epoch": 3.01673640167364,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044007459870472706,
|
|
"loss": 4.7413,
|
|
"mean_token_accuracy": 0.21011182963848113,
|
|
"num_tokens": 6664524.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.067801189422608,
|
|
"epoch": 3.02092050209205,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004398551054815859,
|
|
"loss": 4.6902,
|
|
"mean_token_accuracy": 0.21025084406137468,
|
|
"num_tokens": 6674330.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.150098466873169,
|
|
"epoch": 3.0251046025104604,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004396352730131422,
|
|
"loss": 4.7795,
|
|
"mean_token_accuracy": 0.20670407265424728,
|
|
"num_tokens": 6683121.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.119746398925781,
|
|
"epoch": 3.0292887029288704,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004394151017517758,
|
|
"loss": 4.7442,
|
|
"mean_token_accuracy": 0.2111794278025627,
|
|
"num_tokens": 6692566.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.190159320831299,
|
|
"epoch": 3.0334728033472804,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004391945921505638,
|
|
"loss": 4.8191,
|
|
"mean_token_accuracy": 0.20630361884832382,
|
|
"num_tokens": 6701577.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.087260055541992,
|
|
"epoch": 3.0376569037656904,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043897374466327944,
|
|
"loss": 4.6864,
|
|
"mean_token_accuracy": 0.21943875551223754,
|
|
"num_tokens": 6710672.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.058188581466675,
|
|
"epoch": 3.0418410041841004,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043875255974439145,
|
|
"loss": 4.7054,
|
|
"mean_token_accuracy": 0.20946837812662125,
|
|
"num_tokens": 6719412.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.0682212829589846,
|
|
"epoch": 3.0460251046025104,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004385310378490627,
|
|
"loss": 4.6849,
|
|
"mean_token_accuracy": 0.21102710217237472,
|
|
"num_tokens": 6729169.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.074842214584351,
|
|
"epoch": 3.0502092050209204,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043830917943314977,
|
|
"loss": 4.6289,
|
|
"mean_token_accuracy": 0.21655241549015045,
|
|
"num_tokens": 6738549.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.050983333587647,
|
|
"epoch": 3.0543933054393304,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043808698495320154,
|
|
"loss": 4.7103,
|
|
"mean_token_accuracy": 0.21011070013046265,
|
|
"num_tokens": 6747909.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.100692558288574,
|
|
"epoch": 3.0585774058577404,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043786445486645846,
|
|
"loss": 4.7526,
|
|
"mean_token_accuracy": 0.20967238545417785,
|
|
"num_tokens": 6757125.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.098560047149658,
|
|
"epoch": 3.062761506276151,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004376415896308518,
|
|
"loss": 4.7326,
|
|
"mean_token_accuracy": 0.21780457198619843,
|
|
"num_tokens": 6765774.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.121496105194092,
|
|
"epoch": 3.066945606694561,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043741838970500234,
|
|
"loss": 4.7866,
|
|
"mean_token_accuracy": 0.21280336230993271,
|
|
"num_tokens": 6774980.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.107612562179566,
|
|
"epoch": 3.071129707112971,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004371948555482196,
|
|
"loss": 4.7833,
|
|
"mean_token_accuracy": 0.208768767118454,
|
|
"num_tokens": 6784187.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.117950534820556,
|
|
"epoch": 3.075313807531381,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000436970987620501,
|
|
"loss": 4.8072,
|
|
"mean_token_accuracy": 0.20933396965265275,
|
|
"num_tokens": 6792953.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.092345905303955,
|
|
"epoch": 3.079497907949791,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043674678638253066,
|
|
"loss": 4.6683,
|
|
"mean_token_accuracy": 0.21843459457159042,
|
|
"num_tokens": 6803004.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.047590494155884,
|
|
"epoch": 3.083682008368201,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004365222522956788,
|
|
"loss": 4.7256,
|
|
"mean_token_accuracy": 0.20577909350395202,
|
|
"num_tokens": 6811883.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.083851432800293,
|
|
"epoch": 3.087866108786611,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004362973858220004,
|
|
"loss": 4.7632,
|
|
"mean_token_accuracy": 0.2176697075366974,
|
|
"num_tokens": 6820327.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.108062410354615,
|
|
"epoch": 3.092050209205021,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043607218742423447,
|
|
"loss": 4.7482,
|
|
"mean_token_accuracy": 0.2086792394518852,
|
|
"num_tokens": 6829412.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.044431877136231,
|
|
"epoch": 3.096234309623431,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043584665756580323,
|
|
"loss": 4.7343,
|
|
"mean_token_accuracy": 0.21022414267063141,
|
|
"num_tokens": 6838997.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.154791688919067,
|
|
"epoch": 3.100418410041841,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004356207967108106,
|
|
"loss": 4.7925,
|
|
"mean_token_accuracy": 0.20132009834051132,
|
|
"num_tokens": 6848957.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.0912672996521,
|
|
"epoch": 3.104602510460251,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004353946053240422,
|
|
"loss": 4.8299,
|
|
"mean_token_accuracy": 0.20455342233181,
|
|
"num_tokens": 6859688.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.189009761810302,
|
|
"epoch": 3.1087866108786613,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004351680838709635,
|
|
"loss": 4.8652,
|
|
"mean_token_accuracy": 0.1948360338807106,
|
|
"num_tokens": 6869149.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.1483806610107425,
|
|
"epoch": 3.1129707112970713,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043494123281771915,
|
|
"loss": 4.8375,
|
|
"mean_token_accuracy": 0.19877780228853226,
|
|
"num_tokens": 6879432.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.098960733413696,
|
|
"epoch": 3.1171548117154813,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043471405263113214,
|
|
"loss": 4.7898,
|
|
"mean_token_accuracy": 0.20733687430620193,
|
|
"num_tokens": 6888398.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.14504041671753,
|
|
"epoch": 3.1213389121338913,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004344865437787027,
|
|
"loss": 4.7321,
|
|
"mean_token_accuracy": 0.2048347368836403,
|
|
"num_tokens": 6898082.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.030231094360351,
|
|
"epoch": 3.1255230125523012,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043425870672860774,
|
|
"loss": 4.6984,
|
|
"mean_token_accuracy": 0.21329722702503204,
|
|
"num_tokens": 6907693.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.19649305343628,
|
|
"epoch": 3.1297071129707112,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004340305419496989,
|
|
"loss": 4.8984,
|
|
"mean_token_accuracy": 0.19849764108657836,
|
|
"num_tokens": 6917095.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.134822082519531,
|
|
"epoch": 3.1338912133891212,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004338020499115029,
|
|
"loss": 4.8291,
|
|
"mean_token_accuracy": 0.2039004921913147,
|
|
"num_tokens": 6926952.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.100955867767334,
|
|
"epoch": 3.1380753138075312,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004335732310842196,
|
|
"loss": 4.7791,
|
|
"mean_token_accuracy": 0.20620789825916291,
|
|
"num_tokens": 6936704.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.08941330909729,
|
|
"epoch": 3.1422594142259412,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004333440859387212,
|
|
"loss": 4.7821,
|
|
"mean_token_accuracy": 0.21059116572141648,
|
|
"num_tokens": 6946614.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.145793390274048,
|
|
"epoch": 3.1464435146443517,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043311461494655165,
|
|
"loss": 4.782,
|
|
"mean_token_accuracy": 0.2146619066596031,
|
|
"num_tokens": 6955980.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.103716135025024,
|
|
"epoch": 3.1506276150627617,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004328848185799254,
|
|
"loss": 4.7698,
|
|
"mean_token_accuracy": 0.21150581687688827,
|
|
"num_tokens": 6964108.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.104557275772095,
|
|
"epoch": 3.1548117154811717,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004326546973117264,
|
|
"loss": 4.7649,
|
|
"mean_token_accuracy": 0.220062717795372,
|
|
"num_tokens": 6972895.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.046841621398926,
|
|
"epoch": 3.1589958158995817,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043242425161550723,
|
|
"loss": 4.7917,
|
|
"mean_token_accuracy": 0.21240701228380204,
|
|
"num_tokens": 6982222.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.183664417266845,
|
|
"epoch": 3.1631799163179917,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043219348196548807,
|
|
"loss": 4.7751,
|
|
"mean_token_accuracy": 0.21130128651857377,
|
|
"num_tokens": 6990820.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.09278826713562,
|
|
"epoch": 3.1673640167364017,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004319623888365557,
|
|
"loss": 4.8122,
|
|
"mean_token_accuracy": 0.20467470437288285,
|
|
"num_tokens": 7000846.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.040134286880493,
|
|
"epoch": 3.1715481171548117,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00043173097270426275,
|
|
"loss": 4.7318,
|
|
"mean_token_accuracy": 0.20678451657295227,
|
|
"num_tokens": 7009086.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.200841331481934,
|
|
"epoch": 3.1757322175732217,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043149923404482634,
|
|
"loss": 4.8059,
|
|
"mean_token_accuracy": 0.2062577411532402,
|
|
"num_tokens": 7017752.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.043055582046509,
|
|
"epoch": 3.1799163179916317,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043126717333512733,
|
|
"loss": 4.7963,
|
|
"mean_token_accuracy": 0.21397409290075303,
|
|
"num_tokens": 7026535.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.108639240264893,
|
|
"epoch": 3.1841004184100417,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004310347910527094,
|
|
"loss": 4.841,
|
|
"mean_token_accuracy": 0.20998916625976563,
|
|
"num_tokens": 7035462.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.090834522247315,
|
|
"epoch": 3.1882845188284517,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004308020876757778,
|
|
"loss": 4.7446,
|
|
"mean_token_accuracy": 0.21191719621419908,
|
|
"num_tokens": 7043978.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.031073522567749,
|
|
"epoch": 3.192468619246862,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000430569063683199,
|
|
"loss": 4.7646,
|
|
"mean_token_accuracy": 0.21432637125253678,
|
|
"num_tokens": 7053757.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.021868944168091,
|
|
"epoch": 3.196652719665272,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004303357195544986,
|
|
"loss": 4.6806,
|
|
"mean_token_accuracy": 0.20763231366872786,
|
|
"num_tokens": 7062491.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.03041410446167,
|
|
"epoch": 3.200836820083682,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004301020557698615,
|
|
"loss": 4.7352,
|
|
"mean_token_accuracy": 0.2140173465013504,
|
|
"num_tokens": 7072157.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.011174440383911,
|
|
"epoch": 3.205020920502092,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00042986807281013013,
|
|
"loss": 4.6874,
|
|
"mean_token_accuracy": 0.21297517567873,
|
|
"num_tokens": 7080631.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.092728090286255,
|
|
"epoch": 3.209205020920502,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004296337711568037,
|
|
"loss": 4.7643,
|
|
"mean_token_accuracy": 0.22053097039461136,
|
|
"num_tokens": 7089944.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.103617763519287,
|
|
"epoch": 3.213389121338912,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042939915129203745,
|
|
"loss": 4.8183,
|
|
"mean_token_accuracy": 0.20843600779771804,
|
|
"num_tokens": 7099190.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 4.9565423965454105,
|
|
"epoch": 3.217573221757322,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004291642136986413,
|
|
"loss": 4.7904,
|
|
"mean_token_accuracy": 0.20124946981668473,
|
|
"num_tokens": 7107944.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.029982376098633,
|
|
"epoch": 3.221757322175732,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000428928958860079,
|
|
"loss": 4.6425,
|
|
"mean_token_accuracy": 0.21207152009010316,
|
|
"num_tokens": 7116953.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.028992176055908,
|
|
"epoch": 3.225941422594142,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004286933872604673,
|
|
"loss": 4.6816,
|
|
"mean_token_accuracy": 0.21626634895801544,
|
|
"num_tokens": 7126236.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.052717494964599,
|
|
"epoch": 3.2301255230125525,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004284574993845746,
|
|
"loss": 4.7289,
|
|
"mean_token_accuracy": 0.21213791519403458,
|
|
"num_tokens": 7136286.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.114240503311157,
|
|
"epoch": 3.2343096234309625,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00042822129571782025,
|
|
"loss": 4.8325,
|
|
"mean_token_accuracy": 0.20526242703199388,
|
|
"num_tokens": 7145551.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.149323892593384,
|
|
"epoch": 3.2384937238493725,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042798477674627343,
|
|
"loss": 4.7719,
|
|
"mean_token_accuracy": 0.2100081130862236,
|
|
"num_tokens": 7155486.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.072192573547364,
|
|
"epoch": 3.2426778242677825,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042774794295665213,
|
|
"loss": 4.7613,
|
|
"mean_token_accuracy": 0.21465275585651397,
|
|
"num_tokens": 7163984.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 4.979444456100464,
|
|
"epoch": 3.2468619246861925,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004275107948363223,
|
|
"loss": 4.7418,
|
|
"mean_token_accuracy": 0.20954679995775222,
|
|
"num_tokens": 7172854.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.028590965270996,
|
|
"epoch": 3.2510460251046025,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004272733328732965,
|
|
"loss": 4.7197,
|
|
"mean_token_accuracy": 0.21132369935512543,
|
|
"num_tokens": 7182202.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.0914758205413815,
|
|
"epoch": 3.2552301255230125,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042703555755623337,
|
|
"loss": 4.8512,
|
|
"mean_token_accuracy": 0.19847214967012405,
|
|
"num_tokens": 7192062.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.189693546295166,
|
|
"epoch": 3.2594142259414225,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004267974693744362,
|
|
"loss": 4.8431,
|
|
"mean_token_accuracy": 0.20873661786317826,
|
|
"num_tokens": 7201766.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.098833513259888,
|
|
"epoch": 3.2635983263598325,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004265590688178523,
|
|
"loss": 4.776,
|
|
"mean_token_accuracy": 0.20412208288908004,
|
|
"num_tokens": 7211361.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.025070858001709,
|
|
"epoch": 3.2677824267782425,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004263203563770716,
|
|
"loss": 4.747,
|
|
"mean_token_accuracy": 0.21468700468540192,
|
|
"num_tokens": 7220221.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.058682489395141,
|
|
"epoch": 3.2719665271966525,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000426081332543326,
|
|
"loss": 4.8623,
|
|
"mean_token_accuracy": 0.20202397406101227,
|
|
"num_tokens": 7230134.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.008880710601806,
|
|
"epoch": 3.276150627615063,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004258419978084881,
|
|
"loss": 4.6701,
|
|
"mean_token_accuracy": 0.21683662682771682,
|
|
"num_tokens": 7238713.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.104620313644409,
|
|
"epoch": 3.280334728033473,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004256023526650702,
|
|
"loss": 4.7946,
|
|
"mean_token_accuracy": 0.21614361554384232,
|
|
"num_tokens": 7247603.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.031362915039063,
|
|
"epoch": 3.284518828451883,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004253623976062236,
|
|
"loss": 4.8284,
|
|
"mean_token_accuracy": 0.2112140029668808,
|
|
"num_tokens": 7257452.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.145403671264648,
|
|
"epoch": 3.288702928870293,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004251221331257372,
|
|
"loss": 4.8281,
|
|
"mean_token_accuracy": 0.20674990862607956,
|
|
"num_tokens": 7266401.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.065143775939942,
|
|
"epoch": 3.292887029288703,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004248815597180366,
|
|
"loss": 4.7803,
|
|
"mean_token_accuracy": 0.21541957259178163,
|
|
"num_tokens": 7275836.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.080591487884521,
|
|
"epoch": 3.297071129707113,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042464067787818333,
|
|
"loss": 4.8344,
|
|
"mean_token_accuracy": 0.20549215674400328,
|
|
"num_tokens": 7284854.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.069702911376953,
|
|
"epoch": 3.301255230125523,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004243994881018735,
|
|
"loss": 4.7813,
|
|
"mean_token_accuracy": 0.20480358153581618,
|
|
"num_tokens": 7293421.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.017064380645752,
|
|
"epoch": 3.305439330543933,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004241579908854367,
|
|
"loss": 4.7215,
|
|
"mean_token_accuracy": 0.21256045252084732,
|
|
"num_tokens": 7302184.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.1147784233093265,
|
|
"epoch": 3.309623430962343,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00042391618672583546,
|
|
"loss": 4.8741,
|
|
"mean_token_accuracy": 0.20789355784654617,
|
|
"num_tokens": 7311596.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.076926517486572,
|
|
"epoch": 3.3138075313807533,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00042367407612066383,
|
|
"loss": 4.7326,
|
|
"mean_token_accuracy": 0.21281673312187194,
|
|
"num_tokens": 7319866.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.118884468078614,
|
|
"epoch": 3.3179916317991633,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004234316595681468,
|
|
"loss": 4.7818,
|
|
"mean_token_accuracy": 0.20886427611112596,
|
|
"num_tokens": 7329327.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.213572263717651,
|
|
"epoch": 3.3221757322175733,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004231889375671382,
|
|
"loss": 4.9342,
|
|
"mean_token_accuracy": 0.1930358365178108,
|
|
"num_tokens": 7339717.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.066238975524902,
|
|
"epoch": 3.3263598326359833,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00042294591061712124,
|
|
"loss": 4.7687,
|
|
"mean_token_accuracy": 0.21156300455331803,
|
|
"num_tokens": 7348850.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.01680908203125,
|
|
"epoch": 3.3305439330543933,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00042270257921820623,
|
|
"loss": 4.722,
|
|
"mean_token_accuracy": 0.21210106909275056,
|
|
"num_tokens": 7357081.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 4.942685222625732,
|
|
"epoch": 3.3347280334728033,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00042245894387112987,
|
|
"loss": 4.7343,
|
|
"mean_token_accuracy": 0.21606798470020294,
|
|
"num_tokens": 7365762.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.058714103698731,
|
|
"epoch": 3.3389121338912133,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004222150050772548,
|
|
"loss": 4.8006,
|
|
"mean_token_accuracy": 0.20745150744915009,
|
|
"num_tokens": 7374870.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.03049750328064,
|
|
"epoch": 3.3430962343096233,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004219707633385677,
|
|
"loss": 4.6922,
|
|
"mean_token_accuracy": 0.21497696936130523,
|
|
"num_tokens": 7383836.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.022936248779297,
|
|
"epoch": 3.3472803347280333,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042172621915767886,
|
|
"loss": 4.7849,
|
|
"mean_token_accuracy": 0.2123652830719948,
|
|
"num_tokens": 7394059.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.3472803347280333,
|
|
"eval_entropy": 4.877757190926674,
|
|
"eval_loss": 5.409787178039551,
|
|
"eval_mean_token_accuracy": 0.1900516237857883,
|
|
"eval_num_tokens": 7394059.0,
|
|
"eval_runtime": 2.7297,
|
|
"eval_samples_per_second": 1431.305,
|
|
"eval_steps_per_second": 179.142,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11950,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.0764341692416e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|