741ef3be25
Model: fpadovani/dan-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
18233 lines
498 KiB
JSON
18233 lines
498 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 7.531380753138075,
|
|
"eval_steps": 500,
|
|
"global_step": 9000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 4.993609952926636,
|
|
"epoch": 3.3514644351464433,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042148137303782077,
|
|
"loss": 4.7368,
|
|
"mean_token_accuracy": 0.2144579917192459,
|
|
"num_tokens": 7402947.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.105665397644043,
|
|
"epoch": 3.3556485355648533,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004212362254828475,
|
|
"loss": 4.8052,
|
|
"mean_token_accuracy": 0.20833970457315446,
|
|
"num_tokens": 7413430.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.100651502609253,
|
|
"epoch": 3.3598326359832638,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00042099077699723326,
|
|
"loss": 4.8085,
|
|
"mean_token_accuracy": 0.20969858765602112,
|
|
"num_tokens": 7423239.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.076298809051513,
|
|
"epoch": 3.3640167364016738,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004207450280860716,
|
|
"loss": 4.7965,
|
|
"mean_token_accuracy": 0.2144022911787033,
|
|
"num_tokens": 7432519.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.015933179855347,
|
|
"epoch": 3.3682008368200838,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004204989792550742,
|
|
"loss": 4.7624,
|
|
"mean_token_accuracy": 0.21373432874679565,
|
|
"num_tokens": 7442089.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.068990564346313,
|
|
"epoch": 3.3723849372384938,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00042025263101057016,
|
|
"loss": 4.7727,
|
|
"mean_token_accuracy": 0.2096349060535431,
|
|
"num_tokens": 7451059.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.056183385848999,
|
|
"epoch": 3.3765690376569037,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00042000598385950427,
|
|
"loss": 4.7248,
|
|
"mean_token_accuracy": 0.21488821804523467,
|
|
"num_tokens": 7458960.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.055731344223022,
|
|
"epoch": 3.3807531380753137,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000419759038309437,
|
|
"loss": 4.9144,
|
|
"mean_token_accuracy": 0.19852719753980635,
|
|
"num_tokens": 7468863.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.069086647033691,
|
|
"epoch": 3.3849372384937237,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041951179486854253,
|
|
"loss": 4.8663,
|
|
"mean_token_accuracy": 0.20806890577077866,
|
|
"num_tokens": 7477204.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.062913274765014,
|
|
"epoch": 3.3891213389121337,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000419264254045608,
|
|
"loss": 4.6991,
|
|
"mean_token_accuracy": 0.2156752824783325,
|
|
"num_tokens": 7487367.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 4.937308216094971,
|
|
"epoch": 3.393305439330544,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004190164163500326,
|
|
"loss": 4.6839,
|
|
"mean_token_accuracy": 0.22092362642288207,
|
|
"num_tokens": 7495808.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.063209486007691,
|
|
"epoch": 3.397489539748954,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004187682822918266,
|
|
"loss": 4.8415,
|
|
"mean_token_accuracy": 0.196127949655056,
|
|
"num_tokens": 7506119.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.076227140426636,
|
|
"epoch": 3.401673640167364,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041851985238161,
|
|
"loss": 4.8097,
|
|
"mean_token_accuracy": 0.21152033060789108,
|
|
"num_tokens": 7514931.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 4.931141138076782,
|
|
"epoch": 3.405857740585774,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041827112713061155,
|
|
"loss": 4.6184,
|
|
"mean_token_accuracy": 0.22603769749403,
|
|
"num_tokens": 7523753.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.023400211334229,
|
|
"epoch": 3.410041841004184,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00041802210705066775,
|
|
"loss": 4.6846,
|
|
"mean_token_accuracy": 0.21837748289108277,
|
|
"num_tokens": 7532447.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.028650140762329,
|
|
"epoch": 3.414225941422594,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000417772792654222,
|
|
"loss": 4.8609,
|
|
"mean_token_accuracy": 0.20048442780971526,
|
|
"num_tokens": 7541388.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.0950805187225345,
|
|
"epoch": 3.418410041841004,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00041752318445432323,
|
|
"loss": 4.8386,
|
|
"mean_token_accuracy": 0.19994585514068602,
|
|
"num_tokens": 7550422.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.041496658325196,
|
|
"epoch": 3.422594142259414,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000417273282964625,
|
|
"loss": 4.7069,
|
|
"mean_token_accuracy": 0.21728615313768387,
|
|
"num_tokens": 7559522.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.0843363285064695,
|
|
"epoch": 3.426778242677824,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004170230886993844,
|
|
"loss": 4.8356,
|
|
"mean_token_accuracy": 0.20384197533130646,
|
|
"num_tokens": 7568785.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.010285091400147,
|
|
"epoch": 3.430962343096234,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00041677260217346093,
|
|
"loss": 4.7452,
|
|
"mean_token_accuracy": 0.21137407720088958,
|
|
"num_tokens": 7579013.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.049632740020752,
|
|
"epoch": 3.435146443514644,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041652182390231556,
|
|
"loss": 4.7858,
|
|
"mean_token_accuracy": 0.20622973293066024,
|
|
"num_tokens": 7588577.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.010334253311157,
|
|
"epoch": 3.439330543933054,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041627075440200977,
|
|
"loss": 4.6902,
|
|
"mean_token_accuracy": 0.2165146976709366,
|
|
"num_tokens": 7597702.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 4.999788284301758,
|
|
"epoch": 3.4435146443514646,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004160193941892043,
|
|
"loss": 4.7613,
|
|
"mean_token_accuracy": 0.21234292387962342,
|
|
"num_tokens": 7607156.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.070609045028687,
|
|
"epoch": 3.4476987447698746,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004157677437811578,
|
|
"loss": 4.854,
|
|
"mean_token_accuracy": 0.21513694673776626,
|
|
"num_tokens": 7615936.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.097305202484131,
|
|
"epoch": 3.4518828451882846,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004155158036957265,
|
|
"loss": 4.8185,
|
|
"mean_token_accuracy": 0.20928449630737306,
|
|
"num_tokens": 7626568.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.076028442382812,
|
|
"epoch": 3.4560669456066946,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004152635744513626,
|
|
"loss": 4.7462,
|
|
"mean_token_accuracy": 0.2149614632129669,
|
|
"num_tokens": 7635102.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 4.993189859390259,
|
|
"epoch": 3.4602510460251046,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004150110565671133,
|
|
"loss": 4.6842,
|
|
"mean_token_accuracy": 0.22427398413419725,
|
|
"num_tokens": 7644204.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.047870254516601,
|
|
"epoch": 3.4644351464435146,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00041475825056261985,
|
|
"loss": 4.8064,
|
|
"mean_token_accuracy": 0.20513217896223068,
|
|
"num_tokens": 7653585.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.098278522491455,
|
|
"epoch": 3.4686192468619246,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004145051569581162,
|
|
"loss": 4.8013,
|
|
"mean_token_accuracy": 0.20767406225204468,
|
|
"num_tokens": 7662734.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.119975471496582,
|
|
"epoch": 3.4728033472803346,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004142517762744286,
|
|
"loss": 4.7506,
|
|
"mean_token_accuracy": 0.2178206667304039,
|
|
"num_tokens": 7671682.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.081677865982056,
|
|
"epoch": 3.476987447698745,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004139981090329735,
|
|
"loss": 4.8644,
|
|
"mean_token_accuracy": 0.2072183296084404,
|
|
"num_tokens": 7681551.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.085852479934692,
|
|
"epoch": 3.481171548117155,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004137441557557575,
|
|
"loss": 4.8299,
|
|
"mean_token_accuracy": 0.21125478744506837,
|
|
"num_tokens": 7690943.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 4.946019172668457,
|
|
"epoch": 3.485355648535565,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041348991696537533,
|
|
"loss": 4.7233,
|
|
"mean_token_accuracy": 0.21817297041416167,
|
|
"num_tokens": 7700376.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.079485368728638,
|
|
"epoch": 3.489539748953975,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004132353931850098,
|
|
"loss": 4.8258,
|
|
"mean_token_accuracy": 0.20190587639808655,
|
|
"num_tokens": 7709845.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.1945390701293945,
|
|
"epoch": 3.493723849372385,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004129805849384299,
|
|
"loss": 4.9081,
|
|
"mean_token_accuracy": 0.20315530002117158,
|
|
"num_tokens": 7719908.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.056597280502319,
|
|
"epoch": 3.497907949790795,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004127254927499898,
|
|
"loss": 4.8281,
|
|
"mean_token_accuracy": 0.20524413585662843,
|
|
"num_tokens": 7729543.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.0866728782653805,
|
|
"epoch": 3.502092050209205,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004124701171446286,
|
|
"loss": 4.8428,
|
|
"mean_token_accuracy": 0.20897953808307648,
|
|
"num_tokens": 7739371.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.068806076049805,
|
|
"epoch": 3.506276150627615,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00041221445864786794,
|
|
"loss": 4.7532,
|
|
"mean_token_accuracy": 0.20831515043973922,
|
|
"num_tokens": 7748027.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 4.965265798568725,
|
|
"epoch": 3.510460251046025,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000411958517785812,
|
|
"loss": 4.6876,
|
|
"mean_token_accuracy": 0.21820432543754578,
|
|
"num_tokens": 7756588.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.0286516666412355,
|
|
"epoch": 3.514644351464435,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041170229508514584,
|
|
"loss": 4.6772,
|
|
"mean_token_accuracy": 0.21500977724790574,
|
|
"num_tokens": 7765222.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.058751201629638,
|
|
"epoch": 3.518828451882845,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004114457910731347,
|
|
"loss": 4.7924,
|
|
"mean_token_accuracy": 0.21472012251615524,
|
|
"num_tokens": 7774139.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 4.97785758972168,
|
|
"epoch": 3.523012552301255,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004111890062776226,
|
|
"loss": 4.6607,
|
|
"mean_token_accuracy": 0.22054859399795532,
|
|
"num_tokens": 7783597.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 4.961191606521607,
|
|
"epoch": 3.5271966527196654,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004109319412270314,
|
|
"loss": 4.7664,
|
|
"mean_token_accuracy": 0.21386989206075668,
|
|
"num_tokens": 7792711.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.114382266998291,
|
|
"epoch": 3.5313807531380754,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004106745964503594,
|
|
"loss": 4.8355,
|
|
"mean_token_accuracy": 0.20978693515062333,
|
|
"num_tokens": 7801878.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.0327211856842045,
|
|
"epoch": 3.5355648535564854,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000410416972477181,
|
|
"loss": 4.7233,
|
|
"mean_token_accuracy": 0.21823534667491912,
|
|
"num_tokens": 7810484.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.026141548156739,
|
|
"epoch": 3.5397489539748954,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00041015906983764495,
|
|
"loss": 4.8281,
|
|
"mean_token_accuracy": 0.21025963127613068,
|
|
"num_tokens": 7820521.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.0720508098602295,
|
|
"epoch": 3.5439330543933054,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00040990088906247326,
|
|
"loss": 4.7496,
|
|
"mean_token_accuracy": 0.21724545657634736,
|
|
"num_tokens": 7829218.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.064081144332886,
|
|
"epoch": 3.5481171548117154,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004096424306829605,
|
|
"loss": 4.7762,
|
|
"mean_token_accuracy": 0.2074562832713127,
|
|
"num_tokens": 7837795.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.069561338424682,
|
|
"epoch": 3.5523012552301254,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004093836952309726,
|
|
"loss": 4.7874,
|
|
"mean_token_accuracy": 0.2101425290107727,
|
|
"num_tokens": 7846973.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.097525787353516,
|
|
"epoch": 3.5564853556485354,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004091246832389453,
|
|
"loss": 4.8465,
|
|
"mean_token_accuracy": 0.2080000087618828,
|
|
"num_tokens": 7856627.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.026092576980591,
|
|
"epoch": 3.560669456066946,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004088653952398837,
|
|
"loss": 4.7508,
|
|
"mean_token_accuracy": 0.2151255950331688,
|
|
"num_tokens": 7866308.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 4.95861291885376,
|
|
"epoch": 3.564853556485356,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00040860583176736097,
|
|
"loss": 4.7274,
|
|
"mean_token_accuracy": 0.21694380640983582,
|
|
"num_tokens": 7875350.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.016439533233642,
|
|
"epoch": 3.569037656903766,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004083459933555169,
|
|
"loss": 4.7464,
|
|
"mean_token_accuracy": 0.21901049762964248,
|
|
"num_tokens": 7884276.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.023753213882446,
|
|
"epoch": 3.573221757322176,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004080858805390571,
|
|
"loss": 4.8271,
|
|
"mean_token_accuracy": 0.20734743326902388,
|
|
"num_tokens": 7893836.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.063378524780274,
|
|
"epoch": 3.577405857740586,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000407825493853252,
|
|
"loss": 4.7767,
|
|
"mean_token_accuracy": 0.21605860441923141,
|
|
"num_tokens": 7903249.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.097228908538819,
|
|
"epoch": 3.581589958158996,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040756483383393556,
|
|
"loss": 4.8463,
|
|
"mean_token_accuracy": 0.20450157523155213,
|
|
"num_tokens": 7912217.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.035151529312134,
|
|
"epoch": 3.585774058577406,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004073039010175041,
|
|
"loss": 4.8729,
|
|
"mean_token_accuracy": 0.20275839120149614,
|
|
"num_tokens": 7922241.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.0980524063110355,
|
|
"epoch": 3.589958158995816,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040704269594091556,
|
|
"loss": 4.7877,
|
|
"mean_token_accuracy": 0.20571845918893814,
|
|
"num_tokens": 7931866.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.114587497711182,
|
|
"epoch": 3.594142259414226,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004067812191416878,
|
|
"loss": 4.8265,
|
|
"mean_token_accuracy": 0.20533080101013185,
|
|
"num_tokens": 7941787.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.056887435913086,
|
|
"epoch": 3.598326359832636,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004065194711578982,
|
|
"loss": 4.838,
|
|
"mean_token_accuracy": 0.20451996624469757,
|
|
"num_tokens": 7950832.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.015948104858398,
|
|
"epoch": 3.602510460251046,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004062574525281819,
|
|
"loss": 4.8167,
|
|
"mean_token_accuracy": 0.21464730203151702,
|
|
"num_tokens": 7960059.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.044953775405884,
|
|
"epoch": 3.606694560669456,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00040599516379173113,
|
|
"loss": 4.8152,
|
|
"mean_token_accuracy": 0.20449014902114868,
|
|
"num_tokens": 7969411.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.063703393936157,
|
|
"epoch": 3.6108786610878663,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040573260548829417,
|
|
"loss": 4.7902,
|
|
"mean_token_accuracy": 0.2101280763745308,
|
|
"num_tokens": 7978446.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 4.968495607376099,
|
|
"epoch": 3.6150627615062763,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004054697781581734,
|
|
"loss": 4.694,
|
|
"mean_token_accuracy": 0.21090520024299622,
|
|
"num_tokens": 7987117.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.027896070480347,
|
|
"epoch": 3.6192468619246863,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004052066823422255,
|
|
"loss": 4.7851,
|
|
"mean_token_accuracy": 0.2081789791584015,
|
|
"num_tokens": 7995989.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.006231880187988,
|
|
"epoch": 3.6234309623430963,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004049433185818593,
|
|
"loss": 4.8307,
|
|
"mean_token_accuracy": 0.20475263893604279,
|
|
"num_tokens": 8005486.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 4.999587488174439,
|
|
"epoch": 3.6276150627615062,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040467968741903507,
|
|
"loss": 4.7223,
|
|
"mean_token_accuracy": 0.21311263591051102,
|
|
"num_tokens": 8014853.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.070083856582642,
|
|
"epoch": 3.6317991631799162,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004044157893962633,
|
|
"loss": 4.7561,
|
|
"mean_token_accuracy": 0.21649097204208373,
|
|
"num_tokens": 8023931.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.027342319488525,
|
|
"epoch": 3.6359832635983262,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004041516250566037,
|
|
"loss": 4.7997,
|
|
"mean_token_accuracy": 0.2138889104127884,
|
|
"num_tokens": 8032659.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.076915597915649,
|
|
"epoch": 3.6401673640167362,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040388719494366404,
|
|
"loss": 4.8812,
|
|
"mean_token_accuracy": 0.20516789257526397,
|
|
"num_tokens": 8041857.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.135842609405517,
|
|
"epoch": 3.6443514644351467,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00040362249960159904,
|
|
"loss": 4.8916,
|
|
"mean_token_accuracy": 0.20334802567958832,
|
|
"num_tokens": 8051186.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.051467847824097,
|
|
"epoch": 3.6485355648535567,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00040335753957510906,
|
|
"loss": 4.8108,
|
|
"mean_token_accuracy": 0.19686822444200516,
|
|
"num_tokens": 8060396.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.010056114196777,
|
|
"epoch": 3.6527196652719667,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040309231540943937,
|
|
"loss": 4.7281,
|
|
"mean_token_accuracy": 0.2181757941842079,
|
|
"num_tokens": 8068106.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.07225489616394,
|
|
"epoch": 3.6569037656903767,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004028268276503784,
|
|
"loss": 4.7752,
|
|
"mean_token_accuracy": 0.21034346371889115,
|
|
"num_tokens": 8077878.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.1194242477417,
|
|
"epoch": 3.6610878661087867,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004025610768442576,
|
|
"loss": 4.8956,
|
|
"mean_token_accuracy": 0.2039255127310753,
|
|
"num_tokens": 8086968.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.02964186668396,
|
|
"epoch": 3.6652719665271967,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004022950635379492,
|
|
"loss": 4.6943,
|
|
"mean_token_accuracy": 0.22411167174577712,
|
|
"num_tokens": 8095999.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 4.93168306350708,
|
|
"epoch": 3.6694560669456067,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040202878827886587,
|
|
"loss": 4.7067,
|
|
"mean_token_accuracy": 0.22017838507890702,
|
|
"num_tokens": 8105376.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.064293289184571,
|
|
"epoch": 3.6736401673640167,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004017622516149592,
|
|
"loss": 4.7916,
|
|
"mean_token_accuracy": 0.21099653095006943,
|
|
"num_tokens": 8115024.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.0469403743743895,
|
|
"epoch": 3.6778242677824267,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004014954540947189,
|
|
"loss": 4.8102,
|
|
"mean_token_accuracy": 0.20675877034664153,
|
|
"num_tokens": 8124306.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.030801105499267,
|
|
"epoch": 3.6820083682008367,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040122839626717137,
|
|
"loss": 4.8047,
|
|
"mean_token_accuracy": 0.2138267993927002,
|
|
"num_tokens": 8133151.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.026749897003174,
|
|
"epoch": 3.6861924686192467,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004009610786818786,
|
|
"loss": 4.817,
|
|
"mean_token_accuracy": 0.2109696701169014,
|
|
"num_tokens": 8141745.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 4.979212856292724,
|
|
"epoch": 3.6903765690376567,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00040069350188893735,
|
|
"loss": 4.6452,
|
|
"mean_token_accuracy": 0.2266265034675598,
|
|
"num_tokens": 8150884.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.03546109199524,
|
|
"epoch": 3.694560669456067,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00040042566643897745,
|
|
"loss": 4.7712,
|
|
"mean_token_accuracy": 0.20574879944324492,
|
|
"num_tokens": 8160138.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.07488260269165,
|
|
"epoch": 3.698744769874477,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004001575728831613,
|
|
"loss": 4.8264,
|
|
"mean_token_accuracy": 0.211177359521389,
|
|
"num_tokens": 8169655.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.088115262985229,
|
|
"epoch": 3.702928870292887,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003998892217731824,
|
|
"loss": 4.8383,
|
|
"mean_token_accuracy": 0.21389532834291458,
|
|
"num_tokens": 8178846.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 4.991552209854126,
|
|
"epoch": 3.707112970711297,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000399620613661264,
|
|
"loss": 4.7603,
|
|
"mean_token_accuracy": 0.21337458938360215,
|
|
"num_tokens": 8187525.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.038919067382812,
|
|
"epoch": 3.711297071129707,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003993517491001586,
|
|
"loss": 4.841,
|
|
"mean_token_accuracy": 0.20478158295154572,
|
|
"num_tokens": 8196855.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 4.98480920791626,
|
|
"epoch": 3.715481171548117,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003990826286431462,
|
|
"loss": 4.722,
|
|
"mean_token_accuracy": 0.21002013832330704,
|
|
"num_tokens": 8205530.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.070639991760254,
|
|
"epoch": 3.719665271966527,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003988132528440333,
|
|
"loss": 4.7909,
|
|
"mean_token_accuracy": 0.20910948514938354,
|
|
"num_tokens": 8215066.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.03878698348999,
|
|
"epoch": 3.723849372384937,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003985436222571521,
|
|
"loss": 4.7812,
|
|
"mean_token_accuracy": 0.2042725384235382,
|
|
"num_tokens": 8224585.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.110638809204102,
|
|
"epoch": 3.7280334728033475,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000398273737437359,
|
|
"loss": 4.8531,
|
|
"mean_token_accuracy": 0.20546989440917968,
|
|
"num_tokens": 8233334.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.101515913009644,
|
|
"epoch": 3.7322175732217575,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039800359894003355,
|
|
"loss": 4.7829,
|
|
"mean_token_accuracy": 0.21466459482908248,
|
|
"num_tokens": 8241636.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.046425724029541,
|
|
"epoch": 3.7364016736401675,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003977332073210774,
|
|
"loss": 4.8133,
|
|
"mean_token_accuracy": 0.2069918692111969,
|
|
"num_tokens": 8250273.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.0488715171813965,
|
|
"epoch": 3.7405857740585775,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003974625631369129,
|
|
"loss": 4.835,
|
|
"mean_token_accuracy": 0.20301061868667603,
|
|
"num_tokens": 8260349.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.100539922714233,
|
|
"epoch": 3.7447698744769875,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003971916669444824,
|
|
"loss": 4.8637,
|
|
"mean_token_accuracy": 0.20882807224988936,
|
|
"num_tokens": 8270505.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.1056725025177006,
|
|
"epoch": 3.7489539748953975,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039692051930124663,
|
|
"loss": 4.8123,
|
|
"mean_token_accuracy": 0.20472077578306197,
|
|
"num_tokens": 8278651.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.006243515014648,
|
|
"epoch": 3.7531380753138075,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003966491207651838,
|
|
"loss": 4.7668,
|
|
"mean_token_accuracy": 0.2140067219734192,
|
|
"num_tokens": 8287173.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.110287380218506,
|
|
"epoch": 3.7573221757322175,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00039637747189478866,
|
|
"loss": 4.778,
|
|
"mean_token_accuracy": 0.209299036860466,
|
|
"num_tokens": 8295785.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.0763989925384525,
|
|
"epoch": 3.7615062761506275,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003961055732490708,
|
|
"loss": 4.8719,
|
|
"mean_token_accuracy": 0.20888783037662506,
|
|
"num_tokens": 8305534.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.006777715682984,
|
|
"epoch": 3.7656903765690375,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039583342538755374,
|
|
"loss": 4.7602,
|
|
"mean_token_accuracy": 0.2170412614941597,
|
|
"num_tokens": 8314232.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.7656903765690375,
|
|
"eval_entropy": 4.856396960334544,
|
|
"eval_loss": 5.360367774963379,
|
|
"eval_mean_token_accuracy": 0.1944493100443257,
|
|
"eval_num_tokens": 8314232.0,
|
|
"eval_runtime": 2.7552,
|
|
"eval_samples_per_second": 1418.043,
|
|
"eval_steps_per_second": 177.482,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.063841772079468,
|
|
"epoch": 3.7698744769874475,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039556102887027423,
|
|
"loss": 4.8292,
|
|
"mean_token_accuracy": 0.19938815385103226,
|
|
"num_tokens": 8323230.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.148337936401367,
|
|
"epoch": 3.7740585774058575,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039528838425778053,
|
|
"loss": 4.8697,
|
|
"mean_token_accuracy": 0.20738583654165268,
|
|
"num_tokens": 8333072.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.055735778808594,
|
|
"epoch": 3.778242677824268,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003950154921111313,
|
|
"loss": 4.8257,
|
|
"mean_token_accuracy": 0.20428294986486434,
|
|
"num_tokens": 8342953.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.064943742752075,
|
|
"epoch": 3.782426778242678,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039474235299189484,
|
|
"loss": 4.8126,
|
|
"mean_token_accuracy": 0.20925110280513765,
|
|
"num_tokens": 8351729.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.142831039428711,
|
|
"epoch": 3.786610878661088,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003944689674621474,
|
|
"loss": 4.8689,
|
|
"mean_token_accuracy": 0.20951608270406724,
|
|
"num_tokens": 8360929.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.083317995071411,
|
|
"epoch": 3.790794979079498,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039419533608447244,
|
|
"loss": 4.8488,
|
|
"mean_token_accuracy": 0.209488844871521,
|
|
"num_tokens": 8369506.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.03674201965332,
|
|
"epoch": 3.794979079497908,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003939214594219595,
|
|
"loss": 4.7634,
|
|
"mean_token_accuracy": 0.21465256810188293,
|
|
"num_tokens": 8378583.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.072650480270386,
|
|
"epoch": 3.799163179916318,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039364733803820263,
|
|
"loss": 4.7681,
|
|
"mean_token_accuracy": 0.21140149980783463,
|
|
"num_tokens": 8388078.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 4.981070899963379,
|
|
"epoch": 3.803347280334728,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003933729724972996,
|
|
"loss": 4.7561,
|
|
"mean_token_accuracy": 0.21170564293861388,
|
|
"num_tokens": 8397086.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.030064678192138,
|
|
"epoch": 3.8075313807531384,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039309836336385044,
|
|
"loss": 4.7511,
|
|
"mean_token_accuracy": 0.22283560037612915,
|
|
"num_tokens": 8405621.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.080426549911499,
|
|
"epoch": 3.8117154811715483,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003928235112029568,
|
|
"loss": 4.8225,
|
|
"mean_token_accuracy": 0.20763765275478363,
|
|
"num_tokens": 8415220.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.068022775650024,
|
|
"epoch": 3.8158995815899583,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003925484165802201,
|
|
"loss": 4.8498,
|
|
"mean_token_accuracy": 0.20945118367671967,
|
|
"num_tokens": 8424719.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.032356834411621,
|
|
"epoch": 3.8200836820083683,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003922730800617408,
|
|
"loss": 4.7809,
|
|
"mean_token_accuracy": 0.20876203179359437,
|
|
"num_tokens": 8433844.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 4.902211904525757,
|
|
"epoch": 3.8242677824267783,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003919975022141174,
|
|
"loss": 4.6945,
|
|
"mean_token_accuracy": 0.22155311852693557,
|
|
"num_tokens": 8442700.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.117083787918091,
|
|
"epoch": 3.8284518828451883,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00039172168360444463,
|
|
"loss": 4.9385,
|
|
"mean_token_accuracy": 0.1976716250181198,
|
|
"num_tokens": 8452246.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.074465608596801,
|
|
"epoch": 3.8326359832635983,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039144562480031297,
|
|
"loss": 4.7791,
|
|
"mean_token_accuracy": 0.2151802733540535,
|
|
"num_tokens": 8461977.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.001318168640137,
|
|
"epoch": 3.8368200836820083,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039116932636980685,
|
|
"loss": 4.684,
|
|
"mean_token_accuracy": 0.21847201138734818,
|
|
"num_tokens": 8470124.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.0121293544769285,
|
|
"epoch": 3.8410041841004183,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039089278888150417,
|
|
"loss": 4.835,
|
|
"mean_token_accuracy": 0.20484836399555206,
|
|
"num_tokens": 8479408.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.041266679763794,
|
|
"epoch": 3.8451882845188283,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003906160129044746,
|
|
"loss": 4.7655,
|
|
"mean_token_accuracy": 0.21078479737043382,
|
|
"num_tokens": 8488239.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.079682493209839,
|
|
"epoch": 3.8493723849372383,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003903389990082787,
|
|
"loss": 4.8344,
|
|
"mean_token_accuracy": 0.20453627705574035,
|
|
"num_tokens": 8497690.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.102167701721191,
|
|
"epoch": 3.8535564853556483,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003900617477629663,
|
|
"loss": 4.8918,
|
|
"mean_token_accuracy": 0.20692556500434875,
|
|
"num_tokens": 8506974.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.06046199798584,
|
|
"epoch": 3.8577405857740583,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003897842597390761,
|
|
"loss": 4.8152,
|
|
"mean_token_accuracy": 0.21480900794267654,
|
|
"num_tokens": 8516066.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.052480411529541,
|
|
"epoch": 3.8619246861924688,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003895065355076338,
|
|
"loss": 4.7899,
|
|
"mean_token_accuracy": 0.21002887040376664,
|
|
"num_tokens": 8524747.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 4.999960994720459,
|
|
"epoch": 3.8661087866108788,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003892285756401511,
|
|
"loss": 4.7811,
|
|
"mean_token_accuracy": 0.217883338034153,
|
|
"num_tokens": 8534274.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.066146659851074,
|
|
"epoch": 3.8702928870292888,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003889503807086248,
|
|
"loss": 4.7887,
|
|
"mean_token_accuracy": 0.20365401804447175,
|
|
"num_tokens": 8543453.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.028362226486206,
|
|
"epoch": 3.8744769874476988,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038867195128553536,
|
|
"loss": 4.8727,
|
|
"mean_token_accuracy": 0.21105711609125138,
|
|
"num_tokens": 8553190.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.109710311889648,
|
|
"epoch": 3.8786610878661087,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003883932879438457,
|
|
"loss": 4.7602,
|
|
"mean_token_accuracy": 0.21233774870634078,
|
|
"num_tokens": 8562814.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.003157663345337,
|
|
"epoch": 3.8828451882845187,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003881143912570002,
|
|
"loss": 4.7645,
|
|
"mean_token_accuracy": 0.20913225263357163,
|
|
"num_tokens": 8572131.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 4.998585748672485,
|
|
"epoch": 3.8870292887029287,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003878352617989235,
|
|
"loss": 4.8499,
|
|
"mean_token_accuracy": 0.20043347477912904,
|
|
"num_tokens": 8582325.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.007804298400879,
|
|
"epoch": 3.891213389121339,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003875559001440191,
|
|
"loss": 4.7454,
|
|
"mean_token_accuracy": 0.2162082463502884,
|
|
"num_tokens": 8591704.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.096501350402832,
|
|
"epoch": 3.895397489539749,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003872763068671682,
|
|
"loss": 4.7993,
|
|
"mean_token_accuracy": 0.20613608807325362,
|
|
"num_tokens": 8601136.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.00411286354065,
|
|
"epoch": 3.899581589958159,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003869964825437291,
|
|
"loss": 4.7777,
|
|
"mean_token_accuracy": 0.21378411203622819,
|
|
"num_tokens": 8610491.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.027632713317871,
|
|
"epoch": 3.903765690376569,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038671642774953507,
|
|
"loss": 4.7978,
|
|
"mean_token_accuracy": 0.21232058554887773,
|
|
"num_tokens": 8619199.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.054979562759399,
|
|
"epoch": 3.907949790794979,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003864361430608939,
|
|
"loss": 4.8698,
|
|
"mean_token_accuracy": 0.2062155470252037,
|
|
"num_tokens": 8629291.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.08322582244873,
|
|
"epoch": 3.912133891213389,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003861556290545864,
|
|
"loss": 4.8235,
|
|
"mean_token_accuracy": 0.20892180353403092,
|
|
"num_tokens": 8639212.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.045267772674561,
|
|
"epoch": 3.916317991631799,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038587488630786534,
|
|
"loss": 4.7817,
|
|
"mean_token_accuracy": 0.21108687818050384,
|
|
"num_tokens": 8648351.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.0488951206207275,
|
|
"epoch": 3.920502092050209,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038559391539845413,
|
|
"loss": 4.8246,
|
|
"mean_token_accuracy": 0.20351879745721818,
|
|
"num_tokens": 8657267.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.051149606704712,
|
|
"epoch": 3.924686192468619,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038531271690454567,
|
|
"loss": 4.9604,
|
|
"mean_token_accuracy": 0.20884377509355545,
|
|
"num_tokens": 8666982.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.101274585723877,
|
|
"epoch": 3.928870292887029,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003850312914048015,
|
|
"loss": 4.7387,
|
|
"mean_token_accuracy": 0.21851029098033906,
|
|
"num_tokens": 8676067.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.085463809967041,
|
|
"epoch": 3.933054393305439,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003847496394783496,
|
|
"loss": 4.8127,
|
|
"mean_token_accuracy": 0.21416863203048705,
|
|
"num_tokens": 8685106.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.0263824462890625,
|
|
"epoch": 3.937238493723849,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003844677617047848,
|
|
"loss": 4.866,
|
|
"mean_token_accuracy": 0.20321287363767623,
|
|
"num_tokens": 8695203.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 4.940713119506836,
|
|
"epoch": 3.941422594142259,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000384185658664166,
|
|
"loss": 4.6928,
|
|
"mean_token_accuracy": 0.21624139845371246,
|
|
"num_tokens": 8703700.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.135861539840699,
|
|
"epoch": 3.9456066945606696,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038390333093701596,
|
|
"loss": 4.8301,
|
|
"mean_token_accuracy": 0.20929133147001266,
|
|
"num_tokens": 8713393.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.05723466873169,
|
|
"epoch": 3.9497907949790796,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038362077910431973,
|
|
"loss": 4.7836,
|
|
"mean_token_accuracy": 0.20981893092393875,
|
|
"num_tokens": 8722263.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.015006256103516,
|
|
"epoch": 3.9539748953974896,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003833380037475236,
|
|
"loss": 4.809,
|
|
"mean_token_accuracy": 0.20997167527675628,
|
|
"num_tokens": 8732190.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.074105072021484,
|
|
"epoch": 3.9581589958158996,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038305500544853375,
|
|
"loss": 4.7624,
|
|
"mean_token_accuracy": 0.21410872787237167,
|
|
"num_tokens": 8740995.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.076176452636719,
|
|
"epoch": 3.9623430962343096,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038277178478971527,
|
|
"loss": 4.9067,
|
|
"mean_token_accuracy": 0.20255099684000016,
|
|
"num_tokens": 8750768.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.015133428573608,
|
|
"epoch": 3.9665271966527196,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003824883423538905,
|
|
"loss": 4.8114,
|
|
"mean_token_accuracy": 0.21436353772878647,
|
|
"num_tokens": 8760070.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.013438892364502,
|
|
"epoch": 3.9707112970711296,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038220467872433875,
|
|
"loss": 4.7873,
|
|
"mean_token_accuracy": 0.2172729790210724,
|
|
"num_tokens": 8769313.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.033514595031738,
|
|
"epoch": 3.97489539748954,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003819207944847939,
|
|
"loss": 4.7642,
|
|
"mean_token_accuracy": 0.2136005148291588,
|
|
"num_tokens": 8778328.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.0747950077056885,
|
|
"epoch": 3.97907949790795,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038163669021944413,
|
|
"loss": 4.7895,
|
|
"mean_token_accuracy": 0.2115177646279335,
|
|
"num_tokens": 8787408.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.020825815200806,
|
|
"epoch": 3.98326359832636,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003813523665129304,
|
|
"loss": 4.7446,
|
|
"mean_token_accuracy": 0.21196018606424333,
|
|
"num_tokens": 8796781.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.060452508926391,
|
|
"epoch": 3.98744769874477,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003810678239503452,
|
|
"loss": 4.7941,
|
|
"mean_token_accuracy": 0.21185942739248276,
|
|
"num_tokens": 8806965.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.001754093170166,
|
|
"epoch": 3.99163179916318,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003807830631172312,
|
|
"loss": 4.8172,
|
|
"mean_token_accuracy": 0.20775340050458907,
|
|
"num_tokens": 8817348.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 4.993555116653442,
|
|
"epoch": 3.99581589958159,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038049808459958065,
|
|
"loss": 4.8523,
|
|
"mean_token_accuracy": 0.21119900196790695,
|
|
"num_tokens": 8827304.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.034407329559326,
|
|
"epoch": 4.0,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038021288898383343,
|
|
"loss": 4.7729,
|
|
"mean_token_accuracy": 0.213047394156456,
|
|
"num_tokens": 8837288.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.053593969345092,
|
|
"epoch": 4.00418410041841,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003799274768568761,
|
|
"loss": 4.5487,
|
|
"mean_token_accuracy": 0.22916788309812547,
|
|
"num_tokens": 8846447.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 4.994427728652954,
|
|
"epoch": 4.00836820083682,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037964184880604125,
|
|
"loss": 4.4743,
|
|
"mean_token_accuracy": 0.24118888080120088,
|
|
"num_tokens": 8855990.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 4.947293853759765,
|
|
"epoch": 4.01255230125523,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037935600541910523,
|
|
"loss": 4.4752,
|
|
"mean_token_accuracy": 0.23137906044721604,
|
|
"num_tokens": 8864949.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 4.91449613571167,
|
|
"epoch": 4.01673640167364,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003790699472842879,
|
|
"loss": 4.4264,
|
|
"mean_token_accuracy": 0.2373172402381897,
|
|
"num_tokens": 8873501.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 4.963536596298217,
|
|
"epoch": 4.02092050209205,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037878367499025086,
|
|
"loss": 4.5269,
|
|
"mean_token_accuracy": 0.229280586540699,
|
|
"num_tokens": 8882920.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 4.961940431594849,
|
|
"epoch": 4.02510460251046,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003784971891260965,
|
|
"loss": 4.4661,
|
|
"mean_token_accuracy": 0.2305803582072258,
|
|
"num_tokens": 8893617.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.012513589859009,
|
|
"epoch": 4.02928870292887,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00037821049028136664,
|
|
"loss": 4.4641,
|
|
"mean_token_accuracy": 0.22554513365030288,
|
|
"num_tokens": 8902558.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 4.951485061645508,
|
|
"epoch": 4.03347280334728,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003779235790460414,
|
|
"loss": 4.5002,
|
|
"mean_token_accuracy": 0.23347299993038179,
|
|
"num_tokens": 8912477.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 4.948047494888305,
|
|
"epoch": 4.03765690376569,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003776364560105379,
|
|
"loss": 4.5185,
|
|
"mean_token_accuracy": 0.22958608269691466,
|
|
"num_tokens": 8922313.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 4.998057508468628,
|
|
"epoch": 4.0418410041841,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00037734912176570935,
|
|
"loss": 4.4467,
|
|
"mean_token_accuracy": 0.23608748763799667,
|
|
"num_tokens": 8932011.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 4.991758012771607,
|
|
"epoch": 4.046025104602511,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037706157690284335,
|
|
"loss": 4.5099,
|
|
"mean_token_accuracy": 0.22585823386907578,
|
|
"num_tokens": 8941022.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 4.94198751449585,
|
|
"epoch": 4.050209205020921,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037677382201366103,
|
|
"loss": 4.4981,
|
|
"mean_token_accuracy": 0.23595688492059708,
|
|
"num_tokens": 8950825.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 4.986269998550415,
|
|
"epoch": 4.054393305439331,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037648585769031555,
|
|
"loss": 4.5485,
|
|
"mean_token_accuracy": 0.2197233334183693,
|
|
"num_tokens": 8961035.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.011518478393555,
|
|
"epoch": 4.058577405857741,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003761976845253915,
|
|
"loss": 4.5528,
|
|
"mean_token_accuracy": 0.22383718192577362,
|
|
"num_tokens": 8970535.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 4.905951309204101,
|
|
"epoch": 4.062761506276151,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003759093031119025,
|
|
"loss": 4.5113,
|
|
"mean_token_accuracy": 0.22325447350740432,
|
|
"num_tokens": 8979404.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.033632373809814,
|
|
"epoch": 4.066945606694561,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037562071404329147,
|
|
"loss": 4.4952,
|
|
"mean_token_accuracy": 0.23838532865047454,
|
|
"num_tokens": 8989164.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.07048921585083,
|
|
"epoch": 4.071129707112971,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00037533191791342817,
|
|
"loss": 4.5638,
|
|
"mean_token_accuracy": 0.21892854273319245,
|
|
"num_tokens": 8998630.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 4.968783664703369,
|
|
"epoch": 4.075313807531381,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037504291531660866,
|
|
"loss": 4.4851,
|
|
"mean_token_accuracy": 0.23522623181343078,
|
|
"num_tokens": 9008235.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 4.971577596664429,
|
|
"epoch": 4.079497907949791,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003747537068475538,
|
|
"loss": 4.5919,
|
|
"mean_token_accuracy": 0.22783499211072922,
|
|
"num_tokens": 9018175.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.081594848632813,
|
|
"epoch": 4.083682008368201,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003744642931014082,
|
|
"loss": 4.6051,
|
|
"mean_token_accuracy": 0.22275071144104003,
|
|
"num_tokens": 9028159.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 4.932556581497193,
|
|
"epoch": 4.087866108786611,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003741746746737386,
|
|
"loss": 4.5356,
|
|
"mean_token_accuracy": 0.22377136200666428,
|
|
"num_tokens": 9038263.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 4.969483661651611,
|
|
"epoch": 4.092050209205021,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003738848521605333,
|
|
"loss": 4.5161,
|
|
"mean_token_accuracy": 0.23186618387699126,
|
|
"num_tokens": 9048023.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 4.970028448104858,
|
|
"epoch": 4.096234309623431,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003735948261582005,
|
|
"loss": 4.5876,
|
|
"mean_token_accuracy": 0.21888670772314073,
|
|
"num_tokens": 9056988.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 4.975243854522705,
|
|
"epoch": 4.100418410041841,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000373304597263567,
|
|
"loss": 4.5482,
|
|
"mean_token_accuracy": 0.22196453511714936,
|
|
"num_tokens": 9066595.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 4.94406681060791,
|
|
"epoch": 4.104602510460251,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00037301416607387716,
|
|
"loss": 4.575,
|
|
"mean_token_accuracy": 0.22717536389827728,
|
|
"num_tokens": 9075034.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 4.9102370262146,
|
|
"epoch": 4.108786610878661,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037272353318679185,
|
|
"loss": 4.5327,
|
|
"mean_token_accuracy": 0.23529067188501357,
|
|
"num_tokens": 9084574.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 4.914450550079346,
|
|
"epoch": 4.112970711297071,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003724326992003867,
|
|
"loss": 4.4934,
|
|
"mean_token_accuracy": 0.23602002263069152,
|
|
"num_tokens": 9093591.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 4.962212657928466,
|
|
"epoch": 4.117154811715481,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003721416647131512,
|
|
"loss": 4.5147,
|
|
"mean_token_accuracy": 0.2335410639643669,
|
|
"num_tokens": 9102439.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 4.9085548400878904,
|
|
"epoch": 4.121338912133891,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003718504303239876,
|
|
"loss": 4.4009,
|
|
"mean_token_accuracy": 0.2403297558426857,
|
|
"num_tokens": 9111363.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.019706344604492,
|
|
"epoch": 4.125523012552302,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003715589966322097,
|
|
"loss": 4.5898,
|
|
"mean_token_accuracy": 0.22311563640832902,
|
|
"num_tokens": 9120417.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 4.992297077178955,
|
|
"epoch": 4.129707112970712,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003712673642375409,
|
|
"loss": 4.5518,
|
|
"mean_token_accuracy": 0.21806425899267196,
|
|
"num_tokens": 9130622.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 4.962204027175903,
|
|
"epoch": 4.133891213389122,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003709755337401138,
|
|
"loss": 4.5505,
|
|
"mean_token_accuracy": 0.2274765580892563,
|
|
"num_tokens": 9139992.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 4.898878526687622,
|
|
"epoch": 4.138075313807532,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037068350574046867,
|
|
"loss": 4.5144,
|
|
"mean_token_accuracy": 0.2300298035144806,
|
|
"num_tokens": 9149883.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 4.9406603336334225,
|
|
"epoch": 4.142259414225942,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037039128083955227,
|
|
"loss": 4.4706,
|
|
"mean_token_accuracy": 0.23780096769332887,
|
|
"num_tokens": 9159633.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 4.930354928970337,
|
|
"epoch": 4.146443514644352,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003700988596387163,
|
|
"loss": 4.4665,
|
|
"mean_token_accuracy": 0.22847926318645478,
|
|
"num_tokens": 9167780.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 4.931838417053223,
|
|
"epoch": 4.150627615062762,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003698062427397168,
|
|
"loss": 4.5047,
|
|
"mean_token_accuracy": 0.23294421136379242,
|
|
"num_tokens": 9176808.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.000950813293457,
|
|
"epoch": 4.154811715481172,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036951343074471217,
|
|
"loss": 4.5465,
|
|
"mean_token_accuracy": 0.22885906398296357,
|
|
"num_tokens": 9186097.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.005588245391846,
|
|
"epoch": 4.158995815899582,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003692204242562624,
|
|
"loss": 4.5574,
|
|
"mean_token_accuracy": 0.21918788999319078,
|
|
"num_tokens": 9194428.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 4.951692819595337,
|
|
"epoch": 4.163179916317992,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003689272238773278,
|
|
"loss": 4.5451,
|
|
"mean_token_accuracy": 0.2248445048928261,
|
|
"num_tokens": 9203156.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 4.907498359680176,
|
|
"epoch": 4.167364016736402,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00036863383021126756,
|
|
"loss": 4.5234,
|
|
"mean_token_accuracy": 0.2332322493195534,
|
|
"num_tokens": 9212102.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 4.938442373275757,
|
|
"epoch": 4.171548117154812,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036834024386183875,
|
|
"loss": 4.573,
|
|
"mean_token_accuracy": 0.22472037076950074,
|
|
"num_tokens": 9221507.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 4.936264753341675,
|
|
"epoch": 4.175732217573222,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003680464654331947,
|
|
"loss": 4.4815,
|
|
"mean_token_accuracy": 0.2330555185675621,
|
|
"num_tokens": 9230543.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.002370309829712,
|
|
"epoch": 4.179916317991632,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003677524955298844,
|
|
"loss": 4.5885,
|
|
"mean_token_accuracy": 0.22652823626995086,
|
|
"num_tokens": 9240092.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 4.934385442733765,
|
|
"epoch": 4.184100418410042,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003674583347568505,
|
|
"loss": 4.5355,
|
|
"mean_token_accuracy": 0.22651159912347793,
|
|
"num_tokens": 9250145.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.184100418410042,
|
|
"eval_entropy": 4.787722891337545,
|
|
"eval_loss": 5.353400230407715,
|
|
"eval_mean_token_accuracy": 0.19547097864875032,
|
|
"eval_num_tokens": 9250145.0,
|
|
"eval_runtime": 2.7702,
|
|
"eval_samples_per_second": 1410.359,
|
|
"eval_steps_per_second": 176.52,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 4.9245908737182615,
|
|
"epoch": 4.188284518828452,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036716398371942866,
|
|
"loss": 4.5361,
|
|
"mean_token_accuracy": 0.22549083232879638,
|
|
"num_tokens": 9259643.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 4.977786493301392,
|
|
"epoch": 4.192468619246862,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003668694430233459,
|
|
"loss": 4.552,
|
|
"mean_token_accuracy": 0.22524666041135788,
|
|
"num_tokens": 9268471.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 4.988661432266236,
|
|
"epoch": 4.196652719665272,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003665747132747198,
|
|
"loss": 4.4927,
|
|
"mean_token_accuracy": 0.22485900223255156,
|
|
"num_tokens": 9276916.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 4.874058437347412,
|
|
"epoch": 4.200836820083682,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00036627979508005654,
|
|
"loss": 4.5131,
|
|
"mean_token_accuracy": 0.22951095551252365,
|
|
"num_tokens": 9285867.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 4.993943166732788,
|
|
"epoch": 4.205020920502092,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003659846890462506,
|
|
"loss": 4.6229,
|
|
"mean_token_accuracy": 0.22107454687356948,
|
|
"num_tokens": 9295044.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.047410535812378,
|
|
"epoch": 4.209205020920502,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003656893957805826,
|
|
"loss": 4.631,
|
|
"mean_token_accuracy": 0.22708899974823,
|
|
"num_tokens": 9303744.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 4.959535789489746,
|
|
"epoch": 4.2133891213389125,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003653939158907186,
|
|
"loss": 4.518,
|
|
"mean_token_accuracy": 0.22757700383663176,
|
|
"num_tokens": 9312996.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 4.894146537780761,
|
|
"epoch": 4.2175732217573225,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003650982499847088,
|
|
"loss": 4.5114,
|
|
"mean_token_accuracy": 0.23188632279634475,
|
|
"num_tokens": 9321912.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 4.950233316421508,
|
|
"epoch": 4.2217573221757325,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000364802398670986,
|
|
"loss": 4.5269,
|
|
"mean_token_accuracy": 0.2280694454908371,
|
|
"num_tokens": 9331798.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 4.9599090099334715,
|
|
"epoch": 4.2259414225941425,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036450636255836474,
|
|
"loss": 4.6094,
|
|
"mean_token_accuracy": 0.21910071223974228,
|
|
"num_tokens": 9342064.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 4.9794921875,
|
|
"epoch": 4.2301255230125525,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00036421014225603964,
|
|
"loss": 4.5078,
|
|
"mean_token_accuracy": 0.22316126078367232,
|
|
"num_tokens": 9351494.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 4.866561603546143,
|
|
"epoch": 4.2343096234309625,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00036391373837358455,
|
|
"loss": 4.4613,
|
|
"mean_token_accuracy": 0.23453741669654846,
|
|
"num_tokens": 9360683.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 4.896189785003662,
|
|
"epoch": 4.2384937238493725,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00036361715152095095,
|
|
"loss": 4.5514,
|
|
"mean_token_accuracy": 0.22136776894330978,
|
|
"num_tokens": 9370784.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.004334735870361,
|
|
"epoch": 4.2426778242677825,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003633203823084669,
|
|
"loss": 4.5884,
|
|
"mean_token_accuracy": 0.22848763018846513,
|
|
"num_tokens": 9380673.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 4.97426061630249,
|
|
"epoch": 4.2468619246861925,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00036302343134683566,
|
|
"loss": 4.577,
|
|
"mean_token_accuracy": 0.2269229605793953,
|
|
"num_tokens": 9390260.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.0343505859375,
|
|
"epoch": 4.2510460251046025,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003627262992471348,
|
|
"loss": 4.6334,
|
|
"mean_token_accuracy": 0.21926866620779037,
|
|
"num_tokens": 9400643.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 4.924920225143433,
|
|
"epoch": 4.2552301255230125,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003624289866208142,
|
|
"loss": 4.4859,
|
|
"mean_token_accuracy": 0.2302759975194931,
|
|
"num_tokens": 9408898.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 4.933794450759888,
|
|
"epoch": 4.2594142259414225,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036213149407969547,
|
|
"loss": 4.5497,
|
|
"mean_token_accuracy": 0.2214295580983162,
|
|
"num_tokens": 9418024.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.0673564910888675,
|
|
"epoch": 4.2635983263598325,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003618338222359705,
|
|
"loss": 4.6329,
|
|
"mean_token_accuracy": 0.21673974841833116,
|
|
"num_tokens": 9427548.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 4.953032064437866,
|
|
"epoch": 4.2677824267782425,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036153597170220005,
|
|
"loss": 4.4812,
|
|
"mean_token_accuracy": 0.23158632218837738,
|
|
"num_tokens": 9436370.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 4.96004056930542,
|
|
"epoch": 4.2719665271966525,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003612379430913126,
|
|
"loss": 4.5404,
|
|
"mean_token_accuracy": 0.22605469673871995,
|
|
"num_tokens": 9445506.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 4.952097034454345,
|
|
"epoch": 4.2761506276150625,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003609397370166031,
|
|
"loss": 4.471,
|
|
"mean_token_accuracy": 0.23710474520921707,
|
|
"num_tokens": 9455288.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 4.875089406967163,
|
|
"epoch": 4.2803347280334725,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003606413540917319,
|
|
"loss": 4.5936,
|
|
"mean_token_accuracy": 0.23008251190185547,
|
|
"num_tokens": 9464366.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 4.966262769699097,
|
|
"epoch": 4.2845188284518825,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003603427949307228,
|
|
"loss": 4.5277,
|
|
"mean_token_accuracy": 0.2371608644723892,
|
|
"num_tokens": 9474205.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 4.896690082550049,
|
|
"epoch": 4.2887029288702925,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036004406014796274,
|
|
"loss": 4.5155,
|
|
"mean_token_accuracy": 0.23245862871408463,
|
|
"num_tokens": 9483181.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 4.912990093231201,
|
|
"epoch": 4.292887029288703,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003597451503581998,
|
|
"loss": 4.5148,
|
|
"mean_token_accuracy": 0.23060404062271117,
|
|
"num_tokens": 9492394.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 4.972609329223633,
|
|
"epoch": 4.297071129707113,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035944606617654236,
|
|
"loss": 4.5877,
|
|
"mean_token_accuracy": 0.21716731041669846,
|
|
"num_tokens": 9502184.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 4.9294360160827635,
|
|
"epoch": 4.301255230125523,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00035914680821845746,
|
|
"loss": 4.5731,
|
|
"mean_token_accuracy": 0.22622814774513245,
|
|
"num_tokens": 9511399.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 4.9366601467132565,
|
|
"epoch": 4.305439330543933,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00035884737709976995,
|
|
"loss": 4.5773,
|
|
"mean_token_accuracy": 0.2281072437763214,
|
|
"num_tokens": 9520710.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 4.974839448928833,
|
|
"epoch": 4.309623430962343,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00035854777343666093,
|
|
"loss": 4.4458,
|
|
"mean_token_accuracy": 0.24106617718935014,
|
|
"num_tokens": 9529992.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 4.889826726913452,
|
|
"epoch": 4.313807531380753,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003582479978456665,
|
|
"loss": 4.4613,
|
|
"mean_token_accuracy": 0.23866698741912842,
|
|
"num_tokens": 9538279.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 4.861593389511109,
|
|
"epoch": 4.317991631799163,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003579480509436766,
|
|
"loss": 4.4841,
|
|
"mean_token_accuracy": 0.23682813048362733,
|
|
"num_tokens": 9547425.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 4.923556137084961,
|
|
"epoch": 4.322175732217573,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00035764793334793393,
|
|
"loss": 4.5156,
|
|
"mean_token_accuracy": 0.22888800650835037,
|
|
"num_tokens": 9556657.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 4.949155473709107,
|
|
"epoch": 4.326359832635983,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035734764567603216,
|
|
"loss": 4.5808,
|
|
"mean_token_accuracy": 0.2253078281879425,
|
|
"num_tokens": 9565989.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 4.9176538467407225,
|
|
"epoch": 4.330543933054393,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003570471885459149,
|
|
"loss": 4.5245,
|
|
"mean_token_accuracy": 0.22859694957733154,
|
|
"num_tokens": 9575246.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 4.9070662498474125,
|
|
"epoch": 4.334728033472803,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003567465625758748,
|
|
"loss": 4.4669,
|
|
"mean_token_accuracy": 0.23719528317451477,
|
|
"num_tokens": 9586185.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 4.885430812835693,
|
|
"epoch": 4.338912133891213,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003564457683845516,
|
|
"loss": 4.4634,
|
|
"mean_token_accuracy": 0.2344937413930893,
|
|
"num_tokens": 9595222.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 4.9916849613189695,
|
|
"epoch": 4.343096234309623,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003561448065909315,
|
|
"loss": 4.6281,
|
|
"mean_token_accuracy": 0.23004246950149537,
|
|
"num_tokens": 9604866.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 4.885975408554077,
|
|
"epoch": 4.347280334728033,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003558436778143455,
|
|
"loss": 4.4743,
|
|
"mean_token_accuracy": 0.23645812422037124,
|
|
"num_tokens": 9614522.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 4.9504491806030275,
|
|
"epoch": 4.351464435146443,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003555423826744682,
|
|
"loss": 4.6245,
|
|
"mean_token_accuracy": 0.21740238964557648,
|
|
"num_tokens": 9623308.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 4.992205429077148,
|
|
"epoch": 4.355648535564853,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035524092179131645,
|
|
"loss": 4.6094,
|
|
"mean_token_accuracy": 0.22325228303670883,
|
|
"num_tokens": 9632887.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 4.949487113952637,
|
|
"epoch": 4.359832635983263,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003549392957852485,
|
|
"loss": 4.5811,
|
|
"mean_token_accuracy": 0.22403300404548646,
|
|
"num_tokens": 9643218.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 4.955694484710693,
|
|
"epoch": 4.364016736401673,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003546375052769619,
|
|
"loss": 4.5883,
|
|
"mean_token_accuracy": 0.2183682516217232,
|
|
"num_tokens": 9653175.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 4.874564266204834,
|
|
"epoch": 4.368200836820083,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003543355508874932,
|
|
"loss": 4.4708,
|
|
"mean_token_accuracy": 0.23744809180498122,
|
|
"num_tokens": 9662431.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 4.963984584808349,
|
|
"epoch": 4.372384937238493,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003540334332382159,
|
|
"loss": 4.52,
|
|
"mean_token_accuracy": 0.2345571920275688,
|
|
"num_tokens": 9671191.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 4.869043016433716,
|
|
"epoch": 4.376569037656903,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003537311529508396,
|
|
"loss": 4.4752,
|
|
"mean_token_accuracy": 0.22793266773223878,
|
|
"num_tokens": 9680637.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 4.920653581619263,
|
|
"epoch": 4.380753138075314,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035342871064740846,
|
|
"loss": 4.4451,
|
|
"mean_token_accuracy": 0.22920337170362473,
|
|
"num_tokens": 9689325.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 4.9706339836120605,
|
|
"epoch": 4.384937238493724,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035312610695030017,
|
|
"loss": 4.5639,
|
|
"mean_token_accuracy": 0.2256588190793991,
|
|
"num_tokens": 9698335.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 4.9091246128082275,
|
|
"epoch": 4.389121338912134,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003528233424822245,
|
|
"loss": 4.4885,
|
|
"mean_token_accuracy": 0.23156821727752686,
|
|
"num_tokens": 9706647.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.030884647369385,
|
|
"epoch": 4.393305439330544,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000352520417866222,
|
|
"loss": 4.5348,
|
|
"mean_token_accuracy": 0.23225063234567642,
|
|
"num_tokens": 9715941.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 4.9036924839019775,
|
|
"epoch": 4.397489539748954,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003522173337256628,
|
|
"loss": 4.6075,
|
|
"mean_token_accuracy": 0.2217849761247635,
|
|
"num_tokens": 9724693.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 4.934817123413086,
|
|
"epoch": 4.401673640167364,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003519140906842455,
|
|
"loss": 4.5929,
|
|
"mean_token_accuracy": 0.22499735057353973,
|
|
"num_tokens": 9733314.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.066191291809082,
|
|
"epoch": 4.405857740585774,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003516106893659953,
|
|
"loss": 4.6252,
|
|
"mean_token_accuracy": 0.22578448057174683,
|
|
"num_tokens": 9742844.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 4.905662631988525,
|
|
"epoch": 4.410041841004184,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003513071303952635,
|
|
"loss": 4.505,
|
|
"mean_token_accuracy": 0.2264240249991417,
|
|
"num_tokens": 9752164.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 4.807758045196533,
|
|
"epoch": 4.414225941422594,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003510034143967256,
|
|
"loss": 4.5114,
|
|
"mean_token_accuracy": 0.22810776829719542,
|
|
"num_tokens": 9761483.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 4.897205114364624,
|
|
"epoch": 4.418410041841004,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035069954199538027,
|
|
"loss": 4.4398,
|
|
"mean_token_accuracy": 0.23503599017858506,
|
|
"num_tokens": 9769998.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 4.970601415634155,
|
|
"epoch": 4.422594142259414,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035039551381654816,
|
|
"loss": 4.534,
|
|
"mean_token_accuracy": 0.23068057149648666,
|
|
"num_tokens": 9778965.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 4.960281753540039,
|
|
"epoch": 4.426778242677824,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035009133048587024,
|
|
"loss": 4.568,
|
|
"mean_token_accuracy": 0.22838028371334076,
|
|
"num_tokens": 9788543.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 4.8813621520996096,
|
|
"epoch": 4.430962343096234,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003497869926293069,
|
|
"loss": 4.4337,
|
|
"mean_token_accuracy": 0.23523009121417998,
|
|
"num_tokens": 9797744.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 4.835930013656617,
|
|
"epoch": 4.435146443514644,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003494825008731367,
|
|
"loss": 4.5094,
|
|
"mean_token_accuracy": 0.23088685423135757,
|
|
"num_tokens": 9807528.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 4.8468725204467775,
|
|
"epoch": 4.439330543933054,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034917785584395457,
|
|
"loss": 4.5283,
|
|
"mean_token_accuracy": 0.22365168929100038,
|
|
"num_tokens": 9816078.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 4.959681367874145,
|
|
"epoch": 4.443514644351464,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000348873058168671,
|
|
"loss": 4.4843,
|
|
"mean_token_accuracy": 0.22954466044902802,
|
|
"num_tokens": 9825199.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 4.989903593063355,
|
|
"epoch": 4.447698744769874,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003485681084745106,
|
|
"loss": 4.632,
|
|
"mean_token_accuracy": 0.22501065135002135,
|
|
"num_tokens": 9835294.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 4.889072322845459,
|
|
"epoch": 4.451882845188284,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000348263007389011,
|
|
"loss": 4.4998,
|
|
"mean_token_accuracy": 0.22612627744674682,
|
|
"num_tokens": 9844449.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 4.916523885726929,
|
|
"epoch": 4.456066945606695,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00034795775554002093,
|
|
"loss": 4.533,
|
|
"mean_token_accuracy": 0.2318318083882332,
|
|
"num_tokens": 9852930.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 4.8952912330627445,
|
|
"epoch": 4.460251046025105,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00034765235355569974,
|
|
"loss": 4.4606,
|
|
"mean_token_accuracy": 0.23591193109750747,
|
|
"num_tokens": 9862168.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 4.9487316608428955,
|
|
"epoch": 4.464435146443515,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003473468020645155,
|
|
"loss": 4.6029,
|
|
"mean_token_accuracy": 0.22125384360551834,
|
|
"num_tokens": 9871702.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.008761930465698,
|
|
"epoch": 4.468619246861925,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003470411016952442,
|
|
"loss": 4.6067,
|
|
"mean_token_accuracy": 0.22733209580183028,
|
|
"num_tokens": 9880502.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.009288597106933,
|
|
"epoch": 4.472803347280335,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000346735253076968,
|
|
"loss": 4.5437,
|
|
"mean_token_accuracy": 0.22945308983325957,
|
|
"num_tokens": 9889093.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 4.841006278991699,
|
|
"epoch": 4.476987447698745,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003464292568390743,
|
|
"loss": 4.4426,
|
|
"mean_token_accuracy": 0.23427850902080535,
|
|
"num_tokens": 9897113.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 4.813449192047119,
|
|
"epoch": 4.481171548117155,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003461231136112539,
|
|
"loss": 4.4944,
|
|
"mean_token_accuracy": 0.23093901425600052,
|
|
"num_tokens": 9906478.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 4.971549940109253,
|
|
"epoch": 4.485355648535565,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00034581682402350057,
|
|
"loss": 4.5592,
|
|
"mean_token_accuracy": 0.21604788154363633,
|
|
"num_tokens": 9916143.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 4.96110692024231,
|
|
"epoch": 4.489539748953975,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00034551038870610885,
|
|
"loss": 4.5427,
|
|
"mean_token_accuracy": 0.22704036682844161,
|
|
"num_tokens": 9925675.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 4.874826908111572,
|
|
"epoch": 4.493723849372385,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034520380828967346,
|
|
"loss": 4.4635,
|
|
"mean_token_accuracy": 0.23469811528921128,
|
|
"num_tokens": 9934398.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 4.8507688522338865,
|
|
"epoch": 4.497907949790795,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003448970834050877,
|
|
"loss": 4.5187,
|
|
"mean_token_accuracy": 0.23145101964473724,
|
|
"num_tokens": 9943066.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 4.805233335494995,
|
|
"epoch": 4.502092050209205,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003445902146835419,
|
|
"loss": 4.43,
|
|
"mean_token_accuracy": 0.23833929002285004,
|
|
"num_tokens": 9951565.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 4.904879379272461,
|
|
"epoch": 4.506276150627615,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003442832027565227,
|
|
"loss": 4.5616,
|
|
"mean_token_accuracy": 0.22564203590154647,
|
|
"num_tokens": 9960997.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 4.997456121444702,
|
|
"epoch": 4.510460251046025,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034397604825581115,
|
|
"loss": 4.5607,
|
|
"mean_token_accuracy": 0.22707622051239013,
|
|
"num_tokens": 9970188.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 4.9093372344970705,
|
|
"epoch": 4.514644351464435,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003436687518134819,
|
|
"loss": 4.5365,
|
|
"mean_token_accuracy": 0.2397874653339386,
|
|
"num_tokens": 9978749.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 4.971592426300049,
|
|
"epoch": 4.518828451882845,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003433613140619016,
|
|
"loss": 4.6158,
|
|
"mean_token_accuracy": 0.22887016981840133,
|
|
"num_tokens": 9988107.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 4.882873773574829,
|
|
"epoch": 4.523012552301255,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003430537356337277,
|
|
"loss": 4.4498,
|
|
"mean_token_accuracy": 0.2324157953262329,
|
|
"num_tokens": 9996127.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 4.9393562316894535,
|
|
"epoch": 4.527196652719665,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003427460171619073,
|
|
"loss": 4.562,
|
|
"mean_token_accuracy": 0.22322940528392793,
|
|
"num_tokens": 10005482.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 4.970738649368286,
|
|
"epoch": 4.531380753138075,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034243815927967536,
|
|
"loss": 4.5987,
|
|
"mean_token_accuracy": 0.217702217400074,
|
|
"num_tokens": 10014765.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 4.930580568313599,
|
|
"epoch": 4.535564853556485,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034213016262055396,
|
|
"loss": 4.5575,
|
|
"mean_token_accuracy": 0.22504401504993438,
|
|
"num_tokens": 10022897.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 4.946266174316406,
|
|
"epoch": 4.539748953974895,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003418220278183507,
|
|
"loss": 4.6138,
|
|
"mean_token_accuracy": 0.21606180667877198,
|
|
"num_tokens": 10032281.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 4.877597856521606,
|
|
"epoch": 4.543933054393305,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034151375550715753,
|
|
"loss": 4.4611,
|
|
"mean_token_accuracy": 0.2426906034350395,
|
|
"num_tokens": 10041194.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 4.801550006866455,
|
|
"epoch": 4.548117154811716,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034120534632134917,
|
|
"loss": 4.5292,
|
|
"mean_token_accuracy": 0.22138302773237228,
|
|
"num_tokens": 10050062.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 4.908306407928467,
|
|
"epoch": 4.552301255230126,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034089680089558226,
|
|
"loss": 4.6212,
|
|
"mean_token_accuracy": 0.23345834463834764,
|
|
"num_tokens": 10059747.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 4.970927381515503,
|
|
"epoch": 4.556485355648536,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034058811986479367,
|
|
"loss": 4.6205,
|
|
"mean_token_accuracy": 0.22192314267158508,
|
|
"num_tokens": 10069351.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 4.9670289039611815,
|
|
"epoch": 4.560669456066946,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034027930386419935,
|
|
"loss": 4.5487,
|
|
"mean_token_accuracy": 0.22394704222679138,
|
|
"num_tokens": 10078727.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 4.898691606521607,
|
|
"epoch": 4.564853556485356,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033997035352929285,
|
|
"loss": 4.5119,
|
|
"mean_token_accuracy": 0.23111017793416977,
|
|
"num_tokens": 10088151.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 4.839825057983399,
|
|
"epoch": 4.569037656903766,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003396612694958444,
|
|
"loss": 4.4646,
|
|
"mean_token_accuracy": 0.23166335821151735,
|
|
"num_tokens": 10097311.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 4.9434655666351315,
|
|
"epoch": 4.573221757322176,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033935205239989927,
|
|
"loss": 4.6124,
|
|
"mean_token_accuracy": 0.21592042297124864,
|
|
"num_tokens": 10107474.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 4.940805149078369,
|
|
"epoch": 4.577405857740586,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033904270287777645,
|
|
"loss": 4.5344,
|
|
"mean_token_accuracy": 0.2386762633919716,
|
|
"num_tokens": 10116305.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 4.9668638706207275,
|
|
"epoch": 4.581589958158996,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003387332215660675,
|
|
"loss": 4.5736,
|
|
"mean_token_accuracy": 0.22674793601036072,
|
|
"num_tokens": 10125947.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 4.886618947982788,
|
|
"epoch": 4.585774058577406,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00033842360910163536,
|
|
"loss": 4.5549,
|
|
"mean_token_accuracy": 0.2263866737484932,
|
|
"num_tokens": 10134510.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 4.997741079330444,
|
|
"epoch": 4.589958158995816,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003381138661216125,
|
|
"loss": 4.6097,
|
|
"mean_token_accuracy": 0.22223571091890335,
|
|
"num_tokens": 10143429.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 4.992606782913208,
|
|
"epoch": 4.594142259414226,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00033780399326340026,
|
|
"loss": 4.5982,
|
|
"mean_token_accuracy": 0.22430578470230103,
|
|
"num_tokens": 10153405.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 4.944302988052368,
|
|
"epoch": 4.598326359832636,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00033749399116466723,
|
|
"loss": 4.5421,
|
|
"mean_token_accuracy": 0.2359272837638855,
|
|
"num_tokens": 10162334.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 4.880887413024903,
|
|
"epoch": 4.602510460251046,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003371838604633479,
|
|
"loss": 4.5486,
|
|
"mean_token_accuracy": 0.2290939658880234,
|
|
"num_tokens": 10171604.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.602510460251046,
|
|
"eval_entropy": 4.6893728447350504,
|
|
"eval_loss": 5.326577186584473,
|
|
"eval_mean_token_accuracy": 0.1996338447148327,
|
|
"eval_num_tokens": 10171604.0,
|
|
"eval_runtime": 2.7689,
|
|
"eval_samples_per_second": 1411.007,
|
|
"eval_steps_per_second": 176.602,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 4.870112657546997,
|
|
"epoch": 4.606694560669456,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033687360179764125,
|
|
"loss": 4.5221,
|
|
"mean_token_accuracy": 0.2296696588397026,
|
|
"num_tokens": 10180659.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 4.937135839462281,
|
|
"epoch": 4.610878661087866,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003365632158060098,
|
|
"loss": 4.5848,
|
|
"mean_token_accuracy": 0.22333342880010604,
|
|
"num_tokens": 10190226.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 4.937072706222534,
|
|
"epoch": 4.615062761506276,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033625270312717805,
|
|
"loss": 4.4753,
|
|
"mean_token_accuracy": 0.22886465787887572,
|
|
"num_tokens": 10198806.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 4.889058685302734,
|
|
"epoch": 4.619246861924686,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003359420644001311,
|
|
"loss": 4.4803,
|
|
"mean_token_accuracy": 0.23300485461950302,
|
|
"num_tokens": 10207927.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 4.812423467636108,
|
|
"epoch": 4.623430962343097,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033563130026411357,
|
|
"loss": 4.4295,
|
|
"mean_token_accuracy": 0.23922039270401002,
|
|
"num_tokens": 10216241.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 4.923349905014038,
|
|
"epoch": 4.627615062761507,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033532041135862816,
|
|
"loss": 4.6145,
|
|
"mean_token_accuracy": 0.2218698814511299,
|
|
"num_tokens": 10225241.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 4.92033953666687,
|
|
"epoch": 4.631799163179917,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00033500939832343407,
|
|
"loss": 4.5677,
|
|
"mean_token_accuracy": 0.2273694470524788,
|
|
"num_tokens": 10233968.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.031051874160767,
|
|
"epoch": 4.635983263598327,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003346982617985463,
|
|
"loss": 4.6593,
|
|
"mean_token_accuracy": 0.22181983888149262,
|
|
"num_tokens": 10243589.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 4.8457269191741945,
|
|
"epoch": 4.640167364016737,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003343870024242337,
|
|
"loss": 4.4745,
|
|
"mean_token_accuracy": 0.23231513947248458,
|
|
"num_tokens": 10252416.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 4.924397373199463,
|
|
"epoch": 4.644351464435147,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033407562084101795,
|
|
"loss": 4.616,
|
|
"mean_token_accuracy": 0.2189919039607048,
|
|
"num_tokens": 10262213.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 4.976135301589966,
|
|
"epoch": 4.648535564853557,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033376411768967247,
|
|
"loss": 4.658,
|
|
"mean_token_accuracy": 0.2217055916786194,
|
|
"num_tokens": 10272774.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 4.946973705291748,
|
|
"epoch": 4.652719665271967,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033345249361122056,
|
|
"loss": 4.5096,
|
|
"mean_token_accuracy": 0.2347109854221344,
|
|
"num_tokens": 10281286.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 4.868842554092407,
|
|
"epoch": 4.656903765690377,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033314074924693446,
|
|
"loss": 4.5129,
|
|
"mean_token_accuracy": 0.22932395488023757,
|
|
"num_tokens": 10290101.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 4.849915504455566,
|
|
"epoch": 4.661087866108787,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033282888523833394,
|
|
"loss": 4.5356,
|
|
"mean_token_accuracy": 0.23551418632268906,
|
|
"num_tokens": 10299580.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 4.90254807472229,
|
|
"epoch": 4.665271966527197,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033251690222718504,
|
|
"loss": 4.5306,
|
|
"mean_token_accuracy": 0.23179284185171128,
|
|
"num_tokens": 10307860.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 4.918003892898559,
|
|
"epoch": 4.669456066945607,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003322048008554986,
|
|
"loss": 4.5461,
|
|
"mean_token_accuracy": 0.23381084501743316,
|
|
"num_tokens": 10316546.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.001623487472534,
|
|
"epoch": 4.673640167364017,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003318925817655291,
|
|
"loss": 4.6459,
|
|
"mean_token_accuracy": 0.2272306889295578,
|
|
"num_tokens": 10326012.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.023741817474365,
|
|
"epoch": 4.677824267782427,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003315802455997732,
|
|
"loss": 4.5881,
|
|
"mean_token_accuracy": 0.22142557799816132,
|
|
"num_tokens": 10334771.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 4.967146110534668,
|
|
"epoch": 4.682008368200837,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033126779300096855,
|
|
"loss": 4.5992,
|
|
"mean_token_accuracy": 0.2247091218829155,
|
|
"num_tokens": 10343416.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.008348798751831,
|
|
"epoch": 4.686192468619247,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003309552246120924,
|
|
"loss": 4.6729,
|
|
"mean_token_accuracy": 0.21793688386678695,
|
|
"num_tokens": 10353539.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 4.912502002716065,
|
|
"epoch": 4.690376569037657,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003306425410763601,
|
|
"loss": 4.5116,
|
|
"mean_token_accuracy": 0.22648889422416688,
|
|
"num_tokens": 10362020.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 4.867328786849976,
|
|
"epoch": 4.694560669456067,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003303297430372241,
|
|
"loss": 4.4454,
|
|
"mean_token_accuracy": 0.23655943274497987,
|
|
"num_tokens": 10371277.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 4.860052156448364,
|
|
"epoch": 4.698744769874477,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003300168311383727,
|
|
"loss": 4.5518,
|
|
"mean_token_accuracy": 0.22888745367527008,
|
|
"num_tokens": 10380680.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 4.950733041763305,
|
|
"epoch": 4.702928870292887,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032970380602372805,
|
|
"loss": 4.488,
|
|
"mean_token_accuracy": 0.2389546290040016,
|
|
"num_tokens": 10389576.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 4.929499959945678,
|
|
"epoch": 4.707112970711297,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003293906683374456,
|
|
"loss": 4.5771,
|
|
"mean_token_accuracy": 0.223915196955204,
|
|
"num_tokens": 10399182.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.015476226806641,
|
|
"epoch": 4.711297071129707,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003290774187239124,
|
|
"loss": 4.6909,
|
|
"mean_token_accuracy": 0.22491335570812226,
|
|
"num_tokens": 10409595.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.013520908355713,
|
|
"epoch": 4.7154811715481175,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032876405782774574,
|
|
"loss": 4.6676,
|
|
"mean_token_accuracy": 0.21178410202264786,
|
|
"num_tokens": 10419057.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 4.854910850524902,
|
|
"epoch": 4.7196652719665275,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003284505862937919,
|
|
"loss": 4.439,
|
|
"mean_token_accuracy": 0.23659967631101608,
|
|
"num_tokens": 10428090.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 4.943002223968506,
|
|
"epoch": 4.7238493723849375,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000328137004767125,
|
|
"loss": 4.612,
|
|
"mean_token_accuracy": 0.22136467695236206,
|
|
"num_tokens": 10438088.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 4.896690511703492,
|
|
"epoch": 4.7280334728033475,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003278233138930454,
|
|
"loss": 4.5942,
|
|
"mean_token_accuracy": 0.2258934885263443,
|
|
"num_tokens": 10447934.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 4.882231521606445,
|
|
"epoch": 4.7322175732217575,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003275095143170784,
|
|
"loss": 4.4506,
|
|
"mean_token_accuracy": 0.23515382558107376,
|
|
"num_tokens": 10456088.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 4.891965293884278,
|
|
"epoch": 4.7364016736401675,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032719560668497316,
|
|
"loss": 4.4664,
|
|
"mean_token_accuracy": 0.23839344829320908,
|
|
"num_tokens": 10465096.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 4.830285978317261,
|
|
"epoch": 4.7405857740585775,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00032688159164270114,
|
|
"loss": 4.548,
|
|
"mean_token_accuracy": 0.23039882779121398,
|
|
"num_tokens": 10473516.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 4.88932466506958,
|
|
"epoch": 4.7447698744769875,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032656746983645473,
|
|
"loss": 4.5347,
|
|
"mean_token_accuracy": 0.23420751839876175,
|
|
"num_tokens": 10482651.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 4.961641693115235,
|
|
"epoch": 4.7489539748953975,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003262532419126461,
|
|
"loss": 4.5584,
|
|
"mean_token_accuracy": 0.22702175229787827,
|
|
"num_tokens": 10491831.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 4.889359617233277,
|
|
"epoch": 4.7531380753138075,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032593890851790593,
|
|
"loss": 4.567,
|
|
"mean_token_accuracy": 0.23343465030193328,
|
|
"num_tokens": 10502070.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 4.881148624420166,
|
|
"epoch": 4.7573221757322175,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003256244702990817,
|
|
"loss": 4.5309,
|
|
"mean_token_accuracy": 0.22221818715333938,
|
|
"num_tokens": 10510638.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 4.852502965927124,
|
|
"epoch": 4.7615062761506275,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032530992790323664,
|
|
"loss": 4.482,
|
|
"mean_token_accuracy": 0.23099784106016158,
|
|
"num_tokens": 10519305.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 4.914034748077393,
|
|
"epoch": 4.7656903765690375,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00032499528197764855,
|
|
"loss": 4.5396,
|
|
"mean_token_accuracy": 0.22462785094976426,
|
|
"num_tokens": 10528844.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 4.955566549301148,
|
|
"epoch": 4.7698744769874475,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032468053316980815,
|
|
"loss": 4.6395,
|
|
"mean_token_accuracy": 0.22263478338718415,
|
|
"num_tokens": 10538260.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 4.939757537841797,
|
|
"epoch": 4.7740585774058575,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003243656821274178,
|
|
"loss": 4.5909,
|
|
"mean_token_accuracy": 0.22473777681589127,
|
|
"num_tokens": 10547733.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 4.913278818130493,
|
|
"epoch": 4.7782426778242675,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003240507294983905,
|
|
"loss": 4.5519,
|
|
"mean_token_accuracy": 0.228051495552063,
|
|
"num_tokens": 10558337.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 4.86893138885498,
|
|
"epoch": 4.7824267782426775,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00032373567593084804,
|
|
"loss": 4.5424,
|
|
"mean_token_accuracy": 0.22525518238544465,
|
|
"num_tokens": 10568374.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 4.9043069839477536,
|
|
"epoch": 4.786610878661088,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00032342052207311986,
|
|
"loss": 4.5977,
|
|
"mean_token_accuracy": 0.22403926253318787,
|
|
"num_tokens": 10578448.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 4.9788659572601315,
|
|
"epoch": 4.790794979079498,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032310526857374204,
|
|
"loss": 4.6401,
|
|
"mean_token_accuracy": 0.21871850788593292,
|
|
"num_tokens": 10586803.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.016373014450073,
|
|
"epoch": 4.794979079497908,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003227899160814556,
|
|
"loss": 4.6091,
|
|
"mean_token_accuracy": 0.22968777418136596,
|
|
"num_tokens": 10596596.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 4.922781705856323,
|
|
"epoch": 4.799163179916318,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003224744652452054,
|
|
"loss": 4.4971,
|
|
"mean_token_accuracy": 0.2282332420349121,
|
|
"num_tokens": 10605462.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.009257793426514,
|
|
"epoch": 4.803347280334728,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003221589167141382,
|
|
"loss": 4.6499,
|
|
"mean_token_accuracy": 0.2166450023651123,
|
|
"num_tokens": 10614908.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 4.9261805534362795,
|
|
"epoch": 4.807531380753138,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003218432711376025,
|
|
"loss": 4.6,
|
|
"mean_token_accuracy": 0.23084898442029952,
|
|
"num_tokens": 10624389.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 4.902324676513672,
|
|
"epoch": 4.811715481171548,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032152752916514605,
|
|
"loss": 4.5624,
|
|
"mean_token_accuracy": 0.22766193151473998,
|
|
"num_tokens": 10633228.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 4.917319965362549,
|
|
"epoch": 4.815899581589958,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032121169144651496,
|
|
"loss": 4.6134,
|
|
"mean_token_accuracy": 0.23094120770692825,
|
|
"num_tokens": 10641996.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 4.871028327941895,
|
|
"epoch": 4.820083682008368,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003208957586316525,
|
|
"loss": 4.5292,
|
|
"mean_token_accuracy": 0.22753013670444489,
|
|
"num_tokens": 10651724.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 4.968953084945679,
|
|
"epoch": 4.824267782426778,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003205797313706977,
|
|
"loss": 4.5872,
|
|
"mean_token_accuracy": 0.2233392298221588,
|
|
"num_tokens": 10661150.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 4.907320547103882,
|
|
"epoch": 4.828451882845188,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032026361031398386,
|
|
"loss": 4.5862,
|
|
"mean_token_accuracy": 0.2198406472802162,
|
|
"num_tokens": 10669890.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 4.977698945999146,
|
|
"epoch": 4.832635983263598,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031994739611203716,
|
|
"loss": 4.5229,
|
|
"mean_token_accuracy": 0.2249158129096031,
|
|
"num_tokens": 10678939.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 4.887291479110718,
|
|
"epoch": 4.836820083682008,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00031963108941557576,
|
|
"loss": 4.4864,
|
|
"mean_token_accuracy": 0.23248544782400132,
|
|
"num_tokens": 10687465.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 4.90247950553894,
|
|
"epoch": 4.841004184100418,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003193146908755078,
|
|
"loss": 4.5881,
|
|
"mean_token_accuracy": 0.22761875092983247,
|
|
"num_tokens": 10696930.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 4.8813889503479,
|
|
"epoch": 4.845188284518828,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00031899820114293073,
|
|
"loss": 4.5029,
|
|
"mean_token_accuracy": 0.23495363295078278,
|
|
"num_tokens": 10706820.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 4.875827503204346,
|
|
"epoch": 4.849372384937238,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031868162086912946,
|
|
"loss": 4.4875,
|
|
"mean_token_accuracy": 0.23774133771657943,
|
|
"num_tokens": 10715528.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 4.853024101257324,
|
|
"epoch": 4.853556485355648,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003183649507055754,
|
|
"loss": 4.498,
|
|
"mean_token_accuracy": 0.23070355504751205,
|
|
"num_tokens": 10724807.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 4.953215599060059,
|
|
"epoch": 4.857740585774058,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00031804819130392477,
|
|
"loss": 4.5833,
|
|
"mean_token_accuracy": 0.22348818629980088,
|
|
"num_tokens": 10733754.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 4.863496017456055,
|
|
"epoch": 4.861924686192468,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003177313433160175,
|
|
"loss": 4.586,
|
|
"mean_token_accuracy": 0.22773488909006118,
|
|
"num_tokens": 10742256.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 4.921669387817383,
|
|
"epoch": 4.866108786610878,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003174144073938757,
|
|
"loss": 4.5166,
|
|
"mean_token_accuracy": 0.24482062458992004,
|
|
"num_tokens": 10751025.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 4.943360757827759,
|
|
"epoch": 4.870292887029288,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00031709738418970275,
|
|
"loss": 4.4869,
|
|
"mean_token_accuracy": 0.23366626352071762,
|
|
"num_tokens": 10759970.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 4.8841503143310545,
|
|
"epoch": 4.874476987447698,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003167802743558813,
|
|
"loss": 4.5236,
|
|
"mean_token_accuracy": 0.23017355054616928,
|
|
"num_tokens": 10768531.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 4.946636819839478,
|
|
"epoch": 4.878661087866108,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003164630785449724,
|
|
"loss": 4.6042,
|
|
"mean_token_accuracy": 0.22595052570104598,
|
|
"num_tokens": 10778802.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 4.925518608093261,
|
|
"epoch": 4.882845188284519,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031614579740971415,
|
|
"loss": 4.5438,
|
|
"mean_token_accuracy": 0.22688824236392974,
|
|
"num_tokens": 10788121.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 4.993703031539917,
|
|
"epoch": 4.887029288702929,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00031582843160302,
|
|
"loss": 4.6496,
|
|
"mean_token_accuracy": 0.2283882111310959,
|
|
"num_tokens": 10797652.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 4.944615221023559,
|
|
"epoch": 4.891213389121339,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031551098177797784,
|
|
"loss": 4.5621,
|
|
"mean_token_accuracy": 0.22399260848760605,
|
|
"num_tokens": 10807564.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 4.900060415267944,
|
|
"epoch": 4.895397489539749,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003151934485878482,
|
|
"loss": 4.5384,
|
|
"mean_token_accuracy": 0.22342884838581084,
|
|
"num_tokens": 10816666.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 4.897633218765259,
|
|
"epoch": 4.899581589958159,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031487583268606354,
|
|
"loss": 4.5366,
|
|
"mean_token_accuracy": 0.23138904571533203,
|
|
"num_tokens": 10825639.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 4.913179349899292,
|
|
"epoch": 4.903765690376569,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003145581347262262,
|
|
"loss": 4.582,
|
|
"mean_token_accuracy": 0.22323246151208878,
|
|
"num_tokens": 10834681.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 4.902258348464966,
|
|
"epoch": 4.907949790794979,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031424035536210755,
|
|
"loss": 4.5844,
|
|
"mean_token_accuracy": 0.22328189313411712,
|
|
"num_tokens": 10844465.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 4.918869066238403,
|
|
"epoch": 4.912133891213389,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031392249524764645,
|
|
"loss": 4.6083,
|
|
"mean_token_accuracy": 0.22272249162197114,
|
|
"num_tokens": 10853327.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 4.9055060863494875,
|
|
"epoch": 4.916317991631799,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031360455503694787,
|
|
"loss": 4.5098,
|
|
"mean_token_accuracy": 0.2324555993080139,
|
|
"num_tokens": 10861865.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 4.955350637435913,
|
|
"epoch": 4.920502092050209,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003132865353842816,
|
|
"loss": 4.5663,
|
|
"mean_token_accuracy": 0.23500151038169861,
|
|
"num_tokens": 10870794.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 4.97408881187439,
|
|
"epoch": 4.924686192468619,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00031296843694408107,
|
|
"loss": 4.4918,
|
|
"mean_token_accuracy": 0.23457674235105513,
|
|
"num_tokens": 10879626.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 4.863820505142212,
|
|
"epoch": 4.928870292887029,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003126502603709417,
|
|
"loss": 4.4499,
|
|
"mean_token_accuracy": 0.23502477407455444,
|
|
"num_tokens": 10888359.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 4.8792774200439455,
|
|
"epoch": 4.933054393305439,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003123320063196196,
|
|
"loss": 4.5193,
|
|
"mean_token_accuracy": 0.23233772963285446,
|
|
"num_tokens": 10897174.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 4.888729858398437,
|
|
"epoch": 4.937238493723849,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003120136754450306,
|
|
"loss": 4.5916,
|
|
"mean_token_accuracy": 0.21830611526966096,
|
|
"num_tokens": 10906660.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 4.949666118621826,
|
|
"epoch": 4.941422594142259,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00031169526840224836,
|
|
"loss": 4.6062,
|
|
"mean_token_accuracy": 0.22134806215763092,
|
|
"num_tokens": 10915990.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.040625047683716,
|
|
"epoch": 4.945606694560669,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003113767858465034,
|
|
"loss": 4.7218,
|
|
"mean_token_accuracy": 0.2056598335504532,
|
|
"num_tokens": 10926355.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 4.862787675857544,
|
|
"epoch": 4.949790794979079,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031105822843318173,
|
|
"loss": 4.569,
|
|
"mean_token_accuracy": 0.23403559029102325,
|
|
"num_tokens": 10935854.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 4.880787086486817,
|
|
"epoch": 4.95397489539749,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003107395968178232,
|
|
"loss": 4.5027,
|
|
"mean_token_accuracy": 0.24179714173078537,
|
|
"num_tokens": 10944903.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 4.927484846115112,
|
|
"epoch": 4.9581589958159,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003104208916561203,
|
|
"loss": 4.5303,
|
|
"mean_token_accuracy": 0.22603820413351058,
|
|
"num_tokens": 10954009.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.022607135772705,
|
|
"epoch": 4.96234309623431,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00031010211360391727,
|
|
"loss": 4.6942,
|
|
"mean_token_accuracy": 0.21704623103141785,
|
|
"num_tokens": 10963174.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 4.998005151748657,
|
|
"epoch": 4.96652719665272,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000309783263317208,
|
|
"loss": 4.6075,
|
|
"mean_token_accuracy": 0.221870718896389,
|
|
"num_tokens": 10972822.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 4.9071746349334715,
|
|
"epoch": 4.97071129707113,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003094643414521349,
|
|
"loss": 4.4861,
|
|
"mean_token_accuracy": 0.2292321264743805,
|
|
"num_tokens": 10981758.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 4.9427204608917235,
|
|
"epoch": 4.97489539748954,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003091453486649883,
|
|
"loss": 4.5615,
|
|
"mean_token_accuracy": 0.22526928186416625,
|
|
"num_tokens": 10991730.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 4.95373272895813,
|
|
"epoch": 4.97907949790795,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030882628561220376,
|
|
"loss": 4.5744,
|
|
"mean_token_accuracy": 0.23037707060575485,
|
|
"num_tokens": 11000827.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 4.943409013748169,
|
|
"epoch": 4.98326359832636,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030850715295036187,
|
|
"loss": 4.5163,
|
|
"mean_token_accuracy": 0.2405583679676056,
|
|
"num_tokens": 11010544.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 4.906629371643066,
|
|
"epoch": 4.98744769874477,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003081879513361863,
|
|
"loss": 4.5831,
|
|
"mean_token_accuracy": 0.23016407042741777,
|
|
"num_tokens": 11019826.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 4.975005149841309,
|
|
"epoch": 4.99163179916318,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003078686814265426,
|
|
"loss": 4.5509,
|
|
"mean_token_accuracy": 0.23153901398181914,
|
|
"num_tokens": 11028408.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 4.971848583221435,
|
|
"epoch": 4.99581589958159,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000307549343878437,
|
|
"loss": 4.6171,
|
|
"mean_token_accuracy": 0.22651520669460296,
|
|
"num_tokens": 11038564.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 4.912516736984253,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00030722993934901487,
|
|
"loss": 4.4594,
|
|
"mean_token_accuracy": 0.23922617137432098,
|
|
"num_tokens": 11046610.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 4.765020084381104,
|
|
"epoch": 5.00418410041841,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00030691046849555927,
|
|
"loss": 4.1769,
|
|
"mean_token_accuracy": 0.26099960654973986,
|
|
"num_tokens": 11055929.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 4.893342924118042,
|
|
"epoch": 5.00836820083682,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00030659093197549,
|
|
"loss": 4.2594,
|
|
"mean_token_accuracy": 0.2544790685176849,
|
|
"num_tokens": 11064748.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 4.817290735244751,
|
|
"epoch": 5.01255230125523,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003062713304463616,
|
|
"loss": 4.2392,
|
|
"mean_token_accuracy": 0.2590743362903595,
|
|
"num_tokens": 11073707.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 4.862590408325195,
|
|
"epoch": 5.01673640167364,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003059516645658629,
|
|
"loss": 4.2673,
|
|
"mean_token_accuracy": 0.2506123945116997,
|
|
"num_tokens": 11082967.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 4.823035955429077,
|
|
"epoch": 5.02092050209205,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003056319349918146,
|
|
"loss": 4.2843,
|
|
"mean_token_accuracy": 0.2502159371972084,
|
|
"num_tokens": 11092105.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.02092050209205,
|
|
"eval_entropy": 4.71509728012397,
|
|
"eval_loss": 5.306480884552002,
|
|
"eval_mean_token_accuracy": 0.20097873666169452,
|
|
"eval_num_tokens": 11092105.0,
|
|
"eval_runtime": 2.7445,
|
|
"eval_samples_per_second": 1423.554,
|
|
"eval_steps_per_second": 178.172,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 4.936825227737427,
|
|
"epoch": 5.02510460251046,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030531214238216906,
|
|
"loss": 4.3555,
|
|
"mean_token_accuracy": 0.23532243371009826,
|
|
"num_tokens": 11100806.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 4.818486738204956,
|
|
"epoch": 5.02928870292887,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00030499228739500793,
|
|
"loss": 4.2114,
|
|
"mean_token_accuracy": 0.2589252471923828,
|
|
"num_tokens": 11109779.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 4.855298709869385,
|
|
"epoch": 5.03347280334728,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00030467237068854136,
|
|
"loss": 4.3005,
|
|
"mean_token_accuracy": 0.24639434963464737,
|
|
"num_tokens": 11118838.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 4.833564424514771,
|
|
"epoch": 5.03765690376569,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003043523929211064,
|
|
"loss": 4.2428,
|
|
"mean_token_accuracy": 0.24991558641195297,
|
|
"num_tokens": 11127029.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 4.864570617675781,
|
|
"epoch": 5.0418410041841,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000304032354751166,
|
|
"loss": 4.2685,
|
|
"mean_token_accuracy": 0.2491785004734993,
|
|
"num_tokens": 11136073.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 4.892072582244873,
|
|
"epoch": 5.046025104602511,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003037122568373071,
|
|
"loss": 4.3226,
|
|
"mean_token_accuracy": 0.24485297203063966,
|
|
"num_tokens": 11144807.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 4.894490528106689,
|
|
"epoch": 5.050209205020921,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003033920998382399,
|
|
"loss": 4.312,
|
|
"mean_token_accuracy": 0.24126479178667068,
|
|
"num_tokens": 11154308.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 4.831698989868164,
|
|
"epoch": 5.054393305439331,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003030718844127959,
|
|
"loss": 4.3506,
|
|
"mean_token_accuracy": 0.24017844796180726,
|
|
"num_tokens": 11163752.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 4.8442737579345705,
|
|
"epoch": 5.058577405857741,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003027516112199271,
|
|
"loss": 4.3098,
|
|
"mean_token_accuracy": 0.24209053814411163,
|
|
"num_tokens": 11173763.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 4.923576974868775,
|
|
"epoch": 5.062761506276151,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030243128091870413,
|
|
"loss": 4.3384,
|
|
"mean_token_accuracy": 0.24202877879142762,
|
|
"num_tokens": 11183577.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 4.897979784011841,
|
|
"epoch": 5.066945606694561,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003021108941683152,
|
|
"loss": 4.3065,
|
|
"mean_token_accuracy": 0.24479904621839524,
|
|
"num_tokens": 11192708.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 4.888266468048096,
|
|
"epoch": 5.071129707112971,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030179045162806494,
|
|
"loss": 4.3556,
|
|
"mean_token_accuracy": 0.24091459065675735,
|
|
"num_tokens": 11201803.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 4.802892255783081,
|
|
"epoch": 5.075313807531381,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00030146995395737247,
|
|
"loss": 4.2485,
|
|
"mean_token_accuracy": 0.2512045443058014,
|
|
"num_tokens": 11211208.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 4.863228797912598,
|
|
"epoch": 5.079497907949791,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003011494018157704,
|
|
"loss": 4.2752,
|
|
"mean_token_accuracy": 0.25700006932020186,
|
|
"num_tokens": 11221091.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 4.905969667434692,
|
|
"epoch": 5.083682008368201,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003008287958629036,
|
|
"loss": 4.3413,
|
|
"mean_token_accuracy": 0.24951196014881133,
|
|
"num_tokens": 11230084.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 4.906709241867065,
|
|
"epoch": 5.087866108786611,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00030050813675852747,
|
|
"loss": 4.3829,
|
|
"mean_token_accuracy": 0.23555608540773393,
|
|
"num_tokens": 11239310.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 4.854889154434204,
|
|
"epoch": 5.092050209205021,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00030018742516250697,
|
|
"loss": 4.2569,
|
|
"mean_token_accuracy": 0.25239221155643465,
|
|
"num_tokens": 11248738.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 4.902827501296997,
|
|
"epoch": 5.096234309623431,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000299866661734815,
|
|
"loss": 4.2781,
|
|
"mean_token_accuracy": 0.25164018124341964,
|
|
"num_tokens": 11257813.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 4.916948986053467,
|
|
"epoch": 5.100418410041841,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00029954584713553107,
|
|
"loss": 4.3366,
|
|
"mean_token_accuracy": 0.2394818529486656,
|
|
"num_tokens": 11266475.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 4.839675807952881,
|
|
"epoch": 5.104602510460251,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00029922498202484014,
|
|
"loss": 4.3124,
|
|
"mean_token_accuracy": 0.2514665678143501,
|
|
"num_tokens": 11276011.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 4.905929470062256,
|
|
"epoch": 5.108786610878661,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000298904067063031,
|
|
"loss": 4.3916,
|
|
"mean_token_accuracy": 0.24179730266332627,
|
|
"num_tokens": 11284619.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 4.881288051605225,
|
|
"epoch": 5.112970711297071,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002985831029104949,
|
|
"loss": 4.2942,
|
|
"mean_token_accuracy": 0.25667922347784045,
|
|
"num_tokens": 11293105.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 4.829472398757934,
|
|
"epoch": 5.117154811715481,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002982620902277246,
|
|
"loss": 4.2959,
|
|
"mean_token_accuracy": 0.24353276640176774,
|
|
"num_tokens": 11302717.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 4.92884087562561,
|
|
"epoch": 5.121338912133891,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00029794102967531263,
|
|
"loss": 4.4101,
|
|
"mean_token_accuracy": 0.245023275911808,
|
|
"num_tokens": 11312709.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 4.831700849533081,
|
|
"epoch": 5.125523012552302,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00029761992191394985,
|
|
"loss": 4.3098,
|
|
"mean_token_accuracy": 0.2465708762407303,
|
|
"num_tokens": 11322040.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 4.880761432647705,
|
|
"epoch": 5.129707112970712,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00029729876760442456,
|
|
"loss": 4.3786,
|
|
"mean_token_accuracy": 0.24700639843940736,
|
|
"num_tokens": 11332029.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 4.881295824050904,
|
|
"epoch": 5.133891213389122,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002969775674076206,
|
|
"loss": 4.3617,
|
|
"mean_token_accuracy": 0.23835889250040054,
|
|
"num_tokens": 11340960.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 4.925517892837524,
|
|
"epoch": 5.138075313807532,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002966563219845164,
|
|
"loss": 4.3743,
|
|
"mean_token_accuracy": 0.23851172775030136,
|
|
"num_tokens": 11349497.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 4.915242719650268,
|
|
"epoch": 5.142259414225942,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002963350319961832,
|
|
"loss": 4.3063,
|
|
"mean_token_accuracy": 0.24959479421377181,
|
|
"num_tokens": 11358477.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 4.87174243927002,
|
|
"epoch": 5.146443514644352,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002960136981037844,
|
|
"loss": 4.3169,
|
|
"mean_token_accuracy": 0.24222665280103683,
|
|
"num_tokens": 11367635.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 4.8638941764831545,
|
|
"epoch": 5.150627615062762,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002956923209685733,
|
|
"loss": 4.2779,
|
|
"mean_token_accuracy": 0.24575286656618117,
|
|
"num_tokens": 11376737.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 4.83456768989563,
|
|
"epoch": 5.154811715481172,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002953709012518925,
|
|
"loss": 4.3448,
|
|
"mean_token_accuracy": 0.24238020926713943,
|
|
"num_tokens": 11386756.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 4.798333358764649,
|
|
"epoch": 5.158995815899582,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00029504943961517204,
|
|
"loss": 4.2347,
|
|
"mean_token_accuracy": 0.25625640004873274,
|
|
"num_tokens": 11394889.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 4.898099803924561,
|
|
"epoch": 5.163179916317992,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00029472793671992824,
|
|
"loss": 4.4102,
|
|
"mean_token_accuracy": 0.2461270049214363,
|
|
"num_tokens": 11404759.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 4.978769254684448,
|
|
"epoch": 5.167364016736402,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00029440639322776246,
|
|
"loss": 4.3649,
|
|
"mean_token_accuracy": 0.23961812257766724,
|
|
"num_tokens": 11413739.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 4.84345178604126,
|
|
"epoch": 5.171548117154812,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00029408480980035943,
|
|
"loss": 4.3591,
|
|
"mean_token_accuracy": 0.24829546809196473,
|
|
"num_tokens": 11422835.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 4.900705432891845,
|
|
"epoch": 5.175732217573222,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002937631870994863,
|
|
"loss": 4.349,
|
|
"mean_token_accuracy": 0.2499272957444191,
|
|
"num_tokens": 11432613.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 4.843280172348022,
|
|
"epoch": 5.179916317991632,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00029344152578699074,
|
|
"loss": 4.3107,
|
|
"mean_token_accuracy": 0.2447648286819458,
|
|
"num_tokens": 11441297.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 4.822741889953614,
|
|
"epoch": 5.184100418410042,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00029311982652479993,
|
|
"loss": 4.2598,
|
|
"mean_token_accuracy": 0.24408998787403108,
|
|
"num_tokens": 11450409.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 4.898094129562378,
|
|
"epoch": 5.188284518828452,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002927980899749195,
|
|
"loss": 4.4028,
|
|
"mean_token_accuracy": 0.23193714916706085,
|
|
"num_tokens": 11460132.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 4.848593425750733,
|
|
"epoch": 5.192468619246862,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002924763167994312,
|
|
"loss": 4.3643,
|
|
"mean_token_accuracy": 0.24269140809774398,
|
|
"num_tokens": 11468984.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 4.939100360870361,
|
|
"epoch": 5.196652719665272,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002921545076604928,
|
|
"loss": 4.3896,
|
|
"mean_token_accuracy": 0.24655464589595794,
|
|
"num_tokens": 11478371.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 4.834752941131592,
|
|
"epoch": 5.200836820083682,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00029183266322033565,
|
|
"loss": 4.366,
|
|
"mean_token_accuracy": 0.24623971432447433,
|
|
"num_tokens": 11487924.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 4.890018177032471,
|
|
"epoch": 5.205020920502092,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002915107841412638,
|
|
"loss": 4.3551,
|
|
"mean_token_accuracy": 0.2409139782190323,
|
|
"num_tokens": 11497892.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 4.86919846534729,
|
|
"epoch": 5.209205020920502,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002911888710856528,
|
|
"loss": 4.2484,
|
|
"mean_token_accuracy": 0.25115692615509033,
|
|
"num_tokens": 11507266.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 4.809077024459839,
|
|
"epoch": 5.2133891213389125,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002908669247159477,
|
|
"loss": 4.2275,
|
|
"mean_token_accuracy": 0.2560761526226997,
|
|
"num_tokens": 11516672.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 4.836382818222046,
|
|
"epoch": 5.2175732217573225,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00029054494569466266,
|
|
"loss": 4.3504,
|
|
"mean_token_accuracy": 0.2416721150279045,
|
|
"num_tokens": 11525865.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 4.934778165817261,
|
|
"epoch": 5.2217573221757325,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00029022293468437853,
|
|
"loss": 4.3868,
|
|
"mean_token_accuracy": 0.24373907148838042,
|
|
"num_tokens": 11534910.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 4.80858097076416,
|
|
"epoch": 5.2259414225941425,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00028990089234774236,
|
|
"loss": 4.2597,
|
|
"mean_token_accuracy": 0.2585694804787636,
|
|
"num_tokens": 11543507.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 4.827981662750244,
|
|
"epoch": 5.2301255230125525,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00028957881934746544,
|
|
"loss": 4.269,
|
|
"mean_token_accuracy": 0.25015259385108946,
|
|
"num_tokens": 11552562.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 4.894914865493774,
|
|
"epoch": 5.2343096234309625,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028925671634632216,
|
|
"loss": 4.3278,
|
|
"mean_token_accuracy": 0.24393458068370819,
|
|
"num_tokens": 11561931.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 4.881958770751953,
|
|
"epoch": 5.2384937238493725,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00028893458400714876,
|
|
"loss": 4.3395,
|
|
"mean_token_accuracy": 0.2495964363217354,
|
|
"num_tokens": 11570639.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 4.855148220062256,
|
|
"epoch": 5.2426778242677825,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002886124229928418,
|
|
"loss": 4.2828,
|
|
"mean_token_accuracy": 0.2543689623475075,
|
|
"num_tokens": 11579796.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 4.883868455886841,
|
|
"epoch": 5.2468619246861925,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00028829023396635683,
|
|
"loss": 4.3841,
|
|
"mean_token_accuracy": 0.2466102659702301,
|
|
"num_tokens": 11589343.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 4.887897872924805,
|
|
"epoch": 5.2510460251046025,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002879680175907071,
|
|
"loss": 4.3807,
|
|
"mean_token_accuracy": 0.2473122149705887,
|
|
"num_tokens": 11598633.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 4.866991233825684,
|
|
"epoch": 5.2552301255230125,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002876457745289621,
|
|
"loss": 4.2515,
|
|
"mean_token_accuracy": 0.2539974868297577,
|
|
"num_tokens": 11607746.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 4.872642469406128,
|
|
"epoch": 5.2594142259414225,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002873235054442463,
|
|
"loss": 4.3536,
|
|
"mean_token_accuracy": 0.2418938621878624,
|
|
"num_tokens": 11616574.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 4.850561428070068,
|
|
"epoch": 5.2635983263598325,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00028700121099973765,
|
|
"loss": 4.3595,
|
|
"mean_token_accuracy": 0.24661397486925124,
|
|
"num_tokens": 11625918.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 4.9034099102020265,
|
|
"epoch": 5.2677824267782425,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002866788918586661,
|
|
"loss": 4.3306,
|
|
"mean_token_accuracy": 0.24706787019968032,
|
|
"num_tokens": 11634255.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 4.868001174926758,
|
|
"epoch": 5.2719665271966525,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00028635654868431286,
|
|
"loss": 4.3377,
|
|
"mean_token_accuracy": 0.24009892642498015,
|
|
"num_tokens": 11643532.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 4.833799695968628,
|
|
"epoch": 5.2761506276150625,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00028603418214000827,
|
|
"loss": 4.3127,
|
|
"mean_token_accuracy": 0.24230614304542542,
|
|
"num_tokens": 11653104.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 4.9235687255859375,
|
|
"epoch": 5.2803347280334725,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002857117928891308,
|
|
"loss": 4.3838,
|
|
"mean_token_accuracy": 0.23816751390695573,
|
|
"num_tokens": 11662752.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 4.7752399921417235,
|
|
"epoch": 5.2845188284518825,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00028538938159510587,
|
|
"loss": 4.2635,
|
|
"mean_token_accuracy": 0.25659535378217696,
|
|
"num_tokens": 11671566.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 4.8576243877410885,
|
|
"epoch": 5.2887029288702925,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00028506694892140385,
|
|
"loss": 4.3388,
|
|
"mean_token_accuracy": 0.24985320419073104,
|
|
"num_tokens": 11681770.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 4.8314056396484375,
|
|
"epoch": 5.292887029288703,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00028474449553153944,
|
|
"loss": 4.3428,
|
|
"mean_token_accuracy": 0.24412952363491058,
|
|
"num_tokens": 11691177.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 4.802004289627075,
|
|
"epoch": 5.297071129707113,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00028442202208906993,
|
|
"loss": 4.2858,
|
|
"mean_token_accuracy": 0.2520890235900879,
|
|
"num_tokens": 11700461.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 4.844103097915649,
|
|
"epoch": 5.301255230125523,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002840995292575938,
|
|
"loss": 4.3381,
|
|
"mean_token_accuracy": 0.23787227421998977,
|
|
"num_tokens": 11708783.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 4.860529518127441,
|
|
"epoch": 5.305439330543933,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002837770177007494,
|
|
"loss": 4.3611,
|
|
"mean_token_accuracy": 0.24257854372262955,
|
|
"num_tokens": 11717419.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 4.897275972366333,
|
|
"epoch": 5.309623430962343,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028345448808221374,
|
|
"loss": 4.4053,
|
|
"mean_token_accuracy": 0.23805580288171768,
|
|
"num_tokens": 11727014.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 4.8894736766815186,
|
|
"epoch": 5.313807531380753,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00028313194106570095,
|
|
"loss": 4.3277,
|
|
"mean_token_accuracy": 0.24317242652177812,
|
|
"num_tokens": 11736718.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 4.931209945678711,
|
|
"epoch": 5.317991631799163,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002828093773149607,
|
|
"loss": 4.4314,
|
|
"mean_token_accuracy": 0.23108220249414443,
|
|
"num_tokens": 11747236.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 4.869767951965332,
|
|
"epoch": 5.322175732217573,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00028248679749377754,
|
|
"loss": 4.3317,
|
|
"mean_token_accuracy": 0.244887013733387,
|
|
"num_tokens": 11756912.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 4.816705703735352,
|
|
"epoch": 5.326359832635983,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028216420226596884,
|
|
"loss": 4.2942,
|
|
"mean_token_accuracy": 0.2506855741143227,
|
|
"num_tokens": 11766035.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 4.781054401397705,
|
|
"epoch": 5.330543933054393,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002818415922953837,
|
|
"loss": 4.2461,
|
|
"mean_token_accuracy": 0.25099744498729704,
|
|
"num_tokens": 11775934.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 4.844691276550293,
|
|
"epoch": 5.334728033472803,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00028151896824590156,
|
|
"loss": 4.3216,
|
|
"mean_token_accuracy": 0.24460023790597915,
|
|
"num_tokens": 11784549.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 4.893819427490234,
|
|
"epoch": 5.338912133891213,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002811963307814309,
|
|
"loss": 4.3772,
|
|
"mean_token_accuracy": 0.23641878962516785,
|
|
"num_tokens": 11794369.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 4.837375831604004,
|
|
"epoch": 5.343096234309623,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00028087368056590767,
|
|
"loss": 4.3631,
|
|
"mean_token_accuracy": 0.24134186506271363,
|
|
"num_tokens": 11803221.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 4.85581259727478,
|
|
"epoch": 5.347280334728033,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002805510182632942,
|
|
"loss": 4.3577,
|
|
"mean_token_accuracy": 0.2448490634560585,
|
|
"num_tokens": 11812811.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 4.921184110641479,
|
|
"epoch": 5.351464435146443,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002802283445375775,
|
|
"loss": 4.3688,
|
|
"mean_token_accuracy": 0.2464720830321312,
|
|
"num_tokens": 11821522.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 4.876348447799683,
|
|
"epoch": 5.355648535564853,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00027990566005276846,
|
|
"loss": 4.3078,
|
|
"mean_token_accuracy": 0.25020708441734313,
|
|
"num_tokens": 11830021.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 4.816551923751831,
|
|
"epoch": 5.359832635983263,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027958296547289966,
|
|
"loss": 4.3076,
|
|
"mean_token_accuracy": 0.2480343133211136,
|
|
"num_tokens": 11838612.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 4.813551998138427,
|
|
"epoch": 5.364016736401673,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027926026146202475,
|
|
"loss": 4.2391,
|
|
"mean_token_accuracy": 0.25443035960197447,
|
|
"num_tokens": 11847648.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 4.82069673538208,
|
|
"epoch": 5.368200836820083,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027893754868421665,
|
|
"loss": 4.2863,
|
|
"mean_token_accuracy": 0.250758196413517,
|
|
"num_tokens": 11857008.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 4.846904325485229,
|
|
"epoch": 5.372384937238493,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002786148278035665,
|
|
"loss": 4.3225,
|
|
"mean_token_accuracy": 0.25432037115097045,
|
|
"num_tokens": 11866055.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 4.924081563949585,
|
|
"epoch": 5.376569037656903,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002782920994841819,
|
|
"loss": 4.3749,
|
|
"mean_token_accuracy": 0.2374558925628662,
|
|
"num_tokens": 11875496.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 4.859309148788452,
|
|
"epoch": 5.380753138075314,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002779693643901858,
|
|
"loss": 4.3126,
|
|
"mean_token_accuracy": 0.24579897671937942,
|
|
"num_tokens": 11884115.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 4.862702322006226,
|
|
"epoch": 5.384937238493724,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00027764662318571526,
|
|
"loss": 4.4682,
|
|
"mean_token_accuracy": 0.23752174526453018,
|
|
"num_tokens": 11893930.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 4.882434606552124,
|
|
"epoch": 5.389121338912134,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002773238765349196,
|
|
"loss": 4.3436,
|
|
"mean_token_accuracy": 0.24105090498924256,
|
|
"num_tokens": 11902702.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 4.92301607131958,
|
|
"epoch": 5.393305439330544,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002770011251019598,
|
|
"loss": 4.3591,
|
|
"mean_token_accuracy": 0.24739033877849578,
|
|
"num_tokens": 11911915.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 4.805233335494995,
|
|
"epoch": 5.397489539748954,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00027667836955100615,
|
|
"loss": 4.3112,
|
|
"mean_token_accuracy": 0.2499281421303749,
|
|
"num_tokens": 11920737.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 4.8364111423492435,
|
|
"epoch": 5.401673640167364,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002763556105462378,
|
|
"loss": 4.3462,
|
|
"mean_token_accuracy": 0.24862528294324876,
|
|
"num_tokens": 11929548.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 4.891482591629028,
|
|
"epoch": 5.405857740585774,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027603284875184095,
|
|
"loss": 4.3097,
|
|
"mean_token_accuracy": 0.24115657806396484,
|
|
"num_tokens": 11938459.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 4.925978326797486,
|
|
"epoch": 5.410041841004184,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027571008483200744,
|
|
"loss": 4.3215,
|
|
"mean_token_accuracy": 0.25120749324560165,
|
|
"num_tokens": 11947282.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 4.8306680679321286,
|
|
"epoch": 5.414225941422594,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00027538731945093346,
|
|
"loss": 4.2978,
|
|
"mean_token_accuracy": 0.24287822246551513,
|
|
"num_tokens": 11956209.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 4.825640249252319,
|
|
"epoch": 5.418410041841004,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027506455327281834,
|
|
"loss": 4.3377,
|
|
"mean_token_accuracy": 0.2472826212644577,
|
|
"num_tokens": 11965689.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 4.844741010665894,
|
|
"epoch": 5.422594142259414,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000274741786961863,
|
|
"loss": 4.3619,
|
|
"mean_token_accuracy": 0.24021243900060654,
|
|
"num_tokens": 11975622.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 4.872931432723999,
|
|
"epoch": 5.426778242677824,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002744190211822685,
|
|
"loss": 4.4444,
|
|
"mean_token_accuracy": 0.236192724108696,
|
|
"num_tokens": 11984904.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 4.907651805877686,
|
|
"epoch": 5.430962343096234,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000274096256598235,
|
|
"loss": 4.4377,
|
|
"mean_token_accuracy": 0.23839851468801498,
|
|
"num_tokens": 11995456.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 4.865474081039428,
|
|
"epoch": 5.435146443514644,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00027377349387396017,
|
|
"loss": 4.3172,
|
|
"mean_token_accuracy": 0.2361571043729782,
|
|
"num_tokens": 12004566.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 4.818851900100708,
|
|
"epoch": 5.439330543933054,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00027345073367363775,
|
|
"loss": 4.2441,
|
|
"mean_token_accuracy": 0.24916997402906418,
|
|
"num_tokens": 12013304.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.439330543933054,
|
|
"eval_entropy": 4.732826380641914,
|
|
"eval_loss": 5.299282073974609,
|
|
"eval_mean_token_accuracy": 0.2028583053400424,
|
|
"eval_num_tokens": 12013304.0,
|
|
"eval_runtime": 2.7636,
|
|
"eval_samples_per_second": 1413.726,
|
|
"eval_steps_per_second": 176.942,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 4.810036993026733,
|
|
"epoch": 5.443514644351464,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002731279766614563,
|
|
"loss": 4.3021,
|
|
"mean_token_accuracy": 0.2508931249380112,
|
|
"num_tokens": 12022007.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 4.7806531429290775,
|
|
"epoch": 5.447698744769874,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000272805223501598,
|
|
"loss": 4.3677,
|
|
"mean_token_accuracy": 0.24052975028753282,
|
|
"num_tokens": 12031493.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 4.816061115264892,
|
|
"epoch": 5.451882845188284,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002724824748582369,
|
|
"loss": 4.3271,
|
|
"mean_token_accuracy": 0.25186098664999007,
|
|
"num_tokens": 12040143.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 4.959175395965576,
|
|
"epoch": 5.456066945606695,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002721597313955376,
|
|
"loss": 4.4981,
|
|
"mean_token_accuracy": 0.23517460972070695,
|
|
"num_tokens": 12050136.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 4.8619592666625975,
|
|
"epoch": 5.460251046025105,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002718369937776546,
|
|
"loss": 4.3808,
|
|
"mean_token_accuracy": 0.24261431843042375,
|
|
"num_tokens": 12060065.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 4.93742265701294,
|
|
"epoch": 5.464435146443515,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002715142626687298,
|
|
"loss": 4.3935,
|
|
"mean_token_accuracy": 0.22723212242126464,
|
|
"num_tokens": 12069175.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 4.852547645568848,
|
|
"epoch": 5.468619246861925,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000271191538732892,
|
|
"loss": 4.2954,
|
|
"mean_token_accuracy": 0.2517950892448425,
|
|
"num_tokens": 12077308.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 4.795202922821045,
|
|
"epoch": 5.472803347280335,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00027086882263425527,
|
|
"loss": 4.2739,
|
|
"mean_token_accuracy": 0.2498415604233742,
|
|
"num_tokens": 12085678.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 4.858919095993042,
|
|
"epoch": 5.476987447698745,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027054611503691736,
|
|
"loss": 4.382,
|
|
"mean_token_accuracy": 0.24336835741996765,
|
|
"num_tokens": 12095163.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 4.880091190338135,
|
|
"epoch": 5.481171548117155,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00027022341660495853,
|
|
"loss": 4.3505,
|
|
"mean_token_accuracy": 0.24403632581233978,
|
|
"num_tokens": 12104185.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 4.883603096008301,
|
|
"epoch": 5.485355648535565,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026990072800244045,
|
|
"loss": 4.3981,
|
|
"mean_token_accuracy": 0.23231228142976762,
|
|
"num_tokens": 12114298.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 4.871931076049805,
|
|
"epoch": 5.489539748953975,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00026957804989340427,
|
|
"loss": 4.3508,
|
|
"mean_token_accuracy": 0.2487948939204216,
|
|
"num_tokens": 12122512.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 4.805060577392578,
|
|
"epoch": 5.493723849372385,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002692553829418697,
|
|
"loss": 4.3006,
|
|
"mean_token_accuracy": 0.24979174584150315,
|
|
"num_tokens": 12132133.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 4.862864685058594,
|
|
"epoch": 5.497907949790795,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002689327278118334,
|
|
"loss": 4.3414,
|
|
"mean_token_accuracy": 0.24378337413072587,
|
|
"num_tokens": 12141299.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 4.985860919952392,
|
|
"epoch": 5.502092050209205,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000268610085167268,
|
|
"loss": 4.4382,
|
|
"mean_token_accuracy": 0.23862579464912415,
|
|
"num_tokens": 12150412.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 4.868389129638672,
|
|
"epoch": 5.506276150627615,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002682874556721198,
|
|
"loss": 4.3802,
|
|
"mean_token_accuracy": 0.2411389634013176,
|
|
"num_tokens": 12159754.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 4.8739941120147705,
|
|
"epoch": 5.510460251046025,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002679648399903087,
|
|
"loss": 4.3357,
|
|
"mean_token_accuracy": 0.2515847861766815,
|
|
"num_tokens": 12169098.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 4.842821025848389,
|
|
"epoch": 5.514644351464435,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002676422387857258,
|
|
"loss": 4.3089,
|
|
"mean_token_accuracy": 0.2520672813057899,
|
|
"num_tokens": 12178304.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 4.766682147979736,
|
|
"epoch": 5.518828451882845,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026731965272223246,
|
|
"loss": 4.2998,
|
|
"mean_token_accuracy": 0.2544041082262993,
|
|
"num_tokens": 12188029.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 4.845515441894531,
|
|
"epoch": 5.523012552301255,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002669970824636591,
|
|
"loss": 4.2954,
|
|
"mean_token_accuracy": 0.2415763705968857,
|
|
"num_tokens": 12196663.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 4.9092034816741945,
|
|
"epoch": 5.527196652719665,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002666745286738033,
|
|
"loss": 4.4058,
|
|
"mean_token_accuracy": 0.23879525661468506,
|
|
"num_tokens": 12206496.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 4.911815166473389,
|
|
"epoch": 5.531380753138075,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026635199201642914,
|
|
"loss": 4.4001,
|
|
"mean_token_accuracy": 0.24310670495033265,
|
|
"num_tokens": 12216597.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 4.801329755783081,
|
|
"epoch": 5.535564853556485,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002660294731552649,
|
|
"loss": 4.3279,
|
|
"mean_token_accuracy": 0.2480277344584465,
|
|
"num_tokens": 12225125.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 4.8107579231262205,
|
|
"epoch": 5.539748953974895,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026570697275400267,
|
|
"loss": 4.3155,
|
|
"mean_token_accuracy": 0.24155509173870088,
|
|
"num_tokens": 12234402.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 4.798937225341797,
|
|
"epoch": 5.543933054393305,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002653844914762964,
|
|
"loss": 4.2888,
|
|
"mean_token_accuracy": 0.24243369251489638,
|
|
"num_tokens": 12244136.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 4.893156290054321,
|
|
"epoch": 5.548117154811716,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002650620299857607,
|
|
"loss": 4.3368,
|
|
"mean_token_accuracy": 0.24701797664165498,
|
|
"num_tokens": 12253039.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 4.889049768447876,
|
|
"epoch": 5.552301255230126,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00026473958894596963,
|
|
"loss": 4.3609,
|
|
"mean_token_accuracy": 0.24905621856451035,
|
|
"num_tokens": 12263172.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 4.889519357681275,
|
|
"epoch": 5.556485355648536,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000264417169020455,
|
|
"loss": 4.3863,
|
|
"mean_token_accuracy": 0.24162942469120025,
|
|
"num_tokens": 12273324.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 4.908349561691284,
|
|
"epoch": 5.560669456066946,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002640947708727049,
|
|
"loss": 4.459,
|
|
"mean_token_accuracy": 0.23467103838920594,
|
|
"num_tokens": 12283719.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 4.823739051818848,
|
|
"epoch": 5.564853556485356,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002637723951661631,
|
|
"loss": 4.3773,
|
|
"mean_token_accuracy": 0.23658206611871718,
|
|
"num_tokens": 12293451.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 4.928967809677124,
|
|
"epoch": 5.569037656903766,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002634500425642268,
|
|
"loss": 4.4373,
|
|
"mean_token_accuracy": 0.23489848226308824,
|
|
"num_tokens": 12302560.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 4.924539184570312,
|
|
"epoch": 5.573221757322176,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002631277137302461,
|
|
"loss": 4.3953,
|
|
"mean_token_accuracy": 0.22994994670152663,
|
|
"num_tokens": 12312115.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 4.8200531005859375,
|
|
"epoch": 5.577405857740586,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026280540932752176,
|
|
"loss": 4.2793,
|
|
"mean_token_accuracy": 0.257461816072464,
|
|
"num_tokens": 12320689.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 4.784903717041016,
|
|
"epoch": 5.581589958158996,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026248313001930445,
|
|
"loss": 4.2623,
|
|
"mean_token_accuracy": 0.2559750810265541,
|
|
"num_tokens": 12329643.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 4.859687471389771,
|
|
"epoch": 5.585774058577406,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00026216087646879326,
|
|
"loss": 4.3841,
|
|
"mean_token_accuracy": 0.24211165755987168,
|
|
"num_tokens": 12339267.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 4.922477912902832,
|
|
"epoch": 5.589958158995816,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002618386493391339,
|
|
"loss": 4.4557,
|
|
"mean_token_accuracy": 0.2338041067123413,
|
|
"num_tokens": 12348515.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 4.854750251770019,
|
|
"epoch": 5.594142259414226,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026151644929341833,
|
|
"loss": 4.4181,
|
|
"mean_token_accuracy": 0.2405848026275635,
|
|
"num_tokens": 12358023.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 4.817475414276123,
|
|
"epoch": 5.598326359832636,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026119427699468226,
|
|
"loss": 4.3227,
|
|
"mean_token_accuracy": 0.24857073724269868,
|
|
"num_tokens": 12367129.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 4.834227228164673,
|
|
"epoch": 5.602510460251046,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00026087213310590455,
|
|
"loss": 4.3448,
|
|
"mean_token_accuracy": 0.2517047390341759,
|
|
"num_tokens": 12375933.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 4.800920677185059,
|
|
"epoch": 5.606694560669456,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026055001829000543,
|
|
"loss": 4.3251,
|
|
"mean_token_accuracy": 0.2512205421924591,
|
|
"num_tokens": 12384784.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 4.852274703979492,
|
|
"epoch": 5.610878661087866,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026022793320984553,
|
|
"loss": 4.3511,
|
|
"mean_token_accuracy": 0.24749025851488113,
|
|
"num_tokens": 12393933.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 4.893293285369873,
|
|
"epoch": 5.615062761506276,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002599058785282241,
|
|
"loss": 4.378,
|
|
"mean_token_accuracy": 0.23699659258127212,
|
|
"num_tokens": 12403442.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 4.858484745025635,
|
|
"epoch": 5.619246861924686,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025958385490787797,
|
|
"loss": 4.2641,
|
|
"mean_token_accuracy": 0.25791127383708956,
|
|
"num_tokens": 12412317.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 4.825613069534302,
|
|
"epoch": 5.623430962343097,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002592618630114798,
|
|
"loss": 4.317,
|
|
"mean_token_accuracy": 0.24275425523519517,
|
|
"num_tokens": 12421459.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 4.837831211090088,
|
|
"epoch": 5.627615062761507,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025893990350163724,
|
|
"loss": 4.3794,
|
|
"mean_token_accuracy": 0.2408878192305565,
|
|
"num_tokens": 12431224.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 4.800341033935547,
|
|
"epoch": 5.631799163179917,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025861797704089133,
|
|
"loss": 4.2748,
|
|
"mean_token_accuracy": 0.24880686402320862,
|
|
"num_tokens": 12439891.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 4.808929491043091,
|
|
"epoch": 5.635983263598327,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002582960842917148,
|
|
"loss": 4.2647,
|
|
"mean_token_accuracy": 0.24698888808488845,
|
|
"num_tokens": 12449852.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 4.904977416992187,
|
|
"epoch": 5.640167364016737,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002579742259165113,
|
|
"loss": 4.3594,
|
|
"mean_token_accuracy": 0.24996743500232696,
|
|
"num_tokens": 12459058.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 4.880940389633179,
|
|
"epoch": 5.644351464435147,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002576524025776135,
|
|
"loss": 4.391,
|
|
"mean_token_accuracy": 0.23986391872167587,
|
|
"num_tokens": 12468684.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 4.824934482574463,
|
|
"epoch": 5.648535564853557,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002573306149372821,
|
|
"loss": 4.349,
|
|
"mean_token_accuracy": 0.24743006378412247,
|
|
"num_tokens": 12478102.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 4.85822548866272,
|
|
"epoch": 5.652719665271967,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002570088636577044,
|
|
"loss": 4.3743,
|
|
"mean_token_accuracy": 0.23871159553527832,
|
|
"num_tokens": 12488036.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 4.90421347618103,
|
|
"epoch": 5.656903765690377,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002566871494009928,
|
|
"loss": 4.3502,
|
|
"mean_token_accuracy": 0.24700721204280854,
|
|
"num_tokens": 12496593.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 4.920361089706421,
|
|
"epoch": 5.661087866108787,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002563654728291835,
|
|
"loss": 4.4335,
|
|
"mean_token_accuracy": 0.23586941361427308,
|
|
"num_tokens": 12505752.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 4.840803098678589,
|
|
"epoch": 5.665271966527197,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002560438346042351,
|
|
"loss": 4.3183,
|
|
"mean_token_accuracy": 0.25199994891881944,
|
|
"num_tokens": 12515192.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 4.912207555770874,
|
|
"epoch": 5.669456066945607,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002557222353880275,
|
|
"loss": 4.4186,
|
|
"mean_token_accuracy": 0.24011440724134445,
|
|
"num_tokens": 12525613.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 4.885090732574463,
|
|
"epoch": 5.673640167364017,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00025540067584235997,
|
|
"loss": 4.3649,
|
|
"mean_token_accuracy": 0.24765989035367966,
|
|
"num_tokens": 12535076.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 4.822365570068359,
|
|
"epoch": 5.677824267782427,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00025507915662895055,
|
|
"loss": 4.3575,
|
|
"mean_token_accuracy": 0.24410198181867598,
|
|
"num_tokens": 12544333.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 4.8724061965942385,
|
|
"epoch": 5.682008368200837,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00025475767840943403,
|
|
"loss": 4.3326,
|
|
"mean_token_accuracy": 0.24656929224729537,
|
|
"num_tokens": 12554292.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 4.903566503524781,
|
|
"epoch": 5.686192468619247,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002544362418453608,
|
|
"loss": 4.4079,
|
|
"mean_token_accuracy": 0.23687848895788194,
|
|
"num_tokens": 12564240.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 4.876070499420166,
|
|
"epoch": 5.690376569037657,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00025411484759819584,
|
|
"loss": 4.3148,
|
|
"mean_token_accuracy": 0.25219092667102816,
|
|
"num_tokens": 12573280.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 4.886744546890259,
|
|
"epoch": 5.694560669456067,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00025379349632931666,
|
|
"loss": 4.4458,
|
|
"mean_token_accuracy": 0.23624920696020127,
|
|
"num_tokens": 12583133.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 4.899893045425415,
|
|
"epoch": 5.698744769874477,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002534721887000127,
|
|
"loss": 4.3711,
|
|
"mean_token_accuracy": 0.24379950761795044,
|
|
"num_tokens": 12591709.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 4.88656177520752,
|
|
"epoch": 5.702928870292887,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002531509253714831,
|
|
"loss": 4.3077,
|
|
"mean_token_accuracy": 0.2527281329035759,
|
|
"num_tokens": 12600187.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 4.745095729827881,
|
|
"epoch": 5.707112970711297,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002528297070048364,
|
|
"loss": 4.288,
|
|
"mean_token_accuracy": 0.25166100859642027,
|
|
"num_tokens": 12608910.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 4.767807579040527,
|
|
"epoch": 5.711297071129707,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00025250853426108814,
|
|
"loss": 4.287,
|
|
"mean_token_accuracy": 0.25341928452253343,
|
|
"num_tokens": 12617370.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 4.86973729133606,
|
|
"epoch": 5.7154811715481175,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025218740780116037,
|
|
"loss": 4.3475,
|
|
"mean_token_accuracy": 0.24931021630764008,
|
|
"num_tokens": 12626146.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 4.871671199798584,
|
|
"epoch": 5.7196652719665275,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025186632828587977,
|
|
"loss": 4.3172,
|
|
"mean_token_accuracy": 0.24676501750946045,
|
|
"num_tokens": 12635565.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 4.920233106613159,
|
|
"epoch": 5.7238493723849375,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002515452963759761,
|
|
"loss": 4.3946,
|
|
"mean_token_accuracy": 0.23562282770872117,
|
|
"num_tokens": 12644869.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 4.890485334396362,
|
|
"epoch": 5.7280334728033475,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002512243127320817,
|
|
"loss": 4.3326,
|
|
"mean_token_accuracy": 0.24792505353689193,
|
|
"num_tokens": 12655215.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 4.819408178329468,
|
|
"epoch": 5.7322175732217575,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002509033780147291,
|
|
"loss": 4.2739,
|
|
"mean_token_accuracy": 0.2491241827607155,
|
|
"num_tokens": 12663684.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 4.8276762008667,
|
|
"epoch": 5.7364016736401675,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025058249288435037,
|
|
"loss": 4.3702,
|
|
"mean_token_accuracy": 0.23822278380393982,
|
|
"num_tokens": 12673367.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 4.844202709197998,
|
|
"epoch": 5.7405857740585775,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002502616580012756,
|
|
"loss": 4.3069,
|
|
"mean_token_accuracy": 0.2525498569011688,
|
|
"num_tokens": 12682472.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 4.811185598373413,
|
|
"epoch": 5.7447698744769875,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024994087402573133,
|
|
"loss": 4.2964,
|
|
"mean_token_accuracy": 0.2535608813166618,
|
|
"num_tokens": 12691215.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 4.773161220550537,
|
|
"epoch": 5.7489539748953975,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024962014161783946,
|
|
"loss": 4.2843,
|
|
"mean_token_accuracy": 0.24100177735090256,
|
|
"num_tokens": 12700332.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 4.828914737701416,
|
|
"epoch": 5.7531380753138075,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024929946143761585,
|
|
"loss": 4.3239,
|
|
"mean_token_accuracy": 0.2410706639289856,
|
|
"num_tokens": 12708886.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 4.86767349243164,
|
|
"epoch": 5.7573221757322175,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00024897883414496844,
|
|
"loss": 4.3041,
|
|
"mean_token_accuracy": 0.25252724140882493,
|
|
"num_tokens": 12717476.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 4.991838026046753,
|
|
"epoch": 5.7615062761506275,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024865826039969683,
|
|
"loss": 4.4561,
|
|
"mean_token_accuracy": 0.23524223864078522,
|
|
"num_tokens": 12726737.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 4.823116111755371,
|
|
"epoch": 5.7656903765690375,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002483377408614903,
|
|
"loss": 4.3356,
|
|
"mean_token_accuracy": 0.25061206370592115,
|
|
"num_tokens": 12735750.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 4.768174314498902,
|
|
"epoch": 5.7698744769874475,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00024801727618992635,
|
|
"loss": 4.3337,
|
|
"mean_token_accuracy": 0.24487781077623366,
|
|
"num_tokens": 12745143.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 4.82567458152771,
|
|
"epoch": 5.7740585774058575,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00024769686704446996,
|
|
"loss": 4.3269,
|
|
"mean_token_accuracy": 0.24116038680076599,
|
|
"num_tokens": 12754623.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 4.950635766983032,
|
|
"epoch": 5.7782426778242675,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024737651408447155,
|
|
"loss": 4.3612,
|
|
"mean_token_accuracy": 0.25221560448408126,
|
|
"num_tokens": 12762921.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 4.885443496704101,
|
|
"epoch": 5.7824267782426775,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000247056217969166,
|
|
"loss": 4.3938,
|
|
"mean_token_accuracy": 0.24490638226270675,
|
|
"num_tokens": 12772159.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 4.822702932357788,
|
|
"epoch": 5.786610878661088,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00024673597935767123,
|
|
"loss": 4.3791,
|
|
"mean_token_accuracy": 0.2413331851363182,
|
|
"num_tokens": 12781916.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 4.845784902572632,
|
|
"epoch": 5.790794979079498,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002464157989089867,
|
|
"loss": 4.3666,
|
|
"mean_token_accuracy": 0.24063920676708223,
|
|
"num_tokens": 12791275.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 4.83692364692688,
|
|
"epoch": 5.794979079497908,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002460956772819925,
|
|
"loss": 4.3377,
|
|
"mean_token_accuracy": 0.24959323853254317,
|
|
"num_tokens": 12800667.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 4.825749349594116,
|
|
"epoch": 5.799163179916318,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024577561513544724,
|
|
"loss": 4.3388,
|
|
"mean_token_accuracy": 0.24317169189453125,
|
|
"num_tokens": 12808986.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 4.857340526580811,
|
|
"epoch": 5.803347280334728,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00024545561312798756,
|
|
"loss": 4.3735,
|
|
"mean_token_accuracy": 0.2508589461445808,
|
|
"num_tokens": 12817658.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 4.8928533554077145,
|
|
"epoch": 5.807531380753138,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024513567191812607,
|
|
"loss": 4.3702,
|
|
"mean_token_accuracy": 0.24639694690704345,
|
|
"num_tokens": 12826301.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 4.927112674713134,
|
|
"epoch": 5.811715481171548,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002448157921642505,
|
|
"loss": 4.398,
|
|
"mean_token_accuracy": 0.2369797110557556,
|
|
"num_tokens": 12836295.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 4.830571985244751,
|
|
"epoch": 5.815899581589958,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00024449597452462165,
|
|
"loss": 4.352,
|
|
"mean_token_accuracy": 0.2420512393116951,
|
|
"num_tokens": 12847128.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 4.827253437042236,
|
|
"epoch": 5.820083682008368,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002441762196573729,
|
|
"loss": 4.3841,
|
|
"mean_token_accuracy": 0.2383188486099243,
|
|
"num_tokens": 12856198.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 4.888354730606079,
|
|
"epoch": 5.824267782426778,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00024385652822050846,
|
|
"loss": 4.3025,
|
|
"mean_token_accuracy": 0.24968597590923308,
|
|
"num_tokens": 12865513.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 4.891372919082642,
|
|
"epoch": 5.828451882845188,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024353690087190188,
|
|
"loss": 4.3281,
|
|
"mean_token_accuracy": 0.24556171000003815,
|
|
"num_tokens": 12874086.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 4.8227348804473875,
|
|
"epoch": 5.832635983263598,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024321733826929487,
|
|
"loss": 4.3556,
|
|
"mean_token_accuracy": 0.2466444417834282,
|
|
"num_tokens": 12883382.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 4.919569587707519,
|
|
"epoch": 5.836820083682008,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00024289784107029588,
|
|
"loss": 4.3923,
|
|
"mean_token_accuracy": 0.24815418422222138,
|
|
"num_tokens": 12893510.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 4.926452159881592,
|
|
"epoch": 5.841004184100418,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00024257840993237862,
|
|
"loss": 4.4242,
|
|
"mean_token_accuracy": 0.23840502947568892,
|
|
"num_tokens": 12902719.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 4.915157175064087,
|
|
"epoch": 5.845188284518828,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00024225904551288124,
|
|
"loss": 4.4426,
|
|
"mean_token_accuracy": 0.23536291122436523,
|
|
"num_tokens": 12912889.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 4.851192283630371,
|
|
"epoch": 5.849372384937238,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024193974846900424,
|
|
"loss": 4.2861,
|
|
"mean_token_accuracy": 0.25262907296419146,
|
|
"num_tokens": 12922454.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 4.850388383865356,
|
|
"epoch": 5.853556485355648,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024162051945780967,
|
|
"loss": 4.3135,
|
|
"mean_token_accuracy": 0.2501085877418518,
|
|
"num_tokens": 12931038.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 4.814456033706665,
|
|
"epoch": 5.857740585774058,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002413013591362195,
|
|
"loss": 4.386,
|
|
"mean_token_accuracy": 0.24322551488876343,
|
|
"num_tokens": 12940830.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 5.857740585774058,
|
|
"eval_entropy": 4.686463615889442,
|
|
"eval_loss": 5.272697925567627,
|
|
"eval_mean_token_accuracy": 0.20518442406366696,
|
|
"eval_num_tokens": 12940830.0,
|
|
"eval_runtime": 2.755,
|
|
"eval_samples_per_second": 1418.148,
|
|
"eval_steps_per_second": 177.495,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 4.876022148132324,
|
|
"epoch": 5.861924686192468,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00024098226816101427,
|
|
"loss": 4.3956,
|
|
"mean_token_accuracy": 0.2402362823486328,
|
|
"num_tokens": 12949159.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 4.948538970947266,
|
|
"epoch": 5.866108786610878,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00024066324718883205,
|
|
"loss": 4.4167,
|
|
"mean_token_accuracy": 0.2419355720281601,
|
|
"num_tokens": 12957818.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 4.921017980575561,
|
|
"epoch": 5.870292887029288,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002403442968761666,
|
|
"loss": 4.4405,
|
|
"mean_token_accuracy": 0.23638582676649095,
|
|
"num_tokens": 12966667.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 4.898641777038574,
|
|
"epoch": 5.874476987447698,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00024002541787936644,
|
|
"loss": 4.3121,
|
|
"mean_token_accuracy": 0.24076194167137147,
|
|
"num_tokens": 12976081.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 4.853302001953125,
|
|
"epoch": 5.878661087866108,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002397066108546332,
|
|
"loss": 4.4079,
|
|
"mean_token_accuracy": 0.2389953002333641,
|
|
"num_tokens": 12985006.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 4.836811351776123,
|
|
"epoch": 5.882845188284519,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023938787645802056,
|
|
"loss": 4.3333,
|
|
"mean_token_accuracy": 0.24848998337984085,
|
|
"num_tokens": 12993862.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 4.814244651794434,
|
|
"epoch": 5.887029288702929,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023906921534543269,
|
|
"loss": 4.359,
|
|
"mean_token_accuracy": 0.24551441073417662,
|
|
"num_tokens": 13003637.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 4.8850359439849855,
|
|
"epoch": 5.891213389121339,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023875062817262284,
|
|
"loss": 4.4024,
|
|
"mean_token_accuracy": 0.2458694264292717,
|
|
"num_tokens": 13013364.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 4.82855749130249,
|
|
"epoch": 5.895397489539749,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023843211559519228,
|
|
"loss": 4.3653,
|
|
"mean_token_accuracy": 0.24699858129024505,
|
|
"num_tokens": 13023432.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 4.904236125946045,
|
|
"epoch": 5.899581589958159,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023811367826858854,
|
|
"loss": 4.393,
|
|
"mean_token_accuracy": 0.24079672247171402,
|
|
"num_tokens": 13034293.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 4.8956193923950195,
|
|
"epoch": 5.903765690376569,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023779531684810462,
|
|
"loss": 4.3925,
|
|
"mean_token_accuracy": 0.24243488311767578,
|
|
"num_tokens": 13043985.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 4.773081541061401,
|
|
"epoch": 5.907949790794979,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023747703198887698,
|
|
"loss": 4.2279,
|
|
"mean_token_accuracy": 0.25836834162473676,
|
|
"num_tokens": 13052339.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 4.8416666984558105,
|
|
"epoch": 5.912133891213389,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023715882434588483,
|
|
"loss": 4.3078,
|
|
"mean_token_accuracy": 0.24352894425392152,
|
|
"num_tokens": 13060858.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 4.850029325485229,
|
|
"epoch": 5.916317991631799,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023684069457394826,
|
|
"loss": 4.2871,
|
|
"mean_token_accuracy": 0.2580322936177254,
|
|
"num_tokens": 13069542.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 4.772866106033325,
|
|
"epoch": 5.920502092050209,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023652264332772721,
|
|
"loss": 4.2365,
|
|
"mean_token_accuracy": 0.2581642419099808,
|
|
"num_tokens": 13078868.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 4.881327533721924,
|
|
"epoch": 5.924686192468619,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023620467126172001,
|
|
"loss": 4.4164,
|
|
"mean_token_accuracy": 0.23502912670373916,
|
|
"num_tokens": 13088573.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 4.8626360416412355,
|
|
"epoch": 5.928870292887029,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023588677903026202,
|
|
"loss": 4.3384,
|
|
"mean_token_accuracy": 0.24132190346717836,
|
|
"num_tokens": 13097723.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 4.8994691371917725,
|
|
"epoch": 5.933054393305439,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023556896728752435,
|
|
"loss": 4.2996,
|
|
"mean_token_accuracy": 0.2460368901491165,
|
|
"num_tokens": 13106071.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 4.795166969299316,
|
|
"epoch": 5.937238493723849,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002352512366875125,
|
|
"loss": 4.3414,
|
|
"mean_token_accuracy": 0.2419857680797577,
|
|
"num_tokens": 13115150.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 4.821146726608276,
|
|
"epoch": 5.941422594142259,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000234933587884065,
|
|
"loss": 4.3986,
|
|
"mean_token_accuracy": 0.2459180861711502,
|
|
"num_tokens": 13124833.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 4.907818174362182,
|
|
"epoch": 5.945606694560669,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000234616021530852,
|
|
"loss": 4.4259,
|
|
"mean_token_accuracy": 0.23949419260025023,
|
|
"num_tokens": 13135751.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 4.816760444641114,
|
|
"epoch": 5.949790794979079,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023429853828137393,
|
|
"loss": 4.2509,
|
|
"mean_token_accuracy": 0.25490049719810487,
|
|
"num_tokens": 13144285.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 4.835776376724243,
|
|
"epoch": 5.95397489539749,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023398113878896033,
|
|
"loss": 4.3465,
|
|
"mean_token_accuracy": 0.24473825097084045,
|
|
"num_tokens": 13152856.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 4.901019191741943,
|
|
"epoch": 5.9581589958159,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023366382370676824,
|
|
"loss": 4.375,
|
|
"mean_token_accuracy": 0.2433757022023201,
|
|
"num_tokens": 13162665.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 4.850583076477051,
|
|
"epoch": 5.96234309623431,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023334659368778117,
|
|
"loss": 4.2994,
|
|
"mean_token_accuracy": 0.2559014156460762,
|
|
"num_tokens": 13171110.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 4.845103311538696,
|
|
"epoch": 5.96652719665272,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023302944938480752,
|
|
"loss": 4.3733,
|
|
"mean_token_accuracy": 0.2388714149594307,
|
|
"num_tokens": 13179948.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 4.920997762680054,
|
|
"epoch": 5.97071129707113,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023271239145047917,
|
|
"loss": 4.3941,
|
|
"mean_token_accuracy": 0.24668112099170686,
|
|
"num_tokens": 13190214.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 4.747384977340698,
|
|
"epoch": 5.97489539748954,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023239542053725044,
|
|
"loss": 4.2515,
|
|
"mean_token_accuracy": 0.26136850118637084,
|
|
"num_tokens": 13199048.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 4.831013011932373,
|
|
"epoch": 5.97907949790795,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00023207853729739663,
|
|
"loss": 4.3441,
|
|
"mean_token_accuracy": 0.24796175062656403,
|
|
"num_tokens": 13208553.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 4.918143939971924,
|
|
"epoch": 5.98326359832636,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023176174238301228,
|
|
"loss": 4.4328,
|
|
"mean_token_accuracy": 0.2364555835723877,
|
|
"num_tokens": 13218151.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 4.836220216751099,
|
|
"epoch": 5.98744769874477,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002314450364460105,
|
|
"loss": 4.3249,
|
|
"mean_token_accuracy": 0.25092939883470533,
|
|
"num_tokens": 13227730.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 4.839579248428345,
|
|
"epoch": 5.99163179916318,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002311284201381212,
|
|
"loss": 4.4176,
|
|
"mean_token_accuracy": 0.23747602105140686,
|
|
"num_tokens": 13237449.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 4.856749629974365,
|
|
"epoch": 5.99581589958159,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023081189411088972,
|
|
"loss": 4.3246,
|
|
"mean_token_accuracy": 0.24239298105239868,
|
|
"num_tokens": 13246662.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 4.905408620834351,
|
|
"epoch": 6.0,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023049545901567586,
|
|
"loss": 4.3816,
|
|
"mean_token_accuracy": 0.24112735986709594,
|
|
"num_tokens": 13255932.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 4.810820245742798,
|
|
"epoch": 6.00418410041841,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023017911550365233,
|
|
"loss": 4.1216,
|
|
"mean_token_accuracy": 0.26554025411605836,
|
|
"num_tokens": 13264600.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 4.921321249008178,
|
|
"epoch": 6.00836820083682,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022986286422580288,
|
|
"loss": 4.2186,
|
|
"mean_token_accuracy": 0.25846633315086365,
|
|
"num_tokens": 13274261.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 4.866313457489014,
|
|
"epoch": 6.01255230125523,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022954670583292196,
|
|
"loss": 4.1636,
|
|
"mean_token_accuracy": 0.26621998846530914,
|
|
"num_tokens": 13284150.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 4.895343208312989,
|
|
"epoch": 6.01673640167364,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022923064097561265,
|
|
"loss": 4.1466,
|
|
"mean_token_accuracy": 0.2531894937157631,
|
|
"num_tokens": 13293114.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 4.826653575897216,
|
|
"epoch": 6.02092050209205,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002289146703042857,
|
|
"loss": 4.2067,
|
|
"mean_token_accuracy": 0.25964345782995224,
|
|
"num_tokens": 13303391.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 4.888730859756469,
|
|
"epoch": 6.02510460251046,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022859879446915788,
|
|
"loss": 4.2651,
|
|
"mean_token_accuracy": 0.25642198622226714,
|
|
"num_tokens": 13312372.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 4.892520618438721,
|
|
"epoch": 6.02928870292887,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022828301412025094,
|
|
"loss": 4.1932,
|
|
"mean_token_accuracy": 0.26270593851804735,
|
|
"num_tokens": 13321543.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 4.8548243045806885,
|
|
"epoch": 6.03347280334728,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002279673299073901,
|
|
"loss": 4.1554,
|
|
"mean_token_accuracy": 0.2666091278195381,
|
|
"num_tokens": 13330179.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 4.837588262557984,
|
|
"epoch": 6.03765690376569,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022765174248020258,
|
|
"loss": 4.1208,
|
|
"mean_token_accuracy": 0.2617232441902161,
|
|
"num_tokens": 13339116.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 4.766485357284546,
|
|
"epoch": 6.0418410041841,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00022733625248811656,
|
|
"loss": 4.1511,
|
|
"mean_token_accuracy": 0.26560292541980746,
|
|
"num_tokens": 13347908.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 4.749566221237183,
|
|
"epoch": 6.046025104602511,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022702086058035985,
|
|
"loss": 4.1204,
|
|
"mean_token_accuracy": 0.2679804444313049,
|
|
"num_tokens": 13356840.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 4.822508335113525,
|
|
"epoch": 6.050209205020921,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022670556740595827,
|
|
"loss": 4.1559,
|
|
"mean_token_accuracy": 0.26190525889396665,
|
|
"num_tokens": 13365288.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 4.77788200378418,
|
|
"epoch": 6.054393305439331,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00022639037361373444,
|
|
"loss": 4.1187,
|
|
"mean_token_accuracy": 0.2753115326166153,
|
|
"num_tokens": 13374612.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 4.841673135757446,
|
|
"epoch": 6.058577405857741,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022607527985230648,
|
|
"loss": 4.1546,
|
|
"mean_token_accuracy": 0.2624566093087196,
|
|
"num_tokens": 13384307.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 4.832222938537598,
|
|
"epoch": 6.062761506276151,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00022576028677008688,
|
|
"loss": 4.1739,
|
|
"mean_token_accuracy": 0.26002573072910307,
|
|
"num_tokens": 13392626.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 4.846151828765869,
|
|
"epoch": 6.066945606694561,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002254453950152804,
|
|
"loss": 4.157,
|
|
"mean_token_accuracy": 0.26350108832120894,
|
|
"num_tokens": 13401694.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 4.8211322784423825,
|
|
"epoch": 6.071129707112971,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002251306052358839,
|
|
"loss": 4.2036,
|
|
"mean_token_accuracy": 0.26554072052240374,
|
|
"num_tokens": 13410330.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 4.782461166381836,
|
|
"epoch": 6.075313807531381,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022481591807968417,
|
|
"loss": 4.1103,
|
|
"mean_token_accuracy": 0.26845871061086657,
|
|
"num_tokens": 13418988.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 4.853209686279297,
|
|
"epoch": 6.079497907949791,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002245013341942566,
|
|
"loss": 4.225,
|
|
"mean_token_accuracy": 0.25537670105695726,
|
|
"num_tokens": 13427487.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 4.846694755554199,
|
|
"epoch": 6.083682008368201,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022418685422696444,
|
|
"loss": 4.1484,
|
|
"mean_token_accuracy": 0.2699266761541367,
|
|
"num_tokens": 13436450.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 4.805011129379272,
|
|
"epoch": 6.087866108786611,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002238724788249568,
|
|
"loss": 4.1324,
|
|
"mean_token_accuracy": 0.2593142405152321,
|
|
"num_tokens": 13445363.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 4.803239107131958,
|
|
"epoch": 6.092050209205021,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022355820863516774,
|
|
"loss": 4.0705,
|
|
"mean_token_accuracy": 0.27310639023780825,
|
|
"num_tokens": 13453960.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 4.865917539596557,
|
|
"epoch": 6.096234309623431,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002232440443043147,
|
|
"loss": 4.204,
|
|
"mean_token_accuracy": 0.2665314942598343,
|
|
"num_tokens": 13463619.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 4.8380975246429445,
|
|
"epoch": 6.100418410041841,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002229299864788975,
|
|
"loss": 4.1675,
|
|
"mean_token_accuracy": 0.2650081068277359,
|
|
"num_tokens": 13473566.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 4.785891771316528,
|
|
"epoch": 6.104602510460251,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00022261603580519656,
|
|
"loss": 4.1001,
|
|
"mean_token_accuracy": 0.2666401967406273,
|
|
"num_tokens": 13481848.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 4.898573875427246,
|
|
"epoch": 6.108786610878661,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00022230219292927188,
|
|
"loss": 4.2678,
|
|
"mean_token_accuracy": 0.25051648169755936,
|
|
"num_tokens": 13491600.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 4.895514583587646,
|
|
"epoch": 6.112970711297071,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00022198845849696175,
|
|
"loss": 4.2426,
|
|
"mean_token_accuracy": 0.2568917855620384,
|
|
"num_tokens": 13500888.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 4.853113889694214,
|
|
"epoch": 6.117154811715481,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022167483315388108,
|
|
"loss": 4.1907,
|
|
"mean_token_accuracy": 0.26801941990852357,
|
|
"num_tokens": 13511031.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 4.793132781982422,
|
|
"epoch": 6.121338912133891,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022136131754542048,
|
|
"loss": 4.2185,
|
|
"mean_token_accuracy": 0.24858860671520233,
|
|
"num_tokens": 13520704.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 4.786058568954468,
|
|
"epoch": 6.125523012552302,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022104791231674465,
|
|
"loss": 4.1091,
|
|
"mean_token_accuracy": 0.26482129991054537,
|
|
"num_tokens": 13530297.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 4.817823076248169,
|
|
"epoch": 6.129707112970712,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022073461811279106,
|
|
"loss": 4.1192,
|
|
"mean_token_accuracy": 0.265226511657238,
|
|
"num_tokens": 13538796.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 4.707890224456787,
|
|
"epoch": 6.133891213389122,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022042143557826892,
|
|
"loss": 4.0776,
|
|
"mean_token_accuracy": 0.27148049175739286,
|
|
"num_tokens": 13548145.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 4.692135143280029,
|
|
"epoch": 6.138075313807532,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022010836535765749,
|
|
"loss": 4.099,
|
|
"mean_token_accuracy": 0.26237593591213226,
|
|
"num_tokens": 13557025.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 4.74822506904602,
|
|
"epoch": 6.142259414225942,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021979540809520504,
|
|
"loss": 4.1746,
|
|
"mean_token_accuracy": 0.26509158462286,
|
|
"num_tokens": 13566485.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 4.87022819519043,
|
|
"epoch": 6.146443514644352,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002194825644349272,
|
|
"loss": 4.1723,
|
|
"mean_token_accuracy": 0.25893722623586657,
|
|
"num_tokens": 13575167.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 4.928844594955445,
|
|
"epoch": 6.150627615062762,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021916983502060595,
|
|
"loss": 4.1941,
|
|
"mean_token_accuracy": 0.25857413858175277,
|
|
"num_tokens": 13585757.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 4.808111333847046,
|
|
"epoch": 6.154811715481172,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00021885722049578817,
|
|
"loss": 4.2279,
|
|
"mean_token_accuracy": 0.2557098299264908,
|
|
"num_tokens": 13595862.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 4.858302974700928,
|
|
"epoch": 6.158995815899582,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002185447215037842,
|
|
"loss": 4.1789,
|
|
"mean_token_accuracy": 0.2565613254904747,
|
|
"num_tokens": 13605106.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 4.80931978225708,
|
|
"epoch": 6.163179916317992,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021823233868766684,
|
|
"loss": 4.1333,
|
|
"mean_token_accuracy": 0.26499359160661695,
|
|
"num_tokens": 13614317.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 4.7676959991455075,
|
|
"epoch": 6.167364016736402,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002179200726902697,
|
|
"loss": 4.0929,
|
|
"mean_token_accuracy": 0.27164238691329956,
|
|
"num_tokens": 13623196.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 4.777904367446899,
|
|
"epoch": 6.171548117154812,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002176079241541859,
|
|
"loss": 4.1334,
|
|
"mean_token_accuracy": 0.26596971303224565,
|
|
"num_tokens": 13631872.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 4.867989778518677,
|
|
"epoch": 6.175732217573222,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000217295893721767,
|
|
"loss": 4.1602,
|
|
"mean_token_accuracy": 0.260381805896759,
|
|
"num_tokens": 13641240.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 4.747178316116333,
|
|
"epoch": 6.179916317991632,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021698398203512148,
|
|
"loss": 4.1225,
|
|
"mean_token_accuracy": 0.2657135590910912,
|
|
"num_tokens": 13650495.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 4.728201818466187,
|
|
"epoch": 6.184100418410042,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021667218973611346,
|
|
"loss": 4.0847,
|
|
"mean_token_accuracy": 0.27276587039232253,
|
|
"num_tokens": 13660234.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 4.826923561096192,
|
|
"epoch": 6.188284518828452,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021636051746636126,
|
|
"loss": 4.257,
|
|
"mean_token_accuracy": 0.25639844238758086,
|
|
"num_tokens": 13670060.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 4.867006635665893,
|
|
"epoch": 6.192468619246862,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021604896586723644,
|
|
"loss": 4.1985,
|
|
"mean_token_accuracy": 0.2615767255425453,
|
|
"num_tokens": 13679551.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 4.905156421661377,
|
|
"epoch": 6.196652719665272,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021573753557986198,
|
|
"loss": 4.1994,
|
|
"mean_token_accuracy": 0.2638272553682327,
|
|
"num_tokens": 13688772.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 4.915416288375854,
|
|
"epoch": 6.200836820083682,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021542622724511146,
|
|
"loss": 4.2228,
|
|
"mean_token_accuracy": 0.2579490229487419,
|
|
"num_tokens": 13698110.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 4.845192766189575,
|
|
"epoch": 6.205020920502092,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00021511504150360724,
|
|
"loss": 4.1466,
|
|
"mean_token_accuracy": 0.2619974821805954,
|
|
"num_tokens": 13707459.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 4.846059656143188,
|
|
"epoch": 6.209205020920502,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00021480397899571963,
|
|
"loss": 4.2368,
|
|
"mean_token_accuracy": 0.2580647110939026,
|
|
"num_tokens": 13717596.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 4.796379566192627,
|
|
"epoch": 6.2133891213389125,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00021449304036156525,
|
|
"loss": 4.1425,
|
|
"mean_token_accuracy": 0.26614964157342913,
|
|
"num_tokens": 13726871.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 4.83381757736206,
|
|
"epoch": 6.2175732217573225,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00021418222624100587,
|
|
"loss": 4.1199,
|
|
"mean_token_accuracy": 0.26769975423812864,
|
|
"num_tokens": 13734960.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 4.797351932525634,
|
|
"epoch": 6.2217573221757325,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021387153727364677,
|
|
"loss": 4.1483,
|
|
"mean_token_accuracy": 0.2664002373814583,
|
|
"num_tokens": 13743771.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 4.771816778182983,
|
|
"epoch": 6.2259414225941425,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002135609740988362,
|
|
"loss": 4.106,
|
|
"mean_token_accuracy": 0.2760073482990265,
|
|
"num_tokens": 13753080.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 4.8374475002288815,
|
|
"epoch": 6.2301255230125525,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002132505373556632,
|
|
"loss": 4.2472,
|
|
"mean_token_accuracy": 0.2566974610090256,
|
|
"num_tokens": 13762698.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 4.790829849243164,
|
|
"epoch": 6.2343096234309625,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002129402276829564,
|
|
"loss": 4.1402,
|
|
"mean_token_accuracy": 0.26138503700494764,
|
|
"num_tokens": 13771693.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 4.850272178649902,
|
|
"epoch": 6.2384937238493725,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021263004571928345,
|
|
"loss": 4.247,
|
|
"mean_token_accuracy": 0.257717527449131,
|
|
"num_tokens": 13780939.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 4.787961006164551,
|
|
"epoch": 6.2426778242677825,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021231999210294882,
|
|
"loss": 4.1103,
|
|
"mean_token_accuracy": 0.26710599958896636,
|
|
"num_tokens": 13790001.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 4.888015079498291,
|
|
"epoch": 6.2468619246861925,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00021201006747199313,
|
|
"loss": 4.1908,
|
|
"mean_token_accuracy": 0.26887045949697497,
|
|
"num_tokens": 13799360.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 4.794003677368164,
|
|
"epoch": 6.2510460251046025,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021170027246419137,
|
|
"loss": 4.1757,
|
|
"mean_token_accuracy": 0.26355336159467696,
|
|
"num_tokens": 13808518.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 4.87125678062439,
|
|
"epoch": 6.2552301255230125,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021139060771705205,
|
|
"loss": 4.252,
|
|
"mean_token_accuracy": 0.25477629005908964,
|
|
"num_tokens": 13818144.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 4.837577533721924,
|
|
"epoch": 6.2594142259414225,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021108107386781517,
|
|
"loss": 4.144,
|
|
"mean_token_accuracy": 0.26705680042505264,
|
|
"num_tokens": 13828063.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 4.740513038635254,
|
|
"epoch": 6.2635983263598325,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021077167155345166,
|
|
"loss": 4.1188,
|
|
"mean_token_accuracy": 0.25982263386249543,
|
|
"num_tokens": 13837014.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 4.7825840473175045,
|
|
"epoch": 6.2677824267782425,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002104624014106618,
|
|
"loss": 4.2047,
|
|
"mean_token_accuracy": 0.26653329133987425,
|
|
"num_tokens": 13845518.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 4.8209367275238035,
|
|
"epoch": 6.2719665271966525,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021015326407587389,
|
|
"loss": 4.1977,
|
|
"mean_token_accuracy": 0.2608263731002808,
|
|
"num_tokens": 13854154.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 4.8261792182922365,
|
|
"epoch": 6.2761506276150625,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020984426018524278,
|
|
"loss": 4.192,
|
|
"mean_token_accuracy": 0.26873925477266314,
|
|
"num_tokens": 13863404.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.2761506276150625,
|
|
"eval_entropy": 4.6755887696591865,
|
|
"eval_loss": 5.279442310333252,
|
|
"eval_mean_token_accuracy": 0.20572668390161908,
|
|
"eval_num_tokens": 13863404.0,
|
|
"eval_runtime": 2.757,
|
|
"eval_samples_per_second": 1417.099,
|
|
"eval_steps_per_second": 177.364,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 4.74080262184143,
|
|
"epoch": 6.2803347280334725,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020953539037464882,
|
|
"loss": 4.0508,
|
|
"mean_token_accuracy": 0.27842895686626434,
|
|
"num_tokens": 13872554.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 4.766878318786621,
|
|
"epoch": 6.2845188284518825,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020922665527969646,
|
|
"loss": 4.1532,
|
|
"mean_token_accuracy": 0.26676701158285143,
|
|
"num_tokens": 13881740.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 4.849253606796265,
|
|
"epoch": 6.2887029288702925,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002089180555357128,
|
|
"loss": 4.2248,
|
|
"mean_token_accuracy": 0.2654907241463661,
|
|
"num_tokens": 13890114.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 4.832908868789673,
|
|
"epoch": 6.292887029288703,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020860959177774652,
|
|
"loss": 4.3021,
|
|
"mean_token_accuracy": 0.24932253509759902,
|
|
"num_tokens": 13899744.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 4.89445858001709,
|
|
"epoch": 6.297071129707113,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00020830126464056647,
|
|
"loss": 4.1924,
|
|
"mean_token_accuracy": 0.2563816949725151,
|
|
"num_tokens": 13909011.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 4.887471485137939,
|
|
"epoch": 6.301255230125523,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020799307475866035,
|
|
"loss": 4.1804,
|
|
"mean_token_accuracy": 0.2554214134812355,
|
|
"num_tokens": 13918435.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 4.788136768341064,
|
|
"epoch": 6.305439330543933,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020768502276623336,
|
|
"loss": 4.1788,
|
|
"mean_token_accuracy": 0.2665564477443695,
|
|
"num_tokens": 13927164.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 4.784514284133911,
|
|
"epoch": 6.309623430962343,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020737710929720694,
|
|
"loss": 4.0902,
|
|
"mean_token_accuracy": 0.2790061205625534,
|
|
"num_tokens": 13935694.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 4.9540105819702145,
|
|
"epoch": 6.313807531380753,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00020706933498521767,
|
|
"loss": 4.2693,
|
|
"mean_token_accuracy": 0.25293782651424407,
|
|
"num_tokens": 13946068.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 4.907020616531372,
|
|
"epoch": 6.317991631799163,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002067617004636153,
|
|
"loss": 4.2531,
|
|
"mean_token_accuracy": 0.2560804322361946,
|
|
"num_tokens": 13955238.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 4.823457050323486,
|
|
"epoch": 6.322175732217573,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020645420636546246,
|
|
"loss": 4.1402,
|
|
"mean_token_accuracy": 0.27223782390356066,
|
|
"num_tokens": 13964251.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 4.836338186264038,
|
|
"epoch": 6.326359832635983,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00020614685332353246,
|
|
"loss": 4.2879,
|
|
"mean_token_accuracy": 0.24714537262916564,
|
|
"num_tokens": 13974111.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 4.82497878074646,
|
|
"epoch": 6.330543933054393,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020583964197030852,
|
|
"loss": 4.1407,
|
|
"mean_token_accuracy": 0.2597179844975471,
|
|
"num_tokens": 13982842.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 4.738140010833741,
|
|
"epoch": 6.334728033472803,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020553257293798216,
|
|
"loss": 4.1113,
|
|
"mean_token_accuracy": 0.27079141438007354,
|
|
"num_tokens": 13992010.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 4.793126058578491,
|
|
"epoch": 6.338912133891213,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020522564685845207,
|
|
"loss": 4.0862,
|
|
"mean_token_accuracy": 0.2773148462176323,
|
|
"num_tokens": 14000746.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 4.82713360786438,
|
|
"epoch": 6.343096234309623,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00020491886436332287,
|
|
"loss": 4.1598,
|
|
"mean_token_accuracy": 0.26745514273643495,
|
|
"num_tokens": 14010664.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 4.846192932128906,
|
|
"epoch": 6.347280334728033,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020461222608390346,
|
|
"loss": 4.1925,
|
|
"mean_token_accuracy": 0.254590205848217,
|
|
"num_tokens": 14019473.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 4.797559118270874,
|
|
"epoch": 6.351464435146443,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020430573265120623,
|
|
"loss": 4.1913,
|
|
"mean_token_accuracy": 0.26261122077703475,
|
|
"num_tokens": 14027971.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 4.696042823791504,
|
|
"epoch": 6.355648535564853,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002039993846959453,
|
|
"loss": 4.0189,
|
|
"mean_token_accuracy": 0.27458125203847883,
|
|
"num_tokens": 14036972.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 4.835784149169922,
|
|
"epoch": 6.359832635983263,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020369318284853562,
|
|
"loss": 4.1531,
|
|
"mean_token_accuracy": 0.25532643496990204,
|
|
"num_tokens": 14046486.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 4.799405336380005,
|
|
"epoch": 6.364016736401673,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020338712773909125,
|
|
"loss": 4.1598,
|
|
"mean_token_accuracy": 0.2661730721592903,
|
|
"num_tokens": 14056398.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 4.92598237991333,
|
|
"epoch": 6.368200836820083,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020308121999742458,
|
|
"loss": 4.2931,
|
|
"mean_token_accuracy": 0.248145954310894,
|
|
"num_tokens": 14065517.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 4.8304990291595455,
|
|
"epoch": 6.372384937238493,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020277546025304432,
|
|
"loss": 4.1753,
|
|
"mean_token_accuracy": 0.2619725540280342,
|
|
"num_tokens": 14074563.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 4.812159729003906,
|
|
"epoch": 6.376569037656903,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00020246984913515503,
|
|
"loss": 4.2137,
|
|
"mean_token_accuracy": 0.2603787198662758,
|
|
"num_tokens": 14083125.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 4.819386291503906,
|
|
"epoch": 6.380753138075314,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020216438727265517,
|
|
"loss": 4.1447,
|
|
"mean_token_accuracy": 0.2633805125951767,
|
|
"num_tokens": 14091965.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 4.74980731010437,
|
|
"epoch": 6.384937238493724,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020185907529413622,
|
|
"loss": 4.1406,
|
|
"mean_token_accuracy": 0.25715351700782774,
|
|
"num_tokens": 14100872.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 4.85809965133667,
|
|
"epoch": 6.389121338912134,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020155391382788108,
|
|
"loss": 4.2072,
|
|
"mean_token_accuracy": 0.261527007818222,
|
|
"num_tokens": 14109860.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 4.895865106582642,
|
|
"epoch": 6.393305439330544,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020124890350186304,
|
|
"loss": 4.2353,
|
|
"mean_token_accuracy": 0.2572548881173134,
|
|
"num_tokens": 14119256.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 4.867618036270142,
|
|
"epoch": 6.397489539748954,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002009440449437443,
|
|
"loss": 4.3123,
|
|
"mean_token_accuracy": 0.24977186024188996,
|
|
"num_tokens": 14128160.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 4.832208776473999,
|
|
"epoch": 6.401673640167364,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002006393387808747,
|
|
"loss": 4.2634,
|
|
"mean_token_accuracy": 0.25452412813901903,
|
|
"num_tokens": 14137810.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 4.864672708511352,
|
|
"epoch": 6.405857740585774,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020033478564029063,
|
|
"loss": 4.0768,
|
|
"mean_token_accuracy": 0.27408157140016554,
|
|
"num_tokens": 14145914.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 4.938950300216675,
|
|
"epoch": 6.410041841004184,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002000303861487134,
|
|
"loss": 4.293,
|
|
"mean_token_accuracy": 0.2480102688074112,
|
|
"num_tokens": 14156348.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 4.933863925933838,
|
|
"epoch": 6.414225941422594,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019972614093254822,
|
|
"loss": 4.3071,
|
|
"mean_token_accuracy": 0.25632224678993226,
|
|
"num_tokens": 14166079.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 4.789388227462768,
|
|
"epoch": 6.418410041841004,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00019942205061788292,
|
|
"loss": 4.1843,
|
|
"mean_token_accuracy": 0.25972204506397245,
|
|
"num_tokens": 14175732.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 4.843462371826172,
|
|
"epoch": 6.422594142259414,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00019911811583048645,
|
|
"loss": 4.2016,
|
|
"mean_token_accuracy": 0.261931012570858,
|
|
"num_tokens": 14184935.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 4.850495386123657,
|
|
"epoch": 6.426778242677824,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019881433719580766,
|
|
"loss": 4.2725,
|
|
"mean_token_accuracy": 0.2549947530031204,
|
|
"num_tokens": 14193693.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 4.779506874084473,
|
|
"epoch": 6.430962343096234,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019851071533897425,
|
|
"loss": 4.1666,
|
|
"mean_token_accuracy": 0.26724977046251297,
|
|
"num_tokens": 14203323.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 4.820087003707886,
|
|
"epoch": 6.435146443514644,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00019820725088479107,
|
|
"loss": 4.1402,
|
|
"mean_token_accuracy": 0.2597114399075508,
|
|
"num_tokens": 14211992.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 4.829944705963134,
|
|
"epoch": 6.439330543933054,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001979039444577393,
|
|
"loss": 4.1568,
|
|
"mean_token_accuracy": 0.2684673577547073,
|
|
"num_tokens": 14221578.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 4.842211198806763,
|
|
"epoch": 6.443514644351464,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019760079668197474,
|
|
"loss": 4.1936,
|
|
"mean_token_accuracy": 0.26198330223560334,
|
|
"num_tokens": 14231411.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 4.8337695598602295,
|
|
"epoch": 6.447698744769874,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019729780818132664,
|
|
"loss": 4.1548,
|
|
"mean_token_accuracy": 0.26438852548599245,
|
|
"num_tokens": 14241317.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 4.8444358825683596,
|
|
"epoch": 6.451882845188284,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019699497957929702,
|
|
"loss": 4.2186,
|
|
"mean_token_accuracy": 0.26537605822086335,
|
|
"num_tokens": 14251248.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 4.778241586685181,
|
|
"epoch": 6.456066945606695,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019669231149905804,
|
|
"loss": 4.1675,
|
|
"mean_token_accuracy": 0.25988617837429046,
|
|
"num_tokens": 14261487.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 4.826183032989502,
|
|
"epoch": 6.460251046025105,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019638980456345207,
|
|
"loss": 4.2326,
|
|
"mean_token_accuracy": 0.2471970021724701,
|
|
"num_tokens": 14271438.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 4.832457494735718,
|
|
"epoch": 6.464435146443515,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019608745939498983,
|
|
"loss": 4.2019,
|
|
"mean_token_accuracy": 0.2580079913139343,
|
|
"num_tokens": 14280683.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 4.8046448707580565,
|
|
"epoch": 6.468619246861925,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019578527661584898,
|
|
"loss": 4.2239,
|
|
"mean_token_accuracy": 0.2623331367969513,
|
|
"num_tokens": 14290041.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 4.833163261413574,
|
|
"epoch": 6.472803347280335,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019548325684787311,
|
|
"loss": 4.2428,
|
|
"mean_token_accuracy": 0.25395538806915285,
|
|
"num_tokens": 14298914.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 4.813269233703613,
|
|
"epoch": 6.476987447698745,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001951814007125704,
|
|
"loss": 4.1782,
|
|
"mean_token_accuracy": 0.2535831660032272,
|
|
"num_tokens": 14308347.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 4.796567296981811,
|
|
"epoch": 6.481171548117155,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00019487970883111234,
|
|
"loss": 4.1891,
|
|
"mean_token_accuracy": 0.2578530892729759,
|
|
"num_tokens": 14317839.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 4.797550296783447,
|
|
"epoch": 6.485355648535565,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00019457818182433206,
|
|
"loss": 4.1932,
|
|
"mean_token_accuracy": 0.25826616287231446,
|
|
"num_tokens": 14327037.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 4.80318398475647,
|
|
"epoch": 6.489539748953975,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019427682031272382,
|
|
"loss": 4.1834,
|
|
"mean_token_accuracy": 0.2653974503278732,
|
|
"num_tokens": 14337417.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 4.950545597076416,
|
|
"epoch": 6.493723849372385,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019397562491644112,
|
|
"loss": 4.2943,
|
|
"mean_token_accuracy": 0.25625875145196914,
|
|
"num_tokens": 14347037.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 4.8573352813720705,
|
|
"epoch": 6.497907949790795,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019367459625529564,
|
|
"loss": 4.2364,
|
|
"mean_token_accuracy": 0.2562482461333275,
|
|
"num_tokens": 14356509.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 4.7933619022369385,
|
|
"epoch": 6.502092050209205,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00019337373494875604,
|
|
"loss": 4.1128,
|
|
"mean_token_accuracy": 0.26787490397691727,
|
|
"num_tokens": 14365276.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 4.877886819839477,
|
|
"epoch": 6.506276150627615,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019307304161594657,
|
|
"loss": 4.216,
|
|
"mean_token_accuracy": 0.2640374183654785,
|
|
"num_tokens": 14373546.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 4.804327821731567,
|
|
"epoch": 6.510460251046025,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019277251687564557,
|
|
"loss": 4.1619,
|
|
"mean_token_accuracy": 0.2689590513706207,
|
|
"num_tokens": 14382767.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 4.753371477127075,
|
|
"epoch": 6.514644351464435,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019247216134628476,
|
|
"loss": 4.1786,
|
|
"mean_token_accuracy": 0.25304106175899505,
|
|
"num_tokens": 14392602.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 4.810116863250732,
|
|
"epoch": 6.518828451882845,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019217197564594746,
|
|
"loss": 4.2571,
|
|
"mean_token_accuracy": 0.25515416264533997,
|
|
"num_tokens": 14402191.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 4.852352571487427,
|
|
"epoch": 6.523012552301255,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019187196039236772,
|
|
"loss": 4.1865,
|
|
"mean_token_accuracy": 0.2620208814740181,
|
|
"num_tokens": 14411116.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 4.763415288925171,
|
|
"epoch": 6.527196652719665,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019157211620292853,
|
|
"loss": 4.1293,
|
|
"mean_token_accuracy": 0.26632496118545534,
|
|
"num_tokens": 14419725.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 4.86669430732727,
|
|
"epoch": 6.531380753138075,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019127244369466118,
|
|
"loss": 4.2619,
|
|
"mean_token_accuracy": 0.25516562908887863,
|
|
"num_tokens": 14429904.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 4.792462158203125,
|
|
"epoch": 6.535564853556485,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00019097294348424348,
|
|
"loss": 4.1501,
|
|
"mean_token_accuracy": 0.25916097313165665,
|
|
"num_tokens": 14439384.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 4.806901311874389,
|
|
"epoch": 6.539748953974895,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019067361618799856,
|
|
"loss": 4.1185,
|
|
"mean_token_accuracy": 0.27518945932388306,
|
|
"num_tokens": 14448002.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 4.801858186721802,
|
|
"epoch": 6.543933054393305,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019037446242189398,
|
|
"loss": 4.1596,
|
|
"mean_token_accuracy": 0.2558473840355873,
|
|
"num_tokens": 14456887.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 4.877001619338989,
|
|
"epoch": 6.548117154811716,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00019007548280154007,
|
|
"loss": 4.3219,
|
|
"mean_token_accuracy": 0.24723988473415376,
|
|
"num_tokens": 14466615.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 4.798673152923584,
|
|
"epoch": 6.552301255230126,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00018977667794218894,
|
|
"loss": 4.1275,
|
|
"mean_token_accuracy": 0.26957189440727236,
|
|
"num_tokens": 14476684.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 4.812650918960571,
|
|
"epoch": 6.556485355648536,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001894780484587328,
|
|
"loss": 4.1517,
|
|
"mean_token_accuracy": 0.2675973907113075,
|
|
"num_tokens": 14485364.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 4.823258686065674,
|
|
"epoch": 6.560669456066946,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00018917959496570318,
|
|
"loss": 4.2298,
|
|
"mean_token_accuracy": 0.26263212263584135,
|
|
"num_tokens": 14495099.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 4.843588352203369,
|
|
"epoch": 6.564853556485356,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001888813180772694,
|
|
"loss": 4.1729,
|
|
"mean_token_accuracy": 0.26162929087877274,
|
|
"num_tokens": 14504051.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 4.851773405075074,
|
|
"epoch": 6.569037656903766,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018858321840723724,
|
|
"loss": 4.1698,
|
|
"mean_token_accuracy": 0.2574668675661087,
|
|
"num_tokens": 14513517.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 4.774352884292602,
|
|
"epoch": 6.573221757322176,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018828529656904802,
|
|
"loss": 4.2252,
|
|
"mean_token_accuracy": 0.25952457934617995,
|
|
"num_tokens": 14522504.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 4.850616216659546,
|
|
"epoch": 6.577405857740586,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018798755317577691,
|
|
"loss": 4.2266,
|
|
"mean_token_accuracy": 0.26339261829853056,
|
|
"num_tokens": 14532215.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 4.754785108566284,
|
|
"epoch": 6.581589958158996,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000187689988840132,
|
|
"loss": 4.0541,
|
|
"mean_token_accuracy": 0.27382495552301406,
|
|
"num_tokens": 14540489.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 4.788809728622437,
|
|
"epoch": 6.585774058577406,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001873926041744528,
|
|
"loss": 4.194,
|
|
"mean_token_accuracy": 0.2656221479177475,
|
|
"num_tokens": 14550476.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 4.83154616355896,
|
|
"epoch": 6.589958158995816,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00018709539979070922,
|
|
"loss": 4.1365,
|
|
"mean_token_accuracy": 0.26585482656955717,
|
|
"num_tokens": 14559349.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 4.79423508644104,
|
|
"epoch": 6.594142259414226,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00018679837630049994,
|
|
"loss": 4.1981,
|
|
"mean_token_accuracy": 0.262923526763916,
|
|
"num_tokens": 14568480.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 4.820297431945801,
|
|
"epoch": 6.598326359832636,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00018650153431505167,
|
|
"loss": 4.1854,
|
|
"mean_token_accuracy": 0.26532700955867766,
|
|
"num_tokens": 14578071.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 4.8040741920471195,
|
|
"epoch": 6.602510460251046,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001862048744452175,
|
|
"loss": 4.1675,
|
|
"mean_token_accuracy": 0.2688471347093582,
|
|
"num_tokens": 14586211.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 4.821338891983032,
|
|
"epoch": 6.606694560669456,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001859083973014757,
|
|
"loss": 4.2052,
|
|
"mean_token_accuracy": 0.2586241289973259,
|
|
"num_tokens": 14595206.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 4.80607271194458,
|
|
"epoch": 6.610878661087866,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018561210349392855,
|
|
"loss": 4.1752,
|
|
"mean_token_accuracy": 0.25919564813375473,
|
|
"num_tokens": 14604569.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 4.873177242279053,
|
|
"epoch": 6.615062761506276,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001853159936323011,
|
|
"loss": 4.1885,
|
|
"mean_token_accuracy": 0.26392625123262403,
|
|
"num_tokens": 14614040.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 4.842593955993652,
|
|
"epoch": 6.619246861924686,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00018502006832593992,
|
|
"loss": 4.1715,
|
|
"mean_token_accuracy": 0.26492412090301515,
|
|
"num_tokens": 14624461.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 4.7990423202514645,
|
|
"epoch": 6.623430962343097,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00018472432818381158,
|
|
"loss": 4.2159,
|
|
"mean_token_accuracy": 0.26432661414146424,
|
|
"num_tokens": 14633417.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 4.763973951339722,
|
|
"epoch": 6.627615062761507,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018442877381450175,
|
|
"loss": 4.1094,
|
|
"mean_token_accuracy": 0.26726812273263933,
|
|
"num_tokens": 14642387.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 4.774637031555176,
|
|
"epoch": 6.631799163179917,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001841334058262139,
|
|
"loss": 4.119,
|
|
"mean_token_accuracy": 0.25974465608596803,
|
|
"num_tokens": 14650613.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 4.799201917648316,
|
|
"epoch": 6.635983263598327,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018383822482676777,
|
|
"loss": 4.2293,
|
|
"mean_token_accuracy": 0.2684498310089111,
|
|
"num_tokens": 14659411.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 4.76613039970398,
|
|
"epoch": 6.640167364016737,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00018354323142359848,
|
|
"loss": 4.1538,
|
|
"mean_token_accuracy": 0.2661998808383942,
|
|
"num_tokens": 14668375.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 4.807853889465332,
|
|
"epoch": 6.644351464435147,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018324842622375497,
|
|
"loss": 4.1555,
|
|
"mean_token_accuracy": 0.26919307112693786,
|
|
"num_tokens": 14677608.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 4.85655665397644,
|
|
"epoch": 6.648535564853557,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00018295380983389905,
|
|
"loss": 4.2207,
|
|
"mean_token_accuracy": 0.2597134917974472,
|
|
"num_tokens": 14686646.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 4.751641035079956,
|
|
"epoch": 6.652719665271967,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001826593828603037,
|
|
"loss": 4.1485,
|
|
"mean_token_accuracy": 0.26141815036535265,
|
|
"num_tokens": 14695703.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 4.77737832069397,
|
|
"epoch": 6.656903765690377,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001823651459088524,
|
|
"loss": 4.1535,
|
|
"mean_token_accuracy": 0.2623979106545448,
|
|
"num_tokens": 14704493.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 4.842306900024414,
|
|
"epoch": 6.661087866108787,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00018207109958503747,
|
|
"loss": 4.2243,
|
|
"mean_token_accuracy": 0.2621868535876274,
|
|
"num_tokens": 14713119.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 4.831229782104492,
|
|
"epoch": 6.665271966527197,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000181777244493959,
|
|
"loss": 4.2052,
|
|
"mean_token_accuracy": 0.2579893946647644,
|
|
"num_tokens": 14722360.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 4.841086721420288,
|
|
"epoch": 6.669456066945607,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001814835812403235,
|
|
"loss": 4.2221,
|
|
"mean_token_accuracy": 0.25582362711429596,
|
|
"num_tokens": 14732066.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 4.78739218711853,
|
|
"epoch": 6.673640167364017,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00018119011042844286,
|
|
"loss": 4.1522,
|
|
"mean_token_accuracy": 0.25642079263925555,
|
|
"num_tokens": 14740968.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 4.783497858047485,
|
|
"epoch": 6.677824267782427,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001808968326622325,
|
|
"loss": 4.1831,
|
|
"mean_token_accuracy": 0.2587954804301262,
|
|
"num_tokens": 14750574.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 4.7981781482696535,
|
|
"epoch": 6.682008368200837,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018060374854521116,
|
|
"loss": 4.1462,
|
|
"mean_token_accuracy": 0.26003337651491165,
|
|
"num_tokens": 14760064.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 4.781484365463257,
|
|
"epoch": 6.686192468619247,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018031085868049884,
|
|
"loss": 4.224,
|
|
"mean_token_accuracy": 0.2596814677119255,
|
|
"num_tokens": 14769721.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 4.855734777450562,
|
|
"epoch": 6.690376569037657,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001800181636708157,
|
|
"loss": 4.2227,
|
|
"mean_token_accuracy": 0.2566405892372131,
|
|
"num_tokens": 14778895.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 4.869035100936889,
|
|
"epoch": 6.694560669456067,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017972566411848112,
|
|
"loss": 4.1985,
|
|
"mean_token_accuracy": 0.2642292663455009,
|
|
"num_tokens": 14787541.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 6.694560669456067,
|
|
"eval_entropy": 4.710603593073496,
|
|
"eval_loss": 5.263913154602051,
|
|
"eval_mean_token_accuracy": 0.2071953921413129,
|
|
"eval_num_tokens": 14787541.0,
|
|
"eval_runtime": 2.9357,
|
|
"eval_samples_per_second": 1330.852,
|
|
"eval_steps_per_second": 166.569,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 4.8522069454193115,
|
|
"epoch": 6.698744769874477,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000179433360625412,
|
|
"loss": 4.1862,
|
|
"mean_token_accuracy": 0.25991455018520354,
|
|
"num_tokens": 14795669.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 4.793842220306397,
|
|
"epoch": 6.702928870292887,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00017914125379312208,
|
|
"loss": 4.1606,
|
|
"mean_token_accuracy": 0.2636502906680107,
|
|
"num_tokens": 14804587.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 4.792535495758057,
|
|
"epoch": 6.707112970711297,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00017884934422272026,
|
|
"loss": 4.1576,
|
|
"mean_token_accuracy": 0.2647362411022186,
|
|
"num_tokens": 14813746.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 4.802059173583984,
|
|
"epoch": 6.711297071129707,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017855763251490924,
|
|
"loss": 4.2195,
|
|
"mean_token_accuracy": 0.2548334300518036,
|
|
"num_tokens": 14822906.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 4.816257953643799,
|
|
"epoch": 6.7154811715481175,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00017826611926998514,
|
|
"loss": 4.1913,
|
|
"mean_token_accuracy": 0.26253821402788163,
|
|
"num_tokens": 14832212.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 4.851281452178955,
|
|
"epoch": 6.7196652719665275,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001779748050878352,
|
|
"loss": 4.1849,
|
|
"mean_token_accuracy": 0.2574333190917969,
|
|
"num_tokens": 14841381.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 4.740643119812011,
|
|
"epoch": 6.7238493723849375,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017768369056793727,
|
|
"loss": 4.0682,
|
|
"mean_token_accuracy": 0.26828906387090684,
|
|
"num_tokens": 14850090.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 4.789907503128052,
|
|
"epoch": 6.7280334728033475,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000177392776309358,
|
|
"loss": 4.1827,
|
|
"mean_token_accuracy": 0.261622779071331,
|
|
"num_tokens": 14858885.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 4.7556405544281,
|
|
"epoch": 6.7322175732217575,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017710206291075248,
|
|
"loss": 4.1639,
|
|
"mean_token_accuracy": 0.2618906065821648,
|
|
"num_tokens": 14868173.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 4.779895448684693,
|
|
"epoch": 6.7364016736401675,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017681155097036194,
|
|
"loss": 4.1714,
|
|
"mean_token_accuracy": 0.26035045832395554,
|
|
"num_tokens": 14877164.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 4.806253242492676,
|
|
"epoch": 6.7405857740585775,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001765212410860132,
|
|
"loss": 4.194,
|
|
"mean_token_accuracy": 0.26665230840444565,
|
|
"num_tokens": 14886285.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 4.73868579864502,
|
|
"epoch": 6.7447698744769875,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017623113385511753,
|
|
"loss": 4.1752,
|
|
"mean_token_accuracy": 0.2541866317391396,
|
|
"num_tokens": 14895079.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 4.791785669326782,
|
|
"epoch": 6.7489539748953975,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001759412298746687,
|
|
"loss": 4.1776,
|
|
"mean_token_accuracy": 0.2583383351564407,
|
|
"num_tokens": 14904418.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 4.76855878829956,
|
|
"epoch": 6.7531380753138075,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017565152974124276,
|
|
"loss": 4.2276,
|
|
"mean_token_accuracy": 0.24904851615428925,
|
|
"num_tokens": 14913621.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 4.758419942855835,
|
|
"epoch": 6.7573221757322175,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001753620340509959,
|
|
"loss": 4.1399,
|
|
"mean_token_accuracy": 0.26774251610040667,
|
|
"num_tokens": 14921621.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 4.738441324234008,
|
|
"epoch": 6.7615062761506275,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017507274339966368,
|
|
"loss": 4.1174,
|
|
"mean_token_accuracy": 0.26734609007835386,
|
|
"num_tokens": 14930732.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 4.773867845535278,
|
|
"epoch": 6.7656903765690375,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017478365838255978,
|
|
"loss": 4.2057,
|
|
"mean_token_accuracy": 0.2595567092299461,
|
|
"num_tokens": 14939626.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 4.850559568405151,
|
|
"epoch": 6.7698744769874475,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001744947795945746,
|
|
"loss": 4.1797,
|
|
"mean_token_accuracy": 0.26682859361171724,
|
|
"num_tokens": 14948718.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 4.716847133636475,
|
|
"epoch": 6.7740585774058575,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001742061076301744,
|
|
"loss": 4.0293,
|
|
"mean_token_accuracy": 0.2742579594254494,
|
|
"num_tokens": 14957703.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 4.74285397529602,
|
|
"epoch": 6.7782426778242675,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017391764308339957,
|
|
"loss": 4.1386,
|
|
"mean_token_accuracy": 0.2647492527961731,
|
|
"num_tokens": 14966693.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 4.819277715682984,
|
|
"epoch": 6.7824267782426775,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00017362938654786383,
|
|
"loss": 4.2357,
|
|
"mean_token_accuracy": 0.2602304756641388,
|
|
"num_tokens": 14977267.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 4.878616571426392,
|
|
"epoch": 6.786610878661088,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00017334133861675278,
|
|
"loss": 4.2166,
|
|
"mean_token_accuracy": 0.25429029017686844,
|
|
"num_tokens": 14985648.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 4.7965178966522215,
|
|
"epoch": 6.790794979079498,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001730534998828226,
|
|
"loss": 4.1874,
|
|
"mean_token_accuracy": 0.2608168929815292,
|
|
"num_tokens": 14994915.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 4.79038290977478,
|
|
"epoch": 6.794979079497908,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001727658709383993,
|
|
"loss": 4.1492,
|
|
"mean_token_accuracy": 0.26819644421339034,
|
|
"num_tokens": 15004288.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 4.84480242729187,
|
|
"epoch": 6.799163179916318,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017247845237537693,
|
|
"loss": 4.2125,
|
|
"mean_token_accuracy": 0.2631848469376564,
|
|
"num_tokens": 15012970.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 4.787280416488647,
|
|
"epoch": 6.803347280334728,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017219124478521676,
|
|
"loss": 4.2445,
|
|
"mean_token_accuracy": 0.24735164344310762,
|
|
"num_tokens": 15022989.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 4.866213035583496,
|
|
"epoch": 6.807531380753138,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017190424875894567,
|
|
"loss": 4.2171,
|
|
"mean_token_accuracy": 0.26715106070041655,
|
|
"num_tokens": 15032665.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 4.832035255432129,
|
|
"epoch": 6.811715481171548,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017161746488715546,
|
|
"loss": 4.1746,
|
|
"mean_token_accuracy": 0.2628034114837646,
|
|
"num_tokens": 15041109.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 4.747332620620727,
|
|
"epoch": 6.815899581589958,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017133089376000142,
|
|
"loss": 4.1083,
|
|
"mean_token_accuracy": 0.2738641083240509,
|
|
"num_tokens": 15050080.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 4.7706533908844,
|
|
"epoch": 6.820083682008368,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001710445359672005,
|
|
"loss": 4.1724,
|
|
"mean_token_accuracy": 0.26646038740873335,
|
|
"num_tokens": 15058429.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 4.814546775817871,
|
|
"epoch": 6.824267782426778,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017075839209803117,
|
|
"loss": 4.1268,
|
|
"mean_token_accuracy": 0.2586261138319969,
|
|
"num_tokens": 15068652.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 4.815828323364258,
|
|
"epoch": 6.828451882845188,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017047246274133144,
|
|
"loss": 4.1321,
|
|
"mean_token_accuracy": 0.2673907458782196,
|
|
"num_tokens": 15077643.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 4.815294170379639,
|
|
"epoch": 6.832635983263598,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001701867484854981,
|
|
"loss": 4.1859,
|
|
"mean_token_accuracy": 0.2604037031531334,
|
|
"num_tokens": 15087471.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 4.882863903045655,
|
|
"epoch": 6.836820083682008,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016990124991848497,
|
|
"loss": 4.3044,
|
|
"mean_token_accuracy": 0.25214432626962663,
|
|
"num_tokens": 15096357.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 4.799471616744995,
|
|
"epoch": 6.841004184100418,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001696159676278023,
|
|
"loss": 4.171,
|
|
"mean_token_accuracy": 0.2579039499163628,
|
|
"num_tokens": 15105424.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 4.805128765106201,
|
|
"epoch": 6.845188284518828,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001693309022005153,
|
|
"loss": 4.188,
|
|
"mean_token_accuracy": 0.2650418609380722,
|
|
"num_tokens": 15115641.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 4.801178979873657,
|
|
"epoch": 6.849372384937238,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016904605422324255,
|
|
"loss": 4.1596,
|
|
"mean_token_accuracy": 0.26416143476963044,
|
|
"num_tokens": 15124765.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 4.719380807876587,
|
|
"epoch": 6.853556485355648,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016876142428215552,
|
|
"loss": 4.1233,
|
|
"mean_token_accuracy": 0.27009942978620527,
|
|
"num_tokens": 15133591.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 4.798724508285522,
|
|
"epoch": 6.857740585774058,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001684770129629768,
|
|
"loss": 4.2211,
|
|
"mean_token_accuracy": 0.25618866086006165,
|
|
"num_tokens": 15143291.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 4.865075778961182,
|
|
"epoch": 6.861924686192468,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016819282085097918,
|
|
"loss": 4.2729,
|
|
"mean_token_accuracy": 0.25338044911623003,
|
|
"num_tokens": 15152806.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 4.844187784194946,
|
|
"epoch": 6.866108786610878,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016790884853098441,
|
|
"loss": 4.1818,
|
|
"mean_token_accuracy": 0.25877276957035067,
|
|
"num_tokens": 15163116.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 4.796574592590332,
|
|
"epoch": 6.870292887029288,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001676250965873619,
|
|
"loss": 4.1787,
|
|
"mean_token_accuracy": 0.2648740574717522,
|
|
"num_tokens": 15172916.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 4.816861724853515,
|
|
"epoch": 6.874476987447698,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016734156560402742,
|
|
"loss": 4.1751,
|
|
"mean_token_accuracy": 0.2676452279090881,
|
|
"num_tokens": 15181715.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 4.7736711502075195,
|
|
"epoch": 6.878661087866108,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001670582561644421,
|
|
"loss": 4.1438,
|
|
"mean_token_accuracy": 0.27048349380493164,
|
|
"num_tokens": 15191290.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 4.782187557220459,
|
|
"epoch": 6.882845188284519,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016677516885161127,
|
|
"loss": 4.1449,
|
|
"mean_token_accuracy": 0.2692710056900978,
|
|
"num_tokens": 15200614.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 4.922882699966431,
|
|
"epoch": 6.887029288702929,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001664923042480832,
|
|
"loss": 4.2873,
|
|
"mean_token_accuracy": 0.24913981705904006,
|
|
"num_tokens": 15210793.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 4.779243755340576,
|
|
"epoch": 6.891213389121339,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00016620966293594763,
|
|
"loss": 4.119,
|
|
"mean_token_accuracy": 0.2642215698957443,
|
|
"num_tokens": 15219968.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 4.789557790756225,
|
|
"epoch": 6.895397489539749,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016592724549683514,
|
|
"loss": 4.1657,
|
|
"mean_token_accuracy": 0.26349806040525436,
|
|
"num_tokens": 15229434.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 4.7492125034332275,
|
|
"epoch": 6.899581589958159,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001656450525119152,
|
|
"loss": 4.1066,
|
|
"mean_token_accuracy": 0.2694716051220894,
|
|
"num_tokens": 15238784.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 4.832731819152832,
|
|
"epoch": 6.903765690376569,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016536308456189568,
|
|
"loss": 4.2245,
|
|
"mean_token_accuracy": 0.2582759574055672,
|
|
"num_tokens": 15247926.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 4.88155460357666,
|
|
"epoch": 6.907949790794979,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016508134222702133,
|
|
"loss": 4.2372,
|
|
"mean_token_accuracy": 0.25370642095804213,
|
|
"num_tokens": 15257158.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 4.8245032787322994,
|
|
"epoch": 6.912133891213389,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00016479982608707245,
|
|
"loss": 4.2303,
|
|
"mean_token_accuracy": 0.26123583763837815,
|
|
"num_tokens": 15266639.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 4.8160865783691404,
|
|
"epoch": 6.916317991631799,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016451853672136418,
|
|
"loss": 4.1583,
|
|
"mean_token_accuracy": 0.2597955584526062,
|
|
"num_tokens": 15275446.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 4.815404605865479,
|
|
"epoch": 6.920502092050209,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001642374747087446,
|
|
"loss": 4.2414,
|
|
"mean_token_accuracy": 0.2527800932526588,
|
|
"num_tokens": 15285011.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 4.867608594894409,
|
|
"epoch": 6.924686192468619,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016395664062759442,
|
|
"loss": 4.3506,
|
|
"mean_token_accuracy": 0.24859413057565688,
|
|
"num_tokens": 15295066.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 4.895462512969971,
|
|
"epoch": 6.928870292887029,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016367603505582483,
|
|
"loss": 4.1814,
|
|
"mean_token_accuracy": 0.2635229960083961,
|
|
"num_tokens": 15304660.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 4.788323831558228,
|
|
"epoch": 6.933054393305439,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001633956585708769,
|
|
"loss": 4.1715,
|
|
"mean_token_accuracy": 0.2642997071146965,
|
|
"num_tokens": 15314376.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 4.707601881027221,
|
|
"epoch": 6.937238493723849,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016311551174972062,
|
|
"loss": 4.1135,
|
|
"mean_token_accuracy": 0.27275072038173676,
|
|
"num_tokens": 15323414.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 4.7910772323608395,
|
|
"epoch": 6.941422594142259,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016283559516885285,
|
|
"loss": 4.1999,
|
|
"mean_token_accuracy": 0.2594840586185455,
|
|
"num_tokens": 15332266.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 4.836405944824219,
|
|
"epoch": 6.945606694560669,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016255590940429717,
|
|
"loss": 4.1768,
|
|
"mean_token_accuracy": 0.26194940507411957,
|
|
"num_tokens": 15342238.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 4.886379098892212,
|
|
"epoch": 6.949790794979079,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001622764550316016,
|
|
"loss": 4.2575,
|
|
"mean_token_accuracy": 0.2545137256383896,
|
|
"num_tokens": 15351565.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 4.831608819961548,
|
|
"epoch": 6.95397489539749,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001619972326258386,
|
|
"loss": 4.2116,
|
|
"mean_token_accuracy": 0.25686272978782654,
|
|
"num_tokens": 15361064.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 4.759947776794434,
|
|
"epoch": 6.9581589958159,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016171824276160288,
|
|
"loss": 4.2276,
|
|
"mean_token_accuracy": 0.25028758943080903,
|
|
"num_tokens": 15371152.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 4.827062892913818,
|
|
"epoch": 6.96234309623431,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016143948601301064,
|
|
"loss": 4.2174,
|
|
"mean_token_accuracy": 0.25719341039657595,
|
|
"num_tokens": 15381546.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 4.7676092147827145,
|
|
"epoch": 6.96652719665272,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001611609629536986,
|
|
"loss": 4.1541,
|
|
"mean_token_accuracy": 0.26290807127952576,
|
|
"num_tokens": 15390581.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 4.791965675354004,
|
|
"epoch": 6.97071129707113,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001608826741568223,
|
|
"loss": 4.1666,
|
|
"mean_token_accuracy": 0.2641488239169121,
|
|
"num_tokens": 15400035.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 4.840515851974487,
|
|
"epoch": 6.97489539748954,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001606046201950555,
|
|
"loss": 4.2295,
|
|
"mean_token_accuracy": 0.25318553745746614,
|
|
"num_tokens": 15409199.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 4.755623435974121,
|
|
"epoch": 6.97907949790795,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016032680164058832,
|
|
"loss": 4.1013,
|
|
"mean_token_accuracy": 0.27227047830820084,
|
|
"num_tokens": 15417851.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 4.807187604904175,
|
|
"epoch": 6.98326359832636,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016004921906512699,
|
|
"loss": 4.1379,
|
|
"mean_token_accuracy": 0.26177673488855363,
|
|
"num_tokens": 15427072.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 4.837101936340332,
|
|
"epoch": 6.98744769874477,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001597718730398916,
|
|
"loss": 4.2363,
|
|
"mean_token_accuracy": 0.2522675484418869,
|
|
"num_tokens": 15437486.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 4.809011507034302,
|
|
"epoch": 6.99163179916318,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00015949476413561564,
|
|
"loss": 4.1869,
|
|
"mean_token_accuracy": 0.26844466030597686,
|
|
"num_tokens": 15447650.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 4.735010957717895,
|
|
"epoch": 6.99581589958159,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015921789292254484,
|
|
"loss": 4.1062,
|
|
"mean_token_accuracy": 0.2726203739643097,
|
|
"num_tokens": 15456650.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 4.746040153503418,
|
|
"epoch": 7.0,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00015894125997043551,
|
|
"loss": 4.1419,
|
|
"mean_token_accuracy": 0.2671010971069336,
|
|
"num_tokens": 15465254.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 4.777007722854615,
|
|
"epoch": 7.00418410041841,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00015866486584855397,
|
|
"loss": 4.0216,
|
|
"mean_token_accuracy": 0.2796642929315567,
|
|
"num_tokens": 15474018.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 4.839642810821533,
|
|
"epoch": 7.00836820083682,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00015838871112567465,
|
|
"loss": 4.1127,
|
|
"mean_token_accuracy": 0.26712811291217803,
|
|
"num_tokens": 15483871.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 4.7676016330719,
|
|
"epoch": 7.01255230125523,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001581127963700798,
|
|
"loss": 4.0167,
|
|
"mean_token_accuracy": 0.2798952579498291,
|
|
"num_tokens": 15492996.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 4.855391311645508,
|
|
"epoch": 7.01673640167364,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001578371221495575,
|
|
"loss": 4.0932,
|
|
"mean_token_accuracy": 0.26719109117984774,
|
|
"num_tokens": 15501703.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 4.829797887802124,
|
|
"epoch": 7.02092050209205,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001575616890314009,
|
|
"loss": 4.0566,
|
|
"mean_token_accuracy": 0.27974404841661454,
|
|
"num_tokens": 15510110.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 4.733310699462891,
|
|
"epoch": 7.02510460251046,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015728649758240727,
|
|
"loss": 3.9923,
|
|
"mean_token_accuracy": 0.28292536437511445,
|
|
"num_tokens": 15518352.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 4.730865478515625,
|
|
"epoch": 7.02928870292887,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001570115483688762,
|
|
"loss": 3.9904,
|
|
"mean_token_accuracy": 0.28078057169914244,
|
|
"num_tokens": 15527133.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 4.720526790618896,
|
|
"epoch": 7.03347280334728,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001567368419566091,
|
|
"loss": 3.9812,
|
|
"mean_token_accuracy": 0.2831581518054008,
|
|
"num_tokens": 15536326.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 4.868599367141724,
|
|
"epoch": 7.03765690376569,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00015646237891090749,
|
|
"loss": 4.1019,
|
|
"mean_token_accuracy": 0.2634491354227066,
|
|
"num_tokens": 15546408.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 4.803051710128784,
|
|
"epoch": 7.0418410041841,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001561881597965722,
|
|
"loss": 4.0931,
|
|
"mean_token_accuracy": 0.2684536024928093,
|
|
"num_tokens": 15554988.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 4.777579641342163,
|
|
"epoch": 7.046025104602511,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015591418517790217,
|
|
"loss": 3.9511,
|
|
"mean_token_accuracy": 0.29210805892944336,
|
|
"num_tokens": 15564127.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 4.823932409286499,
|
|
"epoch": 7.050209205020921,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000155640455618693,
|
|
"loss": 4.0717,
|
|
"mean_token_accuracy": 0.2737370550632477,
|
|
"num_tokens": 15573658.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 4.865050649642944,
|
|
"epoch": 7.054393305439331,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00015536697168223626,
|
|
"loss": 4.1043,
|
|
"mean_token_accuracy": 0.270908522605896,
|
|
"num_tokens": 15583728.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 4.7271098613739015,
|
|
"epoch": 7.058577405857741,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001550937339313177,
|
|
"loss": 4.016,
|
|
"mean_token_accuracy": 0.2800894945859909,
|
|
"num_tokens": 15593335.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 4.72858943939209,
|
|
"epoch": 7.062761506276151,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015482074292821687,
|
|
"loss": 3.9393,
|
|
"mean_token_accuracy": 0.2810845822095871,
|
|
"num_tokens": 15603133.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 4.781504440307617,
|
|
"epoch": 7.066945606694561,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001545479992347055,
|
|
"loss": 4.0563,
|
|
"mean_token_accuracy": 0.2764288201928139,
|
|
"num_tokens": 15612509.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 4.7478556632995605,
|
|
"epoch": 7.071129707112971,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00015427550341204585,
|
|
"loss": 4.0324,
|
|
"mean_token_accuracy": 0.28037301898002626,
|
|
"num_tokens": 15622160.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 4.885111093521118,
|
|
"epoch": 7.075313807531381,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00015400325602099086,
|
|
"loss": 4.1549,
|
|
"mean_token_accuracy": 0.26087299883365633,
|
|
"num_tokens": 15631268.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 4.884063577651977,
|
|
"epoch": 7.079497907949791,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00015373125762178168,
|
|
"loss": 4.1071,
|
|
"mean_token_accuracy": 0.2716177314519882,
|
|
"num_tokens": 15640153.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 4.774041652679443,
|
|
"epoch": 7.083682008368201,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015345950877414732,
|
|
"loss": 3.9508,
|
|
"mean_token_accuracy": 0.2866012901067734,
|
|
"num_tokens": 15647752.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 4.857289028167725,
|
|
"epoch": 7.087866108786611,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015318801003730344,
|
|
"loss": 4.1006,
|
|
"mean_token_accuracy": 0.2691643550992012,
|
|
"num_tokens": 15656822.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 4.695055437088013,
|
|
"epoch": 7.092050209205021,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015291676196995047,
|
|
"loss": 3.9693,
|
|
"mean_token_accuracy": 0.2816358804702759,
|
|
"num_tokens": 15665045.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 4.763889217376709,
|
|
"epoch": 7.096234309623431,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00015264576513027367,
|
|
"loss": 4.0505,
|
|
"mean_token_accuracy": 0.26959392577409746,
|
|
"num_tokens": 15674206.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 4.879012250900269,
|
|
"epoch": 7.100418410041841,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001523750200759406,
|
|
"loss": 4.128,
|
|
"mean_token_accuracy": 0.26351071447134017,
|
|
"num_tokens": 15683075.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 4.7556334972381595,
|
|
"epoch": 7.104602510460251,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001521045273641013,
|
|
"loss": 3.9479,
|
|
"mean_token_accuracy": 0.29044548571109774,
|
|
"num_tokens": 15691698.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 4.8200925350189205,
|
|
"epoch": 7.108786610878661,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015183428755138613,
|
|
"loss": 4.1377,
|
|
"mean_token_accuracy": 0.2682234689593315,
|
|
"num_tokens": 15701927.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 4.801332187652588,
|
|
"epoch": 7.112970711297071,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015156430119390524,
|
|
"loss": 4.0782,
|
|
"mean_token_accuracy": 0.2758529409766197,
|
|
"num_tokens": 15710813.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 7.112970711297071,
|
|
"eval_entropy": 4.689213446556669,
|
|
"eval_loss": 5.268126964569092,
|
|
"eval_mean_token_accuracy": 0.2067725897170527,
|
|
"eval_num_tokens": 15710813.0,
|
|
"eval_runtime": 2.7315,
|
|
"eval_samples_per_second": 1430.352,
|
|
"eval_steps_per_second": 179.023,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 4.879066085815429,
|
|
"epoch": 7.117154811715481,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015129456884724723,
|
|
"loss": 4.083,
|
|
"mean_token_accuracy": 0.2725109845399857,
|
|
"num_tokens": 15720559.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 4.787436580657959,
|
|
"epoch": 7.121338912133891,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00015102509106647782,
|
|
"loss": 4.0524,
|
|
"mean_token_accuracy": 0.27712114751338957,
|
|
"num_tokens": 15730649.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 4.793636417388916,
|
|
"epoch": 7.125523012552302,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015075586840613902,
|
|
"loss": 4.0075,
|
|
"mean_token_accuracy": 0.28519228994846346,
|
|
"num_tokens": 15739619.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 4.815764284133911,
|
|
"epoch": 7.129707112970712,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00015048690142024763,
|
|
"loss": 4.087,
|
|
"mean_token_accuracy": 0.2750828206539154,
|
|
"num_tokens": 15748457.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 4.760766792297363,
|
|
"epoch": 7.133891213389122,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00015021819066229452,
|
|
"loss": 4.0166,
|
|
"mean_token_accuracy": 0.2725695803761482,
|
|
"num_tokens": 15758527.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 4.7959856510162355,
|
|
"epoch": 7.138075313807532,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00014994973668524343,
|
|
"loss": 4.1279,
|
|
"mean_token_accuracy": 0.2741900339722633,
|
|
"num_tokens": 15767980.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 4.722530269622803,
|
|
"epoch": 7.142259414225942,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014968154004152922,
|
|
"loss": 4.0268,
|
|
"mean_token_accuracy": 0.28455786406993866,
|
|
"num_tokens": 15777323.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 4.748971557617187,
|
|
"epoch": 7.146443514644352,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00014941360128305765,
|
|
"loss": 4.0335,
|
|
"mean_token_accuracy": 0.2875641018152237,
|
|
"num_tokens": 15787589.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 4.808362579345703,
|
|
"epoch": 7.150627615062762,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00014914592096120348,
|
|
"loss": 4.0673,
|
|
"mean_token_accuracy": 0.27979826033115385,
|
|
"num_tokens": 15797248.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 4.799026298522949,
|
|
"epoch": 7.154811715481172,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00014887849962680978,
|
|
"loss": 4.1162,
|
|
"mean_token_accuracy": 0.25993827283382415,
|
|
"num_tokens": 15806389.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 4.8492834091186525,
|
|
"epoch": 7.158995815899582,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014861133783018678,
|
|
"loss": 4.055,
|
|
"mean_token_accuracy": 0.2800892874598503,
|
|
"num_tokens": 15814719.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 4.861292123794556,
|
|
"epoch": 7.163179916317992,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00014834443612111035,
|
|
"loss": 4.121,
|
|
"mean_token_accuracy": 0.2678546607494354,
|
|
"num_tokens": 15823751.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 4.834306383132935,
|
|
"epoch": 7.167364016736402,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014807779504882147,
|
|
"loss": 4.1093,
|
|
"mean_token_accuracy": 0.2699203848838806,
|
|
"num_tokens": 15832717.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 4.750302505493164,
|
|
"epoch": 7.171548117154812,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00014781141516202439,
|
|
"loss": 4.0435,
|
|
"mean_token_accuracy": 0.2793175846338272,
|
|
"num_tokens": 15841494.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 4.80815110206604,
|
|
"epoch": 7.175732217573222,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001475452970088863,
|
|
"loss": 4.1137,
|
|
"mean_token_accuracy": 0.26973645836114885,
|
|
"num_tokens": 15851619.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 4.812477445602417,
|
|
"epoch": 7.179916317991632,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001472794411370355,
|
|
"loss": 4.1073,
|
|
"mean_token_accuracy": 0.2679521471261978,
|
|
"num_tokens": 15861563.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 4.790307235717774,
|
|
"epoch": 7.184100418410042,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00014701384809356053,
|
|
"loss": 4.0817,
|
|
"mean_token_accuracy": 0.2772174686193466,
|
|
"num_tokens": 15870554.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 4.793127489089966,
|
|
"epoch": 7.188284518828452,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00014674851842500931,
|
|
"loss": 4.0297,
|
|
"mean_token_accuracy": 0.27817195653915405,
|
|
"num_tokens": 15879679.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 4.7401844501495365,
|
|
"epoch": 7.192468619246862,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001464834526773875,
|
|
"loss": 4.0316,
|
|
"mean_token_accuracy": 0.2712982475757599,
|
|
"num_tokens": 15888533.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 4.795364141464233,
|
|
"epoch": 7.196652719665272,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00014621865139615795,
|
|
"loss": 4.0338,
|
|
"mean_token_accuracy": 0.2775416851043701,
|
|
"num_tokens": 15896888.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 4.8199094295501705,
|
|
"epoch": 7.200836820083682,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00014595411512623895,
|
|
"loss": 4.1044,
|
|
"mean_token_accuracy": 0.2735301166772842,
|
|
"num_tokens": 15906843.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 4.682242012023925,
|
|
"epoch": 7.205020920502092,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014568984441200384,
|
|
"loss": 3.9943,
|
|
"mean_token_accuracy": 0.2800831511616707,
|
|
"num_tokens": 15915333.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 4.839969110488892,
|
|
"epoch": 7.209205020920502,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014542583979727908,
|
|
"loss": 4.1156,
|
|
"mean_token_accuracy": 0.26568340212106706,
|
|
"num_tokens": 15924187.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 4.827913904190064,
|
|
"epoch": 7.2133891213389125,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014516210182534373,
|
|
"loss": 4.0755,
|
|
"mean_token_accuracy": 0.2791578248143196,
|
|
"num_tokens": 15932763.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 4.811709451675415,
|
|
"epoch": 7.2175732217573225,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00014489863103892825,
|
|
"loss": 4.0196,
|
|
"mean_token_accuracy": 0.2761692345142365,
|
|
"num_tokens": 15942272.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 4.772731113433838,
|
|
"epoch": 7.2217573221757325,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000144635427980213,
|
|
"loss": 4.1174,
|
|
"mean_token_accuracy": 0.2636865586042404,
|
|
"num_tokens": 15952099.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 4.747719621658325,
|
|
"epoch": 7.2259414225941425,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00014437249319082773,
|
|
"loss": 4.0188,
|
|
"mean_token_accuracy": 0.27362363785505295,
|
|
"num_tokens": 15961206.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 4.761302375793457,
|
|
"epoch": 7.2301255230125525,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014410982721184984,
|
|
"loss": 4.0518,
|
|
"mean_token_accuracy": 0.27920086234807967,
|
|
"num_tokens": 15970694.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 4.790590810775757,
|
|
"epoch": 7.2343096234309625,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00014384743058380373,
|
|
"loss": 4.0315,
|
|
"mean_token_accuracy": 0.2819835513830185,
|
|
"num_tokens": 15979271.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 4.823791790008545,
|
|
"epoch": 7.2384937238493725,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00014358530384665953,
|
|
"loss": 4.0623,
|
|
"mean_token_accuracy": 0.27845172882080077,
|
|
"num_tokens": 15988124.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 4.855895090103149,
|
|
"epoch": 7.2426778242677825,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014332344753983174,
|
|
"loss": 4.0948,
|
|
"mean_token_accuracy": 0.27729480266571044,
|
|
"num_tokens": 15998912.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 4.818265104293824,
|
|
"epoch": 7.2468619246861925,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014306186220217873,
|
|
"loss": 4.0602,
|
|
"mean_token_accuracy": 0.27233550846576693,
|
|
"num_tokens": 16008295.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 4.785058689117432,
|
|
"epoch": 7.2510460251046025,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00014280054837200094,
|
|
"loss": 4.0483,
|
|
"mean_token_accuracy": 0.2869140386581421,
|
|
"num_tokens": 16017120.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 4.860040712356567,
|
|
"epoch": 7.2552301255230125,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00014253950658704034,
|
|
"loss": 4.1433,
|
|
"mean_token_accuracy": 0.2646480366587639,
|
|
"num_tokens": 16026630.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 4.798429536819458,
|
|
"epoch": 7.2594142259414225,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014227873738447883,
|
|
"loss": 4.0689,
|
|
"mean_token_accuracy": 0.27266145348548887,
|
|
"num_tokens": 16034897.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 4.713422393798828,
|
|
"epoch": 7.2635983263598325,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00014201824130093764,
|
|
"loss": 4.0127,
|
|
"mean_token_accuracy": 0.2751693338155746,
|
|
"num_tokens": 16044387.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 4.743470668792725,
|
|
"epoch": 7.2677824267782425,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00014175801887247576,
|
|
"loss": 4.0374,
|
|
"mean_token_accuracy": 0.2751784324645996,
|
|
"num_tokens": 16053659.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 4.786970233917236,
|
|
"epoch": 7.2719665271966525,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014149807063458896,
|
|
"loss": 4.0456,
|
|
"mean_token_accuracy": 0.2806077808141708,
|
|
"num_tokens": 16062997.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 4.83799295425415,
|
|
"epoch": 7.2761506276150625,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014123839712220921,
|
|
"loss": 4.058,
|
|
"mean_token_accuracy": 0.26649373173713686,
|
|
"num_tokens": 16071491.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 4.818638467788697,
|
|
"epoch": 7.2803347280334725,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001409789988697026,
|
|
"loss": 4.1307,
|
|
"mean_token_accuracy": 0.26776303052902223,
|
|
"num_tokens": 16081008.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 4.7656303405761715,
|
|
"epoch": 7.2845188284518825,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00014071987641086926,
|
|
"loss": 4.0313,
|
|
"mean_token_accuracy": 0.2798884600400925,
|
|
"num_tokens": 16090516.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 4.804595470428467,
|
|
"epoch": 7.2887029288702925,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001404610302789414,
|
|
"loss": 4.1445,
|
|
"mean_token_accuracy": 0.2720418944954872,
|
|
"num_tokens": 16100952.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 4.778432369232178,
|
|
"epoch": 7.292887029288703,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00014020246100658268,
|
|
"loss": 4.075,
|
|
"mean_token_accuracy": 0.27684093117713926,
|
|
"num_tokens": 16110493.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 4.7778892993927,
|
|
"epoch": 7.297071129707113,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013994416912588726,
|
|
"loss": 4.0449,
|
|
"mean_token_accuracy": 0.2782765358686447,
|
|
"num_tokens": 16119227.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 4.837534809112549,
|
|
"epoch": 7.301255230125523,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001396861551683781,
|
|
"loss": 4.1609,
|
|
"mean_token_accuracy": 0.2695921793580055,
|
|
"num_tokens": 16129268.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 4.822813558578491,
|
|
"epoch": 7.305439330543933,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001394284196650067,
|
|
"loss": 4.1,
|
|
"mean_token_accuracy": 0.2735602930188179,
|
|
"num_tokens": 16137523.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 4.8397377014160154,
|
|
"epoch": 7.309623430962343,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001391709631461511,
|
|
"loss": 4.1334,
|
|
"mean_token_accuracy": 0.26640370190143586,
|
|
"num_tokens": 16146535.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 4.80674376487732,
|
|
"epoch": 7.313807531380753,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00013891378614161538,
|
|
"loss": 4.0668,
|
|
"mean_token_accuracy": 0.27534746676683425,
|
|
"num_tokens": 16155892.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 4.756717824935913,
|
|
"epoch": 7.317991631799163,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00013865688918062875,
|
|
"loss": 4.049,
|
|
"mean_token_accuracy": 0.2673118829727173,
|
|
"num_tokens": 16164697.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 4.780541133880615,
|
|
"epoch": 7.322175732217573,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013840027279184353,
|
|
"loss": 4.0758,
|
|
"mean_token_accuracy": 0.27218465209007264,
|
|
"num_tokens": 16173387.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 4.777707862854004,
|
|
"epoch": 7.326359832635983,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013814393750333524,
|
|
"loss": 4.0192,
|
|
"mean_token_accuracy": 0.2833226680755615,
|
|
"num_tokens": 16182438.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 4.76727991104126,
|
|
"epoch": 7.330543933054393,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00013788788384260043,
|
|
"loss": 4.0042,
|
|
"mean_token_accuracy": 0.278095069527626,
|
|
"num_tokens": 16193149.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 4.807977867126465,
|
|
"epoch": 7.334728033472803,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013763211233655656,
|
|
"loss": 4.1403,
|
|
"mean_token_accuracy": 0.27345683574676516,
|
|
"num_tokens": 16202354.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 4.781715154647827,
|
|
"epoch": 7.338912133891213,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013737662351154028,
|
|
"loss": 4.082,
|
|
"mean_token_accuracy": 0.2668127715587616,
|
|
"num_tokens": 16210608.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 4.867360210418701,
|
|
"epoch": 7.343096234309623,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013712141789330642,
|
|
"loss": 4.1225,
|
|
"mean_token_accuracy": 0.2781470865011215,
|
|
"num_tokens": 16219544.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 4.802486848831177,
|
|
"epoch": 7.347280334728033,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013686649600702728,
|
|
"loss": 4.0336,
|
|
"mean_token_accuracy": 0.27602696120738984,
|
|
"num_tokens": 16228820.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 4.713297033309937,
|
|
"epoch": 7.351464435146443,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013661185837729085,
|
|
"loss": 3.9289,
|
|
"mean_token_accuracy": 0.2815149039030075,
|
|
"num_tokens": 16237751.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 4.7938422679901125,
|
|
"epoch": 7.355648535564853,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00013635750552810065,
|
|
"loss": 4.0555,
|
|
"mean_token_accuracy": 0.276930496096611,
|
|
"num_tokens": 16246944.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 4.816668844223022,
|
|
"epoch": 7.359832635983263,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00013610343798287376,
|
|
"loss": 4.1374,
|
|
"mean_token_accuracy": 0.2708712428808212,
|
|
"num_tokens": 16256565.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 4.751552724838257,
|
|
"epoch": 7.364016736401673,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013584965626444045,
|
|
"loss": 3.9853,
|
|
"mean_token_accuracy": 0.28116108775138854,
|
|
"num_tokens": 16264886.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 4.887779331207275,
|
|
"epoch": 7.368200836820083,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013559616089504285,
|
|
"loss": 4.1876,
|
|
"mean_token_accuracy": 0.25986612886190413,
|
|
"num_tokens": 16274039.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 4.758947944641113,
|
|
"epoch": 7.372384937238493,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013534295239633342,
|
|
"loss": 4.0774,
|
|
"mean_token_accuracy": 0.27543427646160124,
|
|
"num_tokens": 16283274.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 4.760553979873658,
|
|
"epoch": 7.376569037656903,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001350900312893747,
|
|
"loss": 4.0278,
|
|
"mean_token_accuracy": 0.2797372817993164,
|
|
"num_tokens": 16292048.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 4.778201103210449,
|
|
"epoch": 7.380753138075314,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00013483739809463753,
|
|
"loss": 4.0358,
|
|
"mean_token_accuracy": 0.27722379118204116,
|
|
"num_tokens": 16301910.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 4.787615251541138,
|
|
"epoch": 7.384937238493724,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013458505333200044,
|
|
"loss": 4.0461,
|
|
"mean_token_accuracy": 0.2790739819407463,
|
|
"num_tokens": 16311221.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 4.744237995147705,
|
|
"epoch": 7.389121338912134,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00013433299752074857,
|
|
"loss": 4.0067,
|
|
"mean_token_accuracy": 0.286495552957058,
|
|
"num_tokens": 16319530.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 4.715499973297119,
|
|
"epoch": 7.393305439330544,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000134081231179572,
|
|
"loss": 3.9607,
|
|
"mean_token_accuracy": 0.29102427661418917,
|
|
"num_tokens": 16327927.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 4.854242610931396,
|
|
"epoch": 7.397489539748954,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013382975482656556,
|
|
"loss": 4.1075,
|
|
"mean_token_accuracy": 0.2656175300478935,
|
|
"num_tokens": 16337551.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 4.737663316726684,
|
|
"epoch": 7.401673640167364,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001335785689792271,
|
|
"loss": 4.0479,
|
|
"mean_token_accuracy": 0.27616844326257706,
|
|
"num_tokens": 16345937.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 4.802147769927979,
|
|
"epoch": 7.405857740585774,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001333276741544566,
|
|
"loss": 4.126,
|
|
"mean_token_accuracy": 0.2673077270388603,
|
|
"num_tokens": 16355797.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 4.81158766746521,
|
|
"epoch": 7.410041841004184,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013307707086855546,
|
|
"loss": 4.0559,
|
|
"mean_token_accuracy": 0.27428028434515,
|
|
"num_tokens": 16364547.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 4.744426536560058,
|
|
"epoch": 7.414225941422594,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001328267596372248,
|
|
"loss": 3.9501,
|
|
"mean_token_accuracy": 0.29582202434539795,
|
|
"num_tokens": 16373506.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 4.79816370010376,
|
|
"epoch": 7.418410041841004,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000132576740975565,
|
|
"loss": 4.0311,
|
|
"mean_token_accuracy": 0.27937064319849014,
|
|
"num_tokens": 16383006.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 4.786335611343384,
|
|
"epoch": 7.422594142259414,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00013232701539807415,
|
|
"loss": 4.0991,
|
|
"mean_token_accuracy": 0.2728495329618454,
|
|
"num_tokens": 16392721.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 4.734829664230347,
|
|
"epoch": 7.426778242677824,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001320775834186475,
|
|
"loss": 4.0506,
|
|
"mean_token_accuracy": 0.2751553252339363,
|
|
"num_tokens": 16402404.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 4.831276893615723,
|
|
"epoch": 7.430962343096234,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013182844555057583,
|
|
"loss": 4.0932,
|
|
"mean_token_accuracy": 0.2695514112710953,
|
|
"num_tokens": 16411386.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 4.752012538909912,
|
|
"epoch": 7.435146443514644,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001315796023065448,
|
|
"loss": 4.0281,
|
|
"mean_token_accuracy": 0.2739667952060699,
|
|
"num_tokens": 16420407.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 4.747226428985596,
|
|
"epoch": 7.439330543933054,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00013133105419863394,
|
|
"loss": 3.9426,
|
|
"mean_token_accuracy": 0.29362151622772215,
|
|
"num_tokens": 16429197.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 4.820680713653564,
|
|
"epoch": 7.443514644351464,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013108280173831517,
|
|
"loss": 4.098,
|
|
"mean_token_accuracy": 0.27332728505134585,
|
|
"num_tokens": 16438144.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 4.7628106594085695,
|
|
"epoch": 7.447698744769874,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001308348454364522,
|
|
"loss": 4.1024,
|
|
"mean_token_accuracy": 0.2739241749048233,
|
|
"num_tokens": 16447861.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 4.719743776321411,
|
|
"epoch": 7.451882845188284,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013058718580329917,
|
|
"loss": 3.9425,
|
|
"mean_token_accuracy": 0.29166123270988464,
|
|
"num_tokens": 16456937.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 4.787168312072754,
|
|
"epoch": 7.456066945606695,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001303398233484999,
|
|
"loss": 4.0828,
|
|
"mean_token_accuracy": 0.27813613414764404,
|
|
"num_tokens": 16466714.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 4.782058382034302,
|
|
"epoch": 7.460251046025105,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00013009275858108648,
|
|
"loss": 4.0128,
|
|
"mean_token_accuracy": 0.2803549110889435,
|
|
"num_tokens": 16475219.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 4.727338457107544,
|
|
"epoch": 7.464435146443515,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001298459920094784,
|
|
"loss": 3.9581,
|
|
"mean_token_accuracy": 0.28180397152900694,
|
|
"num_tokens": 16484169.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 4.760374927520752,
|
|
"epoch": 7.468619246861925,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001295995241414818,
|
|
"loss": 4.057,
|
|
"mean_token_accuracy": 0.27491432428359985,
|
|
"num_tokens": 16493413.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 4.750746059417724,
|
|
"epoch": 7.472803347280335,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00012935335548428767,
|
|
"loss": 4.0415,
|
|
"mean_token_accuracy": 0.27418536245822905,
|
|
"num_tokens": 16503195.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 4.72337121963501,
|
|
"epoch": 7.476987447698745,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001291074865444718,
|
|
"loss": 3.9515,
|
|
"mean_token_accuracy": 0.277979364991188,
|
|
"num_tokens": 16511168.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 4.7515229225158695,
|
|
"epoch": 7.481171548117155,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012886191782799267,
|
|
"loss": 4.0198,
|
|
"mean_token_accuracy": 0.28061334490776063,
|
|
"num_tokens": 16520125.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 4.714857292175293,
|
|
"epoch": 7.485355648535565,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012861664984019144,
|
|
"loss": 3.9654,
|
|
"mean_token_accuracy": 0.2826998770236969,
|
|
"num_tokens": 16528121.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 4.730941104888916,
|
|
"epoch": 7.489539748953975,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00012837168308579012,
|
|
"loss": 3.9952,
|
|
"mean_token_accuracy": 0.281155464053154,
|
|
"num_tokens": 16537741.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 4.792245531082154,
|
|
"epoch": 7.493723849372385,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012812701806889077,
|
|
"loss": 4.0875,
|
|
"mean_token_accuracy": 0.274031463265419,
|
|
"num_tokens": 16548091.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 4.822246503829956,
|
|
"epoch": 7.497907949790795,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001278826552929749,
|
|
"loss": 4.0602,
|
|
"mean_token_accuracy": 0.2728355348110199,
|
|
"num_tokens": 16557011.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 4.816597270965576,
|
|
"epoch": 7.502092050209205,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001276385952609016,
|
|
"loss": 4.096,
|
|
"mean_token_accuracy": 0.26269230991601944,
|
|
"num_tokens": 16565868.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 4.7947526454925535,
|
|
"epoch": 7.506276150627615,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00012739483847490734,
|
|
"loss": 4.075,
|
|
"mean_token_accuracy": 0.2765979453921318,
|
|
"num_tokens": 16575131.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 4.799464273452759,
|
|
"epoch": 7.510460251046025,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012715138543660425,
|
|
"loss": 4.1071,
|
|
"mean_token_accuracy": 0.26842520087957383,
|
|
"num_tokens": 16584789.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 4.836958169937134,
|
|
"epoch": 7.514644351464435,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001269082366469797,
|
|
"loss": 4.0913,
|
|
"mean_token_accuracy": 0.27135058790445327,
|
|
"num_tokens": 16594488.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 4.726735353469849,
|
|
"epoch": 7.518828451882845,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00012666539260639475,
|
|
"loss": 3.9856,
|
|
"mean_token_accuracy": 0.2798690527677536,
|
|
"num_tokens": 16602691.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 4.840087938308716,
|
|
"epoch": 7.523012552301255,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001264228538145833,
|
|
"loss": 4.1057,
|
|
"mean_token_accuracy": 0.278209912776947,
|
|
"num_tokens": 16611843.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 4.826506042480469,
|
|
"epoch": 7.527196652719665,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012618062077065126,
|
|
"loss": 4.1022,
|
|
"mean_token_accuracy": 0.2645376518368721,
|
|
"num_tokens": 16620587.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 4.825761556625366,
|
|
"epoch": 7.531380753138075,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00012593869397307527,
|
|
"loss": 4.1619,
|
|
"mean_token_accuracy": 0.2640349090099335,
|
|
"num_tokens": 16631113.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 7.531380753138075,
|
|
"eval_entropy": 4.66939845260667,
|
|
"eval_loss": 5.258347511291504,
|
|
"eval_mean_token_accuracy": 0.20849861754656812,
|
|
"eval_num_tokens": 16631113.0,
|
|
"eval_runtime": 2.7119,
|
|
"eval_samples_per_second": 1440.675,
|
|
"eval_steps_per_second": 180.315,
|
|
"step": 9000
|
|
}
|
|
],
|
|
"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": 2.4229479481344e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|