4035 lines
109 KiB
JSON
4035 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.11183180496533214,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691921043395997,
|
|
"epoch": 0.00027957951241333037,
|
|
"grad_norm": 12.625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8124,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9861.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691890811920166,
|
|
"epoch": 0.0005591590248266607,
|
|
"grad_norm": 13.375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7399,
|
|
"mean_token_accuracy": 0.000473622081335634,
|
|
"num_tokens": 19476.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691054916381836,
|
|
"epoch": 0.0008387385372399911,
|
|
"grad_norm": 9.625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.506,
|
|
"mean_token_accuracy": 0.02479663831181824,
|
|
"num_tokens": 29731.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.682694625854491,
|
|
"epoch": 0.0011183180496533215,
|
|
"grad_norm": 6.4375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1591,
|
|
"mean_token_accuracy": 0.03802054561674595,
|
|
"num_tokens": 38643.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.634562492370605,
|
|
"epoch": 0.0013978975620666518,
|
|
"grad_norm": 4.28125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8669,
|
|
"mean_token_accuracy": 0.04109327364712954,
|
|
"num_tokens": 47707.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.554908943176269,
|
|
"epoch": 0.0016774770744799822,
|
|
"grad_norm": 3.328125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7299,
|
|
"mean_token_accuracy": 0.04091411866247654,
|
|
"num_tokens": 57219.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.570738887786865,
|
|
"epoch": 0.0019570565868933126,
|
|
"grad_norm": 2.90625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6768,
|
|
"mean_token_accuracy": 0.03916576430201531,
|
|
"num_tokens": 66318.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.571315002441406,
|
|
"epoch": 0.002236636099306643,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.5213,
|
|
"mean_token_accuracy": 0.03872227370738983,
|
|
"num_tokens": 74602.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.566377449035645,
|
|
"epoch": 0.0025162156117199733,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.5088,
|
|
"mean_token_accuracy": 0.04258945845067501,
|
|
"num_tokens": 84666.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.548401355743408,
|
|
"epoch": 0.0027957951241333037,
|
|
"grad_norm": 2.671875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.4353,
|
|
"mean_token_accuracy": 0.0500976238399744,
|
|
"num_tokens": 94488.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.471971893310547,
|
|
"epoch": 0.003075374636546634,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.352,
|
|
"mean_token_accuracy": 0.056226722151041034,
|
|
"num_tokens": 104083.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.36090316772461,
|
|
"epoch": 0.0033549541489599644,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.3357,
|
|
"mean_token_accuracy": 0.0560699962079525,
|
|
"num_tokens": 113536.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.390652847290038,
|
|
"epoch": 0.0036345336613732944,
|
|
"grad_norm": 2.859375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.2217,
|
|
"mean_token_accuracy": 0.056196273863315584,
|
|
"num_tokens": 123048.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.416264057159424,
|
|
"epoch": 0.003914113173786625,
|
|
"grad_norm": 2.90625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.0594,
|
|
"mean_token_accuracy": 0.06116741932928562,
|
|
"num_tokens": 131550.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.374309062957764,
|
|
"epoch": 0.004193692686199955,
|
|
"grad_norm": 3.0,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.033,
|
|
"mean_token_accuracy": 0.061420274525880815,
|
|
"num_tokens": 141235.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.379674530029297,
|
|
"epoch": 0.004473272198613286,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 8.9435,
|
|
"mean_token_accuracy": 0.0640605840831995,
|
|
"num_tokens": 150681.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.345459938049316,
|
|
"epoch": 0.004752851711026616,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.8576,
|
|
"mean_token_accuracy": 0.06255088299512863,
|
|
"num_tokens": 160506.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.31331491470337,
|
|
"epoch": 0.005032431223439947,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.7311,
|
|
"mean_token_accuracy": 0.06503418013453484,
|
|
"num_tokens": 169975.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.259163188934327,
|
|
"epoch": 0.005312010735853277,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.6912,
|
|
"mean_token_accuracy": 0.06565042100846767,
|
|
"num_tokens": 178593.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.210173034667969,
|
|
"epoch": 0.005591590248266607,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.5786,
|
|
"mean_token_accuracy": 0.0654834620654583,
|
|
"num_tokens": 188472.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.204802703857421,
|
|
"epoch": 0.005871169760679937,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.4289,
|
|
"mean_token_accuracy": 0.06512044295668602,
|
|
"num_tokens": 197362.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.070987224578857,
|
|
"epoch": 0.006150749273093268,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.3881,
|
|
"mean_token_accuracy": 0.07190057411789894,
|
|
"num_tokens": 207551.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.01644868850708,
|
|
"epoch": 0.006430328785506598,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.1562,
|
|
"mean_token_accuracy": 0.07526783086359501,
|
|
"num_tokens": 218498.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.895391368865967,
|
|
"epoch": 0.006709908297919929,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.0826,
|
|
"mean_token_accuracy": 0.07997448220849038,
|
|
"num_tokens": 228254.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.790292835235595,
|
|
"epoch": 0.006989487810333259,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.9707,
|
|
"mean_token_accuracy": 0.07530777603387832,
|
|
"num_tokens": 237478.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.68728494644165,
|
|
"epoch": 0.007269067322746589,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.9089,
|
|
"mean_token_accuracy": 0.07417170405387878,
|
|
"num_tokens": 246342.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.525614833831787,
|
|
"epoch": 0.0075486468351599195,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.8067,
|
|
"mean_token_accuracy": 0.07382837124168873,
|
|
"num_tokens": 257289.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.321121501922608,
|
|
"epoch": 0.00782822634757325,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.6453,
|
|
"mean_token_accuracy": 0.07587397061288356,
|
|
"num_tokens": 266862.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.158805179595948,
|
|
"epoch": 0.00810780585998658,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.5626,
|
|
"mean_token_accuracy": 0.07878116257488728,
|
|
"num_tokens": 276787.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.895209884643554,
|
|
"epoch": 0.00838738537239991,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.4282,
|
|
"mean_token_accuracy": 0.08020649701356888,
|
|
"num_tokens": 286991.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.710405635833741,
|
|
"epoch": 0.008666964884813241,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.3875,
|
|
"mean_token_accuracy": 0.08550404459238052,
|
|
"num_tokens": 296517.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.512088584899903,
|
|
"epoch": 0.008946544397226572,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.3397,
|
|
"mean_token_accuracy": 0.0807737536728382,
|
|
"num_tokens": 306476.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.354952907562256,
|
|
"epoch": 0.0092261239096399,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.2449,
|
|
"mean_token_accuracy": 0.0858085073530674,
|
|
"num_tokens": 315372.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.175031757354736,
|
|
"epoch": 0.009505703422053232,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.2093,
|
|
"mean_token_accuracy": 0.08027389757335186,
|
|
"num_tokens": 325361.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.012962198257446,
|
|
"epoch": 0.009785282934466562,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.127,
|
|
"mean_token_accuracy": 0.09110699072480202,
|
|
"num_tokens": 335099.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.874845790863037,
|
|
"epoch": 0.010064862446879893,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.037,
|
|
"mean_token_accuracy": 0.08859180435538291,
|
|
"num_tokens": 343850.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.876893758773804,
|
|
"epoch": 0.010344441959293222,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.172,
|
|
"mean_token_accuracy": 0.08158803507685661,
|
|
"num_tokens": 353580.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.905035734176636,
|
|
"epoch": 0.010624021471706553,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.1346,
|
|
"mean_token_accuracy": 0.07959189042448997,
|
|
"num_tokens": 362915.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.806015872955323,
|
|
"epoch": 0.010903600984119884,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.0652,
|
|
"mean_token_accuracy": 0.0788392372429371,
|
|
"num_tokens": 373384.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.610120105743408,
|
|
"epoch": 0.011183180496533215,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.0387,
|
|
"mean_token_accuracy": 0.08487512543797493,
|
|
"num_tokens": 382764.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.788971900939941,
|
|
"epoch": 0.011462760008946544,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.1369,
|
|
"mean_token_accuracy": 0.08739647641777992,
|
|
"num_tokens": 391243.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.615197801589966,
|
|
"epoch": 0.011742339521359875,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.034,
|
|
"mean_token_accuracy": 0.09340409561991692,
|
|
"num_tokens": 400983.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.660063695907593,
|
|
"epoch": 0.012021919033773205,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.8764,
|
|
"mean_token_accuracy": 0.09217005968093872,
|
|
"num_tokens": 410196.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.543980884552002,
|
|
"epoch": 0.012301498546186536,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.9262,
|
|
"mean_token_accuracy": 0.09098378121852875,
|
|
"num_tokens": 419344.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.577476549148559,
|
|
"epoch": 0.012581078058599865,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.0198,
|
|
"mean_token_accuracy": 0.08162087500095368,
|
|
"num_tokens": 429557.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.595211410522461,
|
|
"epoch": 0.012860657571013196,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.038,
|
|
"mean_token_accuracy": 0.07889171466231346,
|
|
"num_tokens": 440474.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.516703319549561,
|
|
"epoch": 0.013140237083426527,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.8554,
|
|
"mean_token_accuracy": 0.09746571034193038,
|
|
"num_tokens": 450552.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.418551683425903,
|
|
"epoch": 0.013419816595839858,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.8744,
|
|
"mean_token_accuracy": 0.08559218943119049,
|
|
"num_tokens": 459501.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.459112453460693,
|
|
"epoch": 0.013699396108253187,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.9676,
|
|
"mean_token_accuracy": 0.08793454244732857,
|
|
"num_tokens": 468807.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.494116115570068,
|
|
"epoch": 0.013978975620666518,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.9437,
|
|
"mean_token_accuracy": 0.08950937315821647,
|
|
"num_tokens": 477386.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.427541923522949,
|
|
"epoch": 0.014258555133079848,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.8711,
|
|
"mean_token_accuracy": 0.08891022279858589,
|
|
"num_tokens": 485894.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.419974088668823,
|
|
"epoch": 0.014538134645493177,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.0094,
|
|
"mean_token_accuracy": 0.08291822224855423,
|
|
"num_tokens": 495083.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.431387090682984,
|
|
"epoch": 0.014817714157906508,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9432,
|
|
"mean_token_accuracy": 0.08412053138017654,
|
|
"num_tokens": 504052.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.427925777435303,
|
|
"epoch": 0.015097293670319839,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.9054,
|
|
"mean_token_accuracy": 0.08690859079360962,
|
|
"num_tokens": 514054.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.391806268692017,
|
|
"epoch": 0.01537687318273317,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.9026,
|
|
"mean_token_accuracy": 0.08920464739203453,
|
|
"num_tokens": 523135.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.383348894119263,
|
|
"epoch": 0.0156564526951465,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.8545,
|
|
"mean_token_accuracy": 0.08667187243700028,
|
|
"num_tokens": 532440.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.335691356658936,
|
|
"epoch": 0.01593603220755983,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.8642,
|
|
"mean_token_accuracy": 0.0919418253004551,
|
|
"num_tokens": 541707.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.2934582233428955,
|
|
"epoch": 0.01621561171997316,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8127,
|
|
"mean_token_accuracy": 0.0894357517361641,
|
|
"num_tokens": 551263.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.366137933731079,
|
|
"epoch": 0.01649519123238649,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.8343,
|
|
"mean_token_accuracy": 0.09377153888344765,
|
|
"num_tokens": 560217.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.273529672622681,
|
|
"epoch": 0.01677477074479982,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.7378,
|
|
"mean_token_accuracy": 0.08633734658360481,
|
|
"num_tokens": 569840.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.181650400161743,
|
|
"epoch": 0.017054350257213153,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.6951,
|
|
"mean_token_accuracy": 0.09873108789324761,
|
|
"num_tokens": 579117.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.249513053894043,
|
|
"epoch": 0.017333929769626482,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.8331,
|
|
"mean_token_accuracy": 0.09003806784749031,
|
|
"num_tokens": 588961.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.089381504058838,
|
|
"epoch": 0.01761350928203981,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.6727,
|
|
"mean_token_accuracy": 0.09398194402456284,
|
|
"num_tokens": 597906.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.242098236083985,
|
|
"epoch": 0.017893088794453144,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.822,
|
|
"mean_token_accuracy": 0.09356562569737434,
|
|
"num_tokens": 607015.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.1716405868530275,
|
|
"epoch": 0.018172668306866473,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.6879,
|
|
"mean_token_accuracy": 0.09605172201991082,
|
|
"num_tokens": 615878.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.05837836265564,
|
|
"epoch": 0.0184522478192798,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.7008,
|
|
"mean_token_accuracy": 0.09321764335036278,
|
|
"num_tokens": 625542.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.229507350921631,
|
|
"epoch": 0.018731827331693134,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.7562,
|
|
"mean_token_accuracy": 0.08855808004736901,
|
|
"num_tokens": 634482.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.227516841888428,
|
|
"epoch": 0.019011406844106463,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.7065,
|
|
"mean_token_accuracy": 0.08909320756793022,
|
|
"num_tokens": 643282.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.060443782806397,
|
|
"epoch": 0.019290986356519796,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.6112,
|
|
"mean_token_accuracy": 0.09391226842999459,
|
|
"num_tokens": 652613.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 6.943112230300903,
|
|
"epoch": 0.019570565868933125,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.5652,
|
|
"mean_token_accuracy": 0.10253061279654503,
|
|
"num_tokens": 662111.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.038354539871216,
|
|
"epoch": 0.019850145381346454,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.7651,
|
|
"mean_token_accuracy": 0.09167603999376298,
|
|
"num_tokens": 670807.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.153097629547119,
|
|
"epoch": 0.020129724893759787,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7888,
|
|
"mean_token_accuracy": 0.0894574448466301,
|
|
"num_tokens": 680447.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.086006307601929,
|
|
"epoch": 0.020409304406173116,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.7339,
|
|
"mean_token_accuracy": 0.08920840471982956,
|
|
"num_tokens": 690818.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.078864479064942,
|
|
"epoch": 0.020688883918586445,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7043,
|
|
"mean_token_accuracy": 0.09464118853211403,
|
|
"num_tokens": 700014.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.129553127288818,
|
|
"epoch": 0.020968463430999777,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7369,
|
|
"mean_token_accuracy": 0.09654519632458687,
|
|
"num_tokens": 710390.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.971775245666504,
|
|
"epoch": 0.021248042943413106,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.5676,
|
|
"mean_token_accuracy": 0.1039146564900875,
|
|
"num_tokens": 719907.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 6.9949382781982425,
|
|
"epoch": 0.021527622455826435,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.6302,
|
|
"mean_token_accuracy": 0.09534877464175225,
|
|
"num_tokens": 730641.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.961507129669189,
|
|
"epoch": 0.021807201968239768,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.5664,
|
|
"mean_token_accuracy": 0.10222888961434365,
|
|
"num_tokens": 739638.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.8918750286102295,
|
|
"epoch": 0.022086781480653097,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.5968,
|
|
"mean_token_accuracy": 0.09721470028162002,
|
|
"num_tokens": 748638.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.056669282913208,
|
|
"epoch": 0.02236636099306643,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.6367,
|
|
"mean_token_accuracy": 0.09450221732258797,
|
|
"num_tokens": 758705.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.024502992630005,
|
|
"epoch": 0.02264594050547976,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.6539,
|
|
"mean_token_accuracy": 0.09404293596744537,
|
|
"num_tokens": 768527.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.913692283630371,
|
|
"epoch": 0.022925520017893088,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.5867,
|
|
"mean_token_accuracy": 0.09760746508836746,
|
|
"num_tokens": 778533.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.921697902679443,
|
|
"epoch": 0.02320509953030642,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.5677,
|
|
"mean_token_accuracy": 0.09978625848889351,
|
|
"num_tokens": 788171.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.961454296112061,
|
|
"epoch": 0.02348467904271975,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.6712,
|
|
"mean_token_accuracy": 0.09329927712678909,
|
|
"num_tokens": 797928.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.977083206176758,
|
|
"epoch": 0.02376425855513308,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.5859,
|
|
"mean_token_accuracy": 0.0953036293387413,
|
|
"num_tokens": 807565.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.022476720809936,
|
|
"epoch": 0.02404383806754641,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6095,
|
|
"mean_token_accuracy": 0.09925682097673416,
|
|
"num_tokens": 817018.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.906362009048462,
|
|
"epoch": 0.02432341757995974,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.5497,
|
|
"mean_token_accuracy": 0.1020013377070427,
|
|
"num_tokens": 826850.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.890042304992676,
|
|
"epoch": 0.024602997092373072,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.5654,
|
|
"mean_token_accuracy": 0.10329198613762855,
|
|
"num_tokens": 835097.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.845738363265991,
|
|
"epoch": 0.0248825766047864,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.5229,
|
|
"mean_token_accuracy": 0.10356307923793792,
|
|
"num_tokens": 844645.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.889461040496826,
|
|
"epoch": 0.02516215611719973,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.5038,
|
|
"mean_token_accuracy": 0.10186199322342873,
|
|
"num_tokens": 853912.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.83258090019226,
|
|
"epoch": 0.025441735629613063,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.5741,
|
|
"mean_token_accuracy": 0.10296021774411201,
|
|
"num_tokens": 863312.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.889059400558471,
|
|
"epoch": 0.025721315142026392,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.5453,
|
|
"mean_token_accuracy": 0.09688765034079552,
|
|
"num_tokens": 873075.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.824527597427368,
|
|
"epoch": 0.02600089465443972,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.4318,
|
|
"mean_token_accuracy": 0.10380328297615052,
|
|
"num_tokens": 880964.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.830307388305664,
|
|
"epoch": 0.026280474166853054,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5615,
|
|
"mean_token_accuracy": 0.09817187637090682,
|
|
"num_tokens": 891148.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.774664688110351,
|
|
"epoch": 0.026560053679266383,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.4576,
|
|
"mean_token_accuracy": 0.09947997033596039,
|
|
"num_tokens": 900635.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.78422474861145,
|
|
"epoch": 0.026839633191679715,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.4409,
|
|
"mean_token_accuracy": 0.10720937103033065,
|
|
"num_tokens": 910454.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.737676954269409,
|
|
"epoch": 0.027119212704093044,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.456,
|
|
"mean_token_accuracy": 0.10532223731279373,
|
|
"num_tokens": 919504.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.7197558879852295,
|
|
"epoch": 0.027398792216506374,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.4703,
|
|
"mean_token_accuracy": 0.10164613649249077,
|
|
"num_tokens": 929255.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.818446922302246,
|
|
"epoch": 0.027678371728919706,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5297,
|
|
"mean_token_accuracy": 0.09887348264455795,
|
|
"num_tokens": 939144.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.689020252227783,
|
|
"epoch": 0.027957951241333035,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.3954,
|
|
"mean_token_accuracy": 0.10200168937444687,
|
|
"num_tokens": 947345.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.830253791809082,
|
|
"epoch": 0.028237530753746364,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.5232,
|
|
"mean_token_accuracy": 0.10504392087459564,
|
|
"num_tokens": 958117.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.81278166770935,
|
|
"epoch": 0.028517110266159697,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.4756,
|
|
"mean_token_accuracy": 0.10109646022319793,
|
|
"num_tokens": 967625.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.846687746047974,
|
|
"epoch": 0.028796689778573026,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5703,
|
|
"mean_token_accuracy": 0.09944209083914757,
|
|
"num_tokens": 978239.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.603201675415039,
|
|
"epoch": 0.029076269290986355,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5511,
|
|
"mean_token_accuracy": 0.09840744361281395,
|
|
"num_tokens": 988944.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.8358293056488035,
|
|
"epoch": 0.029355848803399687,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.4261,
|
|
"mean_token_accuracy": 0.10276176407933235,
|
|
"num_tokens": 998713.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.781039047241211,
|
|
"epoch": 0.029635428315813016,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.4551,
|
|
"mean_token_accuracy": 0.10137152597308159,
|
|
"num_tokens": 1008330.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.695677280426025,
|
|
"epoch": 0.02991500782822635,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.4635,
|
|
"mean_token_accuracy": 0.10742506682872772,
|
|
"num_tokens": 1017890.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.668637800216675,
|
|
"epoch": 0.030194587340639678,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.3999,
|
|
"mean_token_accuracy": 0.10795639976859092,
|
|
"num_tokens": 1028140.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.603444910049438,
|
|
"epoch": 0.030474166853053007,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.3541,
|
|
"mean_token_accuracy": 0.10782371163368225,
|
|
"num_tokens": 1038288.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.673272657394409,
|
|
"epoch": 0.03075374636546634,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.3632,
|
|
"mean_token_accuracy": 0.11124667599797249,
|
|
"num_tokens": 1048035.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.6859392642974855,
|
|
"epoch": 0.03103332587787967,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4133,
|
|
"mean_token_accuracy": 0.10540587455034256,
|
|
"num_tokens": 1057563.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.6857696056365965,
|
|
"epoch": 0.031312905390293,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.385,
|
|
"mean_token_accuracy": 0.11307021751999854,
|
|
"num_tokens": 1066562.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.650216293334961,
|
|
"epoch": 0.03159248490270633,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.3658,
|
|
"mean_token_accuracy": 0.10885544866323471,
|
|
"num_tokens": 1075964.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.616510009765625,
|
|
"epoch": 0.03187206441511966,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4079,
|
|
"mean_token_accuracy": 0.10936605781316758,
|
|
"num_tokens": 1085079.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.689432954788208,
|
|
"epoch": 0.03215164392753299,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.444,
|
|
"mean_token_accuracy": 0.10385599657893181,
|
|
"num_tokens": 1095052.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.681822299957275,
|
|
"epoch": 0.03243122343994632,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.4066,
|
|
"mean_token_accuracy": 0.10496392622590064,
|
|
"num_tokens": 1103730.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.613558197021485,
|
|
"epoch": 0.032710802952359654,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4465,
|
|
"mean_token_accuracy": 0.10784894451498986,
|
|
"num_tokens": 1112988.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.709567213058472,
|
|
"epoch": 0.03299038246477298,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.3832,
|
|
"mean_token_accuracy": 0.1152427189052105,
|
|
"num_tokens": 1121739.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.573565816879272,
|
|
"epoch": 0.03326996197718631,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.3168,
|
|
"mean_token_accuracy": 0.11211551204323769,
|
|
"num_tokens": 1131119.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.604477214813232,
|
|
"epoch": 0.03354954148959964,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4409,
|
|
"mean_token_accuracy": 0.10594701915979385,
|
|
"num_tokens": 1140939.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.550174760818481,
|
|
"epoch": 0.03382912100201297,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.3275,
|
|
"mean_token_accuracy": 0.10821124091744423,
|
|
"num_tokens": 1151084.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.585816049575806,
|
|
"epoch": 0.034108700514426306,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.4047,
|
|
"mean_token_accuracy": 0.10988360941410065,
|
|
"num_tokens": 1161153.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.639032363891602,
|
|
"epoch": 0.034388280026839635,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.3557,
|
|
"mean_token_accuracy": 0.10898142084479331,
|
|
"num_tokens": 1171040.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.541324710845947,
|
|
"epoch": 0.034667859539252964,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.3101,
|
|
"mean_token_accuracy": 0.11271897107362747,
|
|
"num_tokens": 1180090.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.451202249526977,
|
|
"epoch": 0.03494743905166629,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.2325,
|
|
"mean_token_accuracy": 0.11579823195934295,
|
|
"num_tokens": 1188494.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.548623371124267,
|
|
"epoch": 0.03522701856407962,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4326,
|
|
"mean_token_accuracy": 0.10637508630752564,
|
|
"num_tokens": 1198530.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.576272916793823,
|
|
"epoch": 0.03550659807649295,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.3694,
|
|
"mean_token_accuracy": 0.10622836947441101,
|
|
"num_tokens": 1207590.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.6827983379364015,
|
|
"epoch": 0.03578617758890629,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.3672,
|
|
"mean_token_accuracy": 0.1092015527188778,
|
|
"num_tokens": 1216201.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.436986255645752,
|
|
"epoch": 0.036065757101319616,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.234,
|
|
"mean_token_accuracy": 0.12257840186357498,
|
|
"num_tokens": 1225279.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.503623056411743,
|
|
"epoch": 0.036345336613732945,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.3269,
|
|
"mean_token_accuracy": 0.1132922776043415,
|
|
"num_tokens": 1235963.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.565838718414307,
|
|
"epoch": 0.036624916126146274,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.3427,
|
|
"mean_token_accuracy": 0.11427977606654167,
|
|
"num_tokens": 1244859.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.50637845993042,
|
|
"epoch": 0.0369044956385596,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.3278,
|
|
"mean_token_accuracy": 0.11026786714792251,
|
|
"num_tokens": 1254993.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.541554737091064,
|
|
"epoch": 0.03718407515097294,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.3292,
|
|
"mean_token_accuracy": 0.11254208162426949,
|
|
"num_tokens": 1264954.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.491656398773193,
|
|
"epoch": 0.03746365466338627,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.3444,
|
|
"mean_token_accuracy": 0.11582160592079163,
|
|
"num_tokens": 1275133.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.579359912872315,
|
|
"epoch": 0.0377432341757996,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.3413,
|
|
"mean_token_accuracy": 0.1117575354874134,
|
|
"num_tokens": 1285046.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.5179187774658205,
|
|
"epoch": 0.03802281368821293,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.438,
|
|
"mean_token_accuracy": 0.1056198701262474,
|
|
"num_tokens": 1295222.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.391375923156739,
|
|
"epoch": 0.038302393200626256,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.1413,
|
|
"mean_token_accuracy": 0.12317045554518699,
|
|
"num_tokens": 1304162.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.32426347732544,
|
|
"epoch": 0.03858197271303959,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.2517,
|
|
"mean_token_accuracy": 0.11537077352404594,
|
|
"num_tokens": 1312961.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.509217357635498,
|
|
"epoch": 0.03886155222545292,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.1873,
|
|
"mean_token_accuracy": 0.11677368059754371,
|
|
"num_tokens": 1321341.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.537080430984497,
|
|
"epoch": 0.03914113173786625,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3003,
|
|
"mean_token_accuracy": 0.11517599448561669,
|
|
"num_tokens": 1329824.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.424460983276367,
|
|
"epoch": 0.03942071125027958,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.2543,
|
|
"mean_token_accuracy": 0.11960380971431732,
|
|
"num_tokens": 1338502.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.429356813430786,
|
|
"epoch": 0.03970029076269291,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.2961,
|
|
"mean_token_accuracy": 0.10694030821323394,
|
|
"num_tokens": 1347703.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.443802165985107,
|
|
"epoch": 0.03997987027510624,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.3088,
|
|
"mean_token_accuracy": 0.10902290344238282,
|
|
"num_tokens": 1358546.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.435325813293457,
|
|
"epoch": 0.04025944978751957,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.2891,
|
|
"mean_token_accuracy": 0.11368739604949951,
|
|
"num_tokens": 1368176.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.37840690612793,
|
|
"epoch": 0.0405390292999329,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.1848,
|
|
"mean_token_accuracy": 0.11788561046123505,
|
|
"num_tokens": 1377755.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.57690315246582,
|
|
"epoch": 0.04081860881234623,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.251,
|
|
"mean_token_accuracy": 0.12233446016907693,
|
|
"num_tokens": 1386376.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.359943389892578,
|
|
"epoch": 0.04109818832475956,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.2974,
|
|
"mean_token_accuracy": 0.11713942512869835,
|
|
"num_tokens": 1396750.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.4675477027893065,
|
|
"epoch": 0.04137776783717289,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3328,
|
|
"mean_token_accuracy": 0.11545042395591736,
|
|
"num_tokens": 1406439.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.46083631515503,
|
|
"epoch": 0.041657347349586225,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.323,
|
|
"mean_token_accuracy": 0.1170242041349411,
|
|
"num_tokens": 1415853.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.380451917648315,
|
|
"epoch": 0.041936926861999554,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.2084,
|
|
"mean_token_accuracy": 0.1251394920051098,
|
|
"num_tokens": 1424787.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.4544353008270265,
|
|
"epoch": 0.042216506374412884,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.2667,
|
|
"mean_token_accuracy": 0.12094322144985199,
|
|
"num_tokens": 1433747.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.35882077217102,
|
|
"epoch": 0.04249608588682621,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.2451,
|
|
"mean_token_accuracy": 0.1215454287827015,
|
|
"num_tokens": 1443284.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.328829431533814,
|
|
"epoch": 0.04277566539923954,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.2157,
|
|
"mean_token_accuracy": 0.11974160522222518,
|
|
"num_tokens": 1453526.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.357368755340576,
|
|
"epoch": 0.04305524491165287,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.1142,
|
|
"mean_token_accuracy": 0.12436131685972214,
|
|
"num_tokens": 1461719.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.375395154953003,
|
|
"epoch": 0.04333482442406621,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.319,
|
|
"mean_token_accuracy": 0.11257023587822915,
|
|
"num_tokens": 1471879.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.403350496292115,
|
|
"epoch": 0.043614403936479536,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.2563,
|
|
"mean_token_accuracy": 0.12221399322152138,
|
|
"num_tokens": 1480857.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.433130979537964,
|
|
"epoch": 0.043893983448892865,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.1836,
|
|
"mean_token_accuracy": 0.12073334529995919,
|
|
"num_tokens": 1490291.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.377252674102783,
|
|
"epoch": 0.044173562961306194,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3109,
|
|
"mean_token_accuracy": 0.1143433466553688,
|
|
"num_tokens": 1501359.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.186716222763062,
|
|
"epoch": 0.04445314247371952,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.1623,
|
|
"mean_token_accuracy": 0.12547508999705315,
|
|
"num_tokens": 1510773.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.462558174133301,
|
|
"epoch": 0.04473272198613286,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.269,
|
|
"mean_token_accuracy": 0.12258902341127395,
|
|
"num_tokens": 1521653.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.196301889419556,
|
|
"epoch": 0.04501230149854619,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.1371,
|
|
"mean_token_accuracy": 0.12039914578199387,
|
|
"num_tokens": 1530854.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.35107388496399,
|
|
"epoch": 0.04529188101095952,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.2,
|
|
"mean_token_accuracy": 0.12085949927568436,
|
|
"num_tokens": 1539652.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.309471082687378,
|
|
"epoch": 0.045571460523372846,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.0983,
|
|
"mean_token_accuracy": 0.13143228217959405,
|
|
"num_tokens": 1548501.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.33494005203247,
|
|
"epoch": 0.045851040035786175,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2221,
|
|
"mean_token_accuracy": 0.1150277242064476,
|
|
"num_tokens": 1557687.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.201799964904785,
|
|
"epoch": 0.04613061954819951,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2097,
|
|
"mean_token_accuracy": 0.11941343322396278,
|
|
"num_tokens": 1568189.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.377394199371338,
|
|
"epoch": 0.04641019906061284,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.1824,
|
|
"mean_token_accuracy": 0.12035990878939629,
|
|
"num_tokens": 1577773.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.335377168655396,
|
|
"epoch": 0.04668977857302617,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.248,
|
|
"mean_token_accuracy": 0.11886420771479607,
|
|
"num_tokens": 1586780.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.38820219039917,
|
|
"epoch": 0.0469693580854395,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.2367,
|
|
"mean_token_accuracy": 0.12530419155955314,
|
|
"num_tokens": 1595841.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.271928977966309,
|
|
"epoch": 0.04724893759785283,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.0992,
|
|
"mean_token_accuracy": 0.12625835835933685,
|
|
"num_tokens": 1604758.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.234389209747315,
|
|
"epoch": 0.04752851711026616,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.1847,
|
|
"mean_token_accuracy": 0.12268484532833099,
|
|
"num_tokens": 1614594.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.323423290252686,
|
|
"epoch": 0.04780809662267949,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.1528,
|
|
"mean_token_accuracy": 0.12712193951010703,
|
|
"num_tokens": 1623774.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.268539094924927,
|
|
"epoch": 0.04808767613509282,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.2689,
|
|
"mean_token_accuracy": 0.12009239718317985,
|
|
"num_tokens": 1633722.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.398777961730957,
|
|
"epoch": 0.04836725564750615,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.1915,
|
|
"mean_token_accuracy": 0.12330001294612884,
|
|
"num_tokens": 1642661.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.282291746139526,
|
|
"epoch": 0.04864683515991948,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.1266,
|
|
"mean_token_accuracy": 0.12796352654695511,
|
|
"num_tokens": 1651863.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.330269622802734,
|
|
"epoch": 0.04892641467233281,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2741,
|
|
"mean_token_accuracy": 0.11899161115288734,
|
|
"num_tokens": 1662303.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.264892482757569,
|
|
"epoch": 0.049205994184746145,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2517,
|
|
"mean_token_accuracy": 0.11673824712634087,
|
|
"num_tokens": 1672736.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.236477041244507,
|
|
"epoch": 0.049485573697159474,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2377,
|
|
"mean_token_accuracy": 0.12511397823691367,
|
|
"num_tokens": 1682739.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.167548322677613,
|
|
"epoch": 0.0497651532095728,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.0601,
|
|
"mean_token_accuracy": 0.12563416585326195,
|
|
"num_tokens": 1692217.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.289210653305053,
|
|
"epoch": 0.05004473272198613,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.1921,
|
|
"mean_token_accuracy": 0.1288679786026478,
|
|
"num_tokens": 1701902.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.2592321872711185,
|
|
"epoch": 0.05032431223439946,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.1973,
|
|
"mean_token_accuracy": 0.12623300552368164,
|
|
"num_tokens": 1710883.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.2084451675415036,
|
|
"epoch": 0.05060389174681279,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.1473,
|
|
"mean_token_accuracy": 0.12902547046542168,
|
|
"num_tokens": 1720814.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.192575931549072,
|
|
"epoch": 0.050883471259226126,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.0821,
|
|
"mean_token_accuracy": 0.12524855583906175,
|
|
"num_tokens": 1729306.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.173401403427124,
|
|
"epoch": 0.051163050771639455,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.0767,
|
|
"mean_token_accuracy": 0.12862516418099404,
|
|
"num_tokens": 1738054.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.156796979904175,
|
|
"epoch": 0.051442630284052784,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.1263,
|
|
"mean_token_accuracy": 0.1286124937236309,
|
|
"num_tokens": 1747425.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.157821464538574,
|
|
"epoch": 0.05172220979646611,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.0932,
|
|
"mean_token_accuracy": 0.13138109743595122,
|
|
"num_tokens": 1757333.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.214243745803833,
|
|
"epoch": 0.05200178930887944,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.0812,
|
|
"mean_token_accuracy": 0.13311730772256852,
|
|
"num_tokens": 1766502.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.189770793914795,
|
|
"epoch": 0.05228136882129278,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.1559,
|
|
"mean_token_accuracy": 0.12367449924349785,
|
|
"num_tokens": 1776354.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.170189952850341,
|
|
"epoch": 0.05256094833370611,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1648,
|
|
"mean_token_accuracy": 0.12078867107629776,
|
|
"num_tokens": 1785757.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.193646049499511,
|
|
"epoch": 0.05284052784611944,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.0093,
|
|
"mean_token_accuracy": 0.1286062143743038,
|
|
"num_tokens": 1795306.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.227564573287964,
|
|
"epoch": 0.053120107358532766,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.2094,
|
|
"mean_token_accuracy": 0.12222035005688667,
|
|
"num_tokens": 1804694.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.150158262252807,
|
|
"epoch": 0.053399686870946095,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1992,
|
|
"mean_token_accuracy": 0.12697044461965562,
|
|
"num_tokens": 1813684.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.274415349960327,
|
|
"epoch": 0.05367926638335943,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1509,
|
|
"mean_token_accuracy": 0.12724147289991378,
|
|
"num_tokens": 1822343.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.122313642501831,
|
|
"epoch": 0.05395884589577276,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1835,
|
|
"mean_token_accuracy": 0.12922245040535926,
|
|
"num_tokens": 1832629.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.251108312606812,
|
|
"epoch": 0.05423842540818609,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1858,
|
|
"mean_token_accuracy": 0.12649309858679772,
|
|
"num_tokens": 1842430.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.256328678131103,
|
|
"epoch": 0.05451800492059942,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.1895,
|
|
"mean_token_accuracy": 0.11918024644255638,
|
|
"num_tokens": 1852098.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.084271621704102,
|
|
"epoch": 0.05479758443301275,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.1143,
|
|
"mean_token_accuracy": 0.1300920344889164,
|
|
"num_tokens": 1861419.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.2369451999664305,
|
|
"epoch": 0.055077163945426076,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0966,
|
|
"mean_token_accuracy": 0.12017295733094216,
|
|
"num_tokens": 1871050.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.07799916267395,
|
|
"epoch": 0.05535674345783941,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.0793,
|
|
"mean_token_accuracy": 0.12756502851843834,
|
|
"num_tokens": 1879884.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.22581639289856,
|
|
"epoch": 0.05563632297025274,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2094,
|
|
"mean_token_accuracy": 0.12566193267703057,
|
|
"num_tokens": 1889899.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.057796049118042,
|
|
"epoch": 0.05591590248266607,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.028,
|
|
"mean_token_accuracy": 0.12878876328468322,
|
|
"num_tokens": 1899310.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.228454685211181,
|
|
"epoch": 0.0561954819950794,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.2274,
|
|
"mean_token_accuracy": 0.12226422131061554,
|
|
"num_tokens": 1908665.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.183216905593872,
|
|
"epoch": 0.05647506150749273,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.0535,
|
|
"mean_token_accuracy": 0.13170111998915673,
|
|
"num_tokens": 1917976.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 5.981611537933349,
|
|
"epoch": 0.056754641019906064,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.0002,
|
|
"mean_token_accuracy": 0.13305776789784432,
|
|
"num_tokens": 1927782.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.1958225727081295,
|
|
"epoch": 0.057034220532319393,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.116,
|
|
"mean_token_accuracy": 0.13198450952768326,
|
|
"num_tokens": 1936491.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.0382500171661375,
|
|
"epoch": 0.05731380004473272,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 5.9544,
|
|
"mean_token_accuracy": 0.1326470360159874,
|
|
"num_tokens": 1945206.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.076463794708252,
|
|
"epoch": 0.05759337955714605,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.0648,
|
|
"mean_token_accuracy": 0.12909724861383437,
|
|
"num_tokens": 1953484.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.135188627243042,
|
|
"epoch": 0.05787295906955938,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.1867,
|
|
"mean_token_accuracy": 0.12705333158373833,
|
|
"num_tokens": 1963211.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.097447109222412,
|
|
"epoch": 0.05815253858197271,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.0179,
|
|
"mean_token_accuracy": 0.13138104230165482,
|
|
"num_tokens": 1972850.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.0324310779571535,
|
|
"epoch": 0.058432118094386046,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 5.9963,
|
|
"mean_token_accuracy": 0.12721050530672073,
|
|
"num_tokens": 1982250.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 5.9874961376190186,
|
|
"epoch": 0.058711697606799375,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.0296,
|
|
"mean_token_accuracy": 0.13246465846896172,
|
|
"num_tokens": 1990891.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.193099641799927,
|
|
"epoch": 0.058991277119212704,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.0824,
|
|
"mean_token_accuracy": 0.12604981437325477,
|
|
"num_tokens": 2000993.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.022122144699097,
|
|
"epoch": 0.05927085663162603,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.0018,
|
|
"mean_token_accuracy": 0.12894707098603247,
|
|
"num_tokens": 2010190.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.057659482955932,
|
|
"epoch": 0.05955043614403936,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.057,
|
|
"mean_token_accuracy": 0.12649599984288215,
|
|
"num_tokens": 2020169.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.127909851074219,
|
|
"epoch": 0.0598300156564527,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.1296,
|
|
"mean_token_accuracy": 0.13053712248802185,
|
|
"num_tokens": 2029389.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.112711334228516,
|
|
"epoch": 0.06010959516886603,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.1062,
|
|
"mean_token_accuracy": 0.1262196883559227,
|
|
"num_tokens": 2038819.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.189296293258667,
|
|
"epoch": 0.060389174681279356,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.1302,
|
|
"mean_token_accuracy": 0.13122991248965263,
|
|
"num_tokens": 2048431.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.003720903396607,
|
|
"epoch": 0.060668754193692685,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.1277,
|
|
"mean_token_accuracy": 0.12579770088195802,
|
|
"num_tokens": 2057888.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.161061239242554,
|
|
"epoch": 0.060948333706106014,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.0445,
|
|
"mean_token_accuracy": 0.12951943650841713,
|
|
"num_tokens": 2067285.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.078293418884277,
|
|
"epoch": 0.06122791321851935,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.1194,
|
|
"mean_token_accuracy": 0.12521448060870172,
|
|
"num_tokens": 2077593.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.011837148666382,
|
|
"epoch": 0.06150749273093268,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.0021,
|
|
"mean_token_accuracy": 0.1356680363416672,
|
|
"num_tokens": 2086882.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.099917697906494,
|
|
"epoch": 0.06178707224334601,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 5.9472,
|
|
"mean_token_accuracy": 0.13609085083007813,
|
|
"num_tokens": 2095210.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 5.976088380813598,
|
|
"epoch": 0.06206665175575934,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.0378,
|
|
"mean_token_accuracy": 0.1391161397099495,
|
|
"num_tokens": 2104568.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.024702358245849,
|
|
"epoch": 0.06234623126817267,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 5.9214,
|
|
"mean_token_accuracy": 0.1399237684905529,
|
|
"num_tokens": 2112712.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.089004230499268,
|
|
"epoch": 0.062625810780586,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.1019,
|
|
"mean_token_accuracy": 0.12905315309762955,
|
|
"num_tokens": 2121952.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 5.940901374816894,
|
|
"epoch": 0.06290539029299932,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.0644,
|
|
"mean_token_accuracy": 0.13228336945176125,
|
|
"num_tokens": 2131686.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.244207572937012,
|
|
"epoch": 0.06318496980541266,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.0857,
|
|
"mean_token_accuracy": 0.12588815167546272,
|
|
"num_tokens": 2142077.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.0611552715301515,
|
|
"epoch": 0.06346454931782598,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.0251,
|
|
"mean_token_accuracy": 0.12464200705289841,
|
|
"num_tokens": 2151079.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 5.918131542205811,
|
|
"epoch": 0.06374412883023932,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 5.9766,
|
|
"mean_token_accuracy": 0.13423026502132415,
|
|
"num_tokens": 2160350.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.233870983123779,
|
|
"epoch": 0.06402370834265265,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.0632,
|
|
"mean_token_accuracy": 0.1333743192255497,
|
|
"num_tokens": 2169549.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.020859146118164,
|
|
"epoch": 0.06430328785506598,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 5.9857,
|
|
"mean_token_accuracy": 0.13349269032478334,
|
|
"num_tokens": 2178748.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.026945495605469,
|
|
"epoch": 0.06458286736747931,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.0786,
|
|
"mean_token_accuracy": 0.12803979068994523,
|
|
"num_tokens": 2187547.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.127947998046875,
|
|
"epoch": 0.06486244687989264,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.0644,
|
|
"mean_token_accuracy": 0.13440939262509347,
|
|
"num_tokens": 2197232.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 5.9792999744415285,
|
|
"epoch": 0.06514202639230597,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 5.9935,
|
|
"mean_token_accuracy": 0.13207203671336173,
|
|
"num_tokens": 2206582.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.131474590301513,
|
|
"epoch": 0.06542160590471931,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.116,
|
|
"mean_token_accuracy": 0.12943438962101936,
|
|
"num_tokens": 2216930.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 5.9718420028686525,
|
|
"epoch": 0.06570118541713263,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.0046,
|
|
"mean_token_accuracy": 0.13069941401481627,
|
|
"num_tokens": 2227016.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.119905662536621,
|
|
"epoch": 0.06598076492954597,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.014,
|
|
"mean_token_accuracy": 0.14061715230345725,
|
|
"num_tokens": 2236193.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 5.858626651763916,
|
|
"epoch": 0.06626034444195929,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 5.9325,
|
|
"mean_token_accuracy": 0.13983174264431,
|
|
"num_tokens": 2245682.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 5.989790058135986,
|
|
"epoch": 0.06653992395437262,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 5.9007,
|
|
"mean_token_accuracy": 0.13621212393045426,
|
|
"num_tokens": 2255762.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.045697832107544,
|
|
"epoch": 0.06681950346678596,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.0505,
|
|
"mean_token_accuracy": 0.14178939536213875,
|
|
"num_tokens": 2265601.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 5.980489063262939,
|
|
"epoch": 0.06709908297919928,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 5.9794,
|
|
"mean_token_accuracy": 0.1376929186284542,
|
|
"num_tokens": 2274851.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.02736086845398,
|
|
"epoch": 0.06737866249161262,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 5.9045,
|
|
"mean_token_accuracy": 0.13429685235023497,
|
|
"num_tokens": 2284698.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 5.91643123626709,
|
|
"epoch": 0.06765824200402594,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.0598,
|
|
"mean_token_accuracy": 0.13639723956584932,
|
|
"num_tokens": 2293773.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.035378313064575,
|
|
"epoch": 0.06793782151643928,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.0202,
|
|
"mean_token_accuracy": 0.1408623367547989,
|
|
"num_tokens": 2304018.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.070549154281617,
|
|
"epoch": 0.06821740102885261,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 5.9729,
|
|
"mean_token_accuracy": 0.13580385074019433,
|
|
"num_tokens": 2313104.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.0507101058959964,
|
|
"epoch": 0.06849698054126593,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.0819,
|
|
"mean_token_accuracy": 0.1335542805492878,
|
|
"num_tokens": 2321663.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.041438913345337,
|
|
"epoch": 0.06877656005367927,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.13460599184036254,
|
|
"num_tokens": 2330932.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.990973615646363,
|
|
"epoch": 0.06905613956609259,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.0421,
|
|
"mean_token_accuracy": 0.135109806060791,
|
|
"num_tokens": 2341103.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.032921075820923,
|
|
"epoch": 0.06933571907850593,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.014,
|
|
"mean_token_accuracy": 0.13564290553331376,
|
|
"num_tokens": 2350574.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.910844516754151,
|
|
"epoch": 0.06961529859091926,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 5.9394,
|
|
"mean_token_accuracy": 0.14070913344621658,
|
|
"num_tokens": 2359923.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.932819509506226,
|
|
"epoch": 0.06989487810333259,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 5.8602,
|
|
"mean_token_accuracy": 0.1441001147031784,
|
|
"num_tokens": 2369012.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.046881294250488,
|
|
"epoch": 0.07017445761574592,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 5.9666,
|
|
"mean_token_accuracy": 0.1351661428809166,
|
|
"num_tokens": 2379453.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 5.9998232364654545,
|
|
"epoch": 0.07045403712815924,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.0966,
|
|
"mean_token_accuracy": 0.13075585216283797,
|
|
"num_tokens": 2390213.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.958931922912598,
|
|
"epoch": 0.07073361664057258,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 5.8511,
|
|
"mean_token_accuracy": 0.14814525842666626,
|
|
"num_tokens": 2399115.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.002986717224121,
|
|
"epoch": 0.0710131961529859,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 5.9227,
|
|
"mean_token_accuracy": 0.13491304144263266,
|
|
"num_tokens": 2408708.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.939911794662476,
|
|
"epoch": 0.07129277566539924,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 5.8968,
|
|
"mean_token_accuracy": 0.14238273054361344,
|
|
"num_tokens": 2418272.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.945481395721435,
|
|
"epoch": 0.07157235517781257,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 5.9992,
|
|
"mean_token_accuracy": 0.1358647659420967,
|
|
"num_tokens": 2427666.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.06876654624939,
|
|
"epoch": 0.0718519346902259,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 5.9385,
|
|
"mean_token_accuracy": 0.13815628215670586,
|
|
"num_tokens": 2438303.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 5.869863414764405,
|
|
"epoch": 0.07213151420263923,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 5.949,
|
|
"mean_token_accuracy": 0.13556391447782518,
|
|
"num_tokens": 2448355.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.0578197002410885,
|
|
"epoch": 0.07241109371505255,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 5.9861,
|
|
"mean_token_accuracy": 0.1376705937087536,
|
|
"num_tokens": 2457379.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.920118951797486,
|
|
"epoch": 0.07269067322746589,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 5.921,
|
|
"mean_token_accuracy": 0.14255468249320985,
|
|
"num_tokens": 2466167.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.926219415664673,
|
|
"epoch": 0.07297025273987923,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 5.8465,
|
|
"mean_token_accuracy": 0.14156318679451943,
|
|
"num_tokens": 2474585.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.86918363571167,
|
|
"epoch": 0.07324983225229255,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 5.8482,
|
|
"mean_token_accuracy": 0.14929744824767113,
|
|
"num_tokens": 2483340.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.979387664794922,
|
|
"epoch": 0.07352941176470588,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 5.9188,
|
|
"mean_token_accuracy": 0.14429966509342193,
|
|
"num_tokens": 2493329.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 5.933085298538208,
|
|
"epoch": 0.0738089912771192,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 5.953,
|
|
"mean_token_accuracy": 0.13175275921821594,
|
|
"num_tokens": 2502481.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.882027912139892,
|
|
"epoch": 0.07408857078953254,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.14007107689976692,
|
|
"num_tokens": 2512634.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.88964524269104,
|
|
"epoch": 0.07436815030194588,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 5.9019,
|
|
"mean_token_accuracy": 0.13647521212697028,
|
|
"num_tokens": 2521198.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.971690940856933,
|
|
"epoch": 0.0746477298143592,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 5.9338,
|
|
"mean_token_accuracy": 0.14216780960559844,
|
|
"num_tokens": 2531125.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.883842611312867,
|
|
"epoch": 0.07492730932677254,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 5.7639,
|
|
"mean_token_accuracy": 0.1473868668079376,
|
|
"num_tokens": 2540004.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.910669612884521,
|
|
"epoch": 0.07520688883918586,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 5.8575,
|
|
"mean_token_accuracy": 0.14594885036349298,
|
|
"num_tokens": 2549264.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.865490579605103,
|
|
"epoch": 0.0754864683515992,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 5.8441,
|
|
"mean_token_accuracy": 0.14929619431495667,
|
|
"num_tokens": 2557984.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.978964471817017,
|
|
"epoch": 0.07576604786401253,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 5.9768,
|
|
"mean_token_accuracy": 0.13976927995681762,
|
|
"num_tokens": 2566903.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.858755970001221,
|
|
"epoch": 0.07604562737642585,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 5.7985,
|
|
"mean_token_accuracy": 0.14881170094013213,
|
|
"num_tokens": 2575710.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.898805427551269,
|
|
"epoch": 0.07632520688883919,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 5.8861,
|
|
"mean_token_accuracy": 0.1452362820506096,
|
|
"num_tokens": 2585339.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.832397365570069,
|
|
"epoch": 0.07660478640125251,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 5.8063,
|
|
"mean_token_accuracy": 0.14710194766521453,
|
|
"num_tokens": 2594798.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.9450548648834225,
|
|
"epoch": 0.07688436591366585,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 5.9286,
|
|
"mean_token_accuracy": 0.14188876301050185,
|
|
"num_tokens": 2604587.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.890232038497925,
|
|
"epoch": 0.07716394542607918,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 5.8408,
|
|
"mean_token_accuracy": 0.14413416609168053,
|
|
"num_tokens": 2613820.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.886705827713013,
|
|
"epoch": 0.0774435249384925,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 5.9779,
|
|
"mean_token_accuracy": 0.13227701261639596,
|
|
"num_tokens": 2623078.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.841005849838257,
|
|
"epoch": 0.07772310445090584,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 5.8235,
|
|
"mean_token_accuracy": 0.14832434058189392,
|
|
"num_tokens": 2631705.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.837620830535888,
|
|
"epoch": 0.07800268396331916,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 5.8449,
|
|
"mean_token_accuracy": 0.1419344425201416,
|
|
"num_tokens": 2641402.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.930789947509766,
|
|
"epoch": 0.0782822634757325,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 5.7983,
|
|
"mean_token_accuracy": 0.1435087390244007,
|
|
"num_tokens": 2650968.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.829609727859497,
|
|
"epoch": 0.07856184298814582,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 5.8883,
|
|
"mean_token_accuracy": 0.13864338994026185,
|
|
"num_tokens": 2661333.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.890391159057617,
|
|
"epoch": 0.07884142250055916,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 5.8495,
|
|
"mean_token_accuracy": 0.14318352192640305,
|
|
"num_tokens": 2670586.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.840963411331177,
|
|
"epoch": 0.0791210020129725,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 5.8306,
|
|
"mean_token_accuracy": 0.1473466768860817,
|
|
"num_tokens": 2679889.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.98079400062561,
|
|
"epoch": 0.07940058152538582,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 5.8959,
|
|
"mean_token_accuracy": 0.13912662491202354,
|
|
"num_tokens": 2689643.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.936429357528686,
|
|
"epoch": 0.07968016103779915,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 5.9728,
|
|
"mean_token_accuracy": 0.14272105246782302,
|
|
"num_tokens": 2699224.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.9426342964172365,
|
|
"epoch": 0.07995974055021247,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 5.8742,
|
|
"mean_token_accuracy": 0.14053379148244857,
|
|
"num_tokens": 2709131.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.796779680252075,
|
|
"epoch": 0.08023932006262581,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 5.8105,
|
|
"mean_token_accuracy": 0.14447230845689774,
|
|
"num_tokens": 2717838.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.87654972076416,
|
|
"epoch": 0.08051889957503915,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 5.847,
|
|
"mean_token_accuracy": 0.14702155739068984,
|
|
"num_tokens": 2727231.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.88785490989685,
|
|
"epoch": 0.08079847908745247,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.9088,
|
|
"mean_token_accuracy": 0.14593670815229415,
|
|
"num_tokens": 2736576.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.854083585739136,
|
|
"epoch": 0.0810780585998658,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 5.8635,
|
|
"mean_token_accuracy": 0.14281239062547685,
|
|
"num_tokens": 2746906.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.907067108154297,
|
|
"epoch": 0.08135763811227913,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 5.8008,
|
|
"mean_token_accuracy": 0.1396098867058754,
|
|
"num_tokens": 2755787.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.900883293151855,
|
|
"epoch": 0.08163721762469246,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 5.9196,
|
|
"mean_token_accuracy": 0.13834905624389648,
|
|
"num_tokens": 2765423.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.940469598770141,
|
|
"epoch": 0.0819167971371058,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 5.818,
|
|
"mean_token_accuracy": 0.14593163654208183,
|
|
"num_tokens": 2775086.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.755701112747192,
|
|
"epoch": 0.08219637664951912,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 5.7278,
|
|
"mean_token_accuracy": 0.1530797302722931,
|
|
"num_tokens": 2784520.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.844909381866455,
|
|
"epoch": 0.08247595616193246,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 5.8449,
|
|
"mean_token_accuracy": 0.14310653433203696,
|
|
"num_tokens": 2793368.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.952913284301758,
|
|
"epoch": 0.08275553567434578,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 5.8807,
|
|
"mean_token_accuracy": 0.14319991767406465,
|
|
"num_tokens": 2802657.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.893905401229858,
|
|
"epoch": 0.08303511518675911,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 5.8656,
|
|
"mean_token_accuracy": 0.14389365166425705,
|
|
"num_tokens": 2812678.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.835982084274292,
|
|
"epoch": 0.08331469469917245,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 5.9181,
|
|
"mean_token_accuracy": 0.13842813149094582,
|
|
"num_tokens": 2823131.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.865871286392212,
|
|
"epoch": 0.08359427421158577,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 5.8053,
|
|
"mean_token_accuracy": 0.14218075424432755,
|
|
"num_tokens": 2832582.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.8657115459442135,
|
|
"epoch": 0.08387385372399911,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 5.7647,
|
|
"mean_token_accuracy": 0.14778101369738578,
|
|
"num_tokens": 2841597.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.778578901290894,
|
|
"epoch": 0.08415343323641243,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 5.8359,
|
|
"mean_token_accuracy": 0.1389613464474678,
|
|
"num_tokens": 2851508.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.887820529937744,
|
|
"epoch": 0.08443301274882577,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 5.8892,
|
|
"mean_token_accuracy": 0.14238111600279807,
|
|
"num_tokens": 2861121.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.802965974807739,
|
|
"epoch": 0.0847125922612391,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 5.7951,
|
|
"mean_token_accuracy": 0.14955513179302216,
|
|
"num_tokens": 2870291.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.896052312850952,
|
|
"epoch": 0.08499217177365243,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 5.8097,
|
|
"mean_token_accuracy": 0.15359585136175155,
|
|
"num_tokens": 2879365.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.910254335403442,
|
|
"epoch": 0.08527175128606576,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 5.8504,
|
|
"mean_token_accuracy": 0.14722051918506623,
|
|
"num_tokens": 2888286.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.891468858718872,
|
|
"epoch": 0.08555133079847908,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 5.9127,
|
|
"mean_token_accuracy": 0.13912768736481668,
|
|
"num_tokens": 2898017.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.946372413635254,
|
|
"epoch": 0.08583091031089242,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 5.8,
|
|
"mean_token_accuracy": 0.14936287254095076,
|
|
"num_tokens": 2906578.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.8418206691741945,
|
|
"epoch": 0.08611048982330574,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 5.9016,
|
|
"mean_token_accuracy": 0.13408652767539025,
|
|
"num_tokens": 2917047.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.992599439620972,
|
|
"epoch": 0.08639006933571908,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 5.9989,
|
|
"mean_token_accuracy": 0.13653719276189805,
|
|
"num_tokens": 2927741.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.959636974334717,
|
|
"epoch": 0.08666964884813241,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 5.8823,
|
|
"mean_token_accuracy": 0.14125916436314584,
|
|
"num_tokens": 2937452.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.86145601272583,
|
|
"epoch": 0.08694922836054574,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 5.8177,
|
|
"mean_token_accuracy": 0.1483025297522545,
|
|
"num_tokens": 2947663.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.808240222930908,
|
|
"epoch": 0.08722880787295907,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 5.8278,
|
|
"mean_token_accuracy": 0.14839882552623748,
|
|
"num_tokens": 2958224.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.908622932434082,
|
|
"epoch": 0.0875083873853724,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 5.8006,
|
|
"mean_token_accuracy": 0.1478358820080757,
|
|
"num_tokens": 2968205.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.780447483062744,
|
|
"epoch": 0.08778796689778573,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 5.769,
|
|
"mean_token_accuracy": 0.151074181497097,
|
|
"num_tokens": 2977879.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.773506355285645,
|
|
"epoch": 0.08806754641019907,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 5.7644,
|
|
"mean_token_accuracy": 0.15381836965680123,
|
|
"num_tokens": 2987665.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.749718523025512,
|
|
"epoch": 0.08834712592261239,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 5.7969,
|
|
"mean_token_accuracy": 0.14794880151748657,
|
|
"num_tokens": 2998021.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.815536308288574,
|
|
"epoch": 0.08862670543502572,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 5.7095,
|
|
"mean_token_accuracy": 0.15627864003181458,
|
|
"num_tokens": 3007285.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.7407636642456055,
|
|
"epoch": 0.08890628494743905,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 5.7126,
|
|
"mean_token_accuracy": 0.15217786580324172,
|
|
"num_tokens": 3016910.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.947821712493896,
|
|
"epoch": 0.08918586445985238,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 5.7643,
|
|
"mean_token_accuracy": 0.15313432663679122,
|
|
"num_tokens": 3025522.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.773076343536377,
|
|
"epoch": 0.08946544397226572,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 5.7059,
|
|
"mean_token_accuracy": 0.1459415040910244,
|
|
"num_tokens": 3034716.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.861837100982666,
|
|
"epoch": 0.08974502348467904,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 5.8378,
|
|
"mean_token_accuracy": 0.14212724566459656,
|
|
"num_tokens": 3045151.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.750861263275146,
|
|
"epoch": 0.09002460299709238,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 5.8061,
|
|
"mean_token_accuracy": 0.14961824119091033,
|
|
"num_tokens": 3055292.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.882100391387939,
|
|
"epoch": 0.0903041825095057,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 5.7856,
|
|
"mean_token_accuracy": 0.1557572826743126,
|
|
"num_tokens": 3065025.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.795562601089477,
|
|
"epoch": 0.09058376202191903,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 5.7659,
|
|
"mean_token_accuracy": 0.1465141385793686,
|
|
"num_tokens": 3074908.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.670252180099487,
|
|
"epoch": 0.09086334153433237,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 5.6489,
|
|
"mean_token_accuracy": 0.15595338642597198,
|
|
"num_tokens": 3083416.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.859638977050781,
|
|
"epoch": 0.09114292104674569,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 5.7992,
|
|
"mean_token_accuracy": 0.14463712871074677,
|
|
"num_tokens": 3093594.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.8232039451599125,
|
|
"epoch": 0.09142250055915903,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 5.735,
|
|
"mean_token_accuracy": 0.15327433943748475,
|
|
"num_tokens": 3103126.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.77016806602478,
|
|
"epoch": 0.09170208007157235,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 5.6744,
|
|
"mean_token_accuracy": 0.14766955971717835,
|
|
"num_tokens": 3113077.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.687986946105957,
|
|
"epoch": 0.09198165958398569,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 5.7471,
|
|
"mean_token_accuracy": 0.1517716571688652,
|
|
"num_tokens": 3122341.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.802947998046875,
|
|
"epoch": 0.09226123909639902,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 5.7621,
|
|
"mean_token_accuracy": 0.15780353248119355,
|
|
"num_tokens": 3131257.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.762443494796753,
|
|
"epoch": 0.09254081860881234,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 5.7155,
|
|
"mean_token_accuracy": 0.14690385460853578,
|
|
"num_tokens": 3141108.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.803344583511352,
|
|
"epoch": 0.09282039812122568,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 5.6595,
|
|
"mean_token_accuracy": 0.1600823998451233,
|
|
"num_tokens": 3149424.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.783663368225097,
|
|
"epoch": 0.093099977633639,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 5.7027,
|
|
"mean_token_accuracy": 0.15383269190788268,
|
|
"num_tokens": 3158979.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.66107144355774,
|
|
"epoch": 0.09337955714605234,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 5.6317,
|
|
"mean_token_accuracy": 0.15192658454179764,
|
|
"num_tokens": 3168588.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.772243356704712,
|
|
"epoch": 0.09365913665846566,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 5.6617,
|
|
"mean_token_accuracy": 0.159200818836689,
|
|
"num_tokens": 3177914.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.6689084529876705,
|
|
"epoch": 0.093938716170879,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 5.6486,
|
|
"mean_token_accuracy": 0.15601246953010559,
|
|
"num_tokens": 3186290.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.85229721069336,
|
|
"epoch": 0.09421829568329233,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 5.7504,
|
|
"mean_token_accuracy": 0.14570991545915604,
|
|
"num_tokens": 3196094.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.757496166229248,
|
|
"epoch": 0.09449787519570566,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 5.7662,
|
|
"mean_token_accuracy": 0.15119355022907258,
|
|
"num_tokens": 3205508.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.775554513931274,
|
|
"epoch": 0.09477745470811899,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 5.739,
|
|
"mean_token_accuracy": 0.15416313484311103,
|
|
"num_tokens": 3215426.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.8277966499328615,
|
|
"epoch": 0.09505703422053231,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 5.744,
|
|
"mean_token_accuracy": 0.14643914997577667,
|
|
"num_tokens": 3225391.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.747729921340943,
|
|
"epoch": 0.09533661373294565,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 5.721,
|
|
"mean_token_accuracy": 0.1518389865756035,
|
|
"num_tokens": 3235181.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.7080474376678465,
|
|
"epoch": 0.09561619324535899,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 5.7223,
|
|
"mean_token_accuracy": 0.15273163020610808,
|
|
"num_tokens": 3244743.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.901181077957153,
|
|
"epoch": 0.09589577275777231,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 5.751,
|
|
"mean_token_accuracy": 0.1508289471268654,
|
|
"num_tokens": 3255249.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.6971009254455565,
|
|
"epoch": 0.09617535227018564,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 5.7391,
|
|
"mean_token_accuracy": 0.15187209397554396,
|
|
"num_tokens": 3265254.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.836921644210816,
|
|
"epoch": 0.09645493178259897,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 5.799,
|
|
"mean_token_accuracy": 0.1473428189754486,
|
|
"num_tokens": 3274666.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.798277187347412,
|
|
"epoch": 0.0967345112950123,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 5.6828,
|
|
"mean_token_accuracy": 0.15016689151525497,
|
|
"num_tokens": 3284018.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.6422511577606205,
|
|
"epoch": 0.09701409080742564,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 5.7661,
|
|
"mean_token_accuracy": 0.1510323852300644,
|
|
"num_tokens": 3293734.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.922972965240478,
|
|
"epoch": 0.09729367031983896,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 5.8088,
|
|
"mean_token_accuracy": 0.148889422416687,
|
|
"num_tokens": 3303791.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.819644165039063,
|
|
"epoch": 0.0975732498322523,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 5.785,
|
|
"mean_token_accuracy": 0.14719584658741952,
|
|
"num_tokens": 3313693.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.734876728057861,
|
|
"epoch": 0.09785282934466562,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 5.7118,
|
|
"mean_token_accuracy": 0.15313938707113267,
|
|
"num_tokens": 3322565.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.739667510986328,
|
|
"epoch": 0.09813240885707895,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 5.6721,
|
|
"mean_token_accuracy": 0.15636088252067565,
|
|
"num_tokens": 3332094.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.778171157836914,
|
|
"epoch": 0.09841198836949229,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 5.7342,
|
|
"mean_token_accuracy": 0.15363333374261856,
|
|
"num_tokens": 3341984.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.698805952072144,
|
|
"epoch": 0.09869156788190561,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 5.6939,
|
|
"mean_token_accuracy": 0.15260056406259537,
|
|
"num_tokens": 3351694.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.664214658737182,
|
|
"epoch": 0.09897114739431895,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 5.5926,
|
|
"mean_token_accuracy": 0.15653923898935318,
|
|
"num_tokens": 3360841.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.877408599853515,
|
|
"epoch": 0.09925072690673227,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 5.7648,
|
|
"mean_token_accuracy": 0.15017842650413513,
|
|
"num_tokens": 3369774.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.7866393566131595,
|
|
"epoch": 0.0995303064191456,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 5.6508,
|
|
"mean_token_accuracy": 0.1622912108898163,
|
|
"num_tokens": 3378864.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.712972736358642,
|
|
"epoch": 0.09980988593155894,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 5.6368,
|
|
"mean_token_accuracy": 0.16099814474582672,
|
|
"num_tokens": 3387595.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.658095073699951,
|
|
"epoch": 0.10008946544397226,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 5.5784,
|
|
"mean_token_accuracy": 0.1671353504061699,
|
|
"num_tokens": 3396333.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.764716148376465,
|
|
"epoch": 0.1003690449563856,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 5.77,
|
|
"mean_token_accuracy": 0.14844436645507814,
|
|
"num_tokens": 3406576.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.819981527328491,
|
|
"epoch": 0.10064862446879892,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 5.7424,
|
|
"mean_token_accuracy": 0.15551482290029525,
|
|
"num_tokens": 3415736.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.705848598480225,
|
|
"epoch": 0.10092820398121226,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 5.721,
|
|
"mean_token_accuracy": 0.15223492980003356,
|
|
"num_tokens": 3425068.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.768614912033081,
|
|
"epoch": 0.10120778349362558,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 5.6783,
|
|
"mean_token_accuracy": 0.1512942187488079,
|
|
"num_tokens": 3433782.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.6993088722229,
|
|
"epoch": 0.10148736300603892,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 5.6,
|
|
"mean_token_accuracy": 0.16176195442676544,
|
|
"num_tokens": 3442111.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.712747955322266,
|
|
"epoch": 0.10176694251845225,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 5.6413,
|
|
"mean_token_accuracy": 0.15280646234750747,
|
|
"num_tokens": 3451783.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.765063810348511,
|
|
"epoch": 0.10204652203086557,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 5.7592,
|
|
"mean_token_accuracy": 0.15333681851625441,
|
|
"num_tokens": 3461523.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.766888618469238,
|
|
"epoch": 0.10232610154327891,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 5.7204,
|
|
"mean_token_accuracy": 0.15032942295074464,
|
|
"num_tokens": 3471107.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.648005771636963,
|
|
"epoch": 0.10260568105569223,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 5.5929,
|
|
"mean_token_accuracy": 0.15612752139568328,
|
|
"num_tokens": 3480729.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.757135915756225,
|
|
"epoch": 0.10288526056810557,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 5.721,
|
|
"mean_token_accuracy": 0.1431248515844345,
|
|
"num_tokens": 3489804.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.633577156066894,
|
|
"epoch": 0.1031648400805189,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 5.6349,
|
|
"mean_token_accuracy": 0.16112861186265945,
|
|
"num_tokens": 3498656.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.899571895599365,
|
|
"epoch": 0.10344441959293223,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 5.7113,
|
|
"mean_token_accuracy": 0.1511174336075783,
|
|
"num_tokens": 3508855.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.614178609848023,
|
|
"epoch": 0.10372399910534556,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 5.669,
|
|
"mean_token_accuracy": 0.14884379357099534,
|
|
"num_tokens": 3518044.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.808314085006714,
|
|
"epoch": 0.10400357861775889,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 5.7314,
|
|
"mean_token_accuracy": 0.1485123947262764,
|
|
"num_tokens": 3527577.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.624914026260376,
|
|
"epoch": 0.10428315813017222,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 5.5912,
|
|
"mean_token_accuracy": 0.16450488716363906,
|
|
"num_tokens": 3536503.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.813652181625367,
|
|
"epoch": 0.10456273764258556,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 5.6957,
|
|
"mean_token_accuracy": 0.15247510448098184,
|
|
"num_tokens": 3546335.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.7023646354675295,
|
|
"epoch": 0.10484231715499888,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 5.6189,
|
|
"mean_token_accuracy": 0.154330775141716,
|
|
"num_tokens": 3556231.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.7131105899810795,
|
|
"epoch": 0.10512189666741222,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 5.7052,
|
|
"mean_token_accuracy": 0.1487670697271824,
|
|
"num_tokens": 3566232.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.74267086982727,
|
|
"epoch": 0.10540147617982554,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 5.6533,
|
|
"mean_token_accuracy": 0.15348853915929794,
|
|
"num_tokens": 3576475.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.588758516311645,
|
|
"epoch": 0.10568105569223887,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 5.5492,
|
|
"mean_token_accuracy": 0.15948327928781508,
|
|
"num_tokens": 3586412.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.746132516860962,
|
|
"epoch": 0.10596063520465221,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 5.6118,
|
|
"mean_token_accuracy": 0.16162005215883254,
|
|
"num_tokens": 3595358.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.791898727416992,
|
|
"epoch": 0.10624021471706553,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 5.7363,
|
|
"mean_token_accuracy": 0.15219928324222565,
|
|
"num_tokens": 3604970.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.720702505111694,
|
|
"epoch": 0.10651979422947887,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 5.6326,
|
|
"mean_token_accuracy": 0.15265588462352753,
|
|
"num_tokens": 3614550.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.790266990661621,
|
|
"epoch": 0.10679937374189219,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 5.6521,
|
|
"mean_token_accuracy": 0.1665974199771881,
|
|
"num_tokens": 3623289.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.721146583557129,
|
|
"epoch": 0.10707895325430553,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 5.6523,
|
|
"mean_token_accuracy": 0.15115113109350203,
|
|
"num_tokens": 3633273.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.602410936355591,
|
|
"epoch": 0.10735853276671886,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 5.6201,
|
|
"mean_token_accuracy": 0.1511617824435234,
|
|
"num_tokens": 3642622.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.803311014175415,
|
|
"epoch": 0.10763811227913218,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 5.6512,
|
|
"mean_token_accuracy": 0.14831936061382295,
|
|
"num_tokens": 3652449.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.747519779205322,
|
|
"epoch": 0.10791769179154552,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 5.6855,
|
|
"mean_token_accuracy": 0.15171786025166512,
|
|
"num_tokens": 3661963.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.658705377578736,
|
|
"epoch": 0.10819727130395884,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 5.7448,
|
|
"mean_token_accuracy": 0.1576741471886635,
|
|
"num_tokens": 3670598.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.839505052566528,
|
|
"epoch": 0.10847685081637218,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 5.6242,
|
|
"mean_token_accuracy": 0.1517106130719185,
|
|
"num_tokens": 3680006.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.620665264129639,
|
|
"epoch": 0.1087564303287855,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 5.661,
|
|
"mean_token_accuracy": 0.15248942077159883,
|
|
"num_tokens": 3689632.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.734098815917969,
|
|
"epoch": 0.10903600984119884,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 5.6859,
|
|
"mean_token_accuracy": 0.1514577716588974,
|
|
"num_tokens": 3700500.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.748064470291138,
|
|
"epoch": 0.10931558935361217,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 5.6907,
|
|
"mean_token_accuracy": 0.1561550237238407,
|
|
"num_tokens": 3710158.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.755768823623657,
|
|
"epoch": 0.1095951688660255,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 5.5763,
|
|
"mean_token_accuracy": 0.1547596700489521,
|
|
"num_tokens": 3719360.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.7060223579406735,
|
|
"epoch": 0.10987474837843883,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 5.5736,
|
|
"mean_token_accuracy": 0.16160024255514144,
|
|
"num_tokens": 3728776.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.5924194812774655,
|
|
"epoch": 0.11015432789085215,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 5.5894,
|
|
"mean_token_accuracy": 0.15759295225143433,
|
|
"num_tokens": 3738619.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.674139022827148,
|
|
"epoch": 0.11043390740326549,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 5.6345,
|
|
"mean_token_accuracy": 0.1570148304104805,
|
|
"num_tokens": 3747449.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.726358413696289,
|
|
"epoch": 0.11071348691567882,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 5.6917,
|
|
"mean_token_accuracy": 0.15671592950820923,
|
|
"num_tokens": 3757382.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.7088942527771,
|
|
"epoch": 0.11099306642809215,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 5.697,
|
|
"mean_token_accuracy": 0.15646909177303314,
|
|
"num_tokens": 3766970.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.66936674118042,
|
|
"epoch": 0.11127264594050548,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 5.5237,
|
|
"mean_token_accuracy": 0.16630659401416778,
|
|
"num_tokens": 3776203.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.778311347961425,
|
|
"epoch": 0.1115522254529188,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 5.6775,
|
|
"mean_token_accuracy": 0.15549499839544295,
|
|
"num_tokens": 3786054.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.634786367416382,
|
|
"epoch": 0.11183180496533214,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 5.6485,
|
|
"mean_token_accuracy": 0.16290906965732574,
|
|
"num_tokens": 3796203.0,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"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": 5565144674304000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|