2035 lines
54 KiB
JSON
2035 lines
54 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 0.05591590248266607,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 1000,
|
||
|
|
"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
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"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": 2784189081600000.0,
|
||
|
|
"train_batch_size": 16,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|