2057 lines
54 KiB
JSON
2057 lines
54 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 0.9945300845350572,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 1000,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"entropy": 10.69195852279663,
|
||
|
|
"epoch": 0.004972650422675286,
|
||
|
|
"grad_norm": 13.0,
|
||
|
|
"learning_rate": 2e-06,
|
||
|
|
"loss": 10.8154,
|
||
|
|
"mean_token_accuracy": 0.0,
|
||
|
|
"num_tokens": 9089.0,
|
||
|
|
"step": 5
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.69191370010376,
|
||
|
|
"epoch": 0.009945300845350571,
|
||
|
|
"grad_norm": 12.4375,
|
||
|
|
"learning_rate": 4.5e-06,
|
||
|
|
"loss": 10.8258,
|
||
|
|
"mean_token_accuracy": 0.0,
|
||
|
|
"num_tokens": 21650.0,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.69098253250122,
|
||
|
|
"epoch": 0.014917951268025857,
|
||
|
|
"grad_norm": 9.625,
|
||
|
|
"learning_rate": 7e-06,
|
||
|
|
"loss": 10.5784,
|
||
|
|
"mean_token_accuracy": 0.020547608099877834,
|
||
|
|
"num_tokens": 31067.0,
|
||
|
|
"step": 15
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.673026275634765,
|
||
|
|
"epoch": 0.019890601690701143,
|
||
|
|
"grad_norm": 6.40625,
|
||
|
|
"learning_rate": 9.5e-06,
|
||
|
|
"loss": 10.3334,
|
||
|
|
"mean_token_accuracy": 0.041073380038142204,
|
||
|
|
"num_tokens": 40816.0,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.540556049346923,
|
||
|
|
"epoch": 0.02486325211337643,
|
||
|
|
"grad_norm": 4.28125,
|
||
|
|
"learning_rate": 1.2e-05,
|
||
|
|
"loss": 10.1325,
|
||
|
|
"mean_token_accuracy": 0.0424393804743886,
|
||
|
|
"num_tokens": 51065.0,
|
||
|
|
"step": 25
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.523736000061035,
|
||
|
|
"epoch": 0.029835902536051714,
|
||
|
|
"grad_norm": 3.359375,
|
||
|
|
"learning_rate": 1.4500000000000002e-05,
|
||
|
|
"loss": 9.9451,
|
||
|
|
"mean_token_accuracy": 0.044780130684375766,
|
||
|
|
"num_tokens": 60539.0,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.482940578460694,
|
||
|
|
"epoch": 0.034808552958727,
|
||
|
|
"grad_norm": 2.484375,
|
||
|
|
"learning_rate": 1.7000000000000003e-05,
|
||
|
|
"loss": 9.9267,
|
||
|
|
"mean_token_accuracy": 0.04583238400518894,
|
||
|
|
"num_tokens": 71010.0,
|
||
|
|
"step": 35
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.497202396392822,
|
||
|
|
"epoch": 0.039781203381402286,
|
||
|
|
"grad_norm": 2.453125,
|
||
|
|
"learning_rate": 1.95e-05,
|
||
|
|
"loss": 9.8766,
|
||
|
|
"mean_token_accuracy": 0.04677955023944378,
|
||
|
|
"num_tokens": 81394.0,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.495645523071289,
|
||
|
|
"epoch": 0.04475385380407757,
|
||
|
|
"grad_norm": 2.390625,
|
||
|
|
"learning_rate": 2.2e-05,
|
||
|
|
"loss": 9.9242,
|
||
|
|
"mean_token_accuracy": 0.04332513865083456,
|
||
|
|
"num_tokens": 92491.0,
|
||
|
|
"step": 45
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.476744461059571,
|
||
|
|
"epoch": 0.04972650422675286,
|
||
|
|
"grad_norm": 2.265625,
|
||
|
|
"learning_rate": 2.4500000000000003e-05,
|
||
|
|
"loss": 9.8095,
|
||
|
|
"mean_token_accuracy": 0.045694665610790254,
|
||
|
|
"num_tokens": 102113.0,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.468534755706788,
|
||
|
|
"epoch": 0.05469915464942814,
|
||
|
|
"grad_norm": 3.109375,
|
||
|
|
"learning_rate": 2.7e-05,
|
||
|
|
"loss": 9.8159,
|
||
|
|
"mean_token_accuracy": 0.051046956703066824,
|
||
|
|
"num_tokens": 112211.0,
|
||
|
|
"step": 55
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.431697177886964,
|
||
|
|
"epoch": 0.05967180507210343,
|
||
|
|
"grad_norm": 1.8203125,
|
||
|
|
"learning_rate": 2.95e-05,
|
||
|
|
"loss": 9.7102,
|
||
|
|
"mean_token_accuracy": 0.05394814908504486,
|
||
|
|
"num_tokens": 122166.0,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.248080635070801,
|
||
|
|
"epoch": 0.06464445549477872,
|
||
|
|
"grad_norm": 2.125,
|
||
|
|
"learning_rate": 3.2e-05,
|
||
|
|
"loss": 9.6181,
|
||
|
|
"mean_token_accuracy": 0.059726152941584586,
|
||
|
|
"num_tokens": 131016.0,
|
||
|
|
"step": 65
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.313014602661132,
|
||
|
|
"epoch": 0.069617105917454,
|
||
|
|
"grad_norm": 1.984375,
|
||
|
|
"learning_rate": 3.4500000000000005e-05,
|
||
|
|
"loss": 9.587,
|
||
|
|
"mean_token_accuracy": 0.06370560713112354,
|
||
|
|
"num_tokens": 141369.0,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.329190254211426,
|
||
|
|
"epoch": 0.07458975634012929,
|
||
|
|
"grad_norm": 1.9296875,
|
||
|
|
"learning_rate": 3.7e-05,
|
||
|
|
"loss": 9.5355,
|
||
|
|
"mean_token_accuracy": 0.058842772617936134,
|
||
|
|
"num_tokens": 150939.0,
|
||
|
|
"step": 75
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.338141822814942,
|
||
|
|
"epoch": 0.07956240676280457,
|
||
|
|
"grad_norm": 1.890625,
|
||
|
|
"learning_rate": 3.95e-05,
|
||
|
|
"loss": 9.478,
|
||
|
|
"mean_token_accuracy": 0.05786029174923897,
|
||
|
|
"num_tokens": 159981.0,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.291361045837402,
|
||
|
|
"epoch": 0.08453505718547986,
|
||
|
|
"grad_norm": 1.765625,
|
||
|
|
"learning_rate": 4.2000000000000004e-05,
|
||
|
|
"loss": 9.4043,
|
||
|
|
"mean_token_accuracy": 0.05959840677678585,
|
||
|
|
"num_tokens": 170139.0,
|
||
|
|
"step": 85
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.275171947479247,
|
||
|
|
"epoch": 0.08950770760815514,
|
||
|
|
"grad_norm": 1.8046875,
|
||
|
|
"learning_rate": 4.45e-05,
|
||
|
|
"loss": 9.3294,
|
||
|
|
"mean_token_accuracy": 0.06485622599720955,
|
||
|
|
"num_tokens": 180746.0,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.15320930480957,
|
||
|
|
"epoch": 0.09448035803083044,
|
||
|
|
"grad_norm": 1.6328125,
|
||
|
|
"learning_rate": 4.7000000000000004e-05,
|
||
|
|
"loss": 9.1862,
|
||
|
|
"mean_token_accuracy": 0.06401227861642837,
|
||
|
|
"num_tokens": 190250.0,
|
||
|
|
"step": 95
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.223065280914307,
|
||
|
|
"epoch": 0.09945300845350571,
|
||
|
|
"grad_norm": 1.8046875,
|
||
|
|
"learning_rate": 4.9500000000000004e-05,
|
||
|
|
"loss": 9.2406,
|
||
|
|
"mean_token_accuracy": 0.061084074527025224,
|
||
|
|
"num_tokens": 201185.0,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.13264331817627,
|
||
|
|
"epoch": 0.10442565887618101,
|
||
|
|
"grad_norm": 1.46875,
|
||
|
|
"learning_rate": 5.2e-05,
|
||
|
|
"loss": 9.16,
|
||
|
|
"mean_token_accuracy": 0.05885303355753422,
|
||
|
|
"num_tokens": 211476.0,
|
||
|
|
"step": 105
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.12833375930786,
|
||
|
|
"epoch": 0.10939830929885629,
|
||
|
|
"grad_norm": 1.4140625,
|
||
|
|
"learning_rate": 5.45e-05,
|
||
|
|
"loss": 9.0533,
|
||
|
|
"mean_token_accuracy": 0.062058281153440475,
|
||
|
|
"num_tokens": 220965.0,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.080890655517578,
|
||
|
|
"epoch": 0.11437095972153158,
|
||
|
|
"grad_norm": 1.4140625,
|
||
|
|
"learning_rate": 5.7e-05,
|
||
|
|
"loss": 8.9846,
|
||
|
|
"mean_token_accuracy": 0.05859192833304405,
|
||
|
|
"num_tokens": 231539.0,
|
||
|
|
"step": 115
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.94415922164917,
|
||
|
|
"epoch": 0.11934361014420686,
|
||
|
|
"grad_norm": 1.3671875,
|
||
|
|
"learning_rate": 5.9499999999999996e-05,
|
||
|
|
"loss": 8.881,
|
||
|
|
"mean_token_accuracy": 0.06372438780963421,
|
||
|
|
"num_tokens": 241206.0,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.881352806091309,
|
||
|
|
"epoch": 0.12431626056688215,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 6.2e-05,
|
||
|
|
"loss": 8.7735,
|
||
|
|
"mean_token_accuracy": 0.06269489601254463,
|
||
|
|
"num_tokens": 250629.0,
|
||
|
|
"step": 125
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.732184791564942,
|
||
|
|
"epoch": 0.12928891098955744,
|
||
|
|
"grad_norm": 1.2734375,
|
||
|
|
"learning_rate": 6.450000000000001e-05,
|
||
|
|
"loss": 8.7496,
|
||
|
|
"mean_token_accuracy": 0.05945408381521702,
|
||
|
|
"num_tokens": 260705.0,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.715274715423584,
|
||
|
|
"epoch": 0.13426156141223272,
|
||
|
|
"grad_norm": 1.5234375,
|
||
|
|
"learning_rate": 6.7e-05,
|
||
|
|
"loss": 8.6842,
|
||
|
|
"mean_token_accuracy": 0.05957487896084786,
|
||
|
|
"num_tokens": 270499.0,
|
||
|
|
"step": 135
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.532818031311034,
|
||
|
|
"epoch": 0.139234211834908,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 6.950000000000001e-05,
|
||
|
|
"loss": 8.6647,
|
||
|
|
"mean_token_accuracy": 0.05812281183898449,
|
||
|
|
"num_tokens": 279998.0,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.431089496612548,
|
||
|
|
"epoch": 0.14420686225758328,
|
||
|
|
"grad_norm": 1.4140625,
|
||
|
|
"learning_rate": 7.2e-05,
|
||
|
|
"loss": 8.5342,
|
||
|
|
"mean_token_accuracy": 0.06715737730264663,
|
||
|
|
"num_tokens": 290322.0,
|
||
|
|
"step": 145
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.28282127380371,
|
||
|
|
"epoch": 0.14917951268025859,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 7.45e-05,
|
||
|
|
"loss": 8.4927,
|
||
|
|
"mean_token_accuracy": 0.06783988289535045,
|
||
|
|
"num_tokens": 299893.0,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.202215671539307,
|
||
|
|
"epoch": 0.15415216310293386,
|
||
|
|
"grad_norm": 1.2890625,
|
||
|
|
"learning_rate": 7.7e-05,
|
||
|
|
"loss": 8.4742,
|
||
|
|
"mean_token_accuracy": 0.06662740968167782,
|
||
|
|
"num_tokens": 309191.0,
|
||
|
|
"step": 155
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.073563289642333,
|
||
|
|
"epoch": 0.15912481352560914,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 7.950000000000001e-05,
|
||
|
|
"loss": 8.4087,
|
||
|
|
"mean_token_accuracy": 0.06572333909571171,
|
||
|
|
"num_tokens": 318494.0,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.005878067016601,
|
||
|
|
"epoch": 0.16409746394828442,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 8.2e-05,
|
||
|
|
"loss": 8.3906,
|
||
|
|
"mean_token_accuracy": 0.06560419239103794,
|
||
|
|
"num_tokens": 328299.0,
|
||
|
|
"step": 165
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.962747955322266,
|
||
|
|
"epoch": 0.16907011437095973,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 8.450000000000001e-05,
|
||
|
|
"loss": 8.4667,
|
||
|
|
"mean_token_accuracy": 0.06214671768248081,
|
||
|
|
"num_tokens": 338878.0,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.892442321777343,
|
||
|
|
"epoch": 0.174042764793635,
|
||
|
|
"grad_norm": 1.4921875,
|
||
|
|
"learning_rate": 8.7e-05,
|
||
|
|
"loss": 8.3855,
|
||
|
|
"mean_token_accuracy": 0.06676384881138801,
|
||
|
|
"num_tokens": 348508.0,
|
||
|
|
"step": 175
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.82924509048462,
|
||
|
|
"epoch": 0.1790154152163103,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 8.95e-05,
|
||
|
|
"loss": 8.3408,
|
||
|
|
"mean_token_accuracy": 0.06486317589879036,
|
||
|
|
"num_tokens": 358233.0,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.784028911590577,
|
||
|
|
"epoch": 0.1839880656389856,
|
||
|
|
"grad_norm": 1.3125,
|
||
|
|
"learning_rate": 9.2e-05,
|
||
|
|
"loss": 8.2883,
|
||
|
|
"mean_token_accuracy": 0.06572877354919911,
|
||
|
|
"num_tokens": 367297.0,
|
||
|
|
"step": 185
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.75277099609375,
|
||
|
|
"epoch": 0.18896071606166087,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 9.45e-05,
|
||
|
|
"loss": 8.2958,
|
||
|
|
"mean_token_accuracy": 0.06614930555224419,
|
||
|
|
"num_tokens": 376880.0,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.710809421539306,
|
||
|
|
"epoch": 0.19393336648433615,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 9.7e-05,
|
||
|
|
"loss": 8.2937,
|
||
|
|
"mean_token_accuracy": 0.06907838135957718,
|
||
|
|
"num_tokens": 386806.0,
|
||
|
|
"step": 195
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.63250789642334,
|
||
|
|
"epoch": 0.19890601690701143,
|
||
|
|
"grad_norm": 1.3671875,
|
||
|
|
"learning_rate": 9.95e-05,
|
||
|
|
"loss": 8.257,
|
||
|
|
"mean_token_accuracy": 0.07009859867393971,
|
||
|
|
"num_tokens": 396161.0,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.667188930511475,
|
||
|
|
"epoch": 0.20387866732968674,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.000102,
|
||
|
|
"loss": 8.2966,
|
||
|
|
"mean_token_accuracy": 0.0686015360057354,
|
||
|
|
"num_tokens": 405643.0,
|
||
|
|
"step": 205
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.586182785034179,
|
||
|
|
"epoch": 0.20885131775236201,
|
||
|
|
"grad_norm": 1.3671875,
|
||
|
|
"learning_rate": 0.00010449999999999999,
|
||
|
|
"loss": 8.249,
|
||
|
|
"mean_token_accuracy": 0.07686658278107643,
|
||
|
|
"num_tokens": 414367.0,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.671557807922364,
|
||
|
|
"epoch": 0.2138239681750373,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 0.000107,
|
||
|
|
"loss": 8.3009,
|
||
|
|
"mean_token_accuracy": 0.07021823003888131,
|
||
|
|
"num_tokens": 424430.0,
|
||
|
|
"step": 215
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.60766077041626,
|
||
|
|
"epoch": 0.21879661859771257,
|
||
|
|
"grad_norm": 1.3515625,
|
||
|
|
"learning_rate": 0.0001095,
|
||
|
|
"loss": 8.2723,
|
||
|
|
"mean_token_accuracy": 0.07040721997618675,
|
||
|
|
"num_tokens": 435176.0,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.483222579956054,
|
||
|
|
"epoch": 0.22376926902038788,
|
||
|
|
"grad_norm": 1.1640625,
|
||
|
|
"learning_rate": 0.000112,
|
||
|
|
"loss": 8.2963,
|
||
|
|
"mean_token_accuracy": 0.07160340994596481,
|
||
|
|
"num_tokens": 445611.0,
|
||
|
|
"step": 225
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.615241146087646,
|
||
|
|
"epoch": 0.22874191944306316,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.0001145,
|
||
|
|
"loss": 8.2193,
|
||
|
|
"mean_token_accuracy": 0.07408608645200729,
|
||
|
|
"num_tokens": 455143.0,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.600674724578857,
|
||
|
|
"epoch": 0.23371456986573844,
|
||
|
|
"grad_norm": 1.796875,
|
||
|
|
"learning_rate": 0.00011700000000000001,
|
||
|
|
"loss": 8.2898,
|
||
|
|
"mean_token_accuracy": 0.07503409050405026,
|
||
|
|
"num_tokens": 465507.0,
|
||
|
|
"step": 235
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.539074230194093,
|
||
|
|
"epoch": 0.23868722028841372,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.00011949999999999999,
|
||
|
|
"loss": 8.3078,
|
||
|
|
"mean_token_accuracy": 0.0719807006418705,
|
||
|
|
"num_tokens": 476152.0,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.533562564849854,
|
||
|
|
"epoch": 0.24365987071108902,
|
||
|
|
"grad_norm": 0.98046875,
|
||
|
|
"learning_rate": 0.000122,
|
||
|
|
"loss": 8.1356,
|
||
|
|
"mean_token_accuracy": 0.07810752168297767,
|
||
|
|
"num_tokens": 486341.0,
|
||
|
|
"step": 245
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.470274734497071,
|
||
|
|
"epoch": 0.2486325211337643,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.0001245,
|
||
|
|
"loss": 8.0909,
|
||
|
|
"mean_token_accuracy": 0.07914422042667865,
|
||
|
|
"num_tokens": 495915.0,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.434992504119872,
|
||
|
|
"epoch": 0.2536051715564396,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.000127,
|
||
|
|
"loss": 8.1795,
|
||
|
|
"mean_token_accuracy": 0.07838663607835769,
|
||
|
|
"num_tokens": 505666.0,
|
||
|
|
"step": 255
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.52521858215332,
|
||
|
|
"epoch": 0.2585778219791149,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.0001295,
|
||
|
|
"loss": 8.0848,
|
||
|
|
"mean_token_accuracy": 0.07590681686997414,
|
||
|
|
"num_tokens": 515411.0,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.440220355987549,
|
||
|
|
"epoch": 0.26355047240179014,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 0.000132,
|
||
|
|
"loss": 8.1808,
|
||
|
|
"mean_token_accuracy": 0.07005526758730411,
|
||
|
|
"num_tokens": 525018.0,
|
||
|
|
"step": 265
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.485704612731933,
|
||
|
|
"epoch": 0.26852312282446544,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.00013450000000000002,
|
||
|
|
"loss": 8.132,
|
||
|
|
"mean_token_accuracy": 0.07571323961019516,
|
||
|
|
"num_tokens": 535056.0,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.444093799591064,
|
||
|
|
"epoch": 0.27349577324714075,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.00013700000000000002,
|
||
|
|
"loss": 8.0943,
|
||
|
|
"mean_token_accuracy": 0.07863550931215287,
|
||
|
|
"num_tokens": 545115.0,
|
||
|
|
"step": 275
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.46326379776001,
|
||
|
|
"epoch": 0.278468423669816,
|
||
|
|
"grad_norm": 1.3046875,
|
||
|
|
"learning_rate": 0.0001395,
|
||
|
|
"loss": 8.1362,
|
||
|
|
"mean_token_accuracy": 0.07623238041996956,
|
||
|
|
"num_tokens": 554429.0,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.404027462005615,
|
||
|
|
"epoch": 0.2834410740924913,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.00014199999999999998,
|
||
|
|
"loss": 8.0743,
|
||
|
|
"mean_token_accuracy": 0.0847582258284092,
|
||
|
|
"num_tokens": 563357.0,
|
||
|
|
"step": 285
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.44946584701538,
|
||
|
|
"epoch": 0.28841372451516656,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.0001445,
|
||
|
|
"loss": 8.114,
|
||
|
|
"mean_token_accuracy": 0.07873134687542915,
|
||
|
|
"num_tokens": 574480.0,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.350453853607178,
|
||
|
|
"epoch": 0.29338637493784187,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.000147,
|
||
|
|
"loss": 8.1023,
|
||
|
|
"mean_token_accuracy": 0.07694951295852662,
|
||
|
|
"num_tokens": 583722.0,
|
||
|
|
"step": 295
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.420344829559326,
|
||
|
|
"epoch": 0.29835902536051717,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.0001495,
|
||
|
|
"loss": 8.0479,
|
||
|
|
"mean_token_accuracy": 0.08190468847751617,
|
||
|
|
"num_tokens": 593644.0,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.371365547180176,
|
||
|
|
"epoch": 0.3033316757831924,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.000152,
|
||
|
|
"loss": 8.1017,
|
||
|
|
"mean_token_accuracy": 0.07948060184717179,
|
||
|
|
"num_tokens": 603547.0,
|
||
|
|
"step": 305
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.302706527709962,
|
||
|
|
"epoch": 0.30830432620586773,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.00015450000000000001,
|
||
|
|
"loss": 8.1257,
|
||
|
|
"mean_token_accuracy": 0.07789697460830211,
|
||
|
|
"num_tokens": 613880.0,
|
||
|
|
"step": 310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.415989398956299,
|
||
|
|
"epoch": 0.31327697662854304,
|
||
|
|
"grad_norm": 1.328125,
|
||
|
|
"learning_rate": 0.000157,
|
||
|
|
"loss": 8.0974,
|
||
|
|
"mean_token_accuracy": 0.07658020406961441,
|
||
|
|
"num_tokens": 624213.0,
|
||
|
|
"step": 315
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.416389560699463,
|
||
|
|
"epoch": 0.3182496270512183,
|
||
|
|
"grad_norm": 1.2734375,
|
||
|
|
"learning_rate": 0.0001595,
|
||
|
|
"loss": 8.128,
|
||
|
|
"mean_token_accuracy": 0.08172445297241211,
|
||
|
|
"num_tokens": 634350.0,
|
||
|
|
"step": 320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.42117624282837,
|
||
|
|
"epoch": 0.3232222774738936,
|
||
|
|
"grad_norm": 1.390625,
|
||
|
|
"learning_rate": 0.000162,
|
||
|
|
"loss": 8.1272,
|
||
|
|
"mean_token_accuracy": 0.08092540055513382,
|
||
|
|
"num_tokens": 644040.0,
|
||
|
|
"step": 325
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.33439235687256,
|
||
|
|
"epoch": 0.32819492789656884,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.00016450000000000001,
|
||
|
|
"loss": 8.0773,
|
||
|
|
"mean_token_accuracy": 0.0839149996638298,
|
||
|
|
"num_tokens": 652915.0,
|
||
|
|
"step": 330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.363460159301757,
|
||
|
|
"epoch": 0.33316757831924415,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.00016700000000000002,
|
||
|
|
"loss": 8.0669,
|
||
|
|
"mean_token_accuracy": 0.07783392369747162,
|
||
|
|
"num_tokens": 662665.0,
|
||
|
|
"step": 335
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.326759719848633,
|
||
|
|
"epoch": 0.33814022874191946,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.00016950000000000003,
|
||
|
|
"loss": 8.0773,
|
||
|
|
"mean_token_accuracy": 0.08024705946445465,
|
||
|
|
"num_tokens": 671667.0,
|
||
|
|
"step": 340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.384268760681152,
|
||
|
|
"epoch": 0.3431128791645947,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.00017199999999999998,
|
||
|
|
"loss": 8.0868,
|
||
|
|
"mean_token_accuracy": 0.0760427363216877,
|
||
|
|
"num_tokens": 680432.0,
|
||
|
|
"step": 345
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.345528602600098,
|
||
|
|
"epoch": 0.34808552958727,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.00017449999999999999,
|
||
|
|
"loss": 8.0811,
|
||
|
|
"mean_token_accuracy": 0.08016384020447731,
|
||
|
|
"num_tokens": 690184.0,
|
||
|
|
"step": 350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.333233165740968,
|
||
|
|
"epoch": 0.3530581800099453,
|
||
|
|
"grad_norm": 1.34375,
|
||
|
|
"learning_rate": 0.000177,
|
||
|
|
"loss": 7.9808,
|
||
|
|
"mean_token_accuracy": 0.08433079347014427,
|
||
|
|
"num_tokens": 699147.0,
|
||
|
|
"step": 355
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.210507869720459,
|
||
|
|
"epoch": 0.3580308304326206,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.0001795,
|
||
|
|
"loss": 7.9962,
|
||
|
|
"mean_token_accuracy": 0.08303841575980186,
|
||
|
|
"num_tokens": 708696.0,
|
||
|
|
"step": 360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.333552265167237,
|
||
|
|
"epoch": 0.3630034808552959,
|
||
|
|
"grad_norm": 1.2578125,
|
||
|
|
"learning_rate": 0.000182,
|
||
|
|
"loss": 7.9812,
|
||
|
|
"mean_token_accuracy": 0.08292912170290948,
|
||
|
|
"num_tokens": 718438.0,
|
||
|
|
"step": 365
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.339497661590576,
|
||
|
|
"epoch": 0.3679761312779712,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.0001845,
|
||
|
|
"loss": 8.1143,
|
||
|
|
"mean_token_accuracy": 0.07707406133413315,
|
||
|
|
"num_tokens": 728157.0,
|
||
|
|
"step": 370
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.221984767913819,
|
||
|
|
"epoch": 0.37294878170064644,
|
||
|
|
"grad_norm": 1.40625,
|
||
|
|
"learning_rate": 0.000187,
|
||
|
|
"loss": 8.0113,
|
||
|
|
"mean_token_accuracy": 0.083165742456913,
|
||
|
|
"num_tokens": 737007.0,
|
||
|
|
"step": 375
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.301848316192627,
|
||
|
|
"epoch": 0.37792143212332174,
|
||
|
|
"grad_norm": 1.2421875,
|
||
|
|
"learning_rate": 0.0001895,
|
||
|
|
"loss": 7.9361,
|
||
|
|
"mean_token_accuracy": 0.08524221666157246,
|
||
|
|
"num_tokens": 746873.0,
|
||
|
|
"step": 380
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.174521112442017,
|
||
|
|
"epoch": 0.382894082545997,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.000192,
|
||
|
|
"loss": 7.856,
|
||
|
|
"mean_token_accuracy": 0.08981255739927292,
|
||
|
|
"num_tokens": 755486.0,
|
||
|
|
"step": 385
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.252612590789795,
|
||
|
|
"epoch": 0.3878667329686723,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 0.0001945,
|
||
|
|
"loss": 7.9822,
|
||
|
|
"mean_token_accuracy": 0.07825585678219796,
|
||
|
|
"num_tokens": 765481.0,
|
||
|
|
"step": 390
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.236901569366456,
|
||
|
|
"epoch": 0.3928393833913476,
|
||
|
|
"grad_norm": 1.328125,
|
||
|
|
"learning_rate": 0.00019700000000000002,
|
||
|
|
"loss": 7.937,
|
||
|
|
"mean_token_accuracy": 0.08628773540258408,
|
||
|
|
"num_tokens": 774335.0,
|
||
|
|
"step": 395
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.332719326019287,
|
||
|
|
"epoch": 0.39781203381402286,
|
||
|
|
"grad_norm": 1.296875,
|
||
|
|
"learning_rate": 0.00019950000000000002,
|
||
|
|
"loss": 8.0001,
|
||
|
|
"mean_token_accuracy": 0.08222218602895737,
|
||
|
|
"num_tokens": 784307.0,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.230411338806153,
|
||
|
|
"epoch": 0.40278468423669817,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.000202,
|
||
|
|
"loss": 8.0406,
|
||
|
|
"mean_token_accuracy": 0.07930415198206901,
|
||
|
|
"num_tokens": 795058.0,
|
||
|
|
"step": 405
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.313278293609619,
|
||
|
|
"epoch": 0.40775733465937347,
|
||
|
|
"grad_norm": 1.703125,
|
||
|
|
"learning_rate": 0.00020449999999999998,
|
||
|
|
"loss": 7.9507,
|
||
|
|
"mean_token_accuracy": 0.08477597758173942,
|
||
|
|
"num_tokens": 805687.0,
|
||
|
|
"step": 410
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.19376473426819,
|
||
|
|
"epoch": 0.4127299850820487,
|
||
|
|
"grad_norm": 1.1875,
|
||
|
|
"learning_rate": 0.000207,
|
||
|
|
"loss": 7.989,
|
||
|
|
"mean_token_accuracy": 0.08413419276475906,
|
||
|
|
"num_tokens": 815138.0,
|
||
|
|
"step": 415
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.226968097686768,
|
||
|
|
"epoch": 0.41770263550472403,
|
||
|
|
"grad_norm": 1.3125,
|
||
|
|
"learning_rate": 0.0002095,
|
||
|
|
"loss": 7.9451,
|
||
|
|
"mean_token_accuracy": 0.08170225620269775,
|
||
|
|
"num_tokens": 825009.0,
|
||
|
|
"step": 420
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.161935615539551,
|
||
|
|
"epoch": 0.4226752859273993,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.000212,
|
||
|
|
"loss": 7.9202,
|
||
|
|
"mean_token_accuracy": 0.0900972880423069,
|
||
|
|
"num_tokens": 834708.0,
|
||
|
|
"step": 425
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.352538585662842,
|
||
|
|
"epoch": 0.4276479363500746,
|
||
|
|
"grad_norm": 1.296875,
|
||
|
|
"learning_rate": 0.0002145,
|
||
|
|
"loss": 7.9872,
|
||
|
|
"mean_token_accuracy": 0.08343984819948673,
|
||
|
|
"num_tokens": 843888.0,
|
||
|
|
"step": 430
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.144132709503173,
|
||
|
|
"epoch": 0.4326205867727499,
|
||
|
|
"grad_norm": 1.2734375,
|
||
|
|
"learning_rate": 0.00021700000000000002,
|
||
|
|
"loss": 7.9481,
|
||
|
|
"mean_token_accuracy": 0.08321668803691865,
|
||
|
|
"num_tokens": 853057.0,
|
||
|
|
"step": 435
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.240882396697998,
|
||
|
|
"epoch": 0.43759323719542514,
|
||
|
|
"grad_norm": 1.359375,
|
||
|
|
"learning_rate": 0.0002195,
|
||
|
|
"loss": 8.0054,
|
||
|
|
"mean_token_accuracy": 0.07688518613576889,
|
||
|
|
"num_tokens": 862701.0,
|
||
|
|
"step": 440
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.124331283569337,
|
||
|
|
"epoch": 0.44256588761810045,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.000222,
|
||
|
|
"loss": 7.8978,
|
||
|
|
"mean_token_accuracy": 0.0850852258503437,
|
||
|
|
"num_tokens": 872129.0,
|
||
|
|
"step": 445
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.162071371078492,
|
||
|
|
"epoch": 0.44753853804077576,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.0002245,
|
||
|
|
"loss": 7.9163,
|
||
|
|
"mean_token_accuracy": 0.08625531122088433,
|
||
|
|
"num_tokens": 883166.0,
|
||
|
|
"step": 450
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.160173559188843,
|
||
|
|
"epoch": 0.452511188463451,
|
||
|
|
"grad_norm": 1.2890625,
|
||
|
|
"learning_rate": 0.00022700000000000002,
|
||
|
|
"loss": 7.7103,
|
||
|
|
"mean_token_accuracy": 0.10269899219274521,
|
||
|
|
"num_tokens": 894159.0,
|
||
|
|
"step": 455
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.14470615386963,
|
||
|
|
"epoch": 0.4574838388861263,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.00022950000000000002,
|
||
|
|
"loss": 7.9094,
|
||
|
|
"mean_token_accuracy": 0.08693445995450019,
|
||
|
|
"num_tokens": 904588.0,
|
||
|
|
"step": 460
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.188994216918946,
|
||
|
|
"epoch": 0.46245648930880157,
|
||
|
|
"grad_norm": 1.34375,
|
||
|
|
"learning_rate": 0.00023200000000000003,
|
||
|
|
"loss": 7.8764,
|
||
|
|
"mean_token_accuracy": 0.08472439050674438,
|
||
|
|
"num_tokens": 915554.0,
|
||
|
|
"step": 465
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.080974769592284,
|
||
|
|
"epoch": 0.4674291397314769,
|
||
|
|
"grad_norm": 1.2421875,
|
||
|
|
"learning_rate": 0.00023449999999999998,
|
||
|
|
"loss": 7.9153,
|
||
|
|
"mean_token_accuracy": 0.08615779727697373,
|
||
|
|
"num_tokens": 925546.0,
|
||
|
|
"step": 470
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.106526470184326,
|
||
|
|
"epoch": 0.4724017901541522,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.000237,
|
||
|
|
"loss": 7.8827,
|
||
|
|
"mean_token_accuracy": 0.08863506764173508,
|
||
|
|
"num_tokens": 936615.0,
|
||
|
|
"step": 475
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.171993637084961,
|
||
|
|
"epoch": 0.47737444057682743,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.0002395,
|
||
|
|
"loss": 7.9148,
|
||
|
|
"mean_token_accuracy": 0.08405414745211601,
|
||
|
|
"num_tokens": 947374.0,
|
||
|
|
"step": 480
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.07227544784546,
|
||
|
|
"epoch": 0.48234709099950274,
|
||
|
|
"grad_norm": 0.9765625,
|
||
|
|
"learning_rate": 0.000242,
|
||
|
|
"loss": 7.8263,
|
||
|
|
"mean_token_accuracy": 0.0881466455757618,
|
||
|
|
"num_tokens": 957772.0,
|
||
|
|
"step": 485
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.149549674987792,
|
||
|
|
"epoch": 0.48731974142217804,
|
||
|
|
"grad_norm": 1.1875,
|
||
|
|
"learning_rate": 0.0002445,
|
||
|
|
"loss": 7.9265,
|
||
|
|
"mean_token_accuracy": 0.08613847866654396,
|
||
|
|
"num_tokens": 967724.0,
|
||
|
|
"step": 490
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.014617347717286,
|
||
|
|
"epoch": 0.4922923918448533,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.000247,
|
||
|
|
"loss": 7.821,
|
||
|
|
"mean_token_accuracy": 0.08873307779431343,
|
||
|
|
"num_tokens": 976497.0,
|
||
|
|
"step": 495
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.061779451370239,
|
||
|
|
"epoch": 0.4972650422675286,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 0.0002495,
|
||
|
|
"loss": 7.8061,
|
||
|
|
"mean_token_accuracy": 0.08952893353998662,
|
||
|
|
"num_tokens": 986207.0,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4972650422675286,
|
||
|
|
"eval_entropy": 8.12199338789909,
|
||
|
|
"eval_loss": 7.925317764282227,
|
||
|
|
"eval_mean_token_accuracy": 0.0847069110511051,
|
||
|
|
"eval_num_tokens": 986207.0,
|
||
|
|
"eval_runtime": 11.3595,
|
||
|
|
"eval_samples_per_second": 305.031,
|
||
|
|
"eval_steps_per_second": 38.206,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.053079652786256,
|
||
|
|
"epoch": 0.5022376926902039,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.000252,
|
||
|
|
"loss": 7.779,
|
||
|
|
"mean_token_accuracy": 0.09456762075424194,
|
||
|
|
"num_tokens": 995568.0,
|
||
|
|
"step": 505
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.076926040649415,
|
||
|
|
"epoch": 0.5072103431128792,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.0002545,
|
||
|
|
"loss": 7.8379,
|
||
|
|
"mean_token_accuracy": 0.08805601447820663,
|
||
|
|
"num_tokens": 1005829.0,
|
||
|
|
"step": 510
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.036222219467163,
|
||
|
|
"epoch": 0.5121829935355544,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000257,
|
||
|
|
"loss": 7.7909,
|
||
|
|
"mean_token_accuracy": 0.08699572384357453,
|
||
|
|
"num_tokens": 1015290.0,
|
||
|
|
"step": 515
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.061750364303588,
|
||
|
|
"epoch": 0.5171556439582298,
|
||
|
|
"grad_norm": 1.28125,
|
||
|
|
"learning_rate": 0.0002595,
|
||
|
|
"loss": 7.7536,
|
||
|
|
"mean_token_accuracy": 0.09390853121876716,
|
||
|
|
"num_tokens": 1024479.0,
|
||
|
|
"step": 520
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.9042600154876705,
|
||
|
|
"epoch": 0.522128294380905,
|
||
|
|
"grad_norm": 2.078125,
|
||
|
|
"learning_rate": 0.000262,
|
||
|
|
"loss": 7.6663,
|
||
|
|
"mean_token_accuracy": 0.09484521970152855,
|
||
|
|
"num_tokens": 1034092.0,
|
||
|
|
"step": 525
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.166246604919433,
|
||
|
|
"epoch": 0.5271009448035803,
|
||
|
|
"grad_norm": 1.2890625,
|
||
|
|
"learning_rate": 0.00026450000000000003,
|
||
|
|
"loss": 7.9006,
|
||
|
|
"mean_token_accuracy": 0.09057568311691284,
|
||
|
|
"num_tokens": 1043844.0,
|
||
|
|
"step": 530
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.903245353698731,
|
||
|
|
"epoch": 0.5320735952262556,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.00026700000000000004,
|
||
|
|
"loss": 7.7407,
|
||
|
|
"mean_token_accuracy": 0.09305151179432869,
|
||
|
|
"num_tokens": 1052646.0,
|
||
|
|
"step": 535
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.107067775726318,
|
||
|
|
"epoch": 0.5370462456489309,
|
||
|
|
"grad_norm": 1.2578125,
|
||
|
|
"learning_rate": 0.00026950000000000005,
|
||
|
|
"loss": 7.7945,
|
||
|
|
"mean_token_accuracy": 0.08951174244284629,
|
||
|
|
"num_tokens": 1062345.0,
|
||
|
|
"step": 540
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.024613332748412,
|
||
|
|
"epoch": 0.5420188960716061,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.00027200000000000005,
|
||
|
|
"loss": 7.7886,
|
||
|
|
"mean_token_accuracy": 0.08819786384701729,
|
||
|
|
"num_tokens": 1071116.0,
|
||
|
|
"step": 545
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.918569803237915,
|
||
|
|
"epoch": 0.5469915464942815,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.0002745,
|
||
|
|
"loss": 7.7358,
|
||
|
|
"mean_token_accuracy": 0.09049084335565567,
|
||
|
|
"num_tokens": 1080562.0,
|
||
|
|
"step": 550
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.018861722946166,
|
||
|
|
"epoch": 0.5519641969169568,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.000277,
|
||
|
|
"loss": 7.7855,
|
||
|
|
"mean_token_accuracy": 0.08982611894607544,
|
||
|
|
"num_tokens": 1091944.0,
|
||
|
|
"step": 555
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.988247442245483,
|
||
|
|
"epoch": 0.556936847339632,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.0002795,
|
||
|
|
"loss": 7.7154,
|
||
|
|
"mean_token_accuracy": 0.09307126253843308,
|
||
|
|
"num_tokens": 1101412.0,
|
||
|
|
"step": 560
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.006781101226807,
|
||
|
|
"epoch": 0.5619094977623074,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 0.00028199999999999997,
|
||
|
|
"loss": 7.8081,
|
||
|
|
"mean_token_accuracy": 0.09186437204480172,
|
||
|
|
"num_tokens": 1110652.0,
|
||
|
|
"step": 565
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.9736639022827145,
|
||
|
|
"epoch": 0.5668821481849826,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.0002845,
|
||
|
|
"loss": 7.7146,
|
||
|
|
"mean_token_accuracy": 0.09523341283202172,
|
||
|
|
"num_tokens": 1120420.0,
|
||
|
|
"step": 570
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.896536588668823,
|
||
|
|
"epoch": 0.5718547986076579,
|
||
|
|
"grad_norm": 1.2890625,
|
||
|
|
"learning_rate": 0.000287,
|
||
|
|
"loss": 7.775,
|
||
|
|
"mean_token_accuracy": 0.08853160478174686,
|
||
|
|
"num_tokens": 1130046.0,
|
||
|
|
"step": 575
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.949215126037598,
|
||
|
|
"epoch": 0.5768274490303331,
|
||
|
|
"grad_norm": 1.421875,
|
||
|
|
"learning_rate": 0.0002895,
|
||
|
|
"loss": 7.7521,
|
||
|
|
"mean_token_accuracy": 0.09078693836927414,
|
||
|
|
"num_tokens": 1138946.0,
|
||
|
|
"step": 580
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.021563529968262,
|
||
|
|
"epoch": 0.5818000994530085,
|
||
|
|
"grad_norm": 1.2890625,
|
||
|
|
"learning_rate": 0.000292,
|
||
|
|
"loss": 7.4967,
|
||
|
|
"mean_token_accuracy": 0.10361211746931076,
|
||
|
|
"num_tokens": 1149440.0,
|
||
|
|
"step": 585
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.954723596572876,
|
||
|
|
"epoch": 0.5867727498756837,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.0002945,
|
||
|
|
"loss": 7.7857,
|
||
|
|
"mean_token_accuracy": 0.08751340731978416,
|
||
|
|
"num_tokens": 1158636.0,
|
||
|
|
"step": 590
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.810801792144775,
|
||
|
|
"epoch": 0.591745400298359,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 0.000297,
|
||
|
|
"loss": 7.6241,
|
||
|
|
"mean_token_accuracy": 0.0969457134604454,
|
||
|
|
"num_tokens": 1167776.0,
|
||
|
|
"step": 595
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.93477840423584,
|
||
|
|
"epoch": 0.5967180507210343,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.0002995,
|
||
|
|
"loss": 7.6271,
|
||
|
|
"mean_token_accuracy": 0.0945424608886242,
|
||
|
|
"num_tokens": 1177919.0,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.868533611297607,
|
||
|
|
"epoch": 0.6016907011437096,
|
||
|
|
"grad_norm": 1.2421875,
|
||
|
|
"learning_rate": 0.000302,
|
||
|
|
"loss": 7.695,
|
||
|
|
"mean_token_accuracy": 0.10151977390050888,
|
||
|
|
"num_tokens": 1187363.0,
|
||
|
|
"step": 605
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.040846824645996,
|
||
|
|
"epoch": 0.6066633515663848,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.0003045,
|
||
|
|
"loss": 7.8651,
|
||
|
|
"mean_token_accuracy": 0.08632402569055557,
|
||
|
|
"num_tokens": 1197329.0,
|
||
|
|
"step": 610
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.835066843032837,
|
||
|
|
"epoch": 0.6116360019890602,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.000307,
|
||
|
|
"loss": 7.7643,
|
||
|
|
"mean_token_accuracy": 0.08976817056536675,
|
||
|
|
"num_tokens": 1207408.0,
|
||
|
|
"step": 615
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.966505718231201,
|
||
|
|
"epoch": 0.6166086524117355,
|
||
|
|
"grad_norm": 1.390625,
|
||
|
|
"learning_rate": 0.0003095,
|
||
|
|
"loss": 7.7505,
|
||
|
|
"mean_token_accuracy": 0.08791251480579376,
|
||
|
|
"num_tokens": 1219258.0,
|
||
|
|
"step": 620
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.945406055450439,
|
||
|
|
"epoch": 0.6215813028344107,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.000312,
|
||
|
|
"loss": 7.7417,
|
||
|
|
"mean_token_accuracy": 0.08997605070471763,
|
||
|
|
"num_tokens": 1228561.0,
|
||
|
|
"step": 625
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.796209669113159,
|
||
|
|
"epoch": 0.6265539532570861,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 0.0003145,
|
||
|
|
"loss": 7.6714,
|
||
|
|
"mean_token_accuracy": 0.09316384196281433,
|
||
|
|
"num_tokens": 1239665.0,
|
||
|
|
"step": 630
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.944826412200928,
|
||
|
|
"epoch": 0.6315266036797613,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.000317,
|
||
|
|
"loss": 7.6703,
|
||
|
|
"mean_token_accuracy": 0.09130050241947174,
|
||
|
|
"num_tokens": 1249490.0,
|
||
|
|
"step": 635
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.864301347732544,
|
||
|
|
"epoch": 0.6364992541024366,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.0003195,
|
||
|
|
"loss": 7.6962,
|
||
|
|
"mean_token_accuracy": 0.09296346306800843,
|
||
|
|
"num_tokens": 1259328.0,
|
||
|
|
"step": 640
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.925337028503418,
|
||
|
|
"epoch": 0.6414719045251119,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.000322,
|
||
|
|
"loss": 7.6193,
|
||
|
|
"mean_token_accuracy": 0.0926283597946167,
|
||
|
|
"num_tokens": 1269523.0,
|
||
|
|
"step": 645
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.766702365875244,
|
||
|
|
"epoch": 0.6464445549477872,
|
||
|
|
"grad_norm": 1.28125,
|
||
|
|
"learning_rate": 0.00032450000000000003,
|
||
|
|
"loss": 7.7288,
|
||
|
|
"mean_token_accuracy": 0.08978976607322693,
|
||
|
|
"num_tokens": 1277961.0,
|
||
|
|
"step": 650
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.95000376701355,
|
||
|
|
"epoch": 0.6514172053704624,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.00032700000000000003,
|
||
|
|
"loss": 7.6807,
|
||
|
|
"mean_token_accuracy": 0.09115450382232666,
|
||
|
|
"num_tokens": 1288228.0,
|
||
|
|
"step": 655
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.832488107681274,
|
||
|
|
"epoch": 0.6563898557931377,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.00032950000000000004,
|
||
|
|
"loss": 7.6293,
|
||
|
|
"mean_token_accuracy": 0.09410025104880333,
|
||
|
|
"num_tokens": 1299636.0,
|
||
|
|
"step": 660
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.785907840728759,
|
||
|
|
"epoch": 0.661362506215813,
|
||
|
|
"grad_norm": 1.296875,
|
||
|
|
"learning_rate": 0.00033200000000000005,
|
||
|
|
"loss": 7.6553,
|
||
|
|
"mean_token_accuracy": 0.09513049125671387,
|
||
|
|
"num_tokens": 1310035.0,
|
||
|
|
"step": 665
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.74755220413208,
|
||
|
|
"epoch": 0.6663351566384883,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.00033450000000000005,
|
||
|
|
"loss": 7.6673,
|
||
|
|
"mean_token_accuracy": 0.09929112568497658,
|
||
|
|
"num_tokens": 1320751.0,
|
||
|
|
"step": 670
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.734870958328247,
|
||
|
|
"epoch": 0.6713078070611636,
|
||
|
|
"grad_norm": 1.71875,
|
||
|
|
"learning_rate": 0.000337,
|
||
|
|
"loss": 7.5335,
|
||
|
|
"mean_token_accuracy": 0.09909648820757866,
|
||
|
|
"num_tokens": 1331023.0,
|
||
|
|
"step": 675
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.7550742626190186,
|
||
|
|
"epoch": 0.6762804574838389,
|
||
|
|
"grad_norm": 1.1875,
|
||
|
|
"learning_rate": 0.0003395,
|
||
|
|
"loss": 7.6342,
|
||
|
|
"mean_token_accuracy": 0.0891692079603672,
|
||
|
|
"num_tokens": 1341814.0,
|
||
|
|
"step": 680
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.792105770111084,
|
||
|
|
"epoch": 0.6812531079065142,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.000342,
|
||
|
|
"loss": 7.5512,
|
||
|
|
"mean_token_accuracy": 0.09285351634025574,
|
||
|
|
"num_tokens": 1351026.0,
|
||
|
|
"step": 685
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.822172403335571,
|
||
|
|
"epoch": 0.6862257583291894,
|
||
|
|
"grad_norm": 1.1640625,
|
||
|
|
"learning_rate": 0.00034449999999999997,
|
||
|
|
"loss": 7.7478,
|
||
|
|
"mean_token_accuracy": 0.08240504153072833,
|
||
|
|
"num_tokens": 1361004.0,
|
||
|
|
"step": 690
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.7296113014221195,
|
||
|
|
"epoch": 0.6911984087518648,
|
||
|
|
"grad_norm": 1.328125,
|
||
|
|
"learning_rate": 0.000347,
|
||
|
|
"loss": 7.6089,
|
||
|
|
"mean_token_accuracy": 0.09175579845905305,
|
||
|
|
"num_tokens": 1370729.0,
|
||
|
|
"step": 695
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.822756958007813,
|
||
|
|
"epoch": 0.69617105917454,
|
||
|
|
"grad_norm": 1.328125,
|
||
|
|
"learning_rate": 0.0003495,
|
||
|
|
"loss": 7.5643,
|
||
|
|
"mean_token_accuracy": 0.09511974975466728,
|
||
|
|
"num_tokens": 1379525.0,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.708429861068725,
|
||
|
|
"epoch": 0.7011437095972153,
|
||
|
|
"grad_norm": 1.296875,
|
||
|
|
"learning_rate": 0.000352,
|
||
|
|
"loss": 7.6297,
|
||
|
|
"mean_token_accuracy": 0.09320607557892799,
|
||
|
|
"num_tokens": 1389863.0,
|
||
|
|
"step": 705
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.726954174041748,
|
||
|
|
"epoch": 0.7061163600198906,
|
||
|
|
"grad_norm": 1.34375,
|
||
|
|
"learning_rate": 0.0003545,
|
||
|
|
"loss": 7.7481,
|
||
|
|
"mean_token_accuracy": 0.08981630355119705,
|
||
|
|
"num_tokens": 1399468.0,
|
||
|
|
"step": 710
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.754483079910278,
|
||
|
|
"epoch": 0.7110890104425659,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.000357,
|
||
|
|
"loss": 7.5528,
|
||
|
|
"mean_token_accuracy": 0.09678518697619438,
|
||
|
|
"num_tokens": 1408785.0,
|
||
|
|
"step": 715
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.69726414680481,
|
||
|
|
"epoch": 0.7160616608652411,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.0003595,
|
||
|
|
"loss": 7.5883,
|
||
|
|
"mean_token_accuracy": 0.09907481074333191,
|
||
|
|
"num_tokens": 1418666.0,
|
||
|
|
"step": 720
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.866272592544556,
|
||
|
|
"epoch": 0.7210343112879165,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.000362,
|
||
|
|
"loss": 7.7005,
|
||
|
|
"mean_token_accuracy": 0.09374464005231857,
|
||
|
|
"num_tokens": 1428699.0,
|
||
|
|
"step": 725
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.772790002822876,
|
||
|
|
"epoch": 0.7260069617105918,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.0003645,
|
||
|
|
"loss": 7.5955,
|
||
|
|
"mean_token_accuracy": 0.09428214207291603,
|
||
|
|
"num_tokens": 1437354.0,
|
||
|
|
"step": 730
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.768170261383057,
|
||
|
|
"epoch": 0.730979612133267,
|
||
|
|
"grad_norm": 1.1875,
|
||
|
|
"learning_rate": 0.000367,
|
||
|
|
"loss": 7.5997,
|
||
|
|
"mean_token_accuracy": 0.09616516754031182,
|
||
|
|
"num_tokens": 1447327.0,
|
||
|
|
"step": 735
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.792877292633056,
|
||
|
|
"epoch": 0.7359522625559424,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 0.0003695,
|
||
|
|
"loss": 7.6569,
|
||
|
|
"mean_token_accuracy": 0.09555381312966346,
|
||
|
|
"num_tokens": 1457793.0,
|
||
|
|
"step": 740
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.718319368362427,
|
||
|
|
"epoch": 0.7409249129786176,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.000372,
|
||
|
|
"loss": 7.5646,
|
||
|
|
"mean_token_accuracy": 0.09642104059457779,
|
||
|
|
"num_tokens": 1467143.0,
|
||
|
|
"step": 745
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.69613127708435,
|
||
|
|
"epoch": 0.7458975634012929,
|
||
|
|
"grad_norm": 1.265625,
|
||
|
|
"learning_rate": 0.0003745,
|
||
|
|
"loss": 7.6171,
|
||
|
|
"mean_token_accuracy": 0.09400601610541344,
|
||
|
|
"num_tokens": 1476422.0,
|
||
|
|
"step": 750
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.6746134757995605,
|
||
|
|
"epoch": 0.7508702138239681,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.000377,
|
||
|
|
"loss": 7.55,
|
||
|
|
"mean_token_accuracy": 0.10190361589193345,
|
||
|
|
"num_tokens": 1486253.0,
|
||
|
|
"step": 755
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.683488607406616,
|
||
|
|
"epoch": 0.7558428642466435,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.0003795,
|
||
|
|
"loss": 7.5676,
|
||
|
|
"mean_token_accuracy": 0.0934619203209877,
|
||
|
|
"num_tokens": 1496080.0,
|
||
|
|
"step": 760
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.70647873878479,
|
||
|
|
"epoch": 0.7608155146693187,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 0.000382,
|
||
|
|
"loss": 7.5191,
|
||
|
|
"mean_token_accuracy": 0.09906428828835487,
|
||
|
|
"num_tokens": 1505532.0,
|
||
|
|
"step": 765
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.636048603057861,
|
||
|
|
"epoch": 0.765788165091994,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.0003845,
|
||
|
|
"loss": 7.4547,
|
||
|
|
"mean_token_accuracy": 0.09559990018606186,
|
||
|
|
"num_tokens": 1515705.0,
|
||
|
|
"step": 770
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.626402521133423,
|
||
|
|
"epoch": 0.7707608155146694,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.00038700000000000003,
|
||
|
|
"loss": 7.515,
|
||
|
|
"mean_token_accuracy": 0.09692738726735114,
|
||
|
|
"num_tokens": 1525651.0,
|
||
|
|
"step": 775
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.607789754867554,
|
||
|
|
"epoch": 0.7757334659373446,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.00038950000000000003,
|
||
|
|
"loss": 7.5051,
|
||
|
|
"mean_token_accuracy": 0.09369031935930253,
|
||
|
|
"num_tokens": 1534576.0,
|
||
|
|
"step": 780
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.578536558151245,
|
||
|
|
"epoch": 0.7807061163600199,
|
||
|
|
"grad_norm": 2.8125,
|
||
|
|
"learning_rate": 0.00039200000000000004,
|
||
|
|
"loss": 7.3647,
|
||
|
|
"mean_token_accuracy": 0.10836331695318221,
|
||
|
|
"num_tokens": 1544765.0,
|
||
|
|
"step": 785
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.696880531311035,
|
||
|
|
"epoch": 0.7856787667826952,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.00039450000000000005,
|
||
|
|
"loss": 7.4927,
|
||
|
|
"mean_token_accuracy": 0.10022924914956093,
|
||
|
|
"num_tokens": 1553636.0,
|
||
|
|
"step": 790
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.581558704376221,
|
||
|
|
"epoch": 0.7906514172053705,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.00039700000000000005,
|
||
|
|
"loss": 7.4583,
|
||
|
|
"mean_token_accuracy": 0.09628499075770378,
|
||
|
|
"num_tokens": 1563356.0,
|
||
|
|
"step": 795
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.529683589935303,
|
||
|
|
"epoch": 0.7956240676280457,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 0.0003995,
|
||
|
|
"loss": 7.4794,
|
||
|
|
"mean_token_accuracy": 0.10385911986231804,
|
||
|
|
"num_tokens": 1572508.0,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.677089595794678,
|
||
|
|
"epoch": 0.8005967180507211,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 0.000402,
|
||
|
|
"loss": 7.5022,
|
||
|
|
"mean_token_accuracy": 0.09889172688126564,
|
||
|
|
"num_tokens": 1581583.0,
|
||
|
|
"step": 805
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.620496559143066,
|
||
|
|
"epoch": 0.8055693684733963,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.0004045,
|
||
|
|
"loss": 7.5028,
|
||
|
|
"mean_token_accuracy": 0.0993174895644188,
|
||
|
|
"num_tokens": 1592114.0,
|
||
|
|
"step": 810
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.58529052734375,
|
||
|
|
"epoch": 0.8105420188960716,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.00040699999999999997,
|
||
|
|
"loss": 7.414,
|
||
|
|
"mean_token_accuracy": 0.10554423704743385,
|
||
|
|
"num_tokens": 1602417.0,
|
||
|
|
"step": 815
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.617566537857056,
|
||
|
|
"epoch": 0.8155146693187469,
|
||
|
|
"grad_norm": 1.3828125,
|
||
|
|
"learning_rate": 0.0004095,
|
||
|
|
"loss": 7.5095,
|
||
|
|
"mean_token_accuracy": 0.09752819314599037,
|
||
|
|
"num_tokens": 1611753.0,
|
||
|
|
"step": 820
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.640080833435059,
|
||
|
|
"epoch": 0.8204873197414222,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.000412,
|
||
|
|
"loss": 7.565,
|
||
|
|
"mean_token_accuracy": 0.09824235141277313,
|
||
|
|
"num_tokens": 1621959.0,
|
||
|
|
"step": 825
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.517613458633423,
|
||
|
|
"epoch": 0.8254599701640974,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.0004145,
|
||
|
|
"loss": 7.4706,
|
||
|
|
"mean_token_accuracy": 0.09650598615407943,
|
||
|
|
"num_tokens": 1630909.0,
|
||
|
|
"step": 830
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.668545198440552,
|
||
|
|
"epoch": 0.8304326205867727,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 0.000417,
|
||
|
|
"loss": 7.4442,
|
||
|
|
"mean_token_accuracy": 0.09847426488995552,
|
||
|
|
"num_tokens": 1641316.0,
|
||
|
|
"step": 835
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.467155838012696,
|
||
|
|
"epoch": 0.8354052710094481,
|
||
|
|
"grad_norm": 1.2421875,
|
||
|
|
"learning_rate": 0.0004195,
|
||
|
|
"loss": 7.3595,
|
||
|
|
"mean_token_accuracy": 0.10589170381426811,
|
||
|
|
"num_tokens": 1651506.0,
|
||
|
|
"step": 840
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.445871210098266,
|
||
|
|
"epoch": 0.8403779214321233,
|
||
|
|
"grad_norm": 1.2890625,
|
||
|
|
"learning_rate": 0.000422,
|
||
|
|
"loss": 7.3611,
|
||
|
|
"mean_token_accuracy": 0.1002371571958065,
|
||
|
|
"num_tokens": 1659871.0,
|
||
|
|
"step": 845
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.649908971786499,
|
||
|
|
"epoch": 0.8453505718547986,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.0004245,
|
||
|
|
"loss": 7.578,
|
||
|
|
"mean_token_accuracy": 0.09238532930612564,
|
||
|
|
"num_tokens": 1670178.0,
|
||
|
|
"step": 850
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.569011211395264,
|
||
|
|
"epoch": 0.8503232222774739,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000427,
|
||
|
|
"loss": 7.4567,
|
||
|
|
"mean_token_accuracy": 0.09851670637726784,
|
||
|
|
"num_tokens": 1680131.0,
|
||
|
|
"step": 855
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.358003282546997,
|
||
|
|
"epoch": 0.8552958727001492,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 0.0004295,
|
||
|
|
"loss": 7.3453,
|
||
|
|
"mean_token_accuracy": 0.1004187822341919,
|
||
|
|
"num_tokens": 1689568.0,
|
||
|
|
"step": 860
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.614206790924072,
|
||
|
|
"epoch": 0.8602685231228244,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.000432,
|
||
|
|
"loss": 7.3861,
|
||
|
|
"mean_token_accuracy": 0.1079413764178753,
|
||
|
|
"num_tokens": 1698039.0,
|
||
|
|
"step": 865
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.502506685256958,
|
||
|
|
"epoch": 0.8652411735454998,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.0004345,
|
||
|
|
"loss": 7.4282,
|
||
|
|
"mean_token_accuracy": 0.10273555293679237,
|
||
|
|
"num_tokens": 1707659.0,
|
||
|
|
"step": 870
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.472733020782471,
|
||
|
|
"epoch": 0.870213823968175,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.000437,
|
||
|
|
"loss": 7.3861,
|
||
|
|
"mean_token_accuracy": 0.10378114208579063,
|
||
|
|
"num_tokens": 1716952.0,
|
||
|
|
"step": 875
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.560409307479858,
|
||
|
|
"epoch": 0.8751864743908503,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.0004395,
|
||
|
|
"loss": 7.451,
|
||
|
|
"mean_token_accuracy": 0.09781106933951378,
|
||
|
|
"num_tokens": 1726323.0,
|
||
|
|
"step": 880
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.568419170379639,
|
||
|
|
"epoch": 0.8801591248135257,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.000442,
|
||
|
|
"loss": 7.4508,
|
||
|
|
"mean_token_accuracy": 0.09186950996518135,
|
||
|
|
"num_tokens": 1736265.0,
|
||
|
|
"step": 885
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.458628368377686,
|
||
|
|
"epoch": 0.8851317752362009,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.0004445,
|
||
|
|
"loss": 7.3973,
|
||
|
|
"mean_token_accuracy": 0.09887660518288613,
|
||
|
|
"num_tokens": 1745150.0,
|
||
|
|
"step": 890
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.540029096603393,
|
||
|
|
"epoch": 0.8901044256588762,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.000447,
|
||
|
|
"loss": 7.5139,
|
||
|
|
"mean_token_accuracy": 0.09781100377440452,
|
||
|
|
"num_tokens": 1754641.0,
|
||
|
|
"step": 895
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.446507453918457,
|
||
|
|
"epoch": 0.8950770760815515,
|
||
|
|
"grad_norm": 1.1640625,
|
||
|
|
"learning_rate": 0.00044950000000000003,
|
||
|
|
"loss": 7.3119,
|
||
|
|
"mean_token_accuracy": 0.10973177030682564,
|
||
|
|
"num_tokens": 1763468.0,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.422563886642456,
|
||
|
|
"epoch": 0.9000497265042268,
|
||
|
|
"grad_norm": 0.94921875,
|
||
|
|
"learning_rate": 0.00045200000000000004,
|
||
|
|
"loss": 7.4432,
|
||
|
|
"mean_token_accuracy": 0.10403756350278855,
|
||
|
|
"num_tokens": 1774845.0,
|
||
|
|
"step": 905
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.505294227600098,
|
||
|
|
"epoch": 0.905022376926902,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 0.00045450000000000004,
|
||
|
|
"loss": 7.3637,
|
||
|
|
"mean_token_accuracy": 0.10615181997418403,
|
||
|
|
"num_tokens": 1784606.0,
|
||
|
|
"step": 910
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.428920269012451,
|
||
|
|
"epoch": 0.9099950273495774,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.00045700000000000005,
|
||
|
|
"loss": 7.4024,
|
||
|
|
"mean_token_accuracy": 0.10240766182541847,
|
||
|
|
"num_tokens": 1793893.0,
|
||
|
|
"step": 915
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.477800798416138,
|
||
|
|
"epoch": 0.9149676777722526,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.00045950000000000006,
|
||
|
|
"loss": 7.4009,
|
||
|
|
"mean_token_accuracy": 0.1039421871304512,
|
||
|
|
"num_tokens": 1804095.0,
|
||
|
|
"step": 920
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.540898036956787,
|
||
|
|
"epoch": 0.9199403281949279,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.000462,
|
||
|
|
"loss": 7.3965,
|
||
|
|
"mean_token_accuracy": 0.10489938706159592,
|
||
|
|
"num_tokens": 1813475.0,
|
||
|
|
"step": 925
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.317395401000977,
|
||
|
|
"epoch": 0.9249129786176031,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.0004645,
|
||
|
|
"loss": 7.3203,
|
||
|
|
"mean_token_accuracy": 0.109952911734581,
|
||
|
|
"num_tokens": 1822779.0,
|
||
|
|
"step": 930
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.5778134822845455,
|
||
|
|
"epoch": 0.9298856290402785,
|
||
|
|
"grad_norm": 1.265625,
|
||
|
|
"learning_rate": 0.000467,
|
||
|
|
"loss": 7.349,
|
||
|
|
"mean_token_accuracy": 0.10500052943825722,
|
||
|
|
"num_tokens": 1832239.0,
|
||
|
|
"step": 935
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.2777961730957035,
|
||
|
|
"epoch": 0.9348582794629537,
|
||
|
|
"grad_norm": 1.5703125,
|
||
|
|
"learning_rate": 0.0004695,
|
||
|
|
"loss": 7.3235,
|
||
|
|
"mean_token_accuracy": 0.10185129940509796,
|
||
|
|
"num_tokens": 1843395.0,
|
||
|
|
"step": 940
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.460053205490112,
|
||
|
|
"epoch": 0.939830929885629,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.000472,
|
||
|
|
"loss": 7.442,
|
||
|
|
"mean_token_accuracy": 0.10547303482890129,
|
||
|
|
"num_tokens": 1853861.0,
|
||
|
|
"step": 945
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.4669934749603275,
|
||
|
|
"epoch": 0.9448035803083044,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.0004745,
|
||
|
|
"loss": 7.3512,
|
||
|
|
"mean_token_accuracy": 0.10421937704086304,
|
||
|
|
"num_tokens": 1862992.0,
|
||
|
|
"step": 950
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.427795505523681,
|
||
|
|
"epoch": 0.9497762307309796,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000477,
|
||
|
|
"loss": 7.372,
|
||
|
|
"mean_token_accuracy": 0.10107123926281929,
|
||
|
|
"num_tokens": 1871684.0,
|
||
|
|
"step": 955
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.4264805793762205,
|
||
|
|
"epoch": 0.9547488811536549,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 0.0004795,
|
||
|
|
"loss": 7.3865,
|
||
|
|
"mean_token_accuracy": 0.09964759126305581,
|
||
|
|
"num_tokens": 1881956.0,
|
||
|
|
"step": 960
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.41358003616333,
|
||
|
|
"epoch": 0.9597215315763302,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.000482,
|
||
|
|
"loss": 7.462,
|
||
|
|
"mean_token_accuracy": 0.09731132462620735,
|
||
|
|
"num_tokens": 1891860.0,
|
||
|
|
"step": 965
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.419972658157349,
|
||
|
|
"epoch": 0.9646941819990055,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.0004845,
|
||
|
|
"loss": 7.3645,
|
||
|
|
"mean_token_accuracy": 0.10411853864789009,
|
||
|
|
"num_tokens": 1901521.0,
|
||
|
|
"step": 970
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.428276777267456,
|
||
|
|
"epoch": 0.9696668324216807,
|
||
|
|
"grad_norm": 1.2578125,
|
||
|
|
"learning_rate": 0.000487,
|
||
|
|
"loss": 7.3269,
|
||
|
|
"mean_token_accuracy": 0.10307972654700279,
|
||
|
|
"num_tokens": 1910438.0,
|
||
|
|
"step": 975
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.316250371932983,
|
||
|
|
"epoch": 0.9746394828443561,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.0004895,
|
||
|
|
"loss": 7.3629,
|
||
|
|
"mean_token_accuracy": 0.1025395892560482,
|
||
|
|
"num_tokens": 1922045.0,
|
||
|
|
"step": 980
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.4339062690734865,
|
||
|
|
"epoch": 0.9796121332670313,
|
||
|
|
"grad_norm": 1.1640625,
|
||
|
|
"learning_rate": 0.000492,
|
||
|
|
"loss": 7.1867,
|
||
|
|
"mean_token_accuracy": 0.11498386934399604,
|
||
|
|
"num_tokens": 1930330.0,
|
||
|
|
"step": 985
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.193427228927613,
|
||
|
|
"epoch": 0.9845847836897066,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 0.0004945,
|
||
|
|
"loss": 7.1948,
|
||
|
|
"mean_token_accuracy": 0.10757651999592781,
|
||
|
|
"num_tokens": 1939485.0,
|
||
|
|
"step": 990
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.486000108718872,
|
||
|
|
"epoch": 0.989557434112382,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.000497,
|
||
|
|
"loss": 7.2966,
|
||
|
|
"mean_token_accuracy": 0.10132434442639351,
|
||
|
|
"num_tokens": 1949448.0,
|
||
|
|
"step": 995
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.363285207748413,
|
||
|
|
"epoch": 0.9945300845350572,
|
||
|
|
"grad_norm": 0.94921875,
|
||
|
|
"learning_rate": 0.0004995,
|
||
|
|
"loss": 7.3883,
|
||
|
|
"mean_token_accuracy": 0.09904719069600106,
|
||
|
|
"num_tokens": 1959923.0,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9945300845350572,
|
||
|
|
"eval_entropy": 7.191312780028664,
|
||
|
|
"eval_loss": 7.426286697387695,
|
||
|
|
"eval_mean_token_accuracy": 0.09612512418974517,
|
||
|
|
"eval_num_tokens": 1959923.0,
|
||
|
|
"eval_runtime": 13.1868,
|
||
|
|
"eval_samples_per_second": 262.762,
|
||
|
|
"eval_steps_per_second": 32.912,
|
||
|
|
"step": 1000
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 5,
|
||
|
|
"max_steps": 10050,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 10,
|
||
|
|
"save_steps": 500,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": false
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 3220538609664000.0,
|
||
|
|
"train_batch_size": 16,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|