1475 lines
44 KiB
JSON
1475 lines
44 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 0.890224225226729,
|
||
|
|
"eval_steps": 100,
|
||
|
|
"global_step": 1000,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"entropy": 1.0013165885582567,
|
||
|
|
"epoch": 0.00890224225226729,
|
||
|
|
"grad_norm": 4.25,
|
||
|
|
"learning_rate": 6.000000000000001e-07,
|
||
|
|
"loss": 2.1036,
|
||
|
|
"mean_token_accuracy": 0.5934873595833778,
|
||
|
|
"num_tokens": 128102.0,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0327254055067896,
|
||
|
|
"epoch": 0.01780448450453458,
|
||
|
|
"grad_norm": 3.75,
|
||
|
|
"learning_rate": 1.2666666666666669e-06,
|
||
|
|
"loss": 2.1318,
|
||
|
|
"mean_token_accuracy": 0.5815463203936815,
|
||
|
|
"num_tokens": 255758.0,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0127216134220363,
|
||
|
|
"epoch": 0.026706726756801868,
|
||
|
|
"grad_norm": 3.4375,
|
||
|
|
"learning_rate": 1.9333333333333336e-06,
|
||
|
|
"loss": 2.0994,
|
||
|
|
"mean_token_accuracy": 0.5902250200510025,
|
||
|
|
"num_tokens": 392254.0,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9930799461901187,
|
||
|
|
"epoch": 0.03560896900906916,
|
||
|
|
"grad_norm": 3.4375,
|
||
|
|
"learning_rate": 2.6e-06,
|
||
|
|
"loss": 2.0443,
|
||
|
|
"mean_token_accuracy": 0.5956933092325926,
|
||
|
|
"num_tokens": 525545.0,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.011460742354393,
|
||
|
|
"epoch": 0.04451121126133645,
|
||
|
|
"grad_norm": 3.59375,
|
||
|
|
"learning_rate": 3.266666666666667e-06,
|
||
|
|
"loss": 2.0996,
|
||
|
|
"mean_token_accuracy": 0.5882656103000045,
|
||
|
|
"num_tokens": 659462.0,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.030308809503913,
|
||
|
|
"epoch": 0.053413453513603736,
|
||
|
|
"grad_norm": 3.5,
|
||
|
|
"learning_rate": 3.9333333333333335e-06,
|
||
|
|
"loss": 2.0506,
|
||
|
|
"mean_token_accuracy": 0.5891991708427667,
|
||
|
|
"num_tokens": 797469.0,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0320250753313303,
|
||
|
|
"epoch": 0.06231569576587103,
|
||
|
|
"grad_norm": 3.484375,
|
||
|
|
"learning_rate": 4.600000000000001e-06,
|
||
|
|
"loss": 2.018,
|
||
|
|
"mean_token_accuracy": 0.5905353711917997,
|
||
|
|
"num_tokens": 926574.0,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0368458257988096,
|
||
|
|
"epoch": 0.07121793801813832,
|
||
|
|
"grad_norm": 3.0,
|
||
|
|
"learning_rate": 5.2666666666666665e-06,
|
||
|
|
"loss": 1.9841,
|
||
|
|
"mean_token_accuracy": 0.5963305428624153,
|
||
|
|
"num_tokens": 1056145.0,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9638574384152889,
|
||
|
|
"epoch": 0.08012018027040561,
|
||
|
|
"grad_norm": 2.75,
|
||
|
|
"learning_rate": 5.933333333333335e-06,
|
||
|
|
"loss": 1.8996,
|
||
|
|
"mean_token_accuracy": 0.6114491229876876,
|
||
|
|
"num_tokens": 1195941.0,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.98968401607126,
|
||
|
|
"epoch": 0.0890224225226729,
|
||
|
|
"grad_norm": 2.609375,
|
||
|
|
"learning_rate": 6.600000000000001e-06,
|
||
|
|
"loss": 1.855,
|
||
|
|
"mean_token_accuracy": 0.6124794160947203,
|
||
|
|
"num_tokens": 1322822.0,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0890224225226729,
|
||
|
|
"eval_biology_entropy": 1.7304411175251007,
|
||
|
|
"eval_biology_loss": 3.744920492172241,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.3840184355974197,
|
||
|
|
"eval_biology_num_tokens": 1322822.0,
|
||
|
|
"eval_biology_runtime": 19.0021,
|
||
|
|
"eval_biology_samples_per_second": 26.313,
|
||
|
|
"eval_biology_steps_per_second": 13.156,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0890224225226729,
|
||
|
|
"eval_chemistry_entropy": 0.9838921461105347,
|
||
|
|
"eval_chemistry_loss": 1.8955185413360596,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.6147462700605393,
|
||
|
|
"eval_chemistry_num_tokens": 1322822.0,
|
||
|
|
"eval_chemistry_runtime": 24.0155,
|
||
|
|
"eval_chemistry_samples_per_second": 20.82,
|
||
|
|
"eval_chemistry_steps_per_second": 10.41,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0890224225226729,
|
||
|
|
"eval_math_entropy": 0.754765457034111,
|
||
|
|
"eval_math_loss": 2.326155424118042,
|
||
|
|
"eval_math_mean_token_accuracy": 0.591344870209694,
|
||
|
|
"eval_math_num_tokens": 1322822.0,
|
||
|
|
"eval_math_runtime": 24.4976,
|
||
|
|
"eval_math_samples_per_second": 20.41,
|
||
|
|
"eval_math_steps_per_second": 10.205,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0890224225226729,
|
||
|
|
"eval_physics_entropy": 0.9091537948846817,
|
||
|
|
"eval_physics_loss": 1.9409757852554321,
|
||
|
|
"eval_physics_mean_token_accuracy": 0.6192746138572693,
|
||
|
|
"eval_physics_num_tokens": 1322822.0,
|
||
|
|
"eval_physics_runtime": 29.069,
|
||
|
|
"eval_physics_samples_per_second": 17.2,
|
||
|
|
"eval_physics_steps_per_second": 8.6,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0018818765878676,
|
||
|
|
"epoch": 0.09792466477494019,
|
||
|
|
"grad_norm": 2.359375,
|
||
|
|
"learning_rate": 7.266666666666668e-06,
|
||
|
|
"loss": 1.8412,
|
||
|
|
"mean_token_accuracy": 0.6112866658717394,
|
||
|
|
"num_tokens": 1455622.0,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.013153598085046,
|
||
|
|
"epoch": 0.10682690702720747,
|
||
|
|
"grad_norm": 2.078125,
|
||
|
|
"learning_rate": 7.933333333333334e-06,
|
||
|
|
"loss": 1.7484,
|
||
|
|
"mean_token_accuracy": 0.6222657721489668,
|
||
|
|
"num_tokens": 1592594.0,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0879363805055617,
|
||
|
|
"epoch": 0.11572914927947477,
|
||
|
|
"grad_norm": 2.0,
|
||
|
|
"learning_rate": 8.6e-06,
|
||
|
|
"loss": 1.7365,
|
||
|
|
"mean_token_accuracy": 0.6194966306909919,
|
||
|
|
"num_tokens": 1723223.0,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.090317297540605,
|
||
|
|
"epoch": 0.12463139153174206,
|
||
|
|
"grad_norm": 1.859375,
|
||
|
|
"learning_rate": 9.266666666666667e-06,
|
||
|
|
"loss": 1.5882,
|
||
|
|
"mean_token_accuracy": 0.6410850418731571,
|
||
|
|
"num_tokens": 1854187.0,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1811468597501515,
|
||
|
|
"epoch": 0.13353363378400934,
|
||
|
|
"grad_norm": 1.7421875,
|
||
|
|
"learning_rate": 9.933333333333334e-06,
|
||
|
|
"loss": 1.5436,
|
||
|
|
"mean_token_accuracy": 0.6420513158664107,
|
||
|
|
"num_tokens": 1980984.0,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2093722522258759,
|
||
|
|
"epoch": 0.14243587603627664,
|
||
|
|
"grad_norm": 1.359375,
|
||
|
|
"learning_rate": 1.0600000000000002e-05,
|
||
|
|
"loss": 1.4029,
|
||
|
|
"mean_token_accuracy": 0.6615621751174331,
|
||
|
|
"num_tokens": 2114069.0,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2789871104061603,
|
||
|
|
"epoch": 0.15133811828854393,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 1.1266666666666668e-05,
|
||
|
|
"loss": 1.3526,
|
||
|
|
"mean_token_accuracy": 0.673838303796947,
|
||
|
|
"num_tokens": 2248456.0,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.3190820906311274,
|
||
|
|
"epoch": 0.16024036054081123,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 1.1933333333333335e-05,
|
||
|
|
"loss": 1.3278,
|
||
|
|
"mean_token_accuracy": 0.6813167624175549,
|
||
|
|
"num_tokens": 2384381.0,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.3156713977456094,
|
||
|
|
"epoch": 0.1691426027930785,
|
||
|
|
"grad_norm": 0.984375,
|
||
|
|
"learning_rate": 1.2600000000000001e-05,
|
||
|
|
"loss": 1.273,
|
||
|
|
"mean_token_accuracy": 0.6898887110874057,
|
||
|
|
"num_tokens": 2510885.0,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.3284545745700598,
|
||
|
|
"epoch": 0.1780448450453458,
|
||
|
|
"grad_norm": 0.9375,
|
||
|
|
"learning_rate": 1.3266666666666668e-05,
|
||
|
|
"loss": 1.2541,
|
||
|
|
"mean_token_accuracy": 0.6897036660462618,
|
||
|
|
"num_tokens": 2640613.0,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1780448450453458,
|
||
|
|
"eval_biology_entropy": 2.3898276572227477,
|
||
|
|
"eval_biology_loss": 2.6390066146850586,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.4844123523235321,
|
||
|
|
"eval_biology_num_tokens": 2640613.0,
|
||
|
|
"eval_biology_runtime": 18.9294,
|
||
|
|
"eval_biology_samples_per_second": 26.414,
|
||
|
|
"eval_biology_steps_per_second": 13.207,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1780448450453458,
|
||
|
|
"eval_chemistry_entropy": 1.2916195030212403,
|
||
|
|
"eval_chemistry_loss": 1.269665241241455,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.6933187863826752,
|
||
|
|
"eval_chemistry_num_tokens": 2640613.0,
|
||
|
|
"eval_chemistry_runtime": 23.9969,
|
||
|
|
"eval_chemistry_samples_per_second": 20.836,
|
||
|
|
"eval_chemistry_steps_per_second": 10.418,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1780448450453458,
|
||
|
|
"eval_math_entropy": 1.0382833310365678,
|
||
|
|
"eval_math_loss": 1.4177786111831665,
|
||
|
|
"eval_math_mean_token_accuracy": 0.6727467269897461,
|
||
|
|
"eval_math_num_tokens": 2640613.0,
|
||
|
|
"eval_math_runtime": 24.4826,
|
||
|
|
"eval_math_samples_per_second": 20.423,
|
||
|
|
"eval_math_steps_per_second": 10.211,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1780448450453458,
|
||
|
|
"eval_physics_entropy": 1.2059782876968383,
|
||
|
|
"eval_physics_loss": 1.2801809310913086,
|
||
|
|
"eval_physics_mean_token_accuracy": 0.6974429616928101,
|
||
|
|
"eval_physics_num_tokens": 2640613.0,
|
||
|
|
"eval_physics_runtime": 28.6856,
|
||
|
|
"eval_physics_samples_per_second": 17.43,
|
||
|
|
"eval_physics_steps_per_second": 8.715,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.235218785330653,
|
||
|
|
"epoch": 0.18694708729761308,
|
||
|
|
"grad_norm": 0.90234375,
|
||
|
|
"learning_rate": 1.3933333333333334e-05,
|
||
|
|
"loss": 1.1936,
|
||
|
|
"mean_token_accuracy": 0.7037674445658922,
|
||
|
|
"num_tokens": 2779589.0,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2612369772046805,
|
||
|
|
"epoch": 0.19584932954988038,
|
||
|
|
"grad_norm": 0.92578125,
|
||
|
|
"learning_rate": 1.46e-05,
|
||
|
|
"loss": 1.1973,
|
||
|
|
"mean_token_accuracy": 0.6993423495441675,
|
||
|
|
"num_tokens": 2909779.0,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.224467835202813,
|
||
|
|
"epoch": 0.20475157180214767,
|
||
|
|
"grad_norm": 0.9609375,
|
||
|
|
"learning_rate": 1.5266666666666667e-05,
|
||
|
|
"loss": 1.1453,
|
||
|
|
"mean_token_accuracy": 0.710259860381484,
|
||
|
|
"num_tokens": 3045030.0,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2651627618819474,
|
||
|
|
"epoch": 0.21365381405441494,
|
||
|
|
"grad_norm": 0.8515625,
|
||
|
|
"learning_rate": 1.5933333333333336e-05,
|
||
|
|
"loss": 1.1668,
|
||
|
|
"mean_token_accuracy": 0.7021638717502355,
|
||
|
|
"num_tokens": 3180769.0,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2010735258460046,
|
||
|
|
"epoch": 0.22255605630668224,
|
||
|
|
"grad_norm": 0.828125,
|
||
|
|
"learning_rate": 1.66e-05,
|
||
|
|
"loss": 1.1257,
|
||
|
|
"mean_token_accuracy": 0.7146535400301218,
|
||
|
|
"num_tokens": 3317450.0,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2229063987731934,
|
||
|
|
"epoch": 0.23145829855894953,
|
||
|
|
"grad_norm": 0.984375,
|
||
|
|
"learning_rate": 1.726666666666667e-05,
|
||
|
|
"loss": 1.161,
|
||
|
|
"mean_token_accuracy": 0.7094106379896402,
|
||
|
|
"num_tokens": 3448841.0,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1914008114486934,
|
||
|
|
"epoch": 0.24036054081121683,
|
||
|
|
"grad_norm": 0.8828125,
|
||
|
|
"learning_rate": 1.7933333333333333e-05,
|
||
|
|
"loss": 1.091,
|
||
|
|
"mean_token_accuracy": 0.7163884485140443,
|
||
|
|
"num_tokens": 3582810.0,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2218066401779653,
|
||
|
|
"epoch": 0.24926278306348412,
|
||
|
|
"grad_norm": 0.8984375,
|
||
|
|
"learning_rate": 1.86e-05,
|
||
|
|
"loss": 1.1147,
|
||
|
|
"mean_token_accuracy": 0.7146718475967646,
|
||
|
|
"num_tokens": 3713242.0,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2504509769380092,
|
||
|
|
"epoch": 0.2581650253157514,
|
||
|
|
"grad_norm": 0.94140625,
|
||
|
|
"learning_rate": 1.926666666666667e-05,
|
||
|
|
"loss": 1.1535,
|
||
|
|
"mean_token_accuracy": 0.7055120941251516,
|
||
|
|
"num_tokens": 3846944.0,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1762213349342345,
|
||
|
|
"epoch": 0.2670672675680187,
|
||
|
|
"grad_norm": 0.9296875,
|
||
|
|
"learning_rate": 1.9933333333333334e-05,
|
||
|
|
"loss": 1.0892,
|
||
|
|
"mean_token_accuracy": 0.7208757739514112,
|
||
|
|
"num_tokens": 3984712.0,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2670672675680187,
|
||
|
|
"eval_biology_entropy": 2.2724351935386657,
|
||
|
|
"eval_biology_loss": 2.3849706649780273,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.5212258485555649,
|
||
|
|
"eval_biology_num_tokens": 3984712.0,
|
||
|
|
"eval_biology_runtime": 18.9323,
|
||
|
|
"eval_biology_samples_per_second": 26.41,
|
||
|
|
"eval_biology_steps_per_second": 13.205,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2670672675680187,
|
||
|
|
"eval_chemistry_entropy": 1.172815949201584,
|
||
|
|
"eval_chemistry_loss": 1.1251518726348877,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7213010289669037,
|
||
|
|
"eval_chemistry_num_tokens": 3984712.0,
|
||
|
|
"eval_chemistry_runtime": 24.0284,
|
||
|
|
"eval_chemistry_samples_per_second": 20.809,
|
||
|
|
"eval_chemistry_steps_per_second": 10.404,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2670672675680187,
|
||
|
|
"eval_math_entropy": 1.013439424276352,
|
||
|
|
"eval_math_loss": 1.2600808143615723,
|
||
|
|
"eval_math_mean_token_accuracy": 0.7004911189079285,
|
||
|
|
"eval_math_num_tokens": 3984712.0,
|
||
|
|
"eval_math_runtime": 24.488,
|
||
|
|
"eval_math_samples_per_second": 20.418,
|
||
|
|
"eval_math_steps_per_second": 10.209,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2670672675680187,
|
||
|
|
"eval_physics_entropy": 1.146550934791565,
|
||
|
|
"eval_physics_loss": 1.1405712366104126,
|
||
|
|
"eval_physics_mean_token_accuracy": 0.7223960258960724,
|
||
|
|
"eval_physics_num_tokens": 3984712.0,
|
||
|
|
"eval_physics_runtime": 28.6721,
|
||
|
|
"eval_physics_samples_per_second": 17.439,
|
||
|
|
"eval_physics_steps_per_second": 8.719,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1852697040885687,
|
||
|
|
"epoch": 0.275969509820286,
|
||
|
|
"grad_norm": 0.875,
|
||
|
|
"learning_rate": 1.9933333333333334e-05,
|
||
|
|
"loss": 1.0941,
|
||
|
|
"mean_token_accuracy": 0.7201298415660858,
|
||
|
|
"num_tokens": 4126779.0,
|
||
|
|
"step": 310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1581873616203666,
|
||
|
|
"epoch": 0.2848717520725533,
|
||
|
|
"grad_norm": 0.88671875,
|
||
|
|
"learning_rate": 1.985925925925926e-05,
|
||
|
|
"loss": 1.0739,
|
||
|
|
"mean_token_accuracy": 0.7240971345454454,
|
||
|
|
"num_tokens": 4259617.0,
|
||
|
|
"step": 320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1453685130923987,
|
||
|
|
"epoch": 0.29377399432482054,
|
||
|
|
"grad_norm": 0.8359375,
|
||
|
|
"learning_rate": 1.9785185185185187e-05,
|
||
|
|
"loss": 1.0543,
|
||
|
|
"mean_token_accuracy": 0.7301174711436034,
|
||
|
|
"num_tokens": 4391745.0,
|
||
|
|
"step": 330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1414084695279598,
|
||
|
|
"epoch": 0.30267623657708786,
|
||
|
|
"grad_norm": 0.83984375,
|
||
|
|
"learning_rate": 1.971111111111111e-05,
|
||
|
|
"loss": 1.0486,
|
||
|
|
"mean_token_accuracy": 0.7298893980681896,
|
||
|
|
"num_tokens": 4525776.0,
|
||
|
|
"step": 340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.096273798495531,
|
||
|
|
"epoch": 0.31157847882935513,
|
||
|
|
"grad_norm": 0.765625,
|
||
|
|
"learning_rate": 1.963703703703704e-05,
|
||
|
|
"loss": 1.0093,
|
||
|
|
"mean_token_accuracy": 0.7381538521498442,
|
||
|
|
"num_tokens": 4653203.0,
|
||
|
|
"step": 350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.194446151703596,
|
||
|
|
"epoch": 0.32048072108162245,
|
||
|
|
"grad_norm": 0.86328125,
|
||
|
|
"learning_rate": 1.9562962962962964e-05,
|
||
|
|
"loss": 1.0889,
|
||
|
|
"mean_token_accuracy": 0.7179776296019554,
|
||
|
|
"num_tokens": 4786866.0,
|
||
|
|
"step": 360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1112803112715484,
|
||
|
|
"epoch": 0.3293829633338897,
|
||
|
|
"grad_norm": 0.890625,
|
||
|
|
"learning_rate": 1.948888888888889e-05,
|
||
|
|
"loss": 1.019,
|
||
|
|
"mean_token_accuracy": 0.7352581065148115,
|
||
|
|
"num_tokens": 4921793.0,
|
||
|
|
"step": 370
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.145531201735139,
|
||
|
|
"epoch": 0.338285205586157,
|
||
|
|
"grad_norm": 0.71484375,
|
||
|
|
"learning_rate": 1.9414814814814817e-05,
|
||
|
|
"loss": 1.0382,
|
||
|
|
"mean_token_accuracy": 0.7289914164692164,
|
||
|
|
"num_tokens": 5055990.0,
|
||
|
|
"step": 380
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1610372081398963,
|
||
|
|
"epoch": 0.3471874478384243,
|
||
|
|
"grad_norm": 0.78515625,
|
||
|
|
"learning_rate": 1.9340740740740743e-05,
|
||
|
|
"loss": 1.0628,
|
||
|
|
"mean_token_accuracy": 0.7227894127368927,
|
||
|
|
"num_tokens": 5182611.0,
|
||
|
|
"step": 390
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1239100523293017,
|
||
|
|
"epoch": 0.3560896900906916,
|
||
|
|
"grad_norm": 0.828125,
|
||
|
|
"learning_rate": 1.926666666666667e-05,
|
||
|
|
"loss": 1.0403,
|
||
|
|
"mean_token_accuracy": 0.729755662754178,
|
||
|
|
"num_tokens": 5313795.0,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3560896900906916,
|
||
|
|
"eval_biology_entropy": 2.1833299980163576,
|
||
|
|
"eval_biology_loss": 2.2870590686798096,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.5356344033479691,
|
||
|
|
"eval_biology_num_tokens": 5313795.0,
|
||
|
|
"eval_biology_runtime": 18.975,
|
||
|
|
"eval_biology_samples_per_second": 26.35,
|
||
|
|
"eval_biology_steps_per_second": 13.175,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3560896900906916,
|
||
|
|
"eval_chemistry_entropy": 1.1136225144863128,
|
||
|
|
"eval_chemistry_loss": 1.0691337585449219,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7317903306484222,
|
||
|
|
"eval_chemistry_num_tokens": 5313795.0,
|
||
|
|
"eval_chemistry_runtime": 24.7923,
|
||
|
|
"eval_chemistry_samples_per_second": 20.168,
|
||
|
|
"eval_chemistry_steps_per_second": 10.084,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3560896900906916,
|
||
|
|
"eval_math_entropy": 0.9845062309503555,
|
||
|
|
"eval_math_loss": 1.2056244611740112,
|
||
|
|
"eval_math_mean_token_accuracy": 0.711446602344513,
|
||
|
|
"eval_math_num_tokens": 5313795.0,
|
||
|
|
"eval_math_runtime": 24.5234,
|
||
|
|
"eval_math_samples_per_second": 20.389,
|
||
|
|
"eval_math_steps_per_second": 10.194,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3560896900906916,
|
||
|
|
"eval_physics_entropy": 1.108200379371643,
|
||
|
|
"eval_physics_loss": 1.0871150493621826,
|
||
|
|
"eval_physics_mean_token_accuracy": 0.7332238068580628,
|
||
|
|
"eval_physics_num_tokens": 5313795.0,
|
||
|
|
"eval_physics_runtime": 28.6972,
|
||
|
|
"eval_physics_samples_per_second": 17.423,
|
||
|
|
"eval_physics_steps_per_second": 8.712,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0743986388668418,
|
||
|
|
"epoch": 0.3649919323429589,
|
||
|
|
"grad_norm": 0.86328125,
|
||
|
|
"learning_rate": 1.9192592592592593e-05,
|
||
|
|
"loss": 0.9898,
|
||
|
|
"mean_token_accuracy": 0.7425458256155253,
|
||
|
|
"num_tokens": 5447087.0,
|
||
|
|
"step": 410
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1058754766359926,
|
||
|
|
"epoch": 0.37389417459522617,
|
||
|
|
"grad_norm": 0.72265625,
|
||
|
|
"learning_rate": 1.911851851851852e-05,
|
||
|
|
"loss": 1.0094,
|
||
|
|
"mean_token_accuracy": 0.7370415050536394,
|
||
|
|
"num_tokens": 5573445.0,
|
||
|
|
"step": 420
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0946383344009518,
|
||
|
|
"epoch": 0.38279641684749344,
|
||
|
|
"grad_norm": 0.85546875,
|
||
|
|
"learning_rate": 1.9044444444444446e-05,
|
||
|
|
"loss": 1.0192,
|
||
|
|
"mean_token_accuracy": 0.7376058127731084,
|
||
|
|
"num_tokens": 5703640.0,
|
||
|
|
"step": 430
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1165539000183344,
|
||
|
|
"epoch": 0.39169865909976076,
|
||
|
|
"grad_norm": 0.7109375,
|
||
|
|
"learning_rate": 1.8970370370370372e-05,
|
||
|
|
"loss": 1.0148,
|
||
|
|
"mean_token_accuracy": 0.7342595763504505,
|
||
|
|
"num_tokens": 5836274.0,
|
||
|
|
"step": 440
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1028259133920073,
|
||
|
|
"epoch": 0.400600901352028,
|
||
|
|
"grad_norm": 0.8046875,
|
||
|
|
"learning_rate": 1.8896296296296295e-05,
|
||
|
|
"loss": 1.009,
|
||
|
|
"mean_token_accuracy": 0.7364393822848797,
|
||
|
|
"num_tokens": 5967215.0,
|
||
|
|
"step": 450
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0587953502312302,
|
||
|
|
"epoch": 0.40950314360429535,
|
||
|
|
"grad_norm": 0.734375,
|
||
|
|
"learning_rate": 1.8822222222222225e-05,
|
||
|
|
"loss": 0.9688,
|
||
|
|
"mean_token_accuracy": 0.746485960483551,
|
||
|
|
"num_tokens": 6101944.0,
|
||
|
|
"step": 460
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1212815966457128,
|
||
|
|
"epoch": 0.4184053858565626,
|
||
|
|
"grad_norm": 0.85546875,
|
||
|
|
"learning_rate": 1.874814814814815e-05,
|
||
|
|
"loss": 1.0258,
|
||
|
|
"mean_token_accuracy": 0.733516975864768,
|
||
|
|
"num_tokens": 6231388.0,
|
||
|
|
"step": 470
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0713276099413633,
|
||
|
|
"epoch": 0.4273076281088299,
|
||
|
|
"grad_norm": 0.73828125,
|
||
|
|
"learning_rate": 1.8674074074074075e-05,
|
||
|
|
"loss": 0.9709,
|
||
|
|
"mean_token_accuracy": 0.7428883332759142,
|
||
|
|
"num_tokens": 6363951.0,
|
||
|
|
"step": 480
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.09543631747365,
|
||
|
|
"epoch": 0.4362098703610972,
|
||
|
|
"grad_norm": 0.71875,
|
||
|
|
"learning_rate": 1.86e-05,
|
||
|
|
"loss": 1.01,
|
||
|
|
"mean_token_accuracy": 0.735943990573287,
|
||
|
|
"num_tokens": 6495585.0,
|
||
|
|
"step": 490
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0952462505549192,
|
||
|
|
"epoch": 0.4451121126133645,
|
||
|
|
"grad_norm": 0.76171875,
|
||
|
|
"learning_rate": 1.8525925925925928e-05,
|
||
|
|
"loss": 1.0042,
|
||
|
|
"mean_token_accuracy": 0.7367089048027993,
|
||
|
|
"num_tokens": 6622502.0,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4451121126133645,
|
||
|
|
"eval_biology_entropy": 2.1483897223472597,
|
||
|
|
"eval_biology_loss": 2.2432360649108887,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.5430341627597809,
|
||
|
|
"eval_biology_num_tokens": 6622502.0,
|
||
|
|
"eval_biology_runtime": 19.1686,
|
||
|
|
"eval_biology_samples_per_second": 26.084,
|
||
|
|
"eval_biology_steps_per_second": 13.042,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4451121126133645,
|
||
|
|
"eval_chemistry_entropy": 1.091737098813057,
|
||
|
|
"eval_chemistry_loss": 1.040204644203186,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7374620385169983,
|
||
|
|
"eval_chemistry_num_tokens": 6622502.0,
|
||
|
|
"eval_chemistry_runtime": 23.9529,
|
||
|
|
"eval_chemistry_samples_per_second": 20.874,
|
||
|
|
"eval_chemistry_steps_per_second": 10.437,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4451121126133645,
|
||
|
|
"eval_math_entropy": 0.9775256743431091,
|
||
|
|
"eval_math_loss": 1.186624526977539,
|
||
|
|
"eval_math_mean_token_accuracy": 0.7151267173290252,
|
||
|
|
"eval_math_num_tokens": 6622502.0,
|
||
|
|
"eval_math_runtime": 24.4031,
|
||
|
|
"eval_math_samples_per_second": 20.489,
|
||
|
|
"eval_math_steps_per_second": 10.245,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4451121126133645,
|
||
|
|
"eval_physics_entropy": 1.0940199873447418,
|
||
|
|
"eval_physics_loss": 1.0634400844573975,
|
||
|
|
"eval_physics_mean_token_accuracy": 0.737773185968399,
|
||
|
|
"eval_physics_num_tokens": 6622502.0,
|
||
|
|
"eval_physics_runtime": 28.632,
|
||
|
|
"eval_physics_samples_per_second": 17.463,
|
||
|
|
"eval_physics_steps_per_second": 8.731,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0699838446453214,
|
||
|
|
"epoch": 0.4540143548656318,
|
||
|
|
"grad_norm": 0.75,
|
||
|
|
"learning_rate": 1.8451851851851855e-05,
|
||
|
|
"loss": 0.9686,
|
||
|
|
"mean_token_accuracy": 0.7425621785223484,
|
||
|
|
"num_tokens": 6751161.0,
|
||
|
|
"step": 510
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.086439424008131,
|
||
|
|
"epoch": 0.46291659711789906,
|
||
|
|
"grad_norm": 0.7265625,
|
||
|
|
"learning_rate": 1.8377777777777778e-05,
|
||
|
|
"loss": 0.9859,
|
||
|
|
"mean_token_accuracy": 0.7420764189213515,
|
||
|
|
"num_tokens": 6881616.0,
|
||
|
|
"step": 520
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.067279956676066,
|
||
|
|
"epoch": 0.4718188393701664,
|
||
|
|
"grad_norm": 0.765625,
|
||
|
|
"learning_rate": 1.8303703703703704e-05,
|
||
|
|
"loss": 0.9765,
|
||
|
|
"mean_token_accuracy": 0.7449497696012258,
|
||
|
|
"num_tokens": 7017868.0,
|
||
|
|
"step": 530
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.091121963970363,
|
||
|
|
"epoch": 0.48072108162243365,
|
||
|
|
"grad_norm": 0.7578125,
|
||
|
|
"learning_rate": 1.822962962962963e-05,
|
||
|
|
"loss": 1.0114,
|
||
|
|
"mean_token_accuracy": 0.7365596149116754,
|
||
|
|
"num_tokens": 7151294.0,
|
||
|
|
"step": 540
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0589939182624222,
|
||
|
|
"epoch": 0.4896233238747009,
|
||
|
|
"grad_norm": 0.70703125,
|
||
|
|
"learning_rate": 1.8155555555555557e-05,
|
||
|
|
"loss": 0.9649,
|
||
|
|
"mean_token_accuracy": 0.7470602467656136,
|
||
|
|
"num_tokens": 7295452.0,
|
||
|
|
"step": 550
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0378610417246819,
|
||
|
|
"epoch": 0.49852556612696824,
|
||
|
|
"grad_norm": 0.80078125,
|
||
|
|
"learning_rate": 1.8081481481481484e-05,
|
||
|
|
"loss": 0.9457,
|
||
|
|
"mean_token_accuracy": 0.7502593748271466,
|
||
|
|
"num_tokens": 7427354.0,
|
||
|
|
"step": 560
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0780232414603232,
|
||
|
|
"epoch": 0.5074278083792355,
|
||
|
|
"grad_norm": 0.69921875,
|
||
|
|
"learning_rate": 1.800740740740741e-05,
|
||
|
|
"loss": 0.9837,
|
||
|
|
"mean_token_accuracy": 0.7401763413101434,
|
||
|
|
"num_tokens": 7558252.0,
|
||
|
|
"step": 570
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0984149038791657,
|
||
|
|
"epoch": 0.5163300506315028,
|
||
|
|
"grad_norm": 0.8046875,
|
||
|
|
"learning_rate": 1.7933333333333333e-05,
|
||
|
|
"loss": 0.9951,
|
||
|
|
"mean_token_accuracy": 0.737068784981966,
|
||
|
|
"num_tokens": 7688220.0,
|
||
|
|
"step": 580
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0815258041024207,
|
||
|
|
"epoch": 0.52523229288377,
|
||
|
|
"grad_norm": 0.80859375,
|
||
|
|
"learning_rate": 1.785925925925926e-05,
|
||
|
|
"loss": 0.9935,
|
||
|
|
"mean_token_accuracy": 0.7424198940396309,
|
||
|
|
"num_tokens": 7819813.0,
|
||
|
|
"step": 590
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0907988570630551,
|
||
|
|
"epoch": 0.5341345351360374,
|
||
|
|
"grad_norm": 0.859375,
|
||
|
|
"learning_rate": 1.7785185185185186e-05,
|
||
|
|
"loss": 0.9922,
|
||
|
|
"mean_token_accuracy": 0.739279318973422,
|
||
|
|
"num_tokens": 7951372.0,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5341345351360374,
|
||
|
|
"eval_biology_entropy": 2.1178611373901366,
|
||
|
|
"eval_biology_loss": 2.214088201522827,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.5480384603738785,
|
||
|
|
"eval_biology_num_tokens": 7951372.0,
|
||
|
|
"eval_biology_runtime": 18.9891,
|
||
|
|
"eval_biology_samples_per_second": 26.331,
|
||
|
|
"eval_biology_steps_per_second": 13.165,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5341345351360374,
|
||
|
|
"eval_chemistry_entropy": 1.0694828644990921,
|
||
|
|
"eval_chemistry_loss": 1.0220880508422852,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7412470579147339,
|
||
|
|
"eval_chemistry_num_tokens": 7951372.0,
|
||
|
|
"eval_chemistry_runtime": 24.074,
|
||
|
|
"eval_chemistry_samples_per_second": 20.769,
|
||
|
|
"eval_chemistry_steps_per_second": 10.385,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5341345351360374,
|
||
|
|
"eval_math_entropy": 0.9660351483821868,
|
||
|
|
"eval_math_loss": 1.1651713848114014,
|
||
|
|
"eval_math_mean_token_accuracy": 0.7203534297943115,
|
||
|
|
"eval_math_num_tokens": 7951372.0,
|
||
|
|
"eval_math_runtime": 24.6005,
|
||
|
|
"eval_math_samples_per_second": 20.325,
|
||
|
|
"eval_math_steps_per_second": 10.162,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5341345351360374,
|
||
|
|
"eval_physics_entropy": 1.0763019816875459,
|
||
|
|
"eval_physics_loss": 1.047385573387146,
|
||
|
|
"eval_physics_mean_token_accuracy": 0.741036681652069,
|
||
|
|
"eval_physics_num_tokens": 7951372.0,
|
||
|
|
"eval_physics_runtime": 28.7125,
|
||
|
|
"eval_physics_samples_per_second": 17.414,
|
||
|
|
"eval_physics_steps_per_second": 8.707,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0812974084168672,
|
||
|
|
"epoch": 0.5430367773883047,
|
||
|
|
"grad_norm": 0.79296875,
|
||
|
|
"learning_rate": 1.7711111111111113e-05,
|
||
|
|
"loss": 0.9848,
|
||
|
|
"mean_token_accuracy": 0.7414944332093001,
|
||
|
|
"num_tokens": 8086060.0,
|
||
|
|
"step": 610
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0596148006618022,
|
||
|
|
"epoch": 0.551939019640572,
|
||
|
|
"grad_norm": 0.7265625,
|
||
|
|
"learning_rate": 1.763703703703704e-05,
|
||
|
|
"loss": 0.981,
|
||
|
|
"mean_token_accuracy": 0.7439363609999419,
|
||
|
|
"num_tokens": 8224064.0,
|
||
|
|
"step": 620
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0637529838830233,
|
||
|
|
"epoch": 0.5608412618928392,
|
||
|
|
"grad_norm": 0.8828125,
|
||
|
|
"learning_rate": 1.7562962962962962e-05,
|
||
|
|
"loss": 0.9797,
|
||
|
|
"mean_token_accuracy": 0.743705640360713,
|
||
|
|
"num_tokens": 8347686.0,
|
||
|
|
"step": 630
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.04844657368958,
|
||
|
|
"epoch": 0.5697435041451065,
|
||
|
|
"grad_norm": 0.74609375,
|
||
|
|
"learning_rate": 1.7488888888888892e-05,
|
||
|
|
"loss": 0.9901,
|
||
|
|
"mean_token_accuracy": 0.7446227420121432,
|
||
|
|
"num_tokens": 8479025.0,
|
||
|
|
"step": 640
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0554169865325094,
|
||
|
|
"epoch": 0.5786457463973739,
|
||
|
|
"grad_norm": 0.9453125,
|
||
|
|
"learning_rate": 1.7414814814814815e-05,
|
||
|
|
"loss": 0.963,
|
||
|
|
"mean_token_accuracy": 0.7450377620756626,
|
||
|
|
"num_tokens": 8604855.0,
|
||
|
|
"step": 650
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0970475694164634,
|
||
|
|
"epoch": 0.5875479886496411,
|
||
|
|
"grad_norm": 0.921875,
|
||
|
|
"learning_rate": 1.7340740740740742e-05,
|
||
|
|
"loss": 0.9982,
|
||
|
|
"mean_token_accuracy": 0.7373595230281353,
|
||
|
|
"num_tokens": 8739828.0,
|
||
|
|
"step": 660
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0422217467799784,
|
||
|
|
"epoch": 0.5964502309019084,
|
||
|
|
"grad_norm": 0.77734375,
|
||
|
|
"learning_rate": 1.726666666666667e-05,
|
||
|
|
"loss": 0.96,
|
||
|
|
"mean_token_accuracy": 0.7486237779259681,
|
||
|
|
"num_tokens": 8868152.0,
|
||
|
|
"step": 670
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0182484578341247,
|
||
|
|
"epoch": 0.6053524731541757,
|
||
|
|
"grad_norm": 0.71484375,
|
||
|
|
"learning_rate": 1.7192592592592595e-05,
|
||
|
|
"loss": 0.9422,
|
||
|
|
"mean_token_accuracy": 0.749386464804411,
|
||
|
|
"num_tokens": 9006122.0,
|
||
|
|
"step": 680
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0727280082181097,
|
||
|
|
"epoch": 0.614254715406443,
|
||
|
|
"grad_norm": 0.8046875,
|
||
|
|
"learning_rate": 1.711851851851852e-05,
|
||
|
|
"loss": 0.9849,
|
||
|
|
"mean_token_accuracy": 0.7426282811909914,
|
||
|
|
"num_tokens": 9135332.0,
|
||
|
|
"step": 690
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0424362102523446,
|
||
|
|
"epoch": 0.6231569576587103,
|
||
|
|
"grad_norm": 0.82421875,
|
||
|
|
"learning_rate": 1.7044444444444445e-05,
|
||
|
|
"loss": 0.9488,
|
||
|
|
"mean_token_accuracy": 0.7479008056223393,
|
||
|
|
"num_tokens": 9273088.0,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6231569576587103,
|
||
|
|
"eval_biology_entropy": 2.1061350786685944,
|
||
|
|
"eval_biology_loss": 2.1926093101501465,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.550960058927536,
|
||
|
|
"eval_biology_num_tokens": 9273088.0,
|
||
|
|
"eval_biology_runtime": 19.0758,
|
||
|
|
"eval_biology_samples_per_second": 26.211,
|
||
|
|
"eval_biology_steps_per_second": 13.106,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6231569576587103,
|
||
|
|
"eval_chemistry_entropy": 1.0614726347923278,
|
||
|
|
"eval_chemistry_loss": 1.0082380771636963,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7441184866428375,
|
||
|
|
"eval_chemistry_num_tokens": 9273088.0,
|
||
|
|
"eval_chemistry_runtime": 24.1338,
|
||
|
|
"eval_chemistry_samples_per_second": 20.718,
|
||
|
|
"eval_chemistry_steps_per_second": 10.359,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6231569576587103,
|
||
|
|
"eval_math_entropy": 0.9611164342164993,
|
||
|
|
"eval_math_loss": 1.1547077894210815,
|
||
|
|
"eval_math_mean_token_accuracy": 0.7228360230922699,
|
||
|
|
"eval_math_num_tokens": 9273088.0,
|
||
|
|
"eval_math_runtime": 24.528,
|
||
|
|
"eval_math_samples_per_second": 20.385,
|
||
|
|
"eval_math_steps_per_second": 10.192,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6231569576587103,
|
||
|
|
"eval_physics_entropy": 1.0682205600738526,
|
||
|
|
"eval_physics_loss": 1.0357767343521118,
|
||
|
|
"eval_physics_mean_token_accuracy": 0.743281735420227,
|
||
|
|
"eval_physics_num_tokens": 9273088.0,
|
||
|
|
"eval_physics_runtime": 28.7501,
|
||
|
|
"eval_physics_samples_per_second": 17.391,
|
||
|
|
"eval_physics_steps_per_second": 8.696,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0483794966712594,
|
||
|
|
"epoch": 0.6320591999109776,
|
||
|
|
"grad_norm": 0.8203125,
|
||
|
|
"learning_rate": 1.697037037037037e-05,
|
||
|
|
"loss": 0.9756,
|
||
|
|
"mean_token_accuracy": 0.7465494040399789,
|
||
|
|
"num_tokens": 9405858.0,
|
||
|
|
"step": 710
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0315969336777926,
|
||
|
|
"epoch": 0.6409614421632449,
|
||
|
|
"grad_norm": 0.72265625,
|
||
|
|
"learning_rate": 1.6896296296296298e-05,
|
||
|
|
"loss": 0.9318,
|
||
|
|
"mean_token_accuracy": 0.7522267319262028,
|
||
|
|
"num_tokens": 9540719.0,
|
||
|
|
"step": 720
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0228697720915079,
|
||
|
|
"epoch": 0.6498636844155121,
|
||
|
|
"grad_norm": 0.7109375,
|
||
|
|
"learning_rate": 1.6822222222222224e-05,
|
||
|
|
"loss": 0.9206,
|
||
|
|
"mean_token_accuracy": 0.754125602915883,
|
||
|
|
"num_tokens": 9674589.0,
|
||
|
|
"step": 730
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0417734686285258,
|
||
|
|
"epoch": 0.6587659266677794,
|
||
|
|
"grad_norm": 0.734375,
|
||
|
|
"learning_rate": 1.6748148148148147e-05,
|
||
|
|
"loss": 0.963,
|
||
|
|
"mean_token_accuracy": 0.7486015625298024,
|
||
|
|
"num_tokens": 9805714.0,
|
||
|
|
"step": 740
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0292632792145013,
|
||
|
|
"epoch": 0.6676681689200468,
|
||
|
|
"grad_norm": 0.78125,
|
||
|
|
"learning_rate": 1.6674074074074077e-05,
|
||
|
|
"loss": 0.9268,
|
||
|
|
"mean_token_accuracy": 0.7511202458292245,
|
||
|
|
"num_tokens": 9933853.0,
|
||
|
|
"step": 750
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0266933845356108,
|
||
|
|
"epoch": 0.676570411172314,
|
||
|
|
"grad_norm": 0.77734375,
|
||
|
|
"learning_rate": 1.66e-05,
|
||
|
|
"loss": 0.948,
|
||
|
|
"mean_token_accuracy": 0.7509643025696278,
|
||
|
|
"num_tokens": 10073440.0,
|
||
|
|
"step": 760
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9969334525987505,
|
||
|
|
"epoch": 0.6854726534245813,
|
||
|
|
"grad_norm": 0.80078125,
|
||
|
|
"learning_rate": 1.6525925925925927e-05,
|
||
|
|
"loss": 0.9175,
|
||
|
|
"mean_token_accuracy": 0.7583379317075014,
|
||
|
|
"num_tokens": 10211274.0,
|
||
|
|
"step": 770
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0174086516723038,
|
||
|
|
"epoch": 0.6943748956768486,
|
||
|
|
"grad_norm": 0.75390625,
|
||
|
|
"learning_rate": 1.6451851851851853e-05,
|
||
|
|
"loss": 0.9262,
|
||
|
|
"mean_token_accuracy": 0.75306093133986,
|
||
|
|
"num_tokens": 10342522.0,
|
||
|
|
"step": 780
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0142306819558145,
|
||
|
|
"epoch": 0.703277137929116,
|
||
|
|
"grad_norm": 0.79296875,
|
||
|
|
"learning_rate": 1.637777777777778e-05,
|
||
|
|
"loss": 0.9373,
|
||
|
|
"mean_token_accuracy": 0.7523925706744194,
|
||
|
|
"num_tokens": 10478874.0,
|
||
|
|
"step": 790
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0250910749658941,
|
||
|
|
"epoch": 0.7121793801813832,
|
||
|
|
"grad_norm": 0.73046875,
|
||
|
|
"learning_rate": 1.6303703703703706e-05,
|
||
|
|
"loss": 0.9313,
|
||
|
|
"mean_token_accuracy": 0.7509257387369871,
|
||
|
|
"num_tokens": 10611038.0,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7121793801813832,
|
||
|
|
"eval_biology_entropy": 2.069473657131195,
|
||
|
|
"eval_biology_loss": 2.180377960205078,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.5537625602483749,
|
||
|
|
"eval_biology_num_tokens": 10611038.0,
|
||
|
|
"eval_biology_runtime": 18.989,
|
||
|
|
"eval_biology_samples_per_second": 26.331,
|
||
|
|
"eval_biology_steps_per_second": 13.166,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7121793801813832,
|
||
|
|
"eval_chemistry_entropy": 1.0416202774047851,
|
||
|
|
"eval_chemistry_loss": 0.9986625909805298,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7456459767818451,
|
||
|
|
"eval_chemistry_num_tokens": 10611038.0,
|
||
|
|
"eval_chemistry_runtime": 24.1612,
|
||
|
|
"eval_chemistry_samples_per_second": 20.694,
|
||
|
|
"eval_chemistry_steps_per_second": 10.347,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7121793801813832,
|
||
|
|
"eval_math_entropy": 0.9473755178451538,
|
||
|
|
"eval_math_loss": 1.1492873430252075,
|
||
|
|
"eval_math_mean_token_accuracy": 0.7247429151535034,
|
||
|
|
"eval_math_num_tokens": 10611038.0,
|
||
|
|
"eval_math_runtime": 24.5113,
|
||
|
|
"eval_math_samples_per_second": 20.399,
|
||
|
|
"eval_math_steps_per_second": 10.199,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7121793801813832,
|
||
|
|
"eval_physics_entropy": 1.0511555817127227,
|
||
|
|
"eval_physics_loss": 1.0280060768127441,
|
||
|
|
"eval_physics_mean_token_accuracy": 0.7451158492565155,
|
||
|
|
"eval_physics_num_tokens": 10611038.0,
|
||
|
|
"eval_physics_runtime": 28.6841,
|
||
|
|
"eval_physics_samples_per_second": 17.431,
|
||
|
|
"eval_physics_steps_per_second": 8.716,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9938568271696567,
|
||
|
|
"epoch": 0.7210816224336505,
|
||
|
|
"grad_norm": 0.75390625,
|
||
|
|
"learning_rate": 1.622962962962963e-05,
|
||
|
|
"loss": 0.9057,
|
||
|
|
"mean_token_accuracy": 0.7613154832273722,
|
||
|
|
"num_tokens": 10741681.0,
|
||
|
|
"step": 810
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0242274224758148,
|
||
|
|
"epoch": 0.7299838646859178,
|
||
|
|
"grad_norm": 0.83984375,
|
||
|
|
"learning_rate": 1.6155555555555556e-05,
|
||
|
|
"loss": 0.9589,
|
||
|
|
"mean_token_accuracy": 0.7503436904400587,
|
||
|
|
"num_tokens": 10874184.0,
|
||
|
|
"step": 820
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0636370442807674,
|
||
|
|
"epoch": 0.738886106938185,
|
||
|
|
"grad_norm": 0.73046875,
|
||
|
|
"learning_rate": 1.6081481481481482e-05,
|
||
|
|
"loss": 0.965,
|
||
|
|
"mean_token_accuracy": 0.7441002728417516,
|
||
|
|
"num_tokens": 11006058.0,
|
||
|
|
"step": 830
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.006078546307981,
|
||
|
|
"epoch": 0.7477883491904523,
|
||
|
|
"grad_norm": 0.73828125,
|
||
|
|
"learning_rate": 1.600740740740741e-05,
|
||
|
|
"loss": 0.9198,
|
||
|
|
"mean_token_accuracy": 0.7565505605190992,
|
||
|
|
"num_tokens": 11140344.0,
|
||
|
|
"step": 840
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.041573415696621,
|
||
|
|
"epoch": 0.7566905914427197,
|
||
|
|
"grad_norm": 0.8828125,
|
||
|
|
"learning_rate": 1.5933333333333336e-05,
|
||
|
|
"loss": 0.9585,
|
||
|
|
"mean_token_accuracy": 0.746536037698388,
|
||
|
|
"num_tokens": 11273755.0,
|
||
|
|
"step": 850
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0404196836054325,
|
||
|
|
"epoch": 0.7655928336949869,
|
||
|
|
"grad_norm": 0.75390625,
|
||
|
|
"learning_rate": 1.5859259259259262e-05,
|
||
|
|
"loss": 0.9599,
|
||
|
|
"mean_token_accuracy": 0.7491594787687064,
|
||
|
|
"num_tokens": 11411558.0,
|
||
|
|
"step": 860
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0480362083762884,
|
||
|
|
"epoch": 0.7744950759472542,
|
||
|
|
"grad_norm": 0.71484375,
|
||
|
|
"learning_rate": 1.5785185185185185e-05,
|
||
|
|
"loss": 0.9531,
|
||
|
|
"mean_token_accuracy": 0.7495062164962292,
|
||
|
|
"num_tokens": 11542129.0,
|
||
|
|
"step": 870
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0487740390002727,
|
||
|
|
"epoch": 0.7833973181995215,
|
||
|
|
"grad_norm": 0.77734375,
|
||
|
|
"learning_rate": 1.571111111111111e-05,
|
||
|
|
"loss": 0.9719,
|
||
|
|
"mean_token_accuracy": 0.7440535627305508,
|
||
|
|
"num_tokens": 11676795.0,
|
||
|
|
"step": 880
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9984678329899908,
|
||
|
|
"epoch": 0.7922995604517888,
|
||
|
|
"grad_norm": 0.82421875,
|
||
|
|
"learning_rate": 1.5637037037037038e-05,
|
||
|
|
"loss": 0.9075,
|
||
|
|
"mean_token_accuracy": 0.7598592435941101,
|
||
|
|
"num_tokens": 11813290.0,
|
||
|
|
"step": 890
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.067121583223343,
|
||
|
|
"epoch": 0.801201802704056,
|
||
|
|
"grad_norm": 0.7890625,
|
||
|
|
"learning_rate": 1.5562962962962965e-05,
|
||
|
|
"loss": 0.9839,
|
||
|
|
"mean_token_accuracy": 0.7415191430598498,
|
||
|
|
"num_tokens": 11939057.0,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.801201802704056,
|
||
|
|
"eval_biology_entropy": 2.0663459722995756,
|
||
|
|
"eval_biology_loss": 2.1661489009857178,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.5559320805072785,
|
||
|
|
"eval_biology_num_tokens": 11939057.0,
|
||
|
|
"eval_biology_runtime": 18.9535,
|
||
|
|
"eval_biology_samples_per_second": 26.38,
|
||
|
|
"eval_biology_steps_per_second": 13.19,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.801201802704056,
|
||
|
|
"eval_chemistry_entropy": 1.037217586994171,
|
||
|
|
"eval_chemistry_loss": 0.9907548427581787,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7475485587120057,
|
||
|
|
"eval_chemistry_num_tokens": 11939057.0,
|
||
|
|
"eval_chemistry_runtime": 24.0512,
|
||
|
|
"eval_chemistry_samples_per_second": 20.789,
|
||
|
|
"eval_chemistry_steps_per_second": 10.394,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.801201802704056,
|
||
|
|
"eval_math_entropy": 0.945040546298027,
|
||
|
|
"eval_math_loss": 1.1432464122772217,
|
||
|
|
"eval_math_mean_token_accuracy": 0.7263167641162872,
|
||
|
|
"eval_math_num_tokens": 11939057.0,
|
||
|
|
"eval_math_runtime": 24.5059,
|
||
|
|
"eval_math_samples_per_second": 20.403,
|
||
|
|
"eval_math_steps_per_second": 10.202,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.801201802704056,
|
||
|
|
"eval_physics_entropy": 1.0482188910245895,
|
||
|
|
"eval_physics_loss": 1.0214959383010864,
|
||
|
|
"eval_physics_mean_token_accuracy": 0.7463257222175598,
|
||
|
|
"eval_physics_num_tokens": 11939057.0,
|
||
|
|
"eval_physics_runtime": 28.687,
|
||
|
|
"eval_physics_samples_per_second": 17.43,
|
||
|
|
"eval_physics_steps_per_second": 8.715,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.022822911851108,
|
||
|
|
"epoch": 0.8101040449563234,
|
||
|
|
"grad_norm": 0.66796875,
|
||
|
|
"learning_rate": 1.548888888888889e-05,
|
||
|
|
"loss": 0.9437,
|
||
|
|
"mean_token_accuracy": 0.7512454934418201,
|
||
|
|
"num_tokens": 12071285.0,
|
||
|
|
"step": 910
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0286116361618043,
|
||
|
|
"epoch": 0.8190062872085907,
|
||
|
|
"grad_norm": 0.75,
|
||
|
|
"learning_rate": 1.5414814814814814e-05,
|
||
|
|
"loss": 0.9501,
|
||
|
|
"mean_token_accuracy": 0.7485235787928104,
|
||
|
|
"num_tokens": 12197416.0,
|
||
|
|
"step": 920
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.008828422985971,
|
||
|
|
"epoch": 0.8279085294608579,
|
||
|
|
"grad_norm": 0.734375,
|
||
|
|
"learning_rate": 1.5340740740740744e-05,
|
||
|
|
"loss": 0.9133,
|
||
|
|
"mean_token_accuracy": 0.7554512336850167,
|
||
|
|
"num_tokens": 12338376.0,
|
||
|
|
"step": 930
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9990727547556162,
|
||
|
|
"epoch": 0.8368107717131252,
|
||
|
|
"grad_norm": 0.765625,
|
||
|
|
"learning_rate": 1.5266666666666667e-05,
|
||
|
|
"loss": 0.9257,
|
||
|
|
"mean_token_accuracy": 0.7566340938210487,
|
||
|
|
"num_tokens": 12481276.0,
|
||
|
|
"step": 940
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.023878287896514,
|
||
|
|
"epoch": 0.8457130139653926,
|
||
|
|
"grad_norm": 0.73046875,
|
||
|
|
"learning_rate": 1.5192592592592594e-05,
|
||
|
|
"loss": 0.9283,
|
||
|
|
"mean_token_accuracy": 0.7518570881336928,
|
||
|
|
"num_tokens": 12614378.0,
|
||
|
|
"step": 950
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.019566267542541,
|
||
|
|
"epoch": 0.8546152562176598,
|
||
|
|
"grad_norm": 0.70703125,
|
||
|
|
"learning_rate": 1.5118518518518519e-05,
|
||
|
|
"loss": 0.9469,
|
||
|
|
"mean_token_accuracy": 0.753068869188428,
|
||
|
|
"num_tokens": 12749299.0,
|
||
|
|
"step": 960
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9727985240519047,
|
||
|
|
"epoch": 0.8635174984699271,
|
||
|
|
"grad_norm": 0.69921875,
|
||
|
|
"learning_rate": 1.5044444444444445e-05,
|
||
|
|
"loss": 0.8757,
|
||
|
|
"mean_token_accuracy": 0.7642812915146351,
|
||
|
|
"num_tokens": 12880725.0,
|
||
|
|
"step": 970
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.023832948692143,
|
||
|
|
"epoch": 0.8724197407221944,
|
||
|
|
"grad_norm": 0.74609375,
|
||
|
|
"learning_rate": 1.497037037037037e-05,
|
||
|
|
"loss": 0.9526,
|
||
|
|
"mean_token_accuracy": 0.7505445629358292,
|
||
|
|
"num_tokens": 13016669.0,
|
||
|
|
"step": 980
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.063824800774455,
|
||
|
|
"epoch": 0.8813219829744617,
|
||
|
|
"grad_norm": 0.80859375,
|
||
|
|
"learning_rate": 1.4896296296296298e-05,
|
||
|
|
"loss": 0.9742,
|
||
|
|
"mean_token_accuracy": 0.7439213328063488,
|
||
|
|
"num_tokens": 13137848.0,
|
||
|
|
"step": 990
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0260636931285263,
|
||
|
|
"epoch": 0.890224225226729,
|
||
|
|
"grad_norm": 0.78515625,
|
||
|
|
"learning_rate": 1.4822222222222225e-05,
|
||
|
|
"loss": 0.9381,
|
||
|
|
"mean_token_accuracy": 0.7512999761849641,
|
||
|
|
"num_tokens": 13270597.0,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.890224225226729,
|
||
|
|
"eval_biology_entropy": 2.052604050397873,
|
||
|
|
"eval_biology_loss": 2.1542117595672607,
|
||
|
|
"eval_biology_mean_token_accuracy": 0.557542849779129,
|
||
|
|
"eval_biology_num_tokens": 13270597.0,
|
||
|
|
"eval_biology_runtime": 18.9941,
|
||
|
|
"eval_biology_samples_per_second": 26.324,
|
||
|
|
"eval_biology_steps_per_second": 13.162,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.890224225226729,
|
||
|
|
"eval_chemistry_entropy": 1.0286521710157395,
|
||
|
|
"eval_chemistry_loss": 0.9842055439949036,
|
||
|
|
"eval_chemistry_mean_token_accuracy": 0.7491220705509186,
|
||
|
|
"eval_chemistry_num_tokens": 13270597.0,
|
||
|
|
"eval_chemistry_runtime": 23.8824,
|
||
|
|
"eval_chemistry_samples_per_second": 20.936,
|
||
|
|
"eval_chemistry_steps_per_second": 10.468,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.890224225226729,
|
||
|
|
"eval_math_entropy": 0.9400006744861603,
|
||
|
|
"eval_math_loss": 1.138838768005371,
|
||
|
|
"eval_math_mean_token_accuracy": 0.7270237381458282,
|
||
|
|
"eval_math_num_tokens": 13270597.0,
|
||
|
|
"eval_math_runtime": 24.4346,
|
||
|
|
"eval_math_samples_per_second": 20.463,
|
||
|
|
"eval_math_steps_per_second": 10.231,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.890224225226729,
|
||
|
|
"eval_physics_entropy": 1.0414861241579056,
|
||
|
|
"eval_physics_loss": 1.016461968421936,
|
||
|
|
"eval_physics_mean_token_accuracy": 0.7471425406932831,
|
||
|
|
"eval_physics_num_tokens": 13270597.0,
|
||
|
|
"eval_physics_runtime": 28.7138,
|
||
|
|
"eval_physics_samples_per_second": 17.413,
|
||
|
|
"eval_physics_steps_per_second": 8.707,
|
||
|
|
"step": 1000
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 10,
|
||
|
|
"max_steps": 3000,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 3,
|
||
|
|
"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": 7.021065582294467e+17,
|
||
|
|
"train_batch_size": 2,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|