11156 lines
304 KiB
JSON
11156 lines
304 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.310344827586207,
|
|
"eval_steps": 500,
|
|
"global_step": 5500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742109775543213,
|
|
"epoch": 0.003918495297805642,
|
|
"grad_norm": 4.9375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7674,
|
|
"mean_token_accuracy": 9.596928721293807e-05,
|
|
"num_tokens": 9415.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742099189758301,
|
|
"epoch": 0.007836990595611285,
|
|
"grad_norm": 5.65625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7184,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 17669.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742108917236328,
|
|
"epoch": 0.011755485893416929,
|
|
"grad_norm": 5.0625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7241,
|
|
"mean_token_accuracy": 0.00021240997593849898,
|
|
"num_tokens": 26558.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742095565795898,
|
|
"epoch": 0.01567398119122257,
|
|
"grad_norm": 5.40625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6562,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 34940.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742070865631103,
|
|
"epoch": 0.019592476489028215,
|
|
"grad_norm": 4.40625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5966,
|
|
"mean_token_accuracy": 0.0032189773628488183,
|
|
"num_tokens": 44835.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.741683006286621,
|
|
"epoch": 0.023510971786833857,
|
|
"grad_norm": 4.03125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4748,
|
|
"mean_token_accuracy": 0.02363332137465477,
|
|
"num_tokens": 53623.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740711116790772,
|
|
"epoch": 0.0274294670846395,
|
|
"grad_norm": 3.5,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3625,
|
|
"mean_token_accuracy": 0.03914350755512715,
|
|
"num_tokens": 63198.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.738054847717285,
|
|
"epoch": 0.03134796238244514,
|
|
"grad_norm": 2.828125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2445,
|
|
"mean_token_accuracy": 0.043806169740855695,
|
|
"num_tokens": 72379.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.73265199661255,
|
|
"epoch": 0.03526645768025078,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.1124,
|
|
"mean_token_accuracy": 0.04499626830220223,
|
|
"num_tokens": 81098.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.725791835784912,
|
|
"epoch": 0.03918495297805643,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.0549,
|
|
"mean_token_accuracy": 0.04616287313401699,
|
|
"num_tokens": 90637.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.720808124542236,
|
|
"epoch": 0.04310344827586207,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9466,
|
|
"mean_token_accuracy": 0.04513598307967186,
|
|
"num_tokens": 98505.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.717309474945068,
|
|
"epoch": 0.047021943573667714,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8942,
|
|
"mean_token_accuracy": 0.048630540072917935,
|
|
"num_tokens": 107295.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.71382761001587,
|
|
"epoch": 0.050940438871473356,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8181,
|
|
"mean_token_accuracy": 0.04877537116408348,
|
|
"num_tokens": 116381.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.70920639038086,
|
|
"epoch": 0.054858934169279,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7896,
|
|
"mean_token_accuracy": 0.04598020613193512,
|
|
"num_tokens": 124826.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.703414630889892,
|
|
"epoch": 0.05877742946708464,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7231,
|
|
"mean_token_accuracy": 0.04887317828834057,
|
|
"num_tokens": 133255.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.69604263305664,
|
|
"epoch": 0.06269592476489028,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6411,
|
|
"mean_token_accuracy": 0.05134939663112163,
|
|
"num_tokens": 141626.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.6855975151062,
|
|
"epoch": 0.06661442006269593,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5459,
|
|
"mean_token_accuracy": 0.06545204371213913,
|
|
"num_tokens": 150051.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.666928005218505,
|
|
"epoch": 0.07053291536050156,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.5417,
|
|
"mean_token_accuracy": 0.061743459478020665,
|
|
"num_tokens": 159325.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.645073413848877,
|
|
"epoch": 0.07445141065830721,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4416,
|
|
"mean_token_accuracy": 0.06291964910924434,
|
|
"num_tokens": 168142.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.608468723297118,
|
|
"epoch": 0.07836990595611286,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3001,
|
|
"mean_token_accuracy": 0.06304978355765342,
|
|
"num_tokens": 177033.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.553625392913819,
|
|
"epoch": 0.0822884012539185,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.229,
|
|
"mean_token_accuracy": 0.056359085440635684,
|
|
"num_tokens": 185605.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.505783081054688,
|
|
"epoch": 0.08620689655172414,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0952,
|
|
"mean_token_accuracy": 0.0655987836420536,
|
|
"num_tokens": 193863.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.448099517822266,
|
|
"epoch": 0.09012539184952978,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9999,
|
|
"mean_token_accuracy": 0.0653056938201189,
|
|
"num_tokens": 202504.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.432163047790528,
|
|
"epoch": 0.09404388714733543,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9214,
|
|
"mean_token_accuracy": 0.06190854795277119,
|
|
"num_tokens": 211172.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.393641376495362,
|
|
"epoch": 0.09796238244514106,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.8333,
|
|
"mean_token_accuracy": 0.060395382717251776,
|
|
"num_tokens": 220345.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.306912040710449,
|
|
"epoch": 0.10188087774294671,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6192,
|
|
"mean_token_accuracy": 0.0663460198789835,
|
|
"num_tokens": 228396.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.185036849975585,
|
|
"epoch": 0.10579937304075235,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.5194,
|
|
"mean_token_accuracy": 0.06784758679568767,
|
|
"num_tokens": 236773.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.10704460144043,
|
|
"epoch": 0.109717868338558,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.4114,
|
|
"mean_token_accuracy": 0.06528282575309277,
|
|
"num_tokens": 245279.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.997784900665284,
|
|
"epoch": 0.11363636363636363,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.2873,
|
|
"mean_token_accuracy": 0.06404702439904213,
|
|
"num_tokens": 253965.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.862957191467284,
|
|
"epoch": 0.11755485893416928,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.2398,
|
|
"mean_token_accuracy": 0.061683330684900284,
|
|
"num_tokens": 262859.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.73002061843872,
|
|
"epoch": 0.12147335423197492,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.1396,
|
|
"mean_token_accuracy": 0.06386828869581222,
|
|
"num_tokens": 270757.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.495413970947265,
|
|
"epoch": 0.12539184952978055,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.0706,
|
|
"mean_token_accuracy": 0.060359453037381175,
|
|
"num_tokens": 279737.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.4040997505188,
|
|
"epoch": 0.12931034482758622,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.895,
|
|
"mean_token_accuracy": 0.061074954271316526,
|
|
"num_tokens": 287788.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.116162109375,
|
|
"epoch": 0.13322884012539185,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.8633,
|
|
"mean_token_accuracy": 0.06705749183893203,
|
|
"num_tokens": 296481.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.948561000823975,
|
|
"epoch": 0.1371473354231975,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.76,
|
|
"mean_token_accuracy": 0.06502468548715115,
|
|
"num_tokens": 305412.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.750289058685302,
|
|
"epoch": 0.14106583072100312,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7497,
|
|
"mean_token_accuracy": 0.06412209160625934,
|
|
"num_tokens": 314412.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.61333589553833,
|
|
"epoch": 0.14498432601880878,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.6727,
|
|
"mean_token_accuracy": 0.06552823781967163,
|
|
"num_tokens": 322928.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.46501522064209,
|
|
"epoch": 0.14890282131661442,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.693,
|
|
"mean_token_accuracy": 0.0656155750155449,
|
|
"num_tokens": 331605.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.362890625,
|
|
"epoch": 0.15282131661442006,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.6608,
|
|
"mean_token_accuracy": 0.06488309130072593,
|
|
"num_tokens": 340654.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.327593994140624,
|
|
"epoch": 0.15673981191222572,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.6721,
|
|
"mean_token_accuracy": 0.06592198349535465,
|
|
"num_tokens": 348905.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.273851680755616,
|
|
"epoch": 0.16065830721003135,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5357,
|
|
"mean_token_accuracy": 0.06864950023591518,
|
|
"num_tokens": 357276.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.201261711120605,
|
|
"epoch": 0.164576802507837,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.578,
|
|
"mean_token_accuracy": 0.06853164285421372,
|
|
"num_tokens": 365707.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.147554111480712,
|
|
"epoch": 0.16849529780564262,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.5605,
|
|
"mean_token_accuracy": 0.06806492805480957,
|
|
"num_tokens": 374595.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.051411581039428,
|
|
"epoch": 0.1724137931034483,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.5988,
|
|
"mean_token_accuracy": 0.06651411205530167,
|
|
"num_tokens": 383953.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.147121715545655,
|
|
"epoch": 0.17633228840125392,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.5603,
|
|
"mean_token_accuracy": 0.06881510317325593,
|
|
"num_tokens": 393524.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.113260126113891,
|
|
"epoch": 0.18025078369905956,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.439,
|
|
"mean_token_accuracy": 0.0729409821331501,
|
|
"num_tokens": 402079.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.0263240814209,
|
|
"epoch": 0.1841692789968652,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.4926,
|
|
"mean_token_accuracy": 0.06627442128956318,
|
|
"num_tokens": 410306.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.06832423210144,
|
|
"epoch": 0.18808777429467086,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.508,
|
|
"mean_token_accuracy": 0.07089690454304218,
|
|
"num_tokens": 419157.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.074864530563355,
|
|
"epoch": 0.1920062695924765,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.463,
|
|
"mean_token_accuracy": 0.06991211287677288,
|
|
"num_tokens": 428216.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.9781323909759525,
|
|
"epoch": 0.19592476489028213,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.4362,
|
|
"mean_token_accuracy": 0.0743593417108059,
|
|
"num_tokens": 436695.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.013546228408813,
|
|
"epoch": 0.19984326018808776,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.4641,
|
|
"mean_token_accuracy": 0.07037879787385463,
|
|
"num_tokens": 446801.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.9741381168365475,
|
|
"epoch": 0.20376175548589343,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.4692,
|
|
"mean_token_accuracy": 0.0736465446650982,
|
|
"num_tokens": 454980.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.027658605575562,
|
|
"epoch": 0.20768025078369906,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.4402,
|
|
"mean_token_accuracy": 0.07250457666814328,
|
|
"num_tokens": 463699.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.018376350402832,
|
|
"epoch": 0.2115987460815047,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.334,
|
|
"mean_token_accuracy": 0.07564162723720073,
|
|
"num_tokens": 473005.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.891370153427124,
|
|
"epoch": 0.21551724137931033,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.3218,
|
|
"mean_token_accuracy": 0.07659829519689083,
|
|
"num_tokens": 481104.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.955986738204956,
|
|
"epoch": 0.219435736677116,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.4639,
|
|
"mean_token_accuracy": 0.07242306545376778,
|
|
"num_tokens": 490028.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.935779190063476,
|
|
"epoch": 0.22335423197492163,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.46,
|
|
"mean_token_accuracy": 0.07522198893129825,
|
|
"num_tokens": 499093.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.021220302581787,
|
|
"epoch": 0.22727272727272727,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.4596,
|
|
"mean_token_accuracy": 0.07582265920937062,
|
|
"num_tokens": 508049.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.857068347930908,
|
|
"epoch": 0.23119122257053293,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.276,
|
|
"mean_token_accuracy": 0.080800349265337,
|
|
"num_tokens": 516173.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.936904716491699,
|
|
"epoch": 0.23510971786833856,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.3857,
|
|
"mean_token_accuracy": 0.07901114635169507,
|
|
"num_tokens": 524990.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.885950946807862,
|
|
"epoch": 0.2390282131661442,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.4693,
|
|
"mean_token_accuracy": 0.07808374315500259,
|
|
"num_tokens": 533676.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.922548103332519,
|
|
"epoch": 0.24294670846394983,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.4113,
|
|
"mean_token_accuracy": 0.0751194790005684,
|
|
"num_tokens": 542757.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.911523008346558,
|
|
"epoch": 0.2468652037617555,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.3816,
|
|
"mean_token_accuracy": 0.07483164295554161,
|
|
"num_tokens": 551958.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.87831482887268,
|
|
"epoch": 0.2507836990595611,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.297,
|
|
"mean_token_accuracy": 0.07816217541694641,
|
|
"num_tokens": 560850.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.742319774627686,
|
|
"epoch": 0.2547021943573668,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.2442,
|
|
"mean_token_accuracy": 0.07955843061208726,
|
|
"num_tokens": 569226.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.862335586547852,
|
|
"epoch": 0.25862068965517243,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.2487,
|
|
"mean_token_accuracy": 0.08338543474674225,
|
|
"num_tokens": 577674.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.8038520336151125,
|
|
"epoch": 0.26253918495297807,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.3003,
|
|
"mean_token_accuracy": 0.08087649494409561,
|
|
"num_tokens": 587095.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.815746021270752,
|
|
"epoch": 0.2664576802507837,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.2334,
|
|
"mean_token_accuracy": 0.08010689094662667,
|
|
"num_tokens": 595690.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.813080167770385,
|
|
"epoch": 0.27037617554858934,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.2638,
|
|
"mean_token_accuracy": 0.08633363470435143,
|
|
"num_tokens": 603962.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.854545068740845,
|
|
"epoch": 0.274294670846395,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.2481,
|
|
"mean_token_accuracy": 0.08236029744148254,
|
|
"num_tokens": 612407.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.811877584457397,
|
|
"epoch": 0.2782131661442006,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.3141,
|
|
"mean_token_accuracy": 0.08180430978536606,
|
|
"num_tokens": 620970.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.737411594390869,
|
|
"epoch": 0.28213166144200624,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.2908,
|
|
"mean_token_accuracy": 0.07814904972910881,
|
|
"num_tokens": 630701.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.802158451080322,
|
|
"epoch": 0.28605015673981193,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.2974,
|
|
"mean_token_accuracy": 0.08439014628529548,
|
|
"num_tokens": 640246.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.708374309539795,
|
|
"epoch": 0.28996865203761757,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.2173,
|
|
"mean_token_accuracy": 0.08501755222678184,
|
|
"num_tokens": 648698.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.808612918853759,
|
|
"epoch": 0.2938871473354232,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.2139,
|
|
"mean_token_accuracy": 0.09233642369508743,
|
|
"num_tokens": 658087.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.6729991912841795,
|
|
"epoch": 0.29780564263322884,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1842,
|
|
"mean_token_accuracy": 0.08876543864607811,
|
|
"num_tokens": 666664.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.760433673858643,
|
|
"epoch": 0.3017241379310345,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.2527,
|
|
"mean_token_accuracy": 0.08252104483544827,
|
|
"num_tokens": 675738.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.659246635437012,
|
|
"epoch": 0.3056426332288401,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.1341,
|
|
"mean_token_accuracy": 0.0883593812584877,
|
|
"num_tokens": 684751.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.6855284690856935,
|
|
"epoch": 0.30956112852664575,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.1392,
|
|
"mean_token_accuracy": 0.09001046866178512,
|
|
"num_tokens": 693819.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.675692749023438,
|
|
"epoch": 0.31347962382445144,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.2002,
|
|
"mean_token_accuracy": 0.08597821295261383,
|
|
"num_tokens": 703082.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.6546083927154545,
|
|
"epoch": 0.31739811912225707,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.1151,
|
|
"mean_token_accuracy": 0.08707506135106087,
|
|
"num_tokens": 711721.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.646405744552612,
|
|
"epoch": 0.3213166144200627,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.1256,
|
|
"mean_token_accuracy": 0.08994124978780746,
|
|
"num_tokens": 719825.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.639235925674439,
|
|
"epoch": 0.32523510971786834,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.053,
|
|
"mean_token_accuracy": 0.08337244391441345,
|
|
"num_tokens": 728575.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.728874778747558,
|
|
"epoch": 0.329153605015674,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.2405,
|
|
"mean_token_accuracy": 0.08313387259840965,
|
|
"num_tokens": 738379.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.568649673461914,
|
|
"epoch": 0.3330721003134796,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.0653,
|
|
"mean_token_accuracy": 0.0943116620182991,
|
|
"num_tokens": 747484.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.662053632736206,
|
|
"epoch": 0.33699059561128525,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0976,
|
|
"mean_token_accuracy": 0.09167452827095986,
|
|
"num_tokens": 756250.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.631493139266968,
|
|
"epoch": 0.3409090909090909,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.1553,
|
|
"mean_token_accuracy": 0.09070844799280167,
|
|
"num_tokens": 765393.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.533075618743896,
|
|
"epoch": 0.3448275862068966,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.1282,
|
|
"mean_token_accuracy": 0.09012485444545745,
|
|
"num_tokens": 773726.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.584697198867798,
|
|
"epoch": 0.3487460815047022,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0359,
|
|
"mean_token_accuracy": 0.09066722467541695,
|
|
"num_tokens": 782335.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.568591690063476,
|
|
"epoch": 0.35266457680250785,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9887,
|
|
"mean_token_accuracy": 0.09229951500892639,
|
|
"num_tokens": 791546.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.518783092498779,
|
|
"epoch": 0.3565830721003135,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.964,
|
|
"mean_token_accuracy": 0.09283199757337571,
|
|
"num_tokens": 800650.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.47516098022461,
|
|
"epoch": 0.3605015673981191,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.0328,
|
|
"mean_token_accuracy": 0.09423457458615303,
|
|
"num_tokens": 809968.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.544811105728149,
|
|
"epoch": 0.36442006269592475,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0201,
|
|
"mean_token_accuracy": 0.09420157298445701,
|
|
"num_tokens": 818800.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.581027984619141,
|
|
"epoch": 0.3683385579937304,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.9864,
|
|
"mean_token_accuracy": 0.09675867483019829,
|
|
"num_tokens": 827586.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.303757953643799,
|
|
"epoch": 0.3722570532915361,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.9384,
|
|
"mean_token_accuracy": 0.09580465108156204,
|
|
"num_tokens": 836730.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.696502780914306,
|
|
"epoch": 0.3761755485893417,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.1422,
|
|
"mean_token_accuracy": 0.09025072157382966,
|
|
"num_tokens": 845105.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.521749973297119,
|
|
"epoch": 0.38009404388714735,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.0783,
|
|
"mean_token_accuracy": 0.09295645728707314,
|
|
"num_tokens": 853584.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.556596755981445,
|
|
"epoch": 0.384012539184953,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.0668,
|
|
"mean_token_accuracy": 0.08822977468371392,
|
|
"num_tokens": 862442.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.4543486595153805,
|
|
"epoch": 0.3879310344827586,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.9643,
|
|
"mean_token_accuracy": 0.0951703853905201,
|
|
"num_tokens": 870833.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.596196413040161,
|
|
"epoch": 0.39184952978056425,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.1087,
|
|
"mean_token_accuracy": 0.09554079249501228,
|
|
"num_tokens": 879461.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.39184952978056425,
|
|
"eval_entropy": 7.3791314095489735,
|
|
"eval_loss": 7.098165035247803,
|
|
"eval_mean_token_accuracy": 0.09636555407051892,
|
|
"eval_num_tokens": 879461.0,
|
|
"eval_runtime": 2.2153,
|
|
"eval_samples_per_second": 1860.68,
|
|
"eval_steps_per_second": 232.924,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.323290538787842,
|
|
"epoch": 0.3957680250783699,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.8468,
|
|
"mean_token_accuracy": 0.10375453308224677,
|
|
"num_tokens": 887893.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.584144163131714,
|
|
"epoch": 0.3996865203761755,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.078,
|
|
"mean_token_accuracy": 0.09517347961664199,
|
|
"num_tokens": 896129.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.514800357818603,
|
|
"epoch": 0.4036050156739812,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.0467,
|
|
"mean_token_accuracy": 0.0908431887626648,
|
|
"num_tokens": 904448.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.4618706703186035,
|
|
"epoch": 0.40752351097178685,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.9808,
|
|
"mean_token_accuracy": 0.09821045622229577,
|
|
"num_tokens": 912486.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.48437385559082,
|
|
"epoch": 0.4114420062695925,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.9342,
|
|
"mean_token_accuracy": 0.0975078672170639,
|
|
"num_tokens": 920913.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.505733489990234,
|
|
"epoch": 0.4153605015673981,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.0757,
|
|
"mean_token_accuracy": 0.08893783465027809,
|
|
"num_tokens": 930227.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.452349138259888,
|
|
"epoch": 0.41927899686520376,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.9462,
|
|
"mean_token_accuracy": 0.09549824744462967,
|
|
"num_tokens": 938835.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.436382722854614,
|
|
"epoch": 0.4231974921630094,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.9256,
|
|
"mean_token_accuracy": 0.10100051760673523,
|
|
"num_tokens": 946930.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.43581748008728,
|
|
"epoch": 0.427115987460815,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.9024,
|
|
"mean_token_accuracy": 0.09818659126758575,
|
|
"num_tokens": 955450.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.412800025939942,
|
|
"epoch": 0.43103448275862066,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.9804,
|
|
"mean_token_accuracy": 0.09347106665372848,
|
|
"num_tokens": 964456.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.413184976577758,
|
|
"epoch": 0.43495297805642635,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.8786,
|
|
"mean_token_accuracy": 0.0999380074441433,
|
|
"num_tokens": 973088.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.320905685424805,
|
|
"epoch": 0.438871473354232,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8844,
|
|
"mean_token_accuracy": 0.09666496440768242,
|
|
"num_tokens": 982182.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.376274633407593,
|
|
"epoch": 0.4427899686520376,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.8457,
|
|
"mean_token_accuracy": 0.10074941292405129,
|
|
"num_tokens": 990513.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.300549602508545,
|
|
"epoch": 0.44670846394984326,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.857,
|
|
"mean_token_accuracy": 0.09506906270980835,
|
|
"num_tokens": 999394.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.375359916687012,
|
|
"epoch": 0.4506269592476489,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.8453,
|
|
"mean_token_accuracy": 0.1013949416577816,
|
|
"num_tokens": 1007793.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.34481520652771,
|
|
"epoch": 0.45454545454545453,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.9084,
|
|
"mean_token_accuracy": 0.09777404814958572,
|
|
"num_tokens": 1017198.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.3425328731536865,
|
|
"epoch": 0.45846394984326017,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.8047,
|
|
"mean_token_accuracy": 0.10280357003211975,
|
|
"num_tokens": 1025896.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.445605945587158,
|
|
"epoch": 0.46238244514106586,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.0133,
|
|
"mean_token_accuracy": 0.09696159362792969,
|
|
"num_tokens": 1034540.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.213116216659546,
|
|
"epoch": 0.4663009404388715,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.8321,
|
|
"mean_token_accuracy": 0.10862131938338279,
|
|
"num_tokens": 1043006.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.316764163970947,
|
|
"epoch": 0.4702194357366771,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.8099,
|
|
"mean_token_accuracy": 0.10350289791822434,
|
|
"num_tokens": 1052530.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.323724317550659,
|
|
"epoch": 0.47413793103448276,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.8137,
|
|
"mean_token_accuracy": 0.10041582360863685,
|
|
"num_tokens": 1060822.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.30889310836792,
|
|
"epoch": 0.4780564263322884,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.8802,
|
|
"mean_token_accuracy": 0.10094830319285393,
|
|
"num_tokens": 1069962.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.313723564147949,
|
|
"epoch": 0.48197492163009403,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.806,
|
|
"mean_token_accuracy": 0.10286147370934487,
|
|
"num_tokens": 1078903.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.307069587707519,
|
|
"epoch": 0.48589341692789967,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.8794,
|
|
"mean_token_accuracy": 0.09798636510968209,
|
|
"num_tokens": 1087921.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.2093301773071286,
|
|
"epoch": 0.4898119122257053,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7742,
|
|
"mean_token_accuracy": 0.10318923816084861,
|
|
"num_tokens": 1096778.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.321264219284058,
|
|
"epoch": 0.493730407523511,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.8046,
|
|
"mean_token_accuracy": 0.10321516692638397,
|
|
"num_tokens": 1105000.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.175268077850342,
|
|
"epoch": 0.49764890282131663,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6925,
|
|
"mean_token_accuracy": 0.10435841083526612,
|
|
"num_tokens": 1113475.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.1677467823028564,
|
|
"epoch": 0.5015673981191222,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7436,
|
|
"mean_token_accuracy": 0.10689813792705535,
|
|
"num_tokens": 1122188.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.284512710571289,
|
|
"epoch": 0.5054858934169278,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.8446,
|
|
"mean_token_accuracy": 0.10488534867763519,
|
|
"num_tokens": 1130781.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.213377618789673,
|
|
"epoch": 0.5094043887147336,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.771,
|
|
"mean_token_accuracy": 0.09970749393105507,
|
|
"num_tokens": 1139889.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.230893850326538,
|
|
"epoch": 0.5133228840125392,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.7344,
|
|
"mean_token_accuracy": 0.10886386185884475,
|
|
"num_tokens": 1147904.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.088760614395142,
|
|
"epoch": 0.5172413793103449,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.696,
|
|
"mean_token_accuracy": 0.1124557226896286,
|
|
"num_tokens": 1157413.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.306563091278076,
|
|
"epoch": 0.5211598746081505,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.8164,
|
|
"mean_token_accuracy": 0.10038655102252961,
|
|
"num_tokens": 1165635.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.134089422225952,
|
|
"epoch": 0.5250783699059561,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.8298,
|
|
"mean_token_accuracy": 0.09688482508063316,
|
|
"num_tokens": 1174681.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.235197401046753,
|
|
"epoch": 0.5289968652037618,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.8225,
|
|
"mean_token_accuracy": 0.10358083248138428,
|
|
"num_tokens": 1183454.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.254007053375244,
|
|
"epoch": 0.5329153605015674,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.7803,
|
|
"mean_token_accuracy": 0.10250866636633874,
|
|
"num_tokens": 1191836.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.161578273773193,
|
|
"epoch": 0.536833855799373,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.7006,
|
|
"mean_token_accuracy": 0.10597401708364487,
|
|
"num_tokens": 1200280.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.203849601745605,
|
|
"epoch": 0.5407523510971787,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.8066,
|
|
"mean_token_accuracy": 0.09864948689937592,
|
|
"num_tokens": 1208773.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.236851644515991,
|
|
"epoch": 0.5446708463949843,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.795,
|
|
"mean_token_accuracy": 0.10489028915762902,
|
|
"num_tokens": 1218171.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.15051794052124,
|
|
"epoch": 0.54858934169279,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.7532,
|
|
"mean_token_accuracy": 0.10080546215176582,
|
|
"num_tokens": 1226549.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.200709390640259,
|
|
"epoch": 0.5525078369905956,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.7439,
|
|
"mean_token_accuracy": 0.11170587316155434,
|
|
"num_tokens": 1234995.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.17590970993042,
|
|
"epoch": 0.5564263322884012,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.787,
|
|
"mean_token_accuracy": 0.10626475065946579,
|
|
"num_tokens": 1243807.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.112601041793823,
|
|
"epoch": 0.5603448275862069,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.7665,
|
|
"mean_token_accuracy": 0.10710262060165406,
|
|
"num_tokens": 1252773.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.164514064788818,
|
|
"epoch": 0.5642633228840125,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.7316,
|
|
"mean_token_accuracy": 0.09972823187708854,
|
|
"num_tokens": 1262514.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.2263566017150875,
|
|
"epoch": 0.5681818181818182,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.7304,
|
|
"mean_token_accuracy": 0.10351964756846428,
|
|
"num_tokens": 1271329.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.020602416992188,
|
|
"epoch": 0.5721003134796239,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6153,
|
|
"mean_token_accuracy": 0.1127767838537693,
|
|
"num_tokens": 1279563.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.153738784790039,
|
|
"epoch": 0.5760188087774295,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.7494,
|
|
"mean_token_accuracy": 0.10431532934308052,
|
|
"num_tokens": 1288906.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.086837673187256,
|
|
"epoch": 0.5799373040752351,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.6582,
|
|
"mean_token_accuracy": 0.10919351503252983,
|
|
"num_tokens": 1297880.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.083261871337891,
|
|
"epoch": 0.5838557993730408,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.6923,
|
|
"mean_token_accuracy": 0.10958038419485092,
|
|
"num_tokens": 1306740.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.1014440059661865,
|
|
"epoch": 0.5877742946708464,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.6471,
|
|
"mean_token_accuracy": 0.10642243772745133,
|
|
"num_tokens": 1315997.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.118290996551513,
|
|
"epoch": 0.591692789968652,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.634,
|
|
"mean_token_accuracy": 0.11355772763490676,
|
|
"num_tokens": 1324064.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.116359090805053,
|
|
"epoch": 0.5956112852664577,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.7624,
|
|
"mean_token_accuracy": 0.10623705238103867,
|
|
"num_tokens": 1332560.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.0753131866455075,
|
|
"epoch": 0.5995297805642633,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.6853,
|
|
"mean_token_accuracy": 0.10491749420762062,
|
|
"num_tokens": 1341367.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.208959865570068,
|
|
"epoch": 0.603448275862069,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.8643,
|
|
"mean_token_accuracy": 0.09926253855228424,
|
|
"num_tokens": 1350602.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.111758089065551,
|
|
"epoch": 0.6073667711598746,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5959,
|
|
"mean_token_accuracy": 0.11473986878991127,
|
|
"num_tokens": 1358538.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.9510876655578615,
|
|
"epoch": 0.6112852664576802,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5966,
|
|
"mean_token_accuracy": 0.11128686666488648,
|
|
"num_tokens": 1367089.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.095315790176391,
|
|
"epoch": 0.6152037617554859,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.6987,
|
|
"mean_token_accuracy": 0.107613255828619,
|
|
"num_tokens": 1376095.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.0291649341583256,
|
|
"epoch": 0.6191222570532915,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.7278,
|
|
"mean_token_accuracy": 0.10492047220468521,
|
|
"num_tokens": 1385132.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.174719142913818,
|
|
"epoch": 0.6230407523510971,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.8042,
|
|
"mean_token_accuracy": 0.10808081552386284,
|
|
"num_tokens": 1394187.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.009328317642212,
|
|
"epoch": 0.6269592476489029,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.6543,
|
|
"mean_token_accuracy": 0.10319057255983352,
|
|
"num_tokens": 1403250.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.114693927764892,
|
|
"epoch": 0.6308777429467085,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.7523,
|
|
"mean_token_accuracy": 0.10417917445302009,
|
|
"num_tokens": 1412768.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.1693981170654295,
|
|
"epoch": 0.6347962382445141,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.7831,
|
|
"mean_token_accuracy": 0.10479006171226501,
|
|
"num_tokens": 1422437.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.221331596374512,
|
|
"epoch": 0.6387147335423198,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.8352,
|
|
"mean_token_accuracy": 0.10820698514580726,
|
|
"num_tokens": 1432540.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.996674919128418,
|
|
"epoch": 0.6426332288401254,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.6518,
|
|
"mean_token_accuracy": 0.10877189710736275,
|
|
"num_tokens": 1441403.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.932708883285523,
|
|
"epoch": 0.646551724137931,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4891,
|
|
"mean_token_accuracy": 0.10969182550907135,
|
|
"num_tokens": 1450059.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.004702997207642,
|
|
"epoch": 0.6504702194357367,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.6424,
|
|
"mean_token_accuracy": 0.11137095913290977,
|
|
"num_tokens": 1458413.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.982443046569824,
|
|
"epoch": 0.6543887147335423,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.622,
|
|
"mean_token_accuracy": 0.11147240921854973,
|
|
"num_tokens": 1467343.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.956045198440552,
|
|
"epoch": 0.658307210031348,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5607,
|
|
"mean_token_accuracy": 0.11432239189743995,
|
|
"num_tokens": 1476250.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.017556047439575,
|
|
"epoch": 0.6622257053291536,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.5268,
|
|
"mean_token_accuracy": 0.11572841182351112,
|
|
"num_tokens": 1484663.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.062141704559326,
|
|
"epoch": 0.6661442006269592,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.6401,
|
|
"mean_token_accuracy": 0.10651951879262925,
|
|
"num_tokens": 1493898.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.941916561126709,
|
|
"epoch": 0.6700626959247649,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.6059,
|
|
"mean_token_accuracy": 0.1140934869647026,
|
|
"num_tokens": 1502429.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.927605772018433,
|
|
"epoch": 0.6739811912225705,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.6293,
|
|
"mean_token_accuracy": 0.11264230906963349,
|
|
"num_tokens": 1511671.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.936274194717408,
|
|
"epoch": 0.6778996865203761,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.5421,
|
|
"mean_token_accuracy": 0.11239867955446244,
|
|
"num_tokens": 1520481.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.910920095443726,
|
|
"epoch": 0.6818181818181818,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4874,
|
|
"mean_token_accuracy": 0.11565134301781654,
|
|
"num_tokens": 1528589.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.936212968826294,
|
|
"epoch": 0.6857366771159875,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.6584,
|
|
"mean_token_accuracy": 0.108811154961586,
|
|
"num_tokens": 1537555.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.945295000076294,
|
|
"epoch": 0.6896551724137931,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.6032,
|
|
"mean_token_accuracy": 0.11123565956950188,
|
|
"num_tokens": 1546397.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.034606647491455,
|
|
"epoch": 0.6935736677115988,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.6331,
|
|
"mean_token_accuracy": 0.11215913444757461,
|
|
"num_tokens": 1555603.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.92329592704773,
|
|
"epoch": 0.6974921630094044,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.5938,
|
|
"mean_token_accuracy": 0.11114713251590728,
|
|
"num_tokens": 1564688.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.946131467819214,
|
|
"epoch": 0.70141065830721,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.5727,
|
|
"mean_token_accuracy": 0.1131402924656868,
|
|
"num_tokens": 1573490.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.9099291324615475,
|
|
"epoch": 0.7053291536050157,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5425,
|
|
"mean_token_accuracy": 0.11272019520401955,
|
|
"num_tokens": 1582668.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.931637620925903,
|
|
"epoch": 0.7092476489028213,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.6098,
|
|
"mean_token_accuracy": 0.11104563847184182,
|
|
"num_tokens": 1591667.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.885412693023682,
|
|
"epoch": 0.713166144200627,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5154,
|
|
"mean_token_accuracy": 0.11723392531275749,
|
|
"num_tokens": 1600248.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.9636177062988285,
|
|
"epoch": 0.7170846394984326,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.5222,
|
|
"mean_token_accuracy": 0.11748122572898864,
|
|
"num_tokens": 1609094.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.80591516494751,
|
|
"epoch": 0.7210031347962382,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5523,
|
|
"mean_token_accuracy": 0.11386217251420021,
|
|
"num_tokens": 1617845.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.011699295043945,
|
|
"epoch": 0.7249216300940439,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.5865,
|
|
"mean_token_accuracy": 0.11231657788157463,
|
|
"num_tokens": 1626596.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.840612173080444,
|
|
"epoch": 0.7288401253918495,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.5,
|
|
"mean_token_accuracy": 0.11078331917524338,
|
|
"num_tokens": 1635246.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.838626718521118,
|
|
"epoch": 0.7327586206896551,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.556,
|
|
"mean_token_accuracy": 0.11381039470434189,
|
|
"num_tokens": 1645041.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.995179224014282,
|
|
"epoch": 0.7366771159874608,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.6265,
|
|
"mean_token_accuracy": 0.10761500224471092,
|
|
"num_tokens": 1653636.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.835239219665527,
|
|
"epoch": 0.7405956112852664,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.5227,
|
|
"mean_token_accuracy": 0.11463285759091377,
|
|
"num_tokens": 1662485.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.926279592514038,
|
|
"epoch": 0.7445141065830722,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.5461,
|
|
"mean_token_accuracy": 0.12085802108049393,
|
|
"num_tokens": 1670414.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.897070741653442,
|
|
"epoch": 0.7484326018808778,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.5507,
|
|
"mean_token_accuracy": 0.11035623922944068,
|
|
"num_tokens": 1678853.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.948398160934448,
|
|
"epoch": 0.7523510971786834,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4881,
|
|
"mean_token_accuracy": 0.11652906686067581,
|
|
"num_tokens": 1687286.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.789101934432983,
|
|
"epoch": 0.7562695924764891,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.5127,
|
|
"mean_token_accuracy": 0.1135128065943718,
|
|
"num_tokens": 1696302.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.88850679397583,
|
|
"epoch": 0.7601880877742947,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.6328,
|
|
"mean_token_accuracy": 0.10952614322304725,
|
|
"num_tokens": 1704622.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.836090850830078,
|
|
"epoch": 0.7641065830721003,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4682,
|
|
"mean_token_accuracy": 0.11953761354088784,
|
|
"num_tokens": 1713060.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.769405508041382,
|
|
"epoch": 0.768025078369906,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4437,
|
|
"mean_token_accuracy": 0.1201899990439415,
|
|
"num_tokens": 1721654.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.872002363204956,
|
|
"epoch": 0.7719435736677116,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.4967,
|
|
"mean_token_accuracy": 0.11558816209435463,
|
|
"num_tokens": 1730243.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.746385478973389,
|
|
"epoch": 0.7758620689655172,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4224,
|
|
"mean_token_accuracy": 0.11981726884841919,
|
|
"num_tokens": 1739347.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.771752119064331,
|
|
"epoch": 0.7797805642633229,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.4607,
|
|
"mean_token_accuracy": 0.12106571421027183,
|
|
"num_tokens": 1747802.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.778915500640869,
|
|
"epoch": 0.7836990595611285,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.4581,
|
|
"mean_token_accuracy": 0.11269956901669502,
|
|
"num_tokens": 1757407.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.7836990595611285,
|
|
"eval_entropy": 6.515588334364485,
|
|
"eval_loss": 6.588331699371338,
|
|
"eval_mean_token_accuracy": 0.1149799013077173,
|
|
"eval_num_tokens": 1757407.0,
|
|
"eval_runtime": 2.3882,
|
|
"eval_samples_per_second": 1725.989,
|
|
"eval_steps_per_second": 216.063,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.8325563907623295,
|
|
"epoch": 0.7876175548589341,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999998713243189,
|
|
"loss": 6.5047,
|
|
"mean_token_accuracy": 0.11974181681871414,
|
|
"num_tokens": 1766323.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.736519289016724,
|
|
"epoch": 0.7915360501567398,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999993485796164,
|
|
"loss": 6.4707,
|
|
"mean_token_accuracy": 0.11981728523969651,
|
|
"num_tokens": 1776193.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.85446834564209,
|
|
"epoch": 0.7954545454545454,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999984237245962,
|
|
"loss": 6.5192,
|
|
"mean_token_accuracy": 0.1184984378516674,
|
|
"num_tokens": 1785813.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.7683358669281,
|
|
"epoch": 0.799373040752351,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999970967609109,
|
|
"loss": 6.4234,
|
|
"mean_token_accuracy": 0.11528945416212082,
|
|
"num_tokens": 1794448.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.738554382324219,
|
|
"epoch": 0.8032915360501567,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999953676909322,
|
|
"loss": 6.4485,
|
|
"mean_token_accuracy": 0.11520285606384277,
|
|
"num_tokens": 1803091.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.897971105575562,
|
|
"epoch": 0.8072100313479624,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999932365177502,
|
|
"loss": 6.5432,
|
|
"mean_token_accuracy": 0.11414010152220726,
|
|
"num_tokens": 1811649.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.877712345123291,
|
|
"epoch": 0.8111285266457681,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999907032451735,
|
|
"loss": 6.5178,
|
|
"mean_token_accuracy": 0.11190005615353585,
|
|
"num_tokens": 1820060.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.638827466964722,
|
|
"epoch": 0.8150470219435737,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999877678777296,
|
|
"loss": 6.381,
|
|
"mean_token_accuracy": 0.12132669016718864,
|
|
"num_tokens": 1828917.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.85034146308899,
|
|
"epoch": 0.8189655172413793,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999844304206645,
|
|
"loss": 6.4692,
|
|
"mean_token_accuracy": 0.11515972465276718,
|
|
"num_tokens": 1838166.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.693251895904541,
|
|
"epoch": 0.822884012539185,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999806908799428,
|
|
"loss": 6.3677,
|
|
"mean_token_accuracy": 0.12894104793667793,
|
|
"num_tokens": 1846752.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.830075883865357,
|
|
"epoch": 0.8268025078369906,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999765492622474,
|
|
"loss": 6.4093,
|
|
"mean_token_accuracy": 0.11808309555053711,
|
|
"num_tokens": 1855250.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.702435922622681,
|
|
"epoch": 0.8307210031347962,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999720055749804,
|
|
"loss": 6.4262,
|
|
"mean_token_accuracy": 0.11298789232969284,
|
|
"num_tokens": 1863809.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.728408670425415,
|
|
"epoch": 0.8346394984326019,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999670598262619,
|
|
"loss": 6.407,
|
|
"mean_token_accuracy": 0.11927128955721855,
|
|
"num_tokens": 1872308.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.6889567375183105,
|
|
"epoch": 0.8385579937304075,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999617120249308,
|
|
"loss": 6.3375,
|
|
"mean_token_accuracy": 0.1220373697578907,
|
|
"num_tokens": 1881107.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.789124965667725,
|
|
"epoch": 0.8424764890282131,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999559621805445,
|
|
"loss": 6.4623,
|
|
"mean_token_accuracy": 0.11761095523834228,
|
|
"num_tokens": 1889961.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.689903688430786,
|
|
"epoch": 0.8463949843260188,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999498103033788,
|
|
"loss": 6.3445,
|
|
"mean_token_accuracy": 0.11976620554924011,
|
|
"num_tokens": 1898432.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.650704622268677,
|
|
"epoch": 0.8503134796238244,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999432564044284,
|
|
"loss": 6.3578,
|
|
"mean_token_accuracy": 0.12074211835861207,
|
|
"num_tokens": 1907236.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.6976752281188965,
|
|
"epoch": 0.85423197492163,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999363004954058,
|
|
"loss": 6.3497,
|
|
"mean_token_accuracy": 0.1198844201862812,
|
|
"num_tokens": 1916388.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.725913238525391,
|
|
"epoch": 0.8581504702194357,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.456,
|
|
"mean_token_accuracy": 0.11865570768713951,
|
|
"num_tokens": 1926150.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.795915842056274,
|
|
"epoch": 0.8620689655172413,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999211826975884,
|
|
"loss": 6.477,
|
|
"mean_token_accuracy": 0.11915030851960182,
|
|
"num_tokens": 1935113.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.656408739089966,
|
|
"epoch": 0.8659874608150471,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999130208358114,
|
|
"loss": 6.4021,
|
|
"mean_token_accuracy": 0.11451392620801926,
|
|
"num_tokens": 1944470.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.738942337036133,
|
|
"epoch": 0.8699059561128527,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999044570179983,
|
|
"loss": 6.4227,
|
|
"mean_token_accuracy": 0.11082111597061158,
|
|
"num_tokens": 1954327.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.575215244293213,
|
|
"epoch": 0.8738244514106583,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998954912594541,
|
|
"loss": 6.3189,
|
|
"mean_token_accuracy": 0.1276112236082554,
|
|
"num_tokens": 1962860.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.6690638065338135,
|
|
"epoch": 0.877742946708464,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004998861235762018,
|
|
"loss": 6.3139,
|
|
"mean_token_accuracy": 0.12368955090641975,
|
|
"num_tokens": 1971212.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.662813901901245,
|
|
"epoch": 0.8816614420062696,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998763539849832,
|
|
"loss": 6.2525,
|
|
"mean_token_accuracy": 0.11725214570760727,
|
|
"num_tokens": 1980489.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.6482361316680905,
|
|
"epoch": 0.8855799373040752,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998661825032579,
|
|
"loss": 6.4021,
|
|
"mean_token_accuracy": 0.1187636524438858,
|
|
"num_tokens": 1989326.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.722862958908081,
|
|
"epoch": 0.8894984326018809,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998556091492043,
|
|
"loss": 6.459,
|
|
"mean_token_accuracy": 0.11343813464045524,
|
|
"num_tokens": 1997888.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.655751085281372,
|
|
"epoch": 0.8934169278996865,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998446339417184,
|
|
"loss": 6.3462,
|
|
"mean_token_accuracy": 0.12144071459770203,
|
|
"num_tokens": 2006815.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.685620832443237,
|
|
"epoch": 0.8973354231974922,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998332569004149,
|
|
"loss": 6.3128,
|
|
"mean_token_accuracy": 0.12164490222930908,
|
|
"num_tokens": 2015684.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.632381391525269,
|
|
"epoch": 0.9012539184952978,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004998214780456263,
|
|
"loss": 6.2757,
|
|
"mean_token_accuracy": 0.12474701330065727,
|
|
"num_tokens": 2024148.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.573228025436402,
|
|
"epoch": 0.9051724137931034,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998092973984033,
|
|
"loss": 6.3772,
|
|
"mean_token_accuracy": 0.12780468836426734,
|
|
"num_tokens": 2032287.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.785788917541504,
|
|
"epoch": 0.9090909090909091,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997967149805147,
|
|
"loss": 6.4894,
|
|
"mean_token_accuracy": 0.11504697501659393,
|
|
"num_tokens": 2041663.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.793114900588989,
|
|
"epoch": 0.9130094043887147,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997837308144474,
|
|
"loss": 6.4433,
|
|
"mean_token_accuracy": 0.11913444623351097,
|
|
"num_tokens": 2049756.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.65000319480896,
|
|
"epoch": 0.9169278996865203,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997703449234062,
|
|
"loss": 6.454,
|
|
"mean_token_accuracy": 0.11475523188710213,
|
|
"num_tokens": 2058626.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.735292673110962,
|
|
"epoch": 0.920846394984326,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997565573313139,
|
|
"loss": 6.366,
|
|
"mean_token_accuracy": 0.11419134736061096,
|
|
"num_tokens": 2067395.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.603668308258056,
|
|
"epoch": 0.9247648902821317,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997423680628109,
|
|
"loss": 6.3787,
|
|
"mean_token_accuracy": 0.12259217575192452,
|
|
"num_tokens": 2075651.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.688165903091431,
|
|
"epoch": 0.9286833855799373,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499727777143256,
|
|
"loss": 6.4043,
|
|
"mean_token_accuracy": 0.11774512678384781,
|
|
"num_tokens": 2084458.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.804714918136597,
|
|
"epoch": 0.932601880877743,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997127845987255,
|
|
"loss": 6.4884,
|
|
"mean_token_accuracy": 0.11993765234947204,
|
|
"num_tokens": 2092413.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.611707592010498,
|
|
"epoch": 0.9365203761755486,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996973904560135,
|
|
"loss": 6.3697,
|
|
"mean_token_accuracy": 0.1176217883825302,
|
|
"num_tokens": 2100799.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.671212148666382,
|
|
"epoch": 0.9404388714733543,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996815947426317,
|
|
"loss": 6.2887,
|
|
"mean_token_accuracy": 0.12218108549714088,
|
|
"num_tokens": 2109357.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.574631977081299,
|
|
"epoch": 0.9443573667711599,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996653974868097,
|
|
"loss": 6.2794,
|
|
"mean_token_accuracy": 0.11755874156951904,
|
|
"num_tokens": 2117642.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.538684892654419,
|
|
"epoch": 0.9482758620689655,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996487987174946,
|
|
"loss": 6.3098,
|
|
"mean_token_accuracy": 0.12178566977381707,
|
|
"num_tokens": 2126512.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.6026019096374515,
|
|
"epoch": 0.9521943573667712,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996317984643512,
|
|
"loss": 6.3228,
|
|
"mean_token_accuracy": 0.12325456142425537,
|
|
"num_tokens": 2135559.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.681765937805176,
|
|
"epoch": 0.9561128526645768,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996143967577615,
|
|
"loss": 6.3066,
|
|
"mean_token_accuracy": 0.12556551843881608,
|
|
"num_tokens": 2144436.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.583171606063843,
|
|
"epoch": 0.9600313479623824,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995965936288254,
|
|
"loss": 6.2765,
|
|
"mean_token_accuracy": 0.12223444804549218,
|
|
"num_tokens": 2152625.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.714479923248291,
|
|
"epoch": 0.9639498432601881,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995783891093597,
|
|
"loss": 6.4299,
|
|
"mean_token_accuracy": 0.1136121116578579,
|
|
"num_tokens": 2162405.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.595222473144531,
|
|
"epoch": 0.9678683385579937,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499559783231899,
|
|
"loss": 6.3848,
|
|
"mean_token_accuracy": 0.11950541213154793,
|
|
"num_tokens": 2171376.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.623385429382324,
|
|
"epoch": 0.9717868338557993,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995407760296952,
|
|
"loss": 6.2444,
|
|
"mean_token_accuracy": 0.125845268368721,
|
|
"num_tokens": 2180343.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.456481504440307,
|
|
"epoch": 0.975705329153605,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995213675367169,
|
|
"loss": 6.2042,
|
|
"mean_token_accuracy": 0.129156381636858,
|
|
"num_tokens": 2189277.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.577784681320191,
|
|
"epoch": 0.9796238244514106,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004995015577876504,
|
|
"loss": 6.1903,
|
|
"mean_token_accuracy": 0.12381277903914452,
|
|
"num_tokens": 2197614.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.570421743392944,
|
|
"epoch": 0.9835423197492164,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000499481346817899,
|
|
"loss": 6.2956,
|
|
"mean_token_accuracy": 0.12848778143525125,
|
|
"num_tokens": 2205589.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.496778583526611,
|
|
"epoch": 0.987460815047022,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004994607346635829,
|
|
"loss": 6.2882,
|
|
"mean_token_accuracy": 0.1286270134150982,
|
|
"num_tokens": 2214409.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.625440311431885,
|
|
"epoch": 0.9913793103448276,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004994397213615394,
|
|
"loss": 6.3662,
|
|
"mean_token_accuracy": 0.11870254352688789,
|
|
"num_tokens": 2223624.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.606072521209716,
|
|
"epoch": 0.9952978056426333,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004994183069493225,
|
|
"loss": 6.3187,
|
|
"mean_token_accuracy": 0.11965643540024758,
|
|
"num_tokens": 2231878.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.662061548233032,
|
|
"epoch": 0.9992163009404389,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004993964914652034,
|
|
"loss": 6.2584,
|
|
"mean_token_accuracy": 0.1252451278269291,
|
|
"num_tokens": 2240684.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.574692583084106,
|
|
"epoch": 1.0031347962382444,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004993742749481699,
|
|
"loss": 6.1326,
|
|
"mean_token_accuracy": 0.13472574055194855,
|
|
"num_tokens": 2248776.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.412195444107056,
|
|
"epoch": 1.0070532915360502,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004993516574379265,
|
|
"loss": 5.9829,
|
|
"mean_token_accuracy": 0.13876268342137338,
|
|
"num_tokens": 2257637.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.596172285079956,
|
|
"epoch": 1.0109717868338557,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004993286389748943,
|
|
"loss": 6.183,
|
|
"mean_token_accuracy": 0.1223673589527607,
|
|
"num_tokens": 2266602.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.4536356925964355,
|
|
"epoch": 1.0148902821316614,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004993052196002112,
|
|
"loss": 6.0342,
|
|
"mean_token_accuracy": 0.13276253119111062,
|
|
"num_tokens": 2275323.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.496698045730591,
|
|
"epoch": 1.0188087774294672,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004992813993557312,
|
|
"loss": 6.0591,
|
|
"mean_token_accuracy": 0.12783461958169937,
|
|
"num_tokens": 2284455.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.431064796447754,
|
|
"epoch": 1.0227272727272727,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000499257178284025,
|
|
"loss": 6.0945,
|
|
"mean_token_accuracy": 0.1290090449154377,
|
|
"num_tokens": 2293210.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.506451511383057,
|
|
"epoch": 1.0266457680250785,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004992325564283797,
|
|
"loss": 6.1456,
|
|
"mean_token_accuracy": 0.12814865931868552,
|
|
"num_tokens": 2302645.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.554348945617676,
|
|
"epoch": 1.030564263322884,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004992075338327984,
|
|
"loss": 6.1251,
|
|
"mean_token_accuracy": 0.13093000575900077,
|
|
"num_tokens": 2310590.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.577063179016113,
|
|
"epoch": 1.0344827586206897,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004991821105420006,
|
|
"loss": 6.16,
|
|
"mean_token_accuracy": 0.12589022517204285,
|
|
"num_tokens": 2319539.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.535883665084839,
|
|
"epoch": 1.0384012539184952,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004991562866014219,
|
|
"loss": 6.1524,
|
|
"mean_token_accuracy": 0.126813293248415,
|
|
"num_tokens": 2328934.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.542779064178466,
|
|
"epoch": 1.042319749216301,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.1033,
|
|
"mean_token_accuracy": 0.13062299489974977,
|
|
"num_tokens": 2338123.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.488020515441894,
|
|
"epoch": 1.0462382445141065,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004991034369562438,
|
|
"loss": 6.1583,
|
|
"mean_token_accuracy": 0.13101462051272392,
|
|
"num_tokens": 2346535.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.596034717559815,
|
|
"epoch": 1.0501567398119123,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004990764113460953,
|
|
"loss": 6.0472,
|
|
"mean_token_accuracy": 0.13994293436408042,
|
|
"num_tokens": 2355391.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.387597560882568,
|
|
"epoch": 1.0540752351097178,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004990489852750675,
|
|
"loss": 6.089,
|
|
"mean_token_accuracy": 0.13415274173021316,
|
|
"num_tokens": 2364325.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.458016014099121,
|
|
"epoch": 1.0579937304075235,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004990211587921751,
|
|
"loss": 6.0354,
|
|
"mean_token_accuracy": 0.13103872239589692,
|
|
"num_tokens": 2373615.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.614066696166992,
|
|
"epoch": 1.061912225705329,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004989929319471487,
|
|
"loss": 6.1897,
|
|
"mean_token_accuracy": 0.1241883710026741,
|
|
"num_tokens": 2382335.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.386125135421753,
|
|
"epoch": 1.0658307210031348,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004989643047904341,
|
|
"loss": 6.0857,
|
|
"mean_token_accuracy": 0.12894076779484748,
|
|
"num_tokens": 2391448.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.49520092010498,
|
|
"epoch": 1.0697492163009406,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004989352773731928,
|
|
"loss": 6.051,
|
|
"mean_token_accuracy": 0.13576387241482735,
|
|
"num_tokens": 2399697.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.450667142868042,
|
|
"epoch": 1.073667711598746,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004989058497473014,
|
|
"loss": 6.0102,
|
|
"mean_token_accuracy": 0.14315034449100494,
|
|
"num_tokens": 2408704.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.297135686874389,
|
|
"epoch": 1.0775862068965518,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004988760219653518,
|
|
"loss": 5.9451,
|
|
"mean_token_accuracy": 0.13236987814307213,
|
|
"num_tokens": 2417035.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.4906352996826175,
|
|
"epoch": 1.0815047021943573,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004988457940806513,
|
|
"loss": 6.0613,
|
|
"mean_token_accuracy": 0.1321863353252411,
|
|
"num_tokens": 2426226.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.316649675369263,
|
|
"epoch": 1.085423197492163,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004988151661472218,
|
|
"loss": 6.0177,
|
|
"mean_token_accuracy": 0.1341511443257332,
|
|
"num_tokens": 2435226.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.468252182006836,
|
|
"epoch": 1.0893416927899686,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004987841382198006,
|
|
"loss": 6.1007,
|
|
"mean_token_accuracy": 0.13451739475131036,
|
|
"num_tokens": 2444344.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.4722333431243895,
|
|
"epoch": 1.0932601880877744,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004987527103538394,
|
|
"loss": 6.0977,
|
|
"mean_token_accuracy": 0.1346490688621998,
|
|
"num_tokens": 2453965.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.517949199676513,
|
|
"epoch": 1.09717868338558,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000498720882605505,
|
|
"loss": 6.1152,
|
|
"mean_token_accuracy": 0.1274041309952736,
|
|
"num_tokens": 2462929.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.403267765045166,
|
|
"epoch": 1.1010971786833856,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004986886550316788,
|
|
"loss": 6.0646,
|
|
"mean_token_accuracy": 0.13348740786314012,
|
|
"num_tokens": 2471050.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.464239931106567,
|
|
"epoch": 1.1050156739811912,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004986560276899565,
|
|
"loss": 6.0249,
|
|
"mean_token_accuracy": 0.12662593945860862,
|
|
"num_tokens": 2479963.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.395254039764405,
|
|
"epoch": 1.108934169278997,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004986230006386488,
|
|
"loss": 6.0502,
|
|
"mean_token_accuracy": 0.13782202005386351,
|
|
"num_tokens": 2488525.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.461355304718017,
|
|
"epoch": 1.1128526645768024,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004985895739367802,
|
|
"loss": 6.1372,
|
|
"mean_token_accuracy": 0.13265417739748955,
|
|
"num_tokens": 2496892.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.468255186080933,
|
|
"epoch": 1.1167711598746082,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004985557476440895,
|
|
"loss": 6.1618,
|
|
"mean_token_accuracy": 0.12906021475791932,
|
|
"num_tokens": 2506004.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.465244293212891,
|
|
"epoch": 1.1206896551724137,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00049852152182103,
|
|
"loss": 6.0595,
|
|
"mean_token_accuracy": 0.13433735817670822,
|
|
"num_tokens": 2515068.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.416254186630249,
|
|
"epoch": 1.1246081504702194,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004984868965287686,
|
|
"loss": 6.0067,
|
|
"mean_token_accuracy": 0.13896693661808968,
|
|
"num_tokens": 2523908.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.395039653778076,
|
|
"epoch": 1.1285266457680252,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004984518718291864,
|
|
"loss": 6.0145,
|
|
"mean_token_accuracy": 0.13567812219262124,
|
|
"num_tokens": 2532187.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.39695463180542,
|
|
"epoch": 1.1324451410658307,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004984164477848783,
|
|
"loss": 5.9564,
|
|
"mean_token_accuracy": 0.14016151651740075,
|
|
"num_tokens": 2540615.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.3852643966674805,
|
|
"epoch": 1.1363636363636362,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004983806244591528,
|
|
"loss": 6.0761,
|
|
"mean_token_accuracy": 0.13532559722661971,
|
|
"num_tokens": 2549414.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.389466953277588,
|
|
"epoch": 1.140282131661442,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004983444019160318,
|
|
"loss": 6.0829,
|
|
"mean_token_accuracy": 0.13113784193992614,
|
|
"num_tokens": 2558496.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.3495019435882565,
|
|
"epoch": 1.1442006269592477,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004983077802202511,
|
|
"loss": 5.9905,
|
|
"mean_token_accuracy": 0.13659327030181884,
|
|
"num_tokens": 2566810.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.481125736236573,
|
|
"epoch": 1.1481191222570533,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004982707594372595,
|
|
"loss": 6.1072,
|
|
"mean_token_accuracy": 0.1357683263719082,
|
|
"num_tokens": 2576166.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.357020330429077,
|
|
"epoch": 1.152037617554859,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000498233339633219,
|
|
"loss": 5.9931,
|
|
"mean_token_accuracy": 0.14374880045652388,
|
|
"num_tokens": 2583954.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.3597229480743405,
|
|
"epoch": 1.1559561128526645,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000498195520875005,
|
|
"loss": 6.0535,
|
|
"mean_token_accuracy": 0.13409585878252983,
|
|
"num_tokens": 2593216.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.390771007537841,
|
|
"epoch": 1.1598746081504703,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004981573032302059,
|
|
"loss": 6.0011,
|
|
"mean_token_accuracy": 0.13834288418293,
|
|
"num_tokens": 2601294.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.356062984466552,
|
|
"epoch": 1.1637931034482758,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004981186867671225,
|
|
"loss": 6.0337,
|
|
"mean_token_accuracy": 0.1364253617823124,
|
|
"num_tokens": 2609932.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.325963640213013,
|
|
"epoch": 1.1677115987460815,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004980796715547691,
|
|
"loss": 5.9638,
|
|
"mean_token_accuracy": 0.1368013933300972,
|
|
"num_tokens": 2618287.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.45850830078125,
|
|
"epoch": 1.171630094043887,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004980402576628717,
|
|
"loss": 6.0361,
|
|
"mean_token_accuracy": 0.1395344130694866,
|
|
"num_tokens": 2626778.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.269816255569458,
|
|
"epoch": 1.1755485893416928,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004980004451618697,
|
|
"loss": 5.995,
|
|
"mean_token_accuracy": 0.13766313791275026,
|
|
"num_tokens": 2636166.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.1755485893416928,
|
|
"eval_entropy": 6.173720941987148,
|
|
"eval_loss": 6.259021759033203,
|
|
"eval_mean_token_accuracy": 0.13140278101660485,
|
|
"eval_num_tokens": 2636166.0,
|
|
"eval_runtime": 2.187,
|
|
"eval_samples_per_second": 1884.745,
|
|
"eval_steps_per_second": 235.936,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.4017609596252445,
|
|
"epoch": 1.1794670846394983,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004979602341229144,
|
|
"loss": 6.0458,
|
|
"mean_token_accuracy": 0.13523489534854888,
|
|
"num_tokens": 2645591.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.234326887130737,
|
|
"epoch": 1.183385579937304,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004979196246178694,
|
|
"loss": 5.9691,
|
|
"mean_token_accuracy": 0.13330597504973413,
|
|
"num_tokens": 2654531.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.397192525863647,
|
|
"epoch": 1.1873040752351098,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004978786167193105,
|
|
"loss": 6.0096,
|
|
"mean_token_accuracy": 0.13553697988390923,
|
|
"num_tokens": 2663097.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.267082834243775,
|
|
"epoch": 1.1912225705329154,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004978372105005254,
|
|
"loss": 5.9934,
|
|
"mean_token_accuracy": 0.13450803980231285,
|
|
"num_tokens": 2672413.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.4453002452850345,
|
|
"epoch": 1.1951410658307209,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004977954060355139,
|
|
"loss": 6.1125,
|
|
"mean_token_accuracy": 0.12924333065748214,
|
|
"num_tokens": 2680757.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.361589860916138,
|
|
"epoch": 1.1990595611285266,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004977532033989871,
|
|
"loss": 5.9663,
|
|
"mean_token_accuracy": 0.13531593978405,
|
|
"num_tokens": 2689551.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.324816799163818,
|
|
"epoch": 1.2029780564263324,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004977106026663681,
|
|
"loss": 5.9869,
|
|
"mean_token_accuracy": 0.13481266647577286,
|
|
"num_tokens": 2698518.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.3654069900512695,
|
|
"epoch": 1.206896551724138,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976676039137914,
|
|
"loss": 6.067,
|
|
"mean_token_accuracy": 0.13320212066173553,
|
|
"num_tokens": 2706833.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.337685918807983,
|
|
"epoch": 1.2108150470219436,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004976242072181026,
|
|
"loss": 6.0533,
|
|
"mean_token_accuracy": 0.12881046459078788,
|
|
"num_tokens": 2715752.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.467659616470337,
|
|
"epoch": 1.2147335423197492,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004975804126568587,
|
|
"loss": 6.1418,
|
|
"mean_token_accuracy": 0.13140107765793801,
|
|
"num_tokens": 2724394.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.332508373260498,
|
|
"epoch": 1.218652037617555,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004975362203083277,
|
|
"loss": 5.992,
|
|
"mean_token_accuracy": 0.1313839450478554,
|
|
"num_tokens": 2733456.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.30330719947815,
|
|
"epoch": 1.2225705329153604,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004974916302514886,
|
|
"loss": 5.9127,
|
|
"mean_token_accuracy": 0.14520679116249086,
|
|
"num_tokens": 2742059.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.368285703659057,
|
|
"epoch": 1.2264890282131662,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.0924,
|
|
"mean_token_accuracy": 0.13415124639868736,
|
|
"num_tokens": 2751046.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.326376438140869,
|
|
"epoch": 1.2304075235109717,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004974012573323545,
|
|
"loss": 5.9736,
|
|
"mean_token_accuracy": 0.1416211947798729,
|
|
"num_tokens": 2759234.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.4169456481933596,
|
|
"epoch": 1.2343260188087775,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004973554746315709,
|
|
"loss": 6.0346,
|
|
"mean_token_accuracy": 0.13679490983486176,
|
|
"num_tokens": 2767605.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.341928386688233,
|
|
"epoch": 1.238244514106583,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004973092945455009,
|
|
"loss": 5.9528,
|
|
"mean_token_accuracy": 0.13543061465024947,
|
|
"num_tokens": 2775749.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.245615148544312,
|
|
"epoch": 1.2421630094043887,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004972627171566757,
|
|
"loss": 6.0146,
|
|
"mean_token_accuracy": 0.13063447698950767,
|
|
"num_tokens": 2785097.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.32657527923584,
|
|
"epoch": 1.2460815047021945,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004972157425483368,
|
|
"loss": 6.0887,
|
|
"mean_token_accuracy": 0.1307818368077278,
|
|
"num_tokens": 2793953.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.417240858078003,
|
|
"epoch": 1.25,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004971683708044353,
|
|
"loss": 6.0422,
|
|
"mean_token_accuracy": 0.13612862303853035,
|
|
"num_tokens": 2802357.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.254428434371948,
|
|
"epoch": 1.2539184952978055,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004971206020096323,
|
|
"loss": 5.9901,
|
|
"mean_token_accuracy": 0.13007952123880387,
|
|
"num_tokens": 2812387.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.435764122009277,
|
|
"epoch": 1.2578369905956113,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004970724362492984,
|
|
"loss": 6.0676,
|
|
"mean_token_accuracy": 0.1369776926934719,
|
|
"num_tokens": 2821506.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.22492208480835,
|
|
"epoch": 1.261755485893417,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004970238736095135,
|
|
"loss": 5.9362,
|
|
"mean_token_accuracy": 0.13869768530130386,
|
|
"num_tokens": 2829820.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.296129131317139,
|
|
"epoch": 1.2656739811912225,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004969749141770671,
|
|
"loss": 5.8468,
|
|
"mean_token_accuracy": 0.13930305540561677,
|
|
"num_tokens": 2838337.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.229622411727905,
|
|
"epoch": 1.2695924764890283,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004969255580394575,
|
|
"loss": 5.9923,
|
|
"mean_token_accuracy": 0.1379152424633503,
|
|
"num_tokens": 2846919.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.421029376983642,
|
|
"epoch": 1.2735109717868338,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000496875805284892,
|
|
"loss": 6.0319,
|
|
"mean_token_accuracy": 0.13647375255823135,
|
|
"num_tokens": 2855447.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.268044948577881,
|
|
"epoch": 1.2774294670846396,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004968256560022871,
|
|
"loss": 5.9751,
|
|
"mean_token_accuracy": 0.14186585023999215,
|
|
"num_tokens": 2863581.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.380617952346801,
|
|
"epoch": 1.281347962382445,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004967751102812676,
|
|
"loss": 6.1009,
|
|
"mean_token_accuracy": 0.13440172523260116,
|
|
"num_tokens": 2871968.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.368293046951294,
|
|
"epoch": 1.2852664576802508,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004967241682121669,
|
|
"loss": 6.0673,
|
|
"mean_token_accuracy": 0.12853180840611458,
|
|
"num_tokens": 2881142.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.3551513671875,
|
|
"epoch": 1.2891849529780564,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004966728298860267,
|
|
"loss": 6.0005,
|
|
"mean_token_accuracy": 0.1313441164791584,
|
|
"num_tokens": 2889243.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.305865955352783,
|
|
"epoch": 1.293103448275862,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004966210953945969,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.13025051727890968,
|
|
"num_tokens": 2898068.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.292298889160156,
|
|
"epoch": 1.2970219435736676,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004965689648303355,
|
|
"loss": 5.9281,
|
|
"mean_token_accuracy": 0.13824338987469673,
|
|
"num_tokens": 2907011.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.277693414688111,
|
|
"epoch": 1.3009404388714734,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004965164382864083,
|
|
"loss": 5.9943,
|
|
"mean_token_accuracy": 0.13214352130889892,
|
|
"num_tokens": 2916449.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.307842111587524,
|
|
"epoch": 1.3048589341692791,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004964635158566886,
|
|
"loss": 5.8851,
|
|
"mean_token_accuracy": 0.1405140846967697,
|
|
"num_tokens": 2924944.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.2674915313720705,
|
|
"epoch": 1.3087774294670846,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004964101976357574,
|
|
"loss": 5.9389,
|
|
"mean_token_accuracy": 0.14335716813802718,
|
|
"num_tokens": 2933691.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.21014518737793,
|
|
"epoch": 1.3126959247648902,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000496356483718903,
|
|
"loss": 5.9835,
|
|
"mean_token_accuracy": 0.1383981078863144,
|
|
"num_tokens": 2942270.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.325891971588135,
|
|
"epoch": 1.316614420062696,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004963023742021211,
|
|
"loss": 5.9468,
|
|
"mean_token_accuracy": 0.14205208867788316,
|
|
"num_tokens": 2951106.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.258494567871094,
|
|
"epoch": 1.3205329153605017,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000496247869182114,
|
|
"loss": 5.9536,
|
|
"mean_token_accuracy": 0.13559493273496628,
|
|
"num_tokens": 2959962.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.24164342880249,
|
|
"epoch": 1.3244514106583072,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004961929687562909,
|
|
"loss": 6.0586,
|
|
"mean_token_accuracy": 0.1302129790186882,
|
|
"num_tokens": 2969191.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.263980293273926,
|
|
"epoch": 1.328369905956113,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004961376730227681,
|
|
"loss": 5.8616,
|
|
"mean_token_accuracy": 0.14607683569192886,
|
|
"num_tokens": 2977811.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.220906591415405,
|
|
"epoch": 1.3322884012539185,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004960819820803675,
|
|
"loss": 5.8989,
|
|
"mean_token_accuracy": 0.14086293578147888,
|
|
"num_tokens": 2986956.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.353243970870972,
|
|
"epoch": 1.3362068965517242,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004960258960286185,
|
|
"loss": 6.0035,
|
|
"mean_token_accuracy": 0.13733470514416696,
|
|
"num_tokens": 2996275.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.239503669738769,
|
|
"epoch": 1.3401253918495297,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004959694149677553,
|
|
"loss": 5.9373,
|
|
"mean_token_accuracy": 0.13988570868968964,
|
|
"num_tokens": 3005231.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.270683622360229,
|
|
"epoch": 1.3440438871473355,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004959125389987193,
|
|
"loss": 5.8965,
|
|
"mean_token_accuracy": 0.13760650977492334,
|
|
"num_tokens": 3013591.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.228696060180664,
|
|
"epoch": 1.347962382445141,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004958552682231567,
|
|
"loss": 5.9571,
|
|
"mean_token_accuracy": 0.13878614008426665,
|
|
"num_tokens": 3021765.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.3308008193969725,
|
|
"epoch": 1.3518808777429467,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004957976027434199,
|
|
"loss": 5.9403,
|
|
"mean_token_accuracy": 0.14569852352142335,
|
|
"num_tokens": 3030269.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.19557523727417,
|
|
"epoch": 1.3557993730407523,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004957395426625663,
|
|
"loss": 6.0358,
|
|
"mean_token_accuracy": 0.1339428573846817,
|
|
"num_tokens": 3039886.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.347655582427978,
|
|
"epoch": 1.359717868338558,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004956810880843587,
|
|
"loss": 6.0358,
|
|
"mean_token_accuracy": 0.1318750225007534,
|
|
"num_tokens": 3049256.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.270250701904297,
|
|
"epoch": 1.3636363636363638,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000495622239113265,
|
|
"loss": 5.9569,
|
|
"mean_token_accuracy": 0.13846379667520523,
|
|
"num_tokens": 3058046.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.2957230567932125,
|
|
"epoch": 1.3675548589341693,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004955629958544577,
|
|
"loss": 6.0219,
|
|
"mean_token_accuracy": 0.1412307657301426,
|
|
"num_tokens": 3066594.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.244470834732056,
|
|
"epoch": 1.3714733542319748,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004955033584138143,
|
|
"loss": 5.8459,
|
|
"mean_token_accuracy": 0.14759366884827613,
|
|
"num_tokens": 3075274.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.258149719238281,
|
|
"epoch": 1.3753918495297806,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004954433268979164,
|
|
"loss": 6.0169,
|
|
"mean_token_accuracy": 0.13895999193191527,
|
|
"num_tokens": 3083937.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.3190703868865965,
|
|
"epoch": 1.3793103448275863,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004953829014140502,
|
|
"loss": 6.0202,
|
|
"mean_token_accuracy": 0.13991375565528869,
|
|
"num_tokens": 3093598.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.115016126632691,
|
|
"epoch": 1.3832288401253918,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004953220820702057,
|
|
"loss": 5.8881,
|
|
"mean_token_accuracy": 0.136306519061327,
|
|
"num_tokens": 3102290.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.317786741256714,
|
|
"epoch": 1.3871473354231976,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004952608689750771,
|
|
"loss": 5.8844,
|
|
"mean_token_accuracy": 0.13860043063759803,
|
|
"num_tokens": 3111847.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.141478586196899,
|
|
"epoch": 1.391065830721003,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004951992622380619,
|
|
"loss": 5.9097,
|
|
"mean_token_accuracy": 0.1414356455206871,
|
|
"num_tokens": 3120755.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.2546045780181885,
|
|
"epoch": 1.3949843260188088,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004951372619692615,
|
|
"loss": 6.031,
|
|
"mean_token_accuracy": 0.1300249621272087,
|
|
"num_tokens": 3129910.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.271708345413208,
|
|
"epoch": 1.3989028213166144,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004950748682794804,
|
|
"loss": 5.884,
|
|
"mean_token_accuracy": 0.1406226009130478,
|
|
"num_tokens": 3138620.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.193848514556885,
|
|
"epoch": 1.40282131661442,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004950120812802262,
|
|
"loss": 5.9795,
|
|
"mean_token_accuracy": 0.14275466799736022,
|
|
"num_tokens": 3147075.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.231546831130982,
|
|
"epoch": 1.4067398119122256,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004949489010837095,
|
|
"loss": 5.897,
|
|
"mean_token_accuracy": 0.13693352341651915,
|
|
"num_tokens": 3156176.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.210837507247925,
|
|
"epoch": 1.4106583072100314,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 5.98,
|
|
"mean_token_accuracy": 0.13918429240584373,
|
|
"num_tokens": 3165056.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.259754753112793,
|
|
"epoch": 1.414576802507837,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000494821361551244,
|
|
"loss": 5.9143,
|
|
"mean_token_accuracy": 0.1359445720911026,
|
|
"num_tokens": 3174240.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.198961973190308,
|
|
"epoch": 1.4184952978056427,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004947570024432291,
|
|
"loss": 5.9332,
|
|
"mean_token_accuracy": 0.13835996463894845,
|
|
"num_tokens": 3182989.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.2340131282806395,
|
|
"epoch": 1.4224137931034484,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004946922505938189,
|
|
"loss": 5.8808,
|
|
"mean_token_accuracy": 0.14476428106427192,
|
|
"num_tokens": 3192000.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.166196298599243,
|
|
"epoch": 1.426332288401254,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004946271061187354,
|
|
"loss": 5.9661,
|
|
"mean_token_accuracy": 0.1432674393057823,
|
|
"num_tokens": 3200310.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.282206439971924,
|
|
"epoch": 1.4302507836990594,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004945615691344025,
|
|
"loss": 5.9586,
|
|
"mean_token_accuracy": 0.14152021259069442,
|
|
"num_tokens": 3208436.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.183050870895386,
|
|
"epoch": 1.4341692789968652,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004944956397579453,
|
|
"loss": 5.8666,
|
|
"mean_token_accuracy": 0.1475970722734928,
|
|
"num_tokens": 3217164.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.276995325088501,
|
|
"epoch": 1.438087774294671,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004944293181071902,
|
|
"loss": 5.8691,
|
|
"mean_token_accuracy": 0.14128537550568582,
|
|
"num_tokens": 3225736.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.09270133972168,
|
|
"epoch": 1.4420062695924765,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004943626043006649,
|
|
"loss": 5.9365,
|
|
"mean_token_accuracy": 0.14045744836330415,
|
|
"num_tokens": 3233772.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.316040945053101,
|
|
"epoch": 1.4459247648902822,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000494295498457598,
|
|
"loss": 5.8955,
|
|
"mean_token_accuracy": 0.13631879016757012,
|
|
"num_tokens": 3243687.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.130023431777954,
|
|
"epoch": 1.4498432601880877,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004942280006979179,
|
|
"loss": 5.9009,
|
|
"mean_token_accuracy": 0.1420040473341942,
|
|
"num_tokens": 3252464.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.223214721679687,
|
|
"epoch": 1.4537617554858935,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004941601111422546,
|
|
"loss": 5.7476,
|
|
"mean_token_accuracy": 0.16175038665533065,
|
|
"num_tokens": 3260402.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.175519418716431,
|
|
"epoch": 1.457680250783699,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004940918299119375,
|
|
"loss": 6.0455,
|
|
"mean_token_accuracy": 0.1407065212726593,
|
|
"num_tokens": 3268838.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.233507490158081,
|
|
"epoch": 1.4615987460815048,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004940231571289962,
|
|
"loss": 5.8681,
|
|
"mean_token_accuracy": 0.14759975522756577,
|
|
"num_tokens": 3277603.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.156610918045044,
|
|
"epoch": 1.4655172413793103,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004939540929161603,
|
|
"loss": 5.9051,
|
|
"mean_token_accuracy": 0.14266798198223113,
|
|
"num_tokens": 3286514.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.179834365844727,
|
|
"epoch": 1.469435736677116,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004938846373968586,
|
|
"loss": 5.9061,
|
|
"mean_token_accuracy": 0.14338089302182197,
|
|
"num_tokens": 3295613.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.231829214096069,
|
|
"epoch": 1.4733542319749215,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004938147906952194,
|
|
"loss": 5.9052,
|
|
"mean_token_accuracy": 0.14310697540640832,
|
|
"num_tokens": 3303543.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.202773523330689,
|
|
"epoch": 1.4772727272727273,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00049374455293607,
|
|
"loss": 5.8316,
|
|
"mean_token_accuracy": 0.14337966665625573,
|
|
"num_tokens": 3311812.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.290017747879029,
|
|
"epoch": 1.481191222570533,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004936739242449369,
|
|
"loss": 6.0038,
|
|
"mean_token_accuracy": 0.13627132177352905,
|
|
"num_tokens": 3319908.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.288265419006348,
|
|
"epoch": 1.4851097178683386,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004936029047480447,
|
|
"loss": 5.9412,
|
|
"mean_token_accuracy": 0.1356036700308323,
|
|
"num_tokens": 3329572.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.154357957839966,
|
|
"epoch": 1.489028213166144,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004935314945723171,
|
|
"loss": 5.7974,
|
|
"mean_token_accuracy": 0.140255356580019,
|
|
"num_tokens": 3338519.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.1016316413879395,
|
|
"epoch": 1.4929467084639498,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004934596938453754,
|
|
"loss": 5.8113,
|
|
"mean_token_accuracy": 0.14027941226959229,
|
|
"num_tokens": 3347309.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.081620407104492,
|
|
"epoch": 1.4968652037617556,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004933875026955391,
|
|
"loss": 5.9604,
|
|
"mean_token_accuracy": 0.13251368328928947,
|
|
"num_tokens": 3356802.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.262990856170655,
|
|
"epoch": 1.500783699059561,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004933149212518258,
|
|
"loss": 5.8623,
|
|
"mean_token_accuracy": 0.14324725568294525,
|
|
"num_tokens": 3365101.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.14485125541687,
|
|
"epoch": 1.5047021943573666,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004932419496439501,
|
|
"loss": 5.8584,
|
|
"mean_token_accuracy": 0.1449149876832962,
|
|
"num_tokens": 3373902.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.166677236557007,
|
|
"epoch": 1.5086206896551724,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000493168588002324,
|
|
"loss": 5.8717,
|
|
"mean_token_accuracy": 0.14935290515422822,
|
|
"num_tokens": 3382021.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.057049131393432,
|
|
"epoch": 1.5125391849529781,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004930948364580569,
|
|
"loss": 5.8447,
|
|
"mean_token_accuracy": 0.14263902455568314,
|
|
"num_tokens": 3390351.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.163201141357422,
|
|
"epoch": 1.5164576802507836,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004930206951429546,
|
|
"loss": 5.7452,
|
|
"mean_token_accuracy": 0.14573136419057847,
|
|
"num_tokens": 3398728.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.141188287734986,
|
|
"epoch": 1.5203761755485894,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004929461641895197,
|
|
"loss": 5.9302,
|
|
"mean_token_accuracy": 0.144403874874115,
|
|
"num_tokens": 3407466.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.2316145420074465,
|
|
"epoch": 1.5242946708463951,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000492871243730951,
|
|
"loss": 5.9656,
|
|
"mean_token_accuracy": 0.13916341140866278,
|
|
"num_tokens": 3416656.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.236765193939209,
|
|
"epoch": 1.5282131661442007,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004927959339011437,
|
|
"loss": 5.9442,
|
|
"mean_token_accuracy": 0.14159360527992249,
|
|
"num_tokens": 3425763.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.100015354156494,
|
|
"epoch": 1.5321316614420062,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004927202348346885,
|
|
"loss": 5.8372,
|
|
"mean_token_accuracy": 0.1559658631682396,
|
|
"num_tokens": 3434246.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.194915390014648,
|
|
"epoch": 1.536050156739812,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000492644146666872,
|
|
"loss": 5.9234,
|
|
"mean_token_accuracy": 0.14269987791776656,
|
|
"num_tokens": 3443323.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.114914894104004,
|
|
"epoch": 1.5399686520376177,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004925676695336761,
|
|
"loss": 5.824,
|
|
"mean_token_accuracy": 0.14860626459121704,
|
|
"num_tokens": 3452713.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.014723396301269,
|
|
"epoch": 1.5438871473354232,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004924908035717777,
|
|
"loss": 5.8561,
|
|
"mean_token_accuracy": 0.13924858197569848,
|
|
"num_tokens": 3461536.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.227632474899292,
|
|
"epoch": 1.5478056426332287,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492413548918549,
|
|
"loss": 5.7851,
|
|
"mean_token_accuracy": 0.14539091512560845,
|
|
"num_tokens": 3470747.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.139095878601074,
|
|
"epoch": 1.5517241379310345,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004923359057120563,
|
|
"loss": 5.887,
|
|
"mean_token_accuracy": 0.14268513917922973,
|
|
"num_tokens": 3479094.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.177683877944946,
|
|
"epoch": 1.5556426332288402,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004922578740910608,
|
|
"loss": 5.8648,
|
|
"mean_token_accuracy": 0.14357393532991408,
|
|
"num_tokens": 3487936.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.128460693359375,
|
|
"epoch": 1.5595611285266457,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004921794541950177,
|
|
"loss": 5.8827,
|
|
"mean_token_accuracy": 0.1405430756509304,
|
|
"num_tokens": 3496777.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.100703239440918,
|
|
"epoch": 1.5634796238244513,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004921006461640758,
|
|
"loss": 5.9223,
|
|
"mean_token_accuracy": 0.1409299850463867,
|
|
"num_tokens": 3506395.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.151146793365479,
|
|
"epoch": 1.567398119122257,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492021450139078,
|
|
"loss": 5.8812,
|
|
"mean_token_accuracy": 0.1376201443374157,
|
|
"num_tokens": 3515715.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.567398119122257,
|
|
"eval_entropy": 5.957889068034268,
|
|
"eval_loss": 6.054129123687744,
|
|
"eval_mean_token_accuracy": 0.14233148432979287,
|
|
"eval_num_tokens": 3515715.0,
|
|
"eval_runtime": 2.3774,
|
|
"eval_samples_per_second": 1733.822,
|
|
"eval_steps_per_second": 217.043,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.229584407806397,
|
|
"epoch": 1.5713166144200628,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004919418662615605,
|
|
"loss": 5.8782,
|
|
"mean_token_accuracy": 0.14555389285087586,
|
|
"num_tokens": 3525634.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.172345399856567,
|
|
"epoch": 1.5752351097178683,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004918618946737525,
|
|
"loss": 5.9547,
|
|
"mean_token_accuracy": 0.14259756803512574,
|
|
"num_tokens": 3534736.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.186705112457275,
|
|
"epoch": 1.579153605015674,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004917815355185762,
|
|
"loss": 5.857,
|
|
"mean_token_accuracy": 0.14343542978167534,
|
|
"num_tokens": 3544168.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.16968264579773,
|
|
"epoch": 1.5830721003134798,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004917007889396464,
|
|
"loss": 5.9045,
|
|
"mean_token_accuracy": 0.14173540771007537,
|
|
"num_tokens": 3553347.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.204205799102783,
|
|
"epoch": 1.5869905956112853,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004916196550812706,
|
|
"loss": 5.8811,
|
|
"mean_token_accuracy": 0.14535140842199326,
|
|
"num_tokens": 3561800.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.11736741065979,
|
|
"epoch": 1.5909090909090908,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004915381340884477,
|
|
"loss": 5.8708,
|
|
"mean_token_accuracy": 0.1454767882823944,
|
|
"num_tokens": 3570446.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.154869079589844,
|
|
"epoch": 1.5948275862068966,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 5.7631,
|
|
"mean_token_accuracy": 0.15665549635887147,
|
|
"num_tokens": 3578463.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.121966695785522,
|
|
"epoch": 1.5987460815047023,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004913739312829181,
|
|
"loss": 5.9,
|
|
"mean_token_accuracy": 0.14602160453796387,
|
|
"num_tokens": 3587792.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.259881401062012,
|
|
"epoch": 1.6026645768025078,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004912912497636683,
|
|
"loss": 5.9925,
|
|
"mean_token_accuracy": 0.13647706136107446,
|
|
"num_tokens": 3596679.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.139722681045532,
|
|
"epoch": 1.6065830721003134,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004912081816968853,
|
|
"loss": 5.7875,
|
|
"mean_token_accuracy": 0.14886388033628464,
|
|
"num_tokens": 3605639.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.983063745498657,
|
|
"epoch": 1.6105015673981191,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000491124727231025,
|
|
"loss": 5.8421,
|
|
"mean_token_accuracy": 0.14174532443284987,
|
|
"num_tokens": 3614337.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.163030338287354,
|
|
"epoch": 1.6144200626959249,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004910408865152343,
|
|
"loss": 5.8499,
|
|
"mean_token_accuracy": 0.14399999976158143,
|
|
"num_tokens": 3622721.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.102843570709228,
|
|
"epoch": 1.6183385579937304,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004909566596993498,
|
|
"loss": 5.826,
|
|
"mean_token_accuracy": 0.15347028970718385,
|
|
"num_tokens": 3631693.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.0675195217132565,
|
|
"epoch": 1.622257053291536,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004908720469338988,
|
|
"loss": 5.7105,
|
|
"mean_token_accuracy": 0.14580216705799104,
|
|
"num_tokens": 3639973.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.116571426391602,
|
|
"epoch": 1.6261755485893417,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004907870483700979,
|
|
"loss": 5.8295,
|
|
"mean_token_accuracy": 0.14759061485528946,
|
|
"num_tokens": 3648214.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.104803371429443,
|
|
"epoch": 1.6300940438871474,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004907016641598534,
|
|
"loss": 5.8842,
|
|
"mean_token_accuracy": 0.14536356553435326,
|
|
"num_tokens": 3657131.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.1064859390258786,
|
|
"epoch": 1.634012539184953,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004906158944557607,
|
|
"loss": 5.8074,
|
|
"mean_token_accuracy": 0.1481236070394516,
|
|
"num_tokens": 3665937.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.076762342453003,
|
|
"epoch": 1.6379310344827587,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000490529739411104,
|
|
"loss": 5.867,
|
|
"mean_token_accuracy": 0.15004584193229675,
|
|
"num_tokens": 3674019.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.974442720413208,
|
|
"epoch": 1.6418495297805644,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004904431991798565,
|
|
"loss": 5.686,
|
|
"mean_token_accuracy": 0.14843811392784118,
|
|
"num_tokens": 3682462.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.106408977508545,
|
|
"epoch": 1.64576802507837,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004903562739166797,
|
|
"loss": 5.7834,
|
|
"mean_token_accuracy": 0.15050000250339507,
|
|
"num_tokens": 3691131.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.08671007156372,
|
|
"epoch": 1.6496865203761755,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004902689637769229,
|
|
"loss": 5.7791,
|
|
"mean_token_accuracy": 0.1494972288608551,
|
|
"num_tokens": 3699376.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.02228045463562,
|
|
"epoch": 1.6536050156739812,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004901812689166237,
|
|
"loss": 5.7203,
|
|
"mean_token_accuracy": 0.14922843724489213,
|
|
"num_tokens": 3708453.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.046120929718017,
|
|
"epoch": 1.657523510971787,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004900931894925069,
|
|
"loss": 5.8269,
|
|
"mean_token_accuracy": 0.14712499976158142,
|
|
"num_tokens": 3716621.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.119449806213379,
|
|
"epoch": 1.6614420062695925,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004900047256619849,
|
|
"loss": 5.7907,
|
|
"mean_token_accuracy": 0.1447051614522934,
|
|
"num_tokens": 3725902.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.04277720451355,
|
|
"epoch": 1.665360501567398,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004899158775831566,
|
|
"loss": 5.7357,
|
|
"mean_token_accuracy": 0.148492743819952,
|
|
"num_tokens": 3734979.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.104778003692627,
|
|
"epoch": 1.6692789968652038,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004898266454148081,
|
|
"loss": 5.8606,
|
|
"mean_token_accuracy": 0.14171048104763032,
|
|
"num_tokens": 3744414.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.047485303878784,
|
|
"epoch": 1.6731974921630095,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004897370293164119,
|
|
"loss": 5.699,
|
|
"mean_token_accuracy": 0.15120823681354523,
|
|
"num_tokens": 3752918.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.0623266220092775,
|
|
"epoch": 1.677115987460815,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004896470294481262,
|
|
"loss": 5.8615,
|
|
"mean_token_accuracy": 0.14743089601397513,
|
|
"num_tokens": 3762708.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.266351175308228,
|
|
"epoch": 1.6810344827586206,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004895566459707954,
|
|
"loss": 5.9974,
|
|
"mean_token_accuracy": 0.14403848201036454,
|
|
"num_tokens": 3770247.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.10488338470459,
|
|
"epoch": 1.6849529780564263,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004894658790459498,
|
|
"loss": 5.789,
|
|
"mean_token_accuracy": 0.14149514809250832,
|
|
"num_tokens": 3779479.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.109644269943237,
|
|
"epoch": 1.688871473354232,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004893747288358041,
|
|
"loss": 5.8015,
|
|
"mean_token_accuracy": 0.1492653086781502,
|
|
"num_tokens": 3787577.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.072000217437744,
|
|
"epoch": 1.6927899686520376,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000489283195503259,
|
|
"loss": 5.8078,
|
|
"mean_token_accuracy": 0.1505455732345581,
|
|
"num_tokens": 3795917.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.088470602035523,
|
|
"epoch": 1.6967084639498433,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000489191279211899,
|
|
"loss": 5.7941,
|
|
"mean_token_accuracy": 0.14202264100313186,
|
|
"num_tokens": 3804688.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.041987323760987,
|
|
"epoch": 1.700626959247649,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004890989801259935,
|
|
"loss": 5.7271,
|
|
"mean_token_accuracy": 0.14793986529111863,
|
|
"num_tokens": 3813144.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.105598306655883,
|
|
"epoch": 1.7045454545454546,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000489006298410496,
|
|
"loss": 5.8236,
|
|
"mean_token_accuracy": 0.14774255454540253,
|
|
"num_tokens": 3821669.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.185619688034057,
|
|
"epoch": 1.70846394984326,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004889132342310438,
|
|
"loss": 5.8982,
|
|
"mean_token_accuracy": 0.1427639827132225,
|
|
"num_tokens": 3830710.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.134215688705444,
|
|
"epoch": 1.7123824451410659,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004888197877539577,
|
|
"loss": 5.8238,
|
|
"mean_token_accuracy": 0.14964752048254013,
|
|
"num_tokens": 3839339.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.0955894947052,
|
|
"epoch": 1.7163009404388716,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004887259591462417,
|
|
"loss": 5.9303,
|
|
"mean_token_accuracy": 0.14224131032824516,
|
|
"num_tokens": 3848852.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.117247819900513,
|
|
"epoch": 1.7202194357366771,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004886317485755825,
|
|
"loss": 5.8524,
|
|
"mean_token_accuracy": 0.14823238849639891,
|
|
"num_tokens": 3857462.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.019983720779419,
|
|
"epoch": 1.7241379310344827,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004885371562103498,
|
|
"loss": 5.737,
|
|
"mean_token_accuracy": 0.14791555404663087,
|
|
"num_tokens": 3865981.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.106775569915771,
|
|
"epoch": 1.7280564263322884,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004884421822195957,
|
|
"loss": 5.7653,
|
|
"mean_token_accuracy": 0.14729381501674652,
|
|
"num_tokens": 3875004.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.128741264343262,
|
|
"epoch": 1.7319749216300941,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004883468267730538,
|
|
"loss": 5.9346,
|
|
"mean_token_accuracy": 0.13758371993899346,
|
|
"num_tokens": 3884206.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.124679040908814,
|
|
"epoch": 1.7358934169278997,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048825109004114006,
|
|
"loss": 5.8245,
|
|
"mean_token_accuracy": 0.14559186846017838,
|
|
"num_tokens": 3892606.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.027120685577392,
|
|
"epoch": 1.7398119122257052,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004881549721949513,
|
|
"loss": 5.8117,
|
|
"mean_token_accuracy": 0.14804096147418022,
|
|
"num_tokens": 3901550.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.100217723846436,
|
|
"epoch": 1.743730407523511,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004880584734062658,
|
|
"loss": 5.7985,
|
|
"mean_token_accuracy": 0.14690453559160233,
|
|
"num_tokens": 3910680.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.952231168746948,
|
|
"epoch": 1.7476489028213167,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004879615938475425,
|
|
"loss": 5.6739,
|
|
"mean_token_accuracy": 0.1547689452767372,
|
|
"num_tokens": 3919414.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.106700086593628,
|
|
"epoch": 1.7515673981191222,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004878643336919209,
|
|
"loss": 5.7861,
|
|
"mean_token_accuracy": 0.1393340364098549,
|
|
"num_tokens": 3928504.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.0178186893463135,
|
|
"epoch": 1.7554858934169277,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004877666931132206,
|
|
"loss": 5.7395,
|
|
"mean_token_accuracy": 0.14381116032600402,
|
|
"num_tokens": 3936886.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.063669204711914,
|
|
"epoch": 1.7594043887147337,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004876686722859413,
|
|
"loss": 5.7721,
|
|
"mean_token_accuracy": 0.15146784335374833,
|
|
"num_tokens": 3945203.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.043780565261841,
|
|
"epoch": 1.7633228840125392,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000487570271385262,
|
|
"loss": 5.7623,
|
|
"mean_token_accuracy": 0.14916835129261016,
|
|
"num_tokens": 3953432.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.060326099395752,
|
|
"epoch": 1.7672413793103448,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004874714905870411,
|
|
"loss": 5.8551,
|
|
"mean_token_accuracy": 0.14015527591109275,
|
|
"num_tokens": 3963318.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.073757219314575,
|
|
"epoch": 1.7711598746081505,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004873723300678159,
|
|
"loss": 5.8078,
|
|
"mean_token_accuracy": 0.1509538471698761,
|
|
"num_tokens": 3971869.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.105118227005005,
|
|
"epoch": 1.7750783699059562,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048727279000480226,
|
|
"loss": 5.9031,
|
|
"mean_token_accuracy": 0.1482295013964176,
|
|
"num_tokens": 3980766.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.1350971221923825,
|
|
"epoch": 1.7789968652037618,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 5.8363,
|
|
"mean_token_accuracy": 0.15101172104477883,
|
|
"num_tokens": 3989875.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.943312263488769,
|
|
"epoch": 1.7829153605015673,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004870725719596648,
|
|
"loss": 5.7263,
|
|
"mean_token_accuracy": 0.1508297622203827,
|
|
"num_tokens": 3998999.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.167447471618653,
|
|
"epoch": 1.786833855799373,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004869718943353631,
|
|
"loss": 5.8371,
|
|
"mean_token_accuracy": 0.14098343029618263,
|
|
"num_tokens": 4007402.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.102819871902466,
|
|
"epoch": 1.7907523510971788,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048687083788291656,
|
|
"loss": 5.8223,
|
|
"mean_token_accuracy": 0.1473478853702545,
|
|
"num_tokens": 4016369.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.0090995788574215,
|
|
"epoch": 1.7946708463949843,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048676940278292953,
|
|
"loss": 5.7481,
|
|
"mean_token_accuracy": 0.15214203149080277,
|
|
"num_tokens": 4024737.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.063549613952636,
|
|
"epoch": 1.7985893416927898,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048666758921668286,
|
|
"loss": 5.8096,
|
|
"mean_token_accuracy": 0.15026896223425865,
|
|
"num_tokens": 4032980.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.232597684860229,
|
|
"epoch": 1.8025078369905956,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048656539736613403,
|
|
"loss": 5.9992,
|
|
"mean_token_accuracy": 0.13911110311746597,
|
|
"num_tokens": 4041609.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.064432430267334,
|
|
"epoch": 1.8064263322884013,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004864628274139164,
|
|
"loss": 5.762,
|
|
"mean_token_accuracy": 0.15377292186021804,
|
|
"num_tokens": 4050012.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.963364505767823,
|
|
"epoch": 1.8103448275862069,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004863598795433391,
|
|
"loss": 5.7969,
|
|
"mean_token_accuracy": 0.1486699603497982,
|
|
"num_tokens": 4059229.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.081252670288086,
|
|
"epoch": 1.8142633228840124,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048625655393838666,
|
|
"loss": 5.7532,
|
|
"mean_token_accuracy": 0.15345848500728607,
|
|
"num_tokens": 4067809.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.0162107944488525,
|
|
"epoch": 1.8181818181818183,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004861528507837186,
|
|
"loss": 5.7906,
|
|
"mean_token_accuracy": 0.14964328408241273,
|
|
"num_tokens": 4076516.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.144069385528565,
|
|
"epoch": 1.8221003134796239,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004860487702646695,
|
|
"loss": 5.8854,
|
|
"mean_token_accuracy": 0.14570486098527907,
|
|
"num_tokens": 4086182.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.115158796310425,
|
|
"epoch": 1.8260188087774294,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004859443125672479,
|
|
"loss": 5.887,
|
|
"mean_token_accuracy": 0.13880274668335915,
|
|
"num_tokens": 4096134.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.041926574707031,
|
|
"epoch": 1.8299373040752351,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004858394778781368,
|
|
"loss": 5.7428,
|
|
"mean_token_accuracy": 0.14992253258824348,
|
|
"num_tokens": 4104933.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.0705280780792235,
|
|
"epoch": 1.8338557993730409,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004857342663846927,
|
|
"loss": 5.8781,
|
|
"mean_token_accuracy": 0.14462052136659623,
|
|
"num_tokens": 4113579.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.056195211410523,
|
|
"epoch": 1.8377742946708464,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004856286782749456,
|
|
"loss": 5.7038,
|
|
"mean_token_accuracy": 0.146064330637455,
|
|
"num_tokens": 4122263.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.050678300857544,
|
|
"epoch": 1.841692789968652,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004855227137375986,
|
|
"loss": 5.8266,
|
|
"mean_token_accuracy": 0.14227414950728418,
|
|
"num_tokens": 4131540.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.064000082015991,
|
|
"epoch": 1.8456112852664577,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004854163729620275,
|
|
"loss": 5.778,
|
|
"mean_token_accuracy": 0.14523486718535422,
|
|
"num_tokens": 4141495.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.964341926574707,
|
|
"epoch": 1.8495297805642634,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004853096561382805,
|
|
"loss": 5.6818,
|
|
"mean_token_accuracy": 0.1521722063422203,
|
|
"num_tokens": 4150522.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.92801456451416,
|
|
"epoch": 1.853448275862069,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004852025634570779,
|
|
"loss": 5.6972,
|
|
"mean_token_accuracy": 0.14957979694008827,
|
|
"num_tokens": 4159477.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.110127210617065,
|
|
"epoch": 1.8573667711598745,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004850950951098116,
|
|
"loss": 5.7867,
|
|
"mean_token_accuracy": 0.15120171010494232,
|
|
"num_tokens": 4167910.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.9060139656066895,
|
|
"epoch": 1.8612852664576802,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004849872512885451,
|
|
"loss": 5.7092,
|
|
"mean_token_accuracy": 0.15535110160708426,
|
|
"num_tokens": 4175965.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.035844087600708,
|
|
"epoch": 1.865203761755486,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004848790321860127,
|
|
"loss": 5.6906,
|
|
"mean_token_accuracy": 0.15866869688034058,
|
|
"num_tokens": 4183992.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.020617151260376,
|
|
"epoch": 1.8691222570532915,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048477043799561946,
|
|
"loss": 5.7431,
|
|
"mean_token_accuracy": 0.15084972977638245,
|
|
"num_tokens": 4192392.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.070355749130249,
|
|
"epoch": 1.873040752351097,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004846614689114409,
|
|
"loss": 5.747,
|
|
"mean_token_accuracy": 0.15154491364955902,
|
|
"num_tokens": 4201258.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.0385133743286135,
|
|
"epoch": 1.876959247648903,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004845521251282223,
|
|
"loss": 5.8575,
|
|
"mean_token_accuracy": 0.1556412935256958,
|
|
"num_tokens": 4210744.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.9759069919586185,
|
|
"epoch": 1.8808777429467085,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004844424068413789,
|
|
"loss": 5.7029,
|
|
"mean_token_accuracy": 0.1579484984278679,
|
|
"num_tokens": 4219321.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.121573209762573,
|
|
"epoch": 1.884796238244514,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048433231424699504,
|
|
"loss": 5.7936,
|
|
"mean_token_accuracy": 0.15319347530603408,
|
|
"num_tokens": 4227785.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.047457981109619,
|
|
"epoch": 1.8887147335423198,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048422184754182384,
|
|
"loss": 5.7001,
|
|
"mean_token_accuracy": 0.15611863061785697,
|
|
"num_tokens": 4235647.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.9391303062438965,
|
|
"epoch": 1.8926332288401255,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004841110069232875,
|
|
"loss": 5.6606,
|
|
"mean_token_accuracy": 0.15901823192834855,
|
|
"num_tokens": 4244267.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.851119041442871,
|
|
"epoch": 1.896551724137931,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048399979258947597,
|
|
"loss": 5.6021,
|
|
"mean_token_accuracy": 0.15520051568746568,
|
|
"num_tokens": 4252440.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.035585689544678,
|
|
"epoch": 1.9004702194357366,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004838882047391474,
|
|
"loss": 5.79,
|
|
"mean_token_accuracy": 0.1520266965031624,
|
|
"num_tokens": 4260459.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.93746976852417,
|
|
"epoch": 1.9043887147335423,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048377624357172724,
|
|
"loss": 5.6757,
|
|
"mean_token_accuracy": 0.15812840983271598,
|
|
"num_tokens": 4269372.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.957759714126587,
|
|
"epoch": 1.908307210031348,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048366390928730843,
|
|
"loss": 5.6772,
|
|
"mean_token_accuracy": 0.15043582171201705,
|
|
"num_tokens": 4279202.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.082476806640625,
|
|
"epoch": 1.9122257053291536,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004835512020866504,
|
|
"loss": 5.7888,
|
|
"mean_token_accuracy": 0.14807590395212172,
|
|
"num_tokens": 4288897.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.999314832687378,
|
|
"epoch": 1.9161442006269591,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048343812217117925,
|
|
"loss": 5.8419,
|
|
"mean_token_accuracy": 0.14902509450912477,
|
|
"num_tokens": 4297983.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.990492725372315,
|
|
"epoch": 1.9200626959247649,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048332466974298723,
|
|
"loss": 5.7552,
|
|
"mean_token_accuracy": 0.15604517757892608,
|
|
"num_tokens": 4306994.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.959328031539917,
|
|
"epoch": 1.9239811912225706,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048321084500483203,
|
|
"loss": 5.6863,
|
|
"mean_token_accuracy": 0.15593681558966638,
|
|
"num_tokens": 4315024.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.061447954177856,
|
|
"epoch": 1.9278996865203761,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000483096648160137,
|
|
"loss": 5.8525,
|
|
"mean_token_accuracy": 0.14402531385421752,
|
|
"num_tokens": 4324051.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.018288660049438,
|
|
"epoch": 1.9318181818181817,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048298207941299047,
|
|
"loss": 5.6597,
|
|
"mean_token_accuracy": 0.1543619751930237,
|
|
"num_tokens": 4332423.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.98842716217041,
|
|
"epoch": 1.9357366771159876,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048286713896814536,
|
|
"loss": 5.7493,
|
|
"mean_token_accuracy": 0.15125339180231095,
|
|
"num_tokens": 4341463.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.981415510177612,
|
|
"epoch": 1.9396551724137931,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048275182703101877,
|
|
"loss": 5.7533,
|
|
"mean_token_accuracy": 0.1530711308121681,
|
|
"num_tokens": 4349091.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.038469505310059,
|
|
"epoch": 1.9435736677115987,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048263614380769193,
|
|
"loss": 5.7897,
|
|
"mean_token_accuracy": 0.14574714750051498,
|
|
"num_tokens": 4358441.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.075689125061035,
|
|
"epoch": 1.9474921630094044,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048252008950490957,
|
|
"loss": 5.7948,
|
|
"mean_token_accuracy": 0.14674443155527114,
|
|
"num_tokens": 4367604.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.043747854232788,
|
|
"epoch": 1.9514106583072102,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048240366433007935,
|
|
"loss": 5.8286,
|
|
"mean_token_accuracy": 0.15071199908852578,
|
|
"num_tokens": 4376890.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.0984001636505125,
|
|
"epoch": 1.9553291536050157,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048228686849127213,
|
|
"loss": 5.9298,
|
|
"mean_token_accuracy": 0.14237690046429635,
|
|
"num_tokens": 4386134.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.058466148376465,
|
|
"epoch": 1.9592476489028212,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004821697021972209,
|
|
"loss": 5.5861,
|
|
"mean_token_accuracy": 0.16591546237468718,
|
|
"num_tokens": 4393856.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 1.9592476489028212,
|
|
"eval_entropy": 5.817647697389588,
|
|
"eval_loss": 5.903265953063965,
|
|
"eval_mean_token_accuracy": 0.14979963585763245,
|
|
"eval_num_tokens": 4393856.0,
|
|
"eval_runtime": 2.1808,
|
|
"eval_samples_per_second": 1890.175,
|
|
"eval_steps_per_second": 236.616,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.973231172561645,
|
|
"epoch": 1.963166144200627,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 5.723,
|
|
"mean_token_accuracy": 0.15404767394065857,
|
|
"num_tokens": 4402404.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.973379373550415,
|
|
"epoch": 1.9670846394984327,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004819342590816288,
|
|
"loss": 5.8137,
|
|
"mean_token_accuracy": 0.1442849263548851,
|
|
"num_tokens": 4411414.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.993176794052124,
|
|
"epoch": 1.9710031347962382,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004818159826808631,
|
|
"loss": 5.681,
|
|
"mean_token_accuracy": 0.14834492802619934,
|
|
"num_tokens": 4420497.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.028544998168945,
|
|
"epoch": 1.9749216300940438,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004816973366664026,
|
|
"loss": 5.7726,
|
|
"mean_token_accuracy": 0.15435664355754852,
|
|
"num_tokens": 4428535.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.979988479614258,
|
|
"epoch": 1.9788401253918495,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004815783212502871,
|
|
"loss": 5.7199,
|
|
"mean_token_accuracy": 0.15469271391630174,
|
|
"num_tokens": 4437282.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.954149484634399,
|
|
"epoch": 1.9827586206896552,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048145893664521645,
|
|
"loss": 5.5252,
|
|
"mean_token_accuracy": 0.16599334478378297,
|
|
"num_tokens": 4445656.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.931991004943848,
|
|
"epoch": 1.9866771159874608,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048133918306455023,
|
|
"loss": 5.6785,
|
|
"mean_token_accuracy": 0.15485763922333717,
|
|
"num_tokens": 4454760.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.9231061935424805,
|
|
"epoch": 1.9905956112852663,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004812190607223075,
|
|
"loss": 5.7537,
|
|
"mean_token_accuracy": 0.14862551391124726,
|
|
"num_tokens": 4464331.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.954615259170533,
|
|
"epoch": 1.9945141065830723,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048109856983316655,
|
|
"loss": 5.7251,
|
|
"mean_token_accuracy": 0.15735211968421936,
|
|
"num_tokens": 4472940.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.000089120864868,
|
|
"epoch": 1.9984326018808778,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000480977710612464,
|
|
"loss": 5.709,
|
|
"mean_token_accuracy": 0.1558776617050171,
|
|
"num_tokens": 4481491.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.000496768951416,
|
|
"epoch": 2.0023510971786833,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004808564832761948,
|
|
"loss": 5.5871,
|
|
"mean_token_accuracy": 0.15766474604606628,
|
|
"num_tokens": 4490032.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.862171173095703,
|
|
"epoch": 2.006269592476489,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004807348880410119,
|
|
"loss": 5.2675,
|
|
"mean_token_accuracy": 0.17061145901679992,
|
|
"num_tokens": 4498358.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.9291160106658936,
|
|
"epoch": 2.010188087774295,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004806129251242258,
|
|
"loss": 5.4568,
|
|
"mean_token_accuracy": 0.16750640720129012,
|
|
"num_tokens": 4506922.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.991434717178345,
|
|
"epoch": 2.0141065830721003,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048049059474380393,
|
|
"loss": 5.4586,
|
|
"mean_token_accuracy": 0.15871975123882293,
|
|
"num_tokens": 4515587.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.8735860824584964,
|
|
"epoch": 2.018025078369906,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048036789711837047,
|
|
"loss": 5.4209,
|
|
"mean_token_accuracy": 0.1666983112692833,
|
|
"num_tokens": 4523872.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.06866455078125,
|
|
"epoch": 2.0219435736677114,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048024483246720607,
|
|
"loss": 5.508,
|
|
"mean_token_accuracy": 0.15984941869974137,
|
|
"num_tokens": 4532656.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.883355331420899,
|
|
"epoch": 2.0258620689655173,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004801214010102472,
|
|
"loss": 5.4386,
|
|
"mean_token_accuracy": 0.1628689780831337,
|
|
"num_tokens": 4540968.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.9069171905517575,
|
|
"epoch": 2.029780564263323,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004799976029680858,
|
|
"loss": 5.4284,
|
|
"mean_token_accuracy": 0.1674864962697029,
|
|
"num_tokens": 4550383.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.93887996673584,
|
|
"epoch": 2.0336990595611284,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004798734385619691,
|
|
"loss": 5.3608,
|
|
"mean_token_accuracy": 0.168947172164917,
|
|
"num_tokens": 4558757.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.91796760559082,
|
|
"epoch": 2.0376175548589344,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000479748908013799,
|
|
"loss": 5.3984,
|
|
"mean_token_accuracy": 0.1602243572473526,
|
|
"num_tokens": 4568342.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.93248610496521,
|
|
"epoch": 2.04153605015674,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004796240115461319,
|
|
"loss": 5.4216,
|
|
"mean_token_accuracy": 0.1644732028245926,
|
|
"num_tokens": 4578036.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.800256395339966,
|
|
"epoch": 2.0454545454545454,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004794987493821779,
|
|
"loss": 5.3672,
|
|
"mean_token_accuracy": 0.164408877491951,
|
|
"num_tokens": 4586709.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.8819561958312985,
|
|
"epoch": 2.049373040752351,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047937312174580084,
|
|
"loss": 5.3587,
|
|
"mean_token_accuracy": 0.17530310153961182,
|
|
"num_tokens": 4595415.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.8860251903533936,
|
|
"epoch": 2.053291536050157,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004792471288615177,
|
|
"loss": 5.3996,
|
|
"mean_token_accuracy": 0.1687674805521965,
|
|
"num_tokens": 4603908.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.84918360710144,
|
|
"epoch": 2.0572100313479624,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004791207709544981,
|
|
"loss": 5.4487,
|
|
"mean_token_accuracy": 0.15825537592172623,
|
|
"num_tokens": 4613633.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.888375091552734,
|
|
"epoch": 2.061128526645768,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047899404825056424,
|
|
"loss": 5.4076,
|
|
"mean_token_accuracy": 0.15994386672973632,
|
|
"num_tokens": 4622812.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.869131326675415,
|
|
"epoch": 2.0650470219435735,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004788669609761901,
|
|
"loss": 5.431,
|
|
"mean_token_accuracy": 0.1673545002937317,
|
|
"num_tokens": 4631169.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.937556743621826,
|
|
"epoch": 2.0689655172413794,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047873950935850107,
|
|
"loss": 5.4496,
|
|
"mean_token_accuracy": 0.16368993669748305,
|
|
"num_tokens": 4639703.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.895373058319092,
|
|
"epoch": 2.072884012539185,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004786116936252742,
|
|
"loss": 5.3794,
|
|
"mean_token_accuracy": 0.16742658764123916,
|
|
"num_tokens": 4648484.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.880811452865601,
|
|
"epoch": 2.0768025078369905,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004784835140049367,
|
|
"loss": 5.4695,
|
|
"mean_token_accuracy": 0.15584376752376555,
|
|
"num_tokens": 4657430.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.933412265777588,
|
|
"epoch": 2.080721003134796,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004783549707265663,
|
|
"loss": 5.4342,
|
|
"mean_token_accuracy": 0.16683154851198195,
|
|
"num_tokens": 4666100.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.88412880897522,
|
|
"epoch": 2.084639498432602,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047822606401989084,
|
|
"loss": 5.4242,
|
|
"mean_token_accuracy": 0.1708218514919281,
|
|
"num_tokens": 4674046.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.889084053039551,
|
|
"epoch": 2.0885579937304075,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004780967941152873,
|
|
"loss": 5.3559,
|
|
"mean_token_accuracy": 0.1766132801771164,
|
|
"num_tokens": 4682580.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.776461362838745,
|
|
"epoch": 2.092476489028213,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004779671612437822,
|
|
"loss": 5.2801,
|
|
"mean_token_accuracy": 0.17160771787166595,
|
|
"num_tokens": 4690615.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.8601878643035885,
|
|
"epoch": 2.096394984326019,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047783716563705035,
|
|
"loss": 5.4132,
|
|
"mean_token_accuracy": 0.16394216269254686,
|
|
"num_tokens": 4699536.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.995702123641967,
|
|
"epoch": 2.1003134796238245,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000477706807527415,
|
|
"loss": 5.5261,
|
|
"mean_token_accuracy": 0.16179091334342957,
|
|
"num_tokens": 4708733.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.84748215675354,
|
|
"epoch": 2.10423197492163,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004775760871478472,
|
|
"loss": 5.3711,
|
|
"mean_token_accuracy": 0.1728486657142639,
|
|
"num_tokens": 4716925.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.936871767044067,
|
|
"epoch": 2.1081504702194356,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047744500473196564,
|
|
"loss": 5.4667,
|
|
"mean_token_accuracy": 0.16030996888875962,
|
|
"num_tokens": 4725817.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.90448637008667,
|
|
"epoch": 2.1120689655172415,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047731356051403556,
|
|
"loss": 5.4625,
|
|
"mean_token_accuracy": 0.1681454673409462,
|
|
"num_tokens": 4734760.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.812362432479858,
|
|
"epoch": 2.115987460815047,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004771817547289693,
|
|
"loss": 5.4245,
|
|
"mean_token_accuracy": 0.1691887065768242,
|
|
"num_tokens": 4742704.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.856250953674317,
|
|
"epoch": 2.1199059561128526,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004770495876123251,
|
|
"loss": 5.3528,
|
|
"mean_token_accuracy": 0.17168572694063186,
|
|
"num_tokens": 4751265.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.859165382385254,
|
|
"epoch": 2.123824451410658,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004769170594003071,
|
|
"loss": 5.3857,
|
|
"mean_token_accuracy": 0.173263618350029,
|
|
"num_tokens": 4760449.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.839176893234253,
|
|
"epoch": 2.127742946708464,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047678417032976457,
|
|
"loss": 5.4475,
|
|
"mean_token_accuracy": 0.16216098815202712,
|
|
"num_tokens": 4768735.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.764931535720825,
|
|
"epoch": 2.1316614420062696,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004766509206381919,
|
|
"loss": 5.2927,
|
|
"mean_token_accuracy": 0.1692401021718979,
|
|
"num_tokens": 4777644.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.870836210250855,
|
|
"epoch": 2.135579937304075,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004765173105637279,
|
|
"loss": 5.4385,
|
|
"mean_token_accuracy": 0.16707490384578705,
|
|
"num_tokens": 4786797.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.811993217468261,
|
|
"epoch": 2.139498432601881,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047638334034515547,
|
|
"loss": 5.3589,
|
|
"mean_token_accuracy": 0.1632653370499611,
|
|
"num_tokens": 4795254.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.879141044616699,
|
|
"epoch": 2.1434169278996866,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047624901022190106,
|
|
"loss": 5.4118,
|
|
"mean_token_accuracy": 0.16706012040376664,
|
|
"num_tokens": 4804101.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.855080270767212,
|
|
"epoch": 2.147335423197492,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.4433,
|
|
"mean_token_accuracy": 0.15685019344091417,
|
|
"num_tokens": 4813395.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.851852655410767,
|
|
"epoch": 2.1512539184952977,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004759792712222679,
|
|
"loss": 5.3683,
|
|
"mean_token_accuracy": 0.16672899425029755,
|
|
"num_tokens": 4822314.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.8839493751525875,
|
|
"epoch": 2.1551724137931036,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004758438628279565,
|
|
"loss": 5.4678,
|
|
"mean_token_accuracy": 0.1641886129975319,
|
|
"num_tokens": 4831131.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.912335395812988,
|
|
"epoch": 2.159090909090909,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047570809549309697,
|
|
"loss": 5.4765,
|
|
"mean_token_accuracy": 0.15812892019748687,
|
|
"num_tokens": 4839378.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.752632427215576,
|
|
"epoch": 2.1630094043887147,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004755719694603275,
|
|
"loss": 5.3884,
|
|
"mean_token_accuracy": 0.16313555985689163,
|
|
"num_tokens": 4849084.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.881686162948609,
|
|
"epoch": 2.16692789968652,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004754354849729274,
|
|
"loss": 5.4025,
|
|
"mean_token_accuracy": 0.16439216434955597,
|
|
"num_tokens": 4858481.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.843069839477539,
|
|
"epoch": 2.170846394984326,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047529864227481653,
|
|
"loss": 5.391,
|
|
"mean_token_accuracy": 0.1695776104927063,
|
|
"num_tokens": 4866981.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.7651002407073975,
|
|
"epoch": 2.1747648902821317,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000475161441610555,
|
|
"loss": 5.373,
|
|
"mean_token_accuracy": 0.17173573970794678,
|
|
"num_tokens": 4875914.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.780478191375733,
|
|
"epoch": 2.1786833855799372,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004750238832253427,
|
|
"loss": 5.3215,
|
|
"mean_token_accuracy": 0.17755378782749176,
|
|
"num_tokens": 4884362.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.829852724075318,
|
|
"epoch": 2.1826018808777428,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004748859673650187,
|
|
"loss": 5.448,
|
|
"mean_token_accuracy": 0.1657805249094963,
|
|
"num_tokens": 4892663.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.819691467285156,
|
|
"epoch": 2.1865203761755487,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047474769427606115,
|
|
"loss": 5.3466,
|
|
"mean_token_accuracy": 0.171503509581089,
|
|
"num_tokens": 4901413.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.757342720031739,
|
|
"epoch": 2.1904388714733543,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004746090642055863,
|
|
"loss": 5.3749,
|
|
"mean_token_accuracy": 0.17233139276504517,
|
|
"num_tokens": 4910601.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.797521448135376,
|
|
"epoch": 2.19435736677116,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047447007740134857,
|
|
"loss": 5.3373,
|
|
"mean_token_accuracy": 0.17241542637348176,
|
|
"num_tokens": 4919446.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.771145629882812,
|
|
"epoch": 2.1982758620689653,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047433073411174,
|
|
"loss": 5.333,
|
|
"mean_token_accuracy": 0.17256265729665757,
|
|
"num_tokens": 4928911.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.779217624664307,
|
|
"epoch": 2.2021943573667713,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004741910345857896,
|
|
"loss": 5.3792,
|
|
"mean_token_accuracy": 0.17059777230024337,
|
|
"num_tokens": 4936934.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.782461404800415,
|
|
"epoch": 2.206112852664577,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047405097907316315,
|
|
"loss": 5.3435,
|
|
"mean_token_accuracy": 0.1692051559686661,
|
|
"num_tokens": 4945243.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.81570463180542,
|
|
"epoch": 2.2100313479623823,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004739105678241625,
|
|
"loss": 5.4276,
|
|
"mean_token_accuracy": 0.16294209510087967,
|
|
"num_tokens": 4954848.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.8940118789672855,
|
|
"epoch": 2.2139498432601883,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004737698010897253,
|
|
"loss": 5.4653,
|
|
"mean_token_accuracy": 0.15729624778032303,
|
|
"num_tokens": 4964401.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.7935137271881105,
|
|
"epoch": 2.217868338557994,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004736286791214245,
|
|
"loss": 5.3733,
|
|
"mean_token_accuracy": 0.1689634695649147,
|
|
"num_tokens": 4972886.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.736187601089478,
|
|
"epoch": 2.2217868338557993,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047348720217146807,
|
|
"loss": 5.424,
|
|
"mean_token_accuracy": 0.1610773101449013,
|
|
"num_tokens": 4982476.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.993754577636719,
|
|
"epoch": 2.225705329153605,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047334537049269806,
|
|
"loss": 5.5357,
|
|
"mean_token_accuracy": 0.1589807778596878,
|
|
"num_tokens": 4991121.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.755010509490967,
|
|
"epoch": 2.229623824451411,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004732031843385909,
|
|
"loss": 5.417,
|
|
"mean_token_accuracy": 0.1702817499637604,
|
|
"num_tokens": 4999602.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.838142681121826,
|
|
"epoch": 2.2335423197492164,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004730606439632562,
|
|
"loss": 5.4393,
|
|
"mean_token_accuracy": 0.1695335566997528,
|
|
"num_tokens": 5008461.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.981766223907471,
|
|
"epoch": 2.237460815047022,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004729177496214367,
|
|
"loss": 5.5342,
|
|
"mean_token_accuracy": 0.15868925750255586,
|
|
"num_tokens": 5017796.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.896777772903443,
|
|
"epoch": 2.2413793103448274,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047277450156850767,
|
|
"loss": 5.4714,
|
|
"mean_token_accuracy": 0.1610400855541229,
|
|
"num_tokens": 5026735.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.778905868530273,
|
|
"epoch": 2.2452978056426334,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004726309000604768,
|
|
"loss": 5.3601,
|
|
"mean_token_accuracy": 0.17411165684461594,
|
|
"num_tokens": 5035254.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.722349405288696,
|
|
"epoch": 2.249216300940439,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004724869453539832,
|
|
"loss": 5.3645,
|
|
"mean_token_accuracy": 0.17299142032861708,
|
|
"num_tokens": 5044117.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.88076434135437,
|
|
"epoch": 2.2531347962382444,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004723426377062972,
|
|
"loss": 5.3983,
|
|
"mean_token_accuracy": 0.16141462028026582,
|
|
"num_tokens": 5052450.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.870799589157104,
|
|
"epoch": 2.2570532915360504,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047219797737532,
|
|
"loss": 5.4363,
|
|
"mean_token_accuracy": 0.16325958967208862,
|
|
"num_tokens": 5061079.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.826492977142334,
|
|
"epoch": 2.260971786833856,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047205296461958314,
|
|
"loss": 5.462,
|
|
"mean_token_accuracy": 0.1586606025695801,
|
|
"num_tokens": 5070140.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.7866026878356935,
|
|
"epoch": 2.2648902821316614,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047190759969824785,
|
|
"loss": 5.424,
|
|
"mean_token_accuracy": 0.16685089766979216,
|
|
"num_tokens": 5078675.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.944733142852783,
|
|
"epoch": 2.268808777429467,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047176188287110485,
|
|
"loss": 5.4863,
|
|
"mean_token_accuracy": 0.16021176874637605,
|
|
"num_tokens": 5087470.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.808992195129394,
|
|
"epoch": 2.2727272727272725,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004716158143985737,
|
|
"loss": 5.4569,
|
|
"mean_token_accuracy": 0.1652986466884613,
|
|
"num_tokens": 5095981.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.713580560684204,
|
|
"epoch": 2.2766457680250785,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047146939454170245,
|
|
"loss": 5.3652,
|
|
"mean_token_accuracy": 0.17279119193553924,
|
|
"num_tokens": 5104913.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.862746810913086,
|
|
"epoch": 2.280564263322884,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004713226235621672,
|
|
"loss": 5.4119,
|
|
"mean_token_accuracy": 0.16582923233509064,
|
|
"num_tokens": 5113834.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.894469547271728,
|
|
"epoch": 2.2844827586206895,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004711755017222714,
|
|
"loss": 5.4924,
|
|
"mean_token_accuracy": 0.1580079674720764,
|
|
"num_tokens": 5122676.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.7889503002166744,
|
|
"epoch": 2.2884012539184955,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047102802928494563,
|
|
"loss": 5.4623,
|
|
"mean_token_accuracy": 0.16928494423627855,
|
|
"num_tokens": 5132033.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.820769596099853,
|
|
"epoch": 2.292319749216301,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004708802065137471,
|
|
"loss": 5.4191,
|
|
"mean_token_accuracy": 0.16982818096876146,
|
|
"num_tokens": 5140305.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.847978830337524,
|
|
"epoch": 2.2962382445141065,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004707320336728591,
|
|
"loss": 5.4536,
|
|
"mean_token_accuracy": 0.1647203728556633,
|
|
"num_tokens": 5148857.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.817395162582398,
|
|
"epoch": 2.300156739811912,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004705835110270904,
|
|
"loss": 5.4159,
|
|
"mean_token_accuracy": 0.16635480672121047,
|
|
"num_tokens": 5157214.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.890399837493897,
|
|
"epoch": 2.304075235109718,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047043463884187517,
|
|
"loss": 5.4794,
|
|
"mean_token_accuracy": 0.1637230858206749,
|
|
"num_tokens": 5166346.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.77665867805481,
|
|
"epoch": 2.3079937304075235,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047028541738327207,
|
|
"loss": 5.3871,
|
|
"mean_token_accuracy": 0.16928445249795915,
|
|
"num_tokens": 5176456.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.8899280548095705,
|
|
"epoch": 2.311912225705329,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004701358469179641,
|
|
"loss": 5.5038,
|
|
"mean_token_accuracy": 0.16095911264419555,
|
|
"num_tokens": 5185804.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.845748662948608,
|
|
"epoch": 2.3158307210031346,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004699859277132578,
|
|
"loss": 5.4517,
|
|
"mean_token_accuracy": 0.16641164124011992,
|
|
"num_tokens": 5194714.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.78071346282959,
|
|
"epoch": 2.3197492163009406,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046983566003708336,
|
|
"loss": 5.4,
|
|
"mean_token_accuracy": 0.1638484075665474,
|
|
"num_tokens": 5203667.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.817977619171143,
|
|
"epoch": 2.323667711598746,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046968504415799325,
|
|
"loss": 5.4332,
|
|
"mean_token_accuracy": 0.16527473628520967,
|
|
"num_tokens": 5211934.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.860213375091552,
|
|
"epoch": 2.3275862068965516,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004695340803451625,
|
|
"loss": 5.488,
|
|
"mean_token_accuracy": 0.16174516528844834,
|
|
"num_tokens": 5221321.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.860506677627564,
|
|
"epoch": 2.3315047021943576,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 5.3469,
|
|
"mean_token_accuracy": 0.17110743820667268,
|
|
"num_tokens": 5230905.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.785662651062012,
|
|
"epoch": 2.335423197492163,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004692311099980878,
|
|
"loss": 5.4401,
|
|
"mean_token_accuracy": 0.16616067737340928,
|
|
"num_tokens": 5240137.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.79237117767334,
|
|
"epoch": 2.3393416927899686,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046907910400530097,
|
|
"loss": 5.3761,
|
|
"mean_token_accuracy": 0.16875567734241487,
|
|
"num_tokens": 5248603.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.767337226867676,
|
|
"epoch": 2.343260188087774,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004689267511616868,
|
|
"loss": 5.4916,
|
|
"mean_token_accuracy": 0.16034385114908217,
|
|
"num_tokens": 5257984.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.8946727275848385,
|
|
"epoch": 2.34717868338558,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046877405173952465,
|
|
"loss": 5.4898,
|
|
"mean_token_accuracy": 0.16577527076005935,
|
|
"num_tokens": 5266160.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.814199829101563,
|
|
"epoch": 2.3510971786833856,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000468621006011713,
|
|
"loss": 5.3751,
|
|
"mean_token_accuracy": 0.1714093044400215,
|
|
"num_tokens": 5273874.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.3510971786833856,
|
|
"eval_entropy": 5.596698675968851,
|
|
"eval_loss": 5.845335483551025,
|
|
"eval_mean_token_accuracy": 0.154815596290106,
|
|
"eval_num_tokens": 5273874.0,
|
|
"eval_runtime": 2.3806,
|
|
"eval_samples_per_second": 1731.47,
|
|
"eval_steps_per_second": 216.749,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.845364332199097,
|
|
"epoch": 2.355015673981191,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046846761425176943,
|
|
"loss": 5.4286,
|
|
"mean_token_accuracy": 0.15899102091789247,
|
|
"num_tokens": 5282276.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.850484085083008,
|
|
"epoch": 2.3589341692789967,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004683138767338299,
|
|
"loss": 5.4203,
|
|
"mean_token_accuracy": 0.16298952996730803,
|
|
"num_tokens": 5291463.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.818532276153564,
|
|
"epoch": 2.3628526645768027,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004681597937326483,
|
|
"loss": 5.4563,
|
|
"mean_token_accuracy": 0.1673190027475357,
|
|
"num_tokens": 5300383.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.687733316421509,
|
|
"epoch": 2.366771159874608,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004680053655235959,
|
|
"loss": 5.3584,
|
|
"mean_token_accuracy": 0.1687045142054558,
|
|
"num_tokens": 5309449.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.880924320220947,
|
|
"epoch": 2.3706896551724137,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004678505923826609,
|
|
"loss": 5.3755,
|
|
"mean_token_accuracy": 0.1722179666161537,
|
|
"num_tokens": 5318139.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.720145750045776,
|
|
"epoch": 2.3746081504702197,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046769547458644817,
|
|
"loss": 5.3279,
|
|
"mean_token_accuracy": 0.16696547716856003,
|
|
"num_tokens": 5326364.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.793128156661988,
|
|
"epoch": 2.378526645768025,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004675400124121782,
|
|
"loss": 5.4491,
|
|
"mean_token_accuracy": 0.1665338709950447,
|
|
"num_tokens": 5335734.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.894374513626099,
|
|
"epoch": 2.3824451410658307,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046738420613768716,
|
|
"loss": 5.3539,
|
|
"mean_token_accuracy": 0.1639840692281723,
|
|
"num_tokens": 5343772.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.694606924057007,
|
|
"epoch": 2.3863636363636362,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046722805604142614,
|
|
"loss": 5.3911,
|
|
"mean_token_accuracy": 0.1658248096704483,
|
|
"num_tokens": 5352012.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.8409017562866214,
|
|
"epoch": 2.3902821316614418,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046707156240246076,
|
|
"loss": 5.4778,
|
|
"mean_token_accuracy": 0.16721124500036239,
|
|
"num_tokens": 5360814.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.862252283096313,
|
|
"epoch": 2.3942006269592477,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046691472550047027,
|
|
"loss": 5.4493,
|
|
"mean_token_accuracy": 0.16908241659402848,
|
|
"num_tokens": 5369958.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.786740255355835,
|
|
"epoch": 2.3981191222570533,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046675754561574783,
|
|
"loss": 5.4631,
|
|
"mean_token_accuracy": 0.16514671146869658,
|
|
"num_tokens": 5379099.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.7936474800109865,
|
|
"epoch": 2.402037617554859,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046660002302919933,
|
|
"loss": 5.2857,
|
|
"mean_token_accuracy": 0.16601007878780366,
|
|
"num_tokens": 5387713.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.796220207214356,
|
|
"epoch": 2.4059561128526648,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004664421580223433,
|
|
"loss": 5.473,
|
|
"mean_token_accuracy": 0.16711254715919494,
|
|
"num_tokens": 5396730.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.690844106674194,
|
|
"epoch": 2.4098746081504703,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046628395087730995,
|
|
"loss": 5.3478,
|
|
"mean_token_accuracy": 0.1701469287276268,
|
|
"num_tokens": 5405378.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.7996903419494625,
|
|
"epoch": 2.413793103448276,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004661254018768411,
|
|
"loss": 5.4444,
|
|
"mean_token_accuracy": 0.17136418521404267,
|
|
"num_tokens": 5413748.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.825985956192016,
|
|
"epoch": 2.4177115987460813,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004659665113042897,
|
|
"loss": 5.436,
|
|
"mean_token_accuracy": 0.16724079102277756,
|
|
"num_tokens": 5423409.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.765198278427124,
|
|
"epoch": 2.4216300940438873,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004658072794436187,
|
|
"loss": 5.4728,
|
|
"mean_token_accuracy": 0.17427219599485397,
|
|
"num_tokens": 5432562.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.83167519569397,
|
|
"epoch": 2.425548589341693,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046564770657940146,
|
|
"loss": 5.4246,
|
|
"mean_token_accuracy": 0.17290456742048263,
|
|
"num_tokens": 5441100.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.7174262523651125,
|
|
"epoch": 2.4294670846394983,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004654877929968205,
|
|
"loss": 5.3752,
|
|
"mean_token_accuracy": 0.16532681584358216,
|
|
"num_tokens": 5450481.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.817539548873901,
|
|
"epoch": 2.433385579937304,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004653275389816673,
|
|
"loss": 5.3674,
|
|
"mean_token_accuracy": 0.1712379351258278,
|
|
"num_tokens": 5458797.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.7961509227752686,
|
|
"epoch": 2.43730407523511,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046516694482034174,
|
|
"loss": 5.3737,
|
|
"mean_token_accuracy": 0.17615118622779846,
|
|
"num_tokens": 5467031.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.796831464767456,
|
|
"epoch": 2.4412225705329154,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046500601079985164,
|
|
"loss": 5.4167,
|
|
"mean_token_accuracy": 0.17120206356048584,
|
|
"num_tokens": 5475549.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.822018098831177,
|
|
"epoch": 2.445141065830721,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004648447372078123,
|
|
"loss": 5.4387,
|
|
"mean_token_accuracy": 0.16877201646566392,
|
|
"num_tokens": 5483837.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.817354822158814,
|
|
"epoch": 2.449059561128527,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004646831243324457,
|
|
"loss": 5.5056,
|
|
"mean_token_accuracy": 0.1648485168814659,
|
|
"num_tokens": 5493522.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.785455560684204,
|
|
"epoch": 2.4529780564263324,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004645211724625802,
|
|
"loss": 5.333,
|
|
"mean_token_accuracy": 0.17399988919496537,
|
|
"num_tokens": 5502504.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.7855524063110355,
|
|
"epoch": 2.456896551724138,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046435888188765015,
|
|
"loss": 5.4151,
|
|
"mean_token_accuracy": 0.16636843979358673,
|
|
"num_tokens": 5511568.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.748692893981934,
|
|
"epoch": 2.4608150470219434,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004641962528976951,
|
|
"loss": 5.3589,
|
|
"mean_token_accuracy": 0.17583060264587402,
|
|
"num_tokens": 5519899.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.8178221702575685,
|
|
"epoch": 2.4647335423197494,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004640332857833593,
|
|
"loss": 5.4246,
|
|
"mean_token_accuracy": 0.16766543835401534,
|
|
"num_tokens": 5528788.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.766494655609131,
|
|
"epoch": 2.468652037617555,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046386998083589156,
|
|
"loss": 5.4033,
|
|
"mean_token_accuracy": 0.17223501950502396,
|
|
"num_tokens": 5537046.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.718023204803467,
|
|
"epoch": 2.4725705329153604,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004637063383471442,
|
|
"loss": 5.4754,
|
|
"mean_token_accuracy": 0.16232676804065704,
|
|
"num_tokens": 5546363.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.795729780197144,
|
|
"epoch": 2.476489028213166,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046354235860957287,
|
|
"loss": 5.4016,
|
|
"mean_token_accuracy": 0.1670736402273178,
|
|
"num_tokens": 5555871.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.7601439476013185,
|
|
"epoch": 2.480407523510972,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004633780419162361,
|
|
"loss": 5.3489,
|
|
"mean_token_accuracy": 0.17568435966968537,
|
|
"num_tokens": 5564701.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.764534664154053,
|
|
"epoch": 2.4843260188087775,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046321338856079435,
|
|
"loss": 5.3802,
|
|
"mean_token_accuracy": 0.1663267880678177,
|
|
"num_tokens": 5573841.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.740782022476196,
|
|
"epoch": 2.488244514106583,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046304839883750987,
|
|
"loss": 5.3828,
|
|
"mean_token_accuracy": 0.1657790407538414,
|
|
"num_tokens": 5582352.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.776612234115601,
|
|
"epoch": 2.492163009404389,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000462883073041246,
|
|
"loss": 5.5099,
|
|
"mean_token_accuracy": 0.15922176986932754,
|
|
"num_tokens": 5591785.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.940705394744873,
|
|
"epoch": 2.4960815047021945,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004627174114674669,
|
|
"loss": 5.3928,
|
|
"mean_token_accuracy": 0.17555987685918809,
|
|
"num_tokens": 5600752.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.770922994613647,
|
|
"epoch": 2.5,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004625514144122365,
|
|
"loss": 5.4297,
|
|
"mean_token_accuracy": 0.16724998950958253,
|
|
"num_tokens": 5609078.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.632586431503296,
|
|
"epoch": 2.5039184952978055,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004623850821722185,
|
|
"loss": 5.3882,
|
|
"mean_token_accuracy": 0.16894710808992386,
|
|
"num_tokens": 5618077.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.792569351196289,
|
|
"epoch": 2.507836990595611,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004622184150446756,
|
|
"loss": 5.4608,
|
|
"mean_token_accuracy": 0.17146307080984116,
|
|
"num_tokens": 5626281.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.672460985183716,
|
|
"epoch": 2.511755485893417,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046205141332746904,
|
|
"loss": 5.387,
|
|
"mean_token_accuracy": 0.1665313258767128,
|
|
"num_tokens": 5634518.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.730130767822265,
|
|
"epoch": 2.5156739811912225,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 5.4276,
|
|
"mean_token_accuracy": 0.16683324575424194,
|
|
"num_tokens": 5643439.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.735629606246948,
|
|
"epoch": 2.519592476489028,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004617164073184987,
|
|
"loss": 5.3301,
|
|
"mean_token_accuracy": 0.1753636807203293,
|
|
"num_tokens": 5652385.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.695278882980347,
|
|
"epoch": 2.523510971786834,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046154840362544496,
|
|
"loss": 5.3527,
|
|
"mean_token_accuracy": 0.16637712866067886,
|
|
"num_tokens": 5660964.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.764346742630005,
|
|
"epoch": 2.5274294670846396,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004613800665401466,
|
|
"loss": 5.4444,
|
|
"mean_token_accuracy": 0.15888694375753404,
|
|
"num_tokens": 5669445.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.87206997871399,
|
|
"epoch": 2.531347962382445,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004612113963634493,
|
|
"loss": 5.5189,
|
|
"mean_token_accuracy": 0.1621658369898796,
|
|
"num_tokens": 5677954.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.807461738586426,
|
|
"epoch": 2.535266457680251,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004610423933967938,
|
|
"loss": 5.4687,
|
|
"mean_token_accuracy": 0.16734565496444703,
|
|
"num_tokens": 5686679.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.851128816604614,
|
|
"epoch": 2.5391849529780566,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000460873057942216,
|
|
"loss": 5.496,
|
|
"mean_token_accuracy": 0.1628448933362961,
|
|
"num_tokens": 5695840.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.8549315452575685,
|
|
"epoch": 2.543103448275862,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004607033903023458,
|
|
"loss": 5.3754,
|
|
"mean_token_accuracy": 0.1689860388636589,
|
|
"num_tokens": 5703899.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.665745973587036,
|
|
"epoch": 2.5470219435736676,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046053339078040674,
|
|
"loss": 5.3848,
|
|
"mean_token_accuracy": 0.1715059071779251,
|
|
"num_tokens": 5712501.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.8268369197845455,
|
|
"epoch": 2.550940438871473,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004603630596802155,
|
|
"loss": 5.4206,
|
|
"mean_token_accuracy": 0.16608440279960632,
|
|
"num_tokens": 5721933.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.792334985733032,
|
|
"epoch": 2.554858934169279,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004601923973061814,
|
|
"loss": 5.4514,
|
|
"mean_token_accuracy": 0.1655258372426033,
|
|
"num_tokens": 5730579.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.825296068191529,
|
|
"epoch": 2.5587774294670846,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046002140396330575,
|
|
"loss": 5.424,
|
|
"mean_token_accuracy": 0.17014774978160857,
|
|
"num_tokens": 5739212.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.821914100646973,
|
|
"epoch": 2.56269592476489,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045985007995718154,
|
|
"loss": 5.3923,
|
|
"mean_token_accuracy": 0.17102758437395096,
|
|
"num_tokens": 5748111.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.719115734100342,
|
|
"epoch": 2.566614420062696,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004596784255939923,
|
|
"loss": 5.4018,
|
|
"mean_token_accuracy": 0.1675566628575325,
|
|
"num_tokens": 5756332.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.771751403808594,
|
|
"epoch": 2.5705329153605017,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004595064411805123,
|
|
"loss": 5.3893,
|
|
"mean_token_accuracy": 0.17710367888212203,
|
|
"num_tokens": 5765537.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.77226152420044,
|
|
"epoch": 2.574451410658307,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004593341270241057,
|
|
"loss": 5.4104,
|
|
"mean_token_accuracy": 0.16590501368045807,
|
|
"num_tokens": 5773789.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.703116703033447,
|
|
"epoch": 2.5783699059561127,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004591614834327257,
|
|
"loss": 5.4002,
|
|
"mean_token_accuracy": 0.16619465351104737,
|
|
"num_tokens": 5782203.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.794832372665406,
|
|
"epoch": 2.5822884012539182,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045898851071491444,
|
|
"loss": 5.3908,
|
|
"mean_token_accuracy": 0.17682743966579437,
|
|
"num_tokens": 5790517.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.680010509490967,
|
|
"epoch": 2.586206896551724,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004588152091798022,
|
|
"loss": 5.424,
|
|
"mean_token_accuracy": 0.1639699563384056,
|
|
"num_tokens": 5799475.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.814670181274414,
|
|
"epoch": 2.5901253918495297,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004586415791371069,
|
|
"loss": 5.3829,
|
|
"mean_token_accuracy": 0.1711047798395157,
|
|
"num_tokens": 5808521.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.716660022735596,
|
|
"epoch": 2.5940438871473352,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004584676208971336,
|
|
"loss": 5.4119,
|
|
"mean_token_accuracy": 0.17039491087198258,
|
|
"num_tokens": 5817391.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.800877094268799,
|
|
"epoch": 2.597962382445141,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045829333477077384,
|
|
"loss": 5.4422,
|
|
"mean_token_accuracy": 0.16900216192007064,
|
|
"num_tokens": 5826483.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.826243543624878,
|
|
"epoch": 2.6018808777429467,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004581187210695053,
|
|
"loss": 5.3816,
|
|
"mean_token_accuracy": 0.16432885080575943,
|
|
"num_tokens": 5835143.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.691102361679077,
|
|
"epoch": 2.6057993730407523,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000457943780105391,
|
|
"loss": 5.3441,
|
|
"mean_token_accuracy": 0.172003573179245,
|
|
"num_tokens": 5843892.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.794711685180664,
|
|
"epoch": 2.6097178683385582,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045776851219107856,
|
|
"loss": 5.4598,
|
|
"mean_token_accuracy": 0.16774493008852004,
|
|
"num_tokens": 5852711.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.785120820999145,
|
|
"epoch": 2.6136363636363638,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045759291763980035,
|
|
"loss": 5.3174,
|
|
"mean_token_accuracy": 0.18098744601011277,
|
|
"num_tokens": 5860947.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.664022159576416,
|
|
"epoch": 2.6175548589341693,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045741699676537227,
|
|
"loss": 5.3895,
|
|
"mean_token_accuracy": 0.16960566192865373,
|
|
"num_tokens": 5869912.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.741370487213135,
|
|
"epoch": 2.621473354231975,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045724074988219343,
|
|
"loss": 5.4208,
|
|
"mean_token_accuracy": 0.1655040666460991,
|
|
"num_tokens": 5879288.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.8213550567626955,
|
|
"epoch": 2.6253918495297803,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045706417730524565,
|
|
"loss": 5.4348,
|
|
"mean_token_accuracy": 0.16033685207366943,
|
|
"num_tokens": 5888549.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.703435897827148,
|
|
"epoch": 2.6293103448275863,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004568872793500927,
|
|
"loss": 5.3204,
|
|
"mean_token_accuracy": 0.17171522378921508,
|
|
"num_tokens": 5896432.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.733251762390137,
|
|
"epoch": 2.633228840125392,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045671005633287986,
|
|
"loss": 5.4213,
|
|
"mean_token_accuracy": 0.17132176607847213,
|
|
"num_tokens": 5905500.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.806640577316284,
|
|
"epoch": 2.6371473354231973,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004565325085703336,
|
|
"loss": 5.413,
|
|
"mean_token_accuracy": 0.1698663964867592,
|
|
"num_tokens": 5913775.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.7962277889251705,
|
|
"epoch": 2.6410658307210033,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004563546363797602,
|
|
"loss": 5.4355,
|
|
"mean_token_accuracy": 0.1695856586098671,
|
|
"num_tokens": 5923502.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.774279403686523,
|
|
"epoch": 2.644984326018809,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004561764400790465,
|
|
"loss": 5.4175,
|
|
"mean_token_accuracy": 0.16693703830242157,
|
|
"num_tokens": 5932325.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.746313571929932,
|
|
"epoch": 2.6489028213166144,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045599791998665796,
|
|
"loss": 5.3918,
|
|
"mean_token_accuracy": 0.17012834995985032,
|
|
"num_tokens": 5941178.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.800010299682617,
|
|
"epoch": 2.6528213166144203,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004558190764216389,
|
|
"loss": 5.5247,
|
|
"mean_token_accuracy": 0.1658524051308632,
|
|
"num_tokens": 5949402.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.743608713150024,
|
|
"epoch": 2.656739811912226,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004556399097036119,
|
|
"loss": 5.3418,
|
|
"mean_token_accuracy": 0.16783516705036164,
|
|
"num_tokens": 5958259.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.673016929626465,
|
|
"epoch": 2.6606583072100314,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004554604201527768,
|
|
"loss": 5.3999,
|
|
"mean_token_accuracy": 0.16999000757932664,
|
|
"num_tokens": 5967082.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.81972017288208,
|
|
"epoch": 2.664576802507837,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00045528060808991075,
|
|
"loss": 5.3649,
|
|
"mean_token_accuracy": 0.16833469569683074,
|
|
"num_tokens": 5975783.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.713375759124756,
|
|
"epoch": 2.6684952978056424,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004551004738363669,
|
|
"loss": 5.3637,
|
|
"mean_token_accuracy": 0.1715414822101593,
|
|
"num_tokens": 5984413.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.647453737258911,
|
|
"epoch": 2.6724137931034484,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045492001771407434,
|
|
"loss": 5.216,
|
|
"mean_token_accuracy": 0.16858605295419693,
|
|
"num_tokens": 5993336.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.68209662437439,
|
|
"epoch": 2.676332288401254,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004547392400455374,
|
|
"loss": 5.3301,
|
|
"mean_token_accuracy": 0.1709536373615265,
|
|
"num_tokens": 6002010.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.720417070388794,
|
|
"epoch": 2.6802507836990594,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004545581411538353,
|
|
"loss": 5.4164,
|
|
"mean_token_accuracy": 0.16374239027500154,
|
|
"num_tokens": 6010688.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.677560710906983,
|
|
"epoch": 2.6841692789968654,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045437672136262083,
|
|
"loss": 5.3702,
|
|
"mean_token_accuracy": 0.17231615036725997,
|
|
"num_tokens": 6019071.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.778875017166138,
|
|
"epoch": 2.688087774294671,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004541949809961208,
|
|
"loss": 5.4131,
|
|
"mean_token_accuracy": 0.16201838254928588,
|
|
"num_tokens": 6028527.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.767816877365112,
|
|
"epoch": 2.6920062695924765,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004540129203791346,
|
|
"loss": 5.5123,
|
|
"mean_token_accuracy": 0.16237059980630875,
|
|
"num_tokens": 6038165.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.723360013961792,
|
|
"epoch": 2.695924764890282,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045383053983703413,
|
|
"loss": 5.3109,
|
|
"mean_token_accuracy": 0.17263842821121217,
|
|
"num_tokens": 6046752.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.713192987442016,
|
|
"epoch": 2.6998432601880875,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 5.3787,
|
|
"mean_token_accuracy": 0.1704839751124382,
|
|
"num_tokens": 6055041.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.776202297210693,
|
|
"epoch": 2.7037617554858935,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045346482028183583,
|
|
"loss": 5.4045,
|
|
"mean_token_accuracy": 0.17221746146678923,
|
|
"num_tokens": 6064443.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.816495704650879,
|
|
"epoch": 2.707680250783699,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045328148192233823,
|
|
"loss": 5.4807,
|
|
"mean_token_accuracy": 0.16861338764429093,
|
|
"num_tokens": 6072761.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.794143867492676,
|
|
"epoch": 2.7115987460815045,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004530978249449254,
|
|
"loss": 5.4105,
|
|
"mean_token_accuracy": 0.16538355499505997,
|
|
"num_tokens": 6082083.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.777885484695434,
|
|
"epoch": 2.7155172413793105,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004529138496778222,
|
|
"loss": 5.4621,
|
|
"mean_token_accuracy": 0.1669634148478508,
|
|
"num_tokens": 6090116.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.781763553619385,
|
|
"epoch": 2.719435736677116,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004527295564498222,
|
|
"loss": 5.3484,
|
|
"mean_token_accuracy": 0.17244348376989366,
|
|
"num_tokens": 6099306.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.680258321762085,
|
|
"epoch": 2.7233542319749215,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004525449455902874,
|
|
"loss": 5.2893,
|
|
"mean_token_accuracy": 0.17130799740552902,
|
|
"num_tokens": 6108898.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.730869197845459,
|
|
"epoch": 2.7272727272727275,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004523600174291473,
|
|
"loss": 5.3252,
|
|
"mean_token_accuracy": 0.17265976071357728,
|
|
"num_tokens": 6117530.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.757677745819092,
|
|
"epoch": 2.731191222570533,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004521747722968985,
|
|
"loss": 5.3856,
|
|
"mean_token_accuracy": 0.17577317953109742,
|
|
"num_tokens": 6126222.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.7311993598937985,
|
|
"epoch": 2.7351097178683386,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045198921052460396,
|
|
"loss": 5.4153,
|
|
"mean_token_accuracy": 0.172159805893898,
|
|
"num_tokens": 6134742.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.659122133255005,
|
|
"epoch": 2.739028213166144,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004518033324438928,
|
|
"loss": 5.3601,
|
|
"mean_token_accuracy": 0.16933524757623672,
|
|
"num_tokens": 6143279.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.803249740600586,
|
|
"epoch": 2.7429467084639496,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004516171383869593,
|
|
"loss": 5.401,
|
|
"mean_token_accuracy": 0.17628977149724961,
|
|
"num_tokens": 6151648.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.7429467084639496,
|
|
"eval_entropy": 5.588564537292303,
|
|
"eval_loss": 5.767583847045898,
|
|
"eval_mean_token_accuracy": 0.15995488200307817,
|
|
"eval_num_tokens": 6151648.0,
|
|
"eval_runtime": 2.1818,
|
|
"eval_samples_per_second": 1889.282,
|
|
"eval_steps_per_second": 236.504,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.731971549987793,
|
|
"epoch": 2.7468652037617556,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045143062868656234,
|
|
"loss": 5.3803,
|
|
"mean_token_accuracy": 0.16510814130306245,
|
|
"num_tokens": 6160132.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.717729568481445,
|
|
"epoch": 2.750783699059561,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000451243803676025,
|
|
"loss": 5.3486,
|
|
"mean_token_accuracy": 0.17129723727703094,
|
|
"num_tokens": 6168562.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.703787040710449,
|
|
"epoch": 2.7547021943573666,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045105666368923397,
|
|
"loss": 5.3269,
|
|
"mean_token_accuracy": 0.17249537110328675,
|
|
"num_tokens": 6176810.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.606800985336304,
|
|
"epoch": 2.7586206896551726,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045086920906063866,
|
|
"loss": 5.2081,
|
|
"mean_token_accuracy": 0.19040362983942033,
|
|
"num_tokens": 6184862.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.682249689102173,
|
|
"epoch": 2.762539184952978,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045068144012525095,
|
|
"loss": 5.3344,
|
|
"mean_token_accuracy": 0.17671616971492768,
|
|
"num_tokens": 6192881.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.713724279403687,
|
|
"epoch": 2.7664576802507836,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045049335721864426,
|
|
"loss": 5.404,
|
|
"mean_token_accuracy": 0.174327252805233,
|
|
"num_tokens": 6200983.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.7311177253723145,
|
|
"epoch": 2.7703761755485896,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045030496067695336,
|
|
"loss": 5.4222,
|
|
"mean_token_accuracy": 0.1692997395992279,
|
|
"num_tokens": 6209540.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.833955240249634,
|
|
"epoch": 2.774294670846395,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004501162508368733,
|
|
"loss": 5.3969,
|
|
"mean_token_accuracy": 0.16795918494462966,
|
|
"num_tokens": 6218743.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.755520009994507,
|
|
"epoch": 2.7782131661442007,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004499272280356594,
|
|
"loss": 5.4261,
|
|
"mean_token_accuracy": 0.1681907966732979,
|
|
"num_tokens": 6227653.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.706432867050171,
|
|
"epoch": 2.782131661442006,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004497378926111257,
|
|
"loss": 5.4458,
|
|
"mean_token_accuracy": 0.16409747153520585,
|
|
"num_tokens": 6236487.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.786357355117798,
|
|
"epoch": 2.7860501567398117,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004495482449016456,
|
|
"loss": 5.4074,
|
|
"mean_token_accuracy": 0.17048471719026564,
|
|
"num_tokens": 6244907.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.718670320510864,
|
|
"epoch": 2.7899686520376177,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004493582852461501,
|
|
"loss": 5.4341,
|
|
"mean_token_accuracy": 0.15754127353429795,
|
|
"num_tokens": 6253749.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.716686630249024,
|
|
"epoch": 2.793887147335423,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004491680139841281,
|
|
"loss": 5.4504,
|
|
"mean_token_accuracy": 0.17363512217998506,
|
|
"num_tokens": 6262767.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.752951765060425,
|
|
"epoch": 2.7978056426332287,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000448977431455625,
|
|
"loss": 5.3158,
|
|
"mean_token_accuracy": 0.18336465805768967,
|
|
"num_tokens": 6271812.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.725945901870728,
|
|
"epoch": 2.8017241379310347,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044878653800124285,
|
|
"loss": 5.3924,
|
|
"mean_token_accuracy": 0.17299481928348542,
|
|
"num_tokens": 6279954.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.713065481185913,
|
|
"epoch": 2.80564263322884,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004485953339621391,
|
|
"loss": 5.4051,
|
|
"mean_token_accuracy": 0.16675891131162643,
|
|
"num_tokens": 6289053.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.804739856719971,
|
|
"epoch": 2.8095611285266457,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004484038196800265,
|
|
"loss": 5.4418,
|
|
"mean_token_accuracy": 0.1640549421310425,
|
|
"num_tokens": 6298137.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.722743177413941,
|
|
"epoch": 2.8134796238244513,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004482119954971719,
|
|
"loss": 5.3341,
|
|
"mean_token_accuracy": 0.16924910247325897,
|
|
"num_tokens": 6306643.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.703741407394409,
|
|
"epoch": 2.817398119122257,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044801986175639635,
|
|
"loss": 5.379,
|
|
"mean_token_accuracy": 0.17104603946208954,
|
|
"num_tokens": 6314767.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.667274332046508,
|
|
"epoch": 2.8213166144200628,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004478274188010741,
|
|
"loss": 5.2905,
|
|
"mean_token_accuracy": 0.1751739501953125,
|
|
"num_tokens": 6323335.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.729800033569336,
|
|
"epoch": 2.8252351097178683,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044763466697513173,
|
|
"loss": 5.4837,
|
|
"mean_token_accuracy": 0.16514647305011748,
|
|
"num_tokens": 6333741.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.733541011810303,
|
|
"epoch": 2.829153605015674,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044744160662304805,
|
|
"loss": 5.3694,
|
|
"mean_token_accuracy": 0.16914508789777755,
|
|
"num_tokens": 6342981.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.651356220245361,
|
|
"epoch": 2.83307210031348,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044724823808985325,
|
|
"loss": 5.3724,
|
|
"mean_token_accuracy": 0.17457425594329834,
|
|
"num_tokens": 6352211.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.821941471099853,
|
|
"epoch": 2.8369905956112853,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004470545617211283,
|
|
"loss": 5.438,
|
|
"mean_token_accuracy": 0.17105107009410858,
|
|
"num_tokens": 6361206.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.779411649703979,
|
|
"epoch": 2.840909090909091,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044686057786300423,
|
|
"loss": 5.3083,
|
|
"mean_token_accuracy": 0.16978153437376023,
|
|
"num_tokens": 6370224.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.642937326431275,
|
|
"epoch": 2.844827586206897,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044666628686216154,
|
|
"loss": 5.4173,
|
|
"mean_token_accuracy": 0.1641511932015419,
|
|
"num_tokens": 6379079.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.727200889587403,
|
|
"epoch": 2.8487460815047023,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044647168906583,
|
|
"loss": 5.3189,
|
|
"mean_token_accuracy": 0.17294642329216003,
|
|
"num_tokens": 6387635.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.746789932250977,
|
|
"epoch": 2.852664576802508,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044627678482178716,
|
|
"loss": 5.4033,
|
|
"mean_token_accuracy": 0.16777294576168061,
|
|
"num_tokens": 6396608.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.726373529434204,
|
|
"epoch": 2.8565830721003134,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004460815744783587,
|
|
"loss": 5.3585,
|
|
"mean_token_accuracy": 0.17138247340917587,
|
|
"num_tokens": 6405592.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.7246177196502686,
|
|
"epoch": 2.860501567398119,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000445886058384417,
|
|
"loss": 5.3463,
|
|
"mean_token_accuracy": 0.16987413614988328,
|
|
"num_tokens": 6414322.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.675555658340454,
|
|
"epoch": 2.864420062695925,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004456902368893811,
|
|
"loss": 5.3825,
|
|
"mean_token_accuracy": 0.16680659651756286,
|
|
"num_tokens": 6423541.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.741702747344971,
|
|
"epoch": 2.8683385579937304,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004454941103432158,
|
|
"loss": 5.3727,
|
|
"mean_token_accuracy": 0.17605556100606917,
|
|
"num_tokens": 6432738.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.67384934425354,
|
|
"epoch": 2.872257053291536,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044529767909643093,
|
|
"loss": 5.3711,
|
|
"mean_token_accuracy": 0.16563960164785385,
|
|
"num_tokens": 6441908.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.707652568817139,
|
|
"epoch": 2.876175548589342,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004451009435000811,
|
|
"loss": 5.4163,
|
|
"mean_token_accuracy": 0.16620732098817825,
|
|
"num_tokens": 6450652.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.781555318832398,
|
|
"epoch": 2.8800940438871474,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004449039039057647,
|
|
"loss": 5.459,
|
|
"mean_token_accuracy": 0.1647094890475273,
|
|
"num_tokens": 6458774.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.728143930435181,
|
|
"epoch": 2.884012539184953,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 5.3631,
|
|
"mean_token_accuracy": 0.1755487710237503,
|
|
"num_tokens": 6467432.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.742747688293457,
|
|
"epoch": 2.887931034482759,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004445089141323415,
|
|
"loss": 5.4097,
|
|
"mean_token_accuracy": 0.16459475159645082,
|
|
"num_tokens": 6477224.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.694394207000732,
|
|
"epoch": 2.8918495297805644,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044431096465914554,
|
|
"loss": 5.389,
|
|
"mean_token_accuracy": 0.17142396569252014,
|
|
"num_tokens": 6486487.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.72460618019104,
|
|
"epoch": 2.89576802507837,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044411271259980315,
|
|
"loss": 5.4143,
|
|
"mean_token_accuracy": 0.17046311050653457,
|
|
"num_tokens": 6495127.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.597875452041626,
|
|
"epoch": 2.8996865203761755,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004439141583086231,
|
|
"loss": 5.2905,
|
|
"mean_token_accuracy": 0.17678283154964447,
|
|
"num_tokens": 6504466.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.677201938629151,
|
|
"epoch": 2.903605015673981,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044371530214045395,
|
|
"loss": 5.2942,
|
|
"mean_token_accuracy": 0.17507077306509017,
|
|
"num_tokens": 6513310.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.63643741607666,
|
|
"epoch": 2.907523510971787,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044351614445068413,
|
|
"loss": 5.3722,
|
|
"mean_token_accuracy": 0.16740810871124268,
|
|
"num_tokens": 6522740.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.751364612579346,
|
|
"epoch": 2.9114420062695925,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044331668559524043,
|
|
"loss": 5.3329,
|
|
"mean_token_accuracy": 0.1740532785654068,
|
|
"num_tokens": 6531194.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.680466079711914,
|
|
"epoch": 2.915360501567398,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004431169259305883,
|
|
"loss": 5.2654,
|
|
"mean_token_accuracy": 0.18289788216352462,
|
|
"num_tokens": 6539631.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.652289009094238,
|
|
"epoch": 2.919278996865204,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004429168658137306,
|
|
"loss": 5.3655,
|
|
"mean_token_accuracy": 0.16826859563589097,
|
|
"num_tokens": 6548728.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.676466703414917,
|
|
"epoch": 2.9231974921630095,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044271650560220746,
|
|
"loss": 5.2156,
|
|
"mean_token_accuracy": 0.17708035856485366,
|
|
"num_tokens": 6557247.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.698669338226319,
|
|
"epoch": 2.927115987460815,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044251584565409464,
|
|
"loss": 5.4719,
|
|
"mean_token_accuracy": 0.16626885086297988,
|
|
"num_tokens": 6566014.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.708856439590454,
|
|
"epoch": 2.9310344827586206,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004423148863280044,
|
|
"loss": 5.3656,
|
|
"mean_token_accuracy": 0.1706095814704895,
|
|
"num_tokens": 6574346.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.729553604125977,
|
|
"epoch": 2.934952978056426,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044211362798308334,
|
|
"loss": 5.4061,
|
|
"mean_token_accuracy": 0.17207707613706588,
|
|
"num_tokens": 6582950.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.665266704559326,
|
|
"epoch": 2.938871473354232,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004419120709790129,
|
|
"loss": 5.3596,
|
|
"mean_token_accuracy": 0.1720501586794853,
|
|
"num_tokens": 6591882.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.691666507720948,
|
|
"epoch": 2.9427899686520376,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044171021567600814,
|
|
"loss": 5.3666,
|
|
"mean_token_accuracy": 0.17571640014648438,
|
|
"num_tokens": 6600346.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.752440738677978,
|
|
"epoch": 2.946708463949843,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044150806243481715,
|
|
"loss": 5.3754,
|
|
"mean_token_accuracy": 0.17345160096883774,
|
|
"num_tokens": 6608874.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.698383092880249,
|
|
"epoch": 2.950626959247649,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004413056116167206,
|
|
"loss": 5.4078,
|
|
"mean_token_accuracy": 0.17090679407119752,
|
|
"num_tokens": 6617201.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.754421091079712,
|
|
"epoch": 2.9545454545454546,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004411028635835309,
|
|
"loss": 5.3109,
|
|
"mean_token_accuracy": 0.1758136436343193,
|
|
"num_tokens": 6626133.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.683559274673462,
|
|
"epoch": 2.95846394984326,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004408998186975917,
|
|
"loss": 5.3229,
|
|
"mean_token_accuracy": 0.17818258404731752,
|
|
"num_tokens": 6634565.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.671072626113892,
|
|
"epoch": 2.962382445141066,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044069647732177696,
|
|
"loss": 5.393,
|
|
"mean_token_accuracy": 0.1764868602156639,
|
|
"num_tokens": 6642697.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.6621253490448,
|
|
"epoch": 2.9663009404388716,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044049283981949103,
|
|
"loss": 5.3014,
|
|
"mean_token_accuracy": 0.17470928728580476,
|
|
"num_tokens": 6650990.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.693866825103759,
|
|
"epoch": 2.970219435736677,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004402889065546667,
|
|
"loss": 5.2847,
|
|
"mean_token_accuracy": 0.1738991007208824,
|
|
"num_tokens": 6659838.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.703425455093384,
|
|
"epoch": 2.9741379310344827,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00044008467789176625,
|
|
"loss": 5.2921,
|
|
"mean_token_accuracy": 0.1823958709836006,
|
|
"num_tokens": 6667804.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.749673318862915,
|
|
"epoch": 2.978056426332288,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004398801541957791,
|
|
"loss": 5.4924,
|
|
"mean_token_accuracy": 0.17257193177938462,
|
|
"num_tokens": 6677510.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.7611347198486325,
|
|
"epoch": 2.981974921630094,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004396753358322223,
|
|
"loss": 5.3348,
|
|
"mean_token_accuracy": 0.17410726845264435,
|
|
"num_tokens": 6685436.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.6394284725189205,
|
|
"epoch": 2.9858934169278997,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004394702231671396,
|
|
"loss": 5.3046,
|
|
"mean_token_accuracy": 0.1758171871304512,
|
|
"num_tokens": 6694247.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.713444375991822,
|
|
"epoch": 2.989811912225705,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004392648165671004,
|
|
"loss": 5.4581,
|
|
"mean_token_accuracy": 0.17517127841711044,
|
|
"num_tokens": 6702949.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.717865371704102,
|
|
"epoch": 2.993730407523511,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004390591163991998,
|
|
"loss": 5.3383,
|
|
"mean_token_accuracy": 0.176737479865551,
|
|
"num_tokens": 6712231.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.7436614513397215,
|
|
"epoch": 2.9976489028213167,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043885312303105725,
|
|
"loss": 5.3923,
|
|
"mean_token_accuracy": 0.1645299732685089,
|
|
"num_tokens": 6721714.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.666221666336059,
|
|
"epoch": 3.001567398119122,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004386468368308163,
|
|
"loss": 5.2456,
|
|
"mean_token_accuracy": 0.1843735009431839,
|
|
"num_tokens": 6730147.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.66095118522644,
|
|
"epoch": 3.0054858934169277,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004384402581671438,
|
|
"loss": 4.9991,
|
|
"mean_token_accuracy": 0.18904417306184768,
|
|
"num_tokens": 6738456.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.63086166381836,
|
|
"epoch": 3.0094043887147337,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004382333874092295,
|
|
"loss": 5.0237,
|
|
"mean_token_accuracy": 0.19114759266376496,
|
|
"num_tokens": 6746723.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.663303375244141,
|
|
"epoch": 3.0133228840125392,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00043802622492678466,
|
|
"loss": 5.1357,
|
|
"mean_token_accuracy": 0.18646033406257628,
|
|
"num_tokens": 6756713.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.63843092918396,
|
|
"epoch": 3.0172413793103448,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004378187710900426,
|
|
"loss": 5.0207,
|
|
"mean_token_accuracy": 0.19423189610242844,
|
|
"num_tokens": 6765533.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.629498767852783,
|
|
"epoch": 3.0211598746081503,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043761102626975674,
|
|
"loss": 4.9801,
|
|
"mean_token_accuracy": 0.1985636606812477,
|
|
"num_tokens": 6774467.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.585388135910034,
|
|
"epoch": 3.0250783699059562,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004374029908372007,
|
|
"loss": 4.977,
|
|
"mean_token_accuracy": 0.19902051538228988,
|
|
"num_tokens": 6784024.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.65576581954956,
|
|
"epoch": 3.0289968652037618,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043719466516416774,
|
|
"loss": 5.1122,
|
|
"mean_token_accuracy": 0.18697987794876098,
|
|
"num_tokens": 6793022.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.724702167510986,
|
|
"epoch": 3.0329153605015673,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043698604962296946,
|
|
"loss": 5.0144,
|
|
"mean_token_accuracy": 0.20193614214658737,
|
|
"num_tokens": 6801299.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.548857641220093,
|
|
"epoch": 3.0368338557993733,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043677714458643566,
|
|
"loss": 5.0753,
|
|
"mean_token_accuracy": 0.18923732489347458,
|
|
"num_tokens": 6809775.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.674850511550903,
|
|
"epoch": 3.040752351097179,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043656795042791357,
|
|
"loss": 5.0856,
|
|
"mean_token_accuracy": 0.1887703686952591,
|
|
"num_tokens": 6818427.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.641047668457031,
|
|
"epoch": 3.0446708463949843,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004363584675212671,
|
|
"loss": 5.0755,
|
|
"mean_token_accuracy": 0.18946368843317032,
|
|
"num_tokens": 6827554.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.612737035751342,
|
|
"epoch": 3.04858934169279,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004361486962408761,
|
|
"loss": 5.055,
|
|
"mean_token_accuracy": 0.1952921763062477,
|
|
"num_tokens": 6836777.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.589520788192749,
|
|
"epoch": 3.052507836990596,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004359386369616359,
|
|
"loss": 5.052,
|
|
"mean_token_accuracy": 0.1927233397960663,
|
|
"num_tokens": 6846043.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.651739311218262,
|
|
"epoch": 3.0564263322884013,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004357282900589565,
|
|
"loss": 5.0407,
|
|
"mean_token_accuracy": 0.19105500131845474,
|
|
"num_tokens": 6855172.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.54166464805603,
|
|
"epoch": 3.060344827586207,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043551765590876183,
|
|
"loss": 4.9465,
|
|
"mean_token_accuracy": 0.20180509388446807,
|
|
"num_tokens": 6863363.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.607993459701538,
|
|
"epoch": 3.0642633228840124,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004353067348874894,
|
|
"loss": 5.0172,
|
|
"mean_token_accuracy": 0.20221378207206725,
|
|
"num_tokens": 6871991.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.625034809112549,
|
|
"epoch": 3.0681818181818183,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 5.0342,
|
|
"mean_token_accuracy": 0.1984927162528038,
|
|
"num_tokens": 6880824.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.588798952102661,
|
|
"epoch": 3.072100313479624,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004348840337400233,
|
|
"loss": 5.0121,
|
|
"mean_token_accuracy": 0.19837913662195206,
|
|
"num_tokens": 6888954.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.6181916236877445,
|
|
"epoch": 3.0760188087774294,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043467225436926517,
|
|
"loss": 5.0158,
|
|
"mean_token_accuracy": 0.1921957403421402,
|
|
"num_tokens": 6897554.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.701251268386841,
|
|
"epoch": 3.079937304075235,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004344601896382988,
|
|
"loss": 5.1925,
|
|
"mean_token_accuracy": 0.18170069456100463,
|
|
"num_tokens": 6906608.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.580873346328735,
|
|
"epoch": 3.083855799373041,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043424783992611837,
|
|
"loss": 5.1074,
|
|
"mean_token_accuracy": 0.19211028516292572,
|
|
"num_tokens": 6915345.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.61749677658081,
|
|
"epoch": 3.0877742946708464,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043403520561222705,
|
|
"loss": 5.0816,
|
|
"mean_token_accuracy": 0.19388263970613479,
|
|
"num_tokens": 6924068.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.569154977798462,
|
|
"epoch": 3.091692789968652,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004338222870766371,
|
|
"loss": 5.1028,
|
|
"mean_token_accuracy": 0.17925256192684175,
|
|
"num_tokens": 6932713.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.551962375640869,
|
|
"epoch": 3.0956112852664575,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043360908469986827,
|
|
"loss": 5.0332,
|
|
"mean_token_accuracy": 0.1869243711233139,
|
|
"num_tokens": 6941361.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.674963855743409,
|
|
"epoch": 3.0995297805642634,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004333955988629478,
|
|
"loss": 5.1594,
|
|
"mean_token_accuracy": 0.18088530004024506,
|
|
"num_tokens": 6949960.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.716828489303589,
|
|
"epoch": 3.103448275862069,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043318182994740945,
|
|
"loss": 5.082,
|
|
"mean_token_accuracy": 0.19028955101966857,
|
|
"num_tokens": 6958539.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.590631628036499,
|
|
"epoch": 3.1073667711598745,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004329677783352931,
|
|
"loss": 5.0163,
|
|
"mean_token_accuracy": 0.20413250625133514,
|
|
"num_tokens": 6966286.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.584048509597778,
|
|
"epoch": 3.1112852664576804,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004327534444091436,
|
|
"loss": 5.1348,
|
|
"mean_token_accuracy": 0.1766098842024803,
|
|
"num_tokens": 6974925.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.715192031860352,
|
|
"epoch": 3.115203761755486,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043253882855201037,
|
|
"loss": 5.1342,
|
|
"mean_token_accuracy": 0.185565946996212,
|
|
"num_tokens": 6983908.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.698682689666748,
|
|
"epoch": 3.1191222570532915,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043232393114744683,
|
|
"loss": 5.1047,
|
|
"mean_token_accuracy": 0.18598025292158127,
|
|
"num_tokens": 6993044.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.559580564498901,
|
|
"epoch": 3.123040752351097,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004321087525795095,
|
|
"loss": 5.0675,
|
|
"mean_token_accuracy": 0.18857765048742295,
|
|
"num_tokens": 7001620.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.602130556106568,
|
|
"epoch": 3.126959247648903,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004318932932327573,
|
|
"loss": 5.0808,
|
|
"mean_token_accuracy": 0.18459939509630202,
|
|
"num_tokens": 7010146.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.61126480102539,
|
|
"epoch": 3.1308777429467085,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000431677553492251,
|
|
"loss": 5.0146,
|
|
"mean_token_accuracy": 0.19256452172994615,
|
|
"num_tokens": 7018885.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.618481540679932,
|
|
"epoch": 3.134796238244514,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043146153374355256,
|
|
"loss": 5.0681,
|
|
"mean_token_accuracy": 0.18420436680316926,
|
|
"num_tokens": 7027220.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.134796238244514,
|
|
"eval_entropy": 5.408100193785143,
|
|
"eval_loss": 5.757437229156494,
|
|
"eval_mean_token_accuracy": 0.16196971209070019,
|
|
"eval_num_tokens": 7027220.0,
|
|
"eval_runtime": 2.3825,
|
|
"eval_samples_per_second": 1730.134,
|
|
"eval_steps_per_second": 216.582,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.540895414352417,
|
|
"epoch": 3.1387147335423196,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043124523437272427,
|
|
"loss": 4.9373,
|
|
"mean_token_accuracy": 0.20068125128746034,
|
|
"num_tokens": 7036333.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.540096426010132,
|
|
"epoch": 3.1426332288401255,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004310286557663282,
|
|
"loss": 5.0241,
|
|
"mean_token_accuracy": 0.19308097362518312,
|
|
"num_tokens": 7044889.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.59888801574707,
|
|
"epoch": 3.146551724137931,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004308117983114254,
|
|
"loss": 5.0926,
|
|
"mean_token_accuracy": 0.19120583236217498,
|
|
"num_tokens": 7054439.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.603954219818116,
|
|
"epoch": 3.1504702194357366,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004305946623955754,
|
|
"loss": 5.1127,
|
|
"mean_token_accuracy": 0.1859748527407646,
|
|
"num_tokens": 7063285.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.69248046875,
|
|
"epoch": 3.1543887147335425,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043037724840683516,
|
|
"loss": 5.112,
|
|
"mean_token_accuracy": 0.1817049890756607,
|
|
"num_tokens": 7071992.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.6307775497436525,
|
|
"epoch": 3.158307210031348,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043015955673375876,
|
|
"loss": 5.0834,
|
|
"mean_token_accuracy": 0.19021794199943542,
|
|
"num_tokens": 7081249.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.644666051864624,
|
|
"epoch": 3.1622257053291536,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004299415877653966,
|
|
"loss": 5.0554,
|
|
"mean_token_accuracy": 0.19068792462348938,
|
|
"num_tokens": 7089465.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.605574607849121,
|
|
"epoch": 3.166144200626959,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004297233418912945,
|
|
"loss": 5.1216,
|
|
"mean_token_accuracy": 0.18485265076160431,
|
|
"num_tokens": 7098281.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.620939636230469,
|
|
"epoch": 3.170062695924765,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004295048195014932,
|
|
"loss": 5.089,
|
|
"mean_token_accuracy": 0.1931542545557022,
|
|
"num_tokens": 7106822.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.6322752952575685,
|
|
"epoch": 3.1739811912225706,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004292860209865277,
|
|
"loss": 5.0968,
|
|
"mean_token_accuracy": 0.19147595167160034,
|
|
"num_tokens": 7115511.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.554740476608276,
|
|
"epoch": 3.177899686520376,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004290669467374263,
|
|
"loss": 5.0114,
|
|
"mean_token_accuracy": 0.1924958646297455,
|
|
"num_tokens": 7124495.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.613631391525269,
|
|
"epoch": 3.1818181818181817,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042884759714571037,
|
|
"loss": 5.0848,
|
|
"mean_token_accuracy": 0.18693549782037736,
|
|
"num_tokens": 7132632.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.583876895904541,
|
|
"epoch": 3.1857366771159876,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004286279726033932,
|
|
"loss": 5.0538,
|
|
"mean_token_accuracy": 0.19116956889629363,
|
|
"num_tokens": 7141715.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.6701226234436035,
|
|
"epoch": 3.189655172413793,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042840807350297933,
|
|
"loss": 5.129,
|
|
"mean_token_accuracy": 0.18312158584594726,
|
|
"num_tokens": 7150322.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.561582374572754,
|
|
"epoch": 3.1935736677115987,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042818790023746407,
|
|
"loss": 5.0524,
|
|
"mean_token_accuracy": 0.18687334507703782,
|
|
"num_tokens": 7159204.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.5666851043701175,
|
|
"epoch": 3.197492163009404,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042796745320033296,
|
|
"loss": 5.0699,
|
|
"mean_token_accuracy": 0.18837817907333373,
|
|
"num_tokens": 7167695.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.580666208267212,
|
|
"epoch": 3.20141065830721,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042774673278556043,
|
|
"loss": 5.118,
|
|
"mean_token_accuracy": 0.1915913537144661,
|
|
"num_tokens": 7176874.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.5748954772949215,
|
|
"epoch": 3.2053291536050157,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004275257393876097,
|
|
"loss": 5.067,
|
|
"mean_token_accuracy": 0.19157032370567323,
|
|
"num_tokens": 7184555.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.590212631225586,
|
|
"epoch": 3.209247648902821,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004273044734014318,
|
|
"loss": 5.0494,
|
|
"mean_token_accuracy": 0.18979770243167876,
|
|
"num_tokens": 7192975.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.681621265411377,
|
|
"epoch": 3.2131661442006267,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042708293522246486,
|
|
"loss": 5.122,
|
|
"mean_token_accuracy": 0.1875492513179779,
|
|
"num_tokens": 7201681.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.584944152832032,
|
|
"epoch": 3.2170846394984327,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004268611252466337,
|
|
"loss": 5.0179,
|
|
"mean_token_accuracy": 0.1960278034210205,
|
|
"num_tokens": 7209930.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.584336662292481,
|
|
"epoch": 3.2210031347962382,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004266390438703486,
|
|
"loss": 5.1265,
|
|
"mean_token_accuracy": 0.17799908071756362,
|
|
"num_tokens": 7218746.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.61042046546936,
|
|
"epoch": 3.2249216300940438,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042641669149050493,
|
|
"loss": 5.0526,
|
|
"mean_token_accuracy": 0.190690341591835,
|
|
"num_tokens": 7227475.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.703586912155151,
|
|
"epoch": 3.2288401253918497,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004261940685044825,
|
|
"loss": 5.1693,
|
|
"mean_token_accuracy": 0.18507494181394576,
|
|
"num_tokens": 7235618.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.641922473907471,
|
|
"epoch": 3.2327586206896552,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042597117531014474,
|
|
"loss": 5.0722,
|
|
"mean_token_accuracy": 0.19189542531967163,
|
|
"num_tokens": 7244571.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.5707978248596195,
|
|
"epoch": 3.2366771159874608,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004257480123058378,
|
|
"loss": 5.1191,
|
|
"mean_token_accuracy": 0.18079930245876313,
|
|
"num_tokens": 7252731.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.566800880432129,
|
|
"epoch": 3.2405956112852663,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042552457989039036,
|
|
"loss": 4.9987,
|
|
"mean_token_accuracy": 0.19003403335809707,
|
|
"num_tokens": 7261547.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.629475879669189,
|
|
"epoch": 3.2445141065830723,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042530087846311213,
|
|
"loss": 5.0987,
|
|
"mean_token_accuracy": 0.19276611655950546,
|
|
"num_tokens": 7269667.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.594238042831421,
|
|
"epoch": 3.248432601880878,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042507690842379396,
|
|
"loss": 5.1011,
|
|
"mean_token_accuracy": 0.19355297982692718,
|
|
"num_tokens": 7278833.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.590455150604248,
|
|
"epoch": 3.2523510971786833,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 5.0749,
|
|
"mean_token_accuracy": 0.1849210813641548,
|
|
"num_tokens": 7287110.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.600765562057495,
|
|
"epoch": 3.256269592476489,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042462816411060025,
|
|
"loss": 5.0571,
|
|
"mean_token_accuracy": 0.19231417775154114,
|
|
"num_tokens": 7295334.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.534113979339599,
|
|
"epoch": 3.260188087774295,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004244033906387035,
|
|
"loss": 5.0293,
|
|
"mean_token_accuracy": 0.19865134805440904,
|
|
"num_tokens": 7303679.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.580339097976685,
|
|
"epoch": 3.2641065830721003,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004241783501587231,
|
|
"loss": 5.0717,
|
|
"mean_token_accuracy": 0.1868339627981186,
|
|
"num_tokens": 7312523.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.6370627880096436,
|
|
"epoch": 3.268025078369906,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042395304307284284,
|
|
"loss": 5.0658,
|
|
"mean_token_accuracy": 0.18846231698989868,
|
|
"num_tokens": 7321154.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.6309895515441895,
|
|
"epoch": 3.271943573667712,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004237274697837229,
|
|
"loss": 5.1125,
|
|
"mean_token_accuracy": 0.1821906626224518,
|
|
"num_tokens": 7330808.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.596603107452393,
|
|
"epoch": 3.2758620689655173,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004235016306944996,
|
|
"loss": 5.0823,
|
|
"mean_token_accuracy": 0.19341420531272888,
|
|
"num_tokens": 7339286.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.580079412460327,
|
|
"epoch": 3.279780564263323,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004232755262087837,
|
|
"loss": 5.0779,
|
|
"mean_token_accuracy": 0.19512812942266464,
|
|
"num_tokens": 7347840.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.591745233535766,
|
|
"epoch": 3.2836990595611284,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042304915673066083,
|
|
"loss": 5.0081,
|
|
"mean_token_accuracy": 0.19409030675888062,
|
|
"num_tokens": 7355934.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.573127031326294,
|
|
"epoch": 3.287617554858934,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042282252266468985,
|
|
"loss": 4.9801,
|
|
"mean_token_accuracy": 0.1978231653571129,
|
|
"num_tokens": 7365126.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.574249172210694,
|
|
"epoch": 3.29153605015674,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004225956244159025,
|
|
"loss": 5.0863,
|
|
"mean_token_accuracy": 0.18711487352848052,
|
|
"num_tokens": 7373954.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.614953184127808,
|
|
"epoch": 3.2954545454545454,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004223684623898029,
|
|
"loss": 5.024,
|
|
"mean_token_accuracy": 0.19016447961330413,
|
|
"num_tokens": 7382032.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.602483654022217,
|
|
"epoch": 3.299373040752351,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004221410369923662,
|
|
"loss": 5.1059,
|
|
"mean_token_accuracy": 0.18691661804914475,
|
|
"num_tokens": 7390700.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.564588308334351,
|
|
"epoch": 3.303291536050157,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042191334863003873,
|
|
"loss": 5.0962,
|
|
"mean_token_accuracy": 0.1806216135621071,
|
|
"num_tokens": 7400007.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.594666290283203,
|
|
"epoch": 3.3072100313479624,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004216853977097363,
|
|
"loss": 5.0038,
|
|
"mean_token_accuracy": 0.19468976110219954,
|
|
"num_tokens": 7409161.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.551057910919189,
|
|
"epoch": 3.311128526645768,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004214571846388442,
|
|
"loss": 5.1233,
|
|
"mean_token_accuracy": 0.18430280238389968,
|
|
"num_tokens": 7418320.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.6062404155731205,
|
|
"epoch": 3.3150470219435735,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004212287098252164,
|
|
"loss": 4.9956,
|
|
"mean_token_accuracy": 0.18815011531114578,
|
|
"num_tokens": 7427736.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.532464027404785,
|
|
"epoch": 3.3189655172413794,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004209999736771742,
|
|
"loss": 4.9711,
|
|
"mean_token_accuracy": 0.19824261218309402,
|
|
"num_tokens": 7436412.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.510031461715698,
|
|
"epoch": 3.322884012539185,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004207709766035063,
|
|
"loss": 5.0629,
|
|
"mean_token_accuracy": 0.1840561419725418,
|
|
"num_tokens": 7445227.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.685104608535767,
|
|
"epoch": 3.3268025078369905,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004205417190134674,
|
|
"loss": 5.0824,
|
|
"mean_token_accuracy": 0.19221481531858445,
|
|
"num_tokens": 7453108.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.618586778640747,
|
|
"epoch": 3.330721003134796,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004203122013167783,
|
|
"loss": 5.1159,
|
|
"mean_token_accuracy": 0.18530269861221313,
|
|
"num_tokens": 7461348.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.541234302520752,
|
|
"epoch": 3.334639498432602,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042008242392362413,
|
|
"loss": 5.076,
|
|
"mean_token_accuracy": 0.18670390844345092,
|
|
"num_tokens": 7470553.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.651902389526367,
|
|
"epoch": 3.3385579937304075,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041985238724465433,
|
|
"loss": 5.1359,
|
|
"mean_token_accuracy": 0.1860004961490631,
|
|
"num_tokens": 7479865.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.602338027954102,
|
|
"epoch": 3.342476489028213,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041962209169098193,
|
|
"loss": 5.1564,
|
|
"mean_token_accuracy": 0.18450002521276473,
|
|
"num_tokens": 7488809.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.529464292526245,
|
|
"epoch": 3.346394984326019,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004193915376741823,
|
|
"loss": 4.9972,
|
|
"mean_token_accuracy": 0.1907404065132141,
|
|
"num_tokens": 7497845.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.549796867370605,
|
|
"epoch": 3.3503134796238245,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004191607256062928,
|
|
"loss": 5.0411,
|
|
"mean_token_accuracy": 0.19499893635511398,
|
|
"num_tokens": 7507192.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.548674154281616,
|
|
"epoch": 3.35423197492163,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004189296558998121,
|
|
"loss": 5.0669,
|
|
"mean_token_accuracy": 0.19317741096019744,
|
|
"num_tokens": 7516266.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.560562705993652,
|
|
"epoch": 3.3581504702194356,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004186983289676992,
|
|
"loss": 5.0759,
|
|
"mean_token_accuracy": 0.19128655642271042,
|
|
"num_tokens": 7525013.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.617566013336182,
|
|
"epoch": 3.3620689655172415,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041846674522337296,
|
|
"loss": 5.1596,
|
|
"mean_token_accuracy": 0.1860888570547104,
|
|
"num_tokens": 7534872.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.541995859146118,
|
|
"epoch": 3.365987460815047,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041823490508071076,
|
|
"loss": 5.0972,
|
|
"mean_token_accuracy": 0.19157399237155914,
|
|
"num_tokens": 7544156.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.526349401473999,
|
|
"epoch": 3.3699059561128526,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000418002808954049,
|
|
"loss": 5.0414,
|
|
"mean_token_accuracy": 0.19094870537519454,
|
|
"num_tokens": 7553193.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.6064964771270756,
|
|
"epoch": 3.373824451410658,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041777045725818057,
|
|
"loss": 5.0603,
|
|
"mean_token_accuracy": 0.19296390265226365,
|
|
"num_tokens": 7561965.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.576651859283447,
|
|
"epoch": 3.377742946708464,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000417537850408356,
|
|
"loss": 5.1304,
|
|
"mean_token_accuracy": 0.1822283685207367,
|
|
"num_tokens": 7570233.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.5880059719085695,
|
|
"epoch": 3.3816614420062696,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004173049888202814,
|
|
"loss": 5.0447,
|
|
"mean_token_accuracy": 0.19474864155054092,
|
|
"num_tokens": 7578575.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.632257032394409,
|
|
"epoch": 3.385579937304075,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004170718729101179,
|
|
"loss": 5.1276,
|
|
"mean_token_accuracy": 0.18952592015266417,
|
|
"num_tokens": 7587771.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.552601432800293,
|
|
"epoch": 3.389498432601881,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041683850309448187,
|
|
"loss": 4.9755,
|
|
"mean_token_accuracy": 0.19489433765411376,
|
|
"num_tokens": 7595715.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.52767243385315,
|
|
"epoch": 3.3934169278996866,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041660487979044264,
|
|
"loss": 5.0386,
|
|
"mean_token_accuracy": 0.18672572523355485,
|
|
"num_tokens": 7604358.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.559977054595947,
|
|
"epoch": 3.397335423197492,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004163710034155231,
|
|
"loss": 5.0547,
|
|
"mean_token_accuracy": 0.19408606439828874,
|
|
"num_tokens": 7612797.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.545515775680542,
|
|
"epoch": 3.4012539184952977,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004161368743876982,
|
|
"loss": 5.1187,
|
|
"mean_token_accuracy": 0.1853936180472374,
|
|
"num_tokens": 7621687.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.523568201065063,
|
|
"epoch": 3.405172413793103,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004159024931253945,
|
|
"loss": 5.0805,
|
|
"mean_token_accuracy": 0.18181275725364685,
|
|
"num_tokens": 7630353.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.574881458282471,
|
|
"epoch": 3.409090909090909,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041566786004748943,
|
|
"loss": 5.0375,
|
|
"mean_token_accuracy": 0.19578725397586821,
|
|
"num_tokens": 7639114.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.639018249511719,
|
|
"epoch": 3.4130094043887147,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041543297557331015,
|
|
"loss": 5.0994,
|
|
"mean_token_accuracy": 0.18374128192663192,
|
|
"num_tokens": 7648405.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.537978935241699,
|
|
"epoch": 3.41692789968652,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004151978401226335,
|
|
"loss": 5.0276,
|
|
"mean_token_accuracy": 0.18742780536413192,
|
|
"num_tokens": 7656501.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.48979344367981,
|
|
"epoch": 3.420846394984326,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00041496245411568435,
|
|
"loss": 5.0415,
|
|
"mean_token_accuracy": 0.18944326043128967,
|
|
"num_tokens": 7665230.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.4636577606201175,
|
|
"epoch": 3.4247648902821317,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004147268179731359,
|
|
"loss": 5.0444,
|
|
"mean_token_accuracy": 0.19116580337285996,
|
|
"num_tokens": 7674027.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.594990873336792,
|
|
"epoch": 3.4286833855799372,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041449093211610815,
|
|
"loss": 5.0671,
|
|
"mean_token_accuracy": 0.18775967210531236,
|
|
"num_tokens": 7682622.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.564218425750733,
|
|
"epoch": 3.4326018808777428,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041425479696616734,
|
|
"loss": 5.0313,
|
|
"mean_token_accuracy": 0.19167483747005462,
|
|
"num_tokens": 7691837.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.4585918426513675,
|
|
"epoch": 3.4365203761755487,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 5.0324,
|
|
"mean_token_accuracy": 0.19637792110443114,
|
|
"num_tokens": 7700437.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.555591869354248,
|
|
"epoch": 3.4404388714733543,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041378178047603845,
|
|
"loss": 5.061,
|
|
"mean_token_accuracy": 0.19042093455791473,
|
|
"num_tokens": 7708960.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.580157899856568,
|
|
"epoch": 3.44435736677116,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004135448999812074,
|
|
"loss": 5.1258,
|
|
"mean_token_accuracy": 0.18463489264249802,
|
|
"num_tokens": 7717467.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.624056100845337,
|
|
"epoch": 3.4482758620689653,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004133077718841763,
|
|
"loss": 5.2073,
|
|
"mean_token_accuracy": 0.18424418568611145,
|
|
"num_tokens": 7726356.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.605796194076538,
|
|
"epoch": 3.4521943573667713,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041307039660873113,
|
|
"loss": 5.1763,
|
|
"mean_token_accuracy": 0.19027031809091569,
|
|
"num_tokens": 7735821.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.621857452392578,
|
|
"epoch": 3.456112852664577,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004128327745791002,
|
|
"loss": 5.1307,
|
|
"mean_token_accuracy": 0.18760121911764144,
|
|
"num_tokens": 7744614.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.69903302192688,
|
|
"epoch": 3.4600313479623823,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004125949062199526,
|
|
"loss": 5.2115,
|
|
"mean_token_accuracy": 0.17842825055122374,
|
|
"num_tokens": 7754042.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.607316684722901,
|
|
"epoch": 3.4639498432601883,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041235679195639766,
|
|
"loss": 5.0772,
|
|
"mean_token_accuracy": 0.18724562376737594,
|
|
"num_tokens": 7762269.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.631308460235596,
|
|
"epoch": 3.467868338557994,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041211843221398406,
|
|
"loss": 5.1696,
|
|
"mean_token_accuracy": 0.18203292787075043,
|
|
"num_tokens": 7771083.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.621017932891846,
|
|
"epoch": 3.4717868338557993,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004118798274186994,
|
|
"loss": 5.1278,
|
|
"mean_token_accuracy": 0.1891021251678467,
|
|
"num_tokens": 7780470.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.543201971054077,
|
|
"epoch": 3.475705329153605,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004116409779969691,
|
|
"loss": 4.9987,
|
|
"mean_token_accuracy": 0.19531961679458618,
|
|
"num_tokens": 7789363.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.567298460006714,
|
|
"epoch": 3.479623824451411,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041140188437565586,
|
|
"loss": 5.0855,
|
|
"mean_token_accuracy": 0.19768042117357254,
|
|
"num_tokens": 7797505.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.63401985168457,
|
|
"epoch": 3.4835423197492164,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041116254698205873,
|
|
"loss": 5.1812,
|
|
"mean_token_accuracy": 0.1824115112423897,
|
|
"num_tokens": 7806843.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.591701698303223,
|
|
"epoch": 3.487460815047022,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041092296624391244,
|
|
"loss": 5.1181,
|
|
"mean_token_accuracy": 0.18469711244106293,
|
|
"num_tokens": 7815661.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.523241710662842,
|
|
"epoch": 3.4913793103448274,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004106831425893865,
|
|
"loss": 5.0917,
|
|
"mean_token_accuracy": 0.19193295389413834,
|
|
"num_tokens": 7824301.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.593221807479859,
|
|
"epoch": 3.4952978056426334,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004104430764470849,
|
|
"loss": 5.132,
|
|
"mean_token_accuracy": 0.18751560300588607,
|
|
"num_tokens": 7833054.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.583597946166992,
|
|
"epoch": 3.499216300940439,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004102027682460445,
|
|
"loss": 5.1244,
|
|
"mean_token_accuracy": 0.1876629650592804,
|
|
"num_tokens": 7842080.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.555351209640503,
|
|
"epoch": 3.5031347962382444,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004099622184157353,
|
|
"loss": 5.0528,
|
|
"mean_token_accuracy": 0.18947582989931105,
|
|
"num_tokens": 7851358.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.6697423458099365,
|
|
"epoch": 3.5070532915360504,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004097214273860586,
|
|
"loss": 5.1209,
|
|
"mean_token_accuracy": 0.18580541014671326,
|
|
"num_tokens": 7860445.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.629621982574463,
|
|
"epoch": 3.510971786833856,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004094803955873471,
|
|
"loss": 5.1136,
|
|
"mean_token_accuracy": 0.18376435935497284,
|
|
"num_tokens": 7869184.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.4937145709991455,
|
|
"epoch": 3.5148902821316614,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004092391234503638,
|
|
"loss": 5.051,
|
|
"mean_token_accuracy": 0.18854818046092986,
|
|
"num_tokens": 7877362.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.492244243621826,
|
|
"epoch": 3.518808777429467,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040899761140630094,
|
|
"loss": 5.0571,
|
|
"mean_token_accuracy": 0.190338434278965,
|
|
"num_tokens": 7886782.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.559822845458984,
|
|
"epoch": 3.5227272727272725,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040875585988677985,
|
|
"loss": 4.9724,
|
|
"mean_token_accuracy": 0.20205166190862656,
|
|
"num_tokens": 7894811.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.526244068145752,
|
|
"epoch": 3.5266457680250785,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004085138693238497,
|
|
"loss": 5.1212,
|
|
"mean_token_accuracy": 0.18519109189510347,
|
|
"num_tokens": 7903955.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.5266457680250785,
|
|
"eval_entropy": 5.337847076645193,
|
|
"eval_loss": 5.7148871421813965,
|
|
"eval_mean_token_accuracy": 0.1654274752291367,
|
|
"eval_num_tokens": 7903955.0,
|
|
"eval_runtime": 2.1934,
|
|
"eval_samples_per_second": 1879.305,
|
|
"eval_steps_per_second": 235.255,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.560631513595581,
|
|
"epoch": 3.530564263322884,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004082716401499867,
|
|
"loss": 5.0927,
|
|
"mean_token_accuracy": 0.1909871518611908,
|
|
"num_tokens": 7913327.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.597675848007202,
|
|
"epoch": 3.5344827586206895,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040802917279809383,
|
|
"loss": 5.0784,
|
|
"mean_token_accuracy": 0.19403616935014725,
|
|
"num_tokens": 7921917.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.5416717529296875,
|
|
"epoch": 3.5384012539184955,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040778646770149953,
|
|
"loss": 5.126,
|
|
"mean_token_accuracy": 0.1836579531431198,
|
|
"num_tokens": 7931313.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.63011679649353,
|
|
"epoch": 3.542319749216301,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040754352529395716,
|
|
"loss": 5.1546,
|
|
"mean_token_accuracy": 0.18399414718151091,
|
|
"num_tokens": 7942250.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.534352588653564,
|
|
"epoch": 3.5462382445141065,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040730034600964415,
|
|
"loss": 5.111,
|
|
"mean_token_accuracy": 0.1848112389445305,
|
|
"num_tokens": 7950990.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.482068967819214,
|
|
"epoch": 3.5501567398119125,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004070569302831613,
|
|
"loss": 5.0005,
|
|
"mean_token_accuracy": 0.19020459204912185,
|
|
"num_tokens": 7959425.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.613459157943725,
|
|
"epoch": 3.554075235109718,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000406813278549532,
|
|
"loss": 5.0964,
|
|
"mean_token_accuracy": 0.18970496356487274,
|
|
"num_tokens": 7969007.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.59279727935791,
|
|
"epoch": 3.5579937304075235,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040656939124420144,
|
|
"loss": 5.0685,
|
|
"mean_token_accuracy": 0.1877979651093483,
|
|
"num_tokens": 7977848.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.486853456497192,
|
|
"epoch": 3.561912225705329,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004063252688030358,
|
|
"loss": 5.0978,
|
|
"mean_token_accuracy": 0.18053800612688065,
|
|
"num_tokens": 7986070.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.514698219299317,
|
|
"epoch": 3.5658307210031346,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004060809116623213,
|
|
"loss": 5.0058,
|
|
"mean_token_accuracy": 0.19714925289154053,
|
|
"num_tokens": 7994516.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.603819036483765,
|
|
"epoch": 3.5697492163009406,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000405836320258764,
|
|
"loss": 5.1371,
|
|
"mean_token_accuracy": 0.18588383495807648,
|
|
"num_tokens": 8003965.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.492586994171143,
|
|
"epoch": 3.573667711598746,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004055914950294882,
|
|
"loss": 5.0468,
|
|
"mean_token_accuracy": 0.19173711091279982,
|
|
"num_tokens": 8012277.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.515142154693604,
|
|
"epoch": 3.5775862068965516,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040534643641203645,
|
|
"loss": 5.0076,
|
|
"mean_token_accuracy": 0.19872245490550994,
|
|
"num_tokens": 8020921.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.47769947052002,
|
|
"epoch": 3.5815047021943576,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004051011448443681,
|
|
"loss": 5.0741,
|
|
"mean_token_accuracy": 0.19150445163249968,
|
|
"num_tokens": 8029329.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.516225004196167,
|
|
"epoch": 3.585423197492163,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000404855620764859,
|
|
"loss": 5.0732,
|
|
"mean_token_accuracy": 0.1835448697209358,
|
|
"num_tokens": 8038703.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.492142629623413,
|
|
"epoch": 3.5893416927899686,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004046098646123006,
|
|
"loss": 5.0908,
|
|
"mean_token_accuracy": 0.18822676986455916,
|
|
"num_tokens": 8047339.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.661739492416382,
|
|
"epoch": 3.593260188087774,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040436387682589876,
|
|
"loss": 5.2126,
|
|
"mean_token_accuracy": 0.18189732134342193,
|
|
"num_tokens": 8057331.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.5201507091522215,
|
|
"epoch": 3.5971786833855797,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004041176578452737,
|
|
"loss": 4.9966,
|
|
"mean_token_accuracy": 0.19925398081541063,
|
|
"num_tokens": 8065823.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.52663745880127,
|
|
"epoch": 3.6010971786833856,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004038712081104587,
|
|
"loss": 5.0631,
|
|
"mean_token_accuracy": 0.18425243496894836,
|
|
"num_tokens": 8074299.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.52188925743103,
|
|
"epoch": 3.605015673981191,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00040362452806189927,
|
|
"loss": 5.0317,
|
|
"mean_token_accuracy": 0.1951010763645172,
|
|
"num_tokens": 8082867.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.523621320724487,
|
|
"epoch": 3.6089341692789967,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004033776181404527,
|
|
"loss": 5.0079,
|
|
"mean_token_accuracy": 0.19654287099838258,
|
|
"num_tokens": 8090430.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.524655485153199,
|
|
"epoch": 3.6128526645768027,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040313047878738704,
|
|
"loss": 5.1028,
|
|
"mean_token_accuracy": 0.18588650673627855,
|
|
"num_tokens": 8099581.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.585165119171142,
|
|
"epoch": 3.616771159874608,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004028831104443805,
|
|
"loss": 5.1075,
|
|
"mean_token_accuracy": 0.18863822370767594,
|
|
"num_tokens": 8108689.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.513268947601318,
|
|
"epoch": 3.6206896551724137,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 5.0656,
|
|
"mean_token_accuracy": 0.1890157327055931,
|
|
"num_tokens": 8117556.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.580118560791016,
|
|
"epoch": 3.6246081504702197,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040238768855730214,
|
|
"loss": 5.1701,
|
|
"mean_token_accuracy": 0.18975308984518052,
|
|
"num_tokens": 8126495.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.516307258605957,
|
|
"epoch": 3.628526645768025,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040213963589862963,
|
|
"loss": 5.0259,
|
|
"mean_token_accuracy": 0.19353499710559846,
|
|
"num_tokens": 8134961.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.471398830413818,
|
|
"epoch": 3.6324451410658307,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004018913560208131,
|
|
"loss": 5.073,
|
|
"mean_token_accuracy": 0.18923893719911575,
|
|
"num_tokens": 8143962.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.580399036407471,
|
|
"epoch": 3.6363636363636362,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004016428493675689,
|
|
"loss": 5.1153,
|
|
"mean_token_accuracy": 0.1842251241207123,
|
|
"num_tokens": 8152392.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.441334772109985,
|
|
"epoch": 3.6402821316614418,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004013941163830187,
|
|
"loss": 4.9463,
|
|
"mean_token_accuracy": 0.1974005162715912,
|
|
"num_tokens": 8161329.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.555213975906372,
|
|
"epoch": 3.6442006269592477,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004011451575116887,
|
|
"loss": 5.1142,
|
|
"mean_token_accuracy": 0.1885092154145241,
|
|
"num_tokens": 8170201.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.458886575698853,
|
|
"epoch": 3.6481191222570533,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004008959731985087,
|
|
"loss": 4.9644,
|
|
"mean_token_accuracy": 0.19162468165159224,
|
|
"num_tokens": 8178616.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.495558023452759,
|
|
"epoch": 3.652037617554859,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040064656388881157,
|
|
"loss": 5.0585,
|
|
"mean_token_accuracy": 0.1835464805364609,
|
|
"num_tokens": 8187842.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.615549516677857,
|
|
"epoch": 3.6559561128526648,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004003969300283321,
|
|
"loss": 5.1245,
|
|
"mean_token_accuracy": 0.18563947081565857,
|
|
"num_tokens": 8197063.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.506508731842041,
|
|
"epoch": 3.6598746081504703,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040014707206320653,
|
|
"loss": 4.9651,
|
|
"mean_token_accuracy": 0.19691204726696016,
|
|
"num_tokens": 8205086.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.601326370239258,
|
|
"epoch": 3.663793103448276,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039989699043997153,
|
|
"loss": 5.2003,
|
|
"mean_token_accuracy": 0.1806623011827469,
|
|
"num_tokens": 8213453.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.494664430618286,
|
|
"epoch": 3.6677115987460818,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039964668560556356,
|
|
"loss": 5.1127,
|
|
"mean_token_accuracy": 0.1832705482840538,
|
|
"num_tokens": 8222345.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.571859884262085,
|
|
"epoch": 3.6716300940438873,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039939615800731784,
|
|
"loss": 5.1035,
|
|
"mean_token_accuracy": 0.18808005303144454,
|
|
"num_tokens": 8230207.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.582139253616333,
|
|
"epoch": 3.675548589341693,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039914540809296795,
|
|
"loss": 5.1502,
|
|
"mean_token_accuracy": 0.1848934546113014,
|
|
"num_tokens": 8238919.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.5853917598724365,
|
|
"epoch": 3.6794670846394983,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003988944363106445,
|
|
"loss": 5.0482,
|
|
"mean_token_accuracy": 0.18395280092954636,
|
|
"num_tokens": 8248833.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.590873670578003,
|
|
"epoch": 3.683385579937304,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003986432431088749,
|
|
"loss": 5.0767,
|
|
"mean_token_accuracy": 0.19430365264415742,
|
|
"num_tokens": 8258011.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.59118242263794,
|
|
"epoch": 3.68730407523511,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000398391828936582,
|
|
"loss": 5.1879,
|
|
"mean_token_accuracy": 0.18539660423994064,
|
|
"num_tokens": 8267324.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.577607202529907,
|
|
"epoch": 3.6912225705329154,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003981401942430838,
|
|
"loss": 5.0889,
|
|
"mean_token_accuracy": 0.18853191733360292,
|
|
"num_tokens": 8276312.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.532922458648682,
|
|
"epoch": 3.695141065830721,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039788833947809217,
|
|
"loss": 5.0494,
|
|
"mean_token_accuracy": 0.1891567513346672,
|
|
"num_tokens": 8285238.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.463539743423462,
|
|
"epoch": 3.699059561128527,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003976362650917125,
|
|
"loss": 5.0684,
|
|
"mean_token_accuracy": 0.19253510981798172,
|
|
"num_tokens": 8293842.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.569275283813477,
|
|
"epoch": 3.7029780564263324,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039738397153444264,
|
|
"loss": 5.0588,
|
|
"mean_token_accuracy": 0.1946437120437622,
|
|
"num_tokens": 8302962.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.564457893371582,
|
|
"epoch": 3.706896551724138,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003971314592571719,
|
|
"loss": 5.1428,
|
|
"mean_token_accuracy": 0.18560621440410613,
|
|
"num_tokens": 8311570.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.628694915771485,
|
|
"epoch": 3.7108150470219434,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003968787287111809,
|
|
"loss": 5.1094,
|
|
"mean_token_accuracy": 0.1871411085128784,
|
|
"num_tokens": 8319998.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.595128917694092,
|
|
"epoch": 3.714733542319749,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039662578034814,
|
|
"loss": 5.0694,
|
|
"mean_token_accuracy": 0.18592819422483445,
|
|
"num_tokens": 8328767.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.538923454284668,
|
|
"epoch": 3.718652037617555,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039637261462010886,
|
|
"loss": 5.1529,
|
|
"mean_token_accuracy": 0.18640181571245193,
|
|
"num_tokens": 8337459.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.504244756698609,
|
|
"epoch": 3.7225705329153604,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003961192319795358,
|
|
"loss": 5.004,
|
|
"mean_token_accuracy": 0.19079648554325104,
|
|
"num_tokens": 8346349.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.586766242980957,
|
|
"epoch": 3.726489028213166,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003958656328792565,
|
|
"loss": 5.1173,
|
|
"mean_token_accuracy": 0.1872461199760437,
|
|
"num_tokens": 8355069.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.533196830749512,
|
|
"epoch": 3.730407523510972,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039561181777249396,
|
|
"loss": 4.9997,
|
|
"mean_token_accuracy": 0.19198624789714813,
|
|
"num_tokens": 8363295.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.578617191314697,
|
|
"epoch": 3.7343260188087775,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039535778711285676,
|
|
"loss": 5.141,
|
|
"mean_token_accuracy": 0.18830593377351762,
|
|
"num_tokens": 8372260.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.589736843109131,
|
|
"epoch": 3.738244514106583,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003951035413543388,
|
|
"loss": 4.9942,
|
|
"mean_token_accuracy": 0.1931956574320793,
|
|
"num_tokens": 8380776.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.637099123001098,
|
|
"epoch": 3.742163009404389,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039484908095131874,
|
|
"loss": 5.1688,
|
|
"mean_token_accuracy": 0.18469733893871307,
|
|
"num_tokens": 8389223.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.512900257110596,
|
|
"epoch": 3.7460815047021945,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003945944063585582,
|
|
"loss": 4.9872,
|
|
"mean_token_accuracy": 0.19673277884721757,
|
|
"num_tokens": 8397688.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.5134625911712645,
|
|
"epoch": 3.75,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039433951803120225,
|
|
"loss": 5.0932,
|
|
"mean_token_accuracy": 0.18970392942428588,
|
|
"num_tokens": 8407061.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.493066835403442,
|
|
"epoch": 3.7539184952978055,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039408441642477764,
|
|
"loss": 5.0871,
|
|
"mean_token_accuracy": 0.18374049067497253,
|
|
"num_tokens": 8415473.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.5036272525787355,
|
|
"epoch": 3.757836990595611,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003938291019951922,
|
|
"loss": 5.079,
|
|
"mean_token_accuracy": 0.1960655003786087,
|
|
"num_tokens": 8424439.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.57970085144043,
|
|
"epoch": 3.761755485893417,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003935735751987344,
|
|
"loss": 5.1267,
|
|
"mean_token_accuracy": 0.18465493321418763,
|
|
"num_tokens": 8433283.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.527460527420044,
|
|
"epoch": 3.7656739811912225,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039331783649207175,
|
|
"loss": 5.0048,
|
|
"mean_token_accuracy": 0.19909376353025438,
|
|
"num_tokens": 8441639.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.609852027893067,
|
|
"epoch": 3.769592476489028,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039306188633225097,
|
|
"loss": 5.2305,
|
|
"mean_token_accuracy": 0.17572411745786667,
|
|
"num_tokens": 8451183.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.581537914276123,
|
|
"epoch": 3.773510971786834,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003928057251766965,
|
|
"loss": 5.0765,
|
|
"mean_token_accuracy": 0.1886788100004196,
|
|
"num_tokens": 8459729.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.589933443069458,
|
|
"epoch": 3.7774294670846396,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039254935348320984,
|
|
"loss": 5.1319,
|
|
"mean_token_accuracy": 0.18040645718574524,
|
|
"num_tokens": 8468699.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.5669169425964355,
|
|
"epoch": 3.781347962382445,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039229277170996885,
|
|
"loss": 5.1271,
|
|
"mean_token_accuracy": 0.18908928483724594,
|
|
"num_tokens": 8477824.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.570567083358765,
|
|
"epoch": 3.785266457680251,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003920359803155266,
|
|
"loss": 5.0826,
|
|
"mean_token_accuracy": 0.1925947368144989,
|
|
"num_tokens": 8486966.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.558492851257324,
|
|
"epoch": 3.7891849529780566,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039177897975881115,
|
|
"loss": 5.0875,
|
|
"mean_token_accuracy": 0.18648419827222823,
|
|
"num_tokens": 8495656.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.548500919342041,
|
|
"epoch": 3.793103448275862,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003915217704991239,
|
|
"loss": 5.0354,
|
|
"mean_token_accuracy": 0.1990212008357048,
|
|
"num_tokens": 8504246.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.614151477813721,
|
|
"epoch": 3.7970219435736676,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003912643529961397,
|
|
"loss": 5.1981,
|
|
"mean_token_accuracy": 0.17596834003925324,
|
|
"num_tokens": 8512937.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.6028480052948,
|
|
"epoch": 3.800940438871473,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003910067277099053,
|
|
"loss": 5.0582,
|
|
"mean_token_accuracy": 0.1945752277970314,
|
|
"num_tokens": 8521777.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.539871025085449,
|
|
"epoch": 3.804858934169279,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 5.0819,
|
|
"mean_token_accuracy": 0.18999679237604142,
|
|
"num_tokens": 8530323.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.5041773319244385,
|
|
"epoch": 3.8087774294670846,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003904908556297293,
|
|
"loss": 5.1416,
|
|
"mean_token_accuracy": 0.18803969770669937,
|
|
"num_tokens": 8539221.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.568723535537719,
|
|
"epoch": 3.81269592476489,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003902326097577345,
|
|
"loss": 5.0317,
|
|
"mean_token_accuracy": 0.1918994203209877,
|
|
"num_tokens": 8548537.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.519817638397217,
|
|
"epoch": 3.816614420062696,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038997415794638206,
|
|
"loss": 5.0278,
|
|
"mean_token_accuracy": 0.19718809127807618,
|
|
"num_tokens": 8557893.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.533332347869873,
|
|
"epoch": 3.8205329153605017,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003897155006575672,
|
|
"loss": 5.0361,
|
|
"mean_token_accuracy": 0.190166237950325,
|
|
"num_tokens": 8566297.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.651578378677368,
|
|
"epoch": 3.824451410658307,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038945663835355247,
|
|
"loss": 5.1957,
|
|
"mean_token_accuracy": 0.18009587079286576,
|
|
"num_tokens": 8575898.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.575738382339478,
|
|
"epoch": 3.8283699059561127,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038919757149696665,
|
|
"loss": 5.033,
|
|
"mean_token_accuracy": 0.1972601130604744,
|
|
"num_tokens": 8583875.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.415046453475952,
|
|
"epoch": 3.8322884012539182,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038893830055080437,
|
|
"loss": 4.9701,
|
|
"mean_token_accuracy": 0.1941005438566208,
|
|
"num_tokens": 8592428.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.563708877563476,
|
|
"epoch": 3.836206896551724,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003886788259784248,
|
|
"loss": 5.0665,
|
|
"mean_token_accuracy": 0.18945758044719696,
|
|
"num_tokens": 8600840.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.571075820922852,
|
|
"epoch": 3.8401253918495297,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038841914824355093,
|
|
"loss": 5.0621,
|
|
"mean_token_accuracy": 0.19667091965675354,
|
|
"num_tokens": 8609596.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.5646764755249025,
|
|
"epoch": 3.8440438871473352,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038815926781026914,
|
|
"loss": 5.0619,
|
|
"mean_token_accuracy": 0.18016432076692582,
|
|
"num_tokens": 8618705.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.484356880187988,
|
|
"epoch": 3.847962382445141,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003878991851430279,
|
|
"loss": 5.0296,
|
|
"mean_token_accuracy": 0.18855099231004716,
|
|
"num_tokens": 8627553.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.515388107299804,
|
|
"epoch": 3.8518808777429467,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003876389007066371,
|
|
"loss": 5.1096,
|
|
"mean_token_accuracy": 0.18375999480485916,
|
|
"num_tokens": 8637669.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.573273229598999,
|
|
"epoch": 3.8557993730407523,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003873784149662672,
|
|
"loss": 4.9465,
|
|
"mean_token_accuracy": 0.19470633715391159,
|
|
"num_tokens": 8645545.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.480087518692017,
|
|
"epoch": 3.8597178683385582,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003871177283874484,
|
|
"loss": 5.0718,
|
|
"mean_token_accuracy": 0.18605099320411683,
|
|
"num_tokens": 8654190.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.5290416240692135,
|
|
"epoch": 3.8636363636363638,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038685684143606995,
|
|
"loss": 5.081,
|
|
"mean_token_accuracy": 0.19451049715280533,
|
|
"num_tokens": 8662501.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.643653297424317,
|
|
"epoch": 3.8675548589341693,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003865957545783791,
|
|
"loss": 5.1675,
|
|
"mean_token_accuracy": 0.17904634773731232,
|
|
"num_tokens": 8672016.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.518256425857544,
|
|
"epoch": 3.871473354231975,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038633446828098046,
|
|
"loss": 5.0163,
|
|
"mean_token_accuracy": 0.19363693296909332,
|
|
"num_tokens": 8681127.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.543700218200684,
|
|
"epoch": 3.8753918495297803,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000386072983010835,
|
|
"loss": 5.1465,
|
|
"mean_token_accuracy": 0.18741605281829835,
|
|
"num_tokens": 8690719.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.562393045425415,
|
|
"epoch": 3.8793103448275863,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038581129923525914,
|
|
"loss": 5.1004,
|
|
"mean_token_accuracy": 0.19265996515750886,
|
|
"num_tokens": 8699146.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.641261243820191,
|
|
"epoch": 3.883228840125392,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038554941742192445,
|
|
"loss": 5.1478,
|
|
"mean_token_accuracy": 0.18488254249095917,
|
|
"num_tokens": 8707900.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.531866502761841,
|
|
"epoch": 3.8871473354231973,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003852873380388561,
|
|
"loss": 5.0528,
|
|
"mean_token_accuracy": 0.18934565484523774,
|
|
"num_tokens": 8716430.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.601463890075683,
|
|
"epoch": 3.8910658307210033,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003850250615544323,
|
|
"loss": 5.0652,
|
|
"mean_token_accuracy": 0.1889369696378708,
|
|
"num_tokens": 8725077.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.595002317428589,
|
|
"epoch": 3.894984326018809,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038476258843738386,
|
|
"loss": 5.1036,
|
|
"mean_token_accuracy": 0.18661195039749146,
|
|
"num_tokens": 8734732.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.519152593612671,
|
|
"epoch": 3.8989028213166144,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038449991915679273,
|
|
"loss": 5.0297,
|
|
"mean_token_accuracy": 0.1984531745314598,
|
|
"num_tokens": 8743265.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.504469490051269,
|
|
"epoch": 3.9028213166144203,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003842370541820915,
|
|
"loss": 5.0325,
|
|
"mean_token_accuracy": 0.18722153306007386,
|
|
"num_tokens": 8752652.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.635008716583252,
|
|
"epoch": 3.906739811912226,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038397399398306243,
|
|
"loss": 5.0704,
|
|
"mean_token_accuracy": 0.1879400297999382,
|
|
"num_tokens": 8762181.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.6260457038879395,
|
|
"epoch": 3.9106583072100314,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038371073902983684,
|
|
"loss": 5.1386,
|
|
"mean_token_accuracy": 0.1891830712556839,
|
|
"num_tokens": 8770590.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.616936063766479,
|
|
"epoch": 3.914576802507837,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000383447289792894,
|
|
"loss": 5.214,
|
|
"mean_token_accuracy": 0.18705283254384994,
|
|
"num_tokens": 8779281.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.51376953125,
|
|
"epoch": 3.9184952978056424,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038318364674306036,
|
|
"loss": 5.0836,
|
|
"mean_token_accuracy": 0.18901561051607133,
|
|
"num_tokens": 8788353.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 3.9184952978056424,
|
|
"eval_entropy": 5.371145749276923,
|
|
"eval_loss": 5.673300266265869,
|
|
"eval_mean_token_accuracy": 0.16773399917134482,
|
|
"eval_num_tokens": 8788353.0,
|
|
"eval_runtime": 2.3643,
|
|
"eval_samples_per_second": 1743.458,
|
|
"eval_steps_per_second": 218.25,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.5987523078918455,
|
|
"epoch": 3.9224137931034484,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038291981035150883,
|
|
"loss": 5.1026,
|
|
"mean_token_accuracy": 0.18235832452774048,
|
|
"num_tokens": 8797224.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.511776447296143,
|
|
"epoch": 3.926332288401254,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003826557810897579,
|
|
"loss": 5.1062,
|
|
"mean_token_accuracy": 0.1880456581711769,
|
|
"num_tokens": 8806573.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.511373043060303,
|
|
"epoch": 3.9302507836990594,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003823915594296705,
|
|
"loss": 5.0391,
|
|
"mean_token_accuracy": 0.18769147396087646,
|
|
"num_tokens": 8815826.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.534104251861573,
|
|
"epoch": 3.9341692789968654,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003821271458434538,
|
|
"loss": 5.1066,
|
|
"mean_token_accuracy": 0.19072123914957045,
|
|
"num_tokens": 8823845.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.59644660949707,
|
|
"epoch": 3.938087774294671,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00038186254080365756,
|
|
"loss": 5.1044,
|
|
"mean_token_accuracy": 0.1807522341609001,
|
|
"num_tokens": 8832727.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.612049150466919,
|
|
"epoch": 3.9420062695924765,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000381597744783174,
|
|
"loss": 5.0177,
|
|
"mean_token_accuracy": 0.1927120491862297,
|
|
"num_tokens": 8841165.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.492348527908325,
|
|
"epoch": 3.945924764890282,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038133275825523645,
|
|
"loss": 5.1097,
|
|
"mean_token_accuracy": 0.1859942004084587,
|
|
"num_tokens": 8849838.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.60238881111145,
|
|
"epoch": 3.9498432601880875,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038106758169341887,
|
|
"loss": 5.1255,
|
|
"mean_token_accuracy": 0.18698869794607162,
|
|
"num_tokens": 8858938.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.621122694015503,
|
|
"epoch": 3.9537617554858935,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003808022155716348,
|
|
"loss": 5.1391,
|
|
"mean_token_accuracy": 0.18632098138332367,
|
|
"num_tokens": 8867496.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.508530902862549,
|
|
"epoch": 3.957680250783699,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003805366603641364,
|
|
"loss": 5.0725,
|
|
"mean_token_accuracy": 0.1901547446846962,
|
|
"num_tokens": 8877079.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.588417387008667,
|
|
"epoch": 3.9615987460815045,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038027091654551406,
|
|
"loss": 5.1549,
|
|
"mean_token_accuracy": 0.1827967122197151,
|
|
"num_tokens": 8885403.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.608737802505493,
|
|
"epoch": 3.9655172413793105,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038000498459069487,
|
|
"loss": 5.0756,
|
|
"mean_token_accuracy": 0.1955476775765419,
|
|
"num_tokens": 8893124.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.548772954940796,
|
|
"epoch": 3.969435736677116,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003797388649749426,
|
|
"loss": 5.0182,
|
|
"mean_token_accuracy": 0.20016582310199738,
|
|
"num_tokens": 8901532.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.597880125045776,
|
|
"epoch": 3.9733542319749215,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003794725581738559,
|
|
"loss": 5.1393,
|
|
"mean_token_accuracy": 0.18231671899557114,
|
|
"num_tokens": 8910109.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.642048597335815,
|
|
"epoch": 3.9772727272727275,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00037920606466336834,
|
|
"loss": 5.1433,
|
|
"mean_token_accuracy": 0.18488133251667022,
|
|
"num_tokens": 8918513.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.495293474197387,
|
|
"epoch": 3.981191222570533,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000378939384919747,
|
|
"loss": 5.0965,
|
|
"mean_token_accuracy": 0.18819395303726197,
|
|
"num_tokens": 8926853.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.597048616409301,
|
|
"epoch": 3.9851097178683386,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003786725194195918,
|
|
"loss": 5.1327,
|
|
"mean_token_accuracy": 0.1761031836271286,
|
|
"num_tokens": 8936393.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.595620107650757,
|
|
"epoch": 3.989028213166144,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 5.142,
|
|
"mean_token_accuracy": 0.1907366320490837,
|
|
"num_tokens": 8945224.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.473467540740967,
|
|
"epoch": 3.9929467084639496,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037813823305773883,
|
|
"loss": 4.9863,
|
|
"mean_token_accuracy": 0.1985485374927521,
|
|
"num_tokens": 8954147.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.4523036003112795,
|
|
"epoch": 3.9968652037617556,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003778708131508974,
|
|
"loss": 4.9815,
|
|
"mean_token_accuracy": 0.1961939662694931,
|
|
"num_tokens": 8962454.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.600477266311645,
|
|
"epoch": 4.0007836990595615,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00037760320939723307,
|
|
"loss": 5.014,
|
|
"mean_token_accuracy": 0.19382012784481048,
|
|
"num_tokens": 8971621.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.506287527084351,
|
|
"epoch": 4.004702194357367,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037733542227499727,
|
|
"loss": 4.767,
|
|
"mean_token_accuracy": 0.20831096321344375,
|
|
"num_tokens": 8980158.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.50818133354187,
|
|
"epoch": 4.008620689655173,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037706745226276893,
|
|
"loss": 4.8147,
|
|
"mean_token_accuracy": 0.20869106203317642,
|
|
"num_tokens": 8988293.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.491833209991455,
|
|
"epoch": 4.012539184952978,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003767992998394539,
|
|
"loss": 4.7366,
|
|
"mean_token_accuracy": 0.21957777589559554,
|
|
"num_tokens": 8995907.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.456087923049926,
|
|
"epoch": 4.016457680250784,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003765309654842842,
|
|
"loss": 4.8265,
|
|
"mean_token_accuracy": 0.2065334603190422,
|
|
"num_tokens": 9004394.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.455752944946289,
|
|
"epoch": 4.02037617554859,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003762624496768166,
|
|
"loss": 4.8381,
|
|
"mean_token_accuracy": 0.20486140251159668,
|
|
"num_tokens": 9012575.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.514487171173096,
|
|
"epoch": 4.024294670846395,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003759937528969325,
|
|
"loss": 4.7838,
|
|
"mean_token_accuracy": 0.2132658213376999,
|
|
"num_tokens": 9021585.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.589247798919677,
|
|
"epoch": 4.028213166144201,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00037572487562483666,
|
|
"loss": 4.8522,
|
|
"mean_token_accuracy": 0.20767498910427093,
|
|
"num_tokens": 9030638.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.550300550460816,
|
|
"epoch": 4.032131661442007,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037545581834105623,
|
|
"loss": 4.8844,
|
|
"mean_token_accuracy": 0.200663785636425,
|
|
"num_tokens": 9039193.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.525839376449585,
|
|
"epoch": 4.036050156739812,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00037518658152644035,
|
|
"loss": 4.8218,
|
|
"mean_token_accuracy": 0.21329268515110017,
|
|
"num_tokens": 9047969.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.493017768859863,
|
|
"epoch": 4.039968652037618,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003749171656621588,
|
|
"loss": 4.8454,
|
|
"mean_token_accuracy": 0.20602474212646485,
|
|
"num_tokens": 9056613.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.541411638259888,
|
|
"epoch": 4.043887147335423,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003746475712297014,
|
|
"loss": 4.8282,
|
|
"mean_token_accuracy": 0.20796336233615875,
|
|
"num_tokens": 9064691.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.468442392349243,
|
|
"epoch": 4.047805642633229,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003743777987108771,
|
|
"loss": 4.8505,
|
|
"mean_token_accuracy": 0.20266541838645935,
|
|
"num_tokens": 9073363.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.486063051223755,
|
|
"epoch": 4.051724137931035,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003741078485878132,
|
|
"loss": 4.7751,
|
|
"mean_token_accuracy": 0.20718557089567186,
|
|
"num_tokens": 9082153.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.4818892002105715,
|
|
"epoch": 4.05564263322884,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003738377213429542,
|
|
"loss": 4.7818,
|
|
"mean_token_accuracy": 0.20577536821365355,
|
|
"num_tokens": 9091069.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.5446723937988285,
|
|
"epoch": 4.059561128526646,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003735674174590614,
|
|
"loss": 4.8815,
|
|
"mean_token_accuracy": 0.2012221097946167,
|
|
"num_tokens": 9099823.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.553747606277466,
|
|
"epoch": 4.063479623824452,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00037329693741921166,
|
|
"loss": 4.9143,
|
|
"mean_token_accuracy": 0.19897791594266892,
|
|
"num_tokens": 9108508.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.6101867198944095,
|
|
"epoch": 4.067398119122257,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003730262817067967,
|
|
"loss": 4.9158,
|
|
"mean_token_accuracy": 0.20250056087970733,
|
|
"num_tokens": 9116736.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.433481597900391,
|
|
"epoch": 4.071316614420063,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003727554508055221,
|
|
"loss": 4.8197,
|
|
"mean_token_accuracy": 0.2074613630771637,
|
|
"num_tokens": 9125023.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.473941946029663,
|
|
"epoch": 4.075235109717869,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003724844451994066,
|
|
"loss": 4.8823,
|
|
"mean_token_accuracy": 0.20830427706241608,
|
|
"num_tokens": 9135054.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.5989556312561035,
|
|
"epoch": 4.079153605015674,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037221326537278113,
|
|
"loss": 4.8527,
|
|
"mean_token_accuracy": 0.21032673567533494,
|
|
"num_tokens": 9143651.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.539326095581055,
|
|
"epoch": 4.08307210031348,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037194191181028806,
|
|
"loss": 4.8278,
|
|
"mean_token_accuracy": 0.2080790415406227,
|
|
"num_tokens": 9152423.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.450719833374023,
|
|
"epoch": 4.086990595611285,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037167038499688024,
|
|
"loss": 4.7683,
|
|
"mean_token_accuracy": 0.20537828356027604,
|
|
"num_tokens": 9160698.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.522712564468383,
|
|
"epoch": 4.090909090909091,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037139868541782,
|
|
"loss": 4.844,
|
|
"mean_token_accuracy": 0.20190785974264144,
|
|
"num_tokens": 9169323.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.509297704696655,
|
|
"epoch": 4.094827586206897,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003711268135586787,
|
|
"loss": 4.8888,
|
|
"mean_token_accuracy": 0.20097094923257827,
|
|
"num_tokens": 9178875.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.471201229095459,
|
|
"epoch": 4.098746081504702,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037085476990533523,
|
|
"loss": 4.775,
|
|
"mean_token_accuracy": 0.2124708339571953,
|
|
"num_tokens": 9187304.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.5013893127441404,
|
|
"epoch": 4.102664576802508,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003705825549439759,
|
|
"loss": 4.9145,
|
|
"mean_token_accuracy": 0.19863292574882507,
|
|
"num_tokens": 9196893.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.58682074546814,
|
|
"epoch": 4.106583072100314,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037031016916109295,
|
|
"loss": 4.9614,
|
|
"mean_token_accuracy": 0.1980082467198372,
|
|
"num_tokens": 9205925.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.608582353591919,
|
|
"epoch": 4.110501567398119,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037003761304348374,
|
|
"loss": 4.9023,
|
|
"mean_token_accuracy": 0.2071164220571518,
|
|
"num_tokens": 9215317.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.4616536617279055,
|
|
"epoch": 4.114420062695925,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036976488707825056,
|
|
"loss": 4.765,
|
|
"mean_token_accuracy": 0.2127097874879837,
|
|
"num_tokens": 9223648.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.506093883514405,
|
|
"epoch": 4.118338557993731,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00036949199175279876,
|
|
"loss": 4.9082,
|
|
"mean_token_accuracy": 0.19973835945129395,
|
|
"num_tokens": 9233178.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.514060592651367,
|
|
"epoch": 4.122257053291536,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003692189275548364,
|
|
"loss": 4.8152,
|
|
"mean_token_accuracy": 0.20875712484121323,
|
|
"num_tokens": 9241931.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.511327648162842,
|
|
"epoch": 4.126175548589342,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003689456949723737,
|
|
"loss": 4.9369,
|
|
"mean_token_accuracy": 0.20098417848348618,
|
|
"num_tokens": 9251287.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.461143255233765,
|
|
"epoch": 4.130094043887147,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003686722944937215,
|
|
"loss": 4.7707,
|
|
"mean_token_accuracy": 0.20872875303030014,
|
|
"num_tokens": 9259372.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.450579309463501,
|
|
"epoch": 4.134012539184953,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036839872660749084,
|
|
"loss": 4.8198,
|
|
"mean_token_accuracy": 0.20989379584789275,
|
|
"num_tokens": 9268425.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.444522428512573,
|
|
"epoch": 4.137931034482759,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003681249918025918,
|
|
"loss": 4.8278,
|
|
"mean_token_accuracy": 0.20376764088869095,
|
|
"num_tokens": 9277008.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.4921516418457035,
|
|
"epoch": 4.141849529780564,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00036785109056823297,
|
|
"loss": 4.908,
|
|
"mean_token_accuracy": 0.19986599832773208,
|
|
"num_tokens": 9286937.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.489944410324097,
|
|
"epoch": 4.14576802507837,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036757702339392,
|
|
"loss": 4.8062,
|
|
"mean_token_accuracy": 0.2123652771115303,
|
|
"num_tokens": 9294970.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.5329005241394045,
|
|
"epoch": 4.149686520376176,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036730279076945574,
|
|
"loss": 4.8309,
|
|
"mean_token_accuracy": 0.2112215504050255,
|
|
"num_tokens": 9303704.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.420260286331176,
|
|
"epoch": 4.153605015673981,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003670283931849382,
|
|
"loss": 4.7481,
|
|
"mean_token_accuracy": 0.21377620548009874,
|
|
"num_tokens": 9311867.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.433056879043579,
|
|
"epoch": 4.157523510971787,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00036675383113076023,
|
|
"loss": 4.8113,
|
|
"mean_token_accuracy": 0.20675826370716094,
|
|
"num_tokens": 9320052.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.500449705123901,
|
|
"epoch": 4.161442006269592,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00036647910509760896,
|
|
"loss": 4.832,
|
|
"mean_token_accuracy": 0.21465274095535278,
|
|
"num_tokens": 9329126.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.519782638549804,
|
|
"epoch": 4.165360501567398,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00036620421557646414,
|
|
"loss": 4.8,
|
|
"mean_token_accuracy": 0.21130249351263047,
|
|
"num_tokens": 9337734.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.569190311431885,
|
|
"epoch": 4.169278996865204,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000365929163058598,
|
|
"loss": 4.9434,
|
|
"mean_token_accuracy": 0.19796280711889266,
|
|
"num_tokens": 9347138.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.456249141693116,
|
|
"epoch": 4.173197492163009,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 4.7282,
|
|
"mean_token_accuracy": 0.2221844345331192,
|
|
"num_tokens": 9355776.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.484549808502197,
|
|
"epoch": 4.177115987460815,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003653785709992462,
|
|
"loss": 4.8476,
|
|
"mean_token_accuracy": 0.213798888027668,
|
|
"num_tokens": 9364707.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.4504444122314455,
|
|
"epoch": 4.181034482758621,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00036510303244175775,
|
|
"loss": 4.7747,
|
|
"mean_token_accuracy": 0.21001851707696914,
|
|
"num_tokens": 9373599.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.478919458389282,
|
|
"epoch": 4.184952978056426,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036482733285554073,
|
|
"loss": 4.8736,
|
|
"mean_token_accuracy": 0.20725277364253997,
|
|
"num_tokens": 9382554.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.547554063796997,
|
|
"epoch": 4.188871473354232,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036455147273331507,
|
|
"loss": 4.8766,
|
|
"mean_token_accuracy": 0.2018249213695526,
|
|
"num_tokens": 9390654.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.499206447601319,
|
|
"epoch": 4.192789968652038,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003642754525680876,
|
|
"loss": 4.8086,
|
|
"mean_token_accuracy": 0.2123773366212845,
|
|
"num_tokens": 9399334.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.46196141242981,
|
|
"epoch": 4.196708463949843,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00036399927285315104,
|
|
"loss": 4.8101,
|
|
"mean_token_accuracy": 0.20545402467250823,
|
|
"num_tokens": 9408255.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.437937307357788,
|
|
"epoch": 4.200626959247649,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003637229340820834,
|
|
"loss": 4.7773,
|
|
"mean_token_accuracy": 0.212043459713459,
|
|
"num_tokens": 9416758.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.44543948173523,
|
|
"epoch": 4.204545454545454,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003634464367487467,
|
|
"loss": 4.8552,
|
|
"mean_token_accuracy": 0.20784655064344407,
|
|
"num_tokens": 9426087.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.487284469604492,
|
|
"epoch": 4.20846394984326,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003631697813472867,
|
|
"loss": 4.8353,
|
|
"mean_token_accuracy": 0.21206834316253662,
|
|
"num_tokens": 9434552.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.477179145812988,
|
|
"epoch": 4.212382445141066,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036289296837213136,
|
|
"loss": 4.7976,
|
|
"mean_token_accuracy": 0.21521666198968886,
|
|
"num_tokens": 9444073.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.471901893615723,
|
|
"epoch": 4.216300940438871,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036261599831799034,
|
|
"loss": 4.8715,
|
|
"mean_token_accuracy": 0.20849257111549377,
|
|
"num_tokens": 9452994.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.438826084136963,
|
|
"epoch": 4.220219435736677,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003623388716798541,
|
|
"loss": 4.8402,
|
|
"mean_token_accuracy": 0.20498404949903487,
|
|
"num_tokens": 9461623.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.553812265396118,
|
|
"epoch": 4.224137931034483,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00036206158895299277,
|
|
"loss": 4.8913,
|
|
"mean_token_accuracy": 0.19901519268751144,
|
|
"num_tokens": 9470414.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.505804967880249,
|
|
"epoch": 4.228056426332288,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036178415063295574,
|
|
"loss": 4.8868,
|
|
"mean_token_accuracy": 0.21280975490808487,
|
|
"num_tokens": 9478668.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.462782764434815,
|
|
"epoch": 4.231974921630094,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036150655721557,
|
|
"loss": 4.8294,
|
|
"mean_token_accuracy": 0.2057174801826477,
|
|
"num_tokens": 9487906.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.477683734893799,
|
|
"epoch": 4.2358934169279,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003612288091969402,
|
|
"loss": 4.859,
|
|
"mean_token_accuracy": 0.20768630802631377,
|
|
"num_tokens": 9496849.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.463692045211792,
|
|
"epoch": 4.239811912225705,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036095090707344697,
|
|
"loss": 4.8534,
|
|
"mean_token_accuracy": 0.20733812153339387,
|
|
"num_tokens": 9505243.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.480135583877564,
|
|
"epoch": 4.243730407523511,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036067285134174654,
|
|
"loss": 4.7955,
|
|
"mean_token_accuracy": 0.20832840353250504,
|
|
"num_tokens": 9513822.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.492481851577759,
|
|
"epoch": 4.247648902821316,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003603946424987696,
|
|
"loss": 4.8598,
|
|
"mean_token_accuracy": 0.20426886677742004,
|
|
"num_tokens": 9523150.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.578524017333985,
|
|
"epoch": 4.251567398119122,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036011628104172045,
|
|
"loss": 4.9334,
|
|
"mean_token_accuracy": 0.19984120577573777,
|
|
"num_tokens": 9532126.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.544974231719971,
|
|
"epoch": 4.255485893416928,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035983776746807613,
|
|
"loss": 4.8521,
|
|
"mean_token_accuracy": 0.2010730728507042,
|
|
"num_tokens": 9541064.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.50416784286499,
|
|
"epoch": 4.259404388714733,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003595591022755855,
|
|
"loss": 4.8541,
|
|
"mean_token_accuracy": 0.207574699819088,
|
|
"num_tokens": 9549021.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.497020149230957,
|
|
"epoch": 4.263322884012539,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003592802859622684,
|
|
"loss": 4.8603,
|
|
"mean_token_accuracy": 0.20584329217672348,
|
|
"num_tokens": 9557486.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.461274242401123,
|
|
"epoch": 4.267241379310345,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000359001319026415,
|
|
"loss": 4.8011,
|
|
"mean_token_accuracy": 0.20425123274326323,
|
|
"num_tokens": 9566100.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.487164545059204,
|
|
"epoch": 4.27115987460815,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035872220196658426,
|
|
"loss": 4.9083,
|
|
"mean_token_accuracy": 0.1985386058688164,
|
|
"num_tokens": 9574903.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.48832859992981,
|
|
"epoch": 4.275078369905956,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035844293528160384,
|
|
"loss": 4.8324,
|
|
"mean_token_accuracy": 0.2084584653377533,
|
|
"num_tokens": 9583583.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.449230146408081,
|
|
"epoch": 4.278996865203762,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003581635194705682,
|
|
"loss": 4.7982,
|
|
"mean_token_accuracy": 0.20950127094984056,
|
|
"num_tokens": 9592391.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.495245552062988,
|
|
"epoch": 4.282915360501567,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035788395503283914,
|
|
"loss": 4.8143,
|
|
"mean_token_accuracy": 0.21291882246732713,
|
|
"num_tokens": 9601384.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.4674458503723145,
|
|
"epoch": 4.286833855799373,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035760424246804334,
|
|
"loss": 4.9063,
|
|
"mean_token_accuracy": 0.2039292722940445,
|
|
"num_tokens": 9610475.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.4423424243927006,
|
|
"epoch": 4.290752351097178,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003573243822760726,
|
|
"loss": 4.7799,
|
|
"mean_token_accuracy": 0.21510279327630996,
|
|
"num_tokens": 9618818.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.497783946990967,
|
|
"epoch": 4.294670846394984,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003570443749570825,
|
|
"loss": 4.8556,
|
|
"mean_token_accuracy": 0.2101364716887474,
|
|
"num_tokens": 9628248.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.543870162963867,
|
|
"epoch": 4.29858934169279,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003567642210114914,
|
|
"loss": 4.839,
|
|
"mean_token_accuracy": 0.20688748806715013,
|
|
"num_tokens": 9636551.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.466529512405396,
|
|
"epoch": 4.302507836990595,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035648392093998,
|
|
"loss": 4.9005,
|
|
"mean_token_accuracy": 0.19885916858911515,
|
|
"num_tokens": 9646293.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.443446826934815,
|
|
"epoch": 4.306426332288401,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003562034752434899,
|
|
"loss": 4.8534,
|
|
"mean_token_accuracy": 0.21908959597349167,
|
|
"num_tokens": 9655806.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.4861067771911625,
|
|
"epoch": 4.310344827586207,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003559228844232229,
|
|
"loss": 4.8858,
|
|
"mean_token_accuracy": 0.20368712246418,
|
|
"num_tokens": 9664757.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.310344827586207,
|
|
"eval_entropy": 5.3718279175056045,
|
|
"eval_loss": 5.6871232986450195,
|
|
"eval_mean_token_accuracy": 0.16759934322547543,
|
|
"eval_num_tokens": 9664757.0,
|
|
"eval_runtime": 2.1912,
|
|
"eval_samples_per_second": 1881.166,
|
|
"eval_steps_per_second": 235.488,
|
|
"step": 5500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 12750,
|
|
"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": 1925913071738880.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|