bc58e4528f
Model: fpadovani/ind-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
7112 lines
193 KiB
JSON
7112 lines
193 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.7429467084639496,
|
|
"eval_steps": 500,
|
|
"global_step": 3500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.113325738906861,
|
|
"epoch": 0.003918495297805642,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.8669,
|
|
"mean_token_accuracy": 0.09941852316260338,
|
|
"num_tokens": 9174.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.112642478942871,
|
|
"epoch": 0.007836990595611285,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.8445,
|
|
"mean_token_accuracy": 0.09862779080867767,
|
|
"num_tokens": 17790.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.158182430267334,
|
|
"epoch": 0.011755485893416929,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.9709,
|
|
"mean_token_accuracy": 0.0985221654176712,
|
|
"num_tokens": 25850.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.190377140045166,
|
|
"epoch": 0.01567398119122257,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.0374,
|
|
"mean_token_accuracy": 0.09146212115883827,
|
|
"num_tokens": 35194.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.184538173675537,
|
|
"epoch": 0.019592476489028215,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.9985,
|
|
"mean_token_accuracy": 0.09104539602994918,
|
|
"num_tokens": 44218.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.137639093399048,
|
|
"epoch": 0.023510971786833857,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.942,
|
|
"mean_token_accuracy": 0.09604569971561432,
|
|
"num_tokens": 53397.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.152150678634643,
|
|
"epoch": 0.0274294670846395,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.8465,
|
|
"mean_token_accuracy": 0.09854738116264343,
|
|
"num_tokens": 62749.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.131413555145263,
|
|
"epoch": 0.03134796238244514,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.8427,
|
|
"mean_token_accuracy": 0.10017006173729896,
|
|
"num_tokens": 71721.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.150160312652588,
|
|
"epoch": 0.03526645768025078,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.8265,
|
|
"mean_token_accuracy": 0.1048971325159073,
|
|
"num_tokens": 79844.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.14600887298584,
|
|
"epoch": 0.03918495297805643,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.8045,
|
|
"mean_token_accuracy": 0.10030318275094033,
|
|
"num_tokens": 88866.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.112334775924682,
|
|
"epoch": 0.04310344827586207,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.8137,
|
|
"mean_token_accuracy": 0.10082174241542816,
|
|
"num_tokens": 97918.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.0919252872467045,
|
|
"epoch": 0.047021943573667714,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.8353,
|
|
"mean_token_accuracy": 0.10005640611052513,
|
|
"num_tokens": 107043.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.190604305267334,
|
|
"epoch": 0.050940438871473356,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.8968,
|
|
"mean_token_accuracy": 0.0979295790195465,
|
|
"num_tokens": 116000.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.273019075393677,
|
|
"epoch": 0.054858934169279,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.9081,
|
|
"mean_token_accuracy": 0.10225948318839073,
|
|
"num_tokens": 124559.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.21258225440979,
|
|
"epoch": 0.05877742946708464,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.8128,
|
|
"mean_token_accuracy": 0.10783266574144364,
|
|
"num_tokens": 132868.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.171340751647949,
|
|
"epoch": 0.06269592476489028,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.9077,
|
|
"mean_token_accuracy": 0.09100360721349716,
|
|
"num_tokens": 141286.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.0822936534881595,
|
|
"epoch": 0.06661442006269593,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.8652,
|
|
"mean_token_accuracy": 0.10414223074913025,
|
|
"num_tokens": 150406.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.177085208892822,
|
|
"epoch": 0.07053291536050156,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.913,
|
|
"mean_token_accuracy": 0.09742300063371659,
|
|
"num_tokens": 158770.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.174734401702881,
|
|
"epoch": 0.07445141065830721,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.8071,
|
|
"mean_token_accuracy": 0.1039510190486908,
|
|
"num_tokens": 167763.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.1777242660522464,
|
|
"epoch": 0.07836990595611286,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.8164,
|
|
"mean_token_accuracy": 0.10320222228765488,
|
|
"num_tokens": 177306.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.103335762023926,
|
|
"epoch": 0.0822884012539185,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.8125,
|
|
"mean_token_accuracy": 0.10012169480323792,
|
|
"num_tokens": 186014.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.209170961380005,
|
|
"epoch": 0.08620689655172414,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.7311,
|
|
"mean_token_accuracy": 0.10817711427807808,
|
|
"num_tokens": 194122.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.15381441116333,
|
|
"epoch": 0.09012539184952978,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.7795,
|
|
"mean_token_accuracy": 0.10609495490789414,
|
|
"num_tokens": 203097.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.090649890899658,
|
|
"epoch": 0.09404388714733543,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.7688,
|
|
"mean_token_accuracy": 0.10584712252020836,
|
|
"num_tokens": 211413.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.0910121440887455,
|
|
"epoch": 0.09796238244514106,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.8318,
|
|
"mean_token_accuracy": 0.1045131392776966,
|
|
"num_tokens": 220550.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.155442237854004,
|
|
"epoch": 0.10188087774294671,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.7659,
|
|
"mean_token_accuracy": 0.10155851393938065,
|
|
"num_tokens": 229197.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.165218687057495,
|
|
"epoch": 0.10579937304075235,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.8843,
|
|
"mean_token_accuracy": 0.09892990812659264,
|
|
"num_tokens": 238479.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.259754037857055,
|
|
"epoch": 0.109717868338558,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.8008,
|
|
"mean_token_accuracy": 0.10250233411788941,
|
|
"num_tokens": 246318.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.158761405944825,
|
|
"epoch": 0.11363636363636363,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.8664,
|
|
"mean_token_accuracy": 0.10097494721412659,
|
|
"num_tokens": 254783.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.123447751998901,
|
|
"epoch": 0.11755485893416928,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.7508,
|
|
"mean_token_accuracy": 0.10182412713766098,
|
|
"num_tokens": 263416.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.246314954757691,
|
|
"epoch": 0.12147335423197492,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.8447,
|
|
"mean_token_accuracy": 0.1027430385351181,
|
|
"num_tokens": 271930.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.18838324546814,
|
|
"epoch": 0.12539184952978055,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.7962,
|
|
"mean_token_accuracy": 0.10592841804027557,
|
|
"num_tokens": 280546.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.097708988189697,
|
|
"epoch": 0.12931034482758622,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.8186,
|
|
"mean_token_accuracy": 0.10792352184653282,
|
|
"num_tokens": 288813.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.2543150901794435,
|
|
"epoch": 0.13322884012539185,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.7815,
|
|
"mean_token_accuracy": 0.10464645773172379,
|
|
"num_tokens": 297966.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.13132381439209,
|
|
"epoch": 0.1371473354231975,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.9155,
|
|
"mean_token_accuracy": 0.10553643256425857,
|
|
"num_tokens": 307135.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.204241514205933,
|
|
"epoch": 0.14106583072100312,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.8318,
|
|
"mean_token_accuracy": 0.10002151355147362,
|
|
"num_tokens": 315546.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.187180662155152,
|
|
"epoch": 0.14498432601880878,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.8224,
|
|
"mean_token_accuracy": 0.10502876788377762,
|
|
"num_tokens": 323930.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.087406158447266,
|
|
"epoch": 0.14890282131661442,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.7248,
|
|
"mean_token_accuracy": 0.11350480541586876,
|
|
"num_tokens": 332419.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.082424688339233,
|
|
"epoch": 0.15282131661442006,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.8184,
|
|
"mean_token_accuracy": 0.10288615301251411,
|
|
"num_tokens": 341362.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.1997878074646,
|
|
"epoch": 0.15673981191222572,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.7508,
|
|
"mean_token_accuracy": 0.10806922689080238,
|
|
"num_tokens": 349395.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.0235200881958,
|
|
"epoch": 0.16065830721003135,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.8101,
|
|
"mean_token_accuracy": 0.099128620326519,
|
|
"num_tokens": 358413.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.297369050979614,
|
|
"epoch": 0.164576802507837,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.8827,
|
|
"mean_token_accuracy": 0.10295988991856575,
|
|
"num_tokens": 366489.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.091110467910767,
|
|
"epoch": 0.16849529780564262,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.7112,
|
|
"mean_token_accuracy": 0.10220234468579292,
|
|
"num_tokens": 375335.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.204977703094483,
|
|
"epoch": 0.1724137931034483,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.7757,
|
|
"mean_token_accuracy": 0.10802004486322403,
|
|
"num_tokens": 384276.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.145157670974731,
|
|
"epoch": 0.17633228840125392,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.8694,
|
|
"mean_token_accuracy": 0.09782575145363807,
|
|
"num_tokens": 393571.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.1306267261505125,
|
|
"epoch": 0.18025078369905956,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.8154,
|
|
"mean_token_accuracy": 0.09901047199964523,
|
|
"num_tokens": 401865.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.0903645038604735,
|
|
"epoch": 0.1841692789968652,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.7693,
|
|
"mean_token_accuracy": 0.10383733138442039,
|
|
"num_tokens": 410518.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.176298904418945,
|
|
"epoch": 0.18808777429467086,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.772,
|
|
"mean_token_accuracy": 0.10165056958794594,
|
|
"num_tokens": 419769.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.052699327468872,
|
|
"epoch": 0.1920062695924765,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.7062,
|
|
"mean_token_accuracy": 0.11240611821413041,
|
|
"num_tokens": 428204.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.229116773605346,
|
|
"epoch": 0.19592476489028213,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.8173,
|
|
"mean_token_accuracy": 0.10085652396082878,
|
|
"num_tokens": 436594.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 6.918967628479004,
|
|
"epoch": 0.19984326018808776,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.6093,
|
|
"mean_token_accuracy": 0.1123976968228817,
|
|
"num_tokens": 444645.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.119955444335938,
|
|
"epoch": 0.20376175548589343,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.6713,
|
|
"mean_token_accuracy": 0.1125674806535244,
|
|
"num_tokens": 453016.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.06598129272461,
|
|
"epoch": 0.20768025078369906,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.7522,
|
|
"mean_token_accuracy": 0.1047289289534092,
|
|
"num_tokens": 462116.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.1544452667236325,
|
|
"epoch": 0.2115987460815047,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.7091,
|
|
"mean_token_accuracy": 0.10634818449616432,
|
|
"num_tokens": 470807.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.103619909286499,
|
|
"epoch": 0.21551724137931033,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.8065,
|
|
"mean_token_accuracy": 0.10270731151103973,
|
|
"num_tokens": 479592.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.1156768798828125,
|
|
"epoch": 0.219435736677116,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.7446,
|
|
"mean_token_accuracy": 0.10210504084825515,
|
|
"num_tokens": 488107.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.0720561981201175,
|
|
"epoch": 0.22335423197492163,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.7655,
|
|
"mean_token_accuracy": 0.1056784711778164,
|
|
"num_tokens": 496775.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.222696924209595,
|
|
"epoch": 0.22727272727272727,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.7897,
|
|
"mean_token_accuracy": 0.10648871958255768,
|
|
"num_tokens": 505415.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.079438734054565,
|
|
"epoch": 0.23119122257053293,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.7795,
|
|
"mean_token_accuracy": 0.1061191700398922,
|
|
"num_tokens": 514447.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.112887191772461,
|
|
"epoch": 0.23510971786833856,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.7863,
|
|
"mean_token_accuracy": 0.10735970661044121,
|
|
"num_tokens": 522998.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.187893390655518,
|
|
"epoch": 0.2390282131661442,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.7606,
|
|
"mean_token_accuracy": 0.10097135305404663,
|
|
"num_tokens": 531542.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.002828359603882,
|
|
"epoch": 0.24294670846394983,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.6867,
|
|
"mean_token_accuracy": 0.10434283241629601,
|
|
"num_tokens": 540143.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.160926628112793,
|
|
"epoch": 0.2468652037617555,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.8173,
|
|
"mean_token_accuracy": 0.10131673216819763,
|
|
"num_tokens": 549156.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.030450010299683,
|
|
"epoch": 0.2507836990595611,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.6447,
|
|
"mean_token_accuracy": 0.1034945398569107,
|
|
"num_tokens": 557522.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.135869073867798,
|
|
"epoch": 0.2547021943573668,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.7145,
|
|
"mean_token_accuracy": 0.109796904027462,
|
|
"num_tokens": 566650.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 6.977855205535889,
|
|
"epoch": 0.25862068965517243,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.6739,
|
|
"mean_token_accuracy": 0.1077032022178173,
|
|
"num_tokens": 576219.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.109291315078735,
|
|
"epoch": 0.26253918495297807,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.8204,
|
|
"mean_token_accuracy": 0.1049183763563633,
|
|
"num_tokens": 584304.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.060112428665161,
|
|
"epoch": 0.2664576802507837,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.7326,
|
|
"mean_token_accuracy": 0.10800091400742531,
|
|
"num_tokens": 593163.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.065881586074829,
|
|
"epoch": 0.27037617554858934,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.6391,
|
|
"mean_token_accuracy": 0.10378878638148308,
|
|
"num_tokens": 602077.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.035525178909301,
|
|
"epoch": 0.274294670846395,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.7544,
|
|
"mean_token_accuracy": 0.10282572582364083,
|
|
"num_tokens": 610009.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.141838598251343,
|
|
"epoch": 0.2782131661442006,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.8191,
|
|
"mean_token_accuracy": 0.10182501152157783,
|
|
"num_tokens": 619282.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.106745004653931,
|
|
"epoch": 0.28213166144200624,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7578,
|
|
"mean_token_accuracy": 0.10329684540629387,
|
|
"num_tokens": 628138.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.018217658996582,
|
|
"epoch": 0.28605015673981193,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.579,
|
|
"mean_token_accuracy": 0.11497451290488243,
|
|
"num_tokens": 637021.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.0183099746704105,
|
|
"epoch": 0.28996865203761757,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7904,
|
|
"mean_token_accuracy": 0.09623993113636971,
|
|
"num_tokens": 646703.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.033835697174072,
|
|
"epoch": 0.2938871473354232,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.6457,
|
|
"mean_token_accuracy": 0.10587730780243873,
|
|
"num_tokens": 655616.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.035226058959961,
|
|
"epoch": 0.29780564263322884,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.6675,
|
|
"mean_token_accuracy": 0.11625474393367767,
|
|
"num_tokens": 663783.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.009380054473877,
|
|
"epoch": 0.3017241379310345,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.6976,
|
|
"mean_token_accuracy": 0.10993565618991852,
|
|
"num_tokens": 671855.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.081801080703736,
|
|
"epoch": 0.3056426332288401,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.706,
|
|
"mean_token_accuracy": 0.10679346546530724,
|
|
"num_tokens": 680981.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.960446166992187,
|
|
"epoch": 0.30956112852664575,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.6116,
|
|
"mean_token_accuracy": 0.10894792973995208,
|
|
"num_tokens": 689294.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.911626148223877,
|
|
"epoch": 0.31347962382445144,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.5734,
|
|
"mean_token_accuracy": 0.1067522719502449,
|
|
"num_tokens": 697703.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.081985807418823,
|
|
"epoch": 0.31739811912225707,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7298,
|
|
"mean_token_accuracy": 0.10676686987280845,
|
|
"num_tokens": 706792.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.004686450958252,
|
|
"epoch": 0.3213166144200627,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.6674,
|
|
"mean_token_accuracy": 0.11164869293570519,
|
|
"num_tokens": 715864.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.0087442874908445,
|
|
"epoch": 0.32523510971786834,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7022,
|
|
"mean_token_accuracy": 0.10990188345313072,
|
|
"num_tokens": 723921.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.068296241760254,
|
|
"epoch": 0.329153605015674,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7359,
|
|
"mean_token_accuracy": 0.10363818407058716,
|
|
"num_tokens": 732797.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.906872224807739,
|
|
"epoch": 0.3330721003134796,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.5814,
|
|
"mean_token_accuracy": 0.11549990624189377,
|
|
"num_tokens": 741248.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.061840009689331,
|
|
"epoch": 0.33699059561128525,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6781,
|
|
"mean_token_accuracy": 0.10647747218608856,
|
|
"num_tokens": 749766.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.947984743118286,
|
|
"epoch": 0.3409090909090909,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6594,
|
|
"mean_token_accuracy": 0.10216243714094161,
|
|
"num_tokens": 758695.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.034923553466797,
|
|
"epoch": 0.3448275862068966,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.6444,
|
|
"mean_token_accuracy": 0.10249690040946006,
|
|
"num_tokens": 767624.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.945315933227539,
|
|
"epoch": 0.3487460815047022,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.6862,
|
|
"mean_token_accuracy": 0.10265796408057212,
|
|
"num_tokens": 776616.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.97764482498169,
|
|
"epoch": 0.35266457680250785,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6966,
|
|
"mean_token_accuracy": 0.10413416102528572,
|
|
"num_tokens": 786147.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.103307485580444,
|
|
"epoch": 0.3565830721003135,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7967,
|
|
"mean_token_accuracy": 0.10401362031698227,
|
|
"num_tokens": 795213.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.968210601806641,
|
|
"epoch": 0.3605015673981191,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6085,
|
|
"mean_token_accuracy": 0.10945433229207993,
|
|
"num_tokens": 803118.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.9675391674041744,
|
|
"epoch": 0.36442006269592475,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.5691,
|
|
"mean_token_accuracy": 0.11808251142501831,
|
|
"num_tokens": 811358.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.9308342933654785,
|
|
"epoch": 0.3683385579937304,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.578,
|
|
"mean_token_accuracy": 0.11157155334949494,
|
|
"num_tokens": 819653.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.824059677124024,
|
|
"epoch": 0.3722570532915361,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6342,
|
|
"mean_token_accuracy": 0.10801615715026855,
|
|
"num_tokens": 828462.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.107431173324585,
|
|
"epoch": 0.3761755485893417,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7107,
|
|
"mean_token_accuracy": 0.10842569470405579,
|
|
"num_tokens": 836204.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.875178289413452,
|
|
"epoch": 0.38009404388714735,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6054,
|
|
"mean_token_accuracy": 0.11096373572945595,
|
|
"num_tokens": 845032.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.9599480628967285,
|
|
"epoch": 0.384012539184953,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6359,
|
|
"mean_token_accuracy": 0.11337224692106247,
|
|
"num_tokens": 853324.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.926474952697754,
|
|
"epoch": 0.3879310344827586,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5964,
|
|
"mean_token_accuracy": 0.12043852508068084,
|
|
"num_tokens": 861640.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.9969902515411375,
|
|
"epoch": 0.39184952978056425,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.7759,
|
|
"mean_token_accuracy": 0.10493798106908798,
|
|
"num_tokens": 870758.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.39184952978056425,
|
|
"eval_entropy": 6.9411013653112015,
|
|
"eval_loss": 6.728126049041748,
|
|
"eval_mean_token_accuracy": 0.10773137585772563,
|
|
"eval_num_tokens": 870758.0,
|
|
"eval_runtime": 2.9294,
|
|
"eval_samples_per_second": 1407.095,
|
|
"eval_steps_per_second": 176.143,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.992229700088501,
|
|
"epoch": 0.3957680250783699,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.6957,
|
|
"mean_token_accuracy": 0.10136032626032829,
|
|
"num_tokens": 879968.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.883435440063477,
|
|
"epoch": 0.3996865203761755,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6479,
|
|
"mean_token_accuracy": 0.10431178212165833,
|
|
"num_tokens": 888804.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.971563863754272,
|
|
"epoch": 0.4036050156739812,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6693,
|
|
"mean_token_accuracy": 0.10408427342772483,
|
|
"num_tokens": 898593.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.925919246673584,
|
|
"epoch": 0.40752351097178685,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6901,
|
|
"mean_token_accuracy": 0.10555115193128586,
|
|
"num_tokens": 906873.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.934886932373047,
|
|
"epoch": 0.4114420062695925,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6698,
|
|
"mean_token_accuracy": 0.11156188994646073,
|
|
"num_tokens": 915080.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.947892522811889,
|
|
"epoch": 0.4153605015673981,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6785,
|
|
"mean_token_accuracy": 0.10801282972097397,
|
|
"num_tokens": 923741.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.883993148803711,
|
|
"epoch": 0.41927899686520376,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.564,
|
|
"mean_token_accuracy": 0.11001835316419602,
|
|
"num_tokens": 932455.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.902733898162841,
|
|
"epoch": 0.4231974921630094,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5815,
|
|
"mean_token_accuracy": 0.11325104832649231,
|
|
"num_tokens": 941064.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.83945484161377,
|
|
"epoch": 0.427115987460815,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6085,
|
|
"mean_token_accuracy": 0.109104935079813,
|
|
"num_tokens": 950261.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.026263904571533,
|
|
"epoch": 0.43103448275862066,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6813,
|
|
"mean_token_accuracy": 0.10714460536837578,
|
|
"num_tokens": 959151.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.907926321029663,
|
|
"epoch": 0.43495297805642635,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.7019,
|
|
"mean_token_accuracy": 0.1055656410753727,
|
|
"num_tokens": 968441.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.876136827468872,
|
|
"epoch": 0.438871473354232,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5755,
|
|
"mean_token_accuracy": 0.10419308096170425,
|
|
"num_tokens": 977058.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.904338932037353,
|
|
"epoch": 0.4427899686520376,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.5945,
|
|
"mean_token_accuracy": 0.11105613335967064,
|
|
"num_tokens": 986531.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.768454742431641,
|
|
"epoch": 0.44670846394984326,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4264,
|
|
"mean_token_accuracy": 0.11743755117058755,
|
|
"num_tokens": 995081.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.789128589630127,
|
|
"epoch": 0.4506269592476489,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.4896,
|
|
"mean_token_accuracy": 0.11737141981720925,
|
|
"num_tokens": 1003459.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.81758337020874,
|
|
"epoch": 0.45454545454545453,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.5718,
|
|
"mean_token_accuracy": 0.11360911354422569,
|
|
"num_tokens": 1012420.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.936664247512818,
|
|
"epoch": 0.45846394984326017,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6519,
|
|
"mean_token_accuracy": 0.10682766661047935,
|
|
"num_tokens": 1021198.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.860663843154907,
|
|
"epoch": 0.46238244514106586,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6291,
|
|
"mean_token_accuracy": 0.11134118661284446,
|
|
"num_tokens": 1030023.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.792492532730103,
|
|
"epoch": 0.4663009404388715,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.4932,
|
|
"mean_token_accuracy": 0.11575940474867821,
|
|
"num_tokens": 1038824.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.972049903869629,
|
|
"epoch": 0.4702194357366771,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.7682,
|
|
"mean_token_accuracy": 0.10632524192333222,
|
|
"num_tokens": 1048207.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.017117214202881,
|
|
"epoch": 0.47413793103448276,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.721,
|
|
"mean_token_accuracy": 0.10762688145041466,
|
|
"num_tokens": 1057354.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.904824686050415,
|
|
"epoch": 0.4780564263322884,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6649,
|
|
"mean_token_accuracy": 0.10675990283489227,
|
|
"num_tokens": 1066825.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.836680459976196,
|
|
"epoch": 0.48197492163009403,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.6006,
|
|
"mean_token_accuracy": 0.10550268962979317,
|
|
"num_tokens": 1075752.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.744200849533081,
|
|
"epoch": 0.48589341692789967,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.4958,
|
|
"mean_token_accuracy": 0.11007569953799248,
|
|
"num_tokens": 1084826.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.978894758224487,
|
|
"epoch": 0.4898119122257053,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6386,
|
|
"mean_token_accuracy": 0.10888363420963287,
|
|
"num_tokens": 1093425.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.815886735916138,
|
|
"epoch": 0.493730407523511,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.6631,
|
|
"mean_token_accuracy": 0.10534236505627632,
|
|
"num_tokens": 1102619.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.809355783462524,
|
|
"epoch": 0.49764890282131663,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5432,
|
|
"mean_token_accuracy": 0.11107134073972702,
|
|
"num_tokens": 1111337.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.7085394859313965,
|
|
"epoch": 0.5015673981191222,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.3882,
|
|
"mean_token_accuracy": 0.1193773865699768,
|
|
"num_tokens": 1119381.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.69315505027771,
|
|
"epoch": 0.5054858934169278,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4395,
|
|
"mean_token_accuracy": 0.11702300012111663,
|
|
"num_tokens": 1127825.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.850753402709961,
|
|
"epoch": 0.5094043887147336,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.4943,
|
|
"mean_token_accuracy": 0.10672505050897599,
|
|
"num_tokens": 1136750.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.7495848655700685,
|
|
"epoch": 0.5133228840125392,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5289,
|
|
"mean_token_accuracy": 0.11170963272452354,
|
|
"num_tokens": 1145739.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.791320037841797,
|
|
"epoch": 0.5172413793103449,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5995,
|
|
"mean_token_accuracy": 0.10925844609737397,
|
|
"num_tokens": 1154415.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.774902629852295,
|
|
"epoch": 0.5211598746081505,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4739,
|
|
"mean_token_accuracy": 0.11284942775964737,
|
|
"num_tokens": 1162480.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.759902763366699,
|
|
"epoch": 0.5250783699059561,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.5655,
|
|
"mean_token_accuracy": 0.10794797912240028,
|
|
"num_tokens": 1170774.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.727387142181397,
|
|
"epoch": 0.5289968652037618,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5016,
|
|
"mean_token_accuracy": 0.1089845173060894,
|
|
"num_tokens": 1179629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.78474268913269,
|
|
"epoch": 0.5329153605015674,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.6631,
|
|
"mean_token_accuracy": 0.1076794520020485,
|
|
"num_tokens": 1189111.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.840711498260498,
|
|
"epoch": 0.536833855799373,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6419,
|
|
"mean_token_accuracy": 0.10526446029543876,
|
|
"num_tokens": 1198827.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.920011568069458,
|
|
"epoch": 0.5407523510971787,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.5519,
|
|
"mean_token_accuracy": 0.11411965638399124,
|
|
"num_tokens": 1207089.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.778568458557129,
|
|
"epoch": 0.5446708463949843,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5436,
|
|
"mean_token_accuracy": 0.10897781774401664,
|
|
"num_tokens": 1216509.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.596413469314575,
|
|
"epoch": 0.54858934169279,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3856,
|
|
"mean_token_accuracy": 0.1203030876815319,
|
|
"num_tokens": 1225684.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.760894346237182,
|
|
"epoch": 0.5525078369905956,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4132,
|
|
"mean_token_accuracy": 0.12549414113163948,
|
|
"num_tokens": 1233675.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.648054599761963,
|
|
"epoch": 0.5564263322884012,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4268,
|
|
"mean_token_accuracy": 0.11218728423118592,
|
|
"num_tokens": 1242147.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.716562032699585,
|
|
"epoch": 0.5603448275862069,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5515,
|
|
"mean_token_accuracy": 0.11361126005649566,
|
|
"num_tokens": 1251127.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.667671203613281,
|
|
"epoch": 0.5642633228840125,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4954,
|
|
"mean_token_accuracy": 0.11189050376415252,
|
|
"num_tokens": 1260589.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.848757553100586,
|
|
"epoch": 0.5681818181818182,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5731,
|
|
"mean_token_accuracy": 0.11687926128506661,
|
|
"num_tokens": 1269281.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.655617666244507,
|
|
"epoch": 0.5721003134796239,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.419,
|
|
"mean_token_accuracy": 0.11433188989758492,
|
|
"num_tokens": 1278033.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.798342895507813,
|
|
"epoch": 0.5760188087774295,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5162,
|
|
"mean_token_accuracy": 0.11682824715971947,
|
|
"num_tokens": 1285847.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.71445107460022,
|
|
"epoch": 0.5799373040752351,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.4871,
|
|
"mean_token_accuracy": 0.11209994032979012,
|
|
"num_tokens": 1294770.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.870912408828735,
|
|
"epoch": 0.5838557993730408,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.7032,
|
|
"mean_token_accuracy": 0.10493360385298729,
|
|
"num_tokens": 1304282.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.731619119644165,
|
|
"epoch": 0.5877742946708464,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5341,
|
|
"mean_token_accuracy": 0.10778379514813423,
|
|
"num_tokens": 1313218.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.747414207458496,
|
|
"epoch": 0.591692789968652,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5067,
|
|
"mean_token_accuracy": 0.11040924564003944,
|
|
"num_tokens": 1322093.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.585576343536377,
|
|
"epoch": 0.5956112852664577,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.5403,
|
|
"mean_token_accuracy": 0.11336176246404647,
|
|
"num_tokens": 1330718.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.7226659774780275,
|
|
"epoch": 0.5995297805642633,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3939,
|
|
"mean_token_accuracy": 0.11393325701355934,
|
|
"num_tokens": 1339672.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.641554880142212,
|
|
"epoch": 0.603448275862069,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4174,
|
|
"mean_token_accuracy": 0.11754084452986717,
|
|
"num_tokens": 1348587.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.603374004364014,
|
|
"epoch": 0.6073667711598746,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4169,
|
|
"mean_token_accuracy": 0.11932410523295403,
|
|
"num_tokens": 1357597.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.732632350921631,
|
|
"epoch": 0.6112852664576802,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5225,
|
|
"mean_token_accuracy": 0.11163635179400444,
|
|
"num_tokens": 1366246.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.643588590621948,
|
|
"epoch": 0.6152037617554859,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4231,
|
|
"mean_token_accuracy": 0.11742421090602875,
|
|
"num_tokens": 1374664.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.60569806098938,
|
|
"epoch": 0.6191222570532915,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.4298,
|
|
"mean_token_accuracy": 0.11620132252573967,
|
|
"num_tokens": 1382765.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.606018590927124,
|
|
"epoch": 0.6230407523510971,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4046,
|
|
"mean_token_accuracy": 0.11398342102766038,
|
|
"num_tokens": 1391708.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.664660692214966,
|
|
"epoch": 0.6269592476489029,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5624,
|
|
"mean_token_accuracy": 0.1093956932425499,
|
|
"num_tokens": 1401373.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.67238826751709,
|
|
"epoch": 0.6308777429467085,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3628,
|
|
"mean_token_accuracy": 0.11816710904240608,
|
|
"num_tokens": 1409702.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.693083429336548,
|
|
"epoch": 0.6347962382445141,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5834,
|
|
"mean_token_accuracy": 0.10899265483021736,
|
|
"num_tokens": 1419596.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.636869716644287,
|
|
"epoch": 0.6387147335423198,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5177,
|
|
"mean_token_accuracy": 0.11091106310486794,
|
|
"num_tokens": 1428437.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.675769662857055,
|
|
"epoch": 0.6426332288401254,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4288,
|
|
"mean_token_accuracy": 0.12033025994896888,
|
|
"num_tokens": 1436764.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.624548244476318,
|
|
"epoch": 0.646551724137931,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4725,
|
|
"mean_token_accuracy": 0.11756742298603058,
|
|
"num_tokens": 1445755.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.693640947341919,
|
|
"epoch": 0.6504702194357367,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5313,
|
|
"mean_token_accuracy": 0.1114391803741455,
|
|
"num_tokens": 1454550.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.563589239120484,
|
|
"epoch": 0.6543887147335423,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.418,
|
|
"mean_token_accuracy": 0.11557744145393371,
|
|
"num_tokens": 1463039.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.734728002548218,
|
|
"epoch": 0.658307210031348,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.478,
|
|
"mean_token_accuracy": 0.11269574463367463,
|
|
"num_tokens": 1472029.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.571061801910401,
|
|
"epoch": 0.6622257053291536,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.5079,
|
|
"mean_token_accuracy": 0.11439812406897545,
|
|
"num_tokens": 1481588.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.628999423980713,
|
|
"epoch": 0.6661442006269592,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4059,
|
|
"mean_token_accuracy": 0.11478642746806145,
|
|
"num_tokens": 1489915.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.672647428512573,
|
|
"epoch": 0.6700626959247649,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.4842,
|
|
"mean_token_accuracy": 0.11243890598416328,
|
|
"num_tokens": 1497920.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.510155534744262,
|
|
"epoch": 0.6739811912225705,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3466,
|
|
"mean_token_accuracy": 0.1146190918982029,
|
|
"num_tokens": 1506965.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.6185520648956295,
|
|
"epoch": 0.6778996865203761,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4236,
|
|
"mean_token_accuracy": 0.11795553267002105,
|
|
"num_tokens": 1516028.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.592772388458252,
|
|
"epoch": 0.6818181818181818,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4644,
|
|
"mean_token_accuracy": 0.11529229208827019,
|
|
"num_tokens": 1524579.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.539308547973633,
|
|
"epoch": 0.6857366771159875,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3769,
|
|
"mean_token_accuracy": 0.11740082949399948,
|
|
"num_tokens": 1534285.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.588557434082031,
|
|
"epoch": 0.6896551724137931,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3787,
|
|
"mean_token_accuracy": 0.11641007363796234,
|
|
"num_tokens": 1543701.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.673776912689209,
|
|
"epoch": 0.6935736677115988,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.6596,
|
|
"mean_token_accuracy": 0.10156535282731056,
|
|
"num_tokens": 1552715.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.408450174331665,
|
|
"epoch": 0.6974921630094044,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2622,
|
|
"mean_token_accuracy": 0.11470296010375022,
|
|
"num_tokens": 1561523.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.55662088394165,
|
|
"epoch": 0.70141065830721,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4914,
|
|
"mean_token_accuracy": 0.11589494124054908,
|
|
"num_tokens": 1570057.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.568389081954956,
|
|
"epoch": 0.7053291536050157,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3345,
|
|
"mean_token_accuracy": 0.11779361218214035,
|
|
"num_tokens": 1578795.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.456902074813843,
|
|
"epoch": 0.7092476489028213,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4116,
|
|
"mean_token_accuracy": 0.11231630444526672,
|
|
"num_tokens": 1588108.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.604302167892456,
|
|
"epoch": 0.713166144200627,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3924,
|
|
"mean_token_accuracy": 0.12046928554773331,
|
|
"num_tokens": 1596545.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.536626100540161,
|
|
"epoch": 0.7170846394984326,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3547,
|
|
"mean_token_accuracy": 0.12420721054077148,
|
|
"num_tokens": 1605098.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.524212121963501,
|
|
"epoch": 0.7210031347962382,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4001,
|
|
"mean_token_accuracy": 0.12292874529957772,
|
|
"num_tokens": 1613760.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.498387098312378,
|
|
"epoch": 0.7249216300940439,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.427,
|
|
"mean_token_accuracy": 0.11109715700149536,
|
|
"num_tokens": 1622345.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.643976163864136,
|
|
"epoch": 0.7288401253918495,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4157,
|
|
"mean_token_accuracy": 0.1193670779466629,
|
|
"num_tokens": 1630706.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.5410667896270756,
|
|
"epoch": 0.7327586206896551,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4832,
|
|
"mean_token_accuracy": 0.11565442010760307,
|
|
"num_tokens": 1639880.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.527542972564698,
|
|
"epoch": 0.7366771159874608,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3374,
|
|
"mean_token_accuracy": 0.11161092445254325,
|
|
"num_tokens": 1648325.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.35137448310852,
|
|
"epoch": 0.7405956112852664,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.2914,
|
|
"mean_token_accuracy": 0.12073137611150742,
|
|
"num_tokens": 1656951.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.511531782150269,
|
|
"epoch": 0.7445141065830722,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4261,
|
|
"mean_token_accuracy": 0.11508626267313957,
|
|
"num_tokens": 1665832.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.508042287826538,
|
|
"epoch": 0.7484326018808778,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3476,
|
|
"mean_token_accuracy": 0.11256715059280395,
|
|
"num_tokens": 1675116.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.4066481590271,
|
|
"epoch": 0.7523510971786834,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.3361,
|
|
"mean_token_accuracy": 0.11710453629493714,
|
|
"num_tokens": 1684195.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.511843872070313,
|
|
"epoch": 0.7562695924764891,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3273,
|
|
"mean_token_accuracy": 0.1193136751651764,
|
|
"num_tokens": 1693025.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.339096546173096,
|
|
"epoch": 0.7601880877742947,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3018,
|
|
"mean_token_accuracy": 0.11799748837947846,
|
|
"num_tokens": 1701210.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.4773938179016115,
|
|
"epoch": 0.7641065830721003,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3955,
|
|
"mean_token_accuracy": 0.10802683010697364,
|
|
"num_tokens": 1711026.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.531501007080078,
|
|
"epoch": 0.768025078369906,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4253,
|
|
"mean_token_accuracy": 0.116270412504673,
|
|
"num_tokens": 1719441.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.4567617893219,
|
|
"epoch": 0.7719435736677116,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.4179,
|
|
"mean_token_accuracy": 0.11469775289297104,
|
|
"num_tokens": 1728300.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.554265642166138,
|
|
"epoch": 0.7758620689655172,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3013,
|
|
"mean_token_accuracy": 0.12283783331513405,
|
|
"num_tokens": 1736364.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.199279832839966,
|
|
"epoch": 0.7797805642633229,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1391,
|
|
"mean_token_accuracy": 0.12946273759007454,
|
|
"num_tokens": 1745043.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.48647174835205,
|
|
"epoch": 0.7836990595611285,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2994,
|
|
"mean_token_accuracy": 0.12199065089225769,
|
|
"num_tokens": 1754008.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.7836990595611285,
|
|
"eval_entropy": 6.26603299148323,
|
|
"eval_loss": 6.44874906539917,
|
|
"eval_mean_token_accuracy": 0.11973919128176085,
|
|
"eval_num_tokens": 1754008.0,
|
|
"eval_runtime": 2.8517,
|
|
"eval_samples_per_second": 1445.471,
|
|
"eval_steps_per_second": 180.947,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.291542720794678,
|
|
"epoch": 0.7876175548589341,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999998713243189,
|
|
"loss": 6.3511,
|
|
"mean_token_accuracy": 0.1147798478603363,
|
|
"num_tokens": 1763592.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.429899978637695,
|
|
"epoch": 0.7915360501567398,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999993485796164,
|
|
"loss": 6.2139,
|
|
"mean_token_accuracy": 0.12304601818323135,
|
|
"num_tokens": 1773111.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.388966035842896,
|
|
"epoch": 0.7954545454545454,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999984237245962,
|
|
"loss": 6.2628,
|
|
"mean_token_accuracy": 0.12534007281064988,
|
|
"num_tokens": 1781230.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.55175461769104,
|
|
"epoch": 0.799373040752351,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004999970967609109,
|
|
"loss": 6.5136,
|
|
"mean_token_accuracy": 0.11764922365546227,
|
|
"num_tokens": 1790324.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.400601530075074,
|
|
"epoch": 0.8032915360501567,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004999953676909322,
|
|
"loss": 6.3543,
|
|
"mean_token_accuracy": 0.11396604254841805,
|
|
"num_tokens": 1798849.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.416995716094971,
|
|
"epoch": 0.8072100313479624,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004999932365177502,
|
|
"loss": 6.2295,
|
|
"mean_token_accuracy": 0.12402529492974282,
|
|
"num_tokens": 1808019.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.492223310470581,
|
|
"epoch": 0.8111285266457681,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999907032451735,
|
|
"loss": 6.3318,
|
|
"mean_token_accuracy": 0.11823122128844261,
|
|
"num_tokens": 1816550.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.270820474624633,
|
|
"epoch": 0.8150470219435737,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999877678777296,
|
|
"loss": 6.2441,
|
|
"mean_token_accuracy": 0.127071038633585,
|
|
"num_tokens": 1825238.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.365108633041382,
|
|
"epoch": 0.8189655172413793,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999844304206645,
|
|
"loss": 6.2098,
|
|
"mean_token_accuracy": 0.12079643234610557,
|
|
"num_tokens": 1833645.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.404587888717652,
|
|
"epoch": 0.822884012539185,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999806908799428,
|
|
"loss": 6.2862,
|
|
"mean_token_accuracy": 0.11846316829323769,
|
|
"num_tokens": 1842892.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.343483209609985,
|
|
"epoch": 0.8268025078369906,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004999765492622474,
|
|
"loss": 6.3311,
|
|
"mean_token_accuracy": 0.12146014720201492,
|
|
"num_tokens": 1851400.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.481566047668457,
|
|
"epoch": 0.8307210031347962,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999720055749804,
|
|
"loss": 6.2614,
|
|
"mean_token_accuracy": 0.1218932181596756,
|
|
"num_tokens": 1860088.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.418999338150025,
|
|
"epoch": 0.8346394984326019,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999670598262619,
|
|
"loss": 6.3271,
|
|
"mean_token_accuracy": 0.12037511318922042,
|
|
"num_tokens": 1868558.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.369568681716919,
|
|
"epoch": 0.8385579937304075,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999617120249308,
|
|
"loss": 6.3198,
|
|
"mean_token_accuracy": 0.11883664280176162,
|
|
"num_tokens": 1877624.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.448858261108398,
|
|
"epoch": 0.8424764890282131,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999559621805445,
|
|
"loss": 6.4296,
|
|
"mean_token_accuracy": 0.11337628290057182,
|
|
"num_tokens": 1886505.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.279574394226074,
|
|
"epoch": 0.8463949843260188,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999498103033788,
|
|
"loss": 6.1911,
|
|
"mean_token_accuracy": 0.1266665816307068,
|
|
"num_tokens": 1895701.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.333665657043457,
|
|
"epoch": 0.8503134796238244,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999432564044284,
|
|
"loss": 6.1551,
|
|
"mean_token_accuracy": 0.12536471113562583,
|
|
"num_tokens": 1904640.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.3966066360473635,
|
|
"epoch": 0.85423197492163,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999363004954058,
|
|
"loss": 6.3837,
|
|
"mean_token_accuracy": 0.11640885770320893,
|
|
"num_tokens": 1914216.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.347075700759888,
|
|
"epoch": 0.8581504702194357,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.2932,
|
|
"mean_token_accuracy": 0.12313793152570725,
|
|
"num_tokens": 1923280.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.454287147521972,
|
|
"epoch": 0.8620689655172413,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999211826975884,
|
|
"loss": 6.3132,
|
|
"mean_token_accuracy": 0.126905470341444,
|
|
"num_tokens": 1931403.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.4443706512451175,
|
|
"epoch": 0.8659874608150471,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999130208358114,
|
|
"loss": 6.3174,
|
|
"mean_token_accuracy": 0.12164701670408248,
|
|
"num_tokens": 1940774.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.394332456588745,
|
|
"epoch": 0.8699059561128527,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999044570179983,
|
|
"loss": 6.4178,
|
|
"mean_token_accuracy": 0.11814009100198745,
|
|
"num_tokens": 1950045.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.467902517318725,
|
|
"epoch": 0.8738244514106583,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998954912594541,
|
|
"loss": 6.2796,
|
|
"mean_token_accuracy": 0.12150320187211036,
|
|
"num_tokens": 1959303.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.338294458389282,
|
|
"epoch": 0.877742946708464,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998861235762018,
|
|
"loss": 6.266,
|
|
"mean_token_accuracy": 0.12415505647659301,
|
|
"num_tokens": 1968583.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.4064007759094235,
|
|
"epoch": 0.8816614420062696,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998763539849832,
|
|
"loss": 6.3821,
|
|
"mean_token_accuracy": 0.11770115494728088,
|
|
"num_tokens": 1977489.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.2882476329803465,
|
|
"epoch": 0.8855799373040752,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998661825032579,
|
|
"loss": 6.1443,
|
|
"mean_token_accuracy": 0.1252014622092247,
|
|
"num_tokens": 1985308.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.315972471237183,
|
|
"epoch": 0.8894984326018809,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004998556091492043,
|
|
"loss": 6.3124,
|
|
"mean_token_accuracy": 0.11539808958768845,
|
|
"num_tokens": 1993930.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.357517290115356,
|
|
"epoch": 0.8934169278996865,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998446339417184,
|
|
"loss": 6.302,
|
|
"mean_token_accuracy": 0.11564947813749313,
|
|
"num_tokens": 2003802.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.386947774887085,
|
|
"epoch": 0.8973354231974922,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998332569004149,
|
|
"loss": 6.2131,
|
|
"mean_token_accuracy": 0.12580082193017006,
|
|
"num_tokens": 2011957.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.292334413528442,
|
|
"epoch": 0.9012539184952978,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998214780456263,
|
|
"loss": 6.2042,
|
|
"mean_token_accuracy": 0.12626376748085022,
|
|
"num_tokens": 2020741.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.330618238449096,
|
|
"epoch": 0.9051724137931034,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998092973984033,
|
|
"loss": 6.2945,
|
|
"mean_token_accuracy": 0.12051898464560509,
|
|
"num_tokens": 2029804.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.426625776290893,
|
|
"epoch": 0.9090909090909091,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997967149805147,
|
|
"loss": 6.3338,
|
|
"mean_token_accuracy": 0.12462769895792007,
|
|
"num_tokens": 2038732.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.363294315338135,
|
|
"epoch": 0.9130094043887147,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004997837308144474,
|
|
"loss": 6.2391,
|
|
"mean_token_accuracy": 0.13117346838116645,
|
|
"num_tokens": 2046362.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.337823486328125,
|
|
"epoch": 0.9169278996865203,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997703449234062,
|
|
"loss": 6.2264,
|
|
"mean_token_accuracy": 0.12225590869784356,
|
|
"num_tokens": 2056018.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.2902881622314455,
|
|
"epoch": 0.920846394984326,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997565573313139,
|
|
"loss": 6.209,
|
|
"mean_token_accuracy": 0.11975079849362373,
|
|
"num_tokens": 2065077.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.288452243804931,
|
|
"epoch": 0.9247648902821317,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997423680628109,
|
|
"loss": 6.1988,
|
|
"mean_token_accuracy": 0.12091932371258736,
|
|
"num_tokens": 2073405.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.159915637969971,
|
|
"epoch": 0.9286833855799373,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000499727777143256,
|
|
"loss": 6.159,
|
|
"mean_token_accuracy": 0.12859885394573212,
|
|
"num_tokens": 2081891.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.459240245819092,
|
|
"epoch": 0.932601880877743,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004997127845987255,
|
|
"loss": 6.2215,
|
|
"mean_token_accuracy": 0.13154146820306778,
|
|
"num_tokens": 2090732.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.181884670257569,
|
|
"epoch": 0.9365203761755486,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996973904560135,
|
|
"loss": 6.165,
|
|
"mean_token_accuracy": 0.13058446869254112,
|
|
"num_tokens": 2099352.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.359483480453491,
|
|
"epoch": 0.9404388714733543,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004996815947426317,
|
|
"loss": 6.1959,
|
|
"mean_token_accuracy": 0.12631767764687538,
|
|
"num_tokens": 2107365.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.259723424911499,
|
|
"epoch": 0.9443573667711599,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004996653974868097,
|
|
"loss": 6.0728,
|
|
"mean_token_accuracy": 0.12826675027608872,
|
|
"num_tokens": 2116192.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.181558895111084,
|
|
"epoch": 0.9482758620689655,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996487987174946,
|
|
"loss": 6.1668,
|
|
"mean_token_accuracy": 0.1246504932641983,
|
|
"num_tokens": 2124410.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.323344802856445,
|
|
"epoch": 0.9521943573667712,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996317984643512,
|
|
"loss": 6.1916,
|
|
"mean_token_accuracy": 0.12328546494245529,
|
|
"num_tokens": 2132799.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.306202745437622,
|
|
"epoch": 0.9561128526645768,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996143967577615,
|
|
"loss": 6.3533,
|
|
"mean_token_accuracy": 0.11768218651413917,
|
|
"num_tokens": 2141589.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.337114143371582,
|
|
"epoch": 0.9600313479623824,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004995965936288254,
|
|
"loss": 6.2465,
|
|
"mean_token_accuracy": 0.1237958587706089,
|
|
"num_tokens": 2150704.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.185662794113159,
|
|
"epoch": 0.9639498432601881,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004995783891093597,
|
|
"loss": 6.1759,
|
|
"mean_token_accuracy": 0.12887666672468184,
|
|
"num_tokens": 2159587.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.192087316513062,
|
|
"epoch": 0.9678683385579937,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499559783231899,
|
|
"loss": 6.1353,
|
|
"mean_token_accuracy": 0.12168272957205772,
|
|
"num_tokens": 2169071.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.3143566131591795,
|
|
"epoch": 0.9717868338557993,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995407760296952,
|
|
"loss": 6.2276,
|
|
"mean_token_accuracy": 0.12075230255722999,
|
|
"num_tokens": 2178248.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.30960898399353,
|
|
"epoch": 0.975705329153605,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995213675367169,
|
|
"loss": 6.1378,
|
|
"mean_token_accuracy": 0.11877818033099174,
|
|
"num_tokens": 2187520.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.156886005401612,
|
|
"epoch": 0.9796238244514106,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995015577876504,
|
|
"loss": 6.2089,
|
|
"mean_token_accuracy": 0.12378308922052383,
|
|
"num_tokens": 2196861.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.270166254043579,
|
|
"epoch": 0.9835423197492164,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499481346817899,
|
|
"loss": 6.13,
|
|
"mean_token_accuracy": 0.1273690603673458,
|
|
"num_tokens": 2205523.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.210124588012695,
|
|
"epoch": 0.987460815047022,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004994607346635829,
|
|
"loss": 6.0823,
|
|
"mean_token_accuracy": 0.1346528261899948,
|
|
"num_tokens": 2214291.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.258036661148071,
|
|
"epoch": 0.9913793103448276,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004994397213615394,
|
|
"loss": 6.1947,
|
|
"mean_token_accuracy": 0.1263251446187496,
|
|
"num_tokens": 2223093.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.241307115554809,
|
|
"epoch": 0.9952978056426333,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004994183069493225,
|
|
"loss": 6.1524,
|
|
"mean_token_accuracy": 0.12528934627771376,
|
|
"num_tokens": 2231551.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.143906736373902,
|
|
"epoch": 0.9992163009404389,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004993964914652034,
|
|
"loss": 6.1585,
|
|
"mean_token_accuracy": 0.13556302934885026,
|
|
"num_tokens": 2240650.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.344608116149902,
|
|
"epoch": 1.0031347962382444,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004993742749481699,
|
|
"loss": 6.0429,
|
|
"mean_token_accuracy": 0.12464264035224915,
|
|
"num_tokens": 2249392.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.091021728515625,
|
|
"epoch": 1.0070532915360502,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004993516574379265,
|
|
"loss": 5.8549,
|
|
"mean_token_accuracy": 0.13539742380380632,
|
|
"num_tokens": 2258100.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.187678098678589,
|
|
"epoch": 1.0109717868338557,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004993286389748943,
|
|
"loss": 5.9221,
|
|
"mean_token_accuracy": 0.13306491151452066,
|
|
"num_tokens": 2266661.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.146251630783081,
|
|
"epoch": 1.0148902821316614,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993052196002112,
|
|
"loss": 6.0288,
|
|
"mean_token_accuracy": 0.12982963621616364,
|
|
"num_tokens": 2275475.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.161657190322876,
|
|
"epoch": 1.0188087774294672,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004992813993557312,
|
|
"loss": 5.9267,
|
|
"mean_token_accuracy": 0.13516829237341882,
|
|
"num_tokens": 2284077.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.11398606300354,
|
|
"epoch": 1.0227272727272727,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499257178284025,
|
|
"loss": 5.9547,
|
|
"mean_token_accuracy": 0.12690627500414847,
|
|
"num_tokens": 2293029.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.120842313766479,
|
|
"epoch": 1.0266457680250785,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004992325564283797,
|
|
"loss": 5.8673,
|
|
"mean_token_accuracy": 0.14652334675192832,
|
|
"num_tokens": 2301574.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.083501148223877,
|
|
"epoch": 1.030564263322884,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004992075338327984,
|
|
"loss": 5.8479,
|
|
"mean_token_accuracy": 0.13996412083506585,
|
|
"num_tokens": 2310366.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.188275384902954,
|
|
"epoch": 1.0344827586206897,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004991821105420006,
|
|
"loss": 5.9953,
|
|
"mean_token_accuracy": 0.12889379039406776,
|
|
"num_tokens": 2318943.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.178983068466186,
|
|
"epoch": 1.0384012539184952,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004991562866014219,
|
|
"loss": 5.9852,
|
|
"mean_token_accuracy": 0.13627749904990197,
|
|
"num_tokens": 2327173.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.165381383895874,
|
|
"epoch": 1.042319749216301,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 5.9485,
|
|
"mean_token_accuracy": 0.13148076757788657,
|
|
"num_tokens": 2336194.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.115203237533569,
|
|
"epoch": 1.0462382445141065,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004991034369562438,
|
|
"loss": 6.0124,
|
|
"mean_token_accuracy": 0.13209957554936408,
|
|
"num_tokens": 2344721.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.131257820129394,
|
|
"epoch": 1.0501567398119123,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004990764113460953,
|
|
"loss": 5.9498,
|
|
"mean_token_accuracy": 0.13055929765105248,
|
|
"num_tokens": 2353406.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.2284763813018795,
|
|
"epoch": 1.0540752351097178,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004990489852750675,
|
|
"loss": 6.11,
|
|
"mean_token_accuracy": 0.1256877563893795,
|
|
"num_tokens": 2361896.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.11944842338562,
|
|
"epoch": 1.0579937304075235,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004990211587921751,
|
|
"loss": 5.9414,
|
|
"mean_token_accuracy": 0.13127534314990044,
|
|
"num_tokens": 2371861.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.063726902008057,
|
|
"epoch": 1.061912225705329,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004989929319471487,
|
|
"loss": 5.8775,
|
|
"mean_token_accuracy": 0.13494107648730277,
|
|
"num_tokens": 2381555.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.090635299682617,
|
|
"epoch": 1.0658307210031348,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004989643047904341,
|
|
"loss": 5.8576,
|
|
"mean_token_accuracy": 0.14259101897478105,
|
|
"num_tokens": 2389851.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.012026739120484,
|
|
"epoch": 1.0697492163009406,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989352773731928,
|
|
"loss": 5.8025,
|
|
"mean_token_accuracy": 0.1421963430941105,
|
|
"num_tokens": 2397764.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.188090085983276,
|
|
"epoch": 1.073667711598746,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989058497473014,
|
|
"loss": 5.9497,
|
|
"mean_token_accuracy": 0.13977306112647056,
|
|
"num_tokens": 2406688.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.121643972396851,
|
|
"epoch": 1.0775862068965518,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004988760219653518,
|
|
"loss": 6.0132,
|
|
"mean_token_accuracy": 0.13169477209448816,
|
|
"num_tokens": 2415475.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.123086738586426,
|
|
"epoch": 1.0815047021943573,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004988457940806513,
|
|
"loss": 5.9703,
|
|
"mean_token_accuracy": 0.12858632132411002,
|
|
"num_tokens": 2425719.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.025482082366944,
|
|
"epoch": 1.085423197492163,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004988151661472218,
|
|
"loss": 5.8792,
|
|
"mean_token_accuracy": 0.13557613492012024,
|
|
"num_tokens": 2435138.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.151245784759522,
|
|
"epoch": 1.0893416927899686,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004987841382198006,
|
|
"loss": 6.0136,
|
|
"mean_token_accuracy": 0.13455865979194642,
|
|
"num_tokens": 2443976.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.99044942855835,
|
|
"epoch": 1.0932601880877744,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004987527103538394,
|
|
"loss": 5.9018,
|
|
"mean_token_accuracy": 0.13744083419442177,
|
|
"num_tokens": 2453287.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.198356246948242,
|
|
"epoch": 1.09717868338558,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000498720882605505,
|
|
"loss": 6.0023,
|
|
"mean_token_accuracy": 0.12969460785388948,
|
|
"num_tokens": 2462054.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.005903244018555,
|
|
"epoch": 1.1010971786833856,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004986886550316788,
|
|
"loss": 5.8813,
|
|
"mean_token_accuracy": 0.14015732035040857,
|
|
"num_tokens": 2470333.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.0531473636627195,
|
|
"epoch": 1.1050156739811912,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004986560276899565,
|
|
"loss": 5.9526,
|
|
"mean_token_accuracy": 0.13291476517915726,
|
|
"num_tokens": 2479621.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.125274991989135,
|
|
"epoch": 1.108934169278997,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004986230006386488,
|
|
"loss": 5.9006,
|
|
"mean_token_accuracy": 0.13800743818283082,
|
|
"num_tokens": 2488020.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.111095190048218,
|
|
"epoch": 1.1128526645768024,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004985895739367802,
|
|
"loss": 5.9408,
|
|
"mean_token_accuracy": 0.13480502888560295,
|
|
"num_tokens": 2497779.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.066828060150146,
|
|
"epoch": 1.1167711598746082,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004985557476440895,
|
|
"loss": 5.8817,
|
|
"mean_token_accuracy": 0.1367568626999855,
|
|
"num_tokens": 2506791.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.096922731399536,
|
|
"epoch": 1.1206896551724137,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00049852152182103,
|
|
"loss": 5.9866,
|
|
"mean_token_accuracy": 0.13348396867513657,
|
|
"num_tokens": 2515576.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.147091960906982,
|
|
"epoch": 1.1246081504702194,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004984868965287686,
|
|
"loss": 5.9147,
|
|
"mean_token_accuracy": 0.136567921936512,
|
|
"num_tokens": 2523933.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.00162410736084,
|
|
"epoch": 1.1285266457680252,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004984518718291864,
|
|
"loss": 5.8647,
|
|
"mean_token_accuracy": 0.1318059407174587,
|
|
"num_tokens": 2532453.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.0632120132446286,
|
|
"epoch": 1.1324451410658307,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004984164477848783,
|
|
"loss": 5.9105,
|
|
"mean_token_accuracy": 0.13608234971761704,
|
|
"num_tokens": 2541149.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.078012609481812,
|
|
"epoch": 1.1363636363636362,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004983806244591528,
|
|
"loss": 5.8876,
|
|
"mean_token_accuracy": 0.14004134014248848,
|
|
"num_tokens": 2550302.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.941012239456176,
|
|
"epoch": 1.140282131661442,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004983444019160318,
|
|
"loss": 5.7526,
|
|
"mean_token_accuracy": 0.14576120525598527,
|
|
"num_tokens": 2558507.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.106155681610107,
|
|
"epoch": 1.1442006269592477,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004983077802202511,
|
|
"loss": 6.0189,
|
|
"mean_token_accuracy": 0.13469743952155114,
|
|
"num_tokens": 2567487.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.060013628005981,
|
|
"epoch": 1.1481191222570533,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004982707594372595,
|
|
"loss": 5.8779,
|
|
"mean_token_accuracy": 0.1318593814969063,
|
|
"num_tokens": 2576232.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.010817146301269,
|
|
"epoch": 1.152037617554859,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000498233339633219,
|
|
"loss": 5.8846,
|
|
"mean_token_accuracy": 0.1372211568057537,
|
|
"num_tokens": 2585081.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.013401889801026,
|
|
"epoch": 1.1559561128526645,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000498195520875005,
|
|
"loss": 5.858,
|
|
"mean_token_accuracy": 0.1376308247447014,
|
|
"num_tokens": 2593890.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.076926422119141,
|
|
"epoch": 1.1598746081504703,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004981573032302059,
|
|
"loss": 5.9343,
|
|
"mean_token_accuracy": 0.13154210969805719,
|
|
"num_tokens": 2602011.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.001340532302857,
|
|
"epoch": 1.1637931034482758,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004981186867671225,
|
|
"loss": 5.8382,
|
|
"mean_token_accuracy": 0.1394493967294693,
|
|
"num_tokens": 2610366.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.969131422042847,
|
|
"epoch": 1.1677115987460815,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004980796715547691,
|
|
"loss": 5.8465,
|
|
"mean_token_accuracy": 0.14186024740338327,
|
|
"num_tokens": 2619737.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.0836954593658445,
|
|
"epoch": 1.171630094043887,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004980402576628717,
|
|
"loss": 5.8896,
|
|
"mean_token_accuracy": 0.13914738819003106,
|
|
"num_tokens": 2628176.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.938668918609619,
|
|
"epoch": 1.1755485893416928,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004980004451618697,
|
|
"loss": 5.8927,
|
|
"mean_token_accuracy": 0.13917603492736816,
|
|
"num_tokens": 2636195.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.1755485893416928,
|
|
"eval_entropy": 5.986282704412475,
|
|
"eval_loss": 6.191744327545166,
|
|
"eval_mean_token_accuracy": 0.13095851831657943,
|
|
"eval_num_tokens": 2636195.0,
|
|
"eval_runtime": 2.853,
|
|
"eval_samples_per_second": 1444.817,
|
|
"eval_steps_per_second": 180.865,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.155323219299317,
|
|
"epoch": 1.1794670846394983,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004979602341229144,
|
|
"loss": 5.9544,
|
|
"mean_token_accuracy": 0.13536008521914483,
|
|
"num_tokens": 2645035.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.0477063179016115,
|
|
"epoch": 1.183385579937304,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004979196246178694,
|
|
"loss": 5.9375,
|
|
"mean_token_accuracy": 0.13434076979756354,
|
|
"num_tokens": 2653239.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.962043952941895,
|
|
"epoch": 1.1873040752351098,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004978786167193105,
|
|
"loss": 5.9298,
|
|
"mean_token_accuracy": 0.13612919598817824,
|
|
"num_tokens": 2662532.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.097779035568237,
|
|
"epoch": 1.1912225705329154,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004978372105005254,
|
|
"loss": 5.9252,
|
|
"mean_token_accuracy": 0.1357118271291256,
|
|
"num_tokens": 2671867.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.931209087371826,
|
|
"epoch": 1.1951410658307209,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004977954060355139,
|
|
"loss": 5.8616,
|
|
"mean_token_accuracy": 0.1355850227177143,
|
|
"num_tokens": 2680693.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.132346057891846,
|
|
"epoch": 1.1990595611285266,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004977532033989871,
|
|
"loss": 5.9434,
|
|
"mean_token_accuracy": 0.13678196668624878,
|
|
"num_tokens": 2689228.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.966204833984375,
|
|
"epoch": 1.2029780564263324,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004977106026663681,
|
|
"loss": 5.8242,
|
|
"mean_token_accuracy": 0.13174520581960678,
|
|
"num_tokens": 2698483.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.072604417800903,
|
|
"epoch": 1.206896551724138,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004976676039137914,
|
|
"loss": 5.9645,
|
|
"mean_token_accuracy": 0.1317337729036808,
|
|
"num_tokens": 2706920.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.017421102523803,
|
|
"epoch": 1.2108150470219436,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004976242072181026,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.14151965230703353,
|
|
"num_tokens": 2715954.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.9726134777069095,
|
|
"epoch": 1.2147335423197492,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004975804126568587,
|
|
"loss": 5.8649,
|
|
"mean_token_accuracy": 0.14412466660141945,
|
|
"num_tokens": 2724890.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.04607458114624,
|
|
"epoch": 1.218652037617555,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004975362203083277,
|
|
"loss": 5.8895,
|
|
"mean_token_accuracy": 0.13369582816958428,
|
|
"num_tokens": 2733688.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.079061126708984,
|
|
"epoch": 1.2225705329153604,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004974916302514886,
|
|
"loss": 5.9016,
|
|
"mean_token_accuracy": 0.13256118297576905,
|
|
"num_tokens": 2743364.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.943713235855102,
|
|
"epoch": 1.2264890282131662,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 5.9392,
|
|
"mean_token_accuracy": 0.12959515154361725,
|
|
"num_tokens": 2754276.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.054774904251099,
|
|
"epoch": 1.2304075235109717,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004974012573323545,
|
|
"loss": 5.7447,
|
|
"mean_token_accuracy": 0.1476442575454712,
|
|
"num_tokens": 2762680.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.864394569396973,
|
|
"epoch": 1.2343260188087775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004973554746315709,
|
|
"loss": 5.8482,
|
|
"mean_token_accuracy": 0.14662062674760817,
|
|
"num_tokens": 2771161.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.048675584793091,
|
|
"epoch": 1.238244514106583,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004973092945455009,
|
|
"loss": 5.8051,
|
|
"mean_token_accuracy": 0.14707713723182678,
|
|
"num_tokens": 2779321.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.943394899368286,
|
|
"epoch": 1.2421630094043887,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004972627171566757,
|
|
"loss": 5.9796,
|
|
"mean_token_accuracy": 0.13489320799708365,
|
|
"num_tokens": 2788425.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.981634616851807,
|
|
"epoch": 1.2460815047021945,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004972157425483368,
|
|
"loss": 5.8248,
|
|
"mean_token_accuracy": 0.14254749417304993,
|
|
"num_tokens": 2796747.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.935612487792969,
|
|
"epoch": 1.25,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004971683708044353,
|
|
"loss": 5.8731,
|
|
"mean_token_accuracy": 0.14065025225281716,
|
|
"num_tokens": 2805335.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.01358585357666,
|
|
"epoch": 1.2539184952978055,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004971206020096323,
|
|
"loss": 5.7607,
|
|
"mean_token_accuracy": 0.14524763226509094,
|
|
"num_tokens": 2813350.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.021692228317261,
|
|
"epoch": 1.2578369905956113,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004970724362492984,
|
|
"loss": 5.9608,
|
|
"mean_token_accuracy": 0.13947487324476243,
|
|
"num_tokens": 2821922.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.959397029876709,
|
|
"epoch": 1.261755485893417,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004970238736095135,
|
|
"loss": 5.8589,
|
|
"mean_token_accuracy": 0.1373360872268677,
|
|
"num_tokens": 2830832.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.025397825241089,
|
|
"epoch": 1.2656739811912225,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004969749141770671,
|
|
"loss": 5.8598,
|
|
"mean_token_accuracy": 0.1413205847144127,
|
|
"num_tokens": 2839699.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.997832584381103,
|
|
"epoch": 1.2695924764890283,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004969255580394575,
|
|
"loss": 5.9218,
|
|
"mean_token_accuracy": 0.14140960872173308,
|
|
"num_tokens": 2848293.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.035713386535645,
|
|
"epoch": 1.2735109717868338,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000496875805284892,
|
|
"loss": 5.8727,
|
|
"mean_token_accuracy": 0.14097950309515,
|
|
"num_tokens": 2856468.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.976678323745728,
|
|
"epoch": 1.2774294670846396,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004968256560022871,
|
|
"loss": 5.8147,
|
|
"mean_token_accuracy": 0.14333799257874488,
|
|
"num_tokens": 2865366.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.939387321472168,
|
|
"epoch": 1.281347962382445,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004967751102812676,
|
|
"loss": 5.8761,
|
|
"mean_token_accuracy": 0.14351602345705033,
|
|
"num_tokens": 2873520.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.00247712135315,
|
|
"epoch": 1.2852664576802508,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004967241682121669,
|
|
"loss": 5.9385,
|
|
"mean_token_accuracy": 0.13742373287677764,
|
|
"num_tokens": 2883324.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.022455883026123,
|
|
"epoch": 1.2891849529780564,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004966728298860267,
|
|
"loss": 5.8665,
|
|
"mean_token_accuracy": 0.1368059866130352,
|
|
"num_tokens": 2892200.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.961077690124512,
|
|
"epoch": 1.293103448275862,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004966210953945969,
|
|
"loss": 5.8382,
|
|
"mean_token_accuracy": 0.1419058546423912,
|
|
"num_tokens": 2900597.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.017272663116455,
|
|
"epoch": 1.2970219435736676,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004965689648303355,
|
|
"loss": 5.8791,
|
|
"mean_token_accuracy": 0.1380735568702221,
|
|
"num_tokens": 2909869.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.8662271976470945,
|
|
"epoch": 1.3009404388714734,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004965164382864083,
|
|
"loss": 5.8916,
|
|
"mean_token_accuracy": 0.14124979376792907,
|
|
"num_tokens": 2919166.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.08414101600647,
|
|
"epoch": 1.3048589341692791,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004964635158566886,
|
|
"loss": 5.8954,
|
|
"mean_token_accuracy": 0.14140134826302528,
|
|
"num_tokens": 2927884.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.943005084991455,
|
|
"epoch": 1.3087774294670846,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004964101976357574,
|
|
"loss": 5.8419,
|
|
"mean_token_accuracy": 0.14933276697993278,
|
|
"num_tokens": 2936737.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.990343284606934,
|
|
"epoch": 1.3126959247648902,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000496356483718903,
|
|
"loss": 5.9346,
|
|
"mean_token_accuracy": 0.1422124423086643,
|
|
"num_tokens": 2944203.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.030170917510986,
|
|
"epoch": 1.316614420062696,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004963023742021211,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.13771088048815727,
|
|
"num_tokens": 2952639.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.980809211730957,
|
|
"epoch": 1.3205329153605017,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000496247869182114,
|
|
"loss": 5.8751,
|
|
"mean_token_accuracy": 0.13748376145958902,
|
|
"num_tokens": 2961482.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.976028776168823,
|
|
"epoch": 1.3244514106583072,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004961929687562909,
|
|
"loss": 5.8906,
|
|
"mean_token_accuracy": 0.13681933358311654,
|
|
"num_tokens": 2970927.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.919792747497558,
|
|
"epoch": 1.328369905956113,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004961376730227681,
|
|
"loss": 5.8038,
|
|
"mean_token_accuracy": 0.14354753047227858,
|
|
"num_tokens": 2979329.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.945391416549683,
|
|
"epoch": 1.3322884012539185,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004960819820803675,
|
|
"loss": 5.7922,
|
|
"mean_token_accuracy": 0.14268116950988768,
|
|
"num_tokens": 2986894.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.93856897354126,
|
|
"epoch": 1.3362068965517242,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004960258960286185,
|
|
"loss": 5.7954,
|
|
"mean_token_accuracy": 0.14203962236642836,
|
|
"num_tokens": 2994721.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.9834705829620365,
|
|
"epoch": 1.3401253918495297,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004959694149677553,
|
|
"loss": 5.9459,
|
|
"mean_token_accuracy": 0.13363023549318315,
|
|
"num_tokens": 3003165.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.01661696434021,
|
|
"epoch": 1.3440438871473355,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004959125389987193,
|
|
"loss": 5.9257,
|
|
"mean_token_accuracy": 0.13172226995229722,
|
|
"num_tokens": 3012178.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.932300090789795,
|
|
"epoch": 1.347962382445141,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004958552682231567,
|
|
"loss": 5.8234,
|
|
"mean_token_accuracy": 0.14171281978487968,
|
|
"num_tokens": 3021685.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.928375720977783,
|
|
"epoch": 1.3518808777429467,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004957976027434199,
|
|
"loss": 5.8741,
|
|
"mean_token_accuracy": 0.13452255502343177,
|
|
"num_tokens": 3030667.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.019004249572754,
|
|
"epoch": 1.3557993730407523,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004957395426625663,
|
|
"loss": 5.9333,
|
|
"mean_token_accuracy": 0.13655466735363006,
|
|
"num_tokens": 3039501.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.933786916732788,
|
|
"epoch": 1.359717868338558,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004956810880843587,
|
|
"loss": 5.8624,
|
|
"mean_token_accuracy": 0.13548908457159997,
|
|
"num_tokens": 3048547.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.0302653312683105,
|
|
"epoch": 1.3636363636363638,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000495622239113265,
|
|
"loss": 5.8121,
|
|
"mean_token_accuracy": 0.14338299483060837,
|
|
"num_tokens": 3056521.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.849083042144775,
|
|
"epoch": 1.3675548589341693,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004955629958544577,
|
|
"loss": 5.8126,
|
|
"mean_token_accuracy": 0.14086603075265886,
|
|
"num_tokens": 3066007.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.053999328613282,
|
|
"epoch": 1.3714733542319748,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004955033584138143,
|
|
"loss": 6.0003,
|
|
"mean_token_accuracy": 0.12978003472089766,
|
|
"num_tokens": 3075111.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.875246000289917,
|
|
"epoch": 1.3753918495297806,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004954433268979164,
|
|
"loss": 5.7881,
|
|
"mean_token_accuracy": 0.152919040620327,
|
|
"num_tokens": 3083766.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.886160612106323,
|
|
"epoch": 1.3793103448275863,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004953829014140502,
|
|
"loss": 5.8577,
|
|
"mean_token_accuracy": 0.13894471079111098,
|
|
"num_tokens": 3092103.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.972563314437866,
|
|
"epoch": 1.3832288401253918,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004953220820702057,
|
|
"loss": 5.8714,
|
|
"mean_token_accuracy": 0.13644422814249993,
|
|
"num_tokens": 3101286.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.092113876342774,
|
|
"epoch": 1.3871473354231976,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004952608689750771,
|
|
"loss": 5.8627,
|
|
"mean_token_accuracy": 0.1330120176076889,
|
|
"num_tokens": 3110554.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.918315887451172,
|
|
"epoch": 1.391065830721003,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004951992622380619,
|
|
"loss": 5.7475,
|
|
"mean_token_accuracy": 0.1461511582136154,
|
|
"num_tokens": 3119265.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.870513153076172,
|
|
"epoch": 1.3949843260188088,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004951372619692615,
|
|
"loss": 5.7627,
|
|
"mean_token_accuracy": 0.14474907889962196,
|
|
"num_tokens": 3127962.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.979340076446533,
|
|
"epoch": 1.3989028213166144,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004950748682794804,
|
|
"loss": 5.9175,
|
|
"mean_token_accuracy": 0.13927276879549028,
|
|
"num_tokens": 3137265.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.829528379440307,
|
|
"epoch": 1.40282131661442,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004950120812802262,
|
|
"loss": 5.7332,
|
|
"mean_token_accuracy": 0.14469338655471803,
|
|
"num_tokens": 3146458.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.006521940231323,
|
|
"epoch": 1.4067398119122256,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004949489010837095,
|
|
"loss": 5.8905,
|
|
"mean_token_accuracy": 0.13321248143911363,
|
|
"num_tokens": 3155557.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.917591524124146,
|
|
"epoch": 1.4106583072100314,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 5.7598,
|
|
"mean_token_accuracy": 0.15007541924715043,
|
|
"num_tokens": 3163478.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.903986883163452,
|
|
"epoch": 1.414576802507837,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000494821361551244,
|
|
"loss": 5.8502,
|
|
"mean_token_accuracy": 0.13634541109204293,
|
|
"num_tokens": 3172632.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.879254722595215,
|
|
"epoch": 1.4184952978056427,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004947570024432291,
|
|
"loss": 5.897,
|
|
"mean_token_accuracy": 0.14529131725430489,
|
|
"num_tokens": 3181325.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.980233764648437,
|
|
"epoch": 1.4224137931034484,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004946922505938189,
|
|
"loss": 5.9481,
|
|
"mean_token_accuracy": 0.14158461317420007,
|
|
"num_tokens": 3189783.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.982886266708374,
|
|
"epoch": 1.426332288401254,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004946271061187354,
|
|
"loss": 5.7849,
|
|
"mean_token_accuracy": 0.1370651327073574,
|
|
"num_tokens": 3198605.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.875323295593262,
|
|
"epoch": 1.4302507836990594,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004945615691344025,
|
|
"loss": 5.8894,
|
|
"mean_token_accuracy": 0.13310048058629037,
|
|
"num_tokens": 3208061.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.941305637359619,
|
|
"epoch": 1.4341692789968652,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004944956397579453,
|
|
"loss": 5.7912,
|
|
"mean_token_accuracy": 0.14802727848291397,
|
|
"num_tokens": 3216093.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.901020240783692,
|
|
"epoch": 1.438087774294671,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004944293181071902,
|
|
"loss": 5.7855,
|
|
"mean_token_accuracy": 0.14875897616147996,
|
|
"num_tokens": 3224446.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.942932271957398,
|
|
"epoch": 1.4420062695924765,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004943626043006649,
|
|
"loss": 5.8867,
|
|
"mean_token_accuracy": 0.13621754497289656,
|
|
"num_tokens": 3233398.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.9624693393707275,
|
|
"epoch": 1.4459247648902822,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000494295498457598,
|
|
"loss": 5.9052,
|
|
"mean_token_accuracy": 0.13407745957374573,
|
|
"num_tokens": 3243391.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.926957750320435,
|
|
"epoch": 1.4498432601880877,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004942280006979179,
|
|
"loss": 5.84,
|
|
"mean_token_accuracy": 0.14388877302408218,
|
|
"num_tokens": 3251954.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.921461391448974,
|
|
"epoch": 1.4537617554858935,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004941601111422546,
|
|
"loss": 5.8504,
|
|
"mean_token_accuracy": 0.14268624037504196,
|
|
"num_tokens": 3261121.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.826611518859863,
|
|
"epoch": 1.457680250783699,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004940918299119375,
|
|
"loss": 5.7798,
|
|
"mean_token_accuracy": 0.14540517330169678,
|
|
"num_tokens": 3270179.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.817315769195557,
|
|
"epoch": 1.4615987460815048,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004940231571289962,
|
|
"loss": 5.7913,
|
|
"mean_token_accuracy": 0.13905104100704194,
|
|
"num_tokens": 3279749.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.950955724716186,
|
|
"epoch": 1.4655172413793103,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004939540929161603,
|
|
"loss": 5.7957,
|
|
"mean_token_accuracy": 0.1480885237455368,
|
|
"num_tokens": 3288458.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.870092535018921,
|
|
"epoch": 1.469435736677116,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004938846373968586,
|
|
"loss": 5.7682,
|
|
"mean_token_accuracy": 0.14406080767512322,
|
|
"num_tokens": 3297231.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.94631724357605,
|
|
"epoch": 1.4733542319749215,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004938147906952194,
|
|
"loss": 5.9334,
|
|
"mean_token_accuracy": 0.14153699949383736,
|
|
"num_tokens": 3306915.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.926647710800171,
|
|
"epoch": 1.4772727272727273,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00049374455293607,
|
|
"loss": 5.886,
|
|
"mean_token_accuracy": 0.13828219845891,
|
|
"num_tokens": 3315985.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.947432613372802,
|
|
"epoch": 1.481191222570533,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004936739242449369,
|
|
"loss": 5.8505,
|
|
"mean_token_accuracy": 0.13950431123375892,
|
|
"num_tokens": 3324924.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.935120105743408,
|
|
"epoch": 1.4851097178683386,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004936029047480447,
|
|
"loss": 5.8462,
|
|
"mean_token_accuracy": 0.1395495243370533,
|
|
"num_tokens": 3333464.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.889332723617554,
|
|
"epoch": 1.489028213166144,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004935314945723171,
|
|
"loss": 5.7621,
|
|
"mean_token_accuracy": 0.1458078570663929,
|
|
"num_tokens": 3342441.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.865824174880982,
|
|
"epoch": 1.4929467084639498,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004934596938453754,
|
|
"loss": 5.801,
|
|
"mean_token_accuracy": 0.13902750983834267,
|
|
"num_tokens": 3351338.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.9495453357696535,
|
|
"epoch": 1.4968652037617556,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004933875026955391,
|
|
"loss": 5.7673,
|
|
"mean_token_accuracy": 0.14163718521595,
|
|
"num_tokens": 3360356.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.913040637969971,
|
|
"epoch": 1.500783699059561,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004933149212518258,
|
|
"loss": 5.869,
|
|
"mean_token_accuracy": 0.1361314684152603,
|
|
"num_tokens": 3369485.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.876099491119385,
|
|
"epoch": 1.5047021943573666,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004932419496439501,
|
|
"loss": 5.8161,
|
|
"mean_token_accuracy": 0.1474532313644886,
|
|
"num_tokens": 3378355.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.891150426864624,
|
|
"epoch": 1.5086206896551724,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000493168588002324,
|
|
"loss": 5.7252,
|
|
"mean_token_accuracy": 0.14671744257211686,
|
|
"num_tokens": 3386729.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.020740795135498,
|
|
"epoch": 1.5125391849529781,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004930948364580569,
|
|
"loss": 5.9521,
|
|
"mean_token_accuracy": 0.13383505940437318,
|
|
"num_tokens": 3395822.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.8799830913543705,
|
|
"epoch": 1.5164576802507836,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004930206951429546,
|
|
"loss": 5.8552,
|
|
"mean_token_accuracy": 0.14399714767932892,
|
|
"num_tokens": 3404671.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.891398096084595,
|
|
"epoch": 1.5203761755485894,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004929461641895197,
|
|
"loss": 5.7791,
|
|
"mean_token_accuracy": 0.14337211176753045,
|
|
"num_tokens": 3413289.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.854188442230225,
|
|
"epoch": 1.5242946708463951,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000492871243730951,
|
|
"loss": 5.8445,
|
|
"mean_token_accuracy": 0.14026380628347396,
|
|
"num_tokens": 3421960.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.937834072113037,
|
|
"epoch": 1.5282131661442007,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004927959339011437,
|
|
"loss": 5.839,
|
|
"mean_token_accuracy": 0.13955261707305908,
|
|
"num_tokens": 3431958.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.740097761154175,
|
|
"epoch": 1.5321316614420062,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004927202348346885,
|
|
"loss": 5.6604,
|
|
"mean_token_accuracy": 0.1493469387292862,
|
|
"num_tokens": 3441438.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.925070333480835,
|
|
"epoch": 1.536050156739812,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000492644146666872,
|
|
"loss": 5.7737,
|
|
"mean_token_accuracy": 0.15302247405052186,
|
|
"num_tokens": 3450176.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.784520673751831,
|
|
"epoch": 1.5399686520376177,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004925676695336761,
|
|
"loss": 5.7783,
|
|
"mean_token_accuracy": 0.1460975795984268,
|
|
"num_tokens": 3458685.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.939446878433228,
|
|
"epoch": 1.5438871473354232,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004924908035717777,
|
|
"loss": 5.9218,
|
|
"mean_token_accuracy": 0.13721313923597336,
|
|
"num_tokens": 3468362.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.996397972106934,
|
|
"epoch": 1.5478056426332287,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000492413548918549,
|
|
"loss": 5.8444,
|
|
"mean_token_accuracy": 0.14055879041552544,
|
|
"num_tokens": 3476791.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.791240644454956,
|
|
"epoch": 1.5517241379310345,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004923359057120563,
|
|
"loss": 5.7567,
|
|
"mean_token_accuracy": 0.14039518237113952,
|
|
"num_tokens": 3486486.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.824171447753907,
|
|
"epoch": 1.5556426332288402,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004922578740910608,
|
|
"loss": 5.7176,
|
|
"mean_token_accuracy": 0.15190425664186477,
|
|
"num_tokens": 3496145.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.760051012039185,
|
|
"epoch": 1.5595611285266457,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004921794541950177,
|
|
"loss": 5.7179,
|
|
"mean_token_accuracy": 0.1465641237795353,
|
|
"num_tokens": 3504644.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.8565061569213865,
|
|
"epoch": 1.5634796238244513,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004921006461640758,
|
|
"loss": 5.7313,
|
|
"mean_token_accuracy": 0.14198062419891358,
|
|
"num_tokens": 3513126.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.699295711517334,
|
|
"epoch": 1.567398119122257,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000492021450139078,
|
|
"loss": 5.6865,
|
|
"mean_token_accuracy": 0.1480669930577278,
|
|
"num_tokens": 3521505.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.567398119122257,
|
|
"eval_entropy": 5.672700917073922,
|
|
"eval_loss": 6.029964923858643,
|
|
"eval_mean_token_accuracy": 0.13995708463092646,
|
|
"eval_num_tokens": 3521505.0,
|
|
"eval_runtime": 2.855,
|
|
"eval_samples_per_second": 1443.769,
|
|
"eval_steps_per_second": 180.734,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.9574120998382565,
|
|
"epoch": 1.5713166144200628,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004919418662615605,
|
|
"loss": 5.9376,
|
|
"mean_token_accuracy": 0.13867377862334251,
|
|
"num_tokens": 3530008.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.9297913074493405,
|
|
"epoch": 1.5752351097178683,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004918618946737525,
|
|
"loss": 5.8213,
|
|
"mean_token_accuracy": 0.14330511540174484,
|
|
"num_tokens": 3538778.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.865578556060791,
|
|
"epoch": 1.579153605015674,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004917815355185762,
|
|
"loss": 5.7886,
|
|
"mean_token_accuracy": 0.1410187527537346,
|
|
"num_tokens": 3547108.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.835587406158448,
|
|
"epoch": 1.5830721003134798,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004917007889396464,
|
|
"loss": 5.837,
|
|
"mean_token_accuracy": 0.1368314415216446,
|
|
"num_tokens": 3556094.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.858216381072998,
|
|
"epoch": 1.5869905956112853,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004916196550812706,
|
|
"loss": 5.6885,
|
|
"mean_token_accuracy": 0.1482911989092827,
|
|
"num_tokens": 3564222.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.711031723022461,
|
|
"epoch": 1.5909090909090908,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004915381340884477,
|
|
"loss": 5.7125,
|
|
"mean_token_accuracy": 0.15178643986582757,
|
|
"num_tokens": 3573123.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.912963628768921,
|
|
"epoch": 1.5948275862068966,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 5.7347,
|
|
"mean_token_accuracy": 0.14733487069606782,
|
|
"num_tokens": 3581462.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.7520280361175535,
|
|
"epoch": 1.5987460815047023,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004913739312829181,
|
|
"loss": 5.7551,
|
|
"mean_token_accuracy": 0.1488973803818226,
|
|
"num_tokens": 3590718.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.812071180343628,
|
|
"epoch": 1.6026645768025078,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004912912497636683,
|
|
"loss": 5.7788,
|
|
"mean_token_accuracy": 0.14240214824676514,
|
|
"num_tokens": 3599890.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.9684983253479,
|
|
"epoch": 1.6065830721003134,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004912081816968853,
|
|
"loss": 5.8616,
|
|
"mean_token_accuracy": 0.14224711507558824,
|
|
"num_tokens": 3607927.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.866142892837525,
|
|
"epoch": 1.6105015673981191,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000491124727231025,
|
|
"loss": 5.7005,
|
|
"mean_token_accuracy": 0.15074294060468674,
|
|
"num_tokens": 3615935.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.735476112365722,
|
|
"epoch": 1.6144200626959249,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004910408865152343,
|
|
"loss": 5.7498,
|
|
"mean_token_accuracy": 0.14708520323038102,
|
|
"num_tokens": 3624475.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.000697422027588,
|
|
"epoch": 1.6183385579937304,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004909566596993498,
|
|
"loss": 5.9374,
|
|
"mean_token_accuracy": 0.1381327621638775,
|
|
"num_tokens": 3633995.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.861016702651978,
|
|
"epoch": 1.622257053291536,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004908720469338988,
|
|
"loss": 5.8094,
|
|
"mean_token_accuracy": 0.14452051296830176,
|
|
"num_tokens": 3643121.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.835006523132324,
|
|
"epoch": 1.6261755485893417,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004907870483700979,
|
|
"loss": 5.7401,
|
|
"mean_token_accuracy": 0.1410243555903435,
|
|
"num_tokens": 3652285.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.813826656341552,
|
|
"epoch": 1.6300940438871474,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004907016641598534,
|
|
"loss": 5.8232,
|
|
"mean_token_accuracy": 0.14193576872348784,
|
|
"num_tokens": 3661441.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.849268865585327,
|
|
"epoch": 1.634012539184953,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004906158944557607,
|
|
"loss": 5.6933,
|
|
"mean_token_accuracy": 0.15349719524383545,
|
|
"num_tokens": 3669364.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.878481912612915,
|
|
"epoch": 1.6379310344827587,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000490529739411104,
|
|
"loss": 5.8009,
|
|
"mean_token_accuracy": 0.14380127266049386,
|
|
"num_tokens": 3677848.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.790279960632324,
|
|
"epoch": 1.6418495297805644,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004904431991798565,
|
|
"loss": 5.6863,
|
|
"mean_token_accuracy": 0.15105993747711183,
|
|
"num_tokens": 3686617.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.82244086265564,
|
|
"epoch": 1.64576802507837,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004903562739166797,
|
|
"loss": 5.6836,
|
|
"mean_token_accuracy": 0.14716468155384063,
|
|
"num_tokens": 3695106.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.731297779083252,
|
|
"epoch": 1.6496865203761755,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004902689637769229,
|
|
"loss": 5.6316,
|
|
"mean_token_accuracy": 0.1544244959950447,
|
|
"num_tokens": 3703167.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.850043249130249,
|
|
"epoch": 1.6536050156739812,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004901812689166237,
|
|
"loss": 5.7411,
|
|
"mean_token_accuracy": 0.14633739218115807,
|
|
"num_tokens": 3711597.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.808914756774902,
|
|
"epoch": 1.657523510971787,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004900931894925069,
|
|
"loss": 5.8445,
|
|
"mean_token_accuracy": 0.14314616173505784,
|
|
"num_tokens": 3720277.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.803114509582519,
|
|
"epoch": 1.6614420062695925,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004900047256619849,
|
|
"loss": 5.8091,
|
|
"mean_token_accuracy": 0.1463458240032196,
|
|
"num_tokens": 3729832.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.832270526885987,
|
|
"epoch": 1.665360501567398,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004899158775831566,
|
|
"loss": 5.7648,
|
|
"mean_token_accuracy": 0.15041996985673906,
|
|
"num_tokens": 3738390.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.838447761535645,
|
|
"epoch": 1.6692789968652038,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004898266454148081,
|
|
"loss": 5.8082,
|
|
"mean_token_accuracy": 0.1450522780418396,
|
|
"num_tokens": 3747046.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.7696085453033445,
|
|
"epoch": 1.6731974921630095,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004897370293164119,
|
|
"loss": 5.7837,
|
|
"mean_token_accuracy": 0.13788653165102005,
|
|
"num_tokens": 3755818.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.014819002151489,
|
|
"epoch": 1.677115987460815,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004896470294481262,
|
|
"loss": 5.873,
|
|
"mean_token_accuracy": 0.1447507619857788,
|
|
"num_tokens": 3764123.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.868058919906616,
|
|
"epoch": 1.6810344827586206,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004895566459707954,
|
|
"loss": 5.928,
|
|
"mean_token_accuracy": 0.1357198879122734,
|
|
"num_tokens": 3773105.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.8731566905975345,
|
|
"epoch": 1.6849529780564263,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004894658790459498,
|
|
"loss": 5.7728,
|
|
"mean_token_accuracy": 0.1499609887599945,
|
|
"num_tokens": 3781417.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.944635963439941,
|
|
"epoch": 1.688871473354232,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004893747288358041,
|
|
"loss": 5.8672,
|
|
"mean_token_accuracy": 0.14666299670934677,
|
|
"num_tokens": 3789657.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.897546195983887,
|
|
"epoch": 1.6927899686520376,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000489283195503259,
|
|
"loss": 5.8762,
|
|
"mean_token_accuracy": 0.14236082658171653,
|
|
"num_tokens": 3798700.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.820228528976441,
|
|
"epoch": 1.6967084639498433,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000489191279211899,
|
|
"loss": 5.7223,
|
|
"mean_token_accuracy": 0.14710770919919014,
|
|
"num_tokens": 3807299.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.885312223434449,
|
|
"epoch": 1.700626959247649,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004890989801259935,
|
|
"loss": 5.8276,
|
|
"mean_token_accuracy": 0.1482617050409317,
|
|
"num_tokens": 3816292.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.931986904144287,
|
|
"epoch": 1.7045454545454546,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000489006298410496,
|
|
"loss": 5.7392,
|
|
"mean_token_accuracy": 0.14349312484264373,
|
|
"num_tokens": 3825070.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.804163599014283,
|
|
"epoch": 1.70846394984326,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004889132342310438,
|
|
"loss": 5.7293,
|
|
"mean_token_accuracy": 0.14909339994192122,
|
|
"num_tokens": 3834016.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.880694341659546,
|
|
"epoch": 1.7123824451410659,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004888197877539577,
|
|
"loss": 5.7672,
|
|
"mean_token_accuracy": 0.1458732567727566,
|
|
"num_tokens": 3842172.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.79048523902893,
|
|
"epoch": 1.7163009404388716,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004887259591462417,
|
|
"loss": 5.683,
|
|
"mean_token_accuracy": 0.14992706328630448,
|
|
"num_tokens": 3851109.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.782387828826904,
|
|
"epoch": 1.7202194357366771,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004886317485755825,
|
|
"loss": 5.796,
|
|
"mean_token_accuracy": 0.14876937195658685,
|
|
"num_tokens": 3859656.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.809633064270019,
|
|
"epoch": 1.7241379310344827,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004885371562103498,
|
|
"loss": 5.6571,
|
|
"mean_token_accuracy": 0.15718313902616501,
|
|
"num_tokens": 3868698.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.76657247543335,
|
|
"epoch": 1.7280564263322884,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004884421822195957,
|
|
"loss": 5.7444,
|
|
"mean_token_accuracy": 0.14893446266651153,
|
|
"num_tokens": 3877474.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.85271372795105,
|
|
"epoch": 1.7319749216300941,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004883468267730538,
|
|
"loss": 5.7336,
|
|
"mean_token_accuracy": 0.14580635279417037,
|
|
"num_tokens": 3886328.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.817337465286255,
|
|
"epoch": 1.7358934169278997,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048825109004114006,
|
|
"loss": 5.6926,
|
|
"mean_token_accuracy": 0.15190036296844484,
|
|
"num_tokens": 3894424.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.8421002388000485,
|
|
"epoch": 1.7398119122257052,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004881549721949513,
|
|
"loss": 5.8065,
|
|
"mean_token_accuracy": 0.1421722039580345,
|
|
"num_tokens": 3903746.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.735415267944336,
|
|
"epoch": 1.743730407523511,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004880584734062658,
|
|
"loss": 5.7647,
|
|
"mean_token_accuracy": 0.1507815882563591,
|
|
"num_tokens": 3912033.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.788500356674194,
|
|
"epoch": 1.7476489028213167,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004879615938475425,
|
|
"loss": 5.6733,
|
|
"mean_token_accuracy": 0.14568700939416884,
|
|
"num_tokens": 3920388.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.820400190353394,
|
|
"epoch": 1.7515673981191222,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004878643336919209,
|
|
"loss": 5.8767,
|
|
"mean_token_accuracy": 0.14171729907393454,
|
|
"num_tokens": 3929926.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.996213912963867,
|
|
"epoch": 1.7554858934169277,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004877666931132206,
|
|
"loss": 5.8413,
|
|
"mean_token_accuracy": 0.13652574941515921,
|
|
"num_tokens": 3938696.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.85072169303894,
|
|
"epoch": 1.7594043887147337,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004876686722859413,
|
|
"loss": 5.6819,
|
|
"mean_token_accuracy": 0.1550109453499317,
|
|
"num_tokens": 3947537.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.824322128295899,
|
|
"epoch": 1.7633228840125392,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000487570271385262,
|
|
"loss": 5.6797,
|
|
"mean_token_accuracy": 0.14713258892297745,
|
|
"num_tokens": 3956539.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.770116233825684,
|
|
"epoch": 1.7672413793103448,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004874714905870411,
|
|
"loss": 5.6556,
|
|
"mean_token_accuracy": 0.14795414805412294,
|
|
"num_tokens": 3964927.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.781585693359375,
|
|
"epoch": 1.7711598746081505,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004873723300678159,
|
|
"loss": 5.7993,
|
|
"mean_token_accuracy": 0.14609729051589965,
|
|
"num_tokens": 3973565.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.8436461925506595,
|
|
"epoch": 1.7750783699059562,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048727279000480226,
|
|
"loss": 5.8134,
|
|
"mean_token_accuracy": 0.1419944316148758,
|
|
"num_tokens": 3982683.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.901409387588501,
|
|
"epoch": 1.7789968652037618,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 5.7444,
|
|
"mean_token_accuracy": 0.14614344835281373,
|
|
"num_tokens": 3991462.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.5885340690612795,
|
|
"epoch": 1.7829153605015673,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004870725719596648,
|
|
"loss": 5.4992,
|
|
"mean_token_accuracy": 0.15931711345911026,
|
|
"num_tokens": 3999821.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.817122077941894,
|
|
"epoch": 1.786833855799373,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004869718943353631,
|
|
"loss": 5.7526,
|
|
"mean_token_accuracy": 0.14650990962982177,
|
|
"num_tokens": 4008655.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.852270984649659,
|
|
"epoch": 1.7907523510971788,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048687083788291656,
|
|
"loss": 5.7852,
|
|
"mean_token_accuracy": 0.15168624818325044,
|
|
"num_tokens": 4017317.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.706546688079834,
|
|
"epoch": 1.7946708463949843,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048676940278292953,
|
|
"loss": 5.6183,
|
|
"mean_token_accuracy": 0.15399519652128218,
|
|
"num_tokens": 4025407.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.778053712844849,
|
|
"epoch": 1.7985893416927898,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048666758921668286,
|
|
"loss": 5.6273,
|
|
"mean_token_accuracy": 0.15056194961071015,
|
|
"num_tokens": 4034310.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.775419092178344,
|
|
"epoch": 1.8025078369905956,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048656539736613403,
|
|
"loss": 5.7907,
|
|
"mean_token_accuracy": 0.14853218123316764,
|
|
"num_tokens": 4042326.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.706247758865357,
|
|
"epoch": 1.8064263322884013,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004864628274139164,
|
|
"loss": 5.6481,
|
|
"mean_token_accuracy": 0.15559574365615844,
|
|
"num_tokens": 4051248.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.8236981391906735,
|
|
"epoch": 1.8103448275862069,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004863598795433391,
|
|
"loss": 5.7786,
|
|
"mean_token_accuracy": 0.14639326483011245,
|
|
"num_tokens": 4060227.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.91852650642395,
|
|
"epoch": 1.8142633228840124,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048625655393838666,
|
|
"loss": 5.8551,
|
|
"mean_token_accuracy": 0.1396035686135292,
|
|
"num_tokens": 4069700.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.817505264282227,
|
|
"epoch": 1.8181818181818183,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004861528507837186,
|
|
"loss": 5.7226,
|
|
"mean_token_accuracy": 0.14971303641796113,
|
|
"num_tokens": 4079363.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.909413576126099,
|
|
"epoch": 1.8221003134796239,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004860487702646695,
|
|
"loss": 5.8089,
|
|
"mean_token_accuracy": 0.14138850718736648,
|
|
"num_tokens": 4087944.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.663079452514649,
|
|
"epoch": 1.8260188087774294,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004859443125672479,
|
|
"loss": 5.5141,
|
|
"mean_token_accuracy": 0.16887633204460145,
|
|
"num_tokens": 4096431.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.716325759887695,
|
|
"epoch": 1.8299373040752351,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004858394778781368,
|
|
"loss": 5.6716,
|
|
"mean_token_accuracy": 0.14433413445949556,
|
|
"num_tokens": 4104894.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.810530185699463,
|
|
"epoch": 1.8338557993730409,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004857342663846927,
|
|
"loss": 5.7169,
|
|
"mean_token_accuracy": 0.14703929871320726,
|
|
"num_tokens": 4112878.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.7495544910430905,
|
|
"epoch": 1.8377742946708464,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004856286782749456,
|
|
"loss": 5.7282,
|
|
"mean_token_accuracy": 0.15359484106302262,
|
|
"num_tokens": 4122501.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.722039604187012,
|
|
"epoch": 1.841692789968652,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004855227137375986,
|
|
"loss": 5.6971,
|
|
"mean_token_accuracy": 0.14616180807352067,
|
|
"num_tokens": 4131301.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.7759270668029785,
|
|
"epoch": 1.8456112852664577,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004854163729620275,
|
|
"loss": 5.7346,
|
|
"mean_token_accuracy": 0.15014693140983582,
|
|
"num_tokens": 4140076.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.80101170539856,
|
|
"epoch": 1.8495297805642634,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004853096561382805,
|
|
"loss": 5.6482,
|
|
"mean_token_accuracy": 0.1573817789554596,
|
|
"num_tokens": 4148728.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.722895526885987,
|
|
"epoch": 1.853448275862069,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004852025634570779,
|
|
"loss": 5.817,
|
|
"mean_token_accuracy": 0.14882018342614173,
|
|
"num_tokens": 4157520.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.815022802352905,
|
|
"epoch": 1.8573667711598745,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004850950951098116,
|
|
"loss": 5.6208,
|
|
"mean_token_accuracy": 0.15457617491483688,
|
|
"num_tokens": 4165547.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.7170045375823975,
|
|
"epoch": 1.8612852664576802,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004849872512885451,
|
|
"loss": 5.7669,
|
|
"mean_token_accuracy": 0.14189089238643646,
|
|
"num_tokens": 4174656.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.836828517913818,
|
|
"epoch": 1.865203761755486,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004848790321860127,
|
|
"loss": 5.7994,
|
|
"mean_token_accuracy": 0.1457727000117302,
|
|
"num_tokens": 4183452.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.839574909210205,
|
|
"epoch": 1.8691222570532915,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048477043799561946,
|
|
"loss": 5.7025,
|
|
"mean_token_accuracy": 0.14705116599798201,
|
|
"num_tokens": 4192306.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.765735483169555,
|
|
"epoch": 1.873040752351097,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004846614689114409,
|
|
"loss": 5.7357,
|
|
"mean_token_accuracy": 0.14695977419614792,
|
|
"num_tokens": 4201201.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.742426204681396,
|
|
"epoch": 1.876959247648903,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004845521251282223,
|
|
"loss": 5.6466,
|
|
"mean_token_accuracy": 0.15739916563034057,
|
|
"num_tokens": 4209374.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.652604055404663,
|
|
"epoch": 1.8808777429467085,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004844424068413789,
|
|
"loss": 5.6019,
|
|
"mean_token_accuracy": 0.15140311568975448,
|
|
"num_tokens": 4217258.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.726423931121826,
|
|
"epoch": 1.884796238244514,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048433231424699504,
|
|
"loss": 5.6578,
|
|
"mean_token_accuracy": 0.15339543521404267,
|
|
"num_tokens": 4226484.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.76325569152832,
|
|
"epoch": 1.8887147335423198,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048422184754182384,
|
|
"loss": 5.6794,
|
|
"mean_token_accuracy": 0.1510704293847084,
|
|
"num_tokens": 4235546.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.839502143859863,
|
|
"epoch": 1.8926332288401255,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004841110069232875,
|
|
"loss": 5.8354,
|
|
"mean_token_accuracy": 0.14865082055330275,
|
|
"num_tokens": 4244549.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.769964933395386,
|
|
"epoch": 1.896551724137931,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048399979258947597,
|
|
"loss": 5.7221,
|
|
"mean_token_accuracy": 0.15206404626369477,
|
|
"num_tokens": 4252918.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.813162088394165,
|
|
"epoch": 1.9004702194357366,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004838882047391474,
|
|
"loss": 5.7137,
|
|
"mean_token_accuracy": 0.14775785952806472,
|
|
"num_tokens": 4261674.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.754944515228272,
|
|
"epoch": 1.9043887147335423,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048377624357172724,
|
|
"loss": 5.6275,
|
|
"mean_token_accuracy": 0.15521396249532698,
|
|
"num_tokens": 4269986.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.772288084030151,
|
|
"epoch": 1.908307210031348,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048366390928730843,
|
|
"loss": 5.7277,
|
|
"mean_token_accuracy": 0.15137290358543395,
|
|
"num_tokens": 4278131.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.772933292388916,
|
|
"epoch": 1.9122257053291536,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004835512020866504,
|
|
"loss": 5.6901,
|
|
"mean_token_accuracy": 0.14994694739580156,
|
|
"num_tokens": 4287214.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.747813320159912,
|
|
"epoch": 1.9161442006269591,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048343812217117925,
|
|
"loss": 5.6528,
|
|
"mean_token_accuracy": 0.15408559143543243,
|
|
"num_tokens": 4295949.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.741509056091308,
|
|
"epoch": 1.9200626959247649,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048332466974298723,
|
|
"loss": 5.6356,
|
|
"mean_token_accuracy": 0.15061129480600358,
|
|
"num_tokens": 4305014.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.716789960861206,
|
|
"epoch": 1.9239811912225706,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048321084500483203,
|
|
"loss": 5.6184,
|
|
"mean_token_accuracy": 0.15444232225418092,
|
|
"num_tokens": 4313560.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.753894281387329,
|
|
"epoch": 1.9278996865203761,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000483096648160137,
|
|
"loss": 5.7105,
|
|
"mean_token_accuracy": 0.15123845040798187,
|
|
"num_tokens": 4323280.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.806096601486206,
|
|
"epoch": 1.9318181818181817,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048298207941299047,
|
|
"loss": 5.747,
|
|
"mean_token_accuracy": 0.1527293801307678,
|
|
"num_tokens": 4331757.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.656486320495605,
|
|
"epoch": 1.9357366771159876,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048286713896814536,
|
|
"loss": 5.6845,
|
|
"mean_token_accuracy": 0.15062677338719369,
|
|
"num_tokens": 4341453.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.872777032852173,
|
|
"epoch": 1.9396551724137931,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048275182703101877,
|
|
"loss": 5.6272,
|
|
"mean_token_accuracy": 0.15510055273771287,
|
|
"num_tokens": 4349095.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.647492599487305,
|
|
"epoch": 1.9435736677115987,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048263614380769193,
|
|
"loss": 5.734,
|
|
"mean_token_accuracy": 0.14693398848176004,
|
|
"num_tokens": 4357913.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.7816098690032955,
|
|
"epoch": 1.9474921630094044,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048252008950490957,
|
|
"loss": 5.7042,
|
|
"mean_token_accuracy": 0.1510899305343628,
|
|
"num_tokens": 4366586.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.780308771133423,
|
|
"epoch": 1.9514106583072102,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048240366433007935,
|
|
"loss": 5.6439,
|
|
"mean_token_accuracy": 0.1548413008451462,
|
|
"num_tokens": 4375393.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.719468688964843,
|
|
"epoch": 1.9553291536050157,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048228686849127213,
|
|
"loss": 5.707,
|
|
"mean_token_accuracy": 0.15866673439741136,
|
|
"num_tokens": 4383747.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.827725028991699,
|
|
"epoch": 1.9592476489028212,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004821697021972209,
|
|
"loss": 5.69,
|
|
"mean_token_accuracy": 0.14475997686386108,
|
|
"num_tokens": 4392085.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 1.9592476489028212,
|
|
"eval_entropy": 5.705850722253785,
|
|
"eval_loss": 5.8995537757873535,
|
|
"eval_mean_token_accuracy": 0.14770855615276468,
|
|
"eval_num_tokens": 4392085.0,
|
|
"eval_runtime": 3.0359,
|
|
"eval_samples_per_second": 1357.773,
|
|
"eval_steps_per_second": 169.969,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.789532089233399,
|
|
"epoch": 1.963166144200627,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 5.7037,
|
|
"mean_token_accuracy": 0.1507359117269516,
|
|
"num_tokens": 4400843.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.641516876220703,
|
|
"epoch": 1.9670846394984327,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004819342590816288,
|
|
"loss": 5.664,
|
|
"mean_token_accuracy": 0.15229431241750718,
|
|
"num_tokens": 4409126.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.780962181091309,
|
|
"epoch": 1.9710031347962382,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004818159826808631,
|
|
"loss": 5.7947,
|
|
"mean_token_accuracy": 0.14660751819610596,
|
|
"num_tokens": 4418401.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.7990799903869625,
|
|
"epoch": 1.9749216300940438,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004816973366664026,
|
|
"loss": 5.7212,
|
|
"mean_token_accuracy": 0.1470862217247486,
|
|
"num_tokens": 4428141.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.763505554199218,
|
|
"epoch": 1.9788401253918495,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004815783212502871,
|
|
"loss": 5.7216,
|
|
"mean_token_accuracy": 0.1449996516108513,
|
|
"num_tokens": 4437158.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.798718881607056,
|
|
"epoch": 1.9827586206896552,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00048145893664521645,
|
|
"loss": 5.7848,
|
|
"mean_token_accuracy": 0.1408971928060055,
|
|
"num_tokens": 4445901.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.741735363006592,
|
|
"epoch": 1.9866771159874608,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048133918306455023,
|
|
"loss": 5.6081,
|
|
"mean_token_accuracy": 0.16412827521562576,
|
|
"num_tokens": 4454200.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.753818655014038,
|
|
"epoch": 1.9905956112852663,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004812190607223075,
|
|
"loss": 5.7198,
|
|
"mean_token_accuracy": 0.1448679745197296,
|
|
"num_tokens": 4463484.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.637834119796753,
|
|
"epoch": 1.9945141065830723,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048109856983316655,
|
|
"loss": 5.5579,
|
|
"mean_token_accuracy": 0.15664460957050325,
|
|
"num_tokens": 4472659.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.7259618759155275,
|
|
"epoch": 1.9984326018808778,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000480977710612464,
|
|
"loss": 5.6156,
|
|
"mean_token_accuracy": 0.14887530580163003,
|
|
"num_tokens": 4481355.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.752556180953979,
|
|
"epoch": 2.0023510971786833,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004808564832761948,
|
|
"loss": 5.4781,
|
|
"mean_token_accuracy": 0.15195177793502807,
|
|
"num_tokens": 4490005.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.674878787994385,
|
|
"epoch": 2.006269592476489,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004807348880410119,
|
|
"loss": 5.2969,
|
|
"mean_token_accuracy": 0.163840214908123,
|
|
"num_tokens": 4498645.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.641942977905273,
|
|
"epoch": 2.010188087774295,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004806129251242258,
|
|
"loss": 5.3845,
|
|
"mean_token_accuracy": 0.15954576134681703,
|
|
"num_tokens": 4509675.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.6735498905181885,
|
|
"epoch": 2.0141065830721003,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048049059474380393,
|
|
"loss": 5.2179,
|
|
"mean_token_accuracy": 0.166520294547081,
|
|
"num_tokens": 4518463.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.706894254684448,
|
|
"epoch": 2.018025078369906,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048036789711837047,
|
|
"loss": 5.3649,
|
|
"mean_token_accuracy": 0.15959171056747437,
|
|
"num_tokens": 4527120.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.680517148971558,
|
|
"epoch": 2.0219435736677114,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048024483246720607,
|
|
"loss": 5.397,
|
|
"mean_token_accuracy": 0.15788237601518632,
|
|
"num_tokens": 4535962.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.706279850006103,
|
|
"epoch": 2.0258620689655173,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004801214010102472,
|
|
"loss": 5.3536,
|
|
"mean_token_accuracy": 0.15642207115888596,
|
|
"num_tokens": 4545233.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.604188776016235,
|
|
"epoch": 2.029780564263323,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004799976029680858,
|
|
"loss": 5.227,
|
|
"mean_token_accuracy": 0.16073910295963287,
|
|
"num_tokens": 4554658.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.717430782318115,
|
|
"epoch": 2.0336990595611284,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004798734385619691,
|
|
"loss": 5.4145,
|
|
"mean_token_accuracy": 0.16037535965442656,
|
|
"num_tokens": 4563964.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.666239976882935,
|
|
"epoch": 2.0376175548589344,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000479748908013799,
|
|
"loss": 5.2896,
|
|
"mean_token_accuracy": 0.1706826001405716,
|
|
"num_tokens": 4572865.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.570950031280518,
|
|
"epoch": 2.04153605015674,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004796240115461319,
|
|
"loss": 5.2642,
|
|
"mean_token_accuracy": 0.17284168303012848,
|
|
"num_tokens": 4581021.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.529971885681152,
|
|
"epoch": 2.0454545454545454,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004794987493821779,
|
|
"loss": 5.3565,
|
|
"mean_token_accuracy": 0.16215605586767196,
|
|
"num_tokens": 4590779.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.651610660552978,
|
|
"epoch": 2.049373040752351,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047937312174580084,
|
|
"loss": 5.3414,
|
|
"mean_token_accuracy": 0.16156901866197587,
|
|
"num_tokens": 4599831.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.721800804138184,
|
|
"epoch": 2.053291536050157,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004792471288615177,
|
|
"loss": 5.4688,
|
|
"mean_token_accuracy": 0.1500769391655922,
|
|
"num_tokens": 4609264.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.70226821899414,
|
|
"epoch": 2.0572100313479624,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004791207709544981,
|
|
"loss": 5.374,
|
|
"mean_token_accuracy": 0.1676947593688965,
|
|
"num_tokens": 4617885.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.5007805824279785,
|
|
"epoch": 2.061128526645768,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047899404825056424,
|
|
"loss": 5.3022,
|
|
"mean_token_accuracy": 0.16264875382184982,
|
|
"num_tokens": 4626536.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.664446115493774,
|
|
"epoch": 2.0650470219435735,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004788669609761901,
|
|
"loss": 5.4064,
|
|
"mean_token_accuracy": 0.1559653714299202,
|
|
"num_tokens": 4635563.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.576727724075317,
|
|
"epoch": 2.0689655172413794,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047873950935850107,
|
|
"loss": 5.3001,
|
|
"mean_token_accuracy": 0.15983396619558335,
|
|
"num_tokens": 4644822.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.616795778274536,
|
|
"epoch": 2.072884012539185,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004786116936252742,
|
|
"loss": 5.2834,
|
|
"mean_token_accuracy": 0.16793755143880845,
|
|
"num_tokens": 4652709.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.555015659332275,
|
|
"epoch": 2.0768025078369905,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004784835140049367,
|
|
"loss": 5.3014,
|
|
"mean_token_accuracy": 0.16771620512008667,
|
|
"num_tokens": 4662014.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.66242470741272,
|
|
"epoch": 2.080721003134796,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004783549707265663,
|
|
"loss": 5.2894,
|
|
"mean_token_accuracy": 0.1594408079981804,
|
|
"num_tokens": 4670618.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.5888073444366455,
|
|
"epoch": 2.084639498432602,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047822606401989084,
|
|
"loss": 5.4337,
|
|
"mean_token_accuracy": 0.1533523418009281,
|
|
"num_tokens": 4679937.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.6302934169769285,
|
|
"epoch": 2.0885579937304075,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004780967941152873,
|
|
"loss": 5.238,
|
|
"mean_token_accuracy": 0.17676321119070054,
|
|
"num_tokens": 4688006.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.542519521713257,
|
|
"epoch": 2.092476489028213,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004779671612437822,
|
|
"loss": 5.1899,
|
|
"mean_token_accuracy": 0.173385126888752,
|
|
"num_tokens": 4697106.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.543998193740845,
|
|
"epoch": 2.096394984326019,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047783716563705035,
|
|
"loss": 5.3423,
|
|
"mean_token_accuracy": 0.16200381517410278,
|
|
"num_tokens": 4705582.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.640423583984375,
|
|
"epoch": 2.1003134796238245,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000477706807527415,
|
|
"loss": 5.2754,
|
|
"mean_token_accuracy": 0.1639168754220009,
|
|
"num_tokens": 4713640.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.603535509109497,
|
|
"epoch": 2.10423197492163,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004775760871478472,
|
|
"loss": 5.4069,
|
|
"mean_token_accuracy": 0.160996450483799,
|
|
"num_tokens": 4722729.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.627258968353272,
|
|
"epoch": 2.1081504702194356,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047744500473196564,
|
|
"loss": 5.2855,
|
|
"mean_token_accuracy": 0.1675833374261856,
|
|
"num_tokens": 4732099.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.648689317703247,
|
|
"epoch": 2.1120689655172415,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047731356051403556,
|
|
"loss": 5.3168,
|
|
"mean_token_accuracy": 0.15454934388399125,
|
|
"num_tokens": 4741766.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.469576168060303,
|
|
"epoch": 2.115987460815047,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004771817547289693,
|
|
"loss": 5.2196,
|
|
"mean_token_accuracy": 0.1669618755578995,
|
|
"num_tokens": 4750890.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.580454540252686,
|
|
"epoch": 2.1199059561128526,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004770495876123251,
|
|
"loss": 5.3545,
|
|
"mean_token_accuracy": 0.16359022557735442,
|
|
"num_tokens": 4760063.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.643666553497314,
|
|
"epoch": 2.123824451410658,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004769170594003071,
|
|
"loss": 5.3426,
|
|
"mean_token_accuracy": 0.1646828219294548,
|
|
"num_tokens": 4768928.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.535561656951904,
|
|
"epoch": 2.127742946708464,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047678417032976457,
|
|
"loss": 5.3436,
|
|
"mean_token_accuracy": 0.1597463920712471,
|
|
"num_tokens": 4778129.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.582833862304687,
|
|
"epoch": 2.1316614420062696,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004766509206381919,
|
|
"loss": 5.3256,
|
|
"mean_token_accuracy": 0.16664166301488875,
|
|
"num_tokens": 4786759.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.543987512588501,
|
|
"epoch": 2.135579937304075,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004765173105637279,
|
|
"loss": 5.3209,
|
|
"mean_token_accuracy": 0.16914047598838805,
|
|
"num_tokens": 4795355.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.563296318054199,
|
|
"epoch": 2.139498432601881,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047638334034515547,
|
|
"loss": 5.2819,
|
|
"mean_token_accuracy": 0.16699936389923095,
|
|
"num_tokens": 4804030.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.451429939270019,
|
|
"epoch": 2.1434169278996866,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047624901022190106,
|
|
"loss": 5.2285,
|
|
"mean_token_accuracy": 0.1854039654135704,
|
|
"num_tokens": 4812779.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.612224864959717,
|
|
"epoch": 2.147335423197492,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.349,
|
|
"mean_token_accuracy": 0.16258929520845414,
|
|
"num_tokens": 4822292.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.713485240936279,
|
|
"epoch": 2.1512539184952977,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004759792712222679,
|
|
"loss": 5.408,
|
|
"mean_token_accuracy": 0.16127049475908278,
|
|
"num_tokens": 4830524.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.600223112106323,
|
|
"epoch": 2.1551724137931036,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004758438628279565,
|
|
"loss": 5.4051,
|
|
"mean_token_accuracy": 0.16541811376810073,
|
|
"num_tokens": 4839328.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.586366415023804,
|
|
"epoch": 2.159090909090909,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047570809549309697,
|
|
"loss": 5.2408,
|
|
"mean_token_accuracy": 0.17708596885204314,
|
|
"num_tokens": 4847711.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.651467561721802,
|
|
"epoch": 2.1630094043887147,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004755719694603275,
|
|
"loss": 5.4092,
|
|
"mean_token_accuracy": 0.16118254661560058,
|
|
"num_tokens": 4856706.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.636726808547974,
|
|
"epoch": 2.16692789968652,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004754354849729274,
|
|
"loss": 5.4163,
|
|
"mean_token_accuracy": 0.16455628871917724,
|
|
"num_tokens": 4865250.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.60169677734375,
|
|
"epoch": 2.170846394984326,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047529864227481653,
|
|
"loss": 5.3433,
|
|
"mean_token_accuracy": 0.16435989141464233,
|
|
"num_tokens": 4873826.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.669528484344482,
|
|
"epoch": 2.1747648902821317,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000475161441610555,
|
|
"loss": 5.3172,
|
|
"mean_token_accuracy": 0.16513787358999252,
|
|
"num_tokens": 4882631.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.560564279556274,
|
|
"epoch": 2.1786833855799372,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004750238832253427,
|
|
"loss": 5.3731,
|
|
"mean_token_accuracy": 0.1643209844827652,
|
|
"num_tokens": 4891007.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.567132472991943,
|
|
"epoch": 2.1826018808777428,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004748859673650187,
|
|
"loss": 5.2642,
|
|
"mean_token_accuracy": 0.1682204231619835,
|
|
"num_tokens": 4899692.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.564448595046997,
|
|
"epoch": 2.1865203761755487,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047474769427606115,
|
|
"loss": 5.4068,
|
|
"mean_token_accuracy": 0.1644543319940567,
|
|
"num_tokens": 4909354.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.57516770362854,
|
|
"epoch": 2.1904388714733543,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004746090642055863,
|
|
"loss": 5.3296,
|
|
"mean_token_accuracy": 0.16534655094146727,
|
|
"num_tokens": 4918400.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.628228569030762,
|
|
"epoch": 2.19435736677116,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047447007740134857,
|
|
"loss": 5.4266,
|
|
"mean_token_accuracy": 0.15290777534246444,
|
|
"num_tokens": 4926852.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.629791641235352,
|
|
"epoch": 2.1982758620689653,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047433073411174,
|
|
"loss": 5.3439,
|
|
"mean_token_accuracy": 0.16628239601850509,
|
|
"num_tokens": 4935717.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.503053808212281,
|
|
"epoch": 2.2021943573667713,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004741910345857896,
|
|
"loss": 5.2523,
|
|
"mean_token_accuracy": 0.16592369377613067,
|
|
"num_tokens": 4945389.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.681710863113404,
|
|
"epoch": 2.206112852664577,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047405097907316315,
|
|
"loss": 5.3965,
|
|
"mean_token_accuracy": 0.1633175492286682,
|
|
"num_tokens": 4953670.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.543265533447266,
|
|
"epoch": 2.2100313479623823,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004739105678241625,
|
|
"loss": 5.4136,
|
|
"mean_token_accuracy": 0.16170129925012589,
|
|
"num_tokens": 4962383.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.524320983886719,
|
|
"epoch": 2.2139498432601883,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004737698010897253,
|
|
"loss": 5.2117,
|
|
"mean_token_accuracy": 0.16661869436502458,
|
|
"num_tokens": 4970672.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.6738903522491455,
|
|
"epoch": 2.217868338557994,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004736286791214245,
|
|
"loss": 5.3747,
|
|
"mean_token_accuracy": 0.15929104834795,
|
|
"num_tokens": 4979303.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.573844003677368,
|
|
"epoch": 2.2217868338557993,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047348720217146807,
|
|
"loss": 5.2594,
|
|
"mean_token_accuracy": 0.17054332792758942,
|
|
"num_tokens": 4988322.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.466978693008423,
|
|
"epoch": 2.225705329153605,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047334537049269806,
|
|
"loss": 5.3355,
|
|
"mean_token_accuracy": 0.16777932047843933,
|
|
"num_tokens": 4997250.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.600286865234375,
|
|
"epoch": 2.229623824451411,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004732031843385909,
|
|
"loss": 5.4267,
|
|
"mean_token_accuracy": 0.1585068553686142,
|
|
"num_tokens": 5005459.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.495544385910034,
|
|
"epoch": 2.2335423197492164,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004730606439632562,
|
|
"loss": 5.1551,
|
|
"mean_token_accuracy": 0.1816363513469696,
|
|
"num_tokens": 5013704.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.431286334991455,
|
|
"epoch": 2.237460815047022,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004729177496214367,
|
|
"loss": 5.288,
|
|
"mean_token_accuracy": 0.1695888951420784,
|
|
"num_tokens": 5022545.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.527831506729126,
|
|
"epoch": 2.2413793103448274,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047277450156850767,
|
|
"loss": 5.3118,
|
|
"mean_token_accuracy": 0.17252337634563447,
|
|
"num_tokens": 5031133.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.604853677749634,
|
|
"epoch": 2.2452978056426334,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004726309000604768,
|
|
"loss": 5.3161,
|
|
"mean_token_accuracy": 0.1649239718914032,
|
|
"num_tokens": 5039424.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.648041248321533,
|
|
"epoch": 2.249216300940439,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004724869453539832,
|
|
"loss": 5.5172,
|
|
"mean_token_accuracy": 0.1547361545264721,
|
|
"num_tokens": 5048868.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.618503427505493,
|
|
"epoch": 2.2531347962382444,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004723426377062972,
|
|
"loss": 5.351,
|
|
"mean_token_accuracy": 0.16053935885429382,
|
|
"num_tokens": 5057532.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.597631216049194,
|
|
"epoch": 2.2570532915360504,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047219797737532,
|
|
"loss": 5.3526,
|
|
"mean_token_accuracy": 0.16098810732364655,
|
|
"num_tokens": 5066088.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.480603504180908,
|
|
"epoch": 2.260971786833856,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047205296461958314,
|
|
"loss": 5.2604,
|
|
"mean_token_accuracy": 0.16616066098213195,
|
|
"num_tokens": 5074968.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.47146577835083,
|
|
"epoch": 2.2648902821316614,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047190759969824785,
|
|
"loss": 5.2782,
|
|
"mean_token_accuracy": 0.16563865467905997,
|
|
"num_tokens": 5083605.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.539209842681885,
|
|
"epoch": 2.268808777429467,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047176188287110485,
|
|
"loss": 5.2884,
|
|
"mean_token_accuracy": 0.16735636740922927,
|
|
"num_tokens": 5092182.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.703541231155396,
|
|
"epoch": 2.2727272727272725,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004716158143985737,
|
|
"loss": 5.4828,
|
|
"mean_token_accuracy": 0.1571749359369278,
|
|
"num_tokens": 5100910.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.598054981231689,
|
|
"epoch": 2.2766457680250785,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047146939454170245,
|
|
"loss": 5.3554,
|
|
"mean_token_accuracy": 0.16782473027706146,
|
|
"num_tokens": 5109760.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.595303249359131,
|
|
"epoch": 2.280564263322884,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004713226235621672,
|
|
"loss": 5.3653,
|
|
"mean_token_accuracy": 0.16245563328266144,
|
|
"num_tokens": 5119193.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.53184404373169,
|
|
"epoch": 2.2844827586206895,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004711755017222714,
|
|
"loss": 5.2761,
|
|
"mean_token_accuracy": 0.16629687696695328,
|
|
"num_tokens": 5127480.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.648559045791626,
|
|
"epoch": 2.2884012539184955,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047102802928494563,
|
|
"loss": 5.4764,
|
|
"mean_token_accuracy": 0.15637332051992417,
|
|
"num_tokens": 5136648.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.578768920898438,
|
|
"epoch": 2.292319749216301,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004708802065137471,
|
|
"loss": 5.2737,
|
|
"mean_token_accuracy": 0.16840785294771193,
|
|
"num_tokens": 5144708.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.509908008575439,
|
|
"epoch": 2.2962382445141065,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004707320336728591,
|
|
"loss": 5.3874,
|
|
"mean_token_accuracy": 0.15815174877643584,
|
|
"num_tokens": 5153388.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.623534393310547,
|
|
"epoch": 2.300156739811912,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004705835110270904,
|
|
"loss": 5.3533,
|
|
"mean_token_accuracy": 0.16146910190582275,
|
|
"num_tokens": 5161874.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.560824823379517,
|
|
"epoch": 2.304075235109718,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047043463884187517,
|
|
"loss": 5.3557,
|
|
"mean_token_accuracy": 0.1618855342268944,
|
|
"num_tokens": 5170284.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.497574949264527,
|
|
"epoch": 2.3079937304075235,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047028541738327207,
|
|
"loss": 5.326,
|
|
"mean_token_accuracy": 0.1598228007555008,
|
|
"num_tokens": 5179547.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.571848964691162,
|
|
"epoch": 2.311912225705329,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004701358469179641,
|
|
"loss": 5.3287,
|
|
"mean_token_accuracy": 0.1698288217186928,
|
|
"num_tokens": 5188256.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.594274997711182,
|
|
"epoch": 2.3158307210031346,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004699859277132578,
|
|
"loss": 5.4826,
|
|
"mean_token_accuracy": 0.16385053992271423,
|
|
"num_tokens": 5197464.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.6157807350158695,
|
|
"epoch": 2.3197492163009406,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046983566003708336,
|
|
"loss": 5.2634,
|
|
"mean_token_accuracy": 0.17211963832378388,
|
|
"num_tokens": 5205989.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.490232324600219,
|
|
"epoch": 2.323667711598746,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046968504415799325,
|
|
"loss": 5.3259,
|
|
"mean_token_accuracy": 0.16186022460460664,
|
|
"num_tokens": 5214490.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.529596757888794,
|
|
"epoch": 2.3275862068965516,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004695340803451625,
|
|
"loss": 5.3584,
|
|
"mean_token_accuracy": 0.16376458406448363,
|
|
"num_tokens": 5223335.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.487400341033935,
|
|
"epoch": 2.3315047021943576,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 5.2898,
|
|
"mean_token_accuracy": 0.16245129853487014,
|
|
"num_tokens": 5231665.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.5195292949676515,
|
|
"epoch": 2.335423197492163,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004692311099980878,
|
|
"loss": 5.2879,
|
|
"mean_token_accuracy": 0.16708355247974396,
|
|
"num_tokens": 5239810.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.51229076385498,
|
|
"epoch": 2.3393416927899686,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046907910400530097,
|
|
"loss": 5.2546,
|
|
"mean_token_accuracy": 0.17722575962543488,
|
|
"num_tokens": 5247525.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.4491297721862795,
|
|
"epoch": 2.343260188087774,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004689267511616868,
|
|
"loss": 5.3011,
|
|
"mean_token_accuracy": 0.16801622956991197,
|
|
"num_tokens": 5255528.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.519492959976196,
|
|
"epoch": 2.34717868338558,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046877405173952465,
|
|
"loss": 5.2627,
|
|
"mean_token_accuracy": 0.16878023892641067,
|
|
"num_tokens": 5263944.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.568019199371338,
|
|
"epoch": 2.3510971786833856,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000468621006011713,
|
|
"loss": 5.3662,
|
|
"mean_token_accuracy": 0.16010621339082717,
|
|
"num_tokens": 5271724.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.3510971786833856,
|
|
"eval_entropy": 5.434771962868151,
|
|
"eval_loss": 5.8656816482543945,
|
|
"eval_mean_token_accuracy": 0.15230537186441725,
|
|
"eval_num_tokens": 5271724.0,
|
|
"eval_runtime": 2.8543,
|
|
"eval_samples_per_second": 1444.131,
|
|
"eval_steps_per_second": 180.779,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.572009038925171,
|
|
"epoch": 2.355015673981191,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046846761425176943,
|
|
"loss": 5.3921,
|
|
"mean_token_accuracy": 0.16172949075698853,
|
|
"num_tokens": 5281199.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.65570912361145,
|
|
"epoch": 2.3589341692789967,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004683138767338299,
|
|
"loss": 5.4075,
|
|
"mean_token_accuracy": 0.16199006289243698,
|
|
"num_tokens": 5289782.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.402611207962036,
|
|
"epoch": 2.3628526645768027,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004681597937326483,
|
|
"loss": 5.2169,
|
|
"mean_token_accuracy": 0.17421238571405412,
|
|
"num_tokens": 5297922.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.52378478050232,
|
|
"epoch": 2.366771159874608,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004680053655235959,
|
|
"loss": 5.3576,
|
|
"mean_token_accuracy": 0.16891258656978608,
|
|
"num_tokens": 5306178.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.5647275924682615,
|
|
"epoch": 2.3706896551724137,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004678505923826609,
|
|
"loss": 5.3381,
|
|
"mean_token_accuracy": 0.16613359600305558,
|
|
"num_tokens": 5314936.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.547653722763061,
|
|
"epoch": 2.3746081504702197,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046769547458644817,
|
|
"loss": 5.3539,
|
|
"mean_token_accuracy": 0.16610560715198516,
|
|
"num_tokens": 5323549.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.552595472335815,
|
|
"epoch": 2.378526645768025,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004675400124121782,
|
|
"loss": 5.3373,
|
|
"mean_token_accuracy": 0.16588755548000336,
|
|
"num_tokens": 5332310.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.6628941059112545,
|
|
"epoch": 2.3824451410658307,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046738420613768716,
|
|
"loss": 5.4674,
|
|
"mean_token_accuracy": 0.1585412800312042,
|
|
"num_tokens": 5340757.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.654271364212036,
|
|
"epoch": 2.3863636363636362,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046722805604142614,
|
|
"loss": 5.4677,
|
|
"mean_token_accuracy": 0.1585947036743164,
|
|
"num_tokens": 5349433.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.529625320434571,
|
|
"epoch": 2.3902821316614418,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046707156240246076,
|
|
"loss": 5.4046,
|
|
"mean_token_accuracy": 0.15594652146100998,
|
|
"num_tokens": 5358943.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.524866342544556,
|
|
"epoch": 2.3942006269592477,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00046691472550047027,
|
|
"loss": 5.3264,
|
|
"mean_token_accuracy": 0.16086422502994538,
|
|
"num_tokens": 5367782.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.5129622459411625,
|
|
"epoch": 2.3981191222570533,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046675754561574783,
|
|
"loss": 5.1884,
|
|
"mean_token_accuracy": 0.16960814148187636,
|
|
"num_tokens": 5376186.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.5476171493530275,
|
|
"epoch": 2.402037617554859,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046660002302919933,
|
|
"loss": 5.4609,
|
|
"mean_token_accuracy": 0.15653598308563232,
|
|
"num_tokens": 5384888.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.588283061981201,
|
|
"epoch": 2.4059561128526648,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004664421580223433,
|
|
"loss": 5.3452,
|
|
"mean_token_accuracy": 0.16461212933063507,
|
|
"num_tokens": 5393394.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.627720546722412,
|
|
"epoch": 2.4098746081504703,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046628395087730995,
|
|
"loss": 5.405,
|
|
"mean_token_accuracy": 0.161533859372139,
|
|
"num_tokens": 5402901.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.5674063205719,
|
|
"epoch": 2.413793103448276,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004661254018768411,
|
|
"loss": 5.3989,
|
|
"mean_token_accuracy": 0.1623615577816963,
|
|
"num_tokens": 5412704.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.542704439163208,
|
|
"epoch": 2.4177115987460813,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004659665113042897,
|
|
"loss": 5.3651,
|
|
"mean_token_accuracy": 0.17225506007671357,
|
|
"num_tokens": 5420863.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.480785846710205,
|
|
"epoch": 2.4216300940438873,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004658072794436187,
|
|
"loss": 5.3052,
|
|
"mean_token_accuracy": 0.17093472927808762,
|
|
"num_tokens": 5430002.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.482878065109253,
|
|
"epoch": 2.425548589341693,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00046564770657940146,
|
|
"loss": 5.2993,
|
|
"mean_token_accuracy": 0.17024858742952348,
|
|
"num_tokens": 5438424.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.4856819152832035,
|
|
"epoch": 2.4294670846394983,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004654877929968205,
|
|
"loss": 5.3217,
|
|
"mean_token_accuracy": 0.1706370621919632,
|
|
"num_tokens": 5446970.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.5803690433502195,
|
|
"epoch": 2.433385579937304,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004653275389816673,
|
|
"loss": 5.3947,
|
|
"mean_token_accuracy": 0.15609584301710128,
|
|
"num_tokens": 5456689.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.523205041885376,
|
|
"epoch": 2.43730407523511,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046516694482034174,
|
|
"loss": 5.2773,
|
|
"mean_token_accuracy": 0.1692593887448311,
|
|
"num_tokens": 5465269.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.53827395439148,
|
|
"epoch": 2.4412225705329154,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046500601079985164,
|
|
"loss": 5.3572,
|
|
"mean_token_accuracy": 0.16684028506278992,
|
|
"num_tokens": 5473930.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.512274646759034,
|
|
"epoch": 2.445141065830721,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004648447372078123,
|
|
"loss": 5.3749,
|
|
"mean_token_accuracy": 0.17275755405426024,
|
|
"num_tokens": 5482885.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.504965353012085,
|
|
"epoch": 2.449059561128527,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004646831243324457,
|
|
"loss": 5.3177,
|
|
"mean_token_accuracy": 0.16810073852539062,
|
|
"num_tokens": 5491489.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.583372259140015,
|
|
"epoch": 2.4529780564263324,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004645211724625802,
|
|
"loss": 5.3758,
|
|
"mean_token_accuracy": 0.16375170946121215,
|
|
"num_tokens": 5499694.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.44045352935791,
|
|
"epoch": 2.456896551724138,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046435888188765015,
|
|
"loss": 5.2726,
|
|
"mean_token_accuracy": 0.1700295850634575,
|
|
"num_tokens": 5508037.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.52275652885437,
|
|
"epoch": 2.4608150470219434,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004641962528976951,
|
|
"loss": 5.2497,
|
|
"mean_token_accuracy": 0.17125702649354935,
|
|
"num_tokens": 5516644.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.4940837860107425,
|
|
"epoch": 2.4647335423197494,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004640332857833593,
|
|
"loss": 5.3131,
|
|
"mean_token_accuracy": 0.1654887914657593,
|
|
"num_tokens": 5524789.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.524436664581299,
|
|
"epoch": 2.468652037617555,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046386998083589156,
|
|
"loss": 5.35,
|
|
"mean_token_accuracy": 0.16829695850610732,
|
|
"num_tokens": 5533739.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.635737133026123,
|
|
"epoch": 2.4725705329153604,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004637063383471442,
|
|
"loss": 5.38,
|
|
"mean_token_accuracy": 0.16446826010942459,
|
|
"num_tokens": 5542735.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.472494125366211,
|
|
"epoch": 2.476489028213166,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046354235860957287,
|
|
"loss": 5.2174,
|
|
"mean_token_accuracy": 0.1728504106402397,
|
|
"num_tokens": 5551272.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.540900659561157,
|
|
"epoch": 2.480407523510972,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004633780419162361,
|
|
"loss": 5.4231,
|
|
"mean_token_accuracy": 0.15702152997255325,
|
|
"num_tokens": 5559868.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.465464353561401,
|
|
"epoch": 2.4843260188087775,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046321338856079435,
|
|
"loss": 5.2077,
|
|
"mean_token_accuracy": 0.1737042561173439,
|
|
"num_tokens": 5568276.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.465699577331543,
|
|
"epoch": 2.488244514106583,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046304839883750987,
|
|
"loss": 5.2948,
|
|
"mean_token_accuracy": 0.16228433400392533,
|
|
"num_tokens": 5576462.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.542299032211304,
|
|
"epoch": 2.492163009404389,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000462883073041246,
|
|
"loss": 5.4591,
|
|
"mean_token_accuracy": 0.1608154818415642,
|
|
"num_tokens": 5586044.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.41082797050476,
|
|
"epoch": 2.4960815047021945,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004627174114674669,
|
|
"loss": 5.2255,
|
|
"mean_token_accuracy": 0.17130375206470488,
|
|
"num_tokens": 5594626.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.577104806900024,
|
|
"epoch": 2.5,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004625514144122365,
|
|
"loss": 5.2935,
|
|
"mean_token_accuracy": 0.16783207952976226,
|
|
"num_tokens": 5603268.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.510541343688965,
|
|
"epoch": 2.5039184952978055,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004623850821722185,
|
|
"loss": 5.2773,
|
|
"mean_token_accuracy": 0.1686388924717903,
|
|
"num_tokens": 5612053.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.514151287078858,
|
|
"epoch": 2.507836990595611,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004622184150446756,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.15114220529794692,
|
|
"num_tokens": 5621297.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.616940927505493,
|
|
"epoch": 2.511755485893417,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046205141332746904,
|
|
"loss": 5.3606,
|
|
"mean_token_accuracy": 0.16217829287052155,
|
|
"num_tokens": 5629683.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.43563985824585,
|
|
"epoch": 2.5156739811912225,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 5.2808,
|
|
"mean_token_accuracy": 0.1688768208026886,
|
|
"num_tokens": 5638707.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.549165916442871,
|
|
"epoch": 2.519592476489028,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004617164073184987,
|
|
"loss": 5.3253,
|
|
"mean_token_accuracy": 0.1647924467921257,
|
|
"num_tokens": 5648241.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.498393154144287,
|
|
"epoch": 2.523510971786834,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046154840362544496,
|
|
"loss": 5.3688,
|
|
"mean_token_accuracy": 0.1656247302889824,
|
|
"num_tokens": 5656743.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.500120210647583,
|
|
"epoch": 2.5274294670846396,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004613800665401466,
|
|
"loss": 5.3275,
|
|
"mean_token_accuracy": 0.1651181012392044,
|
|
"num_tokens": 5666418.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.475159645080566,
|
|
"epoch": 2.531347962382445,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004612113963634493,
|
|
"loss": 5.3016,
|
|
"mean_token_accuracy": 0.16872531920671463,
|
|
"num_tokens": 5675404.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.504612731933594,
|
|
"epoch": 2.535266457680251,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004610423933967938,
|
|
"loss": 5.3853,
|
|
"mean_token_accuracy": 0.15886692255735396,
|
|
"num_tokens": 5684699.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.6432898998260494,
|
|
"epoch": 2.5391849529780566,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000460873057942216,
|
|
"loss": 5.3855,
|
|
"mean_token_accuracy": 0.1671795815229416,
|
|
"num_tokens": 5693219.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.5588963508605955,
|
|
"epoch": 2.543103448275862,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004607033903023458,
|
|
"loss": 5.3927,
|
|
"mean_token_accuracy": 0.1640296459197998,
|
|
"num_tokens": 5702005.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.606112480163574,
|
|
"epoch": 2.5470219435736676,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046053339078040674,
|
|
"loss": 5.3993,
|
|
"mean_token_accuracy": 0.16274697929620743,
|
|
"num_tokens": 5710715.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.494814682006836,
|
|
"epoch": 2.550940438871473,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004603630596802155,
|
|
"loss": 5.3251,
|
|
"mean_token_accuracy": 0.1691939041018486,
|
|
"num_tokens": 5719444.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.4016355037689205,
|
|
"epoch": 2.554858934169279,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004601923973061814,
|
|
"loss": 5.3096,
|
|
"mean_token_accuracy": 0.16056416630744935,
|
|
"num_tokens": 5727755.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.500792837142944,
|
|
"epoch": 2.5587774294670846,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046002140396330575,
|
|
"loss": 5.2502,
|
|
"mean_token_accuracy": 0.16782210320234298,
|
|
"num_tokens": 5736933.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.553884267807007,
|
|
"epoch": 2.56269592476489,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045985007995718154,
|
|
"loss": 5.4092,
|
|
"mean_token_accuracy": 0.16856626868247987,
|
|
"num_tokens": 5745831.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.579045343399048,
|
|
"epoch": 2.566614420062696,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004596784255939923,
|
|
"loss": 5.4442,
|
|
"mean_token_accuracy": 0.1629267528653145,
|
|
"num_tokens": 5755419.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.525725603103638,
|
|
"epoch": 2.5705329153605017,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004595064411805123,
|
|
"loss": 5.3328,
|
|
"mean_token_accuracy": 0.17061960101127624,
|
|
"num_tokens": 5764008.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.477341508865356,
|
|
"epoch": 2.574451410658307,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004593341270241057,
|
|
"loss": 5.2416,
|
|
"mean_token_accuracy": 0.16688467860221862,
|
|
"num_tokens": 5772129.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.519706869125367,
|
|
"epoch": 2.5783699059561127,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004591614834327257,
|
|
"loss": 5.3595,
|
|
"mean_token_accuracy": 0.16659962236881257,
|
|
"num_tokens": 5781079.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.471915292739868,
|
|
"epoch": 2.5822884012539182,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045898851071491444,
|
|
"loss": 5.3291,
|
|
"mean_token_accuracy": 0.16874849945306777,
|
|
"num_tokens": 5790444.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.4016204357147215,
|
|
"epoch": 2.586206896551724,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004588152091798022,
|
|
"loss": 5.2211,
|
|
"mean_token_accuracy": 0.1672128587961197,
|
|
"num_tokens": 5799828.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.603023099899292,
|
|
"epoch": 2.5901253918495297,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004586415791371069,
|
|
"loss": 5.4821,
|
|
"mean_token_accuracy": 0.15468620955944062,
|
|
"num_tokens": 5808765.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.571247291564942,
|
|
"epoch": 2.5940438871473352,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004584676208971336,
|
|
"loss": 5.2691,
|
|
"mean_token_accuracy": 0.1765100583434105,
|
|
"num_tokens": 5817281.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.448766040802002,
|
|
"epoch": 2.597962382445141,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045829333477077384,
|
|
"loss": 5.3016,
|
|
"mean_token_accuracy": 0.18109697848558426,
|
|
"num_tokens": 5825741.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.539952754974365,
|
|
"epoch": 2.6018808777429467,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004581187210695053,
|
|
"loss": 5.2821,
|
|
"mean_token_accuracy": 0.17550428509712218,
|
|
"num_tokens": 5834466.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.431042957305908,
|
|
"epoch": 2.6057993730407523,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000457943780105391,
|
|
"loss": 5.3121,
|
|
"mean_token_accuracy": 0.17115687280893327,
|
|
"num_tokens": 5843463.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.540793180465698,
|
|
"epoch": 2.6097178683385582,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045776851219107856,
|
|
"loss": 5.3427,
|
|
"mean_token_accuracy": 0.167041015625,
|
|
"num_tokens": 5851885.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.56136908531189,
|
|
"epoch": 2.6136363636363638,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045759291763980035,
|
|
"loss": 5.3516,
|
|
"mean_token_accuracy": 0.17018300741910936,
|
|
"num_tokens": 5861537.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.4394289493560795,
|
|
"epoch": 2.6175548589341693,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045741699676537227,
|
|
"loss": 5.2759,
|
|
"mean_token_accuracy": 0.1729188710451126,
|
|
"num_tokens": 5870332.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.512276315689087,
|
|
"epoch": 2.621473354231975,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045724074988219343,
|
|
"loss": 5.4836,
|
|
"mean_token_accuracy": 0.16204925179481505,
|
|
"num_tokens": 5879240.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.439336442947388,
|
|
"epoch": 2.6253918495297803,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045706417730524565,
|
|
"loss": 5.239,
|
|
"mean_token_accuracy": 0.1624177500605583,
|
|
"num_tokens": 5887961.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.539043331146241,
|
|
"epoch": 2.6293103448275863,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004568872793500927,
|
|
"loss": 5.4618,
|
|
"mean_token_accuracy": 0.16323541700839997,
|
|
"num_tokens": 5896551.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.431582450866699,
|
|
"epoch": 2.633228840125392,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045671005633287986,
|
|
"loss": 5.2424,
|
|
"mean_token_accuracy": 0.17058624625205993,
|
|
"num_tokens": 5905485.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.482703018188476,
|
|
"epoch": 2.6371473354231973,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004565325085703336,
|
|
"loss": 5.2917,
|
|
"mean_token_accuracy": 0.16687317788600922,
|
|
"num_tokens": 5915166.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.519888830184937,
|
|
"epoch": 2.6410658307210033,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004563546363797602,
|
|
"loss": 5.3259,
|
|
"mean_token_accuracy": 0.17161272913217546,
|
|
"num_tokens": 5923782.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.537396860122681,
|
|
"epoch": 2.644984326018809,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004561764400790465,
|
|
"loss": 5.4332,
|
|
"mean_token_accuracy": 0.16172593981027603,
|
|
"num_tokens": 5931904.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.522799825668335,
|
|
"epoch": 2.6489028213166144,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045599791998665796,
|
|
"loss": 5.2853,
|
|
"mean_token_accuracy": 0.17676885724067687,
|
|
"num_tokens": 5940067.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.4546473026275635,
|
|
"epoch": 2.6528213166144203,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004558190764216389,
|
|
"loss": 5.3056,
|
|
"mean_token_accuracy": 0.1724096104502678,
|
|
"num_tokens": 5948663.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.579775714874268,
|
|
"epoch": 2.656739811912226,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004556399097036119,
|
|
"loss": 5.3241,
|
|
"mean_token_accuracy": 0.17269116044044494,
|
|
"num_tokens": 5957224.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.540528202056885,
|
|
"epoch": 2.6606583072100314,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004554604201527768,
|
|
"loss": 5.4198,
|
|
"mean_token_accuracy": 0.16985767632722854,
|
|
"num_tokens": 5965604.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.461326837539673,
|
|
"epoch": 2.664576802507837,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045528060808991075,
|
|
"loss": 5.2393,
|
|
"mean_token_accuracy": 0.1762665867805481,
|
|
"num_tokens": 5974309.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.470024061203003,
|
|
"epoch": 2.6684952978056424,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004551004738363669,
|
|
"loss": 5.251,
|
|
"mean_token_accuracy": 0.1705390602350235,
|
|
"num_tokens": 5982315.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.452718305587768,
|
|
"epoch": 2.6724137931034484,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045492001771407434,
|
|
"loss": 5.2918,
|
|
"mean_token_accuracy": 0.17495300024747848,
|
|
"num_tokens": 5990515.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.520227289199829,
|
|
"epoch": 2.676332288401254,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004547392400455374,
|
|
"loss": 5.4512,
|
|
"mean_token_accuracy": 0.1606644168496132,
|
|
"num_tokens": 6000431.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.5755597114562985,
|
|
"epoch": 2.6802507836990594,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004545581411538353,
|
|
"loss": 5.3709,
|
|
"mean_token_accuracy": 0.16724153012037277,
|
|
"num_tokens": 6009022.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.514227437973022,
|
|
"epoch": 2.6841692789968654,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045437672136262083,
|
|
"loss": 5.3027,
|
|
"mean_token_accuracy": 0.16582043319940568,
|
|
"num_tokens": 6019330.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.459601211547851,
|
|
"epoch": 2.688087774294671,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004541949809961208,
|
|
"loss": 5.3159,
|
|
"mean_token_accuracy": 0.17988408952951432,
|
|
"num_tokens": 6027492.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.524388551712036,
|
|
"epoch": 2.6920062695924765,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004540129203791346,
|
|
"loss": 5.3378,
|
|
"mean_token_accuracy": 0.1686471864581108,
|
|
"num_tokens": 6036467.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.466223049163818,
|
|
"epoch": 2.695924764890282,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045383053983703413,
|
|
"loss": 5.2957,
|
|
"mean_token_accuracy": 0.1695375457406044,
|
|
"num_tokens": 6045711.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.500885486602783,
|
|
"epoch": 2.6998432601880875,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 5.3736,
|
|
"mean_token_accuracy": 0.1609950840473175,
|
|
"num_tokens": 6054626.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.559614562988282,
|
|
"epoch": 2.7037617554858935,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045346482028183583,
|
|
"loss": 5.4288,
|
|
"mean_token_accuracy": 0.16276882886886596,
|
|
"num_tokens": 6064118.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.523534679412842,
|
|
"epoch": 2.707680250783699,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045328148192233823,
|
|
"loss": 5.3338,
|
|
"mean_token_accuracy": 0.16942061334848404,
|
|
"num_tokens": 6072723.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.520214319229126,
|
|
"epoch": 2.7115987460815045,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004530978249449254,
|
|
"loss": 5.3223,
|
|
"mean_token_accuracy": 0.1724524199962616,
|
|
"num_tokens": 6081535.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.474754762649536,
|
|
"epoch": 2.7155172413793105,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004529138496778222,
|
|
"loss": 5.2672,
|
|
"mean_token_accuracy": 0.16802340596914292,
|
|
"num_tokens": 6090524.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.498336887359619,
|
|
"epoch": 2.719435736677116,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004527295564498222,
|
|
"loss": 5.3441,
|
|
"mean_token_accuracy": 0.166372212767601,
|
|
"num_tokens": 6099870.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.500373983383179,
|
|
"epoch": 2.7233542319749215,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004525449455902874,
|
|
"loss": 5.3115,
|
|
"mean_token_accuracy": 0.16563377976417543,
|
|
"num_tokens": 6107760.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.528516864776611,
|
|
"epoch": 2.7272727272727275,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004523600174291473,
|
|
"loss": 5.3948,
|
|
"mean_token_accuracy": 0.16702790558338165,
|
|
"num_tokens": 6117135.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.451120853424072,
|
|
"epoch": 2.731191222570533,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004521747722968985,
|
|
"loss": 5.3347,
|
|
"mean_token_accuracy": 0.16837533414363862,
|
|
"num_tokens": 6125711.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.553459644317627,
|
|
"epoch": 2.7351097178683386,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045198921052460396,
|
|
"loss": 5.3881,
|
|
"mean_token_accuracy": 0.16275455951690673,
|
|
"num_tokens": 6134623.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.583516311645508,
|
|
"epoch": 2.739028213166144,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004518033324438928,
|
|
"loss": 5.3859,
|
|
"mean_token_accuracy": 0.16478580981492996,
|
|
"num_tokens": 6143228.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.478849411010742,
|
|
"epoch": 2.7429467084639496,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004516171383869593,
|
|
"loss": 5.3613,
|
|
"mean_token_accuracy": 0.16525180339813234,
|
|
"num_tokens": 6152021.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.7429467084639496,
|
|
"eval_entropy": 5.374510490617086,
|
|
"eval_loss": 5.8104634284973145,
|
|
"eval_mean_token_accuracy": 0.15645403335549574,
|
|
"eval_num_tokens": 6152021.0,
|
|
"eval_runtime": 2.8448,
|
|
"eval_samples_per_second": 1448.984,
|
|
"eval_steps_per_second": 181.387,
|
|
"step": 3500
|
|
}
|
|
],
|
|
"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": 8307695952000000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|