bc58e4528f
Model: fpadovani/ind-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.1755485893416928,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 3568490797824000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|