bc58e4528f
Model: fpadovani/ind-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.39184952978056425,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 1156331556864000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|