a45fd99152
Model: fpadovani/ind-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
6101 lines
165 KiB
JSON
6101 lines
165 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.3510971786833856,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.505434942245484,
|
|
"epoch": 0.003918495297805642,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.9614,
|
|
"mean_token_accuracy": 0.0993234820663929,
|
|
"num_tokens": 9174.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.468199586868286,
|
|
"epoch": 0.007836990595611285,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.9456,
|
|
"mean_token_accuracy": 0.097779630869627,
|
|
"num_tokens": 17790.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.504303407669068,
|
|
"epoch": 0.011755485893416929,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.0633,
|
|
"mean_token_accuracy": 0.09743289425969123,
|
|
"num_tokens": 25850.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.537549209594727,
|
|
"epoch": 0.01567398119122257,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.134,
|
|
"mean_token_accuracy": 0.08861189931631089,
|
|
"num_tokens": 35194.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.50921597480774,
|
|
"epoch": 0.019592476489028215,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.093,
|
|
"mean_token_accuracy": 0.09007101207971573,
|
|
"num_tokens": 44218.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.450595808029175,
|
|
"epoch": 0.023510971786833857,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.0503,
|
|
"mean_token_accuracy": 0.08850282579660415,
|
|
"num_tokens": 53397.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.409115219116211,
|
|
"epoch": 0.0274294670846395,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.9462,
|
|
"mean_token_accuracy": 0.09454455524682999,
|
|
"num_tokens": 62749.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.396940851211548,
|
|
"epoch": 0.03134796238244514,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.9541,
|
|
"mean_token_accuracy": 0.09411264210939407,
|
|
"num_tokens": 71721.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.383365297317505,
|
|
"epoch": 0.03526645768025078,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.9327,
|
|
"mean_token_accuracy": 0.09620709493756294,
|
|
"num_tokens": 79844.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.391091680526733,
|
|
"epoch": 0.03918495297805643,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.9128,
|
|
"mean_token_accuracy": 0.10004793405532837,
|
|
"num_tokens": 88866.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.364109182357788,
|
|
"epoch": 0.04310344827586207,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.9315,
|
|
"mean_token_accuracy": 0.09721487462520599,
|
|
"num_tokens": 97918.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.348652935028076,
|
|
"epoch": 0.047021943573667714,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.9589,
|
|
"mean_token_accuracy": 0.09757346361875534,
|
|
"num_tokens": 107043.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.408840799331665,
|
|
"epoch": 0.050940438871473356,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.0024,
|
|
"mean_token_accuracy": 0.0984207846224308,
|
|
"num_tokens": 116000.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.477249097824097,
|
|
"epoch": 0.054858934169279,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.0211,
|
|
"mean_token_accuracy": 0.09341394379734994,
|
|
"num_tokens": 124559.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.440812301635742,
|
|
"epoch": 0.05877742946708464,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.9256,
|
|
"mean_token_accuracy": 0.09969957172870636,
|
|
"num_tokens": 132868.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.4664476871490475,
|
|
"epoch": 0.06269592476489028,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.014,
|
|
"mean_token_accuracy": 0.08899271860718727,
|
|
"num_tokens": 141286.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.387943983078003,
|
|
"epoch": 0.06661442006269593,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.9708,
|
|
"mean_token_accuracy": 0.10166924223303794,
|
|
"num_tokens": 150406.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.404929876327515,
|
|
"epoch": 0.07053291536050156,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.0189,
|
|
"mean_token_accuracy": 0.09384585246443748,
|
|
"num_tokens": 158770.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.3884033203125,
|
|
"epoch": 0.07445141065830721,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.9298,
|
|
"mean_token_accuracy": 0.09883309677243232,
|
|
"num_tokens": 167763.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.446268320083618,
|
|
"epoch": 0.07836990595611286,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.9262,
|
|
"mean_token_accuracy": 0.09872755706310272,
|
|
"num_tokens": 177306.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.385495233535766,
|
|
"epoch": 0.0822884012539185,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.9265,
|
|
"mean_token_accuracy": 0.09789144024252891,
|
|
"num_tokens": 186014.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.39170093536377,
|
|
"epoch": 0.08620689655172414,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.8456,
|
|
"mean_token_accuracy": 0.10207158401608467,
|
|
"num_tokens": 194122.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.412753772735596,
|
|
"epoch": 0.09012539184952978,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.8912,
|
|
"mean_token_accuracy": 0.10196277350187302,
|
|
"num_tokens": 203097.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.368919897079468,
|
|
"epoch": 0.09404388714733543,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.8734,
|
|
"mean_token_accuracy": 0.10027118623256684,
|
|
"num_tokens": 211413.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.324716091156006,
|
|
"epoch": 0.09796238244514106,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.9415,
|
|
"mean_token_accuracy": 0.09764488190412521,
|
|
"num_tokens": 220550.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.366209936141968,
|
|
"epoch": 0.10188087774294671,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.8747,
|
|
"mean_token_accuracy": 0.1016789697110653,
|
|
"num_tokens": 229197.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.4047736644744875,
|
|
"epoch": 0.10579937304075235,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.0097,
|
|
"mean_token_accuracy": 0.09511950314044952,
|
|
"num_tokens": 238479.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.440355205535889,
|
|
"epoch": 0.109717868338558,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.9112,
|
|
"mean_token_accuracy": 0.10273143649101257,
|
|
"num_tokens": 246318.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.443423795700073,
|
|
"epoch": 0.11363636363636363,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.9876,
|
|
"mean_token_accuracy": 0.10033264607191086,
|
|
"num_tokens": 254783.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.378429794311524,
|
|
"epoch": 0.11755485893416928,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.867,
|
|
"mean_token_accuracy": 0.10020775869488716,
|
|
"num_tokens": 263416.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.401640510559082,
|
|
"epoch": 0.12147335423197492,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.9491,
|
|
"mean_token_accuracy": 0.10223579108715057,
|
|
"num_tokens": 271930.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.516922426223755,
|
|
"epoch": 0.12539184952978055,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.9201,
|
|
"mean_token_accuracy": 0.10161197930574417,
|
|
"num_tokens": 280546.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.277652978897095,
|
|
"epoch": 0.12931034482758622,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.9378,
|
|
"mean_token_accuracy": 0.10357455834746361,
|
|
"num_tokens": 288813.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.434544038772583,
|
|
"epoch": 0.13322884012539185,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.9038,
|
|
"mean_token_accuracy": 0.1046725369989872,
|
|
"num_tokens": 297966.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.45661678314209,
|
|
"epoch": 0.1371473354231975,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.0377,
|
|
"mean_token_accuracy": 0.09913579076528549,
|
|
"num_tokens": 307135.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.3250244617462155,
|
|
"epoch": 0.14106583072100312,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.9574,
|
|
"mean_token_accuracy": 0.0996546871960163,
|
|
"num_tokens": 315546.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.480092477798462,
|
|
"epoch": 0.14498432601880878,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.9337,
|
|
"mean_token_accuracy": 0.1008750095963478,
|
|
"num_tokens": 323930.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.33076810836792,
|
|
"epoch": 0.14890282131661442,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.8479,
|
|
"mean_token_accuracy": 0.10825147032737732,
|
|
"num_tokens": 332419.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.28166446685791,
|
|
"epoch": 0.15282131661442006,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.9308,
|
|
"mean_token_accuracy": 0.10017007514834404,
|
|
"num_tokens": 341362.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.500374031066895,
|
|
"epoch": 0.15673981191222572,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.8616,
|
|
"mean_token_accuracy": 0.10328178033232689,
|
|
"num_tokens": 349395.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.2322710990905765,
|
|
"epoch": 0.16065830721003135,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.9086,
|
|
"mean_token_accuracy": 0.09767335131764412,
|
|
"num_tokens": 358413.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.49402232170105,
|
|
"epoch": 0.164576802507837,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.9846,
|
|
"mean_token_accuracy": 0.10406019762158394,
|
|
"num_tokens": 366489.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.382347631454468,
|
|
"epoch": 0.16849529780564262,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.8303,
|
|
"mean_token_accuracy": 0.09956172332167626,
|
|
"num_tokens": 375335.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.36075382232666,
|
|
"epoch": 0.1724137931034483,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.8922,
|
|
"mean_token_accuracy": 0.10197906568646431,
|
|
"num_tokens": 384276.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.486998271942139,
|
|
"epoch": 0.17633228840125392,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.9869,
|
|
"mean_token_accuracy": 0.09691886231303215,
|
|
"num_tokens": 393571.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.307347154617309,
|
|
"epoch": 0.18025078369905956,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.9233,
|
|
"mean_token_accuracy": 0.0977975718677044,
|
|
"num_tokens": 401865.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.42743992805481,
|
|
"epoch": 0.1841692789968652,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.8798,
|
|
"mean_token_accuracy": 0.10318133607506752,
|
|
"num_tokens": 410518.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.2817432403564455,
|
|
"epoch": 0.18808777429467086,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.8894,
|
|
"mean_token_accuracy": 0.0992834985256195,
|
|
"num_tokens": 419769.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.381455421447754,
|
|
"epoch": 0.1920062695924765,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.8163,
|
|
"mean_token_accuracy": 0.10831291303038597,
|
|
"num_tokens": 428204.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.425300884246826,
|
|
"epoch": 0.19592476489028213,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.9337,
|
|
"mean_token_accuracy": 0.09694090113043785,
|
|
"num_tokens": 436594.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.256980228424072,
|
|
"epoch": 0.19984326018808776,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.7228,
|
|
"mean_token_accuracy": 0.10796715542674065,
|
|
"num_tokens": 444645.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.27267370223999,
|
|
"epoch": 0.20376175548589343,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.7866,
|
|
"mean_token_accuracy": 0.10788663849234581,
|
|
"num_tokens": 453016.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.341275072097778,
|
|
"epoch": 0.20768025078369906,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.8743,
|
|
"mean_token_accuracy": 0.10142935216426849,
|
|
"num_tokens": 462116.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.402976942062378,
|
|
"epoch": 0.2115987460815047,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.8313,
|
|
"mean_token_accuracy": 0.10508178547024727,
|
|
"num_tokens": 470807.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.328185224533081,
|
|
"epoch": 0.21551724137931033,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.9207,
|
|
"mean_token_accuracy": 0.10136421322822571,
|
|
"num_tokens": 479592.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.416067266464234,
|
|
"epoch": 0.219435736677116,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.8648,
|
|
"mean_token_accuracy": 0.09813391491770744,
|
|
"num_tokens": 488107.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.215375375747681,
|
|
"epoch": 0.22335423197492163,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.8788,
|
|
"mean_token_accuracy": 0.10034509003162384,
|
|
"num_tokens": 496775.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.533013486862183,
|
|
"epoch": 0.22727272727272727,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.9054,
|
|
"mean_token_accuracy": 0.10370760932564735,
|
|
"num_tokens": 505415.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.286190128326416,
|
|
"epoch": 0.23119122257053293,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.8927,
|
|
"mean_token_accuracy": 0.10432923436164857,
|
|
"num_tokens": 514447.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.416802453994751,
|
|
"epoch": 0.23510971786833856,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9014,
|
|
"mean_token_accuracy": 0.10285199135541916,
|
|
"num_tokens": 522998.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.258221530914307,
|
|
"epoch": 0.2390282131661442,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.8665,
|
|
"mean_token_accuracy": 0.1017219752073288,
|
|
"num_tokens": 531542.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.3578962802886965,
|
|
"epoch": 0.24294670846394983,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.8023,
|
|
"mean_token_accuracy": 0.10435706600546837,
|
|
"num_tokens": 540143.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.363638305664063,
|
|
"epoch": 0.2468652037617555,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9134,
|
|
"mean_token_accuracy": 0.09914149641990662,
|
|
"num_tokens": 549156.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.245448780059815,
|
|
"epoch": 0.2507836990595611,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.7498,
|
|
"mean_token_accuracy": 0.10137966275215149,
|
|
"num_tokens": 557522.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.389043092727661,
|
|
"epoch": 0.2547021943573668,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.8285,
|
|
"mean_token_accuracy": 0.1055278718471527,
|
|
"num_tokens": 566650.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.2701962947845455,
|
|
"epoch": 0.25862068965517243,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.7756,
|
|
"mean_token_accuracy": 0.10573992133140564,
|
|
"num_tokens": 576219.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.332693099975586,
|
|
"epoch": 0.26253918495297807,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9123,
|
|
"mean_token_accuracy": 0.10183104053139687,
|
|
"num_tokens": 584304.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.319992256164551,
|
|
"epoch": 0.2664576802507837,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.8301,
|
|
"mean_token_accuracy": 0.10873695537447929,
|
|
"num_tokens": 593163.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.252126789093017,
|
|
"epoch": 0.27037617554858934,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.7463,
|
|
"mean_token_accuracy": 0.10352636948227882,
|
|
"num_tokens": 602077.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.348601579666138,
|
|
"epoch": 0.274294670846395,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.8758,
|
|
"mean_token_accuracy": 0.10260494649410248,
|
|
"num_tokens": 610009.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.356398010253907,
|
|
"epoch": 0.2782131661442006,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9186,
|
|
"mean_token_accuracy": 0.10047245398163795,
|
|
"num_tokens": 619282.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.314131212234497,
|
|
"epoch": 0.28213166144200624,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8463,
|
|
"mean_token_accuracy": 0.10727460011839866,
|
|
"num_tokens": 628138.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.307407283782959,
|
|
"epoch": 0.28605015673981193,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.7026,
|
|
"mean_token_accuracy": 0.11174972355365753,
|
|
"num_tokens": 637021.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.263330602645874,
|
|
"epoch": 0.28996865203761757,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.8916,
|
|
"mean_token_accuracy": 0.09977171495556832,
|
|
"num_tokens": 646703.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.301595020294189,
|
|
"epoch": 0.2938871473354232,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7644,
|
|
"mean_token_accuracy": 0.10380433276295661,
|
|
"num_tokens": 655616.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.362639045715332,
|
|
"epoch": 0.29780564263322884,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.7801,
|
|
"mean_token_accuracy": 0.11212070435285568,
|
|
"num_tokens": 663783.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.217105197906494,
|
|
"epoch": 0.3017241379310345,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.7884,
|
|
"mean_token_accuracy": 0.11122082397341729,
|
|
"num_tokens": 671855.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.281291818618774,
|
|
"epoch": 0.3056426332288401,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8025,
|
|
"mean_token_accuracy": 0.10285323038697243,
|
|
"num_tokens": 680981.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.2563659191131595,
|
|
"epoch": 0.30956112852664575,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7293,
|
|
"mean_token_accuracy": 0.10659258142113685,
|
|
"num_tokens": 689294.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.190889358520508,
|
|
"epoch": 0.31347962382445144,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.6772,
|
|
"mean_token_accuracy": 0.10824779719114304,
|
|
"num_tokens": 697703.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.26824984550476,
|
|
"epoch": 0.31739811912225707,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8338,
|
|
"mean_token_accuracy": 0.10549585670232772,
|
|
"num_tokens": 706792.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.335061359405517,
|
|
"epoch": 0.3213166144200627,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7495,
|
|
"mean_token_accuracy": 0.10903687998652459,
|
|
"num_tokens": 715864.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.198094320297241,
|
|
"epoch": 0.32523510971786834,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8039,
|
|
"mean_token_accuracy": 0.10607616752386093,
|
|
"num_tokens": 723921.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.3263438701629635,
|
|
"epoch": 0.329153605015674,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.8322,
|
|
"mean_token_accuracy": 0.10520576909184456,
|
|
"num_tokens": 732797.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.203879880905151,
|
|
"epoch": 0.3330721003134796,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6859,
|
|
"mean_token_accuracy": 0.1141057014465332,
|
|
"num_tokens": 741248.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.2253659725189205,
|
|
"epoch": 0.33699059561128525,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7799,
|
|
"mean_token_accuracy": 0.10589491054415703,
|
|
"num_tokens": 749766.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.235883378982544,
|
|
"epoch": 0.3409090909090909,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7563,
|
|
"mean_token_accuracy": 0.1033819131553173,
|
|
"num_tokens": 758695.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.31015157699585,
|
|
"epoch": 0.3448275862068966,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7425,
|
|
"mean_token_accuracy": 0.10435104370117188,
|
|
"num_tokens": 767624.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.151997137069702,
|
|
"epoch": 0.3487460815047022,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7827,
|
|
"mean_token_accuracy": 0.10188787803053856,
|
|
"num_tokens": 776616.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.208565616607666,
|
|
"epoch": 0.35266457680250785,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7977,
|
|
"mean_token_accuracy": 0.10031938627362251,
|
|
"num_tokens": 786147.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.3614260196685795,
|
|
"epoch": 0.3565830721003135,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.8997,
|
|
"mean_token_accuracy": 0.10378970205783844,
|
|
"num_tokens": 795213.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.205408906936645,
|
|
"epoch": 0.3605015673981191,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.712,
|
|
"mean_token_accuracy": 0.11043478101491928,
|
|
"num_tokens": 803118.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.224033689498901,
|
|
"epoch": 0.36442006269592475,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6772,
|
|
"mean_token_accuracy": 0.11600754931569099,
|
|
"num_tokens": 811358.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.1833305835723875,
|
|
"epoch": 0.3683385579937304,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6706,
|
|
"mean_token_accuracy": 0.1100239709019661,
|
|
"num_tokens": 819653.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.133268070220947,
|
|
"epoch": 0.3722570532915361,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7248,
|
|
"mean_token_accuracy": 0.10289673954248428,
|
|
"num_tokens": 828462.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.301584005355835,
|
|
"epoch": 0.3761755485893417,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8085,
|
|
"mean_token_accuracy": 0.10872810781002044,
|
|
"num_tokens": 836204.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.103740882873535,
|
|
"epoch": 0.38009404388714735,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6975,
|
|
"mean_token_accuracy": 0.10832417607307435,
|
|
"num_tokens": 845032.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.280343627929687,
|
|
"epoch": 0.384012539184953,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7194,
|
|
"mean_token_accuracy": 0.11063263043761254,
|
|
"num_tokens": 853324.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.1609453678131105,
|
|
"epoch": 0.3879310344827586,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6956,
|
|
"mean_token_accuracy": 0.11505614593625069,
|
|
"num_tokens": 861640.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.229707145690918,
|
|
"epoch": 0.39184952978056425,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8579,
|
|
"mean_token_accuracy": 0.10259925499558449,
|
|
"num_tokens": 870758.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.39184952978056425,
|
|
"eval_entropy": 7.288463092589563,
|
|
"eval_loss": 6.810089111328125,
|
|
"eval_mean_token_accuracy": 0.10813328818699648,
|
|
"eval_num_tokens": 870758.0,
|
|
"eval_runtime": 2.1909,
|
|
"eval_samples_per_second": 1881.459,
|
|
"eval_steps_per_second": 235.525,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.288912677764893,
|
|
"epoch": 0.3957680250783699,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.773,
|
|
"mean_token_accuracy": 0.10451259464025497,
|
|
"num_tokens": 879968.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.139670133590698,
|
|
"epoch": 0.3996865203761755,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7423,
|
|
"mean_token_accuracy": 0.10636588633060455,
|
|
"num_tokens": 888804.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.235563802719116,
|
|
"epoch": 0.4036050156739812,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7483,
|
|
"mean_token_accuracy": 0.10818563774228096,
|
|
"num_tokens": 898593.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.188141202926635,
|
|
"epoch": 0.40752351097178685,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.7665,
|
|
"mean_token_accuracy": 0.10758788138628006,
|
|
"num_tokens": 906873.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.191415071487427,
|
|
"epoch": 0.4114420062695925,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.7526,
|
|
"mean_token_accuracy": 0.11076318845152855,
|
|
"num_tokens": 915080.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.202563953399658,
|
|
"epoch": 0.4153605015673981,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7624,
|
|
"mean_token_accuracy": 0.1057440385222435,
|
|
"num_tokens": 923741.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.1273657321929935,
|
|
"epoch": 0.41927899686520376,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6411,
|
|
"mean_token_accuracy": 0.10771751850843429,
|
|
"num_tokens": 932455.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.1441364765167235,
|
|
"epoch": 0.4231974921630094,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6634,
|
|
"mean_token_accuracy": 0.11294707357883453,
|
|
"num_tokens": 941064.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.089851665496826,
|
|
"epoch": 0.427115987460815,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6966,
|
|
"mean_token_accuracy": 0.1062391996383667,
|
|
"num_tokens": 950261.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.2247453212738035,
|
|
"epoch": 0.43103448275862066,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7642,
|
|
"mean_token_accuracy": 0.10804740041494369,
|
|
"num_tokens": 959151.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.192678785324096,
|
|
"epoch": 0.43495297805642635,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.7892,
|
|
"mean_token_accuracy": 0.10580966025590896,
|
|
"num_tokens": 968441.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.09973554611206,
|
|
"epoch": 0.438871473354232,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6583,
|
|
"mean_token_accuracy": 0.10889620780944824,
|
|
"num_tokens": 977058.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.1161939144134525,
|
|
"epoch": 0.4427899686520376,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6701,
|
|
"mean_token_accuracy": 0.11010972037911415,
|
|
"num_tokens": 986531.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.034862470626831,
|
|
"epoch": 0.44670846394984326,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.5124,
|
|
"mean_token_accuracy": 0.11886471137404442,
|
|
"num_tokens": 995081.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.107326507568359,
|
|
"epoch": 0.4506269592476489,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5854,
|
|
"mean_token_accuracy": 0.11489994674921036,
|
|
"num_tokens": 1003459.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.113965320587158,
|
|
"epoch": 0.45454545454545453,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6523,
|
|
"mean_token_accuracy": 0.11269798651337623,
|
|
"num_tokens": 1012420.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.152290868759155,
|
|
"epoch": 0.45846394984326017,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.7218,
|
|
"mean_token_accuracy": 0.10892164632678032,
|
|
"num_tokens": 1021198.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.12383131980896,
|
|
"epoch": 0.46238244514106586,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.7097,
|
|
"mean_token_accuracy": 0.11013553664088249,
|
|
"num_tokens": 1030023.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.063526391983032,
|
|
"epoch": 0.4663009404388715,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5702,
|
|
"mean_token_accuracy": 0.11459948495030403,
|
|
"num_tokens": 1038824.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.233054256439209,
|
|
"epoch": 0.4702194357366771,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.8369,
|
|
"mean_token_accuracy": 0.11036593914031982,
|
|
"num_tokens": 1048207.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.266431856155395,
|
|
"epoch": 0.47413793103448276,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.8029,
|
|
"mean_token_accuracy": 0.10950963348150253,
|
|
"num_tokens": 1057354.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.071617269515992,
|
|
"epoch": 0.4780564263322884,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.7325,
|
|
"mean_token_accuracy": 0.10674287006258965,
|
|
"num_tokens": 1066825.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.110283041000367,
|
|
"epoch": 0.48197492163009403,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.6658,
|
|
"mean_token_accuracy": 0.1055957555770874,
|
|
"num_tokens": 1075752.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.0111777782440186,
|
|
"epoch": 0.48589341692789967,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5774,
|
|
"mean_token_accuracy": 0.11234628558158874,
|
|
"num_tokens": 1084826.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.196392774581909,
|
|
"epoch": 0.4898119122257053,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7246,
|
|
"mean_token_accuracy": 0.10648038983345032,
|
|
"num_tokens": 1093425.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.097660875320434,
|
|
"epoch": 0.493730407523511,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.725,
|
|
"mean_token_accuracy": 0.10310131907463074,
|
|
"num_tokens": 1102619.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.025649738311768,
|
|
"epoch": 0.49764890282131663,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6392,
|
|
"mean_token_accuracy": 0.1060211069881916,
|
|
"num_tokens": 1111337.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.025793552398682,
|
|
"epoch": 0.5015673981191222,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4667,
|
|
"mean_token_accuracy": 0.11802783831954003,
|
|
"num_tokens": 1119381.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.967171049118042,
|
|
"epoch": 0.5054858934169278,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5164,
|
|
"mean_token_accuracy": 0.11229672878980637,
|
|
"num_tokens": 1127825.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.02331395149231,
|
|
"epoch": 0.5094043887147336,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5575,
|
|
"mean_token_accuracy": 0.10985610783100128,
|
|
"num_tokens": 1136750.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.101692771911621,
|
|
"epoch": 0.5133228840125392,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6076,
|
|
"mean_token_accuracy": 0.11186198219656944,
|
|
"num_tokens": 1145739.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.043255233764649,
|
|
"epoch": 0.5172413793103449,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.6639,
|
|
"mean_token_accuracy": 0.11381604298949241,
|
|
"num_tokens": 1154415.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.950129413604737,
|
|
"epoch": 0.5211598746081505,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.546,
|
|
"mean_token_accuracy": 0.11080027967691422,
|
|
"num_tokens": 1162480.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.061398935317993,
|
|
"epoch": 0.5250783699059561,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6325,
|
|
"mean_token_accuracy": 0.10909527465701103,
|
|
"num_tokens": 1170774.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.015395069122315,
|
|
"epoch": 0.5289968652037618,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5605,
|
|
"mean_token_accuracy": 0.10989686697721482,
|
|
"num_tokens": 1179629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.973291349411011,
|
|
"epoch": 0.5329153605015674,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.7397,
|
|
"mean_token_accuracy": 0.10960395857691765,
|
|
"num_tokens": 1189111.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.133076858520508,
|
|
"epoch": 0.536833855799373,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.7223,
|
|
"mean_token_accuracy": 0.10501213595271111,
|
|
"num_tokens": 1198827.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.153779125213623,
|
|
"epoch": 0.5407523510971787,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6187,
|
|
"mean_token_accuracy": 0.11944815516471863,
|
|
"num_tokens": 1207089.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.999565696716308,
|
|
"epoch": 0.5446708463949843,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6021,
|
|
"mean_token_accuracy": 0.11203551292419434,
|
|
"num_tokens": 1216509.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.907991218566894,
|
|
"epoch": 0.54858934169279,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4514,
|
|
"mean_token_accuracy": 0.11861641183495522,
|
|
"num_tokens": 1225684.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.967619180679321,
|
|
"epoch": 0.5525078369905956,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4915,
|
|
"mean_token_accuracy": 0.12248942703008651,
|
|
"num_tokens": 1233675.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.910577344894409,
|
|
"epoch": 0.5564263322884012,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5034,
|
|
"mean_token_accuracy": 0.11596095636487007,
|
|
"num_tokens": 1242147.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.959777307510376,
|
|
"epoch": 0.5603448275862069,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.6087,
|
|
"mean_token_accuracy": 0.11405783668160438,
|
|
"num_tokens": 1251127.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.949620103836059,
|
|
"epoch": 0.5642633228840125,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5668,
|
|
"mean_token_accuracy": 0.11413713470101357,
|
|
"num_tokens": 1260589.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.151991510391236,
|
|
"epoch": 0.5681818181818182,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6336,
|
|
"mean_token_accuracy": 0.11564361974596978,
|
|
"num_tokens": 1269281.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.920984840393066,
|
|
"epoch": 0.5721003134796239,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.5,
|
|
"mean_token_accuracy": 0.1117772951722145,
|
|
"num_tokens": 1278033.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.016093635559082,
|
|
"epoch": 0.5760188087774295,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.577,
|
|
"mean_token_accuracy": 0.1150795042514801,
|
|
"num_tokens": 1285847.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.005557060241699,
|
|
"epoch": 0.5799373040752351,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5601,
|
|
"mean_token_accuracy": 0.11328517720103264,
|
|
"num_tokens": 1294770.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.148551988601684,
|
|
"epoch": 0.5838557993730408,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.752,
|
|
"mean_token_accuracy": 0.10918178334832192,
|
|
"num_tokens": 1304282.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.974863862991333,
|
|
"epoch": 0.5877742946708464,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5898,
|
|
"mean_token_accuracy": 0.11170949414372444,
|
|
"num_tokens": 1313218.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.977967596054077,
|
|
"epoch": 0.591692789968652,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5537,
|
|
"mean_token_accuracy": 0.11300176531076431,
|
|
"num_tokens": 1322093.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.865384912490844,
|
|
"epoch": 0.5956112852664577,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.596,
|
|
"mean_token_accuracy": 0.11526871547102928,
|
|
"num_tokens": 1330718.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.960338306427002,
|
|
"epoch": 0.5995297805642633,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4534,
|
|
"mean_token_accuracy": 0.11476444453001022,
|
|
"num_tokens": 1339672.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.9716674327850345,
|
|
"epoch": 0.603448275862069,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4865,
|
|
"mean_token_accuracy": 0.1179777093231678,
|
|
"num_tokens": 1348587.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.843394804000854,
|
|
"epoch": 0.6073667711598746,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4776,
|
|
"mean_token_accuracy": 0.12329629883170128,
|
|
"num_tokens": 1357597.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.021717500686646,
|
|
"epoch": 0.6112852664576802,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5825,
|
|
"mean_token_accuracy": 0.11433721408247947,
|
|
"num_tokens": 1366246.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.862069892883301,
|
|
"epoch": 0.6152037617554859,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4907,
|
|
"mean_token_accuracy": 0.11741773039102554,
|
|
"num_tokens": 1374664.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.841044759750366,
|
|
"epoch": 0.6191222570532915,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.4907,
|
|
"mean_token_accuracy": 0.11631432324647903,
|
|
"num_tokens": 1382765.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.860881376266479,
|
|
"epoch": 0.6230407523510971,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4679,
|
|
"mean_token_accuracy": 0.11437924429774285,
|
|
"num_tokens": 1391708.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.943562173843384,
|
|
"epoch": 0.6269592476489029,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.627,
|
|
"mean_token_accuracy": 0.10611269995570183,
|
|
"num_tokens": 1401373.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.932059621810913,
|
|
"epoch": 0.6308777429467085,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4156,
|
|
"mean_token_accuracy": 0.11710819527506829,
|
|
"num_tokens": 1409702.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.909766006469726,
|
|
"epoch": 0.6347962382445141,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.6343,
|
|
"mean_token_accuracy": 0.11107001528143882,
|
|
"num_tokens": 1419596.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.932200336456299,
|
|
"epoch": 0.6387147335423198,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5701,
|
|
"mean_token_accuracy": 0.10957572385668754,
|
|
"num_tokens": 1428437.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.927781438827514,
|
|
"epoch": 0.6426332288401254,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4785,
|
|
"mean_token_accuracy": 0.11755915731191635,
|
|
"num_tokens": 1436764.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.820908641815185,
|
|
"epoch": 0.646551724137931,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.5205,
|
|
"mean_token_accuracy": 0.11839180067181587,
|
|
"num_tokens": 1445755.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.027404832839966,
|
|
"epoch": 0.6504702194357367,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5715,
|
|
"mean_token_accuracy": 0.11394421234726906,
|
|
"num_tokens": 1454550.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.791588592529297,
|
|
"epoch": 0.6543887147335423,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.465,
|
|
"mean_token_accuracy": 0.12183786556124687,
|
|
"num_tokens": 1463039.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.921341753005981,
|
|
"epoch": 0.658307210031348,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5253,
|
|
"mean_token_accuracy": 0.11498021930456162,
|
|
"num_tokens": 1472029.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.919203472137451,
|
|
"epoch": 0.6622257053291536,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.5541,
|
|
"mean_token_accuracy": 0.1158058650791645,
|
|
"num_tokens": 1481588.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.810756063461303,
|
|
"epoch": 0.6661442006269592,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4606,
|
|
"mean_token_accuracy": 0.11572995036840439,
|
|
"num_tokens": 1489915.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.007041692733765,
|
|
"epoch": 0.6700626959247649,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5469,
|
|
"mean_token_accuracy": 0.11174313724040985,
|
|
"num_tokens": 1497920.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.783069324493408,
|
|
"epoch": 0.6739811912225705,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3784,
|
|
"mean_token_accuracy": 0.1187769241631031,
|
|
"num_tokens": 1506965.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.8709169864654545,
|
|
"epoch": 0.6778996865203761,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4786,
|
|
"mean_token_accuracy": 0.11910211220383644,
|
|
"num_tokens": 1516028.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.849464416503906,
|
|
"epoch": 0.6818181818181818,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.5073,
|
|
"mean_token_accuracy": 0.11529333740472794,
|
|
"num_tokens": 1524579.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.816308784484863,
|
|
"epoch": 0.6857366771159875,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4089,
|
|
"mean_token_accuracy": 0.11816614344716073,
|
|
"num_tokens": 1534285.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.81186728477478,
|
|
"epoch": 0.6896551724137931,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.429,
|
|
"mean_token_accuracy": 0.11657375916838646,
|
|
"num_tokens": 1543701.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.957868671417236,
|
|
"epoch": 0.6935736677115988,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.6989,
|
|
"mean_token_accuracy": 0.10212851464748382,
|
|
"num_tokens": 1552715.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.644842576980591,
|
|
"epoch": 0.6974921630094044,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.301,
|
|
"mean_token_accuracy": 0.11869011968374252,
|
|
"num_tokens": 1561523.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.930772066116333,
|
|
"epoch": 0.70141065830721,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.5282,
|
|
"mean_token_accuracy": 0.11861441135406495,
|
|
"num_tokens": 1570057.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.722787570953369,
|
|
"epoch": 0.7053291536050157,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3797,
|
|
"mean_token_accuracy": 0.11915974989533425,
|
|
"num_tokens": 1578795.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.765309619903564,
|
|
"epoch": 0.7092476489028213,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.463,
|
|
"mean_token_accuracy": 0.11596754267811775,
|
|
"num_tokens": 1588108.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.8899001121521,
|
|
"epoch": 0.713166144200627,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.4225,
|
|
"mean_token_accuracy": 0.12392664477229118,
|
|
"num_tokens": 1596545.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.778941535949707,
|
|
"epoch": 0.7170846394984326,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.413,
|
|
"mean_token_accuracy": 0.1205380491912365,
|
|
"num_tokens": 1605098.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.692997789382934,
|
|
"epoch": 0.7210031347962382,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4289,
|
|
"mean_token_accuracy": 0.12432634681463242,
|
|
"num_tokens": 1613760.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.82074499130249,
|
|
"epoch": 0.7249216300940439,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4713,
|
|
"mean_token_accuracy": 0.11785945072770118,
|
|
"num_tokens": 1622345.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.9439520835876465,
|
|
"epoch": 0.7288401253918495,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4646,
|
|
"mean_token_accuracy": 0.11846572309732437,
|
|
"num_tokens": 1630706.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.778302955627441,
|
|
"epoch": 0.7327586206896551,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.5255,
|
|
"mean_token_accuracy": 0.11318894028663636,
|
|
"num_tokens": 1639880.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.737925910949707,
|
|
"epoch": 0.7366771159874608,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3759,
|
|
"mean_token_accuracy": 0.11056350916624069,
|
|
"num_tokens": 1648325.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.673644399642944,
|
|
"epoch": 0.7405956112852664,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.3296,
|
|
"mean_token_accuracy": 0.1223592646420002,
|
|
"num_tokens": 1656951.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.774504899978638,
|
|
"epoch": 0.7445141065830722,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4633,
|
|
"mean_token_accuracy": 0.11550791412591935,
|
|
"num_tokens": 1665832.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.7116179943084715,
|
|
"epoch": 0.7484326018808778,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3901,
|
|
"mean_token_accuracy": 0.11769320592284202,
|
|
"num_tokens": 1675116.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.6702745914459225,
|
|
"epoch": 0.7523510971786834,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.3906,
|
|
"mean_token_accuracy": 0.11918239742517471,
|
|
"num_tokens": 1684195.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.749203729629516,
|
|
"epoch": 0.7562695924764891,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3553,
|
|
"mean_token_accuracy": 0.11945358514785767,
|
|
"num_tokens": 1693025.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.579623699188232,
|
|
"epoch": 0.7601880877742947,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3574,
|
|
"mean_token_accuracy": 0.11793003380298614,
|
|
"num_tokens": 1701210.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.806158590316772,
|
|
"epoch": 0.7641065830721003,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4399,
|
|
"mean_token_accuracy": 0.10840679034590721,
|
|
"num_tokens": 1711026.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.795991325378418,
|
|
"epoch": 0.768025078369906,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4699,
|
|
"mean_token_accuracy": 0.11842825114727021,
|
|
"num_tokens": 1719441.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.637369060516358,
|
|
"epoch": 0.7719435736677116,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.4382,
|
|
"mean_token_accuracy": 0.1170246496796608,
|
|
"num_tokens": 1728300.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.819873046875,
|
|
"epoch": 0.7758620689655172,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3394,
|
|
"mean_token_accuracy": 0.12422693520784378,
|
|
"num_tokens": 1736364.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.509377813339233,
|
|
"epoch": 0.7797805642633229,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1694,
|
|
"mean_token_accuracy": 0.12769819796085358,
|
|
"num_tokens": 1745043.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.707762241363525,
|
|
"epoch": 0.7836990595611285,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3271,
|
|
"mean_token_accuracy": 0.12413779944181443,
|
|
"num_tokens": 1754008.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.7836990595611285,
|
|
"eval_entropy": 6.428579299948936,
|
|
"eval_loss": 6.458470821380615,
|
|
"eval_mean_token_accuracy": 0.12136510946578527,
|
|
"eval_num_tokens": 1754008.0,
|
|
"eval_runtime": 2.3817,
|
|
"eval_samples_per_second": 1730.672,
|
|
"eval_steps_per_second": 216.649,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.5504334449768065,
|
|
"epoch": 0.7876175548589341,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999998713243189,
|
|
"loss": 6.3782,
|
|
"mean_token_accuracy": 0.1190693736076355,
|
|
"num_tokens": 1763592.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.7233641147613525,
|
|
"epoch": 0.7915360501567398,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999993485796164,
|
|
"loss": 6.2421,
|
|
"mean_token_accuracy": 0.12647797986865045,
|
|
"num_tokens": 1773111.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.679757022857666,
|
|
"epoch": 0.7954545454545454,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999984237245962,
|
|
"loss": 6.2994,
|
|
"mean_token_accuracy": 0.1249285250902176,
|
|
"num_tokens": 1781230.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.7423137664794925,
|
|
"epoch": 0.799373040752351,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004999970967609109,
|
|
"loss": 6.5284,
|
|
"mean_token_accuracy": 0.1177036553621292,
|
|
"num_tokens": 1790324.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.683147621154785,
|
|
"epoch": 0.8032915360501567,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999953676909322,
|
|
"loss": 6.3785,
|
|
"mean_token_accuracy": 0.1153412863612175,
|
|
"num_tokens": 1798849.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.714632272720337,
|
|
"epoch": 0.8072100313479624,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004999932365177502,
|
|
"loss": 6.2756,
|
|
"mean_token_accuracy": 0.12519139871001245,
|
|
"num_tokens": 1808019.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.707686853408814,
|
|
"epoch": 0.8111285266457681,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999907032451735,
|
|
"loss": 6.3873,
|
|
"mean_token_accuracy": 0.11506786793470383,
|
|
"num_tokens": 1816550.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.570829439163208,
|
|
"epoch": 0.8150470219435737,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999877678777296,
|
|
"loss": 6.2688,
|
|
"mean_token_accuracy": 0.12967576310038567,
|
|
"num_tokens": 1825238.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.612727355957031,
|
|
"epoch": 0.8189655172413793,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999844304206645,
|
|
"loss": 6.2453,
|
|
"mean_token_accuracy": 0.12376979440450668,
|
|
"num_tokens": 1833645.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.662022590637207,
|
|
"epoch": 0.822884012539185,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999806908799428,
|
|
"loss": 6.3151,
|
|
"mean_token_accuracy": 0.12326861396431923,
|
|
"num_tokens": 1842892.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.615734004974366,
|
|
"epoch": 0.8268025078369906,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004999765492622474,
|
|
"loss": 6.3655,
|
|
"mean_token_accuracy": 0.12360376715660096,
|
|
"num_tokens": 1851400.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.753599309921265,
|
|
"epoch": 0.8307210031347962,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999720055749804,
|
|
"loss": 6.3077,
|
|
"mean_token_accuracy": 0.12451629340648651,
|
|
"num_tokens": 1860088.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.641028308868409,
|
|
"epoch": 0.8346394984326019,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999670598262619,
|
|
"loss": 6.3669,
|
|
"mean_token_accuracy": 0.12593452781438827,
|
|
"num_tokens": 1868558.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.610889625549317,
|
|
"epoch": 0.8385579937304075,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999617120249308,
|
|
"loss": 6.3557,
|
|
"mean_token_accuracy": 0.12159334719181061,
|
|
"num_tokens": 1877624.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.735355281829834,
|
|
"epoch": 0.8424764890282131,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999559621805445,
|
|
"loss": 6.4559,
|
|
"mean_token_accuracy": 0.12153566628694534,
|
|
"num_tokens": 1886505.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.470803356170654,
|
|
"epoch": 0.8463949843260188,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999498103033788,
|
|
"loss": 6.2173,
|
|
"mean_token_accuracy": 0.12777653336524963,
|
|
"num_tokens": 1895701.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.584574604034424,
|
|
"epoch": 0.8503134796238244,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999432564044284,
|
|
"loss": 6.1704,
|
|
"mean_token_accuracy": 0.12821756675839424,
|
|
"num_tokens": 1904640.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.62569408416748,
|
|
"epoch": 0.85423197492163,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999363004954058,
|
|
"loss": 6.3907,
|
|
"mean_token_accuracy": 0.11882495284080505,
|
|
"num_tokens": 1914216.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.640165042877197,
|
|
"epoch": 0.8581504702194357,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.3239,
|
|
"mean_token_accuracy": 0.12579737603664398,
|
|
"num_tokens": 1923280.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.66733832359314,
|
|
"epoch": 0.8620689655172413,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999211826975884,
|
|
"loss": 6.3342,
|
|
"mean_token_accuracy": 0.1269496500492096,
|
|
"num_tokens": 1931403.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.729711484909058,
|
|
"epoch": 0.8659874608150471,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999130208358114,
|
|
"loss": 6.3518,
|
|
"mean_token_accuracy": 0.12522414475679397,
|
|
"num_tokens": 1940774.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.65296893119812,
|
|
"epoch": 0.8699059561128527,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999044570179983,
|
|
"loss": 6.4568,
|
|
"mean_token_accuracy": 0.12021067664027214,
|
|
"num_tokens": 1950045.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.7069206714630125,
|
|
"epoch": 0.8738244514106583,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998954912594541,
|
|
"loss": 6.301,
|
|
"mean_token_accuracy": 0.12480059787631034,
|
|
"num_tokens": 1959303.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.551336765289307,
|
|
"epoch": 0.877742946708464,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998861235762018,
|
|
"loss": 6.2817,
|
|
"mean_token_accuracy": 0.12389473170042038,
|
|
"num_tokens": 1968583.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.684165620803833,
|
|
"epoch": 0.8816614420062696,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998763539849832,
|
|
"loss": 6.4014,
|
|
"mean_token_accuracy": 0.1191262848675251,
|
|
"num_tokens": 1977489.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.577025747299194,
|
|
"epoch": 0.8855799373040752,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004998661825032579,
|
|
"loss": 6.1669,
|
|
"mean_token_accuracy": 0.12639350816607475,
|
|
"num_tokens": 1985308.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.553803873062134,
|
|
"epoch": 0.8894984326018809,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004998556091492043,
|
|
"loss": 6.3264,
|
|
"mean_token_accuracy": 0.12001217380166054,
|
|
"num_tokens": 1993930.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.5682281970977785,
|
|
"epoch": 0.8934169278996865,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998446339417184,
|
|
"loss": 6.3363,
|
|
"mean_token_accuracy": 0.1184835083782673,
|
|
"num_tokens": 2003802.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.65987720489502,
|
|
"epoch": 0.8973354231974922,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004998332569004149,
|
|
"loss": 6.2649,
|
|
"mean_token_accuracy": 0.13021124452352523,
|
|
"num_tokens": 2011957.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.544471645355225,
|
|
"epoch": 0.9012539184952978,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004998214780456263,
|
|
"loss": 6.2473,
|
|
"mean_token_accuracy": 0.1290348932147026,
|
|
"num_tokens": 2020741.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.585256767272949,
|
|
"epoch": 0.9051724137931034,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998092973984033,
|
|
"loss": 6.3183,
|
|
"mean_token_accuracy": 0.11954792961478233,
|
|
"num_tokens": 2029804.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.665730810165405,
|
|
"epoch": 0.9090909090909091,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997967149805147,
|
|
"loss": 6.3747,
|
|
"mean_token_accuracy": 0.12542299255728723,
|
|
"num_tokens": 2038732.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.597140693664551,
|
|
"epoch": 0.9130094043887147,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004997837308144474,
|
|
"loss": 6.2593,
|
|
"mean_token_accuracy": 0.13063650578260422,
|
|
"num_tokens": 2046362.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.579983711242676,
|
|
"epoch": 0.9169278996865203,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997703449234062,
|
|
"loss": 6.2504,
|
|
"mean_token_accuracy": 0.12467706501483918,
|
|
"num_tokens": 2056018.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.482860279083252,
|
|
"epoch": 0.920846394984326,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997565573313139,
|
|
"loss": 6.2251,
|
|
"mean_token_accuracy": 0.1229049526154995,
|
|
"num_tokens": 2065077.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.643082571029663,
|
|
"epoch": 0.9247648902821317,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997423680628109,
|
|
"loss": 6.2231,
|
|
"mean_token_accuracy": 0.12535155937075615,
|
|
"num_tokens": 2073405.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.363648509979248,
|
|
"epoch": 0.9286833855799373,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000499727777143256,
|
|
"loss": 6.2043,
|
|
"mean_token_accuracy": 0.12422606647014618,
|
|
"num_tokens": 2081891.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.7283204078674315,
|
|
"epoch": 0.932601880877743,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004997127845987255,
|
|
"loss": 6.2677,
|
|
"mean_token_accuracy": 0.1314123049378395,
|
|
"num_tokens": 2090732.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.512512826919556,
|
|
"epoch": 0.9365203761755486,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004996973904560135,
|
|
"loss": 6.2057,
|
|
"mean_token_accuracy": 0.13519041687250138,
|
|
"num_tokens": 2099352.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.588135719299316,
|
|
"epoch": 0.9404388714733543,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004996815947426317,
|
|
"loss": 6.2327,
|
|
"mean_token_accuracy": 0.1259102590382099,
|
|
"num_tokens": 2107365.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.504031848907471,
|
|
"epoch": 0.9443573667711599,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996653974868097,
|
|
"loss": 6.108,
|
|
"mean_token_accuracy": 0.1303439162671566,
|
|
"num_tokens": 2116192.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.478465604782104,
|
|
"epoch": 0.9482758620689655,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996487987174946,
|
|
"loss": 6.2005,
|
|
"mean_token_accuracy": 0.12716533839702607,
|
|
"num_tokens": 2124410.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.561891698837281,
|
|
"epoch": 0.9521943573667712,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996317984643512,
|
|
"loss": 6.2327,
|
|
"mean_token_accuracy": 0.1264433428645134,
|
|
"num_tokens": 2132799.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.604869270324707,
|
|
"epoch": 0.9561128526645768,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996143967577615,
|
|
"loss": 6.3896,
|
|
"mean_token_accuracy": 0.12133820950984955,
|
|
"num_tokens": 2141589.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.596088504791259,
|
|
"epoch": 0.9600313479623824,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004995965936288254,
|
|
"loss": 6.2666,
|
|
"mean_token_accuracy": 0.12742918953299523,
|
|
"num_tokens": 2150704.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.4609973430633545,
|
|
"epoch": 0.9639498432601881,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004995783891093597,
|
|
"loss": 6.1859,
|
|
"mean_token_accuracy": 0.12639283537864685,
|
|
"num_tokens": 2159587.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.416796541213989,
|
|
"epoch": 0.9678683385579937,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499559783231899,
|
|
"loss": 6.1813,
|
|
"mean_token_accuracy": 0.12408439218997955,
|
|
"num_tokens": 2169071.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.5871048927307125,
|
|
"epoch": 0.9717868338557993,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004995407760296952,
|
|
"loss": 6.2625,
|
|
"mean_token_accuracy": 0.12216611951589584,
|
|
"num_tokens": 2178248.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.513316440582275,
|
|
"epoch": 0.975705329153605,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995213675367169,
|
|
"loss": 6.1756,
|
|
"mean_token_accuracy": 0.12269840165972709,
|
|
"num_tokens": 2187520.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.447169017791748,
|
|
"epoch": 0.9796238244514106,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004995015577876504,
|
|
"loss": 6.2341,
|
|
"mean_token_accuracy": 0.12262409701943397,
|
|
"num_tokens": 2196861.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.51647982597351,
|
|
"epoch": 0.9835423197492164,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499481346817899,
|
|
"loss": 6.1516,
|
|
"mean_token_accuracy": 0.12425718829035759,
|
|
"num_tokens": 2205523.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.421361827850342,
|
|
"epoch": 0.987460815047022,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004994607346635829,
|
|
"loss": 6.1037,
|
|
"mean_token_accuracy": 0.13546343818306922,
|
|
"num_tokens": 2214291.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.582326602935791,
|
|
"epoch": 0.9913793103448276,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004994397213615394,
|
|
"loss": 6.2277,
|
|
"mean_token_accuracy": 0.12956432327628137,
|
|
"num_tokens": 2223093.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.4590717315673825,
|
|
"epoch": 0.9952978056426333,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004994183069493225,
|
|
"loss": 6.1765,
|
|
"mean_token_accuracy": 0.12921596318483353,
|
|
"num_tokens": 2231551.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.417230415344238,
|
|
"epoch": 0.9992163009404389,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004993964914652034,
|
|
"loss": 6.1756,
|
|
"mean_token_accuracy": 0.13530100136995316,
|
|
"num_tokens": 2240650.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.566872024536133,
|
|
"epoch": 1.0031347962382444,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004993742749481699,
|
|
"loss": 6.0889,
|
|
"mean_token_accuracy": 0.12914734631776809,
|
|
"num_tokens": 2249392.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.30173978805542,
|
|
"epoch": 1.0070532915360502,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004993516574379265,
|
|
"loss": 5.909,
|
|
"mean_token_accuracy": 0.1383669890463352,
|
|
"num_tokens": 2258100.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.457141256332397,
|
|
"epoch": 1.0109717868338557,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004993286389748943,
|
|
"loss": 5.9916,
|
|
"mean_token_accuracy": 0.14005570858716965,
|
|
"num_tokens": 2266661.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.414223909378052,
|
|
"epoch": 1.0148902821316614,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004993052196002112,
|
|
"loss": 6.0666,
|
|
"mean_token_accuracy": 0.13289112225174904,
|
|
"num_tokens": 2275475.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.4289984703063965,
|
|
"epoch": 1.0188087774294672,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004992813993557312,
|
|
"loss": 5.9928,
|
|
"mean_token_accuracy": 0.13499743342399598,
|
|
"num_tokens": 2284077.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.367063283920288,
|
|
"epoch": 1.0227272727272727,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499257178284025,
|
|
"loss": 5.9951,
|
|
"mean_token_accuracy": 0.12990057617425918,
|
|
"num_tokens": 2293029.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.366022062301636,
|
|
"epoch": 1.0266457680250785,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004992325564283797,
|
|
"loss": 5.9146,
|
|
"mean_token_accuracy": 0.1453743927180767,
|
|
"num_tokens": 2301574.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.361129570007324,
|
|
"epoch": 1.030564263322884,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004992075338327984,
|
|
"loss": 5.9279,
|
|
"mean_token_accuracy": 0.13405117839574815,
|
|
"num_tokens": 2310366.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.398780679702758,
|
|
"epoch": 1.0344827586206897,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004991821105420006,
|
|
"loss": 6.0517,
|
|
"mean_token_accuracy": 0.12896588742733,
|
|
"num_tokens": 2318943.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.513023424148559,
|
|
"epoch": 1.0384012539184952,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004991562866014219,
|
|
"loss": 6.0441,
|
|
"mean_token_accuracy": 0.13917485103011132,
|
|
"num_tokens": 2327173.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.4111004829406735,
|
|
"epoch": 1.042319749216301,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 5.995,
|
|
"mean_token_accuracy": 0.1368990011513233,
|
|
"num_tokens": 2336194.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.3529339790344235,
|
|
"epoch": 1.0462382445141065,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004991034369562438,
|
|
"loss": 6.063,
|
|
"mean_token_accuracy": 0.13536622151732444,
|
|
"num_tokens": 2344721.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.395602416992188,
|
|
"epoch": 1.0501567398119123,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004990764113460953,
|
|
"loss": 5.9979,
|
|
"mean_token_accuracy": 0.13348241671919822,
|
|
"num_tokens": 2353406.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.485130548477173,
|
|
"epoch": 1.0540752351097178,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004990489852750675,
|
|
"loss": 6.1696,
|
|
"mean_token_accuracy": 0.1272469662129879,
|
|
"num_tokens": 2361896.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.432577991485596,
|
|
"epoch": 1.0579937304075235,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004990211587921751,
|
|
"loss": 5.9842,
|
|
"mean_token_accuracy": 0.13497858494520187,
|
|
"num_tokens": 2371861.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.311699247360229,
|
|
"epoch": 1.061912225705329,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004989929319471487,
|
|
"loss": 5.913,
|
|
"mean_token_accuracy": 0.13923537880182266,
|
|
"num_tokens": 2381555.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.296171426773071,
|
|
"epoch": 1.0658307210031348,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989643047904341,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.14336461573839188,
|
|
"num_tokens": 2389851.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.318877696990967,
|
|
"epoch": 1.0697492163009406,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004989352773731928,
|
|
"loss": 5.8554,
|
|
"mean_token_accuracy": 0.14105819165706635,
|
|
"num_tokens": 2397764.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.365135574340821,
|
|
"epoch": 1.073667711598746,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004989058497473014,
|
|
"loss": 6.0056,
|
|
"mean_token_accuracy": 0.13977990821003913,
|
|
"num_tokens": 2406688.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.48265495300293,
|
|
"epoch": 1.0775862068965518,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004988760219653518,
|
|
"loss": 6.0625,
|
|
"mean_token_accuracy": 0.1358851246535778,
|
|
"num_tokens": 2415475.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.324158716201782,
|
|
"epoch": 1.0815047021943573,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004988457940806513,
|
|
"loss": 6.0203,
|
|
"mean_token_accuracy": 0.13343403711915017,
|
|
"num_tokens": 2425719.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.3412618160247805,
|
|
"epoch": 1.085423197492163,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004988151661472218,
|
|
"loss": 5.9256,
|
|
"mean_token_accuracy": 0.1366223305463791,
|
|
"num_tokens": 2435138.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.364029836654663,
|
|
"epoch": 1.0893416927899686,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004987841382198006,
|
|
"loss": 6.0585,
|
|
"mean_token_accuracy": 0.13833310529589654,
|
|
"num_tokens": 2443976.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.256550264358521,
|
|
"epoch": 1.0932601880877744,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004987527103538394,
|
|
"loss": 5.9495,
|
|
"mean_token_accuracy": 0.13751715123653413,
|
|
"num_tokens": 2453287.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.445509147644043,
|
|
"epoch": 1.09717868338558,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000498720882605505,
|
|
"loss": 6.0438,
|
|
"mean_token_accuracy": 0.13279600888490678,
|
|
"num_tokens": 2462054.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.296915054321289,
|
|
"epoch": 1.1010971786833856,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004986886550316788,
|
|
"loss": 5.947,
|
|
"mean_token_accuracy": 0.14234385043382644,
|
|
"num_tokens": 2470333.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.302438354492187,
|
|
"epoch": 1.1050156739811912,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004986560276899565,
|
|
"loss": 5.9828,
|
|
"mean_token_accuracy": 0.1347782127559185,
|
|
"num_tokens": 2479621.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.349420833587646,
|
|
"epoch": 1.108934169278997,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004986230006386488,
|
|
"loss": 5.9529,
|
|
"mean_token_accuracy": 0.14194037169218063,
|
|
"num_tokens": 2488020.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.34328932762146,
|
|
"epoch": 1.1128526645768024,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004985895739367802,
|
|
"loss": 5.9895,
|
|
"mean_token_accuracy": 0.13879080787301062,
|
|
"num_tokens": 2497779.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.391005897521973,
|
|
"epoch": 1.1167711598746082,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004985557476440895,
|
|
"loss": 5.9477,
|
|
"mean_token_accuracy": 0.13822169974446297,
|
|
"num_tokens": 2506791.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.332854890823365,
|
|
"epoch": 1.1206896551724137,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00049852152182103,
|
|
"loss": 6.0383,
|
|
"mean_token_accuracy": 0.1322232812643051,
|
|
"num_tokens": 2515576.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.408337259292603,
|
|
"epoch": 1.1246081504702194,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004984868965287686,
|
|
"loss": 5.97,
|
|
"mean_token_accuracy": 0.133582466840744,
|
|
"num_tokens": 2523933.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.286949396133423,
|
|
"epoch": 1.1285266457680252,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004984518718291864,
|
|
"loss": 5.9119,
|
|
"mean_token_accuracy": 0.13513418883085251,
|
|
"num_tokens": 2532453.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.339997482299805,
|
|
"epoch": 1.1324451410658307,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004984164477848783,
|
|
"loss": 5.9626,
|
|
"mean_token_accuracy": 0.13394998461008073,
|
|
"num_tokens": 2541149.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.28506236076355,
|
|
"epoch": 1.1363636363636362,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004983806244591528,
|
|
"loss": 5.9253,
|
|
"mean_token_accuracy": 0.14137585908174516,
|
|
"num_tokens": 2550302.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.215384769439697,
|
|
"epoch": 1.140282131661442,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004983444019160318,
|
|
"loss": 5.8043,
|
|
"mean_token_accuracy": 0.14588194265961646,
|
|
"num_tokens": 2558507.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.394910097122192,
|
|
"epoch": 1.1442006269592477,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004983077802202511,
|
|
"loss": 6.0643,
|
|
"mean_token_accuracy": 0.1361568197607994,
|
|
"num_tokens": 2567487.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.337320566177368,
|
|
"epoch": 1.1481191222570533,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004982707594372595,
|
|
"loss": 5.9319,
|
|
"mean_token_accuracy": 0.13447902277112006,
|
|
"num_tokens": 2576232.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.251219940185547,
|
|
"epoch": 1.152037617554859,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000498233339633219,
|
|
"loss": 5.9185,
|
|
"mean_token_accuracy": 0.13763391003012657,
|
|
"num_tokens": 2585081.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.252890396118164,
|
|
"epoch": 1.1559561128526645,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000498195520875005,
|
|
"loss": 5.9087,
|
|
"mean_token_accuracy": 0.14007165879011155,
|
|
"num_tokens": 2593890.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.335671854019165,
|
|
"epoch": 1.1598746081504703,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004981573032302059,
|
|
"loss": 5.9845,
|
|
"mean_token_accuracy": 0.1327165886759758,
|
|
"num_tokens": 2602011.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.2285538673400875,
|
|
"epoch": 1.1637931034482758,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004981186867671225,
|
|
"loss": 5.8807,
|
|
"mean_token_accuracy": 0.14279000610113143,
|
|
"num_tokens": 2610366.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.19538049697876,
|
|
"epoch": 1.1677115987460815,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004980796715547691,
|
|
"loss": 5.8856,
|
|
"mean_token_accuracy": 0.13823929131031037,
|
|
"num_tokens": 2619737.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.363335180282593,
|
|
"epoch": 1.171630094043887,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004980402576628717,
|
|
"loss": 5.9276,
|
|
"mean_token_accuracy": 0.13838957920670508,
|
|
"num_tokens": 2628176.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.1801056385040285,
|
|
"epoch": 1.1755485893416928,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004980004451618697,
|
|
"loss": 5.9338,
|
|
"mean_token_accuracy": 0.13763452768325807,
|
|
"num_tokens": 2636195.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.1755485893416928,
|
|
"eval_entropy": 6.09060923073643,
|
|
"eval_loss": 6.193844795227051,
|
|
"eval_mean_token_accuracy": 0.13458612963355096,
|
|
"eval_num_tokens": 2636195.0,
|
|
"eval_runtime": 2.1834,
|
|
"eval_samples_per_second": 1887.848,
|
|
"eval_steps_per_second": 236.324,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.35936918258667,
|
|
"epoch": 1.1794670846394983,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004979602341229144,
|
|
"loss": 6.0075,
|
|
"mean_token_accuracy": 0.13624352514743804,
|
|
"num_tokens": 2645035.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.323285961151123,
|
|
"epoch": 1.183385579937304,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004979196246178694,
|
|
"loss": 5.9708,
|
|
"mean_token_accuracy": 0.13602299019694328,
|
|
"num_tokens": 2653239.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.202506065368652,
|
|
"epoch": 1.1873040752351098,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004978786167193105,
|
|
"loss": 5.9734,
|
|
"mean_token_accuracy": 0.13476402312517166,
|
|
"num_tokens": 2662532.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.307791614532471,
|
|
"epoch": 1.1912225705329154,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004978372105005254,
|
|
"loss": 5.9624,
|
|
"mean_token_accuracy": 0.13529006019234657,
|
|
"num_tokens": 2671867.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.235004043579101,
|
|
"epoch": 1.1951410658307209,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004977954060355139,
|
|
"loss": 5.9168,
|
|
"mean_token_accuracy": 0.13566251918673516,
|
|
"num_tokens": 2680693.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.360129261016846,
|
|
"epoch": 1.1990595611285266,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004977532033989871,
|
|
"loss": 5.9976,
|
|
"mean_token_accuracy": 0.1387275367975235,
|
|
"num_tokens": 2689228.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.196741771697998,
|
|
"epoch": 1.2029780564263324,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004977106026663681,
|
|
"loss": 5.8828,
|
|
"mean_token_accuracy": 0.1363217107951641,
|
|
"num_tokens": 2698483.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.332476854324341,
|
|
"epoch": 1.206896551724138,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004976676039137914,
|
|
"loss": 5.9887,
|
|
"mean_token_accuracy": 0.13287660852074623,
|
|
"num_tokens": 2706920.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.259786128997803,
|
|
"epoch": 1.2108150470219436,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004976242072181026,
|
|
"loss": 5.8938,
|
|
"mean_token_accuracy": 0.14664451107382775,
|
|
"num_tokens": 2715954.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.210068607330323,
|
|
"epoch": 1.2147335423197492,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004975804126568587,
|
|
"loss": 5.8966,
|
|
"mean_token_accuracy": 0.1397111728787422,
|
|
"num_tokens": 2724890.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.3650350093841555,
|
|
"epoch": 1.218652037617555,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004975362203083277,
|
|
"loss": 5.9337,
|
|
"mean_token_accuracy": 0.13924732357263564,
|
|
"num_tokens": 2733688.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.282755756378174,
|
|
"epoch": 1.2225705329153604,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004974916302514886,
|
|
"loss": 5.9404,
|
|
"mean_token_accuracy": 0.13459183499217034,
|
|
"num_tokens": 2743364.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.180372476577759,
|
|
"epoch": 1.2264890282131662,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 5.9722,
|
|
"mean_token_accuracy": 0.1317303143441677,
|
|
"num_tokens": 2754276.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.299848127365112,
|
|
"epoch": 1.2304075235109717,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004974012573323545,
|
|
"loss": 5.7726,
|
|
"mean_token_accuracy": 0.14964899495244027,
|
|
"num_tokens": 2762680.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.078790330886841,
|
|
"epoch": 1.2343260188087775,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004973554746315709,
|
|
"loss": 5.8821,
|
|
"mean_token_accuracy": 0.14441804736852645,
|
|
"num_tokens": 2771161.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.309708786010742,
|
|
"epoch": 1.238244514106583,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004973092945455009,
|
|
"loss": 5.8555,
|
|
"mean_token_accuracy": 0.14814924746751784,
|
|
"num_tokens": 2779321.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.213297414779663,
|
|
"epoch": 1.2421630094043887,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004972627171566757,
|
|
"loss": 6.008,
|
|
"mean_token_accuracy": 0.13737262412905693,
|
|
"num_tokens": 2788425.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.247297859191894,
|
|
"epoch": 1.2460815047021945,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004972157425483368,
|
|
"loss": 5.8721,
|
|
"mean_token_accuracy": 0.1450733035802841,
|
|
"num_tokens": 2796747.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.150865268707276,
|
|
"epoch": 1.25,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004971683708044353,
|
|
"loss": 5.9237,
|
|
"mean_token_accuracy": 0.14032647535204887,
|
|
"num_tokens": 2805335.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.296543598175049,
|
|
"epoch": 1.2539184952978055,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004971206020096323,
|
|
"loss": 5.8147,
|
|
"mean_token_accuracy": 0.1457967095077038,
|
|
"num_tokens": 2813350.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.279100036621093,
|
|
"epoch": 1.2578369905956113,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004970724362492984,
|
|
"loss": 5.9766,
|
|
"mean_token_accuracy": 0.13843000829219818,
|
|
"num_tokens": 2821922.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.185999870300293,
|
|
"epoch": 1.261755485893417,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004970238736095135,
|
|
"loss": 5.9067,
|
|
"mean_token_accuracy": 0.141528981924057,
|
|
"num_tokens": 2830832.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.29846568107605,
|
|
"epoch": 1.2656739811912225,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004969749141770671,
|
|
"loss": 5.8959,
|
|
"mean_token_accuracy": 0.1415686458349228,
|
|
"num_tokens": 2839699.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.213438987731934,
|
|
"epoch": 1.2695924764890283,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004969255580394575,
|
|
"loss": 5.9603,
|
|
"mean_token_accuracy": 0.1428991936147213,
|
|
"num_tokens": 2848293.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.2712019920349125,
|
|
"epoch": 1.2735109717868338,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000496875805284892,
|
|
"loss": 5.9051,
|
|
"mean_token_accuracy": 0.14353382885456084,
|
|
"num_tokens": 2856468.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.231787109375,
|
|
"epoch": 1.2774294670846396,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004968256560022871,
|
|
"loss": 5.8405,
|
|
"mean_token_accuracy": 0.14478754997253418,
|
|
"num_tokens": 2865366.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.161731481552124,
|
|
"epoch": 1.281347962382445,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004967751102812676,
|
|
"loss": 5.8928,
|
|
"mean_token_accuracy": 0.14282298386096953,
|
|
"num_tokens": 2873520.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.248449373245239,
|
|
"epoch": 1.2852664576802508,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004967241682121669,
|
|
"loss": 5.9653,
|
|
"mean_token_accuracy": 0.13820288628339766,
|
|
"num_tokens": 2883324.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.193999719619751,
|
|
"epoch": 1.2891849529780564,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004966728298860267,
|
|
"loss": 5.8941,
|
|
"mean_token_accuracy": 0.1412993147969246,
|
|
"num_tokens": 2892200.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.203446674346924,
|
|
"epoch": 1.293103448275862,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004966210953945969,
|
|
"loss": 5.8757,
|
|
"mean_token_accuracy": 0.1418458916246891,
|
|
"num_tokens": 2900597.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.283433151245117,
|
|
"epoch": 1.2970219435736676,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004965689648303355,
|
|
"loss": 5.9066,
|
|
"mean_token_accuracy": 0.13762896731495858,
|
|
"num_tokens": 2909869.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.118806505203247,
|
|
"epoch": 1.3009404388714734,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004965164382864083,
|
|
"loss": 5.9333,
|
|
"mean_token_accuracy": 0.14366980865597725,
|
|
"num_tokens": 2919166.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.372979736328125,
|
|
"epoch": 1.3048589341692791,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004964635158566886,
|
|
"loss": 5.9487,
|
|
"mean_token_accuracy": 0.1401916429400444,
|
|
"num_tokens": 2927884.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.13509955406189,
|
|
"epoch": 1.3087774294670846,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004964101976357574,
|
|
"loss": 5.8778,
|
|
"mean_token_accuracy": 0.1481625370681286,
|
|
"num_tokens": 2936737.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.248805522918701,
|
|
"epoch": 1.3126959247648902,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000496356483718903,
|
|
"loss": 5.9486,
|
|
"mean_token_accuracy": 0.14187579602003098,
|
|
"num_tokens": 2944203.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.289197778701782,
|
|
"epoch": 1.316614420062696,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004963023742021211,
|
|
"loss": 5.9804,
|
|
"mean_token_accuracy": 0.13468516916036605,
|
|
"num_tokens": 2952639.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.20129885673523,
|
|
"epoch": 1.3205329153605017,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000496247869182114,
|
|
"loss": 5.9024,
|
|
"mean_token_accuracy": 0.141036444157362,
|
|
"num_tokens": 2961482.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.229419851303101,
|
|
"epoch": 1.3244514106583072,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004961929687562909,
|
|
"loss": 5.9176,
|
|
"mean_token_accuracy": 0.13823169395327567,
|
|
"num_tokens": 2970927.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.134537839889527,
|
|
"epoch": 1.328369905956113,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004961376730227681,
|
|
"loss": 5.8578,
|
|
"mean_token_accuracy": 0.1452302888035774,
|
|
"num_tokens": 2979329.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.19011869430542,
|
|
"epoch": 1.3322884012539185,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004960819820803675,
|
|
"loss": 5.8242,
|
|
"mean_token_accuracy": 0.14809555038809777,
|
|
"num_tokens": 2986894.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.190591526031494,
|
|
"epoch": 1.3362068965517242,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004960258960286185,
|
|
"loss": 5.8267,
|
|
"mean_token_accuracy": 0.14454643577337264,
|
|
"num_tokens": 2994721.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.232819938659668,
|
|
"epoch": 1.3401253918495297,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004959694149677553,
|
|
"loss": 5.9791,
|
|
"mean_token_accuracy": 0.1366499751806259,
|
|
"num_tokens": 3003165.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.267041444778442,
|
|
"epoch": 1.3440438871473355,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004959125389987193,
|
|
"loss": 5.962,
|
|
"mean_token_accuracy": 0.13347446471452712,
|
|
"num_tokens": 3012178.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.170607376098633,
|
|
"epoch": 1.347962382445141,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004958552682231567,
|
|
"loss": 5.8462,
|
|
"mean_token_accuracy": 0.14583885446190833,
|
|
"num_tokens": 3021685.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.163441276550293,
|
|
"epoch": 1.3518808777429467,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004957976027434199,
|
|
"loss": 5.905,
|
|
"mean_token_accuracy": 0.13857561498880386,
|
|
"num_tokens": 3030667.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.207879066467285,
|
|
"epoch": 1.3557993730407523,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004957395426625663,
|
|
"loss": 5.9571,
|
|
"mean_token_accuracy": 0.13924508094787597,
|
|
"num_tokens": 3039501.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.172082567214966,
|
|
"epoch": 1.359717868338558,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004956810880843587,
|
|
"loss": 5.8773,
|
|
"mean_token_accuracy": 0.13998890668153763,
|
|
"num_tokens": 3048547.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.289714097976685,
|
|
"epoch": 1.3636363636363638,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000495622239113265,
|
|
"loss": 5.8515,
|
|
"mean_token_accuracy": 0.14459400475025178,
|
|
"num_tokens": 3056521.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.052374267578125,
|
|
"epoch": 1.3675548589341693,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004955629958544577,
|
|
"loss": 5.8331,
|
|
"mean_token_accuracy": 0.14432888403534888,
|
|
"num_tokens": 3066007.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.292523574829102,
|
|
"epoch": 1.3714733542319748,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004955033584138143,
|
|
"loss": 6.021,
|
|
"mean_token_accuracy": 0.13402899950742722,
|
|
"num_tokens": 3075111.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.16716947555542,
|
|
"epoch": 1.3753918495297806,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004954433268979164,
|
|
"loss": 5.8181,
|
|
"mean_token_accuracy": 0.15461900532245637,
|
|
"num_tokens": 3083766.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.132131290435791,
|
|
"epoch": 1.3793103448275863,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004953829014140502,
|
|
"loss": 5.8937,
|
|
"mean_token_accuracy": 0.14321358725428582,
|
|
"num_tokens": 3092103.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.174644136428833,
|
|
"epoch": 1.3832288401253918,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004953220820702057,
|
|
"loss": 5.875,
|
|
"mean_token_accuracy": 0.14289139658212663,
|
|
"num_tokens": 3101286.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.334795522689819,
|
|
"epoch": 1.3871473354231976,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004952608689750771,
|
|
"loss": 5.9036,
|
|
"mean_token_accuracy": 0.14129724800586702,
|
|
"num_tokens": 3110554.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.110700988769532,
|
|
"epoch": 1.391065830721003,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004951992622380619,
|
|
"loss": 5.7763,
|
|
"mean_token_accuracy": 0.1512333407998085,
|
|
"num_tokens": 3119265.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.153517770767212,
|
|
"epoch": 1.3949843260188088,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004951372619692615,
|
|
"loss": 5.8016,
|
|
"mean_token_accuracy": 0.1479355089366436,
|
|
"num_tokens": 3127962.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.194212484359741,
|
|
"epoch": 1.3989028213166144,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004950748682794804,
|
|
"loss": 5.9188,
|
|
"mean_token_accuracy": 0.1445128470659256,
|
|
"num_tokens": 3137265.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.110380983352661,
|
|
"epoch": 1.40282131661442,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004950120812802262,
|
|
"loss": 5.767,
|
|
"mean_token_accuracy": 0.14632658436894416,
|
|
"num_tokens": 3146458.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.19427261352539,
|
|
"epoch": 1.4067398119122256,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004949489010837095,
|
|
"loss": 5.9178,
|
|
"mean_token_accuracy": 0.1385168254375458,
|
|
"num_tokens": 3155557.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.162668991088867,
|
|
"epoch": 1.4106583072100314,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 5.7999,
|
|
"mean_token_accuracy": 0.1494093120098114,
|
|
"num_tokens": 3163478.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.165596389770508,
|
|
"epoch": 1.414576802507837,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000494821361551244,
|
|
"loss": 5.8777,
|
|
"mean_token_accuracy": 0.14174044951796533,
|
|
"num_tokens": 3172632.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.133595561981201,
|
|
"epoch": 1.4184952978056427,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004947570024432291,
|
|
"loss": 5.9101,
|
|
"mean_token_accuracy": 0.14648913517594336,
|
|
"num_tokens": 3181325.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.176501178741455,
|
|
"epoch": 1.4224137931034484,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004946922505938189,
|
|
"loss": 5.9646,
|
|
"mean_token_accuracy": 0.14801075905561448,
|
|
"num_tokens": 3189783.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.189005708694458,
|
|
"epoch": 1.426332288401254,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004946271061187354,
|
|
"loss": 5.81,
|
|
"mean_token_accuracy": 0.13777727410197257,
|
|
"num_tokens": 3198605.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.147968006134033,
|
|
"epoch": 1.4302507836990594,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004945615691344025,
|
|
"loss": 5.9189,
|
|
"mean_token_accuracy": 0.135958693921566,
|
|
"num_tokens": 3208061.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.095811271667481,
|
|
"epoch": 1.4341692789968652,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004944956397579453,
|
|
"loss": 5.8171,
|
|
"mean_token_accuracy": 0.148879474401474,
|
|
"num_tokens": 3216093.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.124493837356567,
|
|
"epoch": 1.438087774294671,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004944293181071902,
|
|
"loss": 5.8245,
|
|
"mean_token_accuracy": 0.14754900187253953,
|
|
"num_tokens": 3224446.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.2404640197753904,
|
|
"epoch": 1.4420062695924765,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004943626043006649,
|
|
"loss": 5.9212,
|
|
"mean_token_accuracy": 0.14390360936522484,
|
|
"num_tokens": 3233398.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.142076921463013,
|
|
"epoch": 1.4459247648902822,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000494295498457598,
|
|
"loss": 5.9383,
|
|
"mean_token_accuracy": 0.1396064966917038,
|
|
"num_tokens": 3243391.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.13278021812439,
|
|
"epoch": 1.4498432601880877,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004942280006979179,
|
|
"loss": 5.8614,
|
|
"mean_token_accuracy": 0.147974494099617,
|
|
"num_tokens": 3251954.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.173754930496216,
|
|
"epoch": 1.4537617554858935,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004941601111422546,
|
|
"loss": 5.8754,
|
|
"mean_token_accuracy": 0.1432081326842308,
|
|
"num_tokens": 3261121.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.064755916595459,
|
|
"epoch": 1.457680250783699,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004940918299119375,
|
|
"loss": 5.7942,
|
|
"mean_token_accuracy": 0.1478736862540245,
|
|
"num_tokens": 3270179.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.044514274597168,
|
|
"epoch": 1.4615987460815048,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004940231571289962,
|
|
"loss": 5.8096,
|
|
"mean_token_accuracy": 0.14012864977121353,
|
|
"num_tokens": 3279749.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.206891250610352,
|
|
"epoch": 1.4655172413793103,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004939540929161603,
|
|
"loss": 5.8172,
|
|
"mean_token_accuracy": 0.14891130328178406,
|
|
"num_tokens": 3288458.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.111593770980835,
|
|
"epoch": 1.469435736677116,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004938846373968586,
|
|
"loss": 5.7945,
|
|
"mean_token_accuracy": 0.1456505686044693,
|
|
"num_tokens": 3297231.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.143124151229858,
|
|
"epoch": 1.4733542319749215,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004938147906952194,
|
|
"loss": 5.954,
|
|
"mean_token_accuracy": 0.14113954305648804,
|
|
"num_tokens": 3306915.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.137325477600098,
|
|
"epoch": 1.4772727272727273,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00049374455293607,
|
|
"loss": 5.8857,
|
|
"mean_token_accuracy": 0.14078657552599907,
|
|
"num_tokens": 3315985.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.214253759384155,
|
|
"epoch": 1.481191222570533,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004936739242449369,
|
|
"loss": 5.8758,
|
|
"mean_token_accuracy": 0.1430355839431286,
|
|
"num_tokens": 3324924.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.155929136276245,
|
|
"epoch": 1.4851097178683386,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004936029047480447,
|
|
"loss": 5.8783,
|
|
"mean_token_accuracy": 0.14333493635058403,
|
|
"num_tokens": 3333464.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.097671461105347,
|
|
"epoch": 1.489028213166144,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004935314945723171,
|
|
"loss": 5.7839,
|
|
"mean_token_accuracy": 0.15236787870526314,
|
|
"num_tokens": 3342441.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.173409175872803,
|
|
"epoch": 1.4929467084639498,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004934596938453754,
|
|
"loss": 5.821,
|
|
"mean_token_accuracy": 0.14437680393457414,
|
|
"num_tokens": 3351338.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.130248737335205,
|
|
"epoch": 1.4968652037617556,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004933875026955391,
|
|
"loss": 5.7985,
|
|
"mean_token_accuracy": 0.1431201823055744,
|
|
"num_tokens": 3360356.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.156800270080566,
|
|
"epoch": 1.500783699059561,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004933149212518258,
|
|
"loss": 5.8815,
|
|
"mean_token_accuracy": 0.13685129433870316,
|
|
"num_tokens": 3369485.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.118210554122925,
|
|
"epoch": 1.5047021943573666,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004932419496439501,
|
|
"loss": 5.8438,
|
|
"mean_token_accuracy": 0.14993089586496353,
|
|
"num_tokens": 3378355.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.071452331542969,
|
|
"epoch": 1.5086206896551724,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000493168588002324,
|
|
"loss": 5.7576,
|
|
"mean_token_accuracy": 0.14710098132491112,
|
|
"num_tokens": 3386729.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.203466367721558,
|
|
"epoch": 1.5125391849529781,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004930948364580569,
|
|
"loss": 5.9736,
|
|
"mean_token_accuracy": 0.1357329525053501,
|
|
"num_tokens": 3395822.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.1295530796051025,
|
|
"epoch": 1.5164576802507836,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004930206951429546,
|
|
"loss": 5.881,
|
|
"mean_token_accuracy": 0.1476832464337349,
|
|
"num_tokens": 3404671.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.156378221511841,
|
|
"epoch": 1.5203761755485894,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004929461641895197,
|
|
"loss": 5.8247,
|
|
"mean_token_accuracy": 0.14363798052072524,
|
|
"num_tokens": 3413289.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.126182746887207,
|
|
"epoch": 1.5242946708463951,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000492871243730951,
|
|
"loss": 5.8695,
|
|
"mean_token_accuracy": 0.14048027992248535,
|
|
"num_tokens": 3421960.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.150514125823975,
|
|
"epoch": 1.5282131661442007,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004927959339011437,
|
|
"loss": 5.8708,
|
|
"mean_token_accuracy": 0.14263076931238175,
|
|
"num_tokens": 3431958.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.982177448272705,
|
|
"epoch": 1.5321316614420062,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004927202348346885,
|
|
"loss": 5.6703,
|
|
"mean_token_accuracy": 0.1512327566742897,
|
|
"num_tokens": 3441438.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.119736242294311,
|
|
"epoch": 1.536050156739812,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000492644146666872,
|
|
"loss": 5.7949,
|
|
"mean_token_accuracy": 0.14976613372564315,
|
|
"num_tokens": 3450176.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.021003246307373,
|
|
"epoch": 1.5399686520376177,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004925676695336761,
|
|
"loss": 5.7823,
|
|
"mean_token_accuracy": 0.14913898408412934,
|
|
"num_tokens": 3458685.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.1155863285064695,
|
|
"epoch": 1.5438871473354232,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004924908035717777,
|
|
"loss": 5.9356,
|
|
"mean_token_accuracy": 0.1392411969602108,
|
|
"num_tokens": 3468362.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.1926249980926515,
|
|
"epoch": 1.5478056426332287,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000492413548918549,
|
|
"loss": 5.8659,
|
|
"mean_token_accuracy": 0.14648685678839685,
|
|
"num_tokens": 3476791.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.058112716674804,
|
|
"epoch": 1.5517241379310345,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004923359057120563,
|
|
"loss": 5.7778,
|
|
"mean_token_accuracy": 0.1415088504552841,
|
|
"num_tokens": 3486486.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.014009571075439,
|
|
"epoch": 1.5556426332288402,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004922578740910608,
|
|
"loss": 5.7326,
|
|
"mean_token_accuracy": 0.15146091133356093,
|
|
"num_tokens": 3496145.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.045761203765869,
|
|
"epoch": 1.5595611285266457,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004921794541950177,
|
|
"loss": 5.7224,
|
|
"mean_token_accuracy": 0.15197996199131011,
|
|
"num_tokens": 3504644.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.051179885864258,
|
|
"epoch": 1.5634796238244513,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004921006461640758,
|
|
"loss": 5.7656,
|
|
"mean_token_accuracy": 0.144666650891304,
|
|
"num_tokens": 3513126.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.922442150115967,
|
|
"epoch": 1.567398119122257,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000492021450139078,
|
|
"loss": 5.7135,
|
|
"mean_token_accuracy": 0.14842752516269683,
|
|
"num_tokens": 3521505.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.567398119122257,
|
|
"eval_entropy": 5.850576069003852,
|
|
"eval_loss": 6.009803295135498,
|
|
"eval_mean_token_accuracy": 0.14355941752130671,
|
|
"eval_num_tokens": 3521505.0,
|
|
"eval_runtime": 2.3602,
|
|
"eval_samples_per_second": 1746.465,
|
|
"eval_steps_per_second": 218.626,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.168898010253907,
|
|
"epoch": 1.5713166144200628,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004919418662615605,
|
|
"loss": 5.9513,
|
|
"mean_token_accuracy": 0.14074478149414063,
|
|
"num_tokens": 3530008.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.191672611236572,
|
|
"epoch": 1.5752351097178683,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004918618946737525,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.146812741458416,
|
|
"num_tokens": 3538778.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.0751337051391605,
|
|
"epoch": 1.579153605015674,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004917815355185762,
|
|
"loss": 5.8072,
|
|
"mean_token_accuracy": 0.14341796189546585,
|
|
"num_tokens": 3547108.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.102761268615723,
|
|
"epoch": 1.5830721003134798,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004917007889396464,
|
|
"loss": 5.8427,
|
|
"mean_token_accuracy": 0.13878570571541787,
|
|
"num_tokens": 3556094.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.042416763305664,
|
|
"epoch": 1.5869905956112853,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004916196550812706,
|
|
"loss": 5.7117,
|
|
"mean_token_accuracy": 0.1559369221329689,
|
|
"num_tokens": 3564222.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.955806732177734,
|
|
"epoch": 1.5909090909090908,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004915381340884477,
|
|
"loss": 5.7144,
|
|
"mean_token_accuracy": 0.15095172077417374,
|
|
"num_tokens": 3573123.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.107495069503784,
|
|
"epoch": 1.5948275862068966,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 5.7444,
|
|
"mean_token_accuracy": 0.15329221189022063,
|
|
"num_tokens": 3581462.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.975087738037109,
|
|
"epoch": 1.5987460815047023,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004913739312829181,
|
|
"loss": 5.7756,
|
|
"mean_token_accuracy": 0.14905810505151748,
|
|
"num_tokens": 3590718.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.060779047012329,
|
|
"epoch": 1.6026645768025078,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004912912497636683,
|
|
"loss": 5.8043,
|
|
"mean_token_accuracy": 0.14534393101930618,
|
|
"num_tokens": 3599890.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.228052043914795,
|
|
"epoch": 1.6065830721003134,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004912081816968853,
|
|
"loss": 5.8721,
|
|
"mean_token_accuracy": 0.14821842312812805,
|
|
"num_tokens": 3607927.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.054880857467651,
|
|
"epoch": 1.6105015673981191,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000491124727231025,
|
|
"loss": 5.7357,
|
|
"mean_token_accuracy": 0.15218927413225175,
|
|
"num_tokens": 3615935.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.981013679504395,
|
|
"epoch": 1.6144200626959249,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004910408865152343,
|
|
"loss": 5.7708,
|
|
"mean_token_accuracy": 0.14986267387866975,
|
|
"num_tokens": 3624475.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.186362886428833,
|
|
"epoch": 1.6183385579937304,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004909566596993498,
|
|
"loss": 5.955,
|
|
"mean_token_accuracy": 0.13607163354754448,
|
|
"num_tokens": 3633995.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.160084772109985,
|
|
"epoch": 1.622257053291536,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004908720469338988,
|
|
"loss": 5.8231,
|
|
"mean_token_accuracy": 0.1483114928007126,
|
|
"num_tokens": 3643121.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.011912488937378,
|
|
"epoch": 1.6261755485893417,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004907870483700979,
|
|
"loss": 5.767,
|
|
"mean_token_accuracy": 0.14670649468898772,
|
|
"num_tokens": 3652285.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.0002540111541744,
|
|
"epoch": 1.6300940438871474,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004907016641598534,
|
|
"loss": 5.8349,
|
|
"mean_token_accuracy": 0.1400149181485176,
|
|
"num_tokens": 3661441.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.07007417678833,
|
|
"epoch": 1.634012539184953,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004906158944557607,
|
|
"loss": 5.724,
|
|
"mean_token_accuracy": 0.15594304949045182,
|
|
"num_tokens": 3669364.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.088734722137451,
|
|
"epoch": 1.6379310344827587,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000490529739411104,
|
|
"loss": 5.8082,
|
|
"mean_token_accuracy": 0.14825862348079683,
|
|
"num_tokens": 3677848.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.017403984069825,
|
|
"epoch": 1.6418495297805644,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004904431991798565,
|
|
"loss": 5.6921,
|
|
"mean_token_accuracy": 0.1556319311261177,
|
|
"num_tokens": 3686617.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.066302490234375,
|
|
"epoch": 1.64576802507837,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004903562739166797,
|
|
"loss": 5.7076,
|
|
"mean_token_accuracy": 0.1490858107805252,
|
|
"num_tokens": 3695106.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.948085021972656,
|
|
"epoch": 1.6496865203761755,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004902689637769229,
|
|
"loss": 5.6303,
|
|
"mean_token_accuracy": 0.1558159440755844,
|
|
"num_tokens": 3703167.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.037562227249145,
|
|
"epoch": 1.6536050156739812,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004901812689166237,
|
|
"loss": 5.7628,
|
|
"mean_token_accuracy": 0.14580009058117865,
|
|
"num_tokens": 3711597.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.018727540969849,
|
|
"epoch": 1.657523510971787,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004900931894925069,
|
|
"loss": 5.8524,
|
|
"mean_token_accuracy": 0.14601681977510453,
|
|
"num_tokens": 3720277.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.954285335540772,
|
|
"epoch": 1.6614420062695925,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004900047256619849,
|
|
"loss": 5.8253,
|
|
"mean_token_accuracy": 0.14634145498275758,
|
|
"num_tokens": 3729832.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.112446212768555,
|
|
"epoch": 1.665360501567398,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004899158775831566,
|
|
"loss": 5.7672,
|
|
"mean_token_accuracy": 0.1511450231075287,
|
|
"num_tokens": 3738390.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.036515760421753,
|
|
"epoch": 1.6692789968652038,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004898266454148081,
|
|
"loss": 5.8373,
|
|
"mean_token_accuracy": 0.14870869666337966,
|
|
"num_tokens": 3747046.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.033357238769531,
|
|
"epoch": 1.6731974921630095,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004897370293164119,
|
|
"loss": 5.7965,
|
|
"mean_token_accuracy": 0.14350008666515351,
|
|
"num_tokens": 3755818.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.17767014503479,
|
|
"epoch": 1.677115987460815,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004896470294481262,
|
|
"loss": 5.8917,
|
|
"mean_token_accuracy": 0.14540392458438872,
|
|
"num_tokens": 3764123.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.047296762466431,
|
|
"epoch": 1.6810344827586206,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004895566459707954,
|
|
"loss": 5.9319,
|
|
"mean_token_accuracy": 0.13513909205794333,
|
|
"num_tokens": 3773105.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.10132737159729,
|
|
"epoch": 1.6849529780564263,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004894658790459498,
|
|
"loss": 5.7868,
|
|
"mean_token_accuracy": 0.15084390938282013,
|
|
"num_tokens": 3781417.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.154627943038941,
|
|
"epoch": 1.688871473354232,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004893747288358041,
|
|
"loss": 5.8618,
|
|
"mean_token_accuracy": 0.150341434776783,
|
|
"num_tokens": 3789657.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.071105241775513,
|
|
"epoch": 1.6927899686520376,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000489283195503259,
|
|
"loss": 5.8906,
|
|
"mean_token_accuracy": 0.1433873564004898,
|
|
"num_tokens": 3798700.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.07456259727478,
|
|
"epoch": 1.6967084639498433,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000489191279211899,
|
|
"loss": 5.7295,
|
|
"mean_token_accuracy": 0.1520608313381672,
|
|
"num_tokens": 3807299.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.024093437194824,
|
|
"epoch": 1.700626959247649,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004890989801259935,
|
|
"loss": 5.8345,
|
|
"mean_token_accuracy": 0.14739376604557036,
|
|
"num_tokens": 3816292.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.15645022392273,
|
|
"epoch": 1.7045454545454546,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000489006298410496,
|
|
"loss": 5.7609,
|
|
"mean_token_accuracy": 0.14776830971240998,
|
|
"num_tokens": 3825070.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.04476408958435,
|
|
"epoch": 1.70846394984326,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004889132342310438,
|
|
"loss": 5.7305,
|
|
"mean_token_accuracy": 0.15002637058496476,
|
|
"num_tokens": 3834016.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.054940891265869,
|
|
"epoch": 1.7123824451410659,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004888197877539577,
|
|
"loss": 5.7812,
|
|
"mean_token_accuracy": 0.15201978236436844,
|
|
"num_tokens": 3842172.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.996536064147949,
|
|
"epoch": 1.7163009404388716,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004887259591462417,
|
|
"loss": 5.6905,
|
|
"mean_token_accuracy": 0.14917295128107072,
|
|
"num_tokens": 3851109.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.021609687805176,
|
|
"epoch": 1.7202194357366771,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004886317485755825,
|
|
"loss": 5.8203,
|
|
"mean_token_accuracy": 0.14999141842126845,
|
|
"num_tokens": 3859656.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.0394346714019775,
|
|
"epoch": 1.7241379310344827,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004885371562103498,
|
|
"loss": 5.6698,
|
|
"mean_token_accuracy": 0.15846703946590424,
|
|
"num_tokens": 3868698.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.975643539428711,
|
|
"epoch": 1.7280564263322884,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004884421822195957,
|
|
"loss": 5.7696,
|
|
"mean_token_accuracy": 0.14646813422441482,
|
|
"num_tokens": 3877474.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.059563016891479,
|
|
"epoch": 1.7319749216300941,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004883468267730538,
|
|
"loss": 5.7372,
|
|
"mean_token_accuracy": 0.15244515836238862,
|
|
"num_tokens": 3886328.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.049549341201782,
|
|
"epoch": 1.7358934169278997,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048825109004114006,
|
|
"loss": 5.7102,
|
|
"mean_token_accuracy": 0.15658637881278992,
|
|
"num_tokens": 3894424.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.026647138595581,
|
|
"epoch": 1.7398119122257052,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004881549721949513,
|
|
"loss": 5.8139,
|
|
"mean_token_accuracy": 0.14736466631293296,
|
|
"num_tokens": 3903746.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.970228815078736,
|
|
"epoch": 1.743730407523511,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004880584734062658,
|
|
"loss": 5.7959,
|
|
"mean_token_accuracy": 0.14859797954559326,
|
|
"num_tokens": 3912033.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.981308507919311,
|
|
"epoch": 1.7476489028213167,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004879615938475425,
|
|
"loss": 5.681,
|
|
"mean_token_accuracy": 0.1470841333270073,
|
|
"num_tokens": 3920388.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.010033988952637,
|
|
"epoch": 1.7515673981191222,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004878643336919209,
|
|
"loss": 5.8717,
|
|
"mean_token_accuracy": 0.14416417405009269,
|
|
"num_tokens": 3929926.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.208255100250244,
|
|
"epoch": 1.7554858934169277,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004877666931132206,
|
|
"loss": 5.852,
|
|
"mean_token_accuracy": 0.13953940123319625,
|
|
"num_tokens": 3938696.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.987451362609863,
|
|
"epoch": 1.7594043887147337,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004876686722859413,
|
|
"loss": 5.6964,
|
|
"mean_token_accuracy": 0.15680383443832396,
|
|
"num_tokens": 3947537.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.061147260665893,
|
|
"epoch": 1.7633228840125392,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000487570271385262,
|
|
"loss": 5.68,
|
|
"mean_token_accuracy": 0.14840978011488914,
|
|
"num_tokens": 3956539.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.9781177043914795,
|
|
"epoch": 1.7672413793103448,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004874714905870411,
|
|
"loss": 5.6705,
|
|
"mean_token_accuracy": 0.15344865769147872,
|
|
"num_tokens": 3964927.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.965769195556641,
|
|
"epoch": 1.7711598746081505,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004873723300678159,
|
|
"loss": 5.8019,
|
|
"mean_token_accuracy": 0.14757918044924737,
|
|
"num_tokens": 3973565.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.0911565780639645,
|
|
"epoch": 1.7750783699059562,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048727279000480226,
|
|
"loss": 5.825,
|
|
"mean_token_accuracy": 0.1453205853700638,
|
|
"num_tokens": 3982683.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.074667072296142,
|
|
"epoch": 1.7789968652037618,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 5.7344,
|
|
"mean_token_accuracy": 0.1473068669438362,
|
|
"num_tokens": 3991462.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.804793357849121,
|
|
"epoch": 1.7829153605015673,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004870725719596648,
|
|
"loss": 5.5014,
|
|
"mean_token_accuracy": 0.16297667622566223,
|
|
"num_tokens": 3999821.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.066985702514648,
|
|
"epoch": 1.786833855799373,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004869718943353631,
|
|
"loss": 5.7702,
|
|
"mean_token_accuracy": 0.14976201504468917,
|
|
"num_tokens": 4008655.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.017048025131226,
|
|
"epoch": 1.7907523510971788,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048687083788291656,
|
|
"loss": 5.8053,
|
|
"mean_token_accuracy": 0.15293228179216384,
|
|
"num_tokens": 4017317.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.907472944259643,
|
|
"epoch": 1.7946708463949843,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048676940278292953,
|
|
"loss": 5.6249,
|
|
"mean_token_accuracy": 0.15246410816907882,
|
|
"num_tokens": 4025407.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.970260381698608,
|
|
"epoch": 1.7985893416927898,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048666758921668286,
|
|
"loss": 5.6445,
|
|
"mean_token_accuracy": 0.15337727665901185,
|
|
"num_tokens": 4034310.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.9996294498443605,
|
|
"epoch": 1.8025078369905956,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048656539736613403,
|
|
"loss": 5.7784,
|
|
"mean_token_accuracy": 0.14906976819038392,
|
|
"num_tokens": 4042326.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.948818683624268,
|
|
"epoch": 1.8064263322884013,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004864628274139164,
|
|
"loss": 5.6542,
|
|
"mean_token_accuracy": 0.15948124080896378,
|
|
"num_tokens": 4051248.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.987146425247192,
|
|
"epoch": 1.8103448275862069,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004863598795433391,
|
|
"loss": 5.7799,
|
|
"mean_token_accuracy": 0.14960964098572732,
|
|
"num_tokens": 4060227.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.133766746520996,
|
|
"epoch": 1.8142633228840124,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048625655393838666,
|
|
"loss": 5.8709,
|
|
"mean_token_accuracy": 0.1416142500936985,
|
|
"num_tokens": 4069700.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.029330253601074,
|
|
"epoch": 1.8181818181818183,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004861528507837186,
|
|
"loss": 5.7513,
|
|
"mean_token_accuracy": 0.15127085000276566,
|
|
"num_tokens": 4079363.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.140486812591552,
|
|
"epoch": 1.8221003134796239,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004860487702646695,
|
|
"loss": 5.8175,
|
|
"mean_token_accuracy": 0.1442444458603859,
|
|
"num_tokens": 4087944.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.896052837371826,
|
|
"epoch": 1.8260188087774294,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004859443125672479,
|
|
"loss": 5.5249,
|
|
"mean_token_accuracy": 0.17257627099752426,
|
|
"num_tokens": 4096431.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.919784927368164,
|
|
"epoch": 1.8299373040752351,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004858394778781368,
|
|
"loss": 5.684,
|
|
"mean_token_accuracy": 0.14755771309137344,
|
|
"num_tokens": 4104894.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.968877601623535,
|
|
"epoch": 1.8338557993730409,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004857342663846927,
|
|
"loss": 5.7398,
|
|
"mean_token_accuracy": 0.15060688853263854,
|
|
"num_tokens": 4112878.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.994266843795776,
|
|
"epoch": 1.8377742946708464,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004856286782749456,
|
|
"loss": 5.7428,
|
|
"mean_token_accuracy": 0.15048128515481948,
|
|
"num_tokens": 4122501.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.917628479003906,
|
|
"epoch": 1.841692789968652,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004855227137375986,
|
|
"loss": 5.7022,
|
|
"mean_token_accuracy": 0.1487029567360878,
|
|
"num_tokens": 4131301.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.992813968658448,
|
|
"epoch": 1.8456112852664577,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004854163729620275,
|
|
"loss": 5.7588,
|
|
"mean_token_accuracy": 0.15141315907239913,
|
|
"num_tokens": 4140076.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.97927041053772,
|
|
"epoch": 1.8495297805642634,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004853096561382805,
|
|
"loss": 5.6662,
|
|
"mean_token_accuracy": 0.16195056587457657,
|
|
"num_tokens": 4148728.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.96298360824585,
|
|
"epoch": 1.853448275862069,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004852025634570779,
|
|
"loss": 5.805,
|
|
"mean_token_accuracy": 0.15149845257401467,
|
|
"num_tokens": 4157520.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.992768335342407,
|
|
"epoch": 1.8573667711598745,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004850950951098116,
|
|
"loss": 5.6198,
|
|
"mean_token_accuracy": 0.1634534031152725,
|
|
"num_tokens": 4165547.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.912991189956665,
|
|
"epoch": 1.8612852664576802,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004849872512885451,
|
|
"loss": 5.7673,
|
|
"mean_token_accuracy": 0.14465269148349763,
|
|
"num_tokens": 4174656.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.001243543624878,
|
|
"epoch": 1.865203761755486,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004848790321860127,
|
|
"loss": 5.8025,
|
|
"mean_token_accuracy": 0.14944776445627211,
|
|
"num_tokens": 4183452.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.004137229919434,
|
|
"epoch": 1.8691222570532915,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048477043799561946,
|
|
"loss": 5.6939,
|
|
"mean_token_accuracy": 0.1496536500751972,
|
|
"num_tokens": 4192306.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.9638097286224365,
|
|
"epoch": 1.873040752351097,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004846614689114409,
|
|
"loss": 5.7456,
|
|
"mean_token_accuracy": 0.14964845776557922,
|
|
"num_tokens": 4201201.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.9588665008544925,
|
|
"epoch": 1.876959247648903,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004845521251282223,
|
|
"loss": 5.6617,
|
|
"mean_token_accuracy": 0.16116371303796767,
|
|
"num_tokens": 4209374.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.862710762023926,
|
|
"epoch": 1.8808777429467085,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004844424068413789,
|
|
"loss": 5.6181,
|
|
"mean_token_accuracy": 0.15533220991492272,
|
|
"num_tokens": 4217258.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.9789571285247805,
|
|
"epoch": 1.884796238244514,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048433231424699504,
|
|
"loss": 5.6504,
|
|
"mean_token_accuracy": 0.15306440740823746,
|
|
"num_tokens": 4226484.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.935569524765015,
|
|
"epoch": 1.8887147335423198,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048422184754182384,
|
|
"loss": 5.6978,
|
|
"mean_token_accuracy": 0.1518302485346794,
|
|
"num_tokens": 4235546.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.019880771636963,
|
|
"epoch": 1.8926332288401255,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004841110069232875,
|
|
"loss": 5.8364,
|
|
"mean_token_accuracy": 0.14888604432344438,
|
|
"num_tokens": 4244549.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.966967248916626,
|
|
"epoch": 1.896551724137931,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048399979258947597,
|
|
"loss": 5.7344,
|
|
"mean_token_accuracy": 0.15552246570587158,
|
|
"num_tokens": 4252918.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.968617105484009,
|
|
"epoch": 1.9004702194357366,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004838882047391474,
|
|
"loss": 5.7245,
|
|
"mean_token_accuracy": 0.15395404100418092,
|
|
"num_tokens": 4261674.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.936679077148438,
|
|
"epoch": 1.9043887147335423,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048377624357172724,
|
|
"loss": 5.6138,
|
|
"mean_token_accuracy": 0.1584752194583416,
|
|
"num_tokens": 4269986.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.952783489227295,
|
|
"epoch": 1.908307210031348,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048366390928730843,
|
|
"loss": 5.7314,
|
|
"mean_token_accuracy": 0.15672740638256072,
|
|
"num_tokens": 4278131.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.04855489730835,
|
|
"epoch": 1.9122257053291536,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004835512020866504,
|
|
"loss": 5.7181,
|
|
"mean_token_accuracy": 0.15229694694280624,
|
|
"num_tokens": 4287214.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.93603549003601,
|
|
"epoch": 1.9161442006269591,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048343812217117925,
|
|
"loss": 5.6532,
|
|
"mean_token_accuracy": 0.1576303869485855,
|
|
"num_tokens": 4295949.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.931463813781738,
|
|
"epoch": 1.9200626959247649,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048332466974298723,
|
|
"loss": 5.6531,
|
|
"mean_token_accuracy": 0.14921436607837676,
|
|
"num_tokens": 4305014.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.882351303100586,
|
|
"epoch": 1.9239811912225706,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048321084500483203,
|
|
"loss": 5.6141,
|
|
"mean_token_accuracy": 0.16436710804700852,
|
|
"num_tokens": 4313560.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.916711711883545,
|
|
"epoch": 1.9278996865203761,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000483096648160137,
|
|
"loss": 5.7265,
|
|
"mean_token_accuracy": 0.1539405345916748,
|
|
"num_tokens": 4323280.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.970920515060425,
|
|
"epoch": 1.9318181818181817,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048298207941299047,
|
|
"loss": 5.7529,
|
|
"mean_token_accuracy": 0.15606391429901123,
|
|
"num_tokens": 4331757.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.849035120010376,
|
|
"epoch": 1.9357366771159876,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048286713896814536,
|
|
"loss": 5.6865,
|
|
"mean_token_accuracy": 0.15091156587004662,
|
|
"num_tokens": 4341453.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.071984434127808,
|
|
"epoch": 1.9396551724137931,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048275182703101877,
|
|
"loss": 5.6375,
|
|
"mean_token_accuracy": 0.15821584388613702,
|
|
"num_tokens": 4349095.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.850894451141357,
|
|
"epoch": 1.9435736677115987,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048263614380769193,
|
|
"loss": 5.7301,
|
|
"mean_token_accuracy": 0.15306517779827117,
|
|
"num_tokens": 4357913.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.972369194030762,
|
|
"epoch": 1.9474921630094044,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048252008950490957,
|
|
"loss": 5.6893,
|
|
"mean_token_accuracy": 0.15192078948020935,
|
|
"num_tokens": 4366586.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.971062660217285,
|
|
"epoch": 1.9514106583072102,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048240366433007935,
|
|
"loss": 5.6544,
|
|
"mean_token_accuracy": 0.15650087743997573,
|
|
"num_tokens": 4375393.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.955308103561402,
|
|
"epoch": 1.9553291536050157,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048228686849127213,
|
|
"loss": 5.7175,
|
|
"mean_token_accuracy": 0.15588297843933105,
|
|
"num_tokens": 4383747.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.943535852432251,
|
|
"epoch": 1.9592476489028212,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004821697021972209,
|
|
"loss": 5.7034,
|
|
"mean_token_accuracy": 0.14956690073013307,
|
|
"num_tokens": 4392085.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 1.9592476489028212,
|
|
"eval_entropy": 5.805665695390036,
|
|
"eval_loss": 5.870630264282227,
|
|
"eval_mean_token_accuracy": 0.1520071476842313,
|
|
"eval_num_tokens": 4392085.0,
|
|
"eval_runtime": 2.1813,
|
|
"eval_samples_per_second": 1889.696,
|
|
"eval_steps_per_second": 236.556,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.96346082687378,
|
|
"epoch": 1.963166144200627,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 5.699,
|
|
"mean_token_accuracy": 0.15488397404551507,
|
|
"num_tokens": 4400843.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.86937952041626,
|
|
"epoch": 1.9670846394984327,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004819342590816288,
|
|
"loss": 5.6587,
|
|
"mean_token_accuracy": 0.15112571716308593,
|
|
"num_tokens": 4409126.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.978038501739502,
|
|
"epoch": 1.9710031347962382,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004818159826808631,
|
|
"loss": 5.8008,
|
|
"mean_token_accuracy": 0.14800565093755721,
|
|
"num_tokens": 4418401.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.965325927734375,
|
|
"epoch": 1.9749216300940438,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004816973366664026,
|
|
"loss": 5.7284,
|
|
"mean_token_accuracy": 0.15000912994146348,
|
|
"num_tokens": 4428141.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.003513240814209,
|
|
"epoch": 1.9788401253918495,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004815783212502871,
|
|
"loss": 5.733,
|
|
"mean_token_accuracy": 0.14810304641723632,
|
|
"num_tokens": 4437158.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.948342418670654,
|
|
"epoch": 1.9827586206896552,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048145893664521645,
|
|
"loss": 5.781,
|
|
"mean_token_accuracy": 0.1501256987452507,
|
|
"num_tokens": 4445901.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.972067403793335,
|
|
"epoch": 1.9866771159874608,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048133918306455023,
|
|
"loss": 5.6102,
|
|
"mean_token_accuracy": 0.16880455911159514,
|
|
"num_tokens": 4454200.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.919958686828613,
|
|
"epoch": 1.9905956112852663,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004812190607223075,
|
|
"loss": 5.7256,
|
|
"mean_token_accuracy": 0.14323122426867485,
|
|
"num_tokens": 4463484.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.7867495059967045,
|
|
"epoch": 1.9945141065830723,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048109856983316655,
|
|
"loss": 5.5479,
|
|
"mean_token_accuracy": 0.16151398718357085,
|
|
"num_tokens": 4472659.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.940403079986572,
|
|
"epoch": 1.9984326018808778,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000480977710612464,
|
|
"loss": 5.6318,
|
|
"mean_token_accuracy": 0.14992604926228523,
|
|
"num_tokens": 4481355.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.8997900009155275,
|
|
"epoch": 2.0023510971786833,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004808564832761948,
|
|
"loss": 5.4913,
|
|
"mean_token_accuracy": 0.16193180680274963,
|
|
"num_tokens": 4490005.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.9053887844085695,
|
|
"epoch": 2.006269592476489,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004807348880410119,
|
|
"loss": 5.3186,
|
|
"mean_token_accuracy": 0.16936820596456528,
|
|
"num_tokens": 4498645.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.80018196105957,
|
|
"epoch": 2.010188087774295,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004806129251242258,
|
|
"loss": 5.4346,
|
|
"mean_token_accuracy": 0.16318274438381195,
|
|
"num_tokens": 4509675.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.838470077514648,
|
|
"epoch": 2.0141065830721003,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048049059474380393,
|
|
"loss": 5.2671,
|
|
"mean_token_accuracy": 0.17468879967927933,
|
|
"num_tokens": 4518463.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.960326480865478,
|
|
"epoch": 2.018025078369906,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048036789711837047,
|
|
"loss": 5.4184,
|
|
"mean_token_accuracy": 0.1616513103246689,
|
|
"num_tokens": 4527120.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.9028702735900875,
|
|
"epoch": 2.0219435736677114,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048024483246720607,
|
|
"loss": 5.4385,
|
|
"mean_token_accuracy": 0.16155478805303575,
|
|
"num_tokens": 4535962.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.918149375915528,
|
|
"epoch": 2.0258620689655173,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004801214010102472,
|
|
"loss": 5.4095,
|
|
"mean_token_accuracy": 0.1586926758289337,
|
|
"num_tokens": 4545233.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.789323329925537,
|
|
"epoch": 2.029780564263323,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004799976029680858,
|
|
"loss": 5.2728,
|
|
"mean_token_accuracy": 0.1645790845155716,
|
|
"num_tokens": 4554658.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.951250314712524,
|
|
"epoch": 2.0336990595611284,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004798734385619691,
|
|
"loss": 5.4719,
|
|
"mean_token_accuracy": 0.16631337106227875,
|
|
"num_tokens": 4563964.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.810087013244629,
|
|
"epoch": 2.0376175548589344,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000479748908013799,
|
|
"loss": 5.3385,
|
|
"mean_token_accuracy": 0.17210058867931366,
|
|
"num_tokens": 4572865.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.794609022140503,
|
|
"epoch": 2.04153605015674,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004796240115461319,
|
|
"loss": 5.3025,
|
|
"mean_token_accuracy": 0.18002838492393494,
|
|
"num_tokens": 4581021.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.792965841293335,
|
|
"epoch": 2.0454545454545454,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004794987493821779,
|
|
"loss": 5.3797,
|
|
"mean_token_accuracy": 0.16931623369455337,
|
|
"num_tokens": 4590779.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.881677770614624,
|
|
"epoch": 2.049373040752351,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047937312174580084,
|
|
"loss": 5.3877,
|
|
"mean_token_accuracy": 0.16402240544557573,
|
|
"num_tokens": 4599831.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.9702167987823485,
|
|
"epoch": 2.053291536050157,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004792471288615177,
|
|
"loss": 5.5211,
|
|
"mean_token_accuracy": 0.15844076126813889,
|
|
"num_tokens": 4609264.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.872013759613037,
|
|
"epoch": 2.0572100313479624,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004791207709544981,
|
|
"loss": 5.418,
|
|
"mean_token_accuracy": 0.16842499375343323,
|
|
"num_tokens": 4617885.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.77409520149231,
|
|
"epoch": 2.061128526645768,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047899404825056424,
|
|
"loss": 5.3439,
|
|
"mean_token_accuracy": 0.16611635237932204,
|
|
"num_tokens": 4626536.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.875023317337036,
|
|
"epoch": 2.0650470219435735,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004788669609761901,
|
|
"loss": 5.4392,
|
|
"mean_token_accuracy": 0.16167420744895936,
|
|
"num_tokens": 4635563.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.799903202056885,
|
|
"epoch": 2.0689655172413794,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047873950935850107,
|
|
"loss": 5.3399,
|
|
"mean_token_accuracy": 0.1623118191957474,
|
|
"num_tokens": 4644822.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.8305216312408445,
|
|
"epoch": 2.072884012539185,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004786116936252742,
|
|
"loss": 5.3216,
|
|
"mean_token_accuracy": 0.17577696740627288,
|
|
"num_tokens": 4652709.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.768903684616089,
|
|
"epoch": 2.0768025078369905,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004784835140049367,
|
|
"loss": 5.366,
|
|
"mean_token_accuracy": 0.16856254041194915,
|
|
"num_tokens": 4662014.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.874498987197876,
|
|
"epoch": 2.080721003134796,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004783549707265663,
|
|
"loss": 5.3103,
|
|
"mean_token_accuracy": 0.16626953333616257,
|
|
"num_tokens": 4670618.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.819876432418823,
|
|
"epoch": 2.084639498432602,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047822606401989084,
|
|
"loss": 5.4749,
|
|
"mean_token_accuracy": 0.15537550970911979,
|
|
"num_tokens": 4679937.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.839358472824097,
|
|
"epoch": 2.0885579937304075,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004780967941152873,
|
|
"loss": 5.2756,
|
|
"mean_token_accuracy": 0.18055983185768126,
|
|
"num_tokens": 4688006.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.7598120212554935,
|
|
"epoch": 2.092476489028213,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004779671612437822,
|
|
"loss": 5.2398,
|
|
"mean_token_accuracy": 0.17702153474092483,
|
|
"num_tokens": 4697106.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.7331727027893065,
|
|
"epoch": 2.096394984326019,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047783716563705035,
|
|
"loss": 5.38,
|
|
"mean_token_accuracy": 0.17051040679216384,
|
|
"num_tokens": 4705582.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.806806945800782,
|
|
"epoch": 2.1003134796238245,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000477706807527415,
|
|
"loss": 5.3185,
|
|
"mean_token_accuracy": 0.17006773352622986,
|
|
"num_tokens": 4713640.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.829270362854004,
|
|
"epoch": 2.10423197492163,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004775760871478472,
|
|
"loss": 5.4248,
|
|
"mean_token_accuracy": 0.1641215831041336,
|
|
"num_tokens": 4722729.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.764045381546021,
|
|
"epoch": 2.1081504702194356,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047744500473196564,
|
|
"loss": 5.3317,
|
|
"mean_token_accuracy": 0.1733331635594368,
|
|
"num_tokens": 4732099.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.841757774353027,
|
|
"epoch": 2.1120689655172415,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047731356051403556,
|
|
"loss": 5.3511,
|
|
"mean_token_accuracy": 0.16420331299304963,
|
|
"num_tokens": 4741766.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.700101089477539,
|
|
"epoch": 2.115987460815047,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004771817547289693,
|
|
"loss": 5.2664,
|
|
"mean_token_accuracy": 0.17011009752750397,
|
|
"num_tokens": 4750890.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.785740327835083,
|
|
"epoch": 2.1199059561128526,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004770495876123251,
|
|
"loss": 5.3856,
|
|
"mean_token_accuracy": 0.1629512369632721,
|
|
"num_tokens": 4760063.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.87955117225647,
|
|
"epoch": 2.123824451410658,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004769170594003071,
|
|
"loss": 5.3793,
|
|
"mean_token_accuracy": 0.1711374670267105,
|
|
"num_tokens": 4768928.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.741111946105957,
|
|
"epoch": 2.127742946708464,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047678417032976457,
|
|
"loss": 5.3538,
|
|
"mean_token_accuracy": 0.16548465490341185,
|
|
"num_tokens": 4778129.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.798362302780151,
|
|
"epoch": 2.1316614420062696,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004766509206381919,
|
|
"loss": 5.3477,
|
|
"mean_token_accuracy": 0.1757352441549301,
|
|
"num_tokens": 4786759.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.739752388000488,
|
|
"epoch": 2.135579937304075,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004765173105637279,
|
|
"loss": 5.3475,
|
|
"mean_token_accuracy": 0.17320456057786943,
|
|
"num_tokens": 4795355.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.737670087814331,
|
|
"epoch": 2.139498432601881,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047638334034515547,
|
|
"loss": 5.3119,
|
|
"mean_token_accuracy": 0.1707861006259918,
|
|
"num_tokens": 4804030.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.65292010307312,
|
|
"epoch": 2.1434169278996866,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047624901022190106,
|
|
"loss": 5.2467,
|
|
"mean_token_accuracy": 0.1885754495859146,
|
|
"num_tokens": 4812779.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.8172399520874025,
|
|
"epoch": 2.147335423197492,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.3923,
|
|
"mean_token_accuracy": 0.16614845097064973,
|
|
"num_tokens": 4822292.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.869544982910156,
|
|
"epoch": 2.1512539184952977,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004759792712222679,
|
|
"loss": 5.4239,
|
|
"mean_token_accuracy": 0.16641001999378205,
|
|
"num_tokens": 4830524.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.875621509552002,
|
|
"epoch": 2.1551724137931036,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004758438628279565,
|
|
"loss": 5.445,
|
|
"mean_token_accuracy": 0.1666880264878273,
|
|
"num_tokens": 4839328.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.746448087692261,
|
|
"epoch": 2.159090909090909,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047570809549309697,
|
|
"loss": 5.2643,
|
|
"mean_token_accuracy": 0.17746930867433547,
|
|
"num_tokens": 4847711.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.874939346313477,
|
|
"epoch": 2.1630094043887147,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004755719694603275,
|
|
"loss": 5.4321,
|
|
"mean_token_accuracy": 0.1654679611325264,
|
|
"num_tokens": 4856706.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.808804845809936,
|
|
"epoch": 2.16692789968652,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004754354849729274,
|
|
"loss": 5.446,
|
|
"mean_token_accuracy": 0.17023264318704606,
|
|
"num_tokens": 4865250.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.751035928726196,
|
|
"epoch": 2.170846394984326,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047529864227481653,
|
|
"loss": 5.3835,
|
|
"mean_token_accuracy": 0.17073231488466262,
|
|
"num_tokens": 4873826.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.906466007232666,
|
|
"epoch": 2.1747648902821317,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000475161441610555,
|
|
"loss": 5.3431,
|
|
"mean_token_accuracy": 0.16954400539398193,
|
|
"num_tokens": 4882631.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.780795669555664,
|
|
"epoch": 2.1786833855799372,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004750238832253427,
|
|
"loss": 5.4098,
|
|
"mean_token_accuracy": 0.1619915932416916,
|
|
"num_tokens": 4891007.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.747603702545166,
|
|
"epoch": 2.1826018808777428,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004748859673650187,
|
|
"loss": 5.3034,
|
|
"mean_token_accuracy": 0.17001678347587584,
|
|
"num_tokens": 4899692.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.763918018341064,
|
|
"epoch": 2.1865203761755487,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047474769427606115,
|
|
"loss": 5.4203,
|
|
"mean_token_accuracy": 0.16626635044813157,
|
|
"num_tokens": 4909354.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.76016640663147,
|
|
"epoch": 2.1904388714733543,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004746090642055863,
|
|
"loss": 5.3697,
|
|
"mean_token_accuracy": 0.16723306030035018,
|
|
"num_tokens": 4918400.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.888517379760742,
|
|
"epoch": 2.19435736677116,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047447007740134857,
|
|
"loss": 5.4667,
|
|
"mean_token_accuracy": 0.15896394550800325,
|
|
"num_tokens": 4926852.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.767556810379029,
|
|
"epoch": 2.1982758620689653,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047433073411174,
|
|
"loss": 5.3604,
|
|
"mean_token_accuracy": 0.17230437994003295,
|
|
"num_tokens": 4935717.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.713468360900879,
|
|
"epoch": 2.2021943573667713,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004741910345857896,
|
|
"loss": 5.287,
|
|
"mean_token_accuracy": 0.1731954351067543,
|
|
"num_tokens": 4945389.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.800408792495728,
|
|
"epoch": 2.206112852664577,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047405097907316315,
|
|
"loss": 5.4231,
|
|
"mean_token_accuracy": 0.1679323822259903,
|
|
"num_tokens": 4953670.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.7536238670349125,
|
|
"epoch": 2.2100313479623823,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004739105678241625,
|
|
"loss": 5.4322,
|
|
"mean_token_accuracy": 0.17051864117383958,
|
|
"num_tokens": 4962383.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.757964658737182,
|
|
"epoch": 2.2139498432601883,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004737698010897253,
|
|
"loss": 5.2435,
|
|
"mean_token_accuracy": 0.1744084358215332,
|
|
"num_tokens": 4970672.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.8481744766235355,
|
|
"epoch": 2.217868338557994,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004736286791214245,
|
|
"loss": 5.4311,
|
|
"mean_token_accuracy": 0.16687192320823668,
|
|
"num_tokens": 4979303.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.7630088329315186,
|
|
"epoch": 2.2217868338557993,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047348720217146807,
|
|
"loss": 5.2769,
|
|
"mean_token_accuracy": 0.17239437848329545,
|
|
"num_tokens": 4988322.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.657605981826782,
|
|
"epoch": 2.225705329153605,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047334537049269806,
|
|
"loss": 5.365,
|
|
"mean_token_accuracy": 0.16851404309272766,
|
|
"num_tokens": 4997250.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.815507698059082,
|
|
"epoch": 2.229623824451411,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004732031843385909,
|
|
"loss": 5.4382,
|
|
"mean_token_accuracy": 0.16451333314180375,
|
|
"num_tokens": 5005459.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.717640113830567,
|
|
"epoch": 2.2335423197492164,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004730606439632562,
|
|
"loss": 5.1924,
|
|
"mean_token_accuracy": 0.1898626208305359,
|
|
"num_tokens": 5013704.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.673655033111572,
|
|
"epoch": 2.237460815047022,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004729177496214367,
|
|
"loss": 5.3333,
|
|
"mean_token_accuracy": 0.16645074784755706,
|
|
"num_tokens": 5022545.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.744983768463134,
|
|
"epoch": 2.2413793103448274,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047277450156850767,
|
|
"loss": 5.3177,
|
|
"mean_token_accuracy": 0.18145410269498824,
|
|
"num_tokens": 5031133.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.769635963439941,
|
|
"epoch": 2.2452978056426334,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004726309000604768,
|
|
"loss": 5.3602,
|
|
"mean_token_accuracy": 0.16541550755500795,
|
|
"num_tokens": 5039424.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.842046117782592,
|
|
"epoch": 2.249216300940439,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004724869453539832,
|
|
"loss": 5.5367,
|
|
"mean_token_accuracy": 0.1631926953792572,
|
|
"num_tokens": 5048868.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.883913040161133,
|
|
"epoch": 2.2531347962382444,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004723426377062972,
|
|
"loss": 5.3841,
|
|
"mean_token_accuracy": 0.16495195180177688,
|
|
"num_tokens": 5057532.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.7595116138458256,
|
|
"epoch": 2.2570532915360504,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047219797737532,
|
|
"loss": 5.3937,
|
|
"mean_token_accuracy": 0.16277937740087509,
|
|
"num_tokens": 5066088.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.702235174179077,
|
|
"epoch": 2.260971786833856,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047205296461958314,
|
|
"loss": 5.2839,
|
|
"mean_token_accuracy": 0.172503100335598,
|
|
"num_tokens": 5074968.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.692880344390869,
|
|
"epoch": 2.2648902821316614,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047190759969824785,
|
|
"loss": 5.3213,
|
|
"mean_token_accuracy": 0.17744118124246597,
|
|
"num_tokens": 5083605.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.721279811859131,
|
|
"epoch": 2.268808777429467,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047176188287110485,
|
|
"loss": 5.3283,
|
|
"mean_token_accuracy": 0.1750294342637062,
|
|
"num_tokens": 5092182.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.863342189788819,
|
|
"epoch": 2.2727272727272725,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004716158143985737,
|
|
"loss": 5.5002,
|
|
"mean_token_accuracy": 0.16113798022270204,
|
|
"num_tokens": 5100910.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.799469614028931,
|
|
"epoch": 2.2766457680250785,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047146939454170245,
|
|
"loss": 5.3893,
|
|
"mean_token_accuracy": 0.17243236750364305,
|
|
"num_tokens": 5109760.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.814022827148437,
|
|
"epoch": 2.280564263322884,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004713226235621672,
|
|
"loss": 5.4155,
|
|
"mean_token_accuracy": 0.16456849724054337,
|
|
"num_tokens": 5119193.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.708380794525146,
|
|
"epoch": 2.2844827586206895,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004711755017222714,
|
|
"loss": 5.3211,
|
|
"mean_token_accuracy": 0.1708238273859024,
|
|
"num_tokens": 5127480.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.841254425048828,
|
|
"epoch": 2.2884012539184955,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047102802928494563,
|
|
"loss": 5.4967,
|
|
"mean_token_accuracy": 0.1584901139140129,
|
|
"num_tokens": 5136648.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.7566770076751705,
|
|
"epoch": 2.292319749216301,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004708802065137471,
|
|
"loss": 5.2715,
|
|
"mean_token_accuracy": 0.17339732944965364,
|
|
"num_tokens": 5144708.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.719892978668213,
|
|
"epoch": 2.2962382445141065,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004707320336728591,
|
|
"loss": 5.4083,
|
|
"mean_token_accuracy": 0.1630084276199341,
|
|
"num_tokens": 5153388.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.879896545410157,
|
|
"epoch": 2.300156739811912,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004705835110270904,
|
|
"loss": 5.3862,
|
|
"mean_token_accuracy": 0.16763117760419846,
|
|
"num_tokens": 5161874.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.786136531829834,
|
|
"epoch": 2.304075235109718,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047043463884187517,
|
|
"loss": 5.3667,
|
|
"mean_token_accuracy": 0.17521363496780396,
|
|
"num_tokens": 5170284.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.668455505371094,
|
|
"epoch": 2.3079937304075235,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047028541738327207,
|
|
"loss": 5.3441,
|
|
"mean_token_accuracy": 0.16458476334810257,
|
|
"num_tokens": 5179547.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.812346982955932,
|
|
"epoch": 2.311912225705329,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004701358469179641,
|
|
"loss": 5.3614,
|
|
"mean_token_accuracy": 0.17573053240776063,
|
|
"num_tokens": 5188256.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.815900039672852,
|
|
"epoch": 2.3158307210031346,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004699859277132578,
|
|
"loss": 5.4924,
|
|
"mean_token_accuracy": 0.17042121291160583,
|
|
"num_tokens": 5197464.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.784163141250611,
|
|
"epoch": 2.3197492163009406,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046983566003708336,
|
|
"loss": 5.2944,
|
|
"mean_token_accuracy": 0.17738744169473647,
|
|
"num_tokens": 5205989.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.738129186630249,
|
|
"epoch": 2.323667711598746,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00046968504415799325,
|
|
"loss": 5.338,
|
|
"mean_token_accuracy": 0.17164268493652343,
|
|
"num_tokens": 5214490.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.717895460128784,
|
|
"epoch": 2.3275862068965516,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004695340803451625,
|
|
"loss": 5.3819,
|
|
"mean_token_accuracy": 0.16975632011890412,
|
|
"num_tokens": 5223335.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.72329626083374,
|
|
"epoch": 2.3315047021943576,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 5.3122,
|
|
"mean_token_accuracy": 0.17277760952711105,
|
|
"num_tokens": 5231665.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.698038911819458,
|
|
"epoch": 2.335423197492163,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004692311099980878,
|
|
"loss": 5.316,
|
|
"mean_token_accuracy": 0.17599530667066574,
|
|
"num_tokens": 5239810.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.676642990112304,
|
|
"epoch": 2.3393416927899686,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046907910400530097,
|
|
"loss": 5.28,
|
|
"mean_token_accuracy": 0.17479155212640762,
|
|
"num_tokens": 5247525.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.615977573394775,
|
|
"epoch": 2.343260188087774,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004689267511616868,
|
|
"loss": 5.331,
|
|
"mean_token_accuracy": 0.17096548080444335,
|
|
"num_tokens": 5255528.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.7472326278686525,
|
|
"epoch": 2.34717868338558,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046877405173952465,
|
|
"loss": 5.297,
|
|
"mean_token_accuracy": 0.17430708557367325,
|
|
"num_tokens": 5263944.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.746042490005493,
|
|
"epoch": 2.3510971786833856,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000468621006011713,
|
|
"loss": 5.3928,
|
|
"mean_token_accuracy": 0.16550849378108978,
|
|
"num_tokens": 5271724.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.3510971786833856,
|
|
"eval_entropy": 5.488131588743639,
|
|
"eval_loss": 5.81321382522583,
|
|
"eval_mean_token_accuracy": 0.15704630561490623,
|
|
"eval_num_tokens": 5271724.0,
|
|
"eval_runtime": 2.3816,
|
|
"eval_samples_per_second": 1730.758,
|
|
"eval_steps_per_second": 216.66,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"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": 1054041072261120.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|