a45fd99152
Model: fpadovani/ind-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
25310 lines
693 KiB
JSON
25310 lines
693 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 9.796238244514107,
|
|
"eval_steps": 500,
|
|
"global_step": 12500,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 5.736651849746704,
|
|
"epoch": 2.355015673981191,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046846761425176943,
|
|
"loss": 5.4188,
|
|
"mean_token_accuracy": 0.16324140429496764,
|
|
"num_tokens": 5281199.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.894945430755615,
|
|
"epoch": 2.3589341692789967,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004683138767338299,
|
|
"loss": 5.436,
|
|
"mean_token_accuracy": 0.16572253853082658,
|
|
"num_tokens": 5289782.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.635149335861206,
|
|
"epoch": 2.3628526645768027,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004681597937326483,
|
|
"loss": 5.2253,
|
|
"mean_token_accuracy": 0.17826488316059114,
|
|
"num_tokens": 5297922.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.673852062225341,
|
|
"epoch": 2.366771159874608,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004680053655235959,
|
|
"loss": 5.3765,
|
|
"mean_token_accuracy": 0.17751118242740632,
|
|
"num_tokens": 5306178.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.702480506896973,
|
|
"epoch": 2.3706896551724137,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004678505923826609,
|
|
"loss": 5.3375,
|
|
"mean_token_accuracy": 0.17454966604709626,
|
|
"num_tokens": 5314936.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.720952796936035,
|
|
"epoch": 2.3746081504702197,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046769547458644817,
|
|
"loss": 5.3837,
|
|
"mean_token_accuracy": 0.1620444968342781,
|
|
"num_tokens": 5323549.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.748093175888061,
|
|
"epoch": 2.378526645768025,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004675400124121782,
|
|
"loss": 5.3798,
|
|
"mean_token_accuracy": 0.16564854681491853,
|
|
"num_tokens": 5332310.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.853538990020752,
|
|
"epoch": 2.3824451410658307,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046738420613768716,
|
|
"loss": 5.4942,
|
|
"mean_token_accuracy": 0.16354199796915053,
|
|
"num_tokens": 5340757.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.836502456665039,
|
|
"epoch": 2.3863636363636362,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046722805604142614,
|
|
"loss": 5.499,
|
|
"mean_token_accuracy": 0.16480527967214584,
|
|
"num_tokens": 5349433.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.7534143924713135,
|
|
"epoch": 2.3902821316614418,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046707156240246076,
|
|
"loss": 5.4366,
|
|
"mean_token_accuracy": 0.1586879551410675,
|
|
"num_tokens": 5358943.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.705928659439087,
|
|
"epoch": 2.3942006269592477,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046691472550047027,
|
|
"loss": 5.3492,
|
|
"mean_token_accuracy": 0.16898986250162124,
|
|
"num_tokens": 5367782.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.687156534194946,
|
|
"epoch": 2.3981191222570533,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046675754561574783,
|
|
"loss": 5.2314,
|
|
"mean_token_accuracy": 0.1744614824652672,
|
|
"num_tokens": 5376186.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.78343448638916,
|
|
"epoch": 2.402037617554859,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046660002302919933,
|
|
"loss": 5.488,
|
|
"mean_token_accuracy": 0.16165671646595,
|
|
"num_tokens": 5384888.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.786775350570679,
|
|
"epoch": 2.4059561128526648,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004664421580223433,
|
|
"loss": 5.3746,
|
|
"mean_token_accuracy": 0.16670224368572234,
|
|
"num_tokens": 5393394.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.791707611083984,
|
|
"epoch": 2.4098746081504703,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046628395087730995,
|
|
"loss": 5.4352,
|
|
"mean_token_accuracy": 0.16912620812654494,
|
|
"num_tokens": 5402901.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.808181190490723,
|
|
"epoch": 2.413793103448276,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004661254018768411,
|
|
"loss": 5.4044,
|
|
"mean_token_accuracy": 0.1639070227742195,
|
|
"num_tokens": 5412704.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.735262537002564,
|
|
"epoch": 2.4177115987460813,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004659665113042897,
|
|
"loss": 5.3775,
|
|
"mean_token_accuracy": 0.17844438254833223,
|
|
"num_tokens": 5420863.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.680651664733887,
|
|
"epoch": 2.4216300940438873,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004658072794436187,
|
|
"loss": 5.3345,
|
|
"mean_token_accuracy": 0.17459718585014344,
|
|
"num_tokens": 5430002.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.689062404632568,
|
|
"epoch": 2.425548589341693,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046564770657940146,
|
|
"loss": 5.3258,
|
|
"mean_token_accuracy": 0.17565509527921677,
|
|
"num_tokens": 5438424.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.696020412445068,
|
|
"epoch": 2.4294670846394983,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004654877929968205,
|
|
"loss": 5.3531,
|
|
"mean_token_accuracy": 0.17165651619434358,
|
|
"num_tokens": 5446970.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.757030630111695,
|
|
"epoch": 2.433385579937304,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004653275389816673,
|
|
"loss": 5.4129,
|
|
"mean_token_accuracy": 0.16429247260093688,
|
|
"num_tokens": 5456689.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.708596324920654,
|
|
"epoch": 2.43730407523511,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046516694482034174,
|
|
"loss": 5.3152,
|
|
"mean_token_accuracy": 0.1773858845233917,
|
|
"num_tokens": 5465269.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.6959387302398685,
|
|
"epoch": 2.4412225705329154,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046500601079985164,
|
|
"loss": 5.3762,
|
|
"mean_token_accuracy": 0.17252293825149537,
|
|
"num_tokens": 5473930.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.691638326644897,
|
|
"epoch": 2.445141065830721,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004648447372078123,
|
|
"loss": 5.4051,
|
|
"mean_token_accuracy": 0.17384545952081681,
|
|
"num_tokens": 5482885.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.648982095718384,
|
|
"epoch": 2.449059561128527,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004646831243324457,
|
|
"loss": 5.336,
|
|
"mean_token_accuracy": 0.17178052812814712,
|
|
"num_tokens": 5491489.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.767613220214844,
|
|
"epoch": 2.4529780564263324,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004645211724625802,
|
|
"loss": 5.3923,
|
|
"mean_token_accuracy": 0.16743856072425842,
|
|
"num_tokens": 5499694.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.6223572254180905,
|
|
"epoch": 2.456896551724138,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046435888188765015,
|
|
"loss": 5.2927,
|
|
"mean_token_accuracy": 0.171783384680748,
|
|
"num_tokens": 5508037.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.661111211776733,
|
|
"epoch": 2.4608150470219434,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004641962528976951,
|
|
"loss": 5.2812,
|
|
"mean_token_accuracy": 0.17228170484304428,
|
|
"num_tokens": 5516644.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.728868722915649,
|
|
"epoch": 2.4647335423197494,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004640332857833593,
|
|
"loss": 5.3309,
|
|
"mean_token_accuracy": 0.17078238278627395,
|
|
"num_tokens": 5524789.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.682557106018066,
|
|
"epoch": 2.468652037617555,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046386998083589156,
|
|
"loss": 5.3641,
|
|
"mean_token_accuracy": 0.16958895027637483,
|
|
"num_tokens": 5533739.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.86026701927185,
|
|
"epoch": 2.4725705329153604,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004637063383471442,
|
|
"loss": 5.3927,
|
|
"mean_token_accuracy": 0.16778089702129365,
|
|
"num_tokens": 5542735.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.637849903106689,
|
|
"epoch": 2.476489028213166,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046354235860957287,
|
|
"loss": 5.2394,
|
|
"mean_token_accuracy": 0.17945276200771332,
|
|
"num_tokens": 5551272.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.702561759948731,
|
|
"epoch": 2.480407523510972,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004633780419162361,
|
|
"loss": 5.4263,
|
|
"mean_token_accuracy": 0.16436652541160585,
|
|
"num_tokens": 5559868.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.721883249282837,
|
|
"epoch": 2.4843260188087775,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046321338856079435,
|
|
"loss": 5.2247,
|
|
"mean_token_accuracy": 0.18002837151288986,
|
|
"num_tokens": 5568276.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.666681814193725,
|
|
"epoch": 2.488244514106583,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046304839883750987,
|
|
"loss": 5.3177,
|
|
"mean_token_accuracy": 0.1675707757472992,
|
|
"num_tokens": 5576462.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.73568959236145,
|
|
"epoch": 2.492163009404389,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000462883073041246,
|
|
"loss": 5.4543,
|
|
"mean_token_accuracy": 0.16734556555747987,
|
|
"num_tokens": 5586044.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.573016691207886,
|
|
"epoch": 2.4960815047021945,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004627174114674669,
|
|
"loss": 5.2508,
|
|
"mean_token_accuracy": 0.1739211156964302,
|
|
"num_tokens": 5594626.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.730077457427979,
|
|
"epoch": 2.5,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004625514144122365,
|
|
"loss": 5.3191,
|
|
"mean_token_accuracy": 0.1721462205052376,
|
|
"num_tokens": 5603268.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.691356801986695,
|
|
"epoch": 2.5039184952978055,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004623850821722185,
|
|
"loss": 5.2864,
|
|
"mean_token_accuracy": 0.17004774510860443,
|
|
"num_tokens": 5612053.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.696454286575317,
|
|
"epoch": 2.507836990595611,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004622184150446756,
|
|
"loss": 5.5434,
|
|
"mean_token_accuracy": 0.1607106164097786,
|
|
"num_tokens": 5621297.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.734076118469238,
|
|
"epoch": 2.511755485893417,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046205141332746904,
|
|
"loss": 5.3562,
|
|
"mean_token_accuracy": 0.16984062343835832,
|
|
"num_tokens": 5629683.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.596327447891236,
|
|
"epoch": 2.5156739811912225,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 5.2973,
|
|
"mean_token_accuracy": 0.17405743151903152,
|
|
"num_tokens": 5638707.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.774000978469848,
|
|
"epoch": 2.519592476489028,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004617164073184987,
|
|
"loss": 5.3318,
|
|
"mean_token_accuracy": 0.17241414338350297,
|
|
"num_tokens": 5648241.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.6768800735473635,
|
|
"epoch": 2.523510971786834,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046154840362544496,
|
|
"loss": 5.3876,
|
|
"mean_token_accuracy": 0.17083406299352646,
|
|
"num_tokens": 5656743.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.726757478713989,
|
|
"epoch": 2.5274294670846396,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004613800665401466,
|
|
"loss": 5.3634,
|
|
"mean_token_accuracy": 0.16949367076158522,
|
|
"num_tokens": 5666418.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.688678121566772,
|
|
"epoch": 2.531347962382445,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004612113963634493,
|
|
"loss": 5.3058,
|
|
"mean_token_accuracy": 0.17605796307325364,
|
|
"num_tokens": 5675404.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.658199167251587,
|
|
"epoch": 2.535266457680251,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004610423933967938,
|
|
"loss": 5.4049,
|
|
"mean_token_accuracy": 0.16016172915697097,
|
|
"num_tokens": 5684699.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.815964269638061,
|
|
"epoch": 2.5391849529780566,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000460873057942216,
|
|
"loss": 5.4143,
|
|
"mean_token_accuracy": 0.17390596270561218,
|
|
"num_tokens": 5693219.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.764363050460815,
|
|
"epoch": 2.543103448275862,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004607033903023458,
|
|
"loss": 5.4439,
|
|
"mean_token_accuracy": 0.16516934484243392,
|
|
"num_tokens": 5702005.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.780673027038574,
|
|
"epoch": 2.5470219435736676,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046053339078040674,
|
|
"loss": 5.4145,
|
|
"mean_token_accuracy": 0.17013225853443145,
|
|
"num_tokens": 5710715.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.7120860576629635,
|
|
"epoch": 2.550940438871473,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004603630596802155,
|
|
"loss": 5.336,
|
|
"mean_token_accuracy": 0.1701664909720421,
|
|
"num_tokens": 5719444.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.632794141769409,
|
|
"epoch": 2.554858934169279,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004601923973061814,
|
|
"loss": 5.3167,
|
|
"mean_token_accuracy": 0.16517282277345657,
|
|
"num_tokens": 5727755.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.641457271575928,
|
|
"epoch": 2.5587774294670846,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046002140396330575,
|
|
"loss": 5.2531,
|
|
"mean_token_accuracy": 0.1755443274974823,
|
|
"num_tokens": 5736933.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.754851961135865,
|
|
"epoch": 2.56269592476489,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045985007995718154,
|
|
"loss": 5.4353,
|
|
"mean_token_accuracy": 0.16590821295976638,
|
|
"num_tokens": 5745831.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.8150616645812985,
|
|
"epoch": 2.566614420062696,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004596784255939923,
|
|
"loss": 5.4516,
|
|
"mean_token_accuracy": 0.16931454241275787,
|
|
"num_tokens": 5755419.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.68149242401123,
|
|
"epoch": 2.5705329153605017,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004595064411805123,
|
|
"loss": 5.3475,
|
|
"mean_token_accuracy": 0.17512361854314804,
|
|
"num_tokens": 5764008.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.654454326629638,
|
|
"epoch": 2.574451410658307,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004593341270241057,
|
|
"loss": 5.2565,
|
|
"mean_token_accuracy": 0.17512470930814744,
|
|
"num_tokens": 5772129.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.665771198272705,
|
|
"epoch": 2.5783699059561127,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004591614834327257,
|
|
"loss": 5.3916,
|
|
"mean_token_accuracy": 0.16697368621826172,
|
|
"num_tokens": 5781079.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.67975869178772,
|
|
"epoch": 2.5822884012539182,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045898851071491444,
|
|
"loss": 5.342,
|
|
"mean_token_accuracy": 0.17753895968198777,
|
|
"num_tokens": 5790444.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.629207563400269,
|
|
"epoch": 2.586206896551724,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004588152091798022,
|
|
"loss": 5.2409,
|
|
"mean_token_accuracy": 0.17479657381772995,
|
|
"num_tokens": 5799828.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.7281801223754885,
|
|
"epoch": 2.5901253918495297,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004586415791371069,
|
|
"loss": 5.5011,
|
|
"mean_token_accuracy": 0.1652237668633461,
|
|
"num_tokens": 5808765.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.803612184524536,
|
|
"epoch": 2.5940438871473352,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004584676208971336,
|
|
"loss": 5.2824,
|
|
"mean_token_accuracy": 0.17455053329467773,
|
|
"num_tokens": 5817281.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.61289701461792,
|
|
"epoch": 2.597962382445141,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045829333477077384,
|
|
"loss": 5.3182,
|
|
"mean_token_accuracy": 0.18395667821168898,
|
|
"num_tokens": 5825741.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.693694019317627,
|
|
"epoch": 2.6018808777429467,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004581187210695053,
|
|
"loss": 5.2915,
|
|
"mean_token_accuracy": 0.17659811675548553,
|
|
"num_tokens": 5834466.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.68879280090332,
|
|
"epoch": 2.6057993730407523,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000457943780105391,
|
|
"loss": 5.32,
|
|
"mean_token_accuracy": 0.1786850169301033,
|
|
"num_tokens": 5843463.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.696980142593384,
|
|
"epoch": 2.6097178683385582,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045776851219107856,
|
|
"loss": 5.3723,
|
|
"mean_token_accuracy": 0.1670466512441635,
|
|
"num_tokens": 5851885.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.7379930973052975,
|
|
"epoch": 2.6136363636363638,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045759291763980035,
|
|
"loss": 5.3459,
|
|
"mean_token_accuracy": 0.1704084724187851,
|
|
"num_tokens": 5861537.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.640843915939331,
|
|
"epoch": 2.6175548589341693,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045741699676537227,
|
|
"loss": 5.2948,
|
|
"mean_token_accuracy": 0.17625776827335357,
|
|
"num_tokens": 5870332.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.715352487564087,
|
|
"epoch": 2.621473354231975,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045724074988219343,
|
|
"loss": 5.4961,
|
|
"mean_token_accuracy": 0.16728264689445496,
|
|
"num_tokens": 5879240.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.6437522888183596,
|
|
"epoch": 2.6253918495297803,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045706417730524565,
|
|
"loss": 5.2414,
|
|
"mean_token_accuracy": 0.16607876867055893,
|
|
"num_tokens": 5887961.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.732382488250733,
|
|
"epoch": 2.6293103448275863,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004568872793500927,
|
|
"loss": 5.4782,
|
|
"mean_token_accuracy": 0.1713300123810768,
|
|
"num_tokens": 5896551.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.635508012771607,
|
|
"epoch": 2.633228840125392,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045671005633287986,
|
|
"loss": 5.2631,
|
|
"mean_token_accuracy": 0.1750209480524063,
|
|
"num_tokens": 5905485.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.677169132232666,
|
|
"epoch": 2.6371473354231973,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004565325085703336,
|
|
"loss": 5.2874,
|
|
"mean_token_accuracy": 0.17128181755542754,
|
|
"num_tokens": 5915166.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.7076586246490475,
|
|
"epoch": 2.6410658307210033,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004563546363797602,
|
|
"loss": 5.3448,
|
|
"mean_token_accuracy": 0.1749894455075264,
|
|
"num_tokens": 5923782.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.715188837051391,
|
|
"epoch": 2.644984326018809,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004561764400790465,
|
|
"loss": 5.4491,
|
|
"mean_token_accuracy": 0.16602206975221634,
|
|
"num_tokens": 5931904.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.74715781211853,
|
|
"epoch": 2.6489028213166144,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045599791998665796,
|
|
"loss": 5.3225,
|
|
"mean_token_accuracy": 0.17717247903347016,
|
|
"num_tokens": 5940067.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.661538982391358,
|
|
"epoch": 2.6528213166144203,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004558190764216389,
|
|
"loss": 5.3408,
|
|
"mean_token_accuracy": 0.1758809953927994,
|
|
"num_tokens": 5948663.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.7242940902709964,
|
|
"epoch": 2.656739811912226,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004556399097036119,
|
|
"loss": 5.3341,
|
|
"mean_token_accuracy": 0.17185231298208237,
|
|
"num_tokens": 5957224.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.708767366409302,
|
|
"epoch": 2.6606583072100314,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004554604201527768,
|
|
"loss": 5.4166,
|
|
"mean_token_accuracy": 0.17407266944646835,
|
|
"num_tokens": 5965604.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.679881572723389,
|
|
"epoch": 2.664576802507837,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045528060808991075,
|
|
"loss": 5.2427,
|
|
"mean_token_accuracy": 0.17856677621603012,
|
|
"num_tokens": 5974309.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.649680519104004,
|
|
"epoch": 2.6684952978056424,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004551004738363669,
|
|
"loss": 5.2692,
|
|
"mean_token_accuracy": 0.17150685638189317,
|
|
"num_tokens": 5982315.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.652850008010864,
|
|
"epoch": 2.6724137931034484,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045492001771407434,
|
|
"loss": 5.2929,
|
|
"mean_token_accuracy": 0.17711860537528992,
|
|
"num_tokens": 5990515.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.710380125045776,
|
|
"epoch": 2.676332288401254,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004547392400455374,
|
|
"loss": 5.4646,
|
|
"mean_token_accuracy": 0.17207272499799728,
|
|
"num_tokens": 6000431.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.721198511123657,
|
|
"epoch": 2.6802507836990594,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004545581411538353,
|
|
"loss": 5.3917,
|
|
"mean_token_accuracy": 0.16679134219884872,
|
|
"num_tokens": 6009022.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.735800123214721,
|
|
"epoch": 2.6841692789968654,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045437672136262083,
|
|
"loss": 5.3171,
|
|
"mean_token_accuracy": 0.17115751206874846,
|
|
"num_tokens": 6019330.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.634147357940674,
|
|
"epoch": 2.688087774294671,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004541949809961208,
|
|
"loss": 5.3136,
|
|
"mean_token_accuracy": 0.18226997256278993,
|
|
"num_tokens": 6027492.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.676221752166748,
|
|
"epoch": 2.6920062695924765,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004540129203791346,
|
|
"loss": 5.3442,
|
|
"mean_token_accuracy": 0.17710426896810533,
|
|
"num_tokens": 6036467.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.673708438873291,
|
|
"epoch": 2.695924764890282,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045383053983703413,
|
|
"loss": 5.3123,
|
|
"mean_token_accuracy": 0.17591846138238906,
|
|
"num_tokens": 6045711.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.65407657623291,
|
|
"epoch": 2.6998432601880875,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 5.3815,
|
|
"mean_token_accuracy": 0.16701181679964067,
|
|
"num_tokens": 6054626.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.714919137954712,
|
|
"epoch": 2.7037617554858935,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045346482028183583,
|
|
"loss": 5.448,
|
|
"mean_token_accuracy": 0.16598840355873107,
|
|
"num_tokens": 6064118.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.7021928310394285,
|
|
"epoch": 2.707680250783699,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045328148192233823,
|
|
"loss": 5.3471,
|
|
"mean_token_accuracy": 0.17301076650619507,
|
|
"num_tokens": 6072723.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.681285285949707,
|
|
"epoch": 2.7115987460815045,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004530978249449254,
|
|
"loss": 5.3193,
|
|
"mean_token_accuracy": 0.18009441643953322,
|
|
"num_tokens": 6081535.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.643833446502685,
|
|
"epoch": 2.7155172413793105,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004529138496778222,
|
|
"loss": 5.2798,
|
|
"mean_token_accuracy": 0.1761292666196823,
|
|
"num_tokens": 6090524.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.702294874191284,
|
|
"epoch": 2.719435736677116,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004527295564498222,
|
|
"loss": 5.3689,
|
|
"mean_token_accuracy": 0.17294523268938064,
|
|
"num_tokens": 6099870.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.660062074661255,
|
|
"epoch": 2.7233542319749215,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004525449455902874,
|
|
"loss": 5.338,
|
|
"mean_token_accuracy": 0.1709714263677597,
|
|
"num_tokens": 6107760.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.695686197280883,
|
|
"epoch": 2.7272727272727275,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004523600174291473,
|
|
"loss": 5.405,
|
|
"mean_token_accuracy": 0.16780283749103547,
|
|
"num_tokens": 6117135.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.632399368286133,
|
|
"epoch": 2.731191222570533,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004521747722968985,
|
|
"loss": 5.3327,
|
|
"mean_token_accuracy": 0.1724711462855339,
|
|
"num_tokens": 6125711.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.735014247894287,
|
|
"epoch": 2.7351097178683386,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045198921052460396,
|
|
"loss": 5.4046,
|
|
"mean_token_accuracy": 0.1653449222445488,
|
|
"num_tokens": 6134623.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.754837465286255,
|
|
"epoch": 2.739028213166144,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004518033324438928,
|
|
"loss": 5.3953,
|
|
"mean_token_accuracy": 0.16823590993881227,
|
|
"num_tokens": 6143228.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.6582499027252195,
|
|
"epoch": 2.7429467084639496,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004516171383869593,
|
|
"loss": 5.3736,
|
|
"mean_token_accuracy": 0.1680469587445259,
|
|
"num_tokens": 6152021.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.7429467084639496,
|
|
"eval_entropy": 5.440021375353022,
|
|
"eval_loss": 5.747522830963135,
|
|
"eval_mean_token_accuracy": 0.16063085169524186,
|
|
"eval_num_tokens": 6152021.0,
|
|
"eval_runtime": 2.1775,
|
|
"eval_samples_per_second": 1893.035,
|
|
"eval_steps_per_second": 236.974,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.722462368011475,
|
|
"epoch": 2.7468652037617556,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045143062868656234,
|
|
"loss": 5.4426,
|
|
"mean_token_accuracy": 0.16724137514829635,
|
|
"num_tokens": 6161389.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.7765895366668705,
|
|
"epoch": 2.750783699059561,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000451243803676025,
|
|
"loss": 5.342,
|
|
"mean_token_accuracy": 0.16516051292419434,
|
|
"num_tokens": 6170381.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.63956241607666,
|
|
"epoch": 2.7547021943573666,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045105666368923397,
|
|
"loss": 5.3055,
|
|
"mean_token_accuracy": 0.17531027495861054,
|
|
"num_tokens": 6179071.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.589923334121704,
|
|
"epoch": 2.7586206896551726,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045086920906063866,
|
|
"loss": 5.3258,
|
|
"mean_token_accuracy": 0.16932518184185028,
|
|
"num_tokens": 6187896.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.6329902648925785,
|
|
"epoch": 2.762539184952978,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045068144012525095,
|
|
"loss": 5.2987,
|
|
"mean_token_accuracy": 0.17633456587791443,
|
|
"num_tokens": 6196328.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.628019046783447,
|
|
"epoch": 2.7664576802507836,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045049335721864426,
|
|
"loss": 5.3316,
|
|
"mean_token_accuracy": 0.16708063334226608,
|
|
"num_tokens": 6204703.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.639265012741089,
|
|
"epoch": 2.7703761755485896,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045030496067695336,
|
|
"loss": 5.3084,
|
|
"mean_token_accuracy": 0.1749113157391548,
|
|
"num_tokens": 6213496.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.658385276794434,
|
|
"epoch": 2.774294670846395,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004501162508368733,
|
|
"loss": 5.2791,
|
|
"mean_token_accuracy": 0.1773057609796524,
|
|
"num_tokens": 6222112.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.688526678085327,
|
|
"epoch": 2.7782131661442007,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004499272280356594,
|
|
"loss": 5.4162,
|
|
"mean_token_accuracy": 0.16061867475509645,
|
|
"num_tokens": 6230288.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.61279525756836,
|
|
"epoch": 2.782131661442006,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004497378926111257,
|
|
"loss": 5.2496,
|
|
"mean_token_accuracy": 0.18118438869714737,
|
|
"num_tokens": 6238608.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.71974925994873,
|
|
"epoch": 2.7860501567398117,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004495482449016456,
|
|
"loss": 5.3788,
|
|
"mean_token_accuracy": 0.17336056828498841,
|
|
"num_tokens": 6247902.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.6707190990448,
|
|
"epoch": 2.7899686520376177,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004493582852461501,
|
|
"loss": 5.3634,
|
|
"mean_token_accuracy": 0.17027460932731628,
|
|
"num_tokens": 6257433.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.598862743377685,
|
|
"epoch": 2.793887147335423,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004491680139841281,
|
|
"loss": 5.3235,
|
|
"mean_token_accuracy": 0.17070114314556123,
|
|
"num_tokens": 6266017.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.721092367172242,
|
|
"epoch": 2.7978056426332287,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000448977431455625,
|
|
"loss": 5.4331,
|
|
"mean_token_accuracy": 0.16319793164730073,
|
|
"num_tokens": 6274759.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.6764976978302,
|
|
"epoch": 2.8017241379310347,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044878653800124285,
|
|
"loss": 5.286,
|
|
"mean_token_accuracy": 0.17585835605859756,
|
|
"num_tokens": 6283699.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.620488548278809,
|
|
"epoch": 2.80564263322884,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004485953339621391,
|
|
"loss": 5.3229,
|
|
"mean_token_accuracy": 0.1738375037908554,
|
|
"num_tokens": 6292562.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.646529817581177,
|
|
"epoch": 2.8095611285266457,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004484038196800265,
|
|
"loss": 5.2896,
|
|
"mean_token_accuracy": 0.17799898087978364,
|
|
"num_tokens": 6300928.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.584728002548218,
|
|
"epoch": 2.8134796238244513,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004482119954971719,
|
|
"loss": 5.2258,
|
|
"mean_token_accuracy": 0.17900719195604325,
|
|
"num_tokens": 6309994.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.705359268188476,
|
|
"epoch": 2.817398119122257,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044801986175639635,
|
|
"loss": 5.3232,
|
|
"mean_token_accuracy": 0.1721549540758133,
|
|
"num_tokens": 6318856.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.656028461456299,
|
|
"epoch": 2.8213166144200628,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004478274188010741,
|
|
"loss": 5.3399,
|
|
"mean_token_accuracy": 0.1725846365094185,
|
|
"num_tokens": 6327274.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.690082025527954,
|
|
"epoch": 2.8252351097178683,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044763466697513173,
|
|
"loss": 5.3985,
|
|
"mean_token_accuracy": 0.17907976508140563,
|
|
"num_tokens": 6335957.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.716768789291382,
|
|
"epoch": 2.829153605015674,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044744160662304805,
|
|
"loss": 5.4175,
|
|
"mean_token_accuracy": 0.17787378430366516,
|
|
"num_tokens": 6344673.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.593239974975586,
|
|
"epoch": 2.83307210031348,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044724823808985325,
|
|
"loss": 5.2631,
|
|
"mean_token_accuracy": 0.1803012177348137,
|
|
"num_tokens": 6353085.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.642696762084961,
|
|
"epoch": 2.8369905956112853,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004470545617211283,
|
|
"loss": 5.3149,
|
|
"mean_token_accuracy": 0.17097059041261672,
|
|
"num_tokens": 6362107.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.653160238265992,
|
|
"epoch": 2.840909090909091,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044686057786300423,
|
|
"loss": 5.3098,
|
|
"mean_token_accuracy": 0.17488847821950912,
|
|
"num_tokens": 6370381.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.674486112594605,
|
|
"epoch": 2.844827586206897,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044666628686216154,
|
|
"loss": 5.3437,
|
|
"mean_token_accuracy": 0.1708572417497635,
|
|
"num_tokens": 6378790.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.595465230941772,
|
|
"epoch": 2.8487460815047023,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044647168906583,
|
|
"loss": 5.3222,
|
|
"mean_token_accuracy": 0.17378978729248046,
|
|
"num_tokens": 6387997.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.710697078704834,
|
|
"epoch": 2.852664576802508,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044627678482178716,
|
|
"loss": 5.3254,
|
|
"mean_token_accuracy": 0.1760479360818863,
|
|
"num_tokens": 6397458.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.686402606964111,
|
|
"epoch": 2.8565830721003134,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004460815744783587,
|
|
"loss": 5.2839,
|
|
"mean_token_accuracy": 0.17581177055835723,
|
|
"num_tokens": 6405481.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.6757673740386965,
|
|
"epoch": 2.860501567398119,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000445886058384417,
|
|
"loss": 5.4029,
|
|
"mean_token_accuracy": 0.16874466836452484,
|
|
"num_tokens": 6414087.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.661119413375855,
|
|
"epoch": 2.864420062695925,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004456902368893811,
|
|
"loss": 5.223,
|
|
"mean_token_accuracy": 0.181645867228508,
|
|
"num_tokens": 6421934.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.649143218994141,
|
|
"epoch": 2.8683385579937304,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004454941103432158,
|
|
"loss": 5.2542,
|
|
"mean_token_accuracy": 0.17862497866153718,
|
|
"num_tokens": 6431028.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.533873605728149,
|
|
"epoch": 2.872257053291536,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044529767909643093,
|
|
"loss": 5.2073,
|
|
"mean_token_accuracy": 0.18256115466356276,
|
|
"num_tokens": 6439913.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.636791181564331,
|
|
"epoch": 2.876175548589342,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004451009435000811,
|
|
"loss": 5.291,
|
|
"mean_token_accuracy": 0.17250144928693772,
|
|
"num_tokens": 6449161.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.687928628921509,
|
|
"epoch": 2.8800940438871474,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004449039039057647,
|
|
"loss": 5.4166,
|
|
"mean_token_accuracy": 0.16819276958703994,
|
|
"num_tokens": 6457834.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.581802749633789,
|
|
"epoch": 2.884012539184953,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 5.2247,
|
|
"mean_token_accuracy": 0.18246169090270997,
|
|
"num_tokens": 6465879.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.6004173278808596,
|
|
"epoch": 2.887931034482759,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004445089141323415,
|
|
"loss": 5.2556,
|
|
"mean_token_accuracy": 0.17483851313591003,
|
|
"num_tokens": 6474801.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.6311845779418945,
|
|
"epoch": 2.8918495297805644,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00044431096465914554,
|
|
"loss": 5.2864,
|
|
"mean_token_accuracy": 0.17921681851148605,
|
|
"num_tokens": 6483532.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.620904111862183,
|
|
"epoch": 2.89576802507837,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044411271259980315,
|
|
"loss": 5.2574,
|
|
"mean_token_accuracy": 0.17421473413705826,
|
|
"num_tokens": 6491372.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.586532068252564,
|
|
"epoch": 2.8996865203761755,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004439141583086231,
|
|
"loss": 5.2676,
|
|
"mean_token_accuracy": 0.17178864032030106,
|
|
"num_tokens": 6500001.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.681151580810547,
|
|
"epoch": 2.903605015673981,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044371530214045395,
|
|
"loss": 5.3878,
|
|
"mean_token_accuracy": 0.17248643040657044,
|
|
"num_tokens": 6507750.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.55568265914917,
|
|
"epoch": 2.907523510971787,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044351614445068413,
|
|
"loss": 5.2497,
|
|
"mean_token_accuracy": 0.17340636253356934,
|
|
"num_tokens": 6517888.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.699483728408813,
|
|
"epoch": 2.9114420062695925,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044331668559524043,
|
|
"loss": 5.4032,
|
|
"mean_token_accuracy": 0.16709693372249604,
|
|
"num_tokens": 6526573.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.609065389633178,
|
|
"epoch": 2.915360501567398,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004431169259305883,
|
|
"loss": 5.3074,
|
|
"mean_token_accuracy": 0.17757185250520707,
|
|
"num_tokens": 6536544.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.679953479766846,
|
|
"epoch": 2.919278996865204,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004429168658137306,
|
|
"loss": 5.3074,
|
|
"mean_token_accuracy": 0.1698578879237175,
|
|
"num_tokens": 6544982.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.626433563232422,
|
|
"epoch": 2.9231974921630095,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044271650560220746,
|
|
"loss": 5.3479,
|
|
"mean_token_accuracy": 0.17322934567928314,
|
|
"num_tokens": 6553664.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.702156734466553,
|
|
"epoch": 2.927115987460815,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044251584565409464,
|
|
"loss": 5.3361,
|
|
"mean_token_accuracy": 0.17482462525367737,
|
|
"num_tokens": 6562234.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.553665113449097,
|
|
"epoch": 2.9310344827586206,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004423148863280044,
|
|
"loss": 5.2901,
|
|
"mean_token_accuracy": 0.1720922291278839,
|
|
"num_tokens": 6570707.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.660065126419068,
|
|
"epoch": 2.934952978056426,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044211362798308334,
|
|
"loss": 5.3358,
|
|
"mean_token_accuracy": 0.17049673646688462,
|
|
"num_tokens": 6580054.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.653960752487182,
|
|
"epoch": 2.938871473354232,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004419120709790129,
|
|
"loss": 5.3787,
|
|
"mean_token_accuracy": 0.17072909623384475,
|
|
"num_tokens": 6588585.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.7185839176177975,
|
|
"epoch": 2.9427899686520376,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044171021567600814,
|
|
"loss": 5.3943,
|
|
"mean_token_accuracy": 0.17066112011671067,
|
|
"num_tokens": 6597290.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.68023567199707,
|
|
"epoch": 2.946708463949843,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044150806243481715,
|
|
"loss": 5.3007,
|
|
"mean_token_accuracy": 0.16959173679351808,
|
|
"num_tokens": 6604966.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.65665636062622,
|
|
"epoch": 2.950626959247649,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004413056116167206,
|
|
"loss": 5.2825,
|
|
"mean_token_accuracy": 0.18088154792785643,
|
|
"num_tokens": 6613588.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.749258708953858,
|
|
"epoch": 2.9545454545454546,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004411028635835309,
|
|
"loss": 5.5125,
|
|
"mean_token_accuracy": 0.1558822587132454,
|
|
"num_tokens": 6622907.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.657208824157715,
|
|
"epoch": 2.95846394984326,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004408998186975917,
|
|
"loss": 5.2985,
|
|
"mean_token_accuracy": 0.17126751989126204,
|
|
"num_tokens": 6632074.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.694626665115356,
|
|
"epoch": 2.962382445141066,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044069647732177696,
|
|
"loss": 5.3112,
|
|
"mean_token_accuracy": 0.17472130060195923,
|
|
"num_tokens": 6640595.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.736867189407349,
|
|
"epoch": 2.9663009404388716,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044049283981949103,
|
|
"loss": 5.3948,
|
|
"mean_token_accuracy": 0.1632532760500908,
|
|
"num_tokens": 6649861.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.70715742111206,
|
|
"epoch": 2.970219435736677,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004402889065546667,
|
|
"loss": 5.4079,
|
|
"mean_token_accuracy": 0.16917193681001663,
|
|
"num_tokens": 6659157.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.660294151306152,
|
|
"epoch": 2.9741379310344827,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044008467789176625,
|
|
"loss": 5.3772,
|
|
"mean_token_accuracy": 0.16952175050973892,
|
|
"num_tokens": 6667979.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.686374378204346,
|
|
"epoch": 2.978056426332288,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004398801541957791,
|
|
"loss": 5.3243,
|
|
"mean_token_accuracy": 0.16979904025793074,
|
|
"num_tokens": 6677239.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.6725075244903564,
|
|
"epoch": 2.981974921630094,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004396753358322223,
|
|
"loss": 5.4237,
|
|
"mean_token_accuracy": 0.16390596181154252,
|
|
"num_tokens": 6687317.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.6351197242736815,
|
|
"epoch": 2.9858934169278997,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004394702231671396,
|
|
"loss": 5.3127,
|
|
"mean_token_accuracy": 0.17069294154644013,
|
|
"num_tokens": 6695732.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.655317068099976,
|
|
"epoch": 2.989811912225705,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004392648165671004,
|
|
"loss": 5.2374,
|
|
"mean_token_accuracy": 0.18597303479909896,
|
|
"num_tokens": 6704272.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.603386545181275,
|
|
"epoch": 2.993730407523511,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004390591163991998,
|
|
"loss": 5.2873,
|
|
"mean_token_accuracy": 0.17730152308940889,
|
|
"num_tokens": 6713355.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.67601900100708,
|
|
"epoch": 2.9976489028213167,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043885312303105725,
|
|
"loss": 5.3685,
|
|
"mean_token_accuracy": 0.16839122921228408,
|
|
"num_tokens": 6721952.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.595555019378662,
|
|
"epoch": 3.001567398119122,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004386468368308163,
|
|
"loss": 5.1853,
|
|
"mean_token_accuracy": 0.17674031257629394,
|
|
"num_tokens": 6730582.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.66121654510498,
|
|
"epoch": 3.0054858934169277,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004384402581671438,
|
|
"loss": 4.9981,
|
|
"mean_token_accuracy": 0.18754991441965102,
|
|
"num_tokens": 6739370.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.6243397235870365,
|
|
"epoch": 3.0094043887147337,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004382333874092295,
|
|
"loss": 4.9652,
|
|
"mean_token_accuracy": 0.195566987991333,
|
|
"num_tokens": 6748282.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.702383422851563,
|
|
"epoch": 3.0133228840125392,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043802622492678466,
|
|
"loss": 5.0951,
|
|
"mean_token_accuracy": 0.18754026293754578,
|
|
"num_tokens": 6757622.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.6851646900177,
|
|
"epoch": 3.0172413793103448,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004378187710900426,
|
|
"loss": 5.0311,
|
|
"mean_token_accuracy": 0.19040577560663224,
|
|
"num_tokens": 6765780.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.578414344787598,
|
|
"epoch": 3.0211598746081503,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043761102626975674,
|
|
"loss": 5.0192,
|
|
"mean_token_accuracy": 0.19095737785100936,
|
|
"num_tokens": 6774343.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.611579513549804,
|
|
"epoch": 3.0250783699059562,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004374029908372007,
|
|
"loss": 5.0398,
|
|
"mean_token_accuracy": 0.19317415207624436,
|
|
"num_tokens": 6783765.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.707476329803467,
|
|
"epoch": 3.0289968652037618,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043719466516416774,
|
|
"loss": 5.0773,
|
|
"mean_token_accuracy": 0.18749720901250838,
|
|
"num_tokens": 6792102.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.66371717453003,
|
|
"epoch": 3.0329153605015673,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043698604962296946,
|
|
"loss": 5.0246,
|
|
"mean_token_accuracy": 0.18686578869819642,
|
|
"num_tokens": 6801435.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.581167364120484,
|
|
"epoch": 3.0368338557993733,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043677714458643566,
|
|
"loss": 5.034,
|
|
"mean_token_accuracy": 0.1861950248479843,
|
|
"num_tokens": 6810409.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.5150903224945065,
|
|
"epoch": 3.040752351097179,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043656795042791357,
|
|
"loss": 4.9093,
|
|
"mean_token_accuracy": 0.2005583241581917,
|
|
"num_tokens": 6818752.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.591932678222657,
|
|
"epoch": 3.0446708463949843,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004363584675212671,
|
|
"loss": 5.0297,
|
|
"mean_token_accuracy": 0.1878923162817955,
|
|
"num_tokens": 6827880.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.5609776973724365,
|
|
"epoch": 3.04858934169279,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004361486962408761,
|
|
"loss": 4.9939,
|
|
"mean_token_accuracy": 0.1979147180914879,
|
|
"num_tokens": 6835865.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.5401147365570065,
|
|
"epoch": 3.052507836990596,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004359386369616359,
|
|
"loss": 4.9777,
|
|
"mean_token_accuracy": 0.19878013879060746,
|
|
"num_tokens": 6843851.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.523969888687134,
|
|
"epoch": 3.0564263322884013,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004357282900589565,
|
|
"loss": 5.0129,
|
|
"mean_token_accuracy": 0.19207025170326233,
|
|
"num_tokens": 6852649.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.557503938674927,
|
|
"epoch": 3.060344827586207,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043551765590876183,
|
|
"loss": 4.9643,
|
|
"mean_token_accuracy": 0.19102779477834703,
|
|
"num_tokens": 6861607.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.536406183242798,
|
|
"epoch": 3.0642633228840124,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004353067348874894,
|
|
"loss": 5.0059,
|
|
"mean_token_accuracy": 0.1851136177778244,
|
|
"num_tokens": 6870650.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.630461740493774,
|
|
"epoch": 3.0681818181818183,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 5.0409,
|
|
"mean_token_accuracy": 0.18416744023561477,
|
|
"num_tokens": 6879820.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.484796476364136,
|
|
"epoch": 3.072100313479624,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004348840337400233,
|
|
"loss": 4.961,
|
|
"mean_token_accuracy": 0.19722549468278885,
|
|
"num_tokens": 6888329.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.502891159057617,
|
|
"epoch": 3.0760188087774294,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043467225436926517,
|
|
"loss": 5.0541,
|
|
"mean_token_accuracy": 0.18794333338737487,
|
|
"num_tokens": 6897635.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.621821498870849,
|
|
"epoch": 3.079937304075235,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004344601896382988,
|
|
"loss": 5.0854,
|
|
"mean_token_accuracy": 0.19702089875936507,
|
|
"num_tokens": 6907275.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.629112243652344,
|
|
"epoch": 3.083855799373041,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043424783992611837,
|
|
"loss": 5.0195,
|
|
"mean_token_accuracy": 0.19181722700595855,
|
|
"num_tokens": 6916377.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.5756913185119625,
|
|
"epoch": 3.0877742946708464,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043403520561222705,
|
|
"loss": 4.9277,
|
|
"mean_token_accuracy": 0.20171520709991456,
|
|
"num_tokens": 6924355.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.553646564483643,
|
|
"epoch": 3.091692789968652,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004338222870766371,
|
|
"loss": 5.046,
|
|
"mean_token_accuracy": 0.19301398545503617,
|
|
"num_tokens": 6933328.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.5955853939056395,
|
|
"epoch": 3.0956112852664575,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043360908469986827,
|
|
"loss": 5.0152,
|
|
"mean_token_accuracy": 0.19658142775297166,
|
|
"num_tokens": 6942189.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.566510963439941,
|
|
"epoch": 3.0995297805642634,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004333955988629478,
|
|
"loss": 4.9948,
|
|
"mean_token_accuracy": 0.19510561972856522,
|
|
"num_tokens": 6951188.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.497058868408203,
|
|
"epoch": 3.103448275862069,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043318182994740945,
|
|
"loss": 4.9456,
|
|
"mean_token_accuracy": 0.20327225476503372,
|
|
"num_tokens": 6958718.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.570499515533447,
|
|
"epoch": 3.1073667711598745,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004329677783352931,
|
|
"loss": 5.0095,
|
|
"mean_token_accuracy": 0.19015664011240005,
|
|
"num_tokens": 6968040.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.61021146774292,
|
|
"epoch": 3.1112852664576804,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004327534444091436,
|
|
"loss": 5.0262,
|
|
"mean_token_accuracy": 0.1878402978181839,
|
|
"num_tokens": 6976977.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.607437610626221,
|
|
"epoch": 3.115203761755486,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043253882855201037,
|
|
"loss": 5.0099,
|
|
"mean_token_accuracy": 0.19021547883749007,
|
|
"num_tokens": 6985859.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.418795251846314,
|
|
"epoch": 3.1191222570532915,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043232393114744683,
|
|
"loss": 4.9207,
|
|
"mean_token_accuracy": 0.20244280397891998,
|
|
"num_tokens": 6994031.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.430121660232544,
|
|
"epoch": 3.123040752351097,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004321087525795095,
|
|
"loss": 4.986,
|
|
"mean_token_accuracy": 0.1924207851290703,
|
|
"num_tokens": 7002527.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.596579837799072,
|
|
"epoch": 3.126959247648903,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004318932932327573,
|
|
"loss": 5.091,
|
|
"mean_token_accuracy": 0.19778767377138137,
|
|
"num_tokens": 7011983.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.604700946807862,
|
|
"epoch": 3.1308777429467085,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000431677553492251,
|
|
"loss": 5.0433,
|
|
"mean_token_accuracy": 0.19013074934482574,
|
|
"num_tokens": 7020819.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.486745882034302,
|
|
"epoch": 3.134796238244514,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043146153374355256,
|
|
"loss": 4.9116,
|
|
"mean_token_accuracy": 0.19927784204483032,
|
|
"num_tokens": 7029257.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.134796238244514,
|
|
"eval_entropy": 5.350714496863905,
|
|
"eval_loss": 5.718877792358398,
|
|
"eval_mean_token_accuracy": 0.16415763132331906,
|
|
"eval_num_tokens": 7029257.0,
|
|
"eval_runtime": 2.359,
|
|
"eval_samples_per_second": 1747.321,
|
|
"eval_steps_per_second": 218.733,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.549456644058227,
|
|
"epoch": 3.1387147335423196,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043124523437272427,
|
|
"loss": 5.0974,
|
|
"mean_token_accuracy": 0.18084846287965775,
|
|
"num_tokens": 7038942.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.568442392349243,
|
|
"epoch": 3.1426332288401255,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004310286557663282,
|
|
"loss": 5.0531,
|
|
"mean_token_accuracy": 0.19351963549852372,
|
|
"num_tokens": 7048074.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.5967950344085695,
|
|
"epoch": 3.146551724137931,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004308117983114254,
|
|
"loss": 4.9814,
|
|
"mean_token_accuracy": 0.19615940153598785,
|
|
"num_tokens": 7056887.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.556157970428467,
|
|
"epoch": 3.1504702194357366,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004305946623955754,
|
|
"loss": 4.955,
|
|
"mean_token_accuracy": 0.20416038930416108,
|
|
"num_tokens": 7065462.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.542744302749634,
|
|
"epoch": 3.1543887147335425,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043037724840683516,
|
|
"loss": 5.0814,
|
|
"mean_token_accuracy": 0.19113944321870804,
|
|
"num_tokens": 7074188.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.571135091781616,
|
|
"epoch": 3.158307210031348,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00043015955673375876,
|
|
"loss": 5.0213,
|
|
"mean_token_accuracy": 0.19301380217075348,
|
|
"num_tokens": 7082282.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.597964191436768,
|
|
"epoch": 3.1622257053291536,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004299415877653966,
|
|
"loss": 5.085,
|
|
"mean_token_accuracy": 0.18788256347179413,
|
|
"num_tokens": 7090631.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.563665771484375,
|
|
"epoch": 3.166144200626959,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004297233418912945,
|
|
"loss": 5.0196,
|
|
"mean_token_accuracy": 0.1908117040991783,
|
|
"num_tokens": 7099546.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.5490429401397705,
|
|
"epoch": 3.170062695924765,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004295048195014932,
|
|
"loss": 5.0729,
|
|
"mean_token_accuracy": 0.18990501761436462,
|
|
"num_tokens": 7109423.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.488223028182984,
|
|
"epoch": 3.1739811912225706,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004292860209865277,
|
|
"loss": 4.9274,
|
|
"mean_token_accuracy": 0.19697259366512299,
|
|
"num_tokens": 7117717.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.566748905181885,
|
|
"epoch": 3.177899686520376,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004290669467374263,
|
|
"loss": 5.1538,
|
|
"mean_token_accuracy": 0.17652396857738495,
|
|
"num_tokens": 7126573.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.521072483062744,
|
|
"epoch": 3.1818181818181817,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042884759714571037,
|
|
"loss": 5.0407,
|
|
"mean_token_accuracy": 0.19642205387353898,
|
|
"num_tokens": 7134471.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.486797571182251,
|
|
"epoch": 3.1857366771159876,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004286279726033932,
|
|
"loss": 4.9861,
|
|
"mean_token_accuracy": 0.19216951876878738,
|
|
"num_tokens": 7143092.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.622600746154785,
|
|
"epoch": 3.189655172413793,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042840807350297933,
|
|
"loss": 5.0591,
|
|
"mean_token_accuracy": 0.19006451964378357,
|
|
"num_tokens": 7152137.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.566015911102295,
|
|
"epoch": 3.1935736677115987,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00042818790023746407,
|
|
"loss": 5.0049,
|
|
"mean_token_accuracy": 0.19681376367807388,
|
|
"num_tokens": 7160037.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.487920665740967,
|
|
"epoch": 3.197492163009404,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042796745320033296,
|
|
"loss": 5.0114,
|
|
"mean_token_accuracy": 0.1900061398744583,
|
|
"num_tokens": 7168354.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.5314219951629635,
|
|
"epoch": 3.20141065830721,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042774673278556043,
|
|
"loss": 5.0697,
|
|
"mean_token_accuracy": 0.18728643506765366,
|
|
"num_tokens": 7176757.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.584316682815552,
|
|
"epoch": 3.2053291536050157,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004275257393876097,
|
|
"loss": 5.0508,
|
|
"mean_token_accuracy": 0.18692827671766282,
|
|
"num_tokens": 7185489.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.602584028244019,
|
|
"epoch": 3.209247648902821,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004273044734014318,
|
|
"loss": 4.9917,
|
|
"mean_token_accuracy": 0.19311882108449935,
|
|
"num_tokens": 7194431.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.447078132629395,
|
|
"epoch": 3.2131661442006267,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042708293522246486,
|
|
"loss": 4.9658,
|
|
"mean_token_accuracy": 0.19534049779176713,
|
|
"num_tokens": 7203074.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.510924673080444,
|
|
"epoch": 3.2170846394984327,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004268611252466337,
|
|
"loss": 5.0097,
|
|
"mean_token_accuracy": 0.19532931745052337,
|
|
"num_tokens": 7211692.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.50434250831604,
|
|
"epoch": 3.2210031347962382,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004266390438703486,
|
|
"loss": 5.005,
|
|
"mean_token_accuracy": 0.1958840921521187,
|
|
"num_tokens": 7220285.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.615520048141479,
|
|
"epoch": 3.2249216300940438,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042641669149050493,
|
|
"loss": 5.1593,
|
|
"mean_token_accuracy": 0.1813105270266533,
|
|
"num_tokens": 7230007.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.556879758834839,
|
|
"epoch": 3.2288401253918497,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004261940685044825,
|
|
"loss": 5.001,
|
|
"mean_token_accuracy": 0.19469877183437348,
|
|
"num_tokens": 7238520.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.4928804397583,
|
|
"epoch": 3.2327586206896552,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042597117531014474,
|
|
"loss": 4.936,
|
|
"mean_token_accuracy": 0.19961730837821962,
|
|
"num_tokens": 7247228.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.579296588897705,
|
|
"epoch": 3.2366771159874608,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004257480123058378,
|
|
"loss": 5.177,
|
|
"mean_token_accuracy": 0.1848059132695198,
|
|
"num_tokens": 7255782.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.613631963729858,
|
|
"epoch": 3.2405956112852663,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042552457989039036,
|
|
"loss": 5.0642,
|
|
"mean_token_accuracy": 0.17927221357822418,
|
|
"num_tokens": 7264946.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.466220235824585,
|
|
"epoch": 3.2445141065830723,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042530087846311213,
|
|
"loss": 4.8972,
|
|
"mean_token_accuracy": 0.2030108392238617,
|
|
"num_tokens": 7273647.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.550145149230957,
|
|
"epoch": 3.248432601880878,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042507690842379396,
|
|
"loss": 5.0308,
|
|
"mean_token_accuracy": 0.1903608649969101,
|
|
"num_tokens": 7283393.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.603197813034058,
|
|
"epoch": 3.2523510971786833,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 5.111,
|
|
"mean_token_accuracy": 0.19307461977005005,
|
|
"num_tokens": 7292368.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.46763277053833,
|
|
"epoch": 3.256269592476489,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042462816411060025,
|
|
"loss": 5.0624,
|
|
"mean_token_accuracy": 0.19133276790380477,
|
|
"num_tokens": 7301505.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.521873140335083,
|
|
"epoch": 3.260188087774295,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004244033906387035,
|
|
"loss": 5.0078,
|
|
"mean_token_accuracy": 0.19893682152032852,
|
|
"num_tokens": 7310011.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.571926927566528,
|
|
"epoch": 3.2641065830721003,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004241783501587231,
|
|
"loss": 5.0461,
|
|
"mean_token_accuracy": 0.19427722543478013,
|
|
"num_tokens": 7319295.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.559621906280517,
|
|
"epoch": 3.268025078369906,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042395304307284284,
|
|
"loss": 5.0692,
|
|
"mean_token_accuracy": 0.18853669911623,
|
|
"num_tokens": 7328390.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.62295355796814,
|
|
"epoch": 3.271943573667712,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004237274697837229,
|
|
"loss": 5.093,
|
|
"mean_token_accuracy": 0.18418903201818465,
|
|
"num_tokens": 7337062.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.529978466033936,
|
|
"epoch": 3.2758620689655173,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004235016306944996,
|
|
"loss": 5.0896,
|
|
"mean_token_accuracy": 0.18724962025880815,
|
|
"num_tokens": 7346364.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.557655334472656,
|
|
"epoch": 3.279780564263323,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004232755262087837,
|
|
"loss": 5.0133,
|
|
"mean_token_accuracy": 0.18999945968389512,
|
|
"num_tokens": 7354896.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.534111404418946,
|
|
"epoch": 3.2836990595611284,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042304915673066083,
|
|
"loss": 5.0661,
|
|
"mean_token_accuracy": 0.1936148852109909,
|
|
"num_tokens": 7363591.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.551254653930664,
|
|
"epoch": 3.287617554858934,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042282252266468985,
|
|
"loss": 5.0644,
|
|
"mean_token_accuracy": 0.18730547279119492,
|
|
"num_tokens": 7372891.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.5315625190734865,
|
|
"epoch": 3.29153605015674,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004225956244159025,
|
|
"loss": 4.9891,
|
|
"mean_token_accuracy": 0.19170905500650406,
|
|
"num_tokens": 7381945.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.639817476272583,
|
|
"epoch": 3.2954545454545454,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004223684623898029,
|
|
"loss": 5.1382,
|
|
"mean_token_accuracy": 0.18904215693473816,
|
|
"num_tokens": 7391449.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.520969581604004,
|
|
"epoch": 3.299373040752351,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004221410369923662,
|
|
"loss": 5.0089,
|
|
"mean_token_accuracy": 0.1959460958838463,
|
|
"num_tokens": 7399762.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.395225477218628,
|
|
"epoch": 3.303291536050157,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042191334863003873,
|
|
"loss": 4.9003,
|
|
"mean_token_accuracy": 0.20033734291791916,
|
|
"num_tokens": 7408711.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.456879329681397,
|
|
"epoch": 3.3072100313479624,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004216853977097363,
|
|
"loss": 4.995,
|
|
"mean_token_accuracy": 0.1994081810116768,
|
|
"num_tokens": 7416868.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.544852018356323,
|
|
"epoch": 3.311128526645768,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004214571846388442,
|
|
"loss": 5.1058,
|
|
"mean_token_accuracy": 0.18360537439584732,
|
|
"num_tokens": 7425687.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.54548487663269,
|
|
"epoch": 3.3150470219435735,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004212287098252164,
|
|
"loss": 4.9955,
|
|
"mean_token_accuracy": 0.19616052657365798,
|
|
"num_tokens": 7433714.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.525017976760864,
|
|
"epoch": 3.3189655172413794,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004209999736771742,
|
|
"loss": 5.0744,
|
|
"mean_token_accuracy": 0.19639303982257844,
|
|
"num_tokens": 7442765.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.503507089614868,
|
|
"epoch": 3.322884012539185,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004207709766035063,
|
|
"loss": 5.0387,
|
|
"mean_token_accuracy": 0.18881263136863707,
|
|
"num_tokens": 7450672.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.544525766372681,
|
|
"epoch": 3.3268025078369905,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004205417190134674,
|
|
"loss": 5.089,
|
|
"mean_token_accuracy": 0.18707315921783446,
|
|
"num_tokens": 7458758.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.548713493347168,
|
|
"epoch": 3.330721003134796,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004203122013167783,
|
|
"loss": 5.0506,
|
|
"mean_token_accuracy": 0.18856921195983886,
|
|
"num_tokens": 7467527.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.554612302780152,
|
|
"epoch": 3.334639498432602,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042008242392362413,
|
|
"loss": 5.0603,
|
|
"mean_token_accuracy": 0.19310801327228547,
|
|
"num_tokens": 7476659.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.474660778045655,
|
|
"epoch": 3.3385579937304075,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041985238724465433,
|
|
"loss": 4.9768,
|
|
"mean_token_accuracy": 0.19619787931442262,
|
|
"num_tokens": 7485813.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.556534481048584,
|
|
"epoch": 3.342476489028213,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041962209169098193,
|
|
"loss": 5.093,
|
|
"mean_token_accuracy": 0.18657441586256027,
|
|
"num_tokens": 7494842.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.584800672531128,
|
|
"epoch": 3.346394984326019,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004193915376741823,
|
|
"loss": 5.0944,
|
|
"mean_token_accuracy": 0.18750968277454377,
|
|
"num_tokens": 7503633.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.4995039939880375,
|
|
"epoch": 3.3503134796238245,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004191607256062928,
|
|
"loss": 5.0326,
|
|
"mean_token_accuracy": 0.1916683256626129,
|
|
"num_tokens": 7512759.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.544700765609742,
|
|
"epoch": 3.35423197492163,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004189296558998121,
|
|
"loss": 5.1243,
|
|
"mean_token_accuracy": 0.18945586830377578,
|
|
"num_tokens": 7522284.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.574970531463623,
|
|
"epoch": 3.3581504702194356,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004186983289676992,
|
|
"loss": 5.0315,
|
|
"mean_token_accuracy": 0.19002804905176163,
|
|
"num_tokens": 7531286.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.532793951034546,
|
|
"epoch": 3.3620689655172415,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041846674522337296,
|
|
"loss": 5.0344,
|
|
"mean_token_accuracy": 0.18470337092876435,
|
|
"num_tokens": 7540450.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.487664794921875,
|
|
"epoch": 3.365987460815047,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041823490508071076,
|
|
"loss": 4.9474,
|
|
"mean_token_accuracy": 0.1925487607717514,
|
|
"num_tokens": 7549386.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.574211311340332,
|
|
"epoch": 3.3699059561128526,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000418002808954049,
|
|
"loss": 5.0422,
|
|
"mean_token_accuracy": 0.1928360015153885,
|
|
"num_tokens": 7558236.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.438048219680786,
|
|
"epoch": 3.373824451410658,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00041777045725818057,
|
|
"loss": 4.9194,
|
|
"mean_token_accuracy": 0.1949421986937523,
|
|
"num_tokens": 7566984.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.473016738891602,
|
|
"epoch": 3.377742946708464,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000417537850408356,
|
|
"loss": 5.0533,
|
|
"mean_token_accuracy": 0.18688634485006334,
|
|
"num_tokens": 7576018.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.518659877777099,
|
|
"epoch": 3.3816614420062696,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004173049888202814,
|
|
"loss": 5.0489,
|
|
"mean_token_accuracy": 0.1891861766576767,
|
|
"num_tokens": 7585294.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.503489255905151,
|
|
"epoch": 3.385579937304075,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004170718729101179,
|
|
"loss": 5.0124,
|
|
"mean_token_accuracy": 0.19261398017406464,
|
|
"num_tokens": 7594634.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.508076286315918,
|
|
"epoch": 3.389498432601881,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041683850309448187,
|
|
"loss": 5.0756,
|
|
"mean_token_accuracy": 0.1867249697446823,
|
|
"num_tokens": 7603153.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.551697874069214,
|
|
"epoch": 3.3934169278996866,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041660487979044264,
|
|
"loss": 5.0034,
|
|
"mean_token_accuracy": 0.19567209482192993,
|
|
"num_tokens": 7612027.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.551417064666748,
|
|
"epoch": 3.397335423197492,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004163710034155231,
|
|
"loss": 5.0202,
|
|
"mean_token_accuracy": 0.19298054426908492,
|
|
"num_tokens": 7620725.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.455377578735352,
|
|
"epoch": 3.4012539184952977,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004161368743876982,
|
|
"loss": 4.9983,
|
|
"mean_token_accuracy": 0.19591687619686127,
|
|
"num_tokens": 7628847.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.522912931442261,
|
|
"epoch": 3.405172413793103,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004159024931253945,
|
|
"loss": 5.0922,
|
|
"mean_token_accuracy": 0.18806445002555847,
|
|
"num_tokens": 7637668.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.594728803634643,
|
|
"epoch": 3.409090909090909,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041566786004748943,
|
|
"loss": 5.1105,
|
|
"mean_token_accuracy": 0.1892433598637581,
|
|
"num_tokens": 7646545.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.6068017959594725,
|
|
"epoch": 3.4130094043887147,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00041543297557331015,
|
|
"loss": 5.1158,
|
|
"mean_token_accuracy": 0.17994534224271774,
|
|
"num_tokens": 7654967.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.470070838928223,
|
|
"epoch": 3.41692789968652,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004151978401226335,
|
|
"loss": 5.0178,
|
|
"mean_token_accuracy": 0.19313666522502898,
|
|
"num_tokens": 7663718.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.543897819519043,
|
|
"epoch": 3.420846394984326,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00041496245411568435,
|
|
"loss": 5.0747,
|
|
"mean_token_accuracy": 0.1927154093980789,
|
|
"num_tokens": 7672430.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.531405067443847,
|
|
"epoch": 3.4247648902821317,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004147268179731359,
|
|
"loss": 5.041,
|
|
"mean_token_accuracy": 0.18746682852506638,
|
|
"num_tokens": 7681270.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.541948699951172,
|
|
"epoch": 3.4286833855799372,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041449093211610815,
|
|
"loss": 5.0392,
|
|
"mean_token_accuracy": 0.18938508331775666,
|
|
"num_tokens": 7690124.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.587382841110229,
|
|
"epoch": 3.4326018808777428,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041425479696616734,
|
|
"loss": 5.0805,
|
|
"mean_token_accuracy": 0.1868514224886894,
|
|
"num_tokens": 7699511.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.502129364013672,
|
|
"epoch": 3.4365203761755487,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 5.0365,
|
|
"mean_token_accuracy": 0.18970204293727874,
|
|
"num_tokens": 7708796.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.538536024093628,
|
|
"epoch": 3.4404388714733543,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00041378178047603845,
|
|
"loss": 5.0294,
|
|
"mean_token_accuracy": 0.19471313655376435,
|
|
"num_tokens": 7717446.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.550191736221313,
|
|
"epoch": 3.44435736677116,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004135448999812074,
|
|
"loss": 5.0737,
|
|
"mean_token_accuracy": 0.18927277326583863,
|
|
"num_tokens": 7726814.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.484743690490722,
|
|
"epoch": 3.4482758620689653,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004133077718841763,
|
|
"loss": 4.9882,
|
|
"mean_token_accuracy": 0.19958024322986603,
|
|
"num_tokens": 7735895.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.488801670074463,
|
|
"epoch": 3.4521943573667713,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041307039660873113,
|
|
"loss": 5.0561,
|
|
"mean_token_accuracy": 0.18797597736120225,
|
|
"num_tokens": 7744160.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.509891843795776,
|
|
"epoch": 3.456112852664577,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004128327745791002,
|
|
"loss": 5.1183,
|
|
"mean_token_accuracy": 0.18300240039825438,
|
|
"num_tokens": 7752844.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.5605815887451175,
|
|
"epoch": 3.4600313479623823,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004125949062199526,
|
|
"loss": 5.086,
|
|
"mean_token_accuracy": 0.18447083234786987,
|
|
"num_tokens": 7762327.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.520887470245361,
|
|
"epoch": 3.4639498432601883,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041235679195639766,
|
|
"loss": 5.0088,
|
|
"mean_token_accuracy": 0.191231606900692,
|
|
"num_tokens": 7770813.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.448841905593872,
|
|
"epoch": 3.467868338557994,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041211843221398406,
|
|
"loss": 5.0363,
|
|
"mean_token_accuracy": 0.18735968470573425,
|
|
"num_tokens": 7779357.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.5421031475067135,
|
|
"epoch": 3.4717868338557993,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004118798274186994,
|
|
"loss": 5.088,
|
|
"mean_token_accuracy": 0.18958608955144882,
|
|
"num_tokens": 7787302.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.511342144012451,
|
|
"epoch": 3.475705329153605,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004116409779969691,
|
|
"loss": 5.0878,
|
|
"mean_token_accuracy": 0.18645029366016388,
|
|
"num_tokens": 7795651.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.514791011810303,
|
|
"epoch": 3.479623824451411,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00041140188437565586,
|
|
"loss": 5.051,
|
|
"mean_token_accuracy": 0.19533909261226653,
|
|
"num_tokens": 7804644.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.606357097625732,
|
|
"epoch": 3.4835423197492164,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041116254698205873,
|
|
"loss": 5.0416,
|
|
"mean_token_accuracy": 0.19350068718194963,
|
|
"num_tokens": 7812479.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.4855146408081055,
|
|
"epoch": 3.487460815047022,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041092296624391244,
|
|
"loss": 5.0371,
|
|
"mean_token_accuracy": 0.1881586492061615,
|
|
"num_tokens": 7820678.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.506537675857544,
|
|
"epoch": 3.4913793103448274,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004106831425893865,
|
|
"loss": 5.0679,
|
|
"mean_token_accuracy": 0.19468591809272767,
|
|
"num_tokens": 7829063.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.556604957580566,
|
|
"epoch": 3.4952978056426334,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004104430764470849,
|
|
"loss": 5.0929,
|
|
"mean_token_accuracy": 0.18826000690460204,
|
|
"num_tokens": 7837274.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.541276693344116,
|
|
"epoch": 3.499216300940439,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004102027682460445,
|
|
"loss": 5.0063,
|
|
"mean_token_accuracy": 0.20001240223646163,
|
|
"num_tokens": 7845379.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.555472564697266,
|
|
"epoch": 3.5031347962382444,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004099622184157353,
|
|
"loss": 5.0385,
|
|
"mean_token_accuracy": 0.18937000930309295,
|
|
"num_tokens": 7853760.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.610234069824219,
|
|
"epoch": 3.5070532915360504,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004097214273860586,
|
|
"loss": 5.1151,
|
|
"mean_token_accuracy": 0.1844977393746376,
|
|
"num_tokens": 7862026.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.509080076217652,
|
|
"epoch": 3.510971786833856,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004094803955873471,
|
|
"loss": 5.0597,
|
|
"mean_token_accuracy": 0.19973956942558288,
|
|
"num_tokens": 7870369.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.539573812484742,
|
|
"epoch": 3.5148902821316614,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004092391234503638,
|
|
"loss": 5.055,
|
|
"mean_token_accuracy": 0.19329393804073333,
|
|
"num_tokens": 7878524.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.546415328979492,
|
|
"epoch": 3.518808777429467,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00040899761140630094,
|
|
"loss": 5.1077,
|
|
"mean_token_accuracy": 0.18202579021453857,
|
|
"num_tokens": 7887369.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.5834942817687985,
|
|
"epoch": 3.5227272727272725,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040875585988677985,
|
|
"loss": 5.0931,
|
|
"mean_token_accuracy": 0.17954394966363907,
|
|
"num_tokens": 7896112.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.499890851974487,
|
|
"epoch": 3.5266457680250785,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004085138693238497,
|
|
"loss": 4.9349,
|
|
"mean_token_accuracy": 0.19304562658071517,
|
|
"num_tokens": 7904580.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.5266457680250785,
|
|
"eval_entropy": 5.323873269465542,
|
|
"eval_loss": 5.686410903930664,
|
|
"eval_mean_token_accuracy": 0.16612362855857657,
|
|
"eval_num_tokens": 7904580.0,
|
|
"eval_runtime": 2.1782,
|
|
"eval_samples_per_second": 1892.416,
|
|
"eval_steps_per_second": 236.896,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.558507633209229,
|
|
"epoch": 3.530564263322884,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004082716401499867,
|
|
"loss": 5.0799,
|
|
"mean_token_accuracy": 0.18950843960046768,
|
|
"num_tokens": 7913755.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.524209976196289,
|
|
"epoch": 3.5344827586206895,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00040802917279809383,
|
|
"loss": 5.0544,
|
|
"mean_token_accuracy": 0.19526356160640718,
|
|
"num_tokens": 7922385.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.502474451065064,
|
|
"epoch": 3.5384012539184955,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040778646770149953,
|
|
"loss": 5.0978,
|
|
"mean_token_accuracy": 0.18355227559804915,
|
|
"num_tokens": 7931634.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.504829216003418,
|
|
"epoch": 3.542319749216301,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040754352529395716,
|
|
"loss": 4.9921,
|
|
"mean_token_accuracy": 0.1861146092414856,
|
|
"num_tokens": 7939866.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.552862501144409,
|
|
"epoch": 3.5462382445141065,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040730034600964415,
|
|
"loss": 5.11,
|
|
"mean_token_accuracy": 0.17732558399438858,
|
|
"num_tokens": 7948961.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.385088968276977,
|
|
"epoch": 3.5501567398119125,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004070569302831613,
|
|
"loss": 4.95,
|
|
"mean_token_accuracy": 0.19414807260036468,
|
|
"num_tokens": 7958526.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.519954013824463,
|
|
"epoch": 3.554075235109718,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000406813278549532,
|
|
"loss": 5.0665,
|
|
"mean_token_accuracy": 0.1948120802640915,
|
|
"num_tokens": 7966887.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.539158153533935,
|
|
"epoch": 3.5579937304075235,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00040656939124420144,
|
|
"loss": 5.0481,
|
|
"mean_token_accuracy": 0.18843865394592285,
|
|
"num_tokens": 7974929.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.4964714527130125,
|
|
"epoch": 3.561912225705329,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004063252688030358,
|
|
"loss": 5.0297,
|
|
"mean_token_accuracy": 0.1928529590368271,
|
|
"num_tokens": 7983821.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.492104482650757,
|
|
"epoch": 3.5658307210031346,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004060809116623213,
|
|
"loss": 5.0289,
|
|
"mean_token_accuracy": 0.18860959857702256,
|
|
"num_tokens": 7992109.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.484443712234497,
|
|
"epoch": 3.5697492163009406,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000405836320258764,
|
|
"loss": 5.0271,
|
|
"mean_token_accuracy": 0.2013839825987816,
|
|
"num_tokens": 8001098.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.574487113952637,
|
|
"epoch": 3.573667711598746,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004055914950294882,
|
|
"loss": 5.0454,
|
|
"mean_token_accuracy": 0.19076309055089952,
|
|
"num_tokens": 8009559.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.474752426147461,
|
|
"epoch": 3.5775862068965516,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040534643641203645,
|
|
"loss": 4.9758,
|
|
"mean_token_accuracy": 0.19323819428682326,
|
|
"num_tokens": 8018021.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.453449058532715,
|
|
"epoch": 3.5815047021943576,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004051011448443681,
|
|
"loss": 4.9941,
|
|
"mean_token_accuracy": 0.19774473160505296,
|
|
"num_tokens": 8026258.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.54755048751831,
|
|
"epoch": 3.585423197492163,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000404855620764859,
|
|
"loss": 5.0457,
|
|
"mean_token_accuracy": 0.19153162091970444,
|
|
"num_tokens": 8036761.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.569961261749268,
|
|
"epoch": 3.5893416927899686,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004046098646123006,
|
|
"loss": 5.0939,
|
|
"mean_token_accuracy": 0.19241693764925002,
|
|
"num_tokens": 8045115.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.590027713775635,
|
|
"epoch": 3.593260188087774,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040436387682589876,
|
|
"loss": 5.1048,
|
|
"mean_token_accuracy": 0.18897724002599717,
|
|
"num_tokens": 8054334.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.452038431167603,
|
|
"epoch": 3.5971786833855797,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004041176578452737,
|
|
"loss": 5.0383,
|
|
"mean_token_accuracy": 0.19179239124059677,
|
|
"num_tokens": 8063165.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.49503264427185,
|
|
"epoch": 3.6010971786833856,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004038712081104587,
|
|
"loss": 5.0307,
|
|
"mean_token_accuracy": 0.19137630760669708,
|
|
"num_tokens": 8072498.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.5195229053497314,
|
|
"epoch": 3.605015673981191,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040362452806189927,
|
|
"loss": 5.0642,
|
|
"mean_token_accuracy": 0.1904445022344589,
|
|
"num_tokens": 8081455.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.452272033691406,
|
|
"epoch": 3.6089341692789967,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004033776181404527,
|
|
"loss": 5.062,
|
|
"mean_token_accuracy": 0.18546648621559142,
|
|
"num_tokens": 8090992.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.556864595413208,
|
|
"epoch": 3.6128526645768027,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040313047878738704,
|
|
"loss": 5.0548,
|
|
"mean_token_accuracy": 0.18403392136096955,
|
|
"num_tokens": 8099976.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.509908437728882,
|
|
"epoch": 3.616771159874608,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004028831104443805,
|
|
"loss": 4.9765,
|
|
"mean_token_accuracy": 0.20540964901447295,
|
|
"num_tokens": 8108275.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.473766756057739,
|
|
"epoch": 3.6206896551724137,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 5.0607,
|
|
"mean_token_accuracy": 0.1897738516330719,
|
|
"num_tokens": 8117441.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.467050218582154,
|
|
"epoch": 3.6246081504702197,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00040238768855730214,
|
|
"loss": 5.0253,
|
|
"mean_token_accuracy": 0.18740377128124236,
|
|
"num_tokens": 8125709.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.486978435516358,
|
|
"epoch": 3.628526645768025,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040213963589862963,
|
|
"loss": 5.0313,
|
|
"mean_token_accuracy": 0.18891195505857467,
|
|
"num_tokens": 8134418.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.5441827297210695,
|
|
"epoch": 3.6324451410658307,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004018913560208131,
|
|
"loss": 5.1141,
|
|
"mean_token_accuracy": 0.1830949753522873,
|
|
"num_tokens": 8143729.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.479403400421143,
|
|
"epoch": 3.6363636363636362,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004016428493675689,
|
|
"loss": 4.9798,
|
|
"mean_token_accuracy": 0.19233753234148027,
|
|
"num_tokens": 8152613.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.487750768661499,
|
|
"epoch": 3.6402821316614418,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004013941163830187,
|
|
"loss": 4.9705,
|
|
"mean_token_accuracy": 0.19886485785245894,
|
|
"num_tokens": 8160862.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.472871160507202,
|
|
"epoch": 3.6442006269592477,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004011451575116887,
|
|
"loss": 5.0681,
|
|
"mean_token_accuracy": 0.1884167015552521,
|
|
"num_tokens": 8169369.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.473298168182373,
|
|
"epoch": 3.6481191222570533,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004008959731985087,
|
|
"loss": 4.9918,
|
|
"mean_token_accuracy": 0.19036366790533066,
|
|
"num_tokens": 8178152.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.474658203125,
|
|
"epoch": 3.652037617554859,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040064656388881157,
|
|
"loss": 4.9687,
|
|
"mean_token_accuracy": 0.19902666062116622,
|
|
"num_tokens": 8186690.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.439276123046875,
|
|
"epoch": 3.6559561128526648,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004003969300283321,
|
|
"loss": 4.9552,
|
|
"mean_token_accuracy": 0.19882189184427262,
|
|
"num_tokens": 8194976.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.4815356731414795,
|
|
"epoch": 3.6598746081504703,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040014707206320653,
|
|
"loss": 5.0378,
|
|
"mean_token_accuracy": 0.1879180282354355,
|
|
"num_tokens": 8202964.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.460087728500366,
|
|
"epoch": 3.663793103448276,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039989699043997153,
|
|
"loss": 5.0531,
|
|
"mean_token_accuracy": 0.18660122007131577,
|
|
"num_tokens": 8213084.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.4686112880706785,
|
|
"epoch": 3.6677115987460818,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00039964668560556356,
|
|
"loss": 4.9666,
|
|
"mean_token_accuracy": 0.2007548078894615,
|
|
"num_tokens": 8221645.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.462784671783448,
|
|
"epoch": 3.6716300940438873,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00039939615800731784,
|
|
"loss": 4.9489,
|
|
"mean_token_accuracy": 0.20153995901346206,
|
|
"num_tokens": 8230195.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.459259939193726,
|
|
"epoch": 3.675548589341693,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039914540809296795,
|
|
"loss": 5.0478,
|
|
"mean_token_accuracy": 0.195230108499527,
|
|
"num_tokens": 8239608.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.563386392593384,
|
|
"epoch": 3.6794670846394983,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003988944363106445,
|
|
"loss": 5.1004,
|
|
"mean_token_accuracy": 0.1923746258020401,
|
|
"num_tokens": 8247993.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.501354217529297,
|
|
"epoch": 3.683385579937304,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003986432431088749,
|
|
"loss": 5.0396,
|
|
"mean_token_accuracy": 0.18640178292989731,
|
|
"num_tokens": 8257279.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.524037933349609,
|
|
"epoch": 3.68730407523511,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000398391828936582,
|
|
"loss": 5.1074,
|
|
"mean_token_accuracy": 0.18100274056196214,
|
|
"num_tokens": 8265166.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.423895835876465,
|
|
"epoch": 3.6912225705329154,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003981401942430838,
|
|
"loss": 4.8869,
|
|
"mean_token_accuracy": 0.20160461217164993,
|
|
"num_tokens": 8273281.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.495573329925537,
|
|
"epoch": 3.695141065830721,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039788833947809217,
|
|
"loss": 5.0807,
|
|
"mean_token_accuracy": 0.1840187892317772,
|
|
"num_tokens": 8282432.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.416339969635009,
|
|
"epoch": 3.699059561128527,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003976362650917125,
|
|
"loss": 4.9121,
|
|
"mean_token_accuracy": 0.20159220993518828,
|
|
"num_tokens": 8290909.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.455284309387207,
|
|
"epoch": 3.7029780564263324,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00039738397153444264,
|
|
"loss": 5.0425,
|
|
"mean_token_accuracy": 0.195650614798069,
|
|
"num_tokens": 8299581.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.502392673492432,
|
|
"epoch": 3.706896551724138,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003971314592571719,
|
|
"loss": 5.0386,
|
|
"mean_token_accuracy": 0.18765258491039277,
|
|
"num_tokens": 8308184.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.498023271560669,
|
|
"epoch": 3.7108150470219434,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003968787287111809,
|
|
"loss": 5.0372,
|
|
"mean_token_accuracy": 0.19046480804681779,
|
|
"num_tokens": 8317054.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.464916038513183,
|
|
"epoch": 3.714733542319749,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039662578034814,
|
|
"loss": 5.0523,
|
|
"mean_token_accuracy": 0.1935029685497284,
|
|
"num_tokens": 8325961.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.601960372924805,
|
|
"epoch": 3.718652037617555,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039637261462010886,
|
|
"loss": 5.1267,
|
|
"mean_token_accuracy": 0.18886515200138093,
|
|
"num_tokens": 8334401.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.5873006820678714,
|
|
"epoch": 3.7225705329153604,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003961192319795358,
|
|
"loss": 5.1082,
|
|
"mean_token_accuracy": 0.18899476826190947,
|
|
"num_tokens": 8343786.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.57136173248291,
|
|
"epoch": 3.726489028213166,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003958656328792565,
|
|
"loss": 5.0836,
|
|
"mean_token_accuracy": 0.19064647555351258,
|
|
"num_tokens": 8352286.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.44538722038269,
|
|
"epoch": 3.730407523510972,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039561181777249396,
|
|
"loss": 5.0,
|
|
"mean_token_accuracy": 0.19029666632413864,
|
|
"num_tokens": 8361274.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.500004291534424,
|
|
"epoch": 3.7343260188087775,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039535778711285676,
|
|
"loss": 5.0566,
|
|
"mean_token_accuracy": 0.19525335282087325,
|
|
"num_tokens": 8369881.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.560193204879761,
|
|
"epoch": 3.738244514106583,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003951035413543388,
|
|
"loss": 5.1105,
|
|
"mean_token_accuracy": 0.17877753973007202,
|
|
"num_tokens": 8378234.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.452444648742675,
|
|
"epoch": 3.742163009404389,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039484908095131874,
|
|
"loss": 5.0313,
|
|
"mean_token_accuracy": 0.19630018174648284,
|
|
"num_tokens": 8387023.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.464573097229004,
|
|
"epoch": 3.7460815047021945,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003945944063585582,
|
|
"loss": 4.9873,
|
|
"mean_token_accuracy": 0.19626669734716415,
|
|
"num_tokens": 8396110.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.449100065231323,
|
|
"epoch": 3.75,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039433951803120225,
|
|
"loss": 4.9573,
|
|
"mean_token_accuracy": 0.20301633179187775,
|
|
"num_tokens": 8405219.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.5450554370880125,
|
|
"epoch": 3.7539184952978055,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039408441642477764,
|
|
"loss": 5.1569,
|
|
"mean_token_accuracy": 0.1859276533126831,
|
|
"num_tokens": 8413996.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.448203134536743,
|
|
"epoch": 3.757836990595611,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003938291019951922,
|
|
"loss": 4.9235,
|
|
"mean_token_accuracy": 0.20487208068370819,
|
|
"num_tokens": 8421931.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.4548468589782715,
|
|
"epoch": 3.761755485893417,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003935735751987344,
|
|
"loss": 5.0094,
|
|
"mean_token_accuracy": 0.18600380122661592,
|
|
"num_tokens": 8430090.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.371029901504516,
|
|
"epoch": 3.7656739811912225,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039331783649207175,
|
|
"loss": 5.0027,
|
|
"mean_token_accuracy": 0.19213851541280746,
|
|
"num_tokens": 8439311.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.5566493511199955,
|
|
"epoch": 3.769592476489028,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039306188633225097,
|
|
"loss": 5.0576,
|
|
"mean_token_accuracy": 0.19423756897449493,
|
|
"num_tokens": 8448952.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.4812651634216305,
|
|
"epoch": 3.773510971786834,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003928057251766965,
|
|
"loss": 4.9864,
|
|
"mean_token_accuracy": 0.19111733436584472,
|
|
"num_tokens": 8457177.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.453255081176758,
|
|
"epoch": 3.7774294670846396,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039254935348320984,
|
|
"loss": 5.026,
|
|
"mean_token_accuracy": 0.19080139100551605,
|
|
"num_tokens": 8465844.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.562940835952759,
|
|
"epoch": 3.781347962382445,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00039229277170996885,
|
|
"loss": 5.1469,
|
|
"mean_token_accuracy": 0.1800929442048073,
|
|
"num_tokens": 8475198.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.564769124984741,
|
|
"epoch": 3.785266457680251,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003920359803155266,
|
|
"loss": 4.9991,
|
|
"mean_token_accuracy": 0.1911545589566231,
|
|
"num_tokens": 8484050.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.438898086547852,
|
|
"epoch": 3.7891849529780566,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039177897975881115,
|
|
"loss": 5.0059,
|
|
"mean_token_accuracy": 0.19609166234731673,
|
|
"num_tokens": 8491825.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.478066778182983,
|
|
"epoch": 3.793103448275862,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003915217704991239,
|
|
"loss": 5.003,
|
|
"mean_token_accuracy": 0.19153099209070207,
|
|
"num_tokens": 8500651.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.490873956680298,
|
|
"epoch": 3.7970219435736676,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003912643529961397,
|
|
"loss": 5.0074,
|
|
"mean_token_accuracy": 0.1905496597290039,
|
|
"num_tokens": 8508464.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.5082885265350345,
|
|
"epoch": 3.800940438871473,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003910067277099053,
|
|
"loss": 5.0441,
|
|
"mean_token_accuracy": 0.198540598154068,
|
|
"num_tokens": 8517676.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.586693048477173,
|
|
"epoch": 3.804858934169279,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 5.0936,
|
|
"mean_token_accuracy": 0.18898268193006515,
|
|
"num_tokens": 8526926.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.440959835052491,
|
|
"epoch": 3.8087774294670846,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003904908556297293,
|
|
"loss": 4.9743,
|
|
"mean_token_accuracy": 0.19295340031385422,
|
|
"num_tokens": 8535814.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.514721012115478,
|
|
"epoch": 3.81269592476489,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003902326097577345,
|
|
"loss": 5.0891,
|
|
"mean_token_accuracy": 0.19375635385513307,
|
|
"num_tokens": 8544814.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.509517192840576,
|
|
"epoch": 3.816614420062696,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038997415794638206,
|
|
"loss": 5.0767,
|
|
"mean_token_accuracy": 0.19104745239019394,
|
|
"num_tokens": 8553455.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.467943239212036,
|
|
"epoch": 3.8205329153605017,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003897155006575672,
|
|
"loss": 4.9841,
|
|
"mean_token_accuracy": 0.19213380068540573,
|
|
"num_tokens": 8562048.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.501840591430664,
|
|
"epoch": 3.824451410658307,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038945663835355247,
|
|
"loss": 5.0431,
|
|
"mean_token_accuracy": 0.191322460770607,
|
|
"num_tokens": 8571855.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.513002634048462,
|
|
"epoch": 3.8283699059561127,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038919757149696665,
|
|
"loss": 5.0397,
|
|
"mean_token_accuracy": 0.18685891926288606,
|
|
"num_tokens": 8580425.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.480809259414673,
|
|
"epoch": 3.8322884012539182,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038893830055080437,
|
|
"loss": 5.0642,
|
|
"mean_token_accuracy": 0.1918909952044487,
|
|
"num_tokens": 8589751.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.489587211608887,
|
|
"epoch": 3.836206896551724,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003886788259784248,
|
|
"loss": 5.076,
|
|
"mean_token_accuracy": 0.18437286019325255,
|
|
"num_tokens": 8598077.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.550845575332642,
|
|
"epoch": 3.8401253918495297,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038841914824355093,
|
|
"loss": 5.093,
|
|
"mean_token_accuracy": 0.18740802854299546,
|
|
"num_tokens": 8606691.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.448281860351562,
|
|
"epoch": 3.8440438871473352,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038815926781026914,
|
|
"loss": 5.0228,
|
|
"mean_token_accuracy": 0.18971820324659347,
|
|
"num_tokens": 8615477.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.427126407623291,
|
|
"epoch": 3.847962382445141,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003878991851430279,
|
|
"loss": 5.0264,
|
|
"mean_token_accuracy": 0.19527520090341569,
|
|
"num_tokens": 8623982.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.478499412536621,
|
|
"epoch": 3.8518808777429467,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003876389007066371,
|
|
"loss": 5.0636,
|
|
"mean_token_accuracy": 0.19111589789390565,
|
|
"num_tokens": 8633017.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.48462142944336,
|
|
"epoch": 3.8557993730407523,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003873784149662672,
|
|
"loss": 5.022,
|
|
"mean_token_accuracy": 0.1917983904480934,
|
|
"num_tokens": 8642299.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.4496711730957035,
|
|
"epoch": 3.8597178683385582,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003871177283874484,
|
|
"loss": 4.9618,
|
|
"mean_token_accuracy": 0.18854192793369293,
|
|
"num_tokens": 8651008.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.457749271392823,
|
|
"epoch": 3.8636363636363638,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038685684143606995,
|
|
"loss": 5.0452,
|
|
"mean_token_accuracy": 0.19297476708889008,
|
|
"num_tokens": 8660103.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.3892570495605465,
|
|
"epoch": 3.8675548589341693,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003865957545783791,
|
|
"loss": 4.9817,
|
|
"mean_token_accuracy": 0.19850101321935654,
|
|
"num_tokens": 8668646.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.4849433422088625,
|
|
"epoch": 3.871473354231975,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038633446828098046,
|
|
"loss": 5.1272,
|
|
"mean_token_accuracy": 0.18707022219896316,
|
|
"num_tokens": 8678190.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.620087099075318,
|
|
"epoch": 3.8753918495297803,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000386072983010835,
|
|
"loss": 5.1386,
|
|
"mean_token_accuracy": 0.1761440023779869,
|
|
"num_tokens": 8687336.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.521949338912964,
|
|
"epoch": 3.8793103448275863,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038581129923525914,
|
|
"loss": 5.0373,
|
|
"mean_token_accuracy": 0.19202707707881927,
|
|
"num_tokens": 8695939.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.477308988571167,
|
|
"epoch": 3.883228840125392,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038554941742192445,
|
|
"loss": 5.0865,
|
|
"mean_token_accuracy": 0.18982657343149184,
|
|
"num_tokens": 8704828.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.510231256484985,
|
|
"epoch": 3.8871473354231973,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003852873380388561,
|
|
"loss": 4.9964,
|
|
"mean_token_accuracy": 0.2003360241651535,
|
|
"num_tokens": 8714427.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.547918224334717,
|
|
"epoch": 3.8910658307210033,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003850250615544323,
|
|
"loss": 5.1004,
|
|
"mean_token_accuracy": 0.19415017068386078,
|
|
"num_tokens": 8723737.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.446525144577026,
|
|
"epoch": 3.894984326018809,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038476258843738386,
|
|
"loss": 5.0533,
|
|
"mean_token_accuracy": 0.1902598187327385,
|
|
"num_tokens": 8733149.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.429483842849732,
|
|
"epoch": 3.8989028213166144,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038449991915679273,
|
|
"loss": 5.0026,
|
|
"mean_token_accuracy": 0.19840745329856874,
|
|
"num_tokens": 8742000.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.496473360061645,
|
|
"epoch": 3.9028213166144203,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003842370541820915,
|
|
"loss": 5.0767,
|
|
"mean_token_accuracy": 0.18689345419406891,
|
|
"num_tokens": 8751069.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.477064847946167,
|
|
"epoch": 3.906739811912226,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038397399398306243,
|
|
"loss": 5.005,
|
|
"mean_token_accuracy": 0.18725234866142274,
|
|
"num_tokens": 8760092.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.495781373977661,
|
|
"epoch": 3.9106583072100314,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038371073902983684,
|
|
"loss": 5.0105,
|
|
"mean_token_accuracy": 0.1943519875407219,
|
|
"num_tokens": 8768039.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.498169326782227,
|
|
"epoch": 3.914576802507837,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000383447289792894,
|
|
"loss": 5.0539,
|
|
"mean_token_accuracy": 0.1849008560180664,
|
|
"num_tokens": 8776630.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.523120498657226,
|
|
"epoch": 3.9184952978056424,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038318364674306036,
|
|
"loss": 5.0155,
|
|
"mean_token_accuracy": 0.1936942920088768,
|
|
"num_tokens": 8785272.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 3.9184952978056424,
|
|
"eval_entropy": 5.30318573833436,
|
|
"eval_loss": 5.639626979827881,
|
|
"eval_mean_token_accuracy": 0.1701374023777339,
|
|
"eval_num_tokens": 8785272.0,
|
|
"eval_runtime": 2.365,
|
|
"eval_samples_per_second": 1742.908,
|
|
"eval_steps_per_second": 218.181,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.4814770221710205,
|
|
"epoch": 3.9224137931034484,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038291981035150883,
|
|
"loss": 5.0268,
|
|
"mean_token_accuracy": 0.19624612778425216,
|
|
"num_tokens": 8794197.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.503341960906982,
|
|
"epoch": 3.926332288401254,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003826557810897579,
|
|
"loss": 5.0466,
|
|
"mean_token_accuracy": 0.19077856093645096,
|
|
"num_tokens": 8803195.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.567094850540161,
|
|
"epoch": 3.9302507836990594,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003823915594296705,
|
|
"loss": 5.0089,
|
|
"mean_token_accuracy": 0.2010100156068802,
|
|
"num_tokens": 8811772.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.357451009750366,
|
|
"epoch": 3.9341692789968654,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003821271458434538,
|
|
"loss": 4.975,
|
|
"mean_token_accuracy": 0.1946580782532692,
|
|
"num_tokens": 8820482.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.400775146484375,
|
|
"epoch": 3.938087774294671,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038186254080365756,
|
|
"loss": 5.0591,
|
|
"mean_token_accuracy": 0.19189645498991012,
|
|
"num_tokens": 8828761.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.431986856460571,
|
|
"epoch": 3.9420062695924765,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000381597744783174,
|
|
"loss": 4.9448,
|
|
"mean_token_accuracy": 0.18972737491130828,
|
|
"num_tokens": 8837165.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.466662168502808,
|
|
"epoch": 3.945924764890282,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038133275825523645,
|
|
"loss": 5.0163,
|
|
"mean_token_accuracy": 0.19124049693346024,
|
|
"num_tokens": 8845472.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.4750594139099125,
|
|
"epoch": 3.9498432601880875,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038106758169341887,
|
|
"loss": 5.0348,
|
|
"mean_token_accuracy": 0.19356803745031356,
|
|
"num_tokens": 8855246.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.50919737815857,
|
|
"epoch": 3.9537617554858935,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003808022155716348,
|
|
"loss": 5.0986,
|
|
"mean_token_accuracy": 0.18736936748027802,
|
|
"num_tokens": 8863667.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.456997346878052,
|
|
"epoch": 3.957680250783699,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003805366603641364,
|
|
"loss": 4.9423,
|
|
"mean_token_accuracy": 0.19759218096733094,
|
|
"num_tokens": 8872306.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.379599094390869,
|
|
"epoch": 3.9615987460815045,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038027091654551406,
|
|
"loss": 4.9604,
|
|
"mean_token_accuracy": 0.19560502618551254,
|
|
"num_tokens": 8882037.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.37017183303833,
|
|
"epoch": 3.9655172413793105,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00038000498459069487,
|
|
"loss": 5.013,
|
|
"mean_token_accuracy": 0.193723164498806,
|
|
"num_tokens": 8890685.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.446093320846558,
|
|
"epoch": 3.969435736677116,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003797388649749426,
|
|
"loss": 5.0048,
|
|
"mean_token_accuracy": 0.19560179561376573,
|
|
"num_tokens": 8899927.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.4495796203613285,
|
|
"epoch": 3.9733542319749215,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003794725581738559,
|
|
"loss": 5.0367,
|
|
"mean_token_accuracy": 0.19382841736078263,
|
|
"num_tokens": 8908571.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.46975417137146,
|
|
"epoch": 3.9772727272727275,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037920606466336834,
|
|
"loss": 4.9526,
|
|
"mean_token_accuracy": 0.19654886424541473,
|
|
"num_tokens": 8917246.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.499976968765258,
|
|
"epoch": 3.981191222570533,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000378939384919747,
|
|
"loss": 5.0647,
|
|
"mean_token_accuracy": 0.19355631172657012,
|
|
"num_tokens": 8926116.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.521025276184082,
|
|
"epoch": 3.9851097178683386,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003786725194195918,
|
|
"loss": 5.0993,
|
|
"mean_token_accuracy": 0.1843150645494461,
|
|
"num_tokens": 8934914.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.483793163299561,
|
|
"epoch": 3.989028213166144,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 4.9842,
|
|
"mean_token_accuracy": 0.18871589452028276,
|
|
"num_tokens": 8943321.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.459476137161255,
|
|
"epoch": 3.9929467084639496,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00037813823305773883,
|
|
"loss": 5.0268,
|
|
"mean_token_accuracy": 0.1932402268052101,
|
|
"num_tokens": 8951848.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.514981985092163,
|
|
"epoch": 3.9968652037617556,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003778708131508974,
|
|
"loss": 5.0542,
|
|
"mean_token_accuracy": 0.19021925181150437,
|
|
"num_tokens": 8961291.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.557011318206787,
|
|
"epoch": 4.0007836990595615,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037760320939723307,
|
|
"loss": 5.0599,
|
|
"mean_token_accuracy": 0.19016327410936357,
|
|
"num_tokens": 8971116.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.437959766387939,
|
|
"epoch": 4.004702194357367,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037733542227499727,
|
|
"loss": 4.7104,
|
|
"mean_token_accuracy": 0.2168150082230568,
|
|
"num_tokens": 8979362.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.519193077087403,
|
|
"epoch": 4.008620689655173,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037706745226276893,
|
|
"loss": 4.8852,
|
|
"mean_token_accuracy": 0.2058098018169403,
|
|
"num_tokens": 8987728.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.462403011322022,
|
|
"epoch": 4.012539184952978,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003767992998394539,
|
|
"loss": 4.7216,
|
|
"mean_token_accuracy": 0.21652127653360367,
|
|
"num_tokens": 8995817.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.470757102966308,
|
|
"epoch": 4.016457680250784,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003765309654842842,
|
|
"loss": 4.8079,
|
|
"mean_token_accuracy": 0.20475055575370787,
|
|
"num_tokens": 9005683.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.395235395431518,
|
|
"epoch": 4.02037617554859,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003762624496768166,
|
|
"loss": 4.7017,
|
|
"mean_token_accuracy": 0.22054595202207566,
|
|
"num_tokens": 9015408.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.510372447967529,
|
|
"epoch": 4.024294670846395,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003759937528969325,
|
|
"loss": 4.852,
|
|
"mean_token_accuracy": 0.20371596962213517,
|
|
"num_tokens": 9024381.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.452179050445556,
|
|
"epoch": 4.028213166144201,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037572487562483666,
|
|
"loss": 4.7302,
|
|
"mean_token_accuracy": 0.21575197577476501,
|
|
"num_tokens": 9033092.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.421417999267578,
|
|
"epoch": 4.032131661442007,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037545581834105623,
|
|
"loss": 4.844,
|
|
"mean_token_accuracy": 0.21289148330688476,
|
|
"num_tokens": 9042251.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.380678367614746,
|
|
"epoch": 4.036050156739812,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037518658152644035,
|
|
"loss": 4.7286,
|
|
"mean_token_accuracy": 0.2122337132692337,
|
|
"num_tokens": 9050628.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.439891719818116,
|
|
"epoch": 4.039968652037618,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003749171656621588,
|
|
"loss": 4.7688,
|
|
"mean_token_accuracy": 0.22236442863941192,
|
|
"num_tokens": 9060080.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.502091360092163,
|
|
"epoch": 4.043887147335423,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003746475712297014,
|
|
"loss": 4.782,
|
|
"mean_token_accuracy": 0.2179523676633835,
|
|
"num_tokens": 9069456.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.480469179153443,
|
|
"epoch": 4.047805642633229,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003743777987108771,
|
|
"loss": 4.7954,
|
|
"mean_token_accuracy": 0.215857695043087,
|
|
"num_tokens": 9078790.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.319148588180542,
|
|
"epoch": 4.051724137931035,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003741078485878132,
|
|
"loss": 4.7769,
|
|
"mean_token_accuracy": 0.2137386530637741,
|
|
"num_tokens": 9086888.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.412674331665039,
|
|
"epoch": 4.05564263322884,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003738377213429542,
|
|
"loss": 4.807,
|
|
"mean_token_accuracy": 0.20727071315050125,
|
|
"num_tokens": 9095139.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.4732200622558596,
|
|
"epoch": 4.059561128526646,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003735674174590614,
|
|
"loss": 4.7351,
|
|
"mean_token_accuracy": 0.21636737287044525,
|
|
"num_tokens": 9103464.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.437044286727906,
|
|
"epoch": 4.063479623824452,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037329693741921166,
|
|
"loss": 4.7471,
|
|
"mean_token_accuracy": 0.21393001526594163,
|
|
"num_tokens": 9111290.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.284782028198242,
|
|
"epoch": 4.067398119122257,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003730262817067967,
|
|
"loss": 4.6237,
|
|
"mean_token_accuracy": 0.21698374003171922,
|
|
"num_tokens": 9120224.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.396932172775268,
|
|
"epoch": 4.071316614420063,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003727554508055221,
|
|
"loss": 4.776,
|
|
"mean_token_accuracy": 0.21098583787679673,
|
|
"num_tokens": 9128524.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.38273777961731,
|
|
"epoch": 4.075235109717869,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003724844451994066,
|
|
"loss": 4.6661,
|
|
"mean_token_accuracy": 0.21724938452243805,
|
|
"num_tokens": 9137026.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.410043001174927,
|
|
"epoch": 4.079153605015674,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037221326537278113,
|
|
"loss": 4.7507,
|
|
"mean_token_accuracy": 0.21346322894096376,
|
|
"num_tokens": 9146312.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.404902219772339,
|
|
"epoch": 4.08307210031348,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037194191181028806,
|
|
"loss": 4.7599,
|
|
"mean_token_accuracy": 0.2111594170331955,
|
|
"num_tokens": 9155226.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.501644420623779,
|
|
"epoch": 4.086990595611285,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00037167038499688024,
|
|
"loss": 4.8874,
|
|
"mean_token_accuracy": 0.20971559286117553,
|
|
"num_tokens": 9164090.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.4440836906433105,
|
|
"epoch": 4.090909090909091,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037139868541782,
|
|
"loss": 4.7504,
|
|
"mean_token_accuracy": 0.2207699164748192,
|
|
"num_tokens": 9173002.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.476128387451172,
|
|
"epoch": 4.094827586206897,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003711268135586787,
|
|
"loss": 4.776,
|
|
"mean_token_accuracy": 0.2099992960691452,
|
|
"num_tokens": 9182020.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.3916144371032715,
|
|
"epoch": 4.098746081504702,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037085476990533523,
|
|
"loss": 4.7987,
|
|
"mean_token_accuracy": 0.2086474135518074,
|
|
"num_tokens": 9190848.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.4069411277771,
|
|
"epoch": 4.102664576802508,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003705825549439759,
|
|
"loss": 4.7216,
|
|
"mean_token_accuracy": 0.22146847993135452,
|
|
"num_tokens": 9198808.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.466919088363648,
|
|
"epoch": 4.106583072100314,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037031016916109295,
|
|
"loss": 4.8486,
|
|
"mean_token_accuracy": 0.20848954916000367,
|
|
"num_tokens": 9207923.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.414611434936523,
|
|
"epoch": 4.110501567398119,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037003761304348374,
|
|
"loss": 4.7909,
|
|
"mean_token_accuracy": 0.21071001887321472,
|
|
"num_tokens": 9216759.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.383496379852295,
|
|
"epoch": 4.114420062695925,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036976488707825056,
|
|
"loss": 4.7472,
|
|
"mean_token_accuracy": 0.20549267679452896,
|
|
"num_tokens": 9225951.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.373333311080932,
|
|
"epoch": 4.118338557993731,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00036949199175279876,
|
|
"loss": 4.797,
|
|
"mean_token_accuracy": 0.20787876546382905,
|
|
"num_tokens": 9234413.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.526278877258301,
|
|
"epoch": 4.122257053291536,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003692189275548364,
|
|
"loss": 4.9363,
|
|
"mean_token_accuracy": 0.19888079911470413,
|
|
"num_tokens": 9244002.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.469981527328491,
|
|
"epoch": 4.126175548589342,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003689456949723737,
|
|
"loss": 4.7585,
|
|
"mean_token_accuracy": 0.2130420610308647,
|
|
"num_tokens": 9252689.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.467747592926026,
|
|
"epoch": 4.130094043887147,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003686722944937215,
|
|
"loss": 4.7604,
|
|
"mean_token_accuracy": 0.213399975001812,
|
|
"num_tokens": 9261118.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.369660139083862,
|
|
"epoch": 4.134012539184953,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036839872660749084,
|
|
"loss": 4.741,
|
|
"mean_token_accuracy": 0.21492306143045425,
|
|
"num_tokens": 9269205.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.391610431671142,
|
|
"epoch": 4.137931034482759,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003681249918025918,
|
|
"loss": 4.7648,
|
|
"mean_token_accuracy": 0.21594133824110032,
|
|
"num_tokens": 9277770.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.339451456069947,
|
|
"epoch": 4.141849529780564,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036785109056823297,
|
|
"loss": 4.7983,
|
|
"mean_token_accuracy": 0.21199292689561844,
|
|
"num_tokens": 9286589.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.434355783462524,
|
|
"epoch": 4.14576802507837,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00036757702339392,
|
|
"loss": 4.7853,
|
|
"mean_token_accuracy": 0.20837830007076263,
|
|
"num_tokens": 9295264.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.40733962059021,
|
|
"epoch": 4.149686520376176,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036730279076945574,
|
|
"loss": 4.7937,
|
|
"mean_token_accuracy": 0.2109970837831497,
|
|
"num_tokens": 9303525.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.410266733169555,
|
|
"epoch": 4.153605015673981,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003670283931849382,
|
|
"loss": 4.8283,
|
|
"mean_token_accuracy": 0.20308942049741746,
|
|
"num_tokens": 9312839.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.419910573959351,
|
|
"epoch": 4.157523510971787,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036675383113076023,
|
|
"loss": 4.7805,
|
|
"mean_token_accuracy": 0.21615110486745834,
|
|
"num_tokens": 9321462.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.385771036148071,
|
|
"epoch": 4.161442006269592,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00036647910509760896,
|
|
"loss": 4.8069,
|
|
"mean_token_accuracy": 0.21139971166849136,
|
|
"num_tokens": 9329315.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.45748348236084,
|
|
"epoch": 4.165360501567398,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00036620421557646414,
|
|
"loss": 4.7926,
|
|
"mean_token_accuracy": 0.20871471017599105,
|
|
"num_tokens": 9336997.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.367199230194092,
|
|
"epoch": 4.169278996865204,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000365929163058598,
|
|
"loss": 4.7359,
|
|
"mean_token_accuracy": 0.21371042728424072,
|
|
"num_tokens": 9345364.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.39553689956665,
|
|
"epoch": 4.173197492163009,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 4.7436,
|
|
"mean_token_accuracy": 0.21399117708206178,
|
|
"num_tokens": 9353961.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.480158901214599,
|
|
"epoch": 4.177115987460815,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003653785709992462,
|
|
"loss": 4.8979,
|
|
"mean_token_accuracy": 0.20475429445505142,
|
|
"num_tokens": 9362901.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.452304077148438,
|
|
"epoch": 4.181034482758621,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036510303244175775,
|
|
"loss": 4.8094,
|
|
"mean_token_accuracy": 0.21008759289979934,
|
|
"num_tokens": 9371800.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.38066873550415,
|
|
"epoch": 4.184952978056426,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036482733285554073,
|
|
"loss": 4.7944,
|
|
"mean_token_accuracy": 0.21615577042102813,
|
|
"num_tokens": 9380033.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.532425498962402,
|
|
"epoch": 4.188871473354232,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00036455147273331507,
|
|
"loss": 4.8657,
|
|
"mean_token_accuracy": 0.2098294749855995,
|
|
"num_tokens": 9389264.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.41873140335083,
|
|
"epoch": 4.192789968652038,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003642754525680876,
|
|
"loss": 4.7706,
|
|
"mean_token_accuracy": 0.21597768664360045,
|
|
"num_tokens": 9398112.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.447447633743286,
|
|
"epoch": 4.196708463949843,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036399927285315104,
|
|
"loss": 4.8971,
|
|
"mean_token_accuracy": 0.20811898559331893,
|
|
"num_tokens": 9407400.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.4437745094299315,
|
|
"epoch": 4.200626959247649,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003637229340820834,
|
|
"loss": 4.7857,
|
|
"mean_token_accuracy": 0.2037736415863037,
|
|
"num_tokens": 9416351.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.4891869068145756,
|
|
"epoch": 4.204545454545454,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003634464367487467,
|
|
"loss": 4.8214,
|
|
"mean_token_accuracy": 0.20914797037839888,
|
|
"num_tokens": 9424577.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.423868322372437,
|
|
"epoch": 4.20846394984326,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003631697813472867,
|
|
"loss": 4.8018,
|
|
"mean_token_accuracy": 0.21426962167024613,
|
|
"num_tokens": 9434242.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.34076566696167,
|
|
"epoch": 4.212382445141066,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036289296837213136,
|
|
"loss": 4.7714,
|
|
"mean_token_accuracy": 0.21370630860328674,
|
|
"num_tokens": 9443247.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.370574712753296,
|
|
"epoch": 4.216300940438871,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00036261599831799034,
|
|
"loss": 4.737,
|
|
"mean_token_accuracy": 0.21416351050138474,
|
|
"num_tokens": 9451998.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.465113735198974,
|
|
"epoch": 4.220219435736677,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003623388716798541,
|
|
"loss": 4.7169,
|
|
"mean_token_accuracy": 0.21369974166154862,
|
|
"num_tokens": 9459993.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.372065544128418,
|
|
"epoch": 4.224137931034483,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036206158895299277,
|
|
"loss": 4.743,
|
|
"mean_token_accuracy": 0.21590758562088014,
|
|
"num_tokens": 9468036.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.327731466293335,
|
|
"epoch": 4.228056426332288,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036178415063295574,
|
|
"loss": 4.8279,
|
|
"mean_token_accuracy": 0.20963084250688552,
|
|
"num_tokens": 9477031.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.4298415184021,
|
|
"epoch": 4.231974921630094,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00036150655721557,
|
|
"loss": 4.7242,
|
|
"mean_token_accuracy": 0.21294908076524735,
|
|
"num_tokens": 9485687.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.407687711715698,
|
|
"epoch": 4.2358934169279,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003612288091969402,
|
|
"loss": 4.7765,
|
|
"mean_token_accuracy": 0.2173039510846138,
|
|
"num_tokens": 9494613.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.408217239379883,
|
|
"epoch": 4.239811912225705,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036095090707344697,
|
|
"loss": 4.7764,
|
|
"mean_token_accuracy": 0.21094192564487457,
|
|
"num_tokens": 9503522.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.413130569458008,
|
|
"epoch": 4.243730407523511,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00036067285134174654,
|
|
"loss": 4.7888,
|
|
"mean_token_accuracy": 0.21247149556875228,
|
|
"num_tokens": 9512293.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.4927815914154055,
|
|
"epoch": 4.247648902821316,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003603946424987696,
|
|
"loss": 4.9094,
|
|
"mean_token_accuracy": 0.20212927758693694,
|
|
"num_tokens": 9520999.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.437141990661621,
|
|
"epoch": 4.251567398119122,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036011628104172045,
|
|
"loss": 4.8212,
|
|
"mean_token_accuracy": 0.20826724022626877,
|
|
"num_tokens": 9529798.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.505478715896606,
|
|
"epoch": 4.255485893416928,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035983776746807613,
|
|
"loss": 4.7794,
|
|
"mean_token_accuracy": 0.20929668843746185,
|
|
"num_tokens": 9538668.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.475000953674316,
|
|
"epoch": 4.259404388714733,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003595591022755855,
|
|
"loss": 4.8193,
|
|
"mean_token_accuracy": 0.209816375374794,
|
|
"num_tokens": 9546973.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.32633147239685,
|
|
"epoch": 4.263322884012539,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003592802859622684,
|
|
"loss": 4.7489,
|
|
"mean_token_accuracy": 0.21123570203781128,
|
|
"num_tokens": 9555947.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.410082244873047,
|
|
"epoch": 4.267241379310345,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000359001319026415,
|
|
"loss": 4.8383,
|
|
"mean_token_accuracy": 0.2066846564412117,
|
|
"num_tokens": 9564730.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.379844903945923,
|
|
"epoch": 4.27115987460815,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035872220196658426,
|
|
"loss": 4.7913,
|
|
"mean_token_accuracy": 0.20767433047294617,
|
|
"num_tokens": 9573470.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.476673173904419,
|
|
"epoch": 4.275078369905956,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035844293528160384,
|
|
"loss": 4.8412,
|
|
"mean_token_accuracy": 0.20179694890975952,
|
|
"num_tokens": 9581607.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.3871252059936525,
|
|
"epoch": 4.278996865203762,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003581635194705682,
|
|
"loss": 4.7834,
|
|
"mean_token_accuracy": 0.20789219290018082,
|
|
"num_tokens": 9589801.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.450583982467651,
|
|
"epoch": 4.282915360501567,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00035788395503283914,
|
|
"loss": 4.8438,
|
|
"mean_token_accuracy": 0.20022940933704375,
|
|
"num_tokens": 9599983.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.449062633514404,
|
|
"epoch": 4.286833855799373,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035760424246804334,
|
|
"loss": 4.7561,
|
|
"mean_token_accuracy": 0.2161661222577095,
|
|
"num_tokens": 9608681.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.363840341567993,
|
|
"epoch": 4.290752351097178,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003573243822760726,
|
|
"loss": 4.7684,
|
|
"mean_token_accuracy": 0.21528727412223816,
|
|
"num_tokens": 9616461.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.423112487792968,
|
|
"epoch": 4.294670846394984,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003570443749570825,
|
|
"loss": 4.8531,
|
|
"mean_token_accuracy": 0.20810550302267075,
|
|
"num_tokens": 9625480.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.471208572387695,
|
|
"epoch": 4.29858934169279,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003567642210114914,
|
|
"loss": 4.8115,
|
|
"mean_token_accuracy": 0.2064492717385292,
|
|
"num_tokens": 9634172.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.359401798248291,
|
|
"epoch": 4.302507836990595,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00035648392093998,
|
|
"loss": 4.8034,
|
|
"mean_token_accuracy": 0.21256857365369797,
|
|
"num_tokens": 9643196.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.355448484420776,
|
|
"epoch": 4.306426332288401,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003562034752434899,
|
|
"loss": 4.716,
|
|
"mean_token_accuracy": 0.2136015772819519,
|
|
"num_tokens": 9652440.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.368397045135498,
|
|
"epoch": 4.310344827586207,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003559228844232229,
|
|
"loss": 4.8079,
|
|
"mean_token_accuracy": 0.21064997017383574,
|
|
"num_tokens": 9661248.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.310344827586207,
|
|
"eval_entropy": 5.197473784749822,
|
|
"eval_loss": 5.659706115722656,
|
|
"eval_mean_token_accuracy": 0.16965749982542308,
|
|
"eval_num_tokens": 9661248.0,
|
|
"eval_runtime": 2.183,
|
|
"eval_samples_per_second": 1888.267,
|
|
"eval_steps_per_second": 236.377,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.443186807632446,
|
|
"epoch": 4.314263322884012,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003556421489806406,
|
|
"loss": 4.8241,
|
|
"mean_token_accuracy": 0.20767869353294371,
|
|
"num_tokens": 9670544.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.434499311447143,
|
|
"epoch": 4.318181818181818,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003553612694174624,
|
|
"loss": 4.7963,
|
|
"mean_token_accuracy": 0.20861081928014755,
|
|
"num_tokens": 9679331.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.541751670837402,
|
|
"epoch": 4.322100313479623,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003550802462356659,
|
|
"loss": 4.8628,
|
|
"mean_token_accuracy": 0.20491621047258377,
|
|
"num_tokens": 9688211.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.3269219398498535,
|
|
"epoch": 4.326018808777429,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00035479907993748484,
|
|
"loss": 4.721,
|
|
"mean_token_accuracy": 0.21988922655582427,
|
|
"num_tokens": 9696231.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.392792463302612,
|
|
"epoch": 4.329937304075235,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035451777102540915,
|
|
"loss": 4.8192,
|
|
"mean_token_accuracy": 0.20204265862703324,
|
|
"num_tokens": 9705577.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.485252141952515,
|
|
"epoch": 4.33385579937304,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003542363200021835,
|
|
"loss": 4.8349,
|
|
"mean_token_accuracy": 0.20878250151872635,
|
|
"num_tokens": 9715003.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.387903308868408,
|
|
"epoch": 4.337774294670846,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003539547273708064,
|
|
"loss": 4.7244,
|
|
"mean_token_accuracy": 0.21347908079624175,
|
|
"num_tokens": 9723513.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.327245473861694,
|
|
"epoch": 4.341692789968652,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00035367299363452966,
|
|
"loss": 4.7922,
|
|
"mean_token_accuracy": 0.2034262865781784,
|
|
"num_tokens": 9731800.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.378030014038086,
|
|
"epoch": 4.3456112852664575,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035339111929685694,
|
|
"loss": 4.7733,
|
|
"mean_token_accuracy": 0.20506692826747894,
|
|
"num_tokens": 9740356.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.455407285690308,
|
|
"epoch": 4.349529780564263,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035310910486154365,
|
|
"loss": 4.8206,
|
|
"mean_token_accuracy": 0.19646463245153428,
|
|
"num_tokens": 9749529.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.461986398696899,
|
|
"epoch": 4.353448275862069,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003528269508325951,
|
|
"loss": 4.794,
|
|
"mean_token_accuracy": 0.21268730461597443,
|
|
"num_tokens": 9758750.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.423493766784668,
|
|
"epoch": 4.3573667711598745,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 4.8772,
|
|
"mean_token_accuracy": 0.20892336517572402,
|
|
"num_tokens": 9766992.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.467726230621338,
|
|
"epoch": 4.36128526645768,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035226222601106105,
|
|
"loss": 4.8806,
|
|
"mean_token_accuracy": 0.2037777841091156,
|
|
"num_tokens": 9777022.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.381536293029785,
|
|
"epoch": 4.3652037617554855,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003519796562277303,
|
|
"loss": 4.6461,
|
|
"mean_token_accuracy": 0.22370839715003968,
|
|
"num_tokens": 9785863.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.386045026779175,
|
|
"epoch": 4.3691222570532915,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003516969488692723,
|
|
"loss": 4.7889,
|
|
"mean_token_accuracy": 0.2056066781282425,
|
|
"num_tokens": 9794357.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.328175163269043,
|
|
"epoch": 4.3730407523510975,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003514141044409308,
|
|
"loss": 4.7736,
|
|
"mean_token_accuracy": 0.2111730009317398,
|
|
"num_tokens": 9803128.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.434909772872925,
|
|
"epoch": 4.3769592476489025,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003511311234481947,
|
|
"loss": 4.8601,
|
|
"mean_token_accuracy": 0.20311858057975768,
|
|
"num_tokens": 9812046.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.378364181518554,
|
|
"epoch": 4.3808777429467085,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035084800639679695,
|
|
"loss": 4.7479,
|
|
"mean_token_accuracy": 0.215052530169487,
|
|
"num_tokens": 9820935.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.4372443675994875,
|
|
"epoch": 4.3847962382445145,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035056475379271356,
|
|
"loss": 4.8642,
|
|
"mean_token_accuracy": 0.2043449714779854,
|
|
"num_tokens": 9829585.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.509523057937622,
|
|
"epoch": 4.38871473354232,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003502813661421629,
|
|
"loss": 4.9328,
|
|
"mean_token_accuracy": 0.19884804785251617,
|
|
"num_tokens": 9839302.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.419396018981933,
|
|
"epoch": 4.3926332288401255,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003499978439516047,
|
|
"loss": 4.8219,
|
|
"mean_token_accuracy": 0.20697667598724365,
|
|
"num_tokens": 9848330.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.346432733535766,
|
|
"epoch": 4.396551724137931,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034971418772773895,
|
|
"loss": 4.7796,
|
|
"mean_token_accuracy": 0.21550718694925308,
|
|
"num_tokens": 9856928.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.448116683959961,
|
|
"epoch": 4.400470219435737,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003494303979775054,
|
|
"loss": 4.8468,
|
|
"mean_token_accuracy": 0.20702877044677734,
|
|
"num_tokens": 9865464.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.446243000030518,
|
|
"epoch": 4.4043887147335425,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034914647520808235,
|
|
"loss": 4.8251,
|
|
"mean_token_accuracy": 0.2114207774400711,
|
|
"num_tokens": 9874345.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.324807357788086,
|
|
"epoch": 4.408307210031348,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003488624199268859,
|
|
"loss": 4.7553,
|
|
"mean_token_accuracy": 0.21617880910634996,
|
|
"num_tokens": 9883135.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.425197410583496,
|
|
"epoch": 4.412225705329154,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003485782326415687,
|
|
"loss": 4.8487,
|
|
"mean_token_accuracy": 0.2090997189283371,
|
|
"num_tokens": 9892229.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.398403215408325,
|
|
"epoch": 4.41614420062696,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003482939138600197,
|
|
"loss": 4.8203,
|
|
"mean_token_accuracy": 0.20667444914579391,
|
|
"num_tokens": 9901044.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.421803951263428,
|
|
"epoch": 4.420062695924765,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003480094640903626,
|
|
"loss": 4.8716,
|
|
"mean_token_accuracy": 0.19587522745132446,
|
|
"num_tokens": 9909818.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.33640685081482,
|
|
"epoch": 4.423981191222571,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034772488384095517,
|
|
"loss": 4.738,
|
|
"mean_token_accuracy": 0.213012632727623,
|
|
"num_tokens": 9919295.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.404332065582276,
|
|
"epoch": 4.427899686520377,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00034744017362038854,
|
|
"loss": 4.8186,
|
|
"mean_token_accuracy": 0.2061314254999161,
|
|
"num_tokens": 9927539.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.397854852676391,
|
|
"epoch": 4.431818181818182,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034715533393748604,
|
|
"loss": 4.8555,
|
|
"mean_token_accuracy": 0.20452551692724227,
|
|
"num_tokens": 9936540.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.390025043487549,
|
|
"epoch": 4.435736677115988,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034687036530130244,
|
|
"loss": 4.7208,
|
|
"mean_token_accuracy": 0.2196156069636345,
|
|
"num_tokens": 9945065.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.389207792282105,
|
|
"epoch": 4.439655172413793,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003465852682211227,
|
|
"loss": 4.7654,
|
|
"mean_token_accuracy": 0.21036447882652282,
|
|
"num_tokens": 9954155.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.379635190963745,
|
|
"epoch": 4.443573667711599,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003463000432064617,
|
|
"loss": 4.8227,
|
|
"mean_token_accuracy": 0.2046230435371399,
|
|
"num_tokens": 9963269.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.466479730606079,
|
|
"epoch": 4.447492163009405,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034601469076706276,
|
|
"loss": 4.8694,
|
|
"mean_token_accuracy": 0.2117801934480667,
|
|
"num_tokens": 9972688.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.490498304367065,
|
|
"epoch": 4.45141065830721,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003457292114128968,
|
|
"loss": 4.8314,
|
|
"mean_token_accuracy": 0.20274507701396943,
|
|
"num_tokens": 9981283.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.3577546119689945,
|
|
"epoch": 4.455329153605016,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000345443605654162,
|
|
"loss": 4.8353,
|
|
"mean_token_accuracy": 0.20901550948619843,
|
|
"num_tokens": 9989870.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.419043922424317,
|
|
"epoch": 4.459247648902822,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00034515787400128195,
|
|
"loss": 4.8526,
|
|
"mean_token_accuracy": 0.20574190318584443,
|
|
"num_tokens": 9998136.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.402076959609985,
|
|
"epoch": 4.463166144200627,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003448720169649054,
|
|
"loss": 4.778,
|
|
"mean_token_accuracy": 0.21142067909240722,
|
|
"num_tokens": 10006730.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.419503211975098,
|
|
"epoch": 4.467084639498433,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003445860350559055,
|
|
"loss": 4.8044,
|
|
"mean_token_accuracy": 0.21988747268915176,
|
|
"num_tokens": 10015517.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.351985931396484,
|
|
"epoch": 4.471003134796238,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000344299928785378,
|
|
"loss": 4.8177,
|
|
"mean_token_accuracy": 0.2043176844716072,
|
|
"num_tokens": 10024552.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.39212384223938,
|
|
"epoch": 4.474921630094044,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00034401369866464133,
|
|
"loss": 4.8127,
|
|
"mean_token_accuracy": 0.20933349430561066,
|
|
"num_tokens": 10033523.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.3613849639892575,
|
|
"epoch": 4.47884012539185,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00034372734520523505,
|
|
"loss": 4.7069,
|
|
"mean_token_accuracy": 0.2153206929564476,
|
|
"num_tokens": 10041988.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.312775182723999,
|
|
"epoch": 4.482758620689655,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003434408689189193,
|
|
"loss": 4.7764,
|
|
"mean_token_accuracy": 0.21232035756111145,
|
|
"num_tokens": 10050543.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.392627859115601,
|
|
"epoch": 4.486677115987461,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003431542703176737,
|
|
"loss": 4.8015,
|
|
"mean_token_accuracy": 0.20773497670888902,
|
|
"num_tokens": 10059300.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.436697673797608,
|
|
"epoch": 4.490595611285267,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003428675499136963,
|
|
"loss": 4.8323,
|
|
"mean_token_accuracy": 0.212333382666111,
|
|
"num_tokens": 10068535.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.4918276309967045,
|
|
"epoch": 4.494514106583072,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000342580708219403,
|
|
"loss": 4.8912,
|
|
"mean_token_accuracy": 0.21103271543979646,
|
|
"num_tokens": 10077731.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.416774320602417,
|
|
"epoch": 4.498432601880878,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003422937457474264,
|
|
"loss": 4.7907,
|
|
"mean_token_accuracy": 0.20488784313201905,
|
|
"num_tokens": 10087146.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.398598146438599,
|
|
"epoch": 4.502351097178684,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034200666301061495,
|
|
"loss": 4.7997,
|
|
"mean_token_accuracy": 0.21014934182167053,
|
|
"num_tokens": 10096659.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.412196683883667,
|
|
"epoch": 4.506269592476489,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034171946052203217,
|
|
"loss": 4.8982,
|
|
"mean_token_accuracy": 0.20096374154090882,
|
|
"num_tokens": 10105403.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.462864255905151,
|
|
"epoch": 4.510188087774295,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034143213879495526,
|
|
"loss": 4.8252,
|
|
"mean_token_accuracy": 0.20886310636997224,
|
|
"num_tokens": 10115410.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.416613912582397,
|
|
"epoch": 4.514106583072101,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000341144698342875,
|
|
"loss": 4.8223,
|
|
"mean_token_accuracy": 0.20548174828290938,
|
|
"num_tokens": 10123800.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.452571201324463,
|
|
"epoch": 4.518025078369906,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003408571396794939,
|
|
"loss": 4.8698,
|
|
"mean_token_accuracy": 0.2006456285715103,
|
|
"num_tokens": 10132775.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.404480457305908,
|
|
"epoch": 4.521943573667712,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003405694633187259,
|
|
"loss": 4.751,
|
|
"mean_token_accuracy": 0.21099176704883577,
|
|
"num_tokens": 10141811.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.324165105819702,
|
|
"epoch": 4.525862068965517,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00034028166977469545,
|
|
"loss": 4.7642,
|
|
"mean_token_accuracy": 0.20946829169988632,
|
|
"num_tokens": 10150823.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.417202186584473,
|
|
"epoch": 4.529780564263323,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033999375956173607,
|
|
"loss": 4.7969,
|
|
"mean_token_accuracy": 0.2077748015522957,
|
|
"num_tokens": 10158983.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.4608917236328125,
|
|
"epoch": 4.533699059561129,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003397057331943902,
|
|
"loss": 4.7856,
|
|
"mean_token_accuracy": 0.2089831918478012,
|
|
"num_tokens": 10167596.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.42900595664978,
|
|
"epoch": 4.537617554858934,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003394175911874076,
|
|
"loss": 4.8835,
|
|
"mean_token_accuracy": 0.2114402547478676,
|
|
"num_tokens": 10176216.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.38248987197876,
|
|
"epoch": 4.54153605015674,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 4.7639,
|
|
"mean_token_accuracy": 0.2099734291434288,
|
|
"num_tokens": 10185685.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.361135387420655,
|
|
"epoch": 4.545454545454545,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003388409623145637,
|
|
"loss": 4.7645,
|
|
"mean_token_accuracy": 0.21032024919986725,
|
|
"num_tokens": 10193679.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.3993510723114015,
|
|
"epoch": 4.549373040752351,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00033855247647923177,
|
|
"loss": 4.8293,
|
|
"mean_token_accuracy": 0.2105245992541313,
|
|
"num_tokens": 10201154.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.386172866821289,
|
|
"epoch": 4.553291536050157,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003382638770653198,
|
|
"loss": 4.7406,
|
|
"mean_token_accuracy": 0.21750179529190064,
|
|
"num_tokens": 10209180.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.454772186279297,
|
|
"epoch": 4.557210031347962,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003379751645886017,
|
|
"loss": 4.8511,
|
|
"mean_token_accuracy": 0.20382963567972184,
|
|
"num_tokens": 10217430.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.356022500991822,
|
|
"epoch": 4.561128526645768,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003376863395650537,
|
|
"loss": 4.8992,
|
|
"mean_token_accuracy": 0.20598388761281966,
|
|
"num_tokens": 10226453.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.5248339653015135,
|
|
"epoch": 4.565047021943574,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00033739740251085263,
|
|
"loss": 4.8969,
|
|
"mean_token_accuracy": 0.20562998950481415,
|
|
"num_tokens": 10235005.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.488627767562866,
|
|
"epoch": 4.568965517241379,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00033710835394237603,
|
|
"loss": 4.8508,
|
|
"mean_token_accuracy": 0.20806352347135543,
|
|
"num_tokens": 10243800.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.410926294326782,
|
|
"epoch": 4.572884012539185,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033681919437620066,
|
|
"loss": 4.7949,
|
|
"mean_token_accuracy": 0.2159759521484375,
|
|
"num_tokens": 10252370.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.399033451080323,
|
|
"epoch": 4.576802507836991,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003365299243291014,
|
|
"loss": 4.861,
|
|
"mean_token_accuracy": 0.20563817024230957,
|
|
"num_tokens": 10260777.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.412040090560913,
|
|
"epoch": 4.580721003134796,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033624054431805067,
|
|
"loss": 4.8813,
|
|
"mean_token_accuracy": 0.1979104220867157,
|
|
"num_tokens": 10270571.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.461255025863648,
|
|
"epoch": 4.584639498432602,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033595105486021763,
|
|
"loss": 4.854,
|
|
"mean_token_accuracy": 0.2091786354780197,
|
|
"num_tokens": 10279284.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.354441165924072,
|
|
"epoch": 4.588557993730408,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033566145647296675,
|
|
"loss": 4.7493,
|
|
"mean_token_accuracy": 0.20902735888957977,
|
|
"num_tokens": 10288610.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.377309942245484,
|
|
"epoch": 4.592476489028213,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033537174967385744,
|
|
"loss": 4.7483,
|
|
"mean_token_accuracy": 0.2120072141289711,
|
|
"num_tokens": 10296603.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.343906354904175,
|
|
"epoch": 4.596394984326019,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033508193498064254,
|
|
"loss": 4.7359,
|
|
"mean_token_accuracy": 0.21689988821744918,
|
|
"num_tokens": 10305199.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.3378440856933596,
|
|
"epoch": 4.600313479623824,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033479201291126807,
|
|
"loss": 4.6964,
|
|
"mean_token_accuracy": 0.2193103089928627,
|
|
"num_tokens": 10313967.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.428088569641114,
|
|
"epoch": 4.60423197492163,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003345019839838717,
|
|
"loss": 4.806,
|
|
"mean_token_accuracy": 0.21562581211328508,
|
|
"num_tokens": 10322532.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.341708564758301,
|
|
"epoch": 4.608150470219436,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003342118487167822,
|
|
"loss": 4.6701,
|
|
"mean_token_accuracy": 0.21567916572093965,
|
|
"num_tokens": 10330631.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.306028890609741,
|
|
"epoch": 4.612068965517241,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033392160762851813,
|
|
"loss": 4.6975,
|
|
"mean_token_accuracy": 0.21092793941497803,
|
|
"num_tokens": 10339229.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.422407865524292,
|
|
"epoch": 4.615987460815047,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003336312612377877,
|
|
"loss": 4.8298,
|
|
"mean_token_accuracy": 0.20483612567186354,
|
|
"num_tokens": 10348882.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.537596225738525,
|
|
"epoch": 4.619905956112853,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033334081006348683,
|
|
"loss": 4.9523,
|
|
"mean_token_accuracy": 0.19908885806798934,
|
|
"num_tokens": 10358154.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.424742126464844,
|
|
"epoch": 4.623824451410658,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003330502546246988,
|
|
"loss": 4.8674,
|
|
"mean_token_accuracy": 0.2078734427690506,
|
|
"num_tokens": 10367100.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.294308710098266,
|
|
"epoch": 4.627742946708464,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00033275959544069335,
|
|
"loss": 4.6675,
|
|
"mean_token_accuracy": 0.22616573870182038,
|
|
"num_tokens": 10375711.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.406526660919189,
|
|
"epoch": 4.631661442006269,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033246883303092564,
|
|
"loss": 4.7837,
|
|
"mean_token_accuracy": 0.21085017621517183,
|
|
"num_tokens": 10384513.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.466464471817017,
|
|
"epoch": 4.635579937304075,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00033217796791503503,
|
|
"loss": 4.8768,
|
|
"mean_token_accuracy": 0.20449079126119613,
|
|
"num_tokens": 10393564.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.448765611648559,
|
|
"epoch": 4.639498432601881,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003318870006128449,
|
|
"loss": 4.8676,
|
|
"mean_token_accuracy": 0.2089618369936943,
|
|
"num_tokens": 10402042.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.364637470245361,
|
|
"epoch": 4.643416927899686,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003315959316443608,
|
|
"loss": 4.7455,
|
|
"mean_token_accuracy": 0.21121509671211242,
|
|
"num_tokens": 10410847.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.398000621795655,
|
|
"epoch": 4.647335423197492,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003313047615297703,
|
|
"loss": 4.8519,
|
|
"mean_token_accuracy": 0.20589703917503357,
|
|
"num_tokens": 10420366.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.397233009338379,
|
|
"epoch": 4.651253918495298,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00033101349078944165,
|
|
"loss": 4.7642,
|
|
"mean_token_accuracy": 0.21116065680980683,
|
|
"num_tokens": 10428870.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.357349300384522,
|
|
"epoch": 4.655172413793103,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003307221199439227,
|
|
"loss": 4.8222,
|
|
"mean_token_accuracy": 0.20543840080499648,
|
|
"num_tokens": 10437714.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.318109178543091,
|
|
"epoch": 4.659090909090909,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033043064951394045,
|
|
"loss": 4.721,
|
|
"mean_token_accuracy": 0.21294464617967607,
|
|
"num_tokens": 10447198.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.397916412353515,
|
|
"epoch": 4.663009404388715,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00033013908002039994,
|
|
"loss": 4.7759,
|
|
"mean_token_accuracy": 0.20848720371723176,
|
|
"num_tokens": 10455790.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.437332725524902,
|
|
"epoch": 4.66692789968652,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032984741198438305,
|
|
"loss": 4.8779,
|
|
"mean_token_accuracy": 0.20919278711080552,
|
|
"num_tokens": 10464094.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.421352291107178,
|
|
"epoch": 4.670846394984326,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003295556459271478,
|
|
"loss": 4.862,
|
|
"mean_token_accuracy": 0.20602272301912308,
|
|
"num_tokens": 10473201.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.454134559631347,
|
|
"epoch": 4.674764890282132,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003292637823701276,
|
|
"loss": 4.8583,
|
|
"mean_token_accuracy": 0.2011372357606888,
|
|
"num_tokens": 10482617.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.3977789878845215,
|
|
"epoch": 4.678683385579937,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003289718218349298,
|
|
"loss": 4.731,
|
|
"mean_token_accuracy": 0.21324052065610885,
|
|
"num_tokens": 10491702.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.4839574813842775,
|
|
"epoch": 4.682601880877743,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003286797648433351,
|
|
"loss": 4.9339,
|
|
"mean_token_accuracy": 0.2025283694267273,
|
|
"num_tokens": 10501189.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.412153005599976,
|
|
"epoch": 4.686520376175548,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000328387611917297,
|
|
"loss": 4.8022,
|
|
"mean_token_accuracy": 0.20796003937721252,
|
|
"num_tokens": 10510272.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.469935417175293,
|
|
"epoch": 4.690438871473354,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003280953635789401,
|
|
"loss": 4.8698,
|
|
"mean_token_accuracy": 0.2106959730386734,
|
|
"num_tokens": 10518782.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.361209344863892,
|
|
"epoch": 4.69435736677116,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003278030203505594,
|
|
"loss": 4.7998,
|
|
"mean_token_accuracy": 0.2106838494539261,
|
|
"num_tokens": 10528262.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.3257811069488525,
|
|
"epoch": 4.698275862068965,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003275105827546197,
|
|
"loss": 4.7997,
|
|
"mean_token_accuracy": 0.20828653126955032,
|
|
"num_tokens": 10536399.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.4399096965789795,
|
|
"epoch": 4.702194357366771,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032721805131375446,
|
|
"loss": 4.8421,
|
|
"mean_token_accuracy": 0.2040537491440773,
|
|
"num_tokens": 10544842.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 4.702194357366771,
|
|
"eval_entropy": 5.266712634138359,
|
|
"eval_loss": 5.6311845779418945,
|
|
"eval_mean_token_accuracy": 0.1715783413008664,
|
|
"eval_num_tokens": 10544842.0,
|
|
"eval_runtime": 2.3789,
|
|
"eval_samples_per_second": 1732.736,
|
|
"eval_steps_per_second": 216.907,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.417787408828735,
|
|
"epoch": 4.706112852664576,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003269254265507647,
|
|
"loss": 4.7702,
|
|
"mean_token_accuracy": 0.21643482744693757,
|
|
"num_tokens": 10553630.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.338301801681519,
|
|
"epoch": 4.710031347962382,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032663270898861845,
|
|
"loss": 4.8111,
|
|
"mean_token_accuracy": 0.2047191381454468,
|
|
"num_tokens": 10561714.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.377836418151856,
|
|
"epoch": 4.713949843260188,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032633989915044944,
|
|
"loss": 4.8705,
|
|
"mean_token_accuracy": 0.20156994611024856,
|
|
"num_tokens": 10571075.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.452186059951782,
|
|
"epoch": 4.717868338557993,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00032604699755955615,
|
|
"loss": 4.758,
|
|
"mean_token_accuracy": 0.2203052446246147,
|
|
"num_tokens": 10579698.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.383743858337402,
|
|
"epoch": 4.721786833855799,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032575400473940135,
|
|
"loss": 4.7792,
|
|
"mean_token_accuracy": 0.21169604659080504,
|
|
"num_tokens": 10588060.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.302184009552002,
|
|
"epoch": 4.725705329153605,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 4.8209,
|
|
"mean_token_accuracy": 0.20800677239894866,
|
|
"num_tokens": 10596482.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.357185173034668,
|
|
"epoch": 4.72962382445141,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003251677475059721,
|
|
"loss": 4.8466,
|
|
"mean_token_accuracy": 0.20605479031801224,
|
|
"num_tokens": 10605630.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.447827386856079,
|
|
"epoch": 4.733542319749216,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032487448414043435,
|
|
"loss": 4.8304,
|
|
"mean_token_accuracy": 0.20992751717567443,
|
|
"num_tokens": 10614772.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.468483638763428,
|
|
"epoch": 4.737460815047022,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032458113164110665,
|
|
"loss": 4.8156,
|
|
"mean_token_accuracy": 0.2175578534603119,
|
|
"num_tokens": 10623583.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.374937963485718,
|
|
"epoch": 4.741379310344827,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003242876905322575,
|
|
"loss": 4.7863,
|
|
"mean_token_accuracy": 0.2208006948232651,
|
|
"num_tokens": 10632136.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.3827714920043945,
|
|
"epoch": 4.745297805642633,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003239941613383138,
|
|
"loss": 4.8854,
|
|
"mean_token_accuracy": 0.2027701660990715,
|
|
"num_tokens": 10641306.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.432511568069458,
|
|
"epoch": 4.749216300940439,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00032370054458385973,
|
|
"loss": 4.8326,
|
|
"mean_token_accuracy": 0.20660535395145416,
|
|
"num_tokens": 10650866.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.454938459396362,
|
|
"epoch": 4.753134796238244,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000323406840793636,
|
|
"loss": 4.8255,
|
|
"mean_token_accuracy": 0.21617360562086105,
|
|
"num_tokens": 10659252.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.360633850097656,
|
|
"epoch": 4.75705329153605,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000323113050492539,
|
|
"loss": 4.7355,
|
|
"mean_token_accuracy": 0.20969840586185456,
|
|
"num_tokens": 10667987.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.417354583740234,
|
|
"epoch": 4.7609717868338555,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00032281917420561967,
|
|
"loss": 4.9156,
|
|
"mean_token_accuracy": 0.200067038834095,
|
|
"num_tokens": 10676524.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.392318964004517,
|
|
"epoch": 4.764890282131661,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003225252124580825,
|
|
"loss": 4.8326,
|
|
"mean_token_accuracy": 0.2103235200047493,
|
|
"num_tokens": 10685940.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.427258110046386,
|
|
"epoch": 4.768808777429467,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032223116577528475,
|
|
"loss": 4.838,
|
|
"mean_token_accuracy": 0.20554986447095872,
|
|
"num_tokens": 10694522.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.424826335906983,
|
|
"epoch": 4.7727272727272725,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003219370346827356,
|
|
"loss": 4.7707,
|
|
"mean_token_accuracy": 0.21283948421478271,
|
|
"num_tokens": 10703868.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.33817687034607,
|
|
"epoch": 4.7766457680250785,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000321642819706095,
|
|
"loss": 4.7567,
|
|
"mean_token_accuracy": 0.21107603013515472,
|
|
"num_tokens": 10712044.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.467048788070679,
|
|
"epoch": 4.7805642633228835,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003213485213711729,
|
|
"loss": 4.9624,
|
|
"mean_token_accuracy": 0.1985192820429802,
|
|
"num_tokens": 10720806.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.344447231292724,
|
|
"epoch": 4.7844827586206895,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000321054140203928,
|
|
"loss": 4.6997,
|
|
"mean_token_accuracy": 0.22239683270454408,
|
|
"num_tokens": 10729238.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.338423061370849,
|
|
"epoch": 4.7884012539184955,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003207596767304672,
|
|
"loss": 4.7608,
|
|
"mean_token_accuracy": 0.21050128042697908,
|
|
"num_tokens": 10738175.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.455464124679565,
|
|
"epoch": 4.7923197492163006,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032046513147704456,
|
|
"loss": 4.84,
|
|
"mean_token_accuracy": 0.20800792425870895,
|
|
"num_tokens": 10747475.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.413299560546875,
|
|
"epoch": 4.7962382445141065,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032017050497006,
|
|
"loss": 4.7922,
|
|
"mean_token_accuracy": 0.208526511490345,
|
|
"num_tokens": 10756739.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.326919937133789,
|
|
"epoch": 4.8001567398119125,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031987579773605913,
|
|
"loss": 4.7901,
|
|
"mean_token_accuracy": 0.2094557985663414,
|
|
"num_tokens": 10765141.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.44666166305542,
|
|
"epoch": 4.804075235109718,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003195810103017314,
|
|
"loss": 4.902,
|
|
"mean_token_accuracy": 0.20108154267072678,
|
|
"num_tokens": 10774759.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.453343152999878,
|
|
"epoch": 4.8079937304075235,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003192861431939098,
|
|
"loss": 4.8589,
|
|
"mean_token_accuracy": 0.20770366936922074,
|
|
"num_tokens": 10783717.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.383384466171265,
|
|
"epoch": 4.8119122257053295,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003189911969395697,
|
|
"loss": 4.8508,
|
|
"mean_token_accuracy": 0.21091044843196868,
|
|
"num_tokens": 10792797.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.326634073257447,
|
|
"epoch": 4.815830721003135,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00031869617206582777,
|
|
"loss": 4.7119,
|
|
"mean_token_accuracy": 0.22026228606700898,
|
|
"num_tokens": 10802249.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.38373875617981,
|
|
"epoch": 4.8197492163009406,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003184010690999413,
|
|
"loss": 4.8081,
|
|
"mean_token_accuracy": 0.2047825500369072,
|
|
"num_tokens": 10810722.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.472630882263184,
|
|
"epoch": 4.8236677115987465,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003181058885693073,
|
|
"loss": 4.8946,
|
|
"mean_token_accuracy": 0.2033669099211693,
|
|
"num_tokens": 10819115.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.4454460620880125,
|
|
"epoch": 4.827586206896552,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003178106310014612,
|
|
"loss": 4.8772,
|
|
"mean_token_accuracy": 0.2043038785457611,
|
|
"num_tokens": 10827419.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.288236808776856,
|
|
"epoch": 4.831504702194358,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003175152969240759,
|
|
"loss": 4.7667,
|
|
"mean_token_accuracy": 0.21056595742702483,
|
|
"num_tokens": 10836610.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.352677297592163,
|
|
"epoch": 4.835423197492163,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00031721988686496156,
|
|
"loss": 4.8426,
|
|
"mean_token_accuracy": 0.2094900503754616,
|
|
"num_tokens": 10845127.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.413989877700805,
|
|
"epoch": 4.839341692789969,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003169244013520638,
|
|
"loss": 4.8443,
|
|
"mean_token_accuracy": 0.21154511272907256,
|
|
"num_tokens": 10853458.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.48058500289917,
|
|
"epoch": 4.843260188087775,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000316628840913463,
|
|
"loss": 4.9444,
|
|
"mean_token_accuracy": 0.19947076886892318,
|
|
"num_tokens": 10862134.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.447872591018677,
|
|
"epoch": 4.84717868338558,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003163332060773738,
|
|
"loss": 4.9425,
|
|
"mean_token_accuracy": 0.20387972593307496,
|
|
"num_tokens": 10872127.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.385355424880982,
|
|
"epoch": 4.851097178683386,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003160374973721434,
|
|
"loss": 4.8268,
|
|
"mean_token_accuracy": 0.20640923827886581,
|
|
"num_tokens": 10880617.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.446012544631958,
|
|
"epoch": 4.855015673981192,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003157417153262513,
|
|
"loss": 4.8445,
|
|
"mean_token_accuracy": 0.20715817511081697,
|
|
"num_tokens": 10888769.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.439696455001831,
|
|
"epoch": 4.858934169278997,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00031544586046830796,
|
|
"loss": 4.7896,
|
|
"mean_token_accuracy": 0.2070804566144943,
|
|
"num_tokens": 10897638.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.386954259872437,
|
|
"epoch": 4.862852664576803,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000315149933327054,
|
|
"loss": 4.7554,
|
|
"mean_token_accuracy": 0.2105443671345711,
|
|
"num_tokens": 10906224.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.279257345199585,
|
|
"epoch": 4.866771159874608,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00031485393443135914,
|
|
"loss": 4.75,
|
|
"mean_token_accuracy": 0.21182352900505066,
|
|
"num_tokens": 10915390.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.397027158737183,
|
|
"epoch": 4.870689655172414,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003145578643102214,
|
|
"loss": 4.8143,
|
|
"mean_token_accuracy": 0.20802035480737685,
|
|
"num_tokens": 10923719.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.362337684631347,
|
|
"epoch": 4.87460815047022,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003142617234927662,
|
|
"loss": 4.7925,
|
|
"mean_token_accuracy": 0.21180858165025712,
|
|
"num_tokens": 10932610.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.399885082244873,
|
|
"epoch": 4.878526645768025,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031396551250824513,
|
|
"loss": 4.7918,
|
|
"mean_token_accuracy": 0.20950356423854827,
|
|
"num_tokens": 10941315.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.285400629043579,
|
|
"epoch": 4.882445141065831,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003136692318860352,
|
|
"loss": 4.7446,
|
|
"mean_token_accuracy": 0.21103135645389556,
|
|
"num_tokens": 10949781.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.462185287475586,
|
|
"epoch": 4.886363636363637,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003133728821556381,
|
|
"loss": 4.9248,
|
|
"mean_token_accuracy": 0.20642747730016708,
|
|
"num_tokens": 10958207.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.400398683547974,
|
|
"epoch": 4.890282131661442,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031307646384667854,
|
|
"loss": 4.8199,
|
|
"mean_token_accuracy": 0.20184200555086135,
|
|
"num_tokens": 10967162.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.475897216796875,
|
|
"epoch": 4.894200626959248,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031277997748890433,
|
|
"loss": 4.897,
|
|
"mean_token_accuracy": 0.20230586975812911,
|
|
"num_tokens": 10975517.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.4033849239349365,
|
|
"epoch": 4.898119122257054,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003124834236121847,
|
|
"loss": 4.7873,
|
|
"mean_token_accuracy": 0.21263960301876067,
|
|
"num_tokens": 10984668.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.321411085128784,
|
|
"epoch": 4.902037617554859,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00031218680274650934,
|
|
"loss": 4.7802,
|
|
"mean_token_accuracy": 0.20472333580255508,
|
|
"num_tokens": 10993475.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.318469667434693,
|
|
"epoch": 4.905956112852665,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031189011542198794,
|
|
"loss": 4.7345,
|
|
"mean_token_accuracy": 0.22124385088682175,
|
|
"num_tokens": 11002419.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.33527774810791,
|
|
"epoch": 4.90987460815047,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 4.7995,
|
|
"mean_token_accuracy": 0.21386657506227494,
|
|
"num_tokens": 11011271.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.461308813095092,
|
|
"epoch": 4.913793103448276,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00031129654351743816,
|
|
"loss": 4.8485,
|
|
"mean_token_accuracy": 0.20318652391433717,
|
|
"num_tokens": 11020461.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.433956909179687,
|
|
"epoch": 4.917711598746082,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00031099965999821906,
|
|
"loss": 4.7603,
|
|
"mean_token_accuracy": 0.21618503481149673,
|
|
"num_tokens": 11029251.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.3851746082305905,
|
|
"epoch": 4.921630094043887,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003107027121417704,
|
|
"loss": 4.7707,
|
|
"mean_token_accuracy": 0.21549003571271896,
|
|
"num_tokens": 11038149.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.432374334335327,
|
|
"epoch": 4.925548589341693,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031040570047878634,
|
|
"loss": 4.8522,
|
|
"mean_token_accuracy": 0.20755261033773423,
|
|
"num_tokens": 11046262.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.422052240371704,
|
|
"epoch": 4.929467084639499,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00031010862554007473,
|
|
"loss": 4.9286,
|
|
"mean_token_accuracy": 0.19925179481506347,
|
|
"num_tokens": 11055755.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.36035795211792,
|
|
"epoch": 4.933385579937304,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003098114878565567,
|
|
"loss": 4.7912,
|
|
"mean_token_accuracy": 0.20568819642066954,
|
|
"num_tokens": 11064534.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.367782497406006,
|
|
"epoch": 4.93730407523511,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00030951428795926536,
|
|
"loss": 4.8619,
|
|
"mean_token_accuracy": 0.20967960357666016,
|
|
"num_tokens": 11073768.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.4065882682800295,
|
|
"epoch": 4.941222570532915,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003092170263793452,
|
|
"loss": 4.8186,
|
|
"mean_token_accuracy": 0.2131222143769264,
|
|
"num_tokens": 11083683.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.455822610855103,
|
|
"epoch": 4.945141065830721,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00030891970364805096,
|
|
"loss": 4.8737,
|
|
"mean_token_accuracy": 0.20590061843395233,
|
|
"num_tokens": 11092368.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.313076686859131,
|
|
"epoch": 4.949059561128527,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003086223202967464,
|
|
"loss": 4.7911,
|
|
"mean_token_accuracy": 0.2104412794113159,
|
|
"num_tokens": 11100544.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.3882715702056885,
|
|
"epoch": 4.952978056426332,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000308324876856904,
|
|
"loss": 4.8278,
|
|
"mean_token_accuracy": 0.2042338952422142,
|
|
"num_tokens": 11108962.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.466433429718018,
|
|
"epoch": 4.956896551724138,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030802737386010314,
|
|
"loss": 4.8876,
|
|
"mean_token_accuracy": 0.2065508782863617,
|
|
"num_tokens": 11117011.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.400778961181641,
|
|
"epoch": 4.960815047021944,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003077298118380301,
|
|
"loss": 4.8236,
|
|
"mean_token_accuracy": 0.2096264973282814,
|
|
"num_tokens": 11126009.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.406397151947021,
|
|
"epoch": 4.964733542319749,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003074321913224765,
|
|
"loss": 4.7214,
|
|
"mean_token_accuracy": 0.21518404185771942,
|
|
"num_tokens": 11134404.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.419449090957642,
|
|
"epoch": 4.968652037617555,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003071345128453382,
|
|
"loss": 4.8569,
|
|
"mean_token_accuracy": 0.20666337609291077,
|
|
"num_tokens": 11142245.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.332555294036865,
|
|
"epoch": 4.972570532915361,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003068367769386153,
|
|
"loss": 4.8218,
|
|
"mean_token_accuracy": 0.20983357578516007,
|
|
"num_tokens": 11150881.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.345796871185303,
|
|
"epoch": 4.976489028213166,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00030653898413440973,
|
|
"loss": 4.7705,
|
|
"mean_token_accuracy": 0.21507095992565156,
|
|
"num_tokens": 11159421.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.414087152481079,
|
|
"epoch": 4.980407523510972,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003062411349649258,
|
|
"loss": 4.8411,
|
|
"mean_token_accuracy": 0.2083553895354271,
|
|
"num_tokens": 11168633.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.356377077102661,
|
|
"epoch": 4.984326018808778,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030594322996246816,
|
|
"loss": 4.6987,
|
|
"mean_token_accuracy": 0.2210353285074234,
|
|
"num_tokens": 11177435.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.464932250976562,
|
|
"epoch": 4.988244514106583,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00030564526965944127,
|
|
"loss": 4.9014,
|
|
"mean_token_accuracy": 0.20801532119512559,
|
|
"num_tokens": 11186195.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.376215887069702,
|
|
"epoch": 4.992163009404389,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003053472545883488,
|
|
"loss": 4.7569,
|
|
"mean_token_accuracy": 0.21403202116489412,
|
|
"num_tokens": 11194087.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.399469614028931,
|
|
"epoch": 4.996081504702194,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00030504918528179164,
|
|
"loss": 4.8469,
|
|
"mean_token_accuracy": 0.2078300431370735,
|
|
"num_tokens": 11202677.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.402387285232544,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00030475106227246824,
|
|
"loss": 4.8641,
|
|
"mean_token_accuracy": 0.1977234423160553,
|
|
"num_tokens": 11211560.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.389865684509277,
|
|
"epoch": 5.003918495297806,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003044528860931726,
|
|
"loss": 4.6246,
|
|
"mean_token_accuracy": 0.2284930855035782,
|
|
"num_tokens": 11220186.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.435016202926636,
|
|
"epoch": 5.007836990595611,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003041546572767939,
|
|
"loss": 4.6418,
|
|
"mean_token_accuracy": 0.22056833207607268,
|
|
"num_tokens": 11229148.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.375746393203736,
|
|
"epoch": 5.011755485893417,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003038563763563156,
|
|
"loss": 4.5574,
|
|
"mean_token_accuracy": 0.23293885886669158,
|
|
"num_tokens": 11237585.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.316483497619629,
|
|
"epoch": 5.015673981191223,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00030355804386481374,
|
|
"loss": 4.5253,
|
|
"mean_token_accuracy": 0.2259256824851036,
|
|
"num_tokens": 11246203.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.377748489379883,
|
|
"epoch": 5.019592476489028,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00030325966033545716,
|
|
"loss": 4.6429,
|
|
"mean_token_accuracy": 0.22607138454914094,
|
|
"num_tokens": 11254999.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.3769059658050535,
|
|
"epoch": 5.023510971786834,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00030296122630150534,
|
|
"loss": 4.6035,
|
|
"mean_token_accuracy": 0.22522027790546417,
|
|
"num_tokens": 11264110.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.405582046508789,
|
|
"epoch": 5.027429467084639,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030266274229630844,
|
|
"loss": 4.5381,
|
|
"mean_token_accuracy": 0.23742579221725463,
|
|
"num_tokens": 11273452.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.275386714935303,
|
|
"epoch": 5.031347962382445,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00030236420885330586,
|
|
"loss": 4.5512,
|
|
"mean_token_accuracy": 0.2311268627643585,
|
|
"num_tokens": 11281926.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.3661580085754395,
|
|
"epoch": 5.035266457680251,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003020656265060251,
|
|
"loss": 4.5984,
|
|
"mean_token_accuracy": 0.23923658579587936,
|
|
"num_tokens": 11290630.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.409304666519165,
|
|
"epoch": 5.039184952978056,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003017669957880813,
|
|
"loss": 4.6223,
|
|
"mean_token_accuracy": 0.22721641957759858,
|
|
"num_tokens": 11299126.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.401488256454468,
|
|
"epoch": 5.043103448275862,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000301468317233176,
|
|
"loss": 4.6987,
|
|
"mean_token_accuracy": 0.21642495542764664,
|
|
"num_tokens": 11308331.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.360465335845947,
|
|
"epoch": 5.047021943573668,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00030116959137509614,
|
|
"loss": 4.5887,
|
|
"mean_token_accuracy": 0.222395159304142,
|
|
"num_tokens": 11316245.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.427735376358032,
|
|
"epoch": 5.050940438871473,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00030087081874771325,
|
|
"loss": 4.6053,
|
|
"mean_token_accuracy": 0.22494175285100937,
|
|
"num_tokens": 11324994.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.301576375961304,
|
|
"epoch": 5.054858934169279,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00030057199988498245,
|
|
"loss": 4.5592,
|
|
"mean_token_accuracy": 0.22560292035341262,
|
|
"num_tokens": 11333255.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.309249401092529,
|
|
"epoch": 5.058777429467085,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003002731353209416,
|
|
"loss": 4.5963,
|
|
"mean_token_accuracy": 0.22171622961759568,
|
|
"num_tokens": 11341857.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.391611337661743,
|
|
"epoch": 5.06269592476489,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002999742255897102,
|
|
"loss": 4.6846,
|
|
"mean_token_accuracy": 0.2169702649116516,
|
|
"num_tokens": 11350933.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.353989887237549,
|
|
"epoch": 5.066614420062696,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00029967527122548826,
|
|
"loss": 4.6017,
|
|
"mean_token_accuracy": 0.22796097099781037,
|
|
"num_tokens": 11360346.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.393879747390747,
|
|
"epoch": 5.070532915360501,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002993762727625556,
|
|
"loss": 4.6303,
|
|
"mean_token_accuracy": 0.22346230447292328,
|
|
"num_tokens": 11369401.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.389891910552978,
|
|
"epoch": 5.074451410658307,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002990772307352712,
|
|
"loss": 4.6027,
|
|
"mean_token_accuracy": 0.22480421811342238,
|
|
"num_tokens": 11377882.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.384570741653443,
|
|
"epoch": 5.078369905956113,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002987781456780715,
|
|
"loss": 4.6134,
|
|
"mean_token_accuracy": 0.22593874037265776,
|
|
"num_tokens": 11386934.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.3440940380096436,
|
|
"epoch": 5.082288401253918,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00029847901812547,
|
|
"loss": 4.6391,
|
|
"mean_token_accuracy": 0.2219812735915184,
|
|
"num_tokens": 11395929.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.370107984542846,
|
|
"epoch": 5.086206896551724,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00029817984861205626,
|
|
"loss": 4.5859,
|
|
"mean_token_accuracy": 0.230423179268837,
|
|
"num_tokens": 11405664.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.365076684951783,
|
|
"epoch": 5.09012539184953,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002978806376724945,
|
|
"loss": 4.6829,
|
|
"mean_token_accuracy": 0.22319872975349425,
|
|
"num_tokens": 11414667.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.393854665756225,
|
|
"epoch": 5.094043887147335,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 4.6516,
|
|
"mean_token_accuracy": 0.22110701501369476,
|
|
"num_tokens": 11423907.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.094043887147335,
|
|
"eval_entropy": 5.188096245129903,
|
|
"eval_loss": 5.632938861846924,
|
|
"eval_mean_token_accuracy": 0.17254990297405756,
|
|
"eval_num_tokens": 11423907.0,
|
|
"eval_runtime": 2.1806,
|
|
"eval_samples_per_second": 1890.285,
|
|
"eval_steps_per_second": 236.63,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.415448331832886,
|
|
"epoch": 5.097962382445141,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00029728209365395433,
|
|
"loss": 4.6883,
|
|
"mean_token_accuracy": 0.228867244720459,
|
|
"num_tokens": 11433824.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.3738768100738525,
|
|
"epoch": 5.101880877742946,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00029698276164467105,
|
|
"loss": 4.5506,
|
|
"mean_token_accuracy": 0.2299201563000679,
|
|
"num_tokens": 11442371.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.36923246383667,
|
|
"epoch": 5.105799373040752,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002966833903486284,
|
|
"loss": 4.5981,
|
|
"mean_token_accuracy": 0.22003730535507202,
|
|
"num_tokens": 11451657.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.2978898048400875,
|
|
"epoch": 5.109717868338558,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00029638398030085134,
|
|
"loss": 4.5283,
|
|
"mean_token_accuracy": 0.2281121701002121,
|
|
"num_tokens": 11460064.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.384819889068604,
|
|
"epoch": 5.113636363636363,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00029608453203643434,
|
|
"loss": 4.6339,
|
|
"mean_token_accuracy": 0.2210537776350975,
|
|
"num_tokens": 11469184.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.428171968460083,
|
|
"epoch": 5.117554858934169,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002957850460905398,
|
|
"loss": 4.5715,
|
|
"mean_token_accuracy": 0.2350159227848053,
|
|
"num_tokens": 11478069.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.405584526062012,
|
|
"epoch": 5.121473354231975,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00029548552299839774,
|
|
"loss": 4.6037,
|
|
"mean_token_accuracy": 0.2310909390449524,
|
|
"num_tokens": 11486572.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.346296119689941,
|
|
"epoch": 5.12539184952978,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002951859632953046,
|
|
"loss": 4.6298,
|
|
"mean_token_accuracy": 0.22505338490009308,
|
|
"num_tokens": 11495857.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.343290090560913,
|
|
"epoch": 5.129310344827586,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00029488636751662196,
|
|
"loss": 4.6175,
|
|
"mean_token_accuracy": 0.22580296993255616,
|
|
"num_tokens": 11504875.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.346730375289917,
|
|
"epoch": 5.133228840125392,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002945867361977762,
|
|
"loss": 4.6287,
|
|
"mean_token_accuracy": 0.21845702975988388,
|
|
"num_tokens": 11514509.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.356408357620239,
|
|
"epoch": 5.137147335423197,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00029428706987425705,
|
|
"loss": 4.6476,
|
|
"mean_token_accuracy": 0.22266778498888015,
|
|
"num_tokens": 11523046.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.352146577835083,
|
|
"epoch": 5.141065830721003,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002939873690816167,
|
|
"loss": 4.5679,
|
|
"mean_token_accuracy": 0.23100701570510865,
|
|
"num_tokens": 11531270.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.375174236297608,
|
|
"epoch": 5.144984326018808,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002936876343554692,
|
|
"loss": 4.6142,
|
|
"mean_token_accuracy": 0.22371602803468704,
|
|
"num_tokens": 11539722.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.3703666687011715,
|
|
"epoch": 5.148902821316614,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000293387866231489,
|
|
"loss": 4.656,
|
|
"mean_token_accuracy": 0.22880760282278062,
|
|
"num_tokens": 11549614.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.283947563171386,
|
|
"epoch": 5.15282131661442,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00029308806524541047,
|
|
"loss": 4.4878,
|
|
"mean_token_accuracy": 0.2296810805797577,
|
|
"num_tokens": 11557674.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.30511703491211,
|
|
"epoch": 5.156739811912225,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00029278823193302643,
|
|
"loss": 4.5557,
|
|
"mean_token_accuracy": 0.2276170000433922,
|
|
"num_tokens": 11566691.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.371000289916992,
|
|
"epoch": 5.160658307210031,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00029248836683018774,
|
|
"loss": 4.5715,
|
|
"mean_token_accuracy": 0.23312035351991653,
|
|
"num_tokens": 11574824.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.30404257774353,
|
|
"epoch": 5.164576802507837,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00029218847047280197,
|
|
"loss": 4.5715,
|
|
"mean_token_accuracy": 0.2318928673863411,
|
|
"num_tokens": 11582988.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.374068832397461,
|
|
"epoch": 5.168495297805642,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002918885433968323,
|
|
"loss": 4.6462,
|
|
"mean_token_accuracy": 0.22530801296234132,
|
|
"num_tokens": 11591553.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.3138810157775875,
|
|
"epoch": 5.172413793103448,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002915885861382974,
|
|
"loss": 4.65,
|
|
"mean_token_accuracy": 0.22298514991998672,
|
|
"num_tokens": 11600347.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.3492213726043705,
|
|
"epoch": 5.1763322884012535,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00029128859923326935,
|
|
"loss": 4.6537,
|
|
"mean_token_accuracy": 0.2246803104877472,
|
|
"num_tokens": 11609217.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.423777484893799,
|
|
"epoch": 5.1802507836990594,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00029098858321787336,
|
|
"loss": 4.6276,
|
|
"mean_token_accuracy": 0.22783687710762024,
|
|
"num_tokens": 11617355.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.348060894012451,
|
|
"epoch": 5.184169278996865,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00029068853862828677,
|
|
"loss": 4.6097,
|
|
"mean_token_accuracy": 0.22840888500213624,
|
|
"num_tokens": 11626152.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.291878652572632,
|
|
"epoch": 5.1880877742946705,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002903884660007378,
|
|
"loss": 4.579,
|
|
"mean_token_accuracy": 0.2272334650158882,
|
|
"num_tokens": 11634868.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.389139175415039,
|
|
"epoch": 5.1920062695924765,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00029008836587150505,
|
|
"loss": 4.6856,
|
|
"mean_token_accuracy": 0.22577418088912965,
|
|
"num_tokens": 11644199.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.466033315658569,
|
|
"epoch": 5.195924764890282,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002897882387769159,
|
|
"loss": 4.7176,
|
|
"mean_token_accuracy": 0.21752761900424958,
|
|
"num_tokens": 11653227.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.286968183517456,
|
|
"epoch": 5.1998432601880875,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002894880852533464,
|
|
"loss": 4.5215,
|
|
"mean_token_accuracy": 0.2373726025223732,
|
|
"num_tokens": 11661628.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.336953639984131,
|
|
"epoch": 5.2037617554858935,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028918790583721936,
|
|
"loss": 4.5799,
|
|
"mean_token_accuracy": 0.2297064870595932,
|
|
"num_tokens": 11670616.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.322612667083741,
|
|
"epoch": 5.2076802507836994,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002888877010650041,
|
|
"loss": 4.5975,
|
|
"mean_token_accuracy": 0.23119895160198212,
|
|
"num_tokens": 11679069.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.344280624389649,
|
|
"epoch": 5.2115987460815045,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028858747147321527,
|
|
"loss": 4.588,
|
|
"mean_token_accuracy": 0.23015287667512893,
|
|
"num_tokens": 11688184.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.355217409133911,
|
|
"epoch": 5.2155172413793105,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002882872175984118,
|
|
"loss": 4.6415,
|
|
"mean_token_accuracy": 0.22452577203512192,
|
|
"num_tokens": 11696389.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.342627716064453,
|
|
"epoch": 5.219435736677116,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002879869399771962,
|
|
"loss": 4.6614,
|
|
"mean_token_accuracy": 0.22327292561531067,
|
|
"num_tokens": 11704881.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.369431066513061,
|
|
"epoch": 5.2233542319749215,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00028768663914621295,
|
|
"loss": 4.6979,
|
|
"mean_token_accuracy": 0.2191631942987442,
|
|
"num_tokens": 11713566.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.3664812564849855,
|
|
"epoch": 5.2272727272727275,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002873863156421486,
|
|
"loss": 4.6432,
|
|
"mean_token_accuracy": 0.2292092725634575,
|
|
"num_tokens": 11722427.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.392437696456909,
|
|
"epoch": 5.231191222570533,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00028708597000172984,
|
|
"loss": 4.6011,
|
|
"mean_token_accuracy": 0.23083536028862,
|
|
"num_tokens": 11731010.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.3115325450897215,
|
|
"epoch": 5.235109717868339,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002867856027617229,
|
|
"loss": 4.6469,
|
|
"mean_token_accuracy": 0.22524809390306472,
|
|
"num_tokens": 11739569.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.3544494152069095,
|
|
"epoch": 5.2390282131661445,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00028648521445893304,
|
|
"loss": 4.6035,
|
|
"mean_token_accuracy": 0.2247149959206581,
|
|
"num_tokens": 11747687.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.365255737304688,
|
|
"epoch": 5.24294670846395,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002861848056302026,
|
|
"loss": 4.6751,
|
|
"mean_token_accuracy": 0.22379979640245437,
|
|
"num_tokens": 11756713.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.351660823822021,
|
|
"epoch": 5.246865203761756,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00028588437681241106,
|
|
"loss": 4.5793,
|
|
"mean_token_accuracy": 0.22728823274374008,
|
|
"num_tokens": 11765625.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.3822410106658936,
|
|
"epoch": 5.250783699059561,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00028558392854247333,
|
|
"loss": 4.6629,
|
|
"mean_token_accuracy": 0.21841179728507995,
|
|
"num_tokens": 11774594.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.347749853134156,
|
|
"epoch": 5.254702194357367,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002852834613573391,
|
|
"loss": 4.6184,
|
|
"mean_token_accuracy": 0.23327738642692566,
|
|
"num_tokens": 11782983.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.31268949508667,
|
|
"epoch": 5.258620689655173,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002849829757939922,
|
|
"loss": 4.5884,
|
|
"mean_token_accuracy": 0.22571429163217543,
|
|
"num_tokens": 11791940.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.364489507675171,
|
|
"epoch": 5.262539184952978,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002846824723894487,
|
|
"loss": 4.6187,
|
|
"mean_token_accuracy": 0.22800127267837525,
|
|
"num_tokens": 11800357.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.39718189239502,
|
|
"epoch": 5.266457680250784,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002843819516807572,
|
|
"loss": 4.6925,
|
|
"mean_token_accuracy": 0.21319834589958192,
|
|
"num_tokens": 11809005.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.425684261322021,
|
|
"epoch": 5.27037617554859,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00028408141420499685,
|
|
"loss": 4.6352,
|
|
"mean_token_accuracy": 0.2174714758992195,
|
|
"num_tokens": 11818021.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.389274740219117,
|
|
"epoch": 5.274294670846395,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002837808604992768,
|
|
"loss": 4.6475,
|
|
"mean_token_accuracy": 0.21976408213377,
|
|
"num_tokens": 11826791.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.3104870319366455,
|
|
"epoch": 5.278213166144201,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 4.5437,
|
|
"mean_token_accuracy": 0.23511412143707275,
|
|
"num_tokens": 11835110.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.331724071502686,
|
|
"epoch": 5.282131661442007,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002831797065465386,
|
|
"loss": 4.6723,
|
|
"mean_token_accuracy": 0.22211682200431823,
|
|
"num_tokens": 11842658.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.3184826374053955,
|
|
"epoch": 5.286050156739812,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028287910737388,
|
|
"loss": 4.6413,
|
|
"mean_token_accuracy": 0.22393565475940705,
|
|
"num_tokens": 11851028.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.3814304828643795,
|
|
"epoch": 5.289968652037618,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000282578494119979,
|
|
"loss": 4.6002,
|
|
"mean_token_accuracy": 0.23092781603336335,
|
|
"num_tokens": 11859471.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.334242963790894,
|
|
"epoch": 5.293887147335424,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00028227786732208014,
|
|
"loss": 4.5515,
|
|
"mean_token_accuracy": 0.2276996046304703,
|
|
"num_tokens": 11868625.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.306391811370849,
|
|
"epoch": 5.297805642633229,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00028197722751745246,
|
|
"loss": 4.5866,
|
|
"mean_token_accuracy": 0.22790590971708297,
|
|
"num_tokens": 11877744.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.340777397155762,
|
|
"epoch": 5.301724137931035,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002816765752433879,
|
|
"loss": 4.5927,
|
|
"mean_token_accuracy": 0.22989525496959687,
|
|
"num_tokens": 11886445.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.241041469573974,
|
|
"epoch": 5.30564263322884,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00028137591103720075,
|
|
"loss": 4.5939,
|
|
"mean_token_accuracy": 0.22844783812761307,
|
|
"num_tokens": 11896040.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.302602338790893,
|
|
"epoch": 5.309561128526646,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002810752354362267,
|
|
"loss": 4.6142,
|
|
"mean_token_accuracy": 0.21471603512763976,
|
|
"num_tokens": 11904996.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.384479236602783,
|
|
"epoch": 5.313479623824452,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000280774548977822,
|
|
"loss": 4.6263,
|
|
"mean_token_accuracy": 0.22510625272989274,
|
|
"num_tokens": 11913357.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.366365146636963,
|
|
"epoch": 5.317398119122257,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028047385219936196,
|
|
"loss": 4.666,
|
|
"mean_token_accuracy": 0.22223397493362426,
|
|
"num_tokens": 11922196.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.293893766403198,
|
|
"epoch": 5.321316614420063,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00028017314563824044,
|
|
"loss": 4.5292,
|
|
"mean_token_accuracy": 0.23568708449602127,
|
|
"num_tokens": 11930682.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.298173999786377,
|
|
"epoch": 5.325235109717869,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00027987242983186896,
|
|
"loss": 4.6285,
|
|
"mean_token_accuracy": 0.2226262152194977,
|
|
"num_tokens": 11939470.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.319783878326416,
|
|
"epoch": 5.329153605015674,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00027957170531767525,
|
|
"loss": 4.6227,
|
|
"mean_token_accuracy": 0.22119760066270827,
|
|
"num_tokens": 11948183.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.392630767822266,
|
|
"epoch": 5.33307210031348,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00027927097263310294,
|
|
"loss": 4.6477,
|
|
"mean_token_accuracy": 0.22783928364515305,
|
|
"num_tokens": 11957027.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.366418504714966,
|
|
"epoch": 5.336990595611285,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002789702323156099,
|
|
"loss": 4.605,
|
|
"mean_token_accuracy": 0.229417285323143,
|
|
"num_tokens": 11965871.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.393637752532959,
|
|
"epoch": 5.340909090909091,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00027866948490266815,
|
|
"loss": 4.7205,
|
|
"mean_token_accuracy": 0.21908856332302093,
|
|
"num_tokens": 11974384.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.369632863998413,
|
|
"epoch": 5.344827586206897,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002783687309317618,
|
|
"loss": 4.6541,
|
|
"mean_token_accuracy": 0.22262327820062638,
|
|
"num_tokens": 11983892.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.464978742599487,
|
|
"epoch": 5.348746081504702,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002780679709403871,
|
|
"loss": 4.6831,
|
|
"mean_token_accuracy": 0.22060687839984894,
|
|
"num_tokens": 11992990.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.3050856590271,
|
|
"epoch": 5.352664576802508,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00027776720546605086,
|
|
"loss": 4.4832,
|
|
"mean_token_accuracy": 0.24133117347955704,
|
|
"num_tokens": 12001679.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.313827323913574,
|
|
"epoch": 5.356583072100314,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002774664350462697,
|
|
"loss": 4.6536,
|
|
"mean_token_accuracy": 0.22566813826560975,
|
|
"num_tokens": 12011300.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.293299531936645,
|
|
"epoch": 5.360501567398119,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00027716566021856933,
|
|
"loss": 4.621,
|
|
"mean_token_accuracy": 0.22668158411979675,
|
|
"num_tokens": 12019776.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.313161611557007,
|
|
"epoch": 5.364420062695925,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000276864881520483,
|
|
"loss": 4.6502,
|
|
"mean_token_accuracy": 0.22606873214244844,
|
|
"num_tokens": 12028838.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.391641855239868,
|
|
"epoch": 5.368338557993731,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002765640994895509,
|
|
"loss": 4.6095,
|
|
"mean_token_accuracy": 0.22615873515605928,
|
|
"num_tokens": 12036903.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.363273429870605,
|
|
"epoch": 5.372257053291536,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00027626331466331955,
|
|
"loss": 4.6072,
|
|
"mean_token_accuracy": 0.22216324061155318,
|
|
"num_tokens": 12045838.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.4339336395263675,
|
|
"epoch": 5.376175548589342,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027596252757933995,
|
|
"loss": 4.7565,
|
|
"mean_token_accuracy": 0.21587950736284256,
|
|
"num_tokens": 12055791.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.309387874603272,
|
|
"epoch": 5.380094043887147,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002756617387751674,
|
|
"loss": 4.5487,
|
|
"mean_token_accuracy": 0.2296561196446419,
|
|
"num_tokens": 12064842.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.313955688476563,
|
|
"epoch": 5.384012539184953,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002753609487883606,
|
|
"loss": 4.6218,
|
|
"mean_token_accuracy": 0.22491351515054703,
|
|
"num_tokens": 12073487.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.251298809051514,
|
|
"epoch": 5.387931034482759,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027506015815647965,
|
|
"loss": 4.5767,
|
|
"mean_token_accuracy": 0.22127929478883743,
|
|
"num_tokens": 12082000.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.305486059188842,
|
|
"epoch": 5.391849529780564,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00027475936741708636,
|
|
"loss": 4.641,
|
|
"mean_token_accuracy": 0.222818985581398,
|
|
"num_tokens": 12090347.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.335762357711792,
|
|
"epoch": 5.39576802507837,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002744585771077425,
|
|
"loss": 4.5951,
|
|
"mean_token_accuracy": 0.23029457181692123,
|
|
"num_tokens": 12098340.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.356241369247437,
|
|
"epoch": 5.399686520376176,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00027415778776600913,
|
|
"loss": 4.6339,
|
|
"mean_token_accuracy": 0.22939420193433763,
|
|
"num_tokens": 12106771.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.3736272811889645,
|
|
"epoch": 5.403605015673981,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002738569999294456,
|
|
"loss": 4.6395,
|
|
"mean_token_accuracy": 0.2201326921582222,
|
|
"num_tokens": 12115436.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.301021766662598,
|
|
"epoch": 5.407523510971787,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002735562141356085,
|
|
"loss": 4.5273,
|
|
"mean_token_accuracy": 0.2356494426727295,
|
|
"num_tokens": 12123939.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.447571277618408,
|
|
"epoch": 5.411442006269592,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002732554309220509,
|
|
"loss": 4.7616,
|
|
"mean_token_accuracy": 0.2201945424079895,
|
|
"num_tokens": 12132803.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.298455095291137,
|
|
"epoch": 5.415360501567398,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002729546508263211,
|
|
"loss": 4.6609,
|
|
"mean_token_accuracy": 0.22638544142246247,
|
|
"num_tokens": 12141875.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.370645380020141,
|
|
"epoch": 5.419278996865204,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002726538743859618,
|
|
"loss": 4.6304,
|
|
"mean_token_accuracy": 0.2304920569062233,
|
|
"num_tokens": 12150685.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.346543407440185,
|
|
"epoch": 5.423197492163009,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002723531021385095,
|
|
"loss": 4.5933,
|
|
"mean_token_accuracy": 0.22665177285671234,
|
|
"num_tokens": 12159070.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.325288438796997,
|
|
"epoch": 5.427115987460815,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002720523346214928,
|
|
"loss": 4.6097,
|
|
"mean_token_accuracy": 0.22591610997915268,
|
|
"num_tokens": 12167676.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.312070369720459,
|
|
"epoch": 5.431034482758621,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00027175157237243204,
|
|
"loss": 4.6232,
|
|
"mean_token_accuracy": 0.22277018427848816,
|
|
"num_tokens": 12176222.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.342537546157837,
|
|
"epoch": 5.434952978056426,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002714508159288382,
|
|
"loss": 4.618,
|
|
"mean_token_accuracy": 0.23263876140117645,
|
|
"num_tokens": 12185171.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.380558967590332,
|
|
"epoch": 5.438871473354232,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002711500658282118,
|
|
"loss": 4.6659,
|
|
"mean_token_accuracy": 0.22947559356689454,
|
|
"num_tokens": 12193875.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.284651756286621,
|
|
"epoch": 5.442789968652038,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00027084932260804193,
|
|
"loss": 4.5991,
|
|
"mean_token_accuracy": 0.2258966699242592,
|
|
"num_tokens": 12202681.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.331091737747192,
|
|
"epoch": 5.446708463949843,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002705485868058056,
|
|
"loss": 4.5891,
|
|
"mean_token_accuracy": 0.23223180323839188,
|
|
"num_tokens": 12210854.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.405281019210816,
|
|
"epoch": 5.450626959247649,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027024785895896644,
|
|
"loss": 4.6457,
|
|
"mean_token_accuracy": 0.22302116602659225,
|
|
"num_tokens": 12219643.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.360724020004272,
|
|
"epoch": 5.454545454545454,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00026994713960497383,
|
|
"loss": 4.6413,
|
|
"mean_token_accuracy": 0.22722207754850388,
|
|
"num_tokens": 12228767.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.343755722045898,
|
|
"epoch": 5.45846394984326,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00026964642928126197,
|
|
"loss": 4.6927,
|
|
"mean_token_accuracy": 0.2221220090985298,
|
|
"num_tokens": 12237468.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.29088282585144,
|
|
"epoch": 5.462382445141066,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 4.6624,
|
|
"mean_token_accuracy": 0.2228047803044319,
|
|
"num_tokens": 12246040.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.403284025192261,
|
|
"epoch": 5.466300940438871,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00026904503787433614,
|
|
"loss": 4.7445,
|
|
"mean_token_accuracy": 0.21147206276655198,
|
|
"num_tokens": 12254646.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.356195783615112,
|
|
"epoch": 5.470219435736677,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000268744357865906,
|
|
"loss": 4.632,
|
|
"mean_token_accuracy": 0.22317475229501724,
|
|
"num_tokens": 12263306.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.308737277984619,
|
|
"epoch": 5.474137931034483,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026844368903732263,
|
|
"loss": 4.6375,
|
|
"mean_token_accuracy": 0.2262572392821312,
|
|
"num_tokens": 12272498.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.265390872955322,
|
|
"epoch": 5.478056426332288,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026814303192593015,
|
|
"loss": 4.5081,
|
|
"mean_token_accuracy": 0.23592832684516907,
|
|
"num_tokens": 12281592.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.348323917388916,
|
|
"epoch": 5.481974921630094,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026784238706905136,
|
|
"loss": 4.6489,
|
|
"mean_token_accuracy": 0.22896516621112822,
|
|
"num_tokens": 12290112.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.402437305450439,
|
|
"epoch": 5.485893416927899,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002675417550039874,
|
|
"loss": 4.6939,
|
|
"mean_token_accuracy": 0.2223354160785675,
|
|
"num_tokens": 12299868.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 5.485893416927899,
|
|
"eval_entropy": 5.1746898529141445,
|
|
"eval_loss": 5.62327241897583,
|
|
"eval_mean_token_accuracy": 0.17388435450288675,
|
|
"eval_num_tokens": 12299868.0,
|
|
"eval_runtime": 2.3286,
|
|
"eval_samples_per_second": 1770.129,
|
|
"eval_steps_per_second": 221.588,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.358647775650025,
|
|
"epoch": 5.489811912225705,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026724113626801655,
|
|
"loss": 4.6141,
|
|
"mean_token_accuracy": 0.22352276295423507,
|
|
"num_tokens": 12307903.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.432712507247925,
|
|
"epoch": 5.493730407523511,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002669405313983932,
|
|
"loss": 4.7111,
|
|
"mean_token_accuracy": 0.22073129713535308,
|
|
"num_tokens": 12317287.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.449434614181518,
|
|
"epoch": 5.497648902821316,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000266639940932347,
|
|
"loss": 4.7169,
|
|
"mean_token_accuracy": 0.22032354176044464,
|
|
"num_tokens": 12325770.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.361948919296265,
|
|
"epoch": 5.501567398119122,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002663393654070819,
|
|
"loss": 4.6416,
|
|
"mean_token_accuracy": 0.2187054932117462,
|
|
"num_tokens": 12334800.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.364303636550903,
|
|
"epoch": 5.505485893416928,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002660388053597752,
|
|
"loss": 4.6734,
|
|
"mean_token_accuracy": 0.2125335857272148,
|
|
"num_tokens": 12343362.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.352037715911865,
|
|
"epoch": 5.509404388714733,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00026573826132757623,
|
|
"loss": 4.6736,
|
|
"mean_token_accuracy": 0.21591694951057433,
|
|
"num_tokens": 12352277.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.298210191726684,
|
|
"epoch": 5.513322884012539,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026543773384760593,
|
|
"loss": 4.531,
|
|
"mean_token_accuracy": 0.24023232460021973,
|
|
"num_tokens": 12361244.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.340992212295532,
|
|
"epoch": 5.517241379310345,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002651372234569557,
|
|
"loss": 4.6244,
|
|
"mean_token_accuracy": 0.21898644417524338,
|
|
"num_tokens": 12370228.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.339789581298828,
|
|
"epoch": 5.52115987460815,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002648367306926864,
|
|
"loss": 4.6262,
|
|
"mean_token_accuracy": 0.22642533779144286,
|
|
"num_tokens": 12379392.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.332865571975708,
|
|
"epoch": 5.525078369905956,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002645362560918271,
|
|
"loss": 4.6681,
|
|
"mean_token_accuracy": 0.2227002814412117,
|
|
"num_tokens": 12387334.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.312329530715942,
|
|
"epoch": 5.528996865203762,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000264235800191375,
|
|
"loss": 4.5801,
|
|
"mean_token_accuracy": 0.22639379501342774,
|
|
"num_tokens": 12396875.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.394218969345093,
|
|
"epoch": 5.532915360501567,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002639353635282932,
|
|
"loss": 4.6063,
|
|
"mean_token_accuracy": 0.23431526720523835,
|
|
"num_tokens": 12405123.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.445225858688355,
|
|
"epoch": 5.536833855799373,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026363494663951076,
|
|
"loss": 4.7232,
|
|
"mean_token_accuracy": 0.21638301759958267,
|
|
"num_tokens": 12413208.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.35152645111084,
|
|
"epoch": 5.540752351097178,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026333455006192146,
|
|
"loss": 4.7181,
|
|
"mean_token_accuracy": 0.21843911409378053,
|
|
"num_tokens": 12421704.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.36061110496521,
|
|
"epoch": 5.544670846394984,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00026303417433238267,
|
|
"loss": 4.6761,
|
|
"mean_token_accuracy": 0.21920190006494522,
|
|
"num_tokens": 12430610.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.341520309448242,
|
|
"epoch": 5.54858934169279,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002627338199877146,
|
|
"loss": 4.574,
|
|
"mean_token_accuracy": 0.22941515296697618,
|
|
"num_tokens": 12439867.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.332852077484131,
|
|
"epoch": 5.552507836990595,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002624334875646989,
|
|
"loss": 4.6539,
|
|
"mean_token_accuracy": 0.22086915522813796,
|
|
"num_tokens": 12449540.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.314187622070312,
|
|
"epoch": 5.556426332288401,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026213317760007864,
|
|
"loss": 4.6754,
|
|
"mean_token_accuracy": 0.22325518429279329,
|
|
"num_tokens": 12458138.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.4030293941497805,
|
|
"epoch": 5.560344827586206,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002618328906305562,
|
|
"loss": 4.7447,
|
|
"mean_token_accuracy": 0.21491104364395142,
|
|
"num_tokens": 12467921.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.416860151290893,
|
|
"epoch": 5.564263322884012,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002615326271927931,
|
|
"loss": 4.6169,
|
|
"mean_token_accuracy": 0.2232373520731926,
|
|
"num_tokens": 12476873.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.425081062316894,
|
|
"epoch": 5.568181818181818,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026123238782340893,
|
|
"loss": 4.6371,
|
|
"mean_token_accuracy": 0.2242217943072319,
|
|
"num_tokens": 12485924.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.345165205001831,
|
|
"epoch": 5.572100313479623,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00026093217305898003,
|
|
"loss": 4.5834,
|
|
"mean_token_accuracy": 0.22803157269954683,
|
|
"num_tokens": 12494533.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.245182657241822,
|
|
"epoch": 5.576018808777429,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00026063198343603887,
|
|
"loss": 4.5329,
|
|
"mean_token_accuracy": 0.23541837632656099,
|
|
"num_tokens": 12503137.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.273105525970459,
|
|
"epoch": 5.579937304075235,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00026033181949107286,
|
|
"loss": 4.6442,
|
|
"mean_token_accuracy": 0.22728424072265624,
|
|
"num_tokens": 12511701.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.30779767036438,
|
|
"epoch": 5.58385579937304,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002600316817605239,
|
|
"loss": 4.6475,
|
|
"mean_token_accuracy": 0.2295583739876747,
|
|
"num_tokens": 12520097.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.339383554458618,
|
|
"epoch": 5.587774294670846,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002597315707807866,
|
|
"loss": 4.6498,
|
|
"mean_token_accuracy": 0.22568415105342865,
|
|
"num_tokens": 12528526.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.392545223236084,
|
|
"epoch": 5.591692789968652,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00025943148708820784,
|
|
"loss": 4.6641,
|
|
"mean_token_accuracy": 0.22573817074298858,
|
|
"num_tokens": 12537888.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.342211151123047,
|
|
"epoch": 5.5956112852664575,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002591314312190861,
|
|
"loss": 4.6456,
|
|
"mean_token_accuracy": 0.22285591214895248,
|
|
"num_tokens": 12546396.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.26887092590332,
|
|
"epoch": 5.599529780564263,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00025883140370966975,
|
|
"loss": 4.5962,
|
|
"mean_token_accuracy": 0.2294570252299309,
|
|
"num_tokens": 12554985.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.342357540130616,
|
|
"epoch": 5.603448275862069,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002585314050961565,
|
|
"loss": 4.5968,
|
|
"mean_token_accuracy": 0.2447383940219879,
|
|
"num_tokens": 12564379.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.3485370635986325,
|
|
"epoch": 5.6073667711598745,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002582314359146927,
|
|
"loss": 4.6492,
|
|
"mean_token_accuracy": 0.22447012960910798,
|
|
"num_tokens": 12572655.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.371013975143432,
|
|
"epoch": 5.61128526645768,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00025793149670137175,
|
|
"loss": 4.6709,
|
|
"mean_token_accuracy": 0.22416313290596007,
|
|
"num_tokens": 12581778.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.267105150222778,
|
|
"epoch": 5.6152037617554855,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002576315879922337,
|
|
"loss": 4.5463,
|
|
"mean_token_accuracy": 0.23166060894727708,
|
|
"num_tokens": 12590324.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.2802191257476805,
|
|
"epoch": 5.6191222570532915,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00025733171032326404,
|
|
"loss": 4.6827,
|
|
"mean_token_accuracy": 0.21613024771213532,
|
|
"num_tokens": 12599761.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.385409355163574,
|
|
"epoch": 5.6230407523510975,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002570318642303929,
|
|
"loss": 4.6846,
|
|
"mean_token_accuracy": 0.22071581929922104,
|
|
"num_tokens": 12608853.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.375567674636841,
|
|
"epoch": 5.6269592476489025,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00025673205024949377,
|
|
"loss": 4.5935,
|
|
"mean_token_accuracy": 0.22949183136224746,
|
|
"num_tokens": 12617434.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.33574538230896,
|
|
"epoch": 5.6308777429467085,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002564322689163828,
|
|
"loss": 4.64,
|
|
"mean_token_accuracy": 0.2272656574845314,
|
|
"num_tokens": 12626021.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.3185278415679935,
|
|
"epoch": 5.6347962382445145,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025613252076681786,
|
|
"loss": 4.625,
|
|
"mean_token_accuracy": 0.225187049806118,
|
|
"num_tokens": 12634952.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.317505884170532,
|
|
"epoch": 5.63871473354232,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002558328063364976,
|
|
"loss": 4.6364,
|
|
"mean_token_accuracy": 0.22270211726427078,
|
|
"num_tokens": 12644094.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.3281786918640135,
|
|
"epoch": 5.6426332288401255,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002555331261610603,
|
|
"loss": 4.6498,
|
|
"mean_token_accuracy": 0.22123885601758958,
|
|
"num_tokens": 12652687.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.320273208618164,
|
|
"epoch": 5.646551724137931,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 4.6026,
|
|
"mean_token_accuracy": 0.23252741247415543,
|
|
"num_tokens": 12661437.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.384975004196167,
|
|
"epoch": 5.650470219435737,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00025493387071708024,
|
|
"loss": 4.6714,
|
|
"mean_token_accuracy": 0.2268464595079422,
|
|
"num_tokens": 12669893.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.366326522827149,
|
|
"epoch": 5.6543887147335425,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025463429651950426,
|
|
"loss": 4.6936,
|
|
"mean_token_accuracy": 0.22118155509233475,
|
|
"num_tokens": 12679349.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.38269305229187,
|
|
"epoch": 5.658307210031348,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002543347587187425,
|
|
"loss": 4.6459,
|
|
"mean_token_accuracy": 0.2166955828666687,
|
|
"num_tokens": 12687163.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.362296390533447,
|
|
"epoch": 5.662225705329154,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025403525785011767,
|
|
"loss": 4.6297,
|
|
"mean_token_accuracy": 0.2217496782541275,
|
|
"num_tokens": 12695646.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.397637510299683,
|
|
"epoch": 5.66614420062696,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025373579444888634,
|
|
"loss": 4.6037,
|
|
"mean_token_accuracy": 0.23816232979297638,
|
|
"num_tokens": 12705144.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.346381664276123,
|
|
"epoch": 5.670062695924765,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000253436369050238,
|
|
"loss": 4.6085,
|
|
"mean_token_accuracy": 0.2248201698064804,
|
|
"num_tokens": 12713881.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.311061096191406,
|
|
"epoch": 5.673981191222571,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00025313698218929474,
|
|
"loss": 4.5925,
|
|
"mean_token_accuracy": 0.2319439396262169,
|
|
"num_tokens": 12722452.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.323856353759766,
|
|
"epoch": 5.677899686520377,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025283763440110923,
|
|
"loss": 4.601,
|
|
"mean_token_accuracy": 0.22589626908302307,
|
|
"num_tokens": 12731105.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.320547389984131,
|
|
"epoch": 5.681818181818182,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002525383262206645,
|
|
"loss": 4.6185,
|
|
"mean_token_accuracy": 0.232512766122818,
|
|
"num_tokens": 12739096.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.336972618103028,
|
|
"epoch": 5.685736677115988,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002522390581828726,
|
|
"loss": 4.6304,
|
|
"mean_token_accuracy": 0.22262786328792572,
|
|
"num_tokens": 12747910.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.273744678497314,
|
|
"epoch": 5.689655172413794,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00025193983082257444,
|
|
"loss": 4.523,
|
|
"mean_token_accuracy": 0.2318722724914551,
|
|
"num_tokens": 12757387.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.315034961700439,
|
|
"epoch": 5.693573667711599,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002516406446745376,
|
|
"loss": 4.6506,
|
|
"mean_token_accuracy": 0.22400361746549607,
|
|
"num_tokens": 12766678.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.286884593963623,
|
|
"epoch": 5.697492163009405,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002513415002734562,
|
|
"loss": 4.5989,
|
|
"mean_token_accuracy": 0.22623259276151658,
|
|
"num_tokens": 12775499.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.348700523376465,
|
|
"epoch": 5.70141065830721,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025104239815394995,
|
|
"loss": 4.6847,
|
|
"mean_token_accuracy": 0.2260138601064682,
|
|
"num_tokens": 12784173.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.3956691265106205,
|
|
"epoch": 5.705329153605016,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002507433388505627,
|
|
"loss": 4.6939,
|
|
"mean_token_accuracy": 0.22593950778245925,
|
|
"num_tokens": 12792991.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.338049840927124,
|
|
"epoch": 5.709247648902822,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025044432289776197,
|
|
"loss": 4.6318,
|
|
"mean_token_accuracy": 0.2222058191895485,
|
|
"num_tokens": 12801868.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.382672119140625,
|
|
"epoch": 5.713166144200627,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00025014535082993765,
|
|
"loss": 4.6564,
|
|
"mean_token_accuracy": 0.2296669602394104,
|
|
"num_tokens": 12810467.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.306714820861816,
|
|
"epoch": 5.717084639498433,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002498464231814015,
|
|
"loss": 4.6504,
|
|
"mean_token_accuracy": 0.22881872653961183,
|
|
"num_tokens": 12818874.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.317830514907837,
|
|
"epoch": 5.721003134796238,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024954754048638555,
|
|
"loss": 4.5833,
|
|
"mean_token_accuracy": 0.22863222956657409,
|
|
"num_tokens": 12828029.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.311804246902466,
|
|
"epoch": 5.724921630094044,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00024924870327904165,
|
|
"loss": 4.6023,
|
|
"mean_token_accuracy": 0.22988693416118622,
|
|
"num_tokens": 12836185.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.401476669311523,
|
|
"epoch": 5.72884012539185,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002489499120934403,
|
|
"loss": 4.762,
|
|
"mean_token_accuracy": 0.21398348957300187,
|
|
"num_tokens": 12845810.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.313670587539673,
|
|
"epoch": 5.732758620689655,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024865116746357004,
|
|
"loss": 4.5043,
|
|
"mean_token_accuracy": 0.2420254796743393,
|
|
"num_tokens": 12854567.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.306736755371094,
|
|
"epoch": 5.736677115987461,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002483524699233356,
|
|
"loss": 4.6173,
|
|
"mean_token_accuracy": 0.22727540880441666,
|
|
"num_tokens": 12863172.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.35368766784668,
|
|
"epoch": 5.740595611285267,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024805382000655824,
|
|
"loss": 4.6979,
|
|
"mean_token_accuracy": 0.22141339182853698,
|
|
"num_tokens": 12873335.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.41165337562561,
|
|
"epoch": 5.744514106583072,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002477552182469736,
|
|
"loss": 4.6591,
|
|
"mean_token_accuracy": 0.2316403791308403,
|
|
"num_tokens": 12882695.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.293678331375122,
|
|
"epoch": 5.748432601880878,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002474566651782315,
|
|
"loss": 4.6513,
|
|
"mean_token_accuracy": 0.22139888852834702,
|
|
"num_tokens": 12890739.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.275561094284058,
|
|
"epoch": 5.752351097178684,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002471581613338946,
|
|
"loss": 4.634,
|
|
"mean_token_accuracy": 0.2255994960665703,
|
|
"num_tokens": 12899503.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.3567564487457275,
|
|
"epoch": 5.756269592476489,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002468597072474378,
|
|
"loss": 4.6675,
|
|
"mean_token_accuracy": 0.22533704936504365,
|
|
"num_tokens": 12908405.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.378401470184326,
|
|
"epoch": 5.760188087774295,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024656130345224685,
|
|
"loss": 4.7206,
|
|
"mean_token_accuracy": 0.21582826226949692,
|
|
"num_tokens": 12917529.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.405226850509644,
|
|
"epoch": 5.764106583072101,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00024626295048161775,
|
|
"loss": 4.6764,
|
|
"mean_token_accuracy": 0.22472936362028123,
|
|
"num_tokens": 12926183.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.449588680267334,
|
|
"epoch": 5.768025078369906,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002459646488687557,
|
|
"loss": 4.7332,
|
|
"mean_token_accuracy": 0.21916427463293076,
|
|
"num_tokens": 12934598.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.337122058868408,
|
|
"epoch": 5.771943573667712,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002456663991467739,
|
|
"loss": 4.6275,
|
|
"mean_token_accuracy": 0.22022423297166824,
|
|
"num_tokens": 12943101.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.364291429519653,
|
|
"epoch": 5.775862068965517,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00024536820184869313,
|
|
"loss": 4.6794,
|
|
"mean_token_accuracy": 0.22441145330667495,
|
|
"num_tokens": 12951717.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.381462621688843,
|
|
"epoch": 5.779780564263323,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002450700575074401,
|
|
"loss": 4.7592,
|
|
"mean_token_accuracy": 0.21652369797229767,
|
|
"num_tokens": 12960761.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.362426710128784,
|
|
"epoch": 5.783699059561129,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00024477196665584735,
|
|
"loss": 4.5659,
|
|
"mean_token_accuracy": 0.2358320951461792,
|
|
"num_tokens": 12969216.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.393388080596924,
|
|
"epoch": 5.787617554858934,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002444739298266513,
|
|
"loss": 4.6356,
|
|
"mean_token_accuracy": 0.21404601484537125,
|
|
"num_tokens": 12977260.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.307894086837768,
|
|
"epoch": 5.79153605015674,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00024417594755249227,
|
|
"loss": 4.6188,
|
|
"mean_token_accuracy": 0.2224228337407112,
|
|
"num_tokens": 12985383.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.354758405685425,
|
|
"epoch": 5.795454545454545,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00024387802036591295,
|
|
"loss": 4.6475,
|
|
"mean_token_accuracy": 0.2246711701154709,
|
|
"num_tokens": 12994675.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.385832023620606,
|
|
"epoch": 5.799373040752351,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002435801487993573,
|
|
"loss": 4.6097,
|
|
"mean_token_accuracy": 0.23065033853054046,
|
|
"num_tokens": 13002979.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.325728893280029,
|
|
"epoch": 5.803291536050157,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00024328233338517024,
|
|
"loss": 4.6258,
|
|
"mean_token_accuracy": 0.2242472931742668,
|
|
"num_tokens": 13011343.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.342053079605103,
|
|
"epoch": 5.807210031347962,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00024298457465559626,
|
|
"loss": 4.7396,
|
|
"mean_token_accuracy": 0.2134027138352394,
|
|
"num_tokens": 13019684.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.323670339584351,
|
|
"epoch": 5.811128526645768,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024268687314277844,
|
|
"loss": 4.6223,
|
|
"mean_token_accuracy": 0.2214588478207588,
|
|
"num_tokens": 13028358.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.263364553451538,
|
|
"epoch": 5.815047021943574,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024238922937875758,
|
|
"loss": 4.586,
|
|
"mean_token_accuracy": 0.22147657722234726,
|
|
"num_tokens": 13037092.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.40320897102356,
|
|
"epoch": 5.818965517241379,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002420916438954714,
|
|
"loss": 4.7445,
|
|
"mean_token_accuracy": 0.22245440930128096,
|
|
"num_tokens": 13045624.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.3761035919189455,
|
|
"epoch": 5.822884012539185,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002417941172247535,
|
|
"loss": 4.6579,
|
|
"mean_token_accuracy": 0.22200690805912018,
|
|
"num_tokens": 13054384.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.321219444274902,
|
|
"epoch": 5.826802507836991,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00024149664989833215,
|
|
"loss": 4.6954,
|
|
"mean_token_accuracy": 0.22286714017391204,
|
|
"num_tokens": 13063786.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.382978820800782,
|
|
"epoch": 5.830721003134796,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 4.7034,
|
|
"mean_token_accuracy": 0.21426378041505814,
|
|
"num_tokens": 13073446.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.311453866958618,
|
|
"epoch": 5.834639498432602,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00024090189540476153,
|
|
"loss": 4.6412,
|
|
"mean_token_accuracy": 0.22102769315242768,
|
|
"num_tokens": 13081780.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.400103044509888,
|
|
"epoch": 5.838557993730408,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00024060460930053513,
|
|
"loss": 4.6636,
|
|
"mean_token_accuracy": 0.22245851010084153,
|
|
"num_tokens": 13089995.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.319657945632935,
|
|
"epoch": 5.842476489028213,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00024030738466644875,
|
|
"loss": 4.6301,
|
|
"mean_token_accuracy": 0.22662611901760102,
|
|
"num_tokens": 13098767.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.308699703216552,
|
|
"epoch": 5.846394984326019,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002400102220336911,
|
|
"loss": 4.6989,
|
|
"mean_token_accuracy": 0.21892803013324738,
|
|
"num_tokens": 13107744.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.324994611740112,
|
|
"epoch": 5.850313479623824,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002397131219333401,
|
|
"loss": 4.603,
|
|
"mean_token_accuracy": 0.2281987652182579,
|
|
"num_tokens": 13116400.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.282018375396729,
|
|
"epoch": 5.85423197492163,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002394160848963617,
|
|
"loss": 4.618,
|
|
"mean_token_accuracy": 0.2301843836903572,
|
|
"num_tokens": 13124973.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.409032678604126,
|
|
"epoch": 5.858150470219436,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023911911145360916,
|
|
"loss": 4.6572,
|
|
"mean_token_accuracy": 0.2195444256067276,
|
|
"num_tokens": 13133061.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.395173597335815,
|
|
"epoch": 5.862068965517241,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023882220213582235,
|
|
"loss": 4.6951,
|
|
"mean_token_accuracy": 0.2222263902425766,
|
|
"num_tokens": 13141322.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.356423807144165,
|
|
"epoch": 5.865987460815047,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002385253574736262,
|
|
"loss": 4.6575,
|
|
"mean_token_accuracy": 0.22061436176300048,
|
|
"num_tokens": 13150553.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.421728134155273,
|
|
"epoch": 5.869905956112853,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023822857799753034,
|
|
"loss": 4.7505,
|
|
"mean_token_accuracy": 0.2184569701552391,
|
|
"num_tokens": 13160308.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.430406141281128,
|
|
"epoch": 5.873824451410658,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023793186423792784,
|
|
"loss": 4.7124,
|
|
"mean_token_accuracy": 0.22216495126485825,
|
|
"num_tokens": 13168219.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.3974040985107425,
|
|
"epoch": 5.877742946708464,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023763521672509432,
|
|
"loss": 4.6984,
|
|
"mean_token_accuracy": 0.21766547113656998,
|
|
"num_tokens": 13176980.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 5.877742946708464,
|
|
"eval_entropy": 5.1821880543878835,
|
|
"eval_loss": 5.606734752655029,
|
|
"eval_mean_token_accuracy": 0.17543860403604286,
|
|
"eval_num_tokens": 13176980.0,
|
|
"eval_runtime": 2.1724,
|
|
"eval_samples_per_second": 1897.46,
|
|
"eval_steps_per_second": 237.528,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.432920837402344,
|
|
"epoch": 5.881661442006269,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023733863598918688,
|
|
"loss": 4.7409,
|
|
"mean_token_accuracy": 0.21577112674713134,
|
|
"num_tokens": 13186267.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.407967805862427,
|
|
"epoch": 5.885579937304075,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002370421225602433,
|
|
"loss": 4.7341,
|
|
"mean_token_accuracy": 0.22114830911159516,
|
|
"num_tokens": 13195457.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.397951745986939,
|
|
"epoch": 5.889498432601881,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023674567696818145,
|
|
"loss": 4.7031,
|
|
"mean_token_accuracy": 0.2139092743396759,
|
|
"num_tokens": 13203707.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.3590295791625975,
|
|
"epoch": 5.893416927899686,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002364492997427975,
|
|
"loss": 4.6316,
|
|
"mean_token_accuracy": 0.22394489049911498,
|
|
"num_tokens": 13213013.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.314443063735962,
|
|
"epoch": 5.897335423197492,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00023615299141376545,
|
|
"loss": 4.6604,
|
|
"mean_token_accuracy": 0.2239106446504593,
|
|
"num_tokens": 13222356.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.310134315490723,
|
|
"epoch": 5.901253918495298,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00023585675251063665,
|
|
"loss": 4.6679,
|
|
"mean_token_accuracy": 0.21980874240398407,
|
|
"num_tokens": 13231791.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.343995571136475,
|
|
"epoch": 5.905172413793103,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002355605835628379,
|
|
"loss": 4.596,
|
|
"mean_token_accuracy": 0.2276252031326294,
|
|
"num_tokens": 13240445.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.370993423461914,
|
|
"epoch": 5.909090909090909,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023526448509967097,
|
|
"loss": 4.6754,
|
|
"mean_token_accuracy": 0.22078887075185777,
|
|
"num_tokens": 13249048.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.316024208068848,
|
|
"epoch": 5.913009404388715,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023496845765031215,
|
|
"loss": 4.5625,
|
|
"mean_token_accuracy": 0.22718962281942368,
|
|
"num_tokens": 13258206.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.376318120956421,
|
|
"epoch": 5.91692789968652,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00023467250174381021,
|
|
"loss": 4.6701,
|
|
"mean_token_accuracy": 0.22041189223527907,
|
|
"num_tokens": 13266756.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.337276315689087,
|
|
"epoch": 5.920846394984326,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00023437661790908648,
|
|
"loss": 4.7182,
|
|
"mean_token_accuracy": 0.21689973920583724,
|
|
"num_tokens": 13276033.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.402188634872436,
|
|
"epoch": 5.924764890282132,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023408080667493332,
|
|
"loss": 4.6608,
|
|
"mean_token_accuracy": 0.22312965840101243,
|
|
"num_tokens": 13285151.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.2969623565673825,
|
|
"epoch": 5.928683385579937,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00023378506857001335,
|
|
"loss": 4.5714,
|
|
"mean_token_accuracy": 0.22940460592508316,
|
|
"num_tokens": 13293412.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.369242238998413,
|
|
"epoch": 5.932601880877743,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023348940412285858,
|
|
"loss": 4.7092,
|
|
"mean_token_accuracy": 0.2116548553109169,
|
|
"num_tokens": 13301982.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.433214998245239,
|
|
"epoch": 5.936520376175548,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023319381386186923,
|
|
"loss": 4.6771,
|
|
"mean_token_accuracy": 0.2268921449780464,
|
|
"num_tokens": 13309766.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.3114025592803955,
|
|
"epoch": 5.940438871473354,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000232898298315313,
|
|
"loss": 4.6292,
|
|
"mean_token_accuracy": 0.23007003217935562,
|
|
"num_tokens": 13317960.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.405201768875122,
|
|
"epoch": 5.94435736677116,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002326028580113243,
|
|
"loss": 4.7132,
|
|
"mean_token_accuracy": 0.2133057862520218,
|
|
"num_tokens": 13326909.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.469137144088745,
|
|
"epoch": 5.948275862068965,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00023230749347790265,
|
|
"loss": 4.722,
|
|
"mean_token_accuracy": 0.22156224101781846,
|
|
"num_tokens": 13335861.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.381613445281983,
|
|
"epoch": 5.952194357366771,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023201220524291232,
|
|
"loss": 4.6859,
|
|
"mean_token_accuracy": 0.2173662379384041,
|
|
"num_tokens": 13345253.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.3534918308258055,
|
|
"epoch": 5.956112852664576,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023171699383408152,
|
|
"loss": 4.6675,
|
|
"mean_token_accuracy": 0.22371300160884858,
|
|
"num_tokens": 13354353.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.372372627258301,
|
|
"epoch": 5.960031347962382,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023142185977900065,
|
|
"loss": 4.733,
|
|
"mean_token_accuracy": 0.21953723281621934,
|
|
"num_tokens": 13363030.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.366191673278808,
|
|
"epoch": 5.963949843260188,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023112680360512218,
|
|
"loss": 4.6654,
|
|
"mean_token_accuracy": 0.22457799166440964,
|
|
"num_tokens": 13372797.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.388562297821045,
|
|
"epoch": 5.967868338557993,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023083182583975942,
|
|
"loss": 4.5999,
|
|
"mean_token_accuracy": 0.21900976598262786,
|
|
"num_tokens": 13381425.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.295317649841309,
|
|
"epoch": 5.971786833855799,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002305369270100855,
|
|
"loss": 4.5844,
|
|
"mean_token_accuracy": 0.2314027100801468,
|
|
"num_tokens": 13390246.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.335381126403808,
|
|
"epoch": 5.975705329153605,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002302421076431322,
|
|
"loss": 4.6265,
|
|
"mean_token_accuracy": 0.22350431680679322,
|
|
"num_tokens": 13398851.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.364312839508057,
|
|
"epoch": 5.97962382445141,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022994736826578976,
|
|
"loss": 4.6935,
|
|
"mean_token_accuracy": 0.22090203166007996,
|
|
"num_tokens": 13407870.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.344297075271607,
|
|
"epoch": 5.983542319749216,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002296527094048052,
|
|
"loss": 4.6937,
|
|
"mean_token_accuracy": 0.21788785308599473,
|
|
"num_tokens": 13416311.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.388355827331543,
|
|
"epoch": 5.987460815047022,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002293581315867817,
|
|
"loss": 4.7613,
|
|
"mean_token_accuracy": 0.21236754804849625,
|
|
"num_tokens": 13425669.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.30972695350647,
|
|
"epoch": 5.991379310344827,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002290636353381774,
|
|
"loss": 4.6215,
|
|
"mean_token_accuracy": 0.22785600870847703,
|
|
"num_tokens": 13434448.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.3782227516174315,
|
|
"epoch": 5.995297805642633,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022876922118530508,
|
|
"loss": 4.6875,
|
|
"mean_token_accuracy": 0.218020898103714,
|
|
"num_tokens": 13443712.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.268182897567749,
|
|
"epoch": 5.999216300940439,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002284748896543305,
|
|
"loss": 4.4829,
|
|
"mean_token_accuracy": 0.23890223056077958,
|
|
"num_tokens": 13452511.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.319502639770508,
|
|
"epoch": 6.003134796238244,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022818064127127163,
|
|
"loss": 4.4697,
|
|
"mean_token_accuracy": 0.23590767979621888,
|
|
"num_tokens": 13460990.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.332038831710816,
|
|
"epoch": 6.00705329153605,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022788647656199834,
|
|
"loss": 4.5987,
|
|
"mean_token_accuracy": 0.2388242095708847,
|
|
"num_tokens": 13470293.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.396560573577881,
|
|
"epoch": 6.0109717868338555,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022759239605223052,
|
|
"loss": 4.5221,
|
|
"mean_token_accuracy": 0.23515519350767136,
|
|
"num_tokens": 13479508.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.310745286941528,
|
|
"epoch": 6.014890282131661,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 4.4305,
|
|
"mean_token_accuracy": 0.24191832691431045,
|
|
"num_tokens": 13488163.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.451779127120972,
|
|
"epoch": 6.018808777429467,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022700448973333808,
|
|
"loss": 4.5829,
|
|
"mean_token_accuracy": 0.23214626908302308,
|
|
"num_tokens": 13497404.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.358430528640747,
|
|
"epoch": 6.0227272727272725,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022671066497489744,
|
|
"loss": 4.5148,
|
|
"mean_token_accuracy": 0.23425958454608917,
|
|
"num_tokens": 13506142.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.305556011199951,
|
|
"epoch": 6.0266457680250785,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002264169265173282,
|
|
"loss": 4.4199,
|
|
"mean_token_accuracy": 0.2472182810306549,
|
|
"num_tokens": 13514525.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.2994238376617435,
|
|
"epoch": 6.030564263322884,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022612327488558866,
|
|
"loss": 4.4294,
|
|
"mean_token_accuracy": 0.2487062469124794,
|
|
"num_tokens": 13523732.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.266369867324829,
|
|
"epoch": 6.0344827586206895,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022582971060448187,
|
|
"loss": 4.444,
|
|
"mean_token_accuracy": 0.24748983830213547,
|
|
"num_tokens": 13532904.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.29990816116333,
|
|
"epoch": 6.0384012539184955,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022553623419865488,
|
|
"loss": 4.4902,
|
|
"mean_token_accuracy": 0.23052548319101335,
|
|
"num_tokens": 13542174.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.313665342330933,
|
|
"epoch": 6.0423197492163006,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022524284619259745,
|
|
"loss": 4.5088,
|
|
"mean_token_accuracy": 0.23471334874629973,
|
|
"num_tokens": 13550591.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.30289044380188,
|
|
"epoch": 6.0462382445141065,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002249495471106417,
|
|
"loss": 4.4602,
|
|
"mean_token_accuracy": 0.2440162032842636,
|
|
"num_tokens": 13559088.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.2433734893798825,
|
|
"epoch": 6.0501567398119125,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022465633747696052,
|
|
"loss": 4.3848,
|
|
"mean_token_accuracy": 0.24363926649093628,
|
|
"num_tokens": 13568511.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.341625833511353,
|
|
"epoch": 6.054075235109718,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022436321781556712,
|
|
"loss": 4.567,
|
|
"mean_token_accuracy": 0.2273353487253189,
|
|
"num_tokens": 13577750.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.35686559677124,
|
|
"epoch": 6.0579937304075235,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002240701886503138,
|
|
"loss": 4.4981,
|
|
"mean_token_accuracy": 0.24087147861719133,
|
|
"num_tokens": 13587138.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.354097986221314,
|
|
"epoch": 6.0619122257053295,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022377725050489128,
|
|
"loss": 4.5139,
|
|
"mean_token_accuracy": 0.24525873363018036,
|
|
"num_tokens": 13595459.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.401451635360718,
|
|
"epoch": 6.065830721003135,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022348440390282753,
|
|
"loss": 4.586,
|
|
"mean_token_accuracy": 0.23739762753248214,
|
|
"num_tokens": 13605227.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.3993853569030765,
|
|
"epoch": 6.0697492163009406,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00022319164936748676,
|
|
"loss": 4.559,
|
|
"mean_token_accuracy": 0.23171960860490798,
|
|
"num_tokens": 13614139.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.365489912033081,
|
|
"epoch": 6.0736677115987465,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00022289898742206904,
|
|
"loss": 4.484,
|
|
"mean_token_accuracy": 0.23841353356838227,
|
|
"num_tokens": 13623420.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.376324701309204,
|
|
"epoch": 6.077586206896552,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002226064185896085,
|
|
"loss": 4.5303,
|
|
"mean_token_accuracy": 0.23920708000659943,
|
|
"num_tokens": 13632655.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.351416730880738,
|
|
"epoch": 6.081504702194358,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022231394339297322,
|
|
"loss": 4.4525,
|
|
"mean_token_accuracy": 0.24024342447519303,
|
|
"num_tokens": 13641782.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.3113054752349855,
|
|
"epoch": 6.085423197492163,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00022202156235486358,
|
|
"loss": 4.4437,
|
|
"mean_token_accuracy": 0.2466355159878731,
|
|
"num_tokens": 13650023.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.3212730884552,
|
|
"epoch": 6.089341692789969,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022172927599781224,
|
|
"loss": 4.4591,
|
|
"mean_token_accuracy": 0.2415129065513611,
|
|
"num_tokens": 13659025.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.386977863311768,
|
|
"epoch": 6.093260188087775,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022143708484418206,
|
|
"loss": 4.5118,
|
|
"mean_token_accuracy": 0.23194327503442763,
|
|
"num_tokens": 13668133.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.33960599899292,
|
|
"epoch": 6.09717868338558,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022114498941616607,
|
|
"loss": 4.4985,
|
|
"mean_token_accuracy": 0.23626119047403335,
|
|
"num_tokens": 13677661.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.300880813598633,
|
|
"epoch": 6.101097178683386,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022085299023578614,
|
|
"loss": 4.5086,
|
|
"mean_token_accuracy": 0.2331958070397377,
|
|
"num_tokens": 13686117.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.30370831489563,
|
|
"epoch": 6.105015673981192,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00022056108782489225,
|
|
"loss": 4.4236,
|
|
"mean_token_accuracy": 0.2435981884598732,
|
|
"num_tokens": 13694792.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.32201623916626,
|
|
"epoch": 6.108934169278997,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022026928270516122,
|
|
"loss": 4.5293,
|
|
"mean_token_accuracy": 0.2330962374806404,
|
|
"num_tokens": 13704038.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.34885802268982,
|
|
"epoch": 6.112852664576803,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021997757539809622,
|
|
"loss": 4.5045,
|
|
"mean_token_accuracy": 0.24041985869407653,
|
|
"num_tokens": 13713266.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.292075729370117,
|
|
"epoch": 6.116771159874608,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021968596642502546,
|
|
"loss": 4.4476,
|
|
"mean_token_accuracy": 0.2474093869328499,
|
|
"num_tokens": 13721847.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.270249223709106,
|
|
"epoch": 6.120689655172414,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00021939445630710138,
|
|
"loss": 4.464,
|
|
"mean_token_accuracy": 0.24181337356567384,
|
|
"num_tokens": 13730925.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.3287537574768065,
|
|
"epoch": 6.12460815047022,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00021910304556529986,
|
|
"loss": 4.5858,
|
|
"mean_token_accuracy": 0.23149532973766326,
|
|
"num_tokens": 13739814.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.383867645263672,
|
|
"epoch": 6.128526645768025,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00021881173472041926,
|
|
"loss": 4.458,
|
|
"mean_token_accuracy": 0.2462196633219719,
|
|
"num_tokens": 13748859.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.295474481582642,
|
|
"epoch": 6.132445141065831,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021852052429307924,
|
|
"loss": 4.5325,
|
|
"mean_token_accuracy": 0.23874372094869614,
|
|
"num_tokens": 13757249.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.3104713439941404,
|
|
"epoch": 6.136363636363637,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00021822941480371995,
|
|
"loss": 4.5026,
|
|
"mean_token_accuracy": 0.23743316531181335,
|
|
"num_tokens": 13765838.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.301988363265991,
|
|
"epoch": 6.140282131661442,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00021793840677260135,
|
|
"loss": 4.4546,
|
|
"mean_token_accuracy": 0.2524898052215576,
|
|
"num_tokens": 13775465.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.447124862670899,
|
|
"epoch": 6.144200626959248,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00021764750071980195,
|
|
"loss": 4.5797,
|
|
"mean_token_accuracy": 0.23247391879558563,
|
|
"num_tokens": 13785373.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.331523847579956,
|
|
"epoch": 6.148119122257054,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00021735669716521806,
|
|
"loss": 4.5117,
|
|
"mean_token_accuracy": 0.2461099162697792,
|
|
"num_tokens": 13794812.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.27855920791626,
|
|
"epoch": 6.152037617554859,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021706599662856265,
|
|
"loss": 4.4949,
|
|
"mean_token_accuracy": 0.2410942941904068,
|
|
"num_tokens": 13803681.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.281662464141846,
|
|
"epoch": 6.155956112852665,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021677539962936494,
|
|
"loss": 4.5797,
|
|
"mean_token_accuracy": 0.22723914980888366,
|
|
"num_tokens": 13812772.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.356380939483643,
|
|
"epoch": 6.15987460815047,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021648490668696875,
|
|
"loss": 4.4866,
|
|
"mean_token_accuracy": 0.23695665299892427,
|
|
"num_tokens": 13821280.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.295778465270996,
|
|
"epoch": 6.163793103448276,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021619451832053199,
|
|
"loss": 4.4682,
|
|
"mean_token_accuracy": 0.24028859734535218,
|
|
"num_tokens": 13829717.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.364227390289306,
|
|
"epoch": 6.167711598746082,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000215904235049026,
|
|
"loss": 4.576,
|
|
"mean_token_accuracy": 0.2322181925177574,
|
|
"num_tokens": 13838723.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.311845779418945,
|
|
"epoch": 6.171630094043887,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002156140573912338,
|
|
"loss": 4.5521,
|
|
"mean_token_accuracy": 0.23699123710393905,
|
|
"num_tokens": 13847081.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.330476140975952,
|
|
"epoch": 6.175548589341693,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00021532398586574997,
|
|
"loss": 4.463,
|
|
"mean_token_accuracy": 0.2453140825033188,
|
|
"num_tokens": 13855142.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.401287269592285,
|
|
"epoch": 6.179467084639499,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00021503402099097959,
|
|
"loss": 4.5698,
|
|
"mean_token_accuracy": 0.23876333981752396,
|
|
"num_tokens": 13863586.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.3503578186035154,
|
|
"epoch": 6.183385579937304,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00021474416328513664,
|
|
"loss": 4.5664,
|
|
"mean_token_accuracy": 0.2353839546442032,
|
|
"num_tokens": 13872286.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.324071884155273,
|
|
"epoch": 6.18730407523511,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002144544132662438,
|
|
"loss": 4.5148,
|
|
"mean_token_accuracy": 0.23538629114627838,
|
|
"num_tokens": 13880866.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.28391056060791,
|
|
"epoch": 6.191222570532915,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002141647714521316,
|
|
"loss": 4.4525,
|
|
"mean_token_accuracy": 0.2439201921224594,
|
|
"num_tokens": 13889015.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.355265998840332,
|
|
"epoch": 6.195141065830721,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00021387523836043676,
|
|
"loss": 4.5447,
|
|
"mean_token_accuracy": 0.2339564487338066,
|
|
"num_tokens": 13898457.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.371236324310303,
|
|
"epoch": 6.199059561128527,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 4.5041,
|
|
"mean_token_accuracy": 0.24937014132738114,
|
|
"num_tokens": 13906722.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.306252431869507,
|
|
"epoch": 6.202978056426332,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002132965004138742,
|
|
"loss": 4.4371,
|
|
"mean_token_accuracy": 0.24976745545864104,
|
|
"num_tokens": 13914714.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.310378742218018,
|
|
"epoch": 6.206896551724138,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002130072965933052,
|
|
"loss": 4.5256,
|
|
"mean_token_accuracy": 0.24162225276231766,
|
|
"num_tokens": 13923378.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.2700111865997314,
|
|
"epoch": 6.210815047021944,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021271820356374876,
|
|
"loss": 4.5064,
|
|
"mean_token_accuracy": 0.23420566320419312,
|
|
"num_tokens": 13931947.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.283775711059571,
|
|
"epoch": 6.214733542319749,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00021242922184186108,
|
|
"loss": 4.4393,
|
|
"mean_token_accuracy": 0.23669122010469437,
|
|
"num_tokens": 13940949.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.3597194194793705,
|
|
"epoch": 6.218652037617555,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002121403519440993,
|
|
"loss": 4.5502,
|
|
"mean_token_accuracy": 0.2297421619296074,
|
|
"num_tokens": 13949944.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.321140766143799,
|
|
"epoch": 6.222570532915361,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00021185159438672085,
|
|
"loss": 4.4867,
|
|
"mean_token_accuracy": 0.23845660090446472,
|
|
"num_tokens": 13958938.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.284685373306274,
|
|
"epoch": 6.226489028213166,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021156294968578232,
|
|
"loss": 4.4427,
|
|
"mean_token_accuracy": 0.24775538444519044,
|
|
"num_tokens": 13967875.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.269733619689942,
|
|
"epoch": 6.230407523510972,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00021127441835713838,
|
|
"loss": 4.4437,
|
|
"mean_token_accuracy": 0.24477414339780806,
|
|
"num_tokens": 13975972.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.279259204864502,
|
|
"epoch": 6.234326018808777,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002109860009164416,
|
|
"loss": 4.4199,
|
|
"mean_token_accuracy": 0.23724733144044877,
|
|
"num_tokens": 13985081.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.242986583709717,
|
|
"epoch": 6.238244514106583,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0002106976978791405,
|
|
"loss": 4.5509,
|
|
"mean_token_accuracy": 0.23463145047426223,
|
|
"num_tokens": 13992426.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.271199369430542,
|
|
"epoch": 6.242163009404389,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021040950976047935,
|
|
"loss": 4.4737,
|
|
"mean_token_accuracy": 0.24121629148721696,
|
|
"num_tokens": 14001328.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.282930755615235,
|
|
"epoch": 6.246081504702194,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021012143707549726,
|
|
"loss": 4.4201,
|
|
"mean_token_accuracy": 0.24700042903423308,
|
|
"num_tokens": 14010340.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.402168321609497,
|
|
"epoch": 6.25,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020983348033902662,
|
|
"loss": 4.6137,
|
|
"mean_token_accuracy": 0.23183374106884003,
|
|
"num_tokens": 14019213.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.34532060623169,
|
|
"epoch": 6.253918495297806,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020954564006569294,
|
|
"loss": 4.5474,
|
|
"mean_token_accuracy": 0.22925416827201844,
|
|
"num_tokens": 14028501.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.253282833099365,
|
|
"epoch": 6.257836990595611,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020925791676991318,
|
|
"loss": 4.4476,
|
|
"mean_token_accuracy": 0.23893422037363052,
|
|
"num_tokens": 14037969.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.36182975769043,
|
|
"epoch": 6.261755485893417,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00020897031096589586,
|
|
"loss": 4.4816,
|
|
"mean_token_accuracy": 0.25348174571990967,
|
|
"num_tokens": 14046987.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.315373134613037,
|
|
"epoch": 6.265673981191223,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020868282316763902,
|
|
"loss": 4.5313,
|
|
"mean_token_accuracy": 0.2344393476843834,
|
|
"num_tokens": 14055894.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.246586179733276,
|
|
"epoch": 6.269592476489028,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020839545388893,
|
|
"loss": 4.4805,
|
|
"mean_token_accuracy": 0.2406961128115654,
|
|
"num_tokens": 14065119.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 6.269592476489028,
|
|
"eval_entropy": 5.176025046858677,
|
|
"eval_loss": 5.628513813018799,
|
|
"eval_mean_token_accuracy": 0.17443832906699458,
|
|
"eval_num_tokens": 14065119.0,
|
|
"eval_runtime": 2.3456,
|
|
"eval_samples_per_second": 1757.341,
|
|
"eval_steps_per_second": 219.987,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.352739381790161,
|
|
"epoch": 6.273510971786834,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020810820364334438,
|
|
"loss": 4.4702,
|
|
"mean_token_accuracy": 0.24658433645963668,
|
|
"num_tokens": 14073320.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.317093420028686,
|
|
"epoch": 6.277429467084639,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020782107294424484,
|
|
"loss": 4.4975,
|
|
"mean_token_accuracy": 0.2425541251897812,
|
|
"num_tokens": 14081730.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.230673837661743,
|
|
"epoch": 6.281347962382445,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002075340623047804,
|
|
"loss": 4.4653,
|
|
"mean_token_accuracy": 0.24055927395820617,
|
|
"num_tokens": 14090738.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.348722171783447,
|
|
"epoch": 6.285266457680251,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020724717223788574,
|
|
"loss": 4.5641,
|
|
"mean_token_accuracy": 0.23002694249153138,
|
|
"num_tokens": 14099412.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.279159879684448,
|
|
"epoch": 6.289184952978056,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00020696040325627991,
|
|
"loss": 4.5202,
|
|
"mean_token_accuracy": 0.243551541864872,
|
|
"num_tokens": 14108334.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.294950151443482,
|
|
"epoch": 6.293103448275862,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020667375587246563,
|
|
"loss": 4.5525,
|
|
"mean_token_accuracy": 0.23692710250616072,
|
|
"num_tokens": 14116853.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.38358063697815,
|
|
"epoch": 6.297021943573668,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020638723059872806,
|
|
"loss": 4.6081,
|
|
"mean_token_accuracy": 0.22968841791152955,
|
|
"num_tokens": 14126735.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.342741823196411,
|
|
"epoch": 6.300940438871473,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00020610082794713454,
|
|
"loss": 4.5819,
|
|
"mean_token_accuracy": 0.24078532606363295,
|
|
"num_tokens": 14136749.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.351387214660645,
|
|
"epoch": 6.304858934169279,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002058145484295329,
|
|
"loss": 4.5338,
|
|
"mean_token_accuracy": 0.23849495947360994,
|
|
"num_tokens": 14145204.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.4068520069122314,
|
|
"epoch": 6.308777429467085,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020552839255755107,
|
|
"loss": 4.6034,
|
|
"mean_token_accuracy": 0.23476792126893997,
|
|
"num_tokens": 14154347.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.311073875427246,
|
|
"epoch": 6.31269592476489,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020524236084259613,
|
|
"loss": 4.5144,
|
|
"mean_token_accuracy": 0.24303431510925294,
|
|
"num_tokens": 14163039.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.28370304107666,
|
|
"epoch": 6.316614420062696,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020495645379585303,
|
|
"loss": 4.4476,
|
|
"mean_token_accuracy": 0.2460635781288147,
|
|
"num_tokens": 14172228.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.341038227081299,
|
|
"epoch": 6.320532915360501,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020467067192828404,
|
|
"loss": 4.5829,
|
|
"mean_token_accuracy": 0.234201580286026,
|
|
"num_tokens": 14181514.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.354066562652588,
|
|
"epoch": 6.324451410658307,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020438501575062752,
|
|
"loss": 4.5737,
|
|
"mean_token_accuracy": 0.2326590597629547,
|
|
"num_tokens": 14190654.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.28126745223999,
|
|
"epoch": 6.328369905956113,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020409948577339765,
|
|
"loss": 4.4576,
|
|
"mean_token_accuracy": 0.2414070412516594,
|
|
"num_tokens": 14199177.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.337183904647827,
|
|
"epoch": 6.332288401253918,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00020381408250688267,
|
|
"loss": 4.5088,
|
|
"mean_token_accuracy": 0.24594120383262635,
|
|
"num_tokens": 14207643.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.380888557434082,
|
|
"epoch": 6.336206896551724,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002035288064611444,
|
|
"loss": 4.5727,
|
|
"mean_token_accuracy": 0.2336971014738083,
|
|
"num_tokens": 14216884.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.3581037521362305,
|
|
"epoch": 6.34012539184953,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002032436581460177,
|
|
"loss": 4.5195,
|
|
"mean_token_accuracy": 0.23178401589393616,
|
|
"num_tokens": 14225324.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.324050569534302,
|
|
"epoch": 6.344043887147335,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020295863807110853,
|
|
"loss": 4.4657,
|
|
"mean_token_accuracy": 0.24401862621307374,
|
|
"num_tokens": 14233732.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.445546722412109,
|
|
"epoch": 6.347962382445141,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020267374674579408,
|
|
"loss": 4.5876,
|
|
"mean_token_accuracy": 0.22843099385499954,
|
|
"num_tokens": 14242080.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.323999500274658,
|
|
"epoch": 6.351880877742946,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020238898467922129,
|
|
"loss": 4.5507,
|
|
"mean_token_accuracy": 0.2460624933242798,
|
|
"num_tokens": 14251140.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.357223081588745,
|
|
"epoch": 6.355799373040752,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020210435238030626,
|
|
"loss": 4.5898,
|
|
"mean_token_accuracy": 0.23144344389438629,
|
|
"num_tokens": 14259900.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.319518661499023,
|
|
"epoch": 6.359717868338558,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00020181985035773304,
|
|
"loss": 4.506,
|
|
"mean_token_accuracy": 0.23310118168592453,
|
|
"num_tokens": 14268414.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.3138526439666744,
|
|
"epoch": 6.363636363636363,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020153547911995272,
|
|
"loss": 4.5685,
|
|
"mean_token_accuracy": 0.23193838745355605,
|
|
"num_tokens": 14277339.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.340259122848511,
|
|
"epoch": 6.367554858934169,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020125123917518308,
|
|
"loss": 4.496,
|
|
"mean_token_accuracy": 0.23792601823806764,
|
|
"num_tokens": 14285710.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.323208284378052,
|
|
"epoch": 6.371473354231975,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020096713103140673,
|
|
"loss": 4.5356,
|
|
"mean_token_accuracy": 0.23741088807582855,
|
|
"num_tokens": 14294625.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.250871372222901,
|
|
"epoch": 6.37539184952978,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00020068315519637103,
|
|
"loss": 4.3674,
|
|
"mean_token_accuracy": 0.25359642654657366,
|
|
"num_tokens": 14303030.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.265396022796631,
|
|
"epoch": 6.379310344827586,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000200399312177587,
|
|
"loss": 4.4647,
|
|
"mean_token_accuracy": 0.24143918305635453,
|
|
"num_tokens": 14311826.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.349812126159668,
|
|
"epoch": 6.383228840125392,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 4.5584,
|
|
"mean_token_accuracy": 0.24072411060333251,
|
|
"num_tokens": 14319600.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.260319185256958,
|
|
"epoch": 6.387147335423197,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001998320266176294,
|
|
"loss": 4.4773,
|
|
"mean_token_accuracy": 0.2419591724872589,
|
|
"num_tokens": 14328768.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.287106275558472,
|
|
"epoch": 6.391065830721003,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000199548585090287,
|
|
"loss": 4.5601,
|
|
"mean_token_accuracy": 0.2301559790968895,
|
|
"num_tokens": 14338236.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.348999404907227,
|
|
"epoch": 6.394984326018808,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000199265278406857,
|
|
"loss": 4.5989,
|
|
"mean_token_accuracy": 0.2328197479248047,
|
|
"num_tokens": 14347190.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.347185039520264,
|
|
"epoch": 6.398902821316614,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00019898210707365428,
|
|
"loss": 4.5655,
|
|
"mean_token_accuracy": 0.23775911778211595,
|
|
"num_tokens": 14355847.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.2725098609924315,
|
|
"epoch": 6.40282131661442,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019869907159675187,
|
|
"loss": 4.4748,
|
|
"mean_token_accuracy": 0.2375943258404732,
|
|
"num_tokens": 14364280.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.329967403411866,
|
|
"epoch": 6.406739811912225,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019841617248198017,
|
|
"loss": 4.5295,
|
|
"mean_token_accuracy": 0.23536909222602845,
|
|
"num_tokens": 14372857.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.337087678909302,
|
|
"epoch": 6.410658307210031,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001981334102349258,
|
|
"loss": 4.4732,
|
|
"mean_token_accuracy": 0.23613303601741792,
|
|
"num_tokens": 14381157.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.382405996322632,
|
|
"epoch": 6.414576802507837,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00019785078536093065,
|
|
"loss": 4.5641,
|
|
"mean_token_accuracy": 0.2300056740641594,
|
|
"num_tokens": 14390404.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.355230903625488,
|
|
"epoch": 6.418495297805642,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019756829836509117,
|
|
"loss": 4.5533,
|
|
"mean_token_accuracy": 0.2327018827199936,
|
|
"num_tokens": 14399426.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.312778759002685,
|
|
"epoch": 6.422413793103448,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001972859497522575,
|
|
"loss": 4.4613,
|
|
"mean_token_accuracy": 0.22979287207126617,
|
|
"num_tokens": 14408233.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.348908948898315,
|
|
"epoch": 6.4263322884012535,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001970037400270323,
|
|
"loss": 4.5222,
|
|
"mean_token_accuracy": 0.23580269664525985,
|
|
"num_tokens": 14416828.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.362889051437378,
|
|
"epoch": 6.4302507836990594,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001967216696937702,
|
|
"loss": 4.5699,
|
|
"mean_token_accuracy": 0.23442524820566177,
|
|
"num_tokens": 14425590.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.362957000732422,
|
|
"epoch": 6.434169278996865,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019643973925657666,
|
|
"loss": 4.6735,
|
|
"mean_token_accuracy": 0.2196706250309944,
|
|
"num_tokens": 14434646.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.433896255493164,
|
|
"epoch": 6.4380877742946705,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019615794921930697,
|
|
"loss": 4.5661,
|
|
"mean_token_accuracy": 0.23435693085193635,
|
|
"num_tokens": 14443862.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.309257364273071,
|
|
"epoch": 6.4420062695924765,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019587630008556563,
|
|
"loss": 4.5016,
|
|
"mean_token_accuracy": 0.23956137299537658,
|
|
"num_tokens": 14452813.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.336293363571167,
|
|
"epoch": 6.445924764890282,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019559479235870537,
|
|
"loss": 4.5718,
|
|
"mean_token_accuracy": 0.23338126093149186,
|
|
"num_tokens": 14461912.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.360854291915894,
|
|
"epoch": 6.4498432601880875,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001953134265418261,
|
|
"loss": 4.5132,
|
|
"mean_token_accuracy": 0.24022517651319503,
|
|
"num_tokens": 14470460.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.330595064163208,
|
|
"epoch": 6.4537617554858935,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0001950322031377742,
|
|
"loss": 4.5163,
|
|
"mean_token_accuracy": 0.23923503756523132,
|
|
"num_tokens": 14479258.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.369436740875244,
|
|
"epoch": 6.4576802507836994,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00019475112264914142,
|
|
"loss": 4.5121,
|
|
"mean_token_accuracy": 0.2423982083797455,
|
|
"num_tokens": 14488209.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.273265314102173,
|
|
"epoch": 6.4615987460815045,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019447018557826424,
|
|
"loss": 4.4726,
|
|
"mean_token_accuracy": 0.24001111090183258,
|
|
"num_tokens": 14497234.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.243346691131592,
|
|
"epoch": 6.4655172413793105,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001941893924272227,
|
|
"loss": 4.3884,
|
|
"mean_token_accuracy": 0.24784687906503677,
|
|
"num_tokens": 14506374.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.344077253341675,
|
|
"epoch": 6.469435736677116,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00019390874369783974,
|
|
"loss": 4.5552,
|
|
"mean_token_accuracy": 0.23302698880434036,
|
|
"num_tokens": 14514796.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.331205129623413,
|
|
"epoch": 6.4733542319749215,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019362823989168006,
|
|
"loss": 4.5215,
|
|
"mean_token_accuracy": 0.23748065084218978,
|
|
"num_tokens": 14523771.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.2481659889221195,
|
|
"epoch": 6.4772727272727275,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00019334788151004962,
|
|
"loss": 4.4306,
|
|
"mean_token_accuracy": 0.24967492371797562,
|
|
"num_tokens": 14532130.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.200548219680786,
|
|
"epoch": 6.481191222570533,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019306766905399415,
|
|
"loss": 4.4331,
|
|
"mean_token_accuracy": 0.23936139643192292,
|
|
"num_tokens": 14540278.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.310354137420655,
|
|
"epoch": 6.485109717868339,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019278760302429872,
|
|
"loss": 4.5391,
|
|
"mean_token_accuracy": 0.22338234186172484,
|
|
"num_tokens": 14549519.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.327473545074463,
|
|
"epoch": 6.4890282131661445,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019250768392148682,
|
|
"loss": 4.5816,
|
|
"mean_token_accuracy": 0.23577690869569778,
|
|
"num_tokens": 14558576.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.318864631652832,
|
|
"epoch": 6.49294670846395,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019222791224581925,
|
|
"loss": 4.4846,
|
|
"mean_token_accuracy": 0.23261725157499313,
|
|
"num_tokens": 14567544.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.293386793136596,
|
|
"epoch": 6.496865203761756,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00019194828849729336,
|
|
"loss": 4.4731,
|
|
"mean_token_accuracy": 0.22724894285202027,
|
|
"num_tokens": 14576662.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.259024381637573,
|
|
"epoch": 6.500783699059561,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019166881317564217,
|
|
"loss": 4.4575,
|
|
"mean_token_accuracy": 0.2372124671936035,
|
|
"num_tokens": 14585163.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.248923301696777,
|
|
"epoch": 6.504702194357367,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00019138948678033336,
|
|
"loss": 4.4193,
|
|
"mean_token_accuracy": 0.24551893025636673,
|
|
"num_tokens": 14593213.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.307544279098511,
|
|
"epoch": 6.508620689655173,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019111030981056832,
|
|
"loss": 4.5032,
|
|
"mean_token_accuracy": 0.2427818149328232,
|
|
"num_tokens": 14601479.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.393671655654908,
|
|
"epoch": 6.512539184952978,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001908312827652819,
|
|
"loss": 4.5188,
|
|
"mean_token_accuracy": 0.23354093432426454,
|
|
"num_tokens": 14610418.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.286554574966431,
|
|
"epoch": 6.516457680250784,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00019055240614314047,
|
|
"loss": 4.5198,
|
|
"mean_token_accuracy": 0.23881785720586776,
|
|
"num_tokens": 14618387.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.267303419113159,
|
|
"epoch": 6.52037617554859,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019027368044254177,
|
|
"loss": 4.4588,
|
|
"mean_token_accuracy": 0.2459932431578636,
|
|
"num_tokens": 14627003.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.302926969528198,
|
|
"epoch": 6.524294670846395,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001899951061616138,
|
|
"loss": 4.5256,
|
|
"mean_token_accuracy": 0.23662308752536773,
|
|
"num_tokens": 14635825.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.32719874382019,
|
|
"epoch": 6.528213166144201,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018971668379821405,
|
|
"loss": 4.4756,
|
|
"mean_token_accuracy": 0.2408345490694046,
|
|
"num_tokens": 14644830.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.280014610290527,
|
|
"epoch": 6.532131661442007,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001894384138499284,
|
|
"loss": 4.5003,
|
|
"mean_token_accuracy": 0.23181315809488295,
|
|
"num_tokens": 14653633.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.280670690536499,
|
|
"epoch": 6.536050156739812,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018916029681407032,
|
|
"loss": 4.3736,
|
|
"mean_token_accuracy": 0.2640697851777077,
|
|
"num_tokens": 14662212.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.352401876449585,
|
|
"epoch": 6.539968652037618,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001888823331876801,
|
|
"loss": 4.5191,
|
|
"mean_token_accuracy": 0.23450336903333663,
|
|
"num_tokens": 14670942.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.356369829177856,
|
|
"epoch": 6.543887147335424,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00018860452346752382,
|
|
"loss": 4.479,
|
|
"mean_token_accuracy": 0.23606017231941223,
|
|
"num_tokens": 14679167.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.256765651702881,
|
|
"epoch": 6.547805642633229,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018832686815009242,
|
|
"loss": 4.5098,
|
|
"mean_token_accuracy": 0.23506882041692734,
|
|
"num_tokens": 14688188.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.361252403259277,
|
|
"epoch": 6.551724137931035,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018804936773160108,
|
|
"loss": 4.5889,
|
|
"mean_token_accuracy": 0.2346164420247078,
|
|
"num_tokens": 14696958.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.281148290634155,
|
|
"epoch": 6.55564263322884,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000187772022707988,
|
|
"loss": 4.5023,
|
|
"mean_token_accuracy": 0.23355162888765335,
|
|
"num_tokens": 14705524.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.305964088439941,
|
|
"epoch": 6.559561128526646,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018749483357491365,
|
|
"loss": 4.5537,
|
|
"mean_token_accuracy": 0.23339749723672867,
|
|
"num_tokens": 14714500.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.336895847320557,
|
|
"epoch": 6.563479623824452,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00018721780082775997,
|
|
"loss": 4.5296,
|
|
"mean_token_accuracy": 0.23777825385332108,
|
|
"num_tokens": 14722816.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.302072191238404,
|
|
"epoch": 6.567398119122257,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 4.5458,
|
|
"mean_token_accuracy": 0.24021702408790588,
|
|
"num_tokens": 14731453.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.359823369979859,
|
|
"epoch": 6.571316614420063,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00018666420647134419,
|
|
"loss": 4.5375,
|
|
"mean_token_accuracy": 0.24263923168182372,
|
|
"num_tokens": 14739924.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.213691997528076,
|
|
"epoch": 6.575235109717868,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001863876458514448,
|
|
"loss": 4.3921,
|
|
"mean_token_accuracy": 0.2528491631150246,
|
|
"num_tokens": 14748356.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.385762643814087,
|
|
"epoch": 6.579153605015674,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018611124359619024,
|
|
"loss": 4.5965,
|
|
"mean_token_accuracy": 0.23197027891874314,
|
|
"num_tokens": 14758846.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.254580307006836,
|
|
"epoch": 6.58307210031348,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00018583500019955595,
|
|
"loss": 4.4235,
|
|
"mean_token_accuracy": 0.2430300831794739,
|
|
"num_tokens": 14767030.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.380947399139404,
|
|
"epoch": 6.586990595611285,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018555891615523374,
|
|
"loss": 4.4854,
|
|
"mean_token_accuracy": 0.24057793617248535,
|
|
"num_tokens": 14776774.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.2723771095275875,
|
|
"epoch": 6.590909090909091,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018528299195663046,
|
|
"loss": 4.4655,
|
|
"mean_token_accuracy": 0.23994943499565125,
|
|
"num_tokens": 14785662.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.383119392395019,
|
|
"epoch": 6.594827586206897,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001850072280968676,
|
|
"loss": 4.6087,
|
|
"mean_token_accuracy": 0.22290503680706025,
|
|
"num_tokens": 14795070.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.331168365478516,
|
|
"epoch": 6.598746081504702,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00018473162506877977,
|
|
"loss": 4.5354,
|
|
"mean_token_accuracy": 0.23633747696876525,
|
|
"num_tokens": 14805452.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.301808834075928,
|
|
"epoch": 6.602664576802508,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018445618336491415,
|
|
"loss": 4.5043,
|
|
"mean_token_accuracy": 0.23962927162647246,
|
|
"num_tokens": 14815632.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.283541011810303,
|
|
"epoch": 6.606583072100314,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018418090347752997,
|
|
"loss": 4.4605,
|
|
"mean_token_accuracy": 0.24028003364801406,
|
|
"num_tokens": 14824228.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.315742444992066,
|
|
"epoch": 6.610501567398119,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001839057858985968,
|
|
"loss": 4.5655,
|
|
"mean_token_accuracy": 0.2324175089597702,
|
|
"num_tokens": 14833609.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.295270538330078,
|
|
"epoch": 6.614420062695925,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00018363083111979452,
|
|
"loss": 4.5168,
|
|
"mean_token_accuracy": 0.23915291875600814,
|
|
"num_tokens": 14841967.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.331786394119263,
|
|
"epoch": 6.618338557993731,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00018335603963251184,
|
|
"loss": 4.5693,
|
|
"mean_token_accuracy": 0.2304493635892868,
|
|
"num_tokens": 14851472.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.410040283203125,
|
|
"epoch": 6.622257053291536,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018308141192784573,
|
|
"loss": 4.5325,
|
|
"mean_token_accuracy": 0.23818644881248474,
|
|
"num_tokens": 14859672.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.350308609008789,
|
|
"epoch": 6.626175548589342,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018280694849660034,
|
|
"loss": 4.4619,
|
|
"mean_token_accuracy": 0.24607438147068023,
|
|
"num_tokens": 14867919.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.302391195297242,
|
|
"epoch": 6.630094043887147,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00018253264982928636,
|
|
"loss": 4.4403,
|
|
"mean_token_accuracy": 0.24121885746717453,
|
|
"num_tokens": 14875670.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.318087005615235,
|
|
"epoch": 6.634012539184953,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00018225851641612007,
|
|
"loss": 4.4871,
|
|
"mean_token_accuracy": 0.23785406947135926,
|
|
"num_tokens": 14883835.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.253361606597901,
|
|
"epoch": 6.637931034482759,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001819845487470223,
|
|
"loss": 4.4347,
|
|
"mean_token_accuracy": 0.24458152949810028,
|
|
"num_tokens": 14892296.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.27231388092041,
|
|
"epoch": 6.641849529780564,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018171074731161758,
|
|
"loss": 4.4751,
|
|
"mean_token_accuracy": 0.2450479820370674,
|
|
"num_tokens": 14901334.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.240012979507446,
|
|
"epoch": 6.64576802507837,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018143711259923363,
|
|
"loss": 4.5046,
|
|
"mean_token_accuracy": 0.24285112768411637,
|
|
"num_tokens": 14910327.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.369470405578613,
|
|
"epoch": 6.649686520376176,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001811636450989,
|
|
"loss": 4.5866,
|
|
"mean_token_accuracy": 0.23242564201354982,
|
|
"num_tokens": 14919513.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.2134678840637205,
|
|
"epoch": 6.653605015673981,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001808903452993474,
|
|
"loss": 4.3721,
|
|
"mean_token_accuracy": 0.24245413839817048,
|
|
"num_tokens": 14928127.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.342692899703979,
|
|
"epoch": 6.657523510971787,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001806172136890068,
|
|
"loss": 4.4863,
|
|
"mean_token_accuracy": 0.244010728597641,
|
|
"num_tokens": 14936771.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.336951780319214,
|
|
"epoch": 6.661442006269592,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001803442507560089,
|
|
"loss": 4.466,
|
|
"mean_token_accuracy": 0.24665225595235823,
|
|
"num_tokens": 14945423.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 6.661442006269592,
|
|
"eval_entropy": 5.141951251399609,
|
|
"eval_loss": 5.620510578155518,
|
|
"eval_mean_token_accuracy": 0.17555470301141574,
|
|
"eval_num_tokens": 14945423.0,
|
|
"eval_runtime": 2.1732,
|
|
"eval_samples_per_second": 1896.756,
|
|
"eval_steps_per_second": 237.44,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.258754014968872,
|
|
"epoch": 6.665360501567398,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001800714569881825,
|
|
"loss": 4.4688,
|
|
"mean_token_accuracy": 0.24992128014564513,
|
|
"num_tokens": 14953859.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.235205030441284,
|
|
"epoch": 6.669278996865204,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017979883287305454,
|
|
"loss": 4.4567,
|
|
"mean_token_accuracy": 0.24309080243110656,
|
|
"num_tokens": 14962598.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.328983879089355,
|
|
"epoch": 6.673197492163009,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017952637889784824,
|
|
"loss": 4.4709,
|
|
"mean_token_accuracy": 0.2452207699418068,
|
|
"num_tokens": 14971606.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.283362913131714,
|
|
"epoch": 6.677115987460815,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00017925409554948329,
|
|
"loss": 4.5021,
|
|
"mean_token_accuracy": 0.24301386177539824,
|
|
"num_tokens": 14979686.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.3306684494018555,
|
|
"epoch": 6.681034482758621,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00017898198331457393,
|
|
"loss": 4.5837,
|
|
"mean_token_accuracy": 0.22964574545621871,
|
|
"num_tokens": 14988977.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.319819116592408,
|
|
"epoch": 6.684952978056426,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00017871004267942908,
|
|
"loss": 4.5968,
|
|
"mean_token_accuracy": 0.2281629592180252,
|
|
"num_tokens": 14996871.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.327080488204956,
|
|
"epoch": 6.688871473354232,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00017843827413005064,
|
|
"loss": 4.5197,
|
|
"mean_token_accuracy": 0.23595262318849564,
|
|
"num_tokens": 15006022.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.279099369049073,
|
|
"epoch": 6.692789968652038,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001781666781521331,
|
|
"loss": 4.5393,
|
|
"mean_token_accuracy": 0.23716886937618256,
|
|
"num_tokens": 15014827.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.328993129730224,
|
|
"epoch": 6.696708463949843,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00017789525523106243,
|
|
"loss": 4.4475,
|
|
"mean_token_accuracy": 0.24349211454391478,
|
|
"num_tokens": 15023157.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.357423877716064,
|
|
"epoch": 6.700626959247649,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017762400585191545,
|
|
"loss": 4.525,
|
|
"mean_token_accuracy": 0.23230668604373933,
|
|
"num_tokens": 15032539.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.395616769790649,
|
|
"epoch": 6.704545454545455,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00017735293049945888,
|
|
"loss": 4.6565,
|
|
"mean_token_accuracy": 0.22531453818082808,
|
|
"num_tokens": 15041527.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.277464151382446,
|
|
"epoch": 6.70846394984326,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017708202965814807,
|
|
"loss": 4.4995,
|
|
"mean_token_accuracy": 0.23695143014192582,
|
|
"num_tokens": 15050794.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.293949460983276,
|
|
"epoch": 6.712382445141066,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00017681130381212707,
|
|
"loss": 4.4924,
|
|
"mean_token_accuracy": 0.24017928391695023,
|
|
"num_tokens": 15059357.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.391639995574951,
|
|
"epoch": 6.716300940438871,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00017654075344522652,
|
|
"loss": 4.5789,
|
|
"mean_token_accuracy": 0.23095313757658004,
|
|
"num_tokens": 15067668.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.326108980178833,
|
|
"epoch": 6.720219435736677,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017627037904096416,
|
|
"loss": 4.4842,
|
|
"mean_token_accuracy": 0.23270707577466965,
|
|
"num_tokens": 15076463.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.245178747177124,
|
|
"epoch": 6.724137931034483,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017600018108254257,
|
|
"loss": 4.4771,
|
|
"mean_token_accuracy": 0.23389974981546402,
|
|
"num_tokens": 15085024.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.2689989566802975,
|
|
"epoch": 6.728056426332288,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017573016005284958,
|
|
"loss": 4.4701,
|
|
"mean_token_accuracy": 0.23898242115974427,
|
|
"num_tokens": 15093735.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.274590682983399,
|
|
"epoch": 6.731974921630094,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00017546031643445636,
|
|
"loss": 4.4845,
|
|
"mean_token_accuracy": 0.23685145676136016,
|
|
"num_tokens": 15102274.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.317244195938111,
|
|
"epoch": 6.735893416927899,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001751906507096172,
|
|
"loss": 4.4922,
|
|
"mean_token_accuracy": 0.24877437353134155,
|
|
"num_tokens": 15110762.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.266567420959473,
|
|
"epoch": 6.739811912225705,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001749211633602687,
|
|
"loss": 4.4713,
|
|
"mean_token_accuracy": 0.24148166328668594,
|
|
"num_tokens": 15119104.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.321460628509522,
|
|
"epoch": 6.743730407523511,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017465185486802826,
|
|
"loss": 4.4701,
|
|
"mean_token_accuracy": 0.24588530361652375,
|
|
"num_tokens": 15127351.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.370578145980835,
|
|
"epoch": 6.747648902821316,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00017438272571419373,
|
|
"loss": 4.6143,
|
|
"mean_token_accuracy": 0.23107295036315917,
|
|
"num_tokens": 15136246.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.301373624801636,
|
|
"epoch": 6.751567398119122,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 4.5477,
|
|
"mean_token_accuracy": 0.23124226927757263,
|
|
"num_tokens": 15144633.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.303641891479492,
|
|
"epoch": 6.755485893416928,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00017384500734533137,
|
|
"loss": 4.5145,
|
|
"mean_token_accuracy": 0.23465587049722672,
|
|
"num_tokens": 15153496.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.275722932815552,
|
|
"epoch": 6.759404388714733,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017357641909129324,
|
|
"loss": 4.5035,
|
|
"mean_token_accuracy": 0.2398765429854393,
|
|
"num_tokens": 15162242.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.339338207244873,
|
|
"epoch": 6.763322884012539,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017330801209763938,
|
|
"loss": 4.5696,
|
|
"mean_token_accuracy": 0.2335802361369133,
|
|
"num_tokens": 15170001.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.358608245849609,
|
|
"epoch": 6.767241379310345,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017303978684405635,
|
|
"loss": 4.4988,
|
|
"mean_token_accuracy": 0.23868305683135987,
|
|
"num_tokens": 15178086.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.328572940826416,
|
|
"epoch": 6.77115987460815,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017277174380990647,
|
|
"loss": 4.5642,
|
|
"mean_token_accuracy": 0.23501645475625993,
|
|
"num_tokens": 15187087.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.3455705642700195,
|
|
"epoch": 6.775078369905956,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00017250388347422578,
|
|
"loss": 4.5248,
|
|
"mean_token_accuracy": 0.24516467303037642,
|
|
"num_tokens": 15195850.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.352624177932739,
|
|
"epoch": 6.778996865203762,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017223620631572433,
|
|
"loss": 4.5446,
|
|
"mean_token_accuracy": 0.24551263004541396,
|
|
"num_tokens": 15204174.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.342869091033935,
|
|
"epoch": 6.782915360501567,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017196871281278465,
|
|
"loss": 4.4417,
|
|
"mean_token_accuracy": 0.2473462700843811,
|
|
"num_tokens": 15212640.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.352691888809204,
|
|
"epoch": 6.786833855799373,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00017170140344346086,
|
|
"loss": 4.4855,
|
|
"mean_token_accuracy": 0.23925651013851165,
|
|
"num_tokens": 15220996.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.275291776657104,
|
|
"epoch": 6.790752351097178,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001714342786854781,
|
|
"loss": 4.5307,
|
|
"mean_token_accuracy": 0.23569363206624985,
|
|
"num_tokens": 15229511.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.302538204193115,
|
|
"epoch": 6.794670846394984,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00017116733901623182,
|
|
"loss": 4.5892,
|
|
"mean_token_accuracy": 0.22992108315229415,
|
|
"num_tokens": 15238673.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.3159160137176515,
|
|
"epoch": 6.79858934169279,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017090058491278637,
|
|
"loss": 4.5093,
|
|
"mean_token_accuracy": 0.23259403854608535,
|
|
"num_tokens": 15246717.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.331652593612671,
|
|
"epoch": 6.802507836990595,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017063401685187457,
|
|
"loss": 4.4979,
|
|
"mean_token_accuracy": 0.23683008402585984,
|
|
"num_tokens": 15255543.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.266857862472534,
|
|
"epoch": 6.806426332288401,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017036763530989688,
|
|
"loss": 4.4281,
|
|
"mean_token_accuracy": 0.24027005583047867,
|
|
"num_tokens": 15264275.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.3299307346344,
|
|
"epoch": 6.810344827586206,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00017010144076292012,
|
|
"loss": 4.5136,
|
|
"mean_token_accuracy": 0.2416551098227501,
|
|
"num_tokens": 15272376.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.353287887573242,
|
|
"epoch": 6.814263322884012,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016983543368667722,
|
|
"loss": 4.5802,
|
|
"mean_token_accuracy": 0.22760017663240434,
|
|
"num_tokens": 15282362.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.308615684509277,
|
|
"epoch": 6.818181818181818,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016956961455656607,
|
|
"loss": 4.5427,
|
|
"mean_token_accuracy": 0.23233690559864045,
|
|
"num_tokens": 15290934.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.300383234024048,
|
|
"epoch": 6.822100313479623,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016930398384764836,
|
|
"loss": 4.4553,
|
|
"mean_token_accuracy": 0.2516689419746399,
|
|
"num_tokens": 15299743.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.350294494628907,
|
|
"epoch": 6.826018808777429,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00016903854203464939,
|
|
"loss": 4.5275,
|
|
"mean_token_accuracy": 0.23712757229804993,
|
|
"num_tokens": 15307858.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.292050409317016,
|
|
"epoch": 6.829937304075235,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00016877328959195653,
|
|
"loss": 4.4819,
|
|
"mean_token_accuracy": 0.23526821732521058,
|
|
"num_tokens": 15316829.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.350876569747925,
|
|
"epoch": 6.83385579937304,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016850822699361913,
|
|
"loss": 4.5894,
|
|
"mean_token_accuracy": 0.2321299657225609,
|
|
"num_tokens": 15325394.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.300871515274048,
|
|
"epoch": 6.837774294670846,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001682433547133469,
|
|
"loss": 4.5192,
|
|
"mean_token_accuracy": 0.23438120186328887,
|
|
"num_tokens": 15333988.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.345728254318237,
|
|
"epoch": 6.841692789968652,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016797867322450948,
|
|
"loss": 4.5501,
|
|
"mean_token_accuracy": 0.23501222729682922,
|
|
"num_tokens": 15343107.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.363087797164917,
|
|
"epoch": 6.8456112852664575,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001677141830001358,
|
|
"loss": 4.4803,
|
|
"mean_token_accuracy": 0.24444160908460616,
|
|
"num_tokens": 15351440.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.262495851516723,
|
|
"epoch": 6.849529780564263,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00016744988451291266,
|
|
"loss": 4.4522,
|
|
"mean_token_accuracy": 0.23560455888509751,
|
|
"num_tokens": 15360426.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.288626194000244,
|
|
"epoch": 6.853448275862069,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016718577823518426,
|
|
"loss": 4.5205,
|
|
"mean_token_accuracy": 0.23618055135011673,
|
|
"num_tokens": 15369252.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.313988256454468,
|
|
"epoch": 6.8573667711598745,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016692186463895136,
|
|
"loss": 4.5421,
|
|
"mean_token_accuracy": 0.22964171767234803,
|
|
"num_tokens": 15377672.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.304130268096924,
|
|
"epoch": 6.86128526645768,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00016665814419587055,
|
|
"loss": 4.4581,
|
|
"mean_token_accuracy": 0.24935263991355897,
|
|
"num_tokens": 15385585.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.35563440322876,
|
|
"epoch": 6.8652037617554855,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001663946173772528,
|
|
"loss": 4.5327,
|
|
"mean_token_accuracy": 0.23600398302078246,
|
|
"num_tokens": 15394193.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.364314842224121,
|
|
"epoch": 6.8691222570532915,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016613128465406329,
|
|
"loss": 4.5501,
|
|
"mean_token_accuracy": 0.23776163458824157,
|
|
"num_tokens": 15402498.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.243097162246704,
|
|
"epoch": 6.8730407523510975,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001658681464969204,
|
|
"loss": 4.4074,
|
|
"mean_token_accuracy": 0.24998972713947296,
|
|
"num_tokens": 15410893.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.3027020454406735,
|
|
"epoch": 6.8769592476489025,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016560520337609465,
|
|
"loss": 4.583,
|
|
"mean_token_accuracy": 0.23093927502632142,
|
|
"num_tokens": 15420024.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.213108921051026,
|
|
"epoch": 6.8808777429467085,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016534245576150787,
|
|
"loss": 4.4912,
|
|
"mean_token_accuracy": 0.2345226839184761,
|
|
"num_tokens": 15428474.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.277981376647949,
|
|
"epoch": 6.8847962382445145,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001650799041227329,
|
|
"loss": 4.4325,
|
|
"mean_token_accuracy": 0.24551166445016862,
|
|
"num_tokens": 15437019.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.375722599029541,
|
|
"epoch": 6.88871473354232,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00016481754892899193,
|
|
"loss": 4.5372,
|
|
"mean_token_accuracy": 0.24411578774452208,
|
|
"num_tokens": 15445838.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.3416351795196535,
|
|
"epoch": 6.8926332288401255,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001645553906491563,
|
|
"loss": 4.5216,
|
|
"mean_token_accuracy": 0.24373453110456467,
|
|
"num_tokens": 15454378.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.3107963562011715,
|
|
"epoch": 6.896551724137931,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001642934297517452,
|
|
"loss": 4.5325,
|
|
"mean_token_accuracy": 0.23505257517099382,
|
|
"num_tokens": 15462447.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.259591770172119,
|
|
"epoch": 6.900470219435737,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001640316667049254,
|
|
"loss": 4.5063,
|
|
"mean_token_accuracy": 0.23882172703742982,
|
|
"num_tokens": 15471646.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.33168478012085,
|
|
"epoch": 6.9043887147335425,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016377010197650993,
|
|
"loss": 4.5224,
|
|
"mean_token_accuracy": 0.23694578558206558,
|
|
"num_tokens": 15481945.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.3250690460205075,
|
|
"epoch": 6.908307210031348,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001635087360339572,
|
|
"loss": 4.4853,
|
|
"mean_token_accuracy": 0.24775975048542023,
|
|
"num_tokens": 15490558.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.339182376861572,
|
|
"epoch": 6.912225705329154,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016324756934437074,
|
|
"loss": 4.478,
|
|
"mean_token_accuracy": 0.23855846971273423,
|
|
"num_tokens": 15499150.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.2792116641998295,
|
|
"epoch": 6.91614420062696,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001629866023744977,
|
|
"loss": 4.4613,
|
|
"mean_token_accuracy": 0.23632628619670867,
|
|
"num_tokens": 15508736.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.371030664443969,
|
|
"epoch": 6.920062695924765,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00016272583559072832,
|
|
"loss": 4.5421,
|
|
"mean_token_accuracy": 0.24001597464084626,
|
|
"num_tokens": 15517445.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.342490577697754,
|
|
"epoch": 6.923981191222571,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00016246526945909505,
|
|
"loss": 4.5448,
|
|
"mean_token_accuracy": 0.24060498774051667,
|
|
"num_tokens": 15526282.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.372104215621948,
|
|
"epoch": 6.927899686520377,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016220490444527198,
|
|
"loss": 4.5379,
|
|
"mean_token_accuracy": 0.24220004677772522,
|
|
"num_tokens": 15534776.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.310294580459595,
|
|
"epoch": 6.931818181818182,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001619447410145735,
|
|
"loss": 4.4952,
|
|
"mean_token_accuracy": 0.2463688850402832,
|
|
"num_tokens": 15543173.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.266290760040283,
|
|
"epoch": 6.935736677115988,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016168477963195382,
|
|
"loss": 4.4218,
|
|
"mean_token_accuracy": 0.24300505518913268,
|
|
"num_tokens": 15551603.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.2556915283203125,
|
|
"epoch": 6.939655172413794,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00016142502076200622,
|
|
"loss": 4.4847,
|
|
"mean_token_accuracy": 0.2408796012401581,
|
|
"num_tokens": 15560412.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.262258148193359,
|
|
"epoch": 6.943573667711599,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016116546486896183,
|
|
"loss": 4.4741,
|
|
"mean_token_accuracy": 0.2376261904835701,
|
|
"num_tokens": 15569781.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.324252891540527,
|
|
"epoch": 6.947492163009405,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00016090611241668904,
|
|
"loss": 4.4619,
|
|
"mean_token_accuracy": 0.24739525020122527,
|
|
"num_tokens": 15577640.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.364842557907105,
|
|
"epoch": 6.95141065830721,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001606469638686928,
|
|
"loss": 4.5848,
|
|
"mean_token_accuracy": 0.23459223061800002,
|
|
"num_tokens": 15586950.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.286621952056885,
|
|
"epoch": 6.955329153605016,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001603880196881137,
|
|
"loss": 4.5483,
|
|
"mean_token_accuracy": 0.24004509150981904,
|
|
"num_tokens": 15595554.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.317348861694336,
|
|
"epoch": 6.959247648902822,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0001601292803377269,
|
|
"loss": 4.4961,
|
|
"mean_token_accuracy": 0.24077611267566681,
|
|
"num_tokens": 15603998.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.370563745498657,
|
|
"epoch": 6.963166144200627,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001598707462799414,
|
|
"loss": 4.58,
|
|
"mean_token_accuracy": 0.2309919461607933,
|
|
"num_tokens": 15612977.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.369113492965698,
|
|
"epoch": 6.967084639498433,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00015961241797679977,
|
|
"loss": 4.5615,
|
|
"mean_token_accuracy": 0.23333321809768676,
|
|
"num_tokens": 15622914.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.376282835006714,
|
|
"epoch": 6.971003134796238,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015935429588997637,
|
|
"loss": 4.5276,
|
|
"mean_token_accuracy": 0.2361798793077469,
|
|
"num_tokens": 15631929.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.242317390441895,
|
|
"epoch": 6.974921630094044,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015909638048077724,
|
|
"loss": 4.3874,
|
|
"mean_token_accuracy": 0.2487921819090843,
|
|
"num_tokens": 15640400.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.3869775295257565,
|
|
"epoch": 6.97884012539185,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015883867221013915,
|
|
"loss": 4.5924,
|
|
"mean_token_accuracy": 0.2395585298538208,
|
|
"num_tokens": 15648934.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.362801218032837,
|
|
"epoch": 6.982758620689655,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001585811715386285,
|
|
"loss": 4.5429,
|
|
"mean_token_accuracy": 0.2313330054283142,
|
|
"num_tokens": 15658461.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.244998979568481,
|
|
"epoch": 6.986677115987461,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015832387892644072,
|
|
"loss": 4.4545,
|
|
"mean_token_accuracy": 0.24810600876808167,
|
|
"num_tokens": 15667182.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.298608779907227,
|
|
"epoch": 6.990595611285267,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001580667948333994,
|
|
"loss": 4.5436,
|
|
"mean_token_accuracy": 0.23584588170051574,
|
|
"num_tokens": 15675951.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.330057621002197,
|
|
"epoch": 6.994514106583072,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001578099197189555,
|
|
"loss": 4.4989,
|
|
"mean_token_accuracy": 0.2392742767930031,
|
|
"num_tokens": 15684171.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.2751157760620115,
|
|
"epoch": 6.998432601880878,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00015755325404218662,
|
|
"loss": 4.4963,
|
|
"mean_token_accuracy": 0.24457567930221558,
|
|
"num_tokens": 15692986.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.377701234817505,
|
|
"epoch": 7.002351097178684,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00015729679826179577,
|
|
"loss": 4.4964,
|
|
"mean_token_accuracy": 0.24556155353784562,
|
|
"num_tokens": 15702591.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.333701324462891,
|
|
"epoch": 7.006269592476489,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015704055283611123,
|
|
"loss": 4.4135,
|
|
"mean_token_accuracy": 0.2451925203204155,
|
|
"num_tokens": 15711186.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.289366769790649,
|
|
"epoch": 7.010188087774295,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00015678451822308495,
|
|
"loss": 4.4303,
|
|
"mean_token_accuracy": 0.2533035665750504,
|
|
"num_tokens": 15719831.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.380713748931885,
|
|
"epoch": 7.0141065830721,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001565286948802923,
|
|
"loss": 4.4771,
|
|
"mean_token_accuracy": 0.24241483509540557,
|
|
"num_tokens": 15727926.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.271444845199585,
|
|
"epoch": 7.018025078369906,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00015627308326493118,
|
|
"loss": 4.3444,
|
|
"mean_token_accuracy": 0.258765310049057,
|
|
"num_tokens": 15735934.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.201727342605591,
|
|
"epoch": 7.021943573667712,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001560176838338209,
|
|
"loss": 4.4143,
|
|
"mean_token_accuracy": 0.24798694252967834,
|
|
"num_tokens": 15745014.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.223081731796265,
|
|
"epoch": 7.025862068965517,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00015576249704340173,
|
|
"loss": 4.3565,
|
|
"mean_token_accuracy": 0.25011841803789137,
|
|
"num_tokens": 15753360.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.277637481689453,
|
|
"epoch": 7.029780564263323,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00015550752334973368,
|
|
"loss": 4.3912,
|
|
"mean_token_accuracy": 0.25387850105762483,
|
|
"num_tokens": 15762395.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.298357391357422,
|
|
"epoch": 7.033699059561129,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015525276320849628,
|
|
"loss": 4.3221,
|
|
"mean_token_accuracy": 0.26269663870334625,
|
|
"num_tokens": 15771292.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.342686319351197,
|
|
"epoch": 7.037617554858934,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015499821707498702,
|
|
"loss": 4.4258,
|
|
"mean_token_accuracy": 0.24899108111858367,
|
|
"num_tokens": 15779879.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.305684423446655,
|
|
"epoch": 7.04153605015674,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00015474388540412138,
|
|
"loss": 4.4079,
|
|
"mean_token_accuracy": 0.249126136302948,
|
|
"num_tokens": 15788960.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.33590726852417,
|
|
"epoch": 7.045454545454546,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000154489768650431,
|
|
"loss": 4.4531,
|
|
"mean_token_accuracy": 0.2427803710103035,
|
|
"num_tokens": 15798260.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.300719499588013,
|
|
"epoch": 7.049373040752351,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00015423586726806394,
|
|
"loss": 4.4418,
|
|
"mean_token_accuracy": 0.24532065093517302,
|
|
"num_tokens": 15807480.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.32033143043518,
|
|
"epoch": 7.053291536050157,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00015398218171078305,
|
|
"loss": 4.51,
|
|
"mean_token_accuracy": 0.24209006130695343,
|
|
"num_tokens": 15816330.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 7.053291536050157,
|
|
"eval_entropy": 5.1433738507041635,
|
|
"eval_loss": 5.6254730224609375,
|
|
"eval_mean_token_accuracy": 0.17532913560091062,
|
|
"eval_num_tokens": 15816330.0,
|
|
"eval_runtime": 2.3531,
|
|
"eval_samples_per_second": 1751.738,
|
|
"eval_steps_per_second": 219.286,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 5.283837175369262,
|
|
"epoch": 7.057210031347962,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015372871243196545,
|
|
"loss": 4.4076,
|
|
"mean_token_accuracy": 0.25260193943977355,
|
|
"num_tokens": 15824600.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 5.327095317840576,
|
|
"epoch": 7.061128526645768,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001534754598846021,
|
|
"loss": 4.4453,
|
|
"mean_token_accuracy": 0.24223601818084717,
|
|
"num_tokens": 15834170.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 5.374682664871216,
|
|
"epoch": 7.065047021943574,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00015322242452129616,
|
|
"loss": 4.4718,
|
|
"mean_token_accuracy": 0.24198867827653886,
|
|
"num_tokens": 15843061.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 5.318432140350342,
|
|
"epoch": 7.068965517241379,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015296960679426285,
|
|
"loss": 4.4219,
|
|
"mean_token_accuracy": 0.2435888648033142,
|
|
"num_tokens": 15851788.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 5.2973731517791744,
|
|
"epoch": 7.072884012539185,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015271700715532854,
|
|
"loss": 4.4528,
|
|
"mean_token_accuracy": 0.24408562928438188,
|
|
"num_tokens": 15860216.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 5.290932750701904,
|
|
"epoch": 7.076802507836991,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015246462605592972,
|
|
"loss": 4.3237,
|
|
"mean_token_accuracy": 0.26170015037059785,
|
|
"num_tokens": 15869498.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 5.293484830856324,
|
|
"epoch": 7.080721003134796,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00015221246394711224,
|
|
"loss": 4.3745,
|
|
"mean_token_accuracy": 0.2565301164984703,
|
|
"num_tokens": 15877981.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 5.334282970428466,
|
|
"epoch": 7.084639498432602,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00015196052127953085,
|
|
"loss": 4.4197,
|
|
"mean_token_accuracy": 0.24756770879030227,
|
|
"num_tokens": 15886454.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 5.242758798599243,
|
|
"epoch": 7.088557993730408,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001517087985034477,
|
|
"loss": 4.3358,
|
|
"mean_token_accuracy": 0.2540945500135422,
|
|
"num_tokens": 15895025.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 5.317369222640991,
|
|
"epoch": 7.092476489028213,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001514572960687325,
|
|
"loss": 4.4811,
|
|
"mean_token_accuracy": 0.24123733639717101,
|
|
"num_tokens": 15903611.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.349587059020996,
|
|
"epoch": 7.096394984326019,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015120601442486061,
|
|
"loss": 4.4602,
|
|
"mean_token_accuracy": 0.24597056806087494,
|
|
"num_tokens": 15912078.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.271403741836548,
|
|
"epoch": 7.100313479623824,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00015095495402091329,
|
|
"loss": 4.3701,
|
|
"mean_token_accuracy": 0.2557546466588974,
|
|
"num_tokens": 15920914.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 5.311720275878907,
|
|
"epoch": 7.10423197492163,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001507041153055761,
|
|
"loss": 4.3442,
|
|
"mean_token_accuracy": 0.2513896390795708,
|
|
"num_tokens": 15929199.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.396252250671386,
|
|
"epoch": 7.108150470219436,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00015045349872713838,
|
|
"loss": 4.5099,
|
|
"mean_token_accuracy": 0.24451833814382554,
|
|
"num_tokens": 15937817.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 5.302476978302002,
|
|
"epoch": 7.112068965517241,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00015020310473349275,
|
|
"loss": 4.4285,
|
|
"mean_token_accuracy": 0.2534029483795166,
|
|
"num_tokens": 15945997.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 5.284367895126342,
|
|
"epoch": 7.115987460815047,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014995293377213388,
|
|
"loss": 4.3841,
|
|
"mean_token_accuracy": 0.2426355317234993,
|
|
"num_tokens": 15954215.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 5.311365270614624,
|
|
"epoch": 7.119905956112853,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0001497029862901578,
|
|
"loss": 4.4325,
|
|
"mean_token_accuracy": 0.2523935303092003,
|
|
"num_tokens": 15962222.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 5.3349365234375,
|
|
"epoch": 7.123824451410658,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00014945326273426102,
|
|
"loss": 4.3736,
|
|
"mean_token_accuracy": 0.2531398072838783,
|
|
"num_tokens": 15970712.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 5.335588645935059,
|
|
"epoch": 7.127742946708464,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00014920376355074034,
|
|
"loss": 4.4512,
|
|
"mean_token_accuracy": 0.2434644028544426,
|
|
"num_tokens": 15979049.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 5.256621217727661,
|
|
"epoch": 7.131661442006269,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014895448918549097,
|
|
"loss": 4.4004,
|
|
"mean_token_accuracy": 0.25194070637226107,
|
|
"num_tokens": 15988221.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 5.300654745101928,
|
|
"epoch": 7.135579937304075,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014870544008400681,
|
|
"loss": 4.3878,
|
|
"mean_token_accuracy": 0.24303680211305617,
|
|
"num_tokens": 15996579.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 5.285613536834717,
|
|
"epoch": 7.139498432601881,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001484566166913789,
|
|
"loss": 4.4196,
|
|
"mean_token_accuracy": 0.2465776026248932,
|
|
"num_tokens": 16005489.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 5.2754134178161625,
|
|
"epoch": 7.143416927899686,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00014820801945229504,
|
|
"loss": 4.3813,
|
|
"mean_token_accuracy": 0.252704431116581,
|
|
"num_tokens": 16014167.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 5.274964666366577,
|
|
"epoch": 7.147335423197492,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001479596488110387,
|
|
"loss": 4.2992,
|
|
"mean_token_accuracy": 0.262321001291275,
|
|
"num_tokens": 16022184.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 5.314197206497193,
|
|
"epoch": 7.151253918495298,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014771150521148863,
|
|
"loss": 4.4353,
|
|
"mean_token_accuracy": 0.2457068994641304,
|
|
"num_tokens": 16031704.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 5.319581413269043,
|
|
"epoch": 7.155172413793103,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001474635890971176,
|
|
"loss": 4.4724,
|
|
"mean_token_accuracy": 0.24409542232751846,
|
|
"num_tokens": 16040913.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 5.274930572509765,
|
|
"epoch": 7.159090909090909,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001472159009109919,
|
|
"loss": 4.3837,
|
|
"mean_token_accuracy": 0.24728970378637313,
|
|
"num_tokens": 16048943.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 5.287972736358642,
|
|
"epoch": 7.163009404388715,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00014696844109577043,
|
|
"loss": 4.3695,
|
|
"mean_token_accuracy": 0.24714432507753373,
|
|
"num_tokens": 16057999.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 5.3387922763824465,
|
|
"epoch": 7.16692789968652,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014672121009370415,
|
|
"loss": 4.5011,
|
|
"mean_token_accuracy": 0.2341273695230484,
|
|
"num_tokens": 16066920.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 5.2621378898620605,
|
|
"epoch": 7.170846394984326,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014647420834663482,
|
|
"loss": 4.4517,
|
|
"mean_token_accuracy": 0.23665154725313187,
|
|
"num_tokens": 16075128.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 5.281534147262573,
|
|
"epoch": 7.174764890282131,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001462274362959946,
|
|
"loss": 4.4244,
|
|
"mean_token_accuracy": 0.25137183517217637,
|
|
"num_tokens": 16083949.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 5.310236120223999,
|
|
"epoch": 7.178683385579937,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001459808943828053,
|
|
"loss": 4.3891,
|
|
"mean_token_accuracy": 0.2532069981098175,
|
|
"num_tokens": 16092891.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 5.3306091785430905,
|
|
"epoch": 7.182601880877743,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014573458304767707,
|
|
"loss": 4.4713,
|
|
"mean_token_accuracy": 0.2345483645796776,
|
|
"num_tokens": 16101602.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 5.33755464553833,
|
|
"epoch": 7.186520376175548,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014548850273080839,
|
|
"loss": 4.4538,
|
|
"mean_token_accuracy": 0.24318726360797882,
|
|
"num_tokens": 16110430.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 5.26176872253418,
|
|
"epoch": 7.190438871473354,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014524265387198465,
|
|
"loss": 4.398,
|
|
"mean_token_accuracy": 0.24676933139562607,
|
|
"num_tokens": 16119444.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 5.264621019363403,
|
|
"epoch": 7.19435736677116,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00014499703691057766,
|
|
"loss": 4.371,
|
|
"mean_token_accuracy": 0.2507361188530922,
|
|
"num_tokens": 16128721.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 5.338419008255005,
|
|
"epoch": 7.198275862068965,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001447516522855447,
|
|
"loss": 4.479,
|
|
"mean_token_accuracy": 0.24437074959278107,
|
|
"num_tokens": 16137508.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 5.3357714176177975,
|
|
"epoch": 7.202194357366771,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00014450650043542775,
|
|
"loss": 4.4077,
|
|
"mean_token_accuracy": 0.2500804975628853,
|
|
"num_tokens": 16146239.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.306744146347046,
|
|
"epoch": 7.206112852664576,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014426158179835317,
|
|
"loss": 4.394,
|
|
"mean_token_accuracy": 0.24388739168643953,
|
|
"num_tokens": 16154769.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 5.221017980575562,
|
|
"epoch": 7.210031347962382,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001440168968120301,
|
|
"loss": 4.294,
|
|
"mean_token_accuracy": 0.26560360193252563,
|
|
"num_tokens": 16163307.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.291001653671264,
|
|
"epoch": 7.213949843260188,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00014377244591375033,
|
|
"loss": 4.3914,
|
|
"mean_token_accuracy": 0.24354729056358337,
|
|
"num_tokens": 16171533.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 5.255368757247925,
|
|
"epoch": 7.217868338557993,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001435282295403873,
|
|
"loss": 4.3825,
|
|
"mean_token_accuracy": 0.25670834332704545,
|
|
"num_tokens": 16180163.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 5.246207857131958,
|
|
"epoch": 7.221786833855799,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014328424812839525,
|
|
"loss": 4.3657,
|
|
"mean_token_accuracy": 0.25452733039855957,
|
|
"num_tokens": 16188060.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 5.273656034469605,
|
|
"epoch": 7.225705329153605,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00014304050211380836,
|
|
"loss": 4.461,
|
|
"mean_token_accuracy": 0.24166370630264283,
|
|
"num_tokens": 16197597.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 5.316214752197266,
|
|
"epoch": 7.22962382445141,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00014279699193224047,
|
|
"loss": 4.4654,
|
|
"mean_token_accuracy": 0.2310787096619606,
|
|
"num_tokens": 16206288.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 5.295995759963989,
|
|
"epoch": 7.233542319749216,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014255371801888379,
|
|
"loss": 4.4282,
|
|
"mean_token_accuracy": 0.2411588177084923,
|
|
"num_tokens": 16215321.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 5.306982898712159,
|
|
"epoch": 7.237460815047022,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00014231068080850813,
|
|
"loss": 4.3422,
|
|
"mean_token_accuracy": 0.2609937027096748,
|
|
"num_tokens": 16224445.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 5.287397336959839,
|
|
"epoch": 7.241379310344827,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014206788073546035,
|
|
"loss": 4.3854,
|
|
"mean_token_accuracy": 0.25974967181682584,
|
|
"num_tokens": 16233189.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 5.262801456451416,
|
|
"epoch": 7.245297805642633,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00014182531823366372,
|
|
"loss": 4.3744,
|
|
"mean_token_accuracy": 0.2589182645082474,
|
|
"num_tokens": 16241664.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 5.251544713973999,
|
|
"epoch": 7.2492163009404385,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014158299373661666,
|
|
"loss": 4.4311,
|
|
"mean_token_accuracy": 0.24906164556741714,
|
|
"num_tokens": 16251293.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 5.333122682571411,
|
|
"epoch": 7.253134796238244,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014134090767739223,
|
|
"loss": 4.3968,
|
|
"mean_token_accuracy": 0.2510964959859848,
|
|
"num_tokens": 16260134.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 5.329854154586792,
|
|
"epoch": 7.25705329153605,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00014109906048863772,
|
|
"loss": 4.4295,
|
|
"mean_token_accuracy": 0.2482133075594902,
|
|
"num_tokens": 16269027.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 5.305157136917114,
|
|
"epoch": 7.2609717868338555,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001408574526025731,
|
|
"loss": 4.4473,
|
|
"mean_token_accuracy": 0.24935528934001921,
|
|
"num_tokens": 16277692.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 5.309747171401978,
|
|
"epoch": 7.264890282131661,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014061608445099083,
|
|
"loss": 4.4446,
|
|
"mean_token_accuracy": 0.23883333802223206,
|
|
"num_tokens": 16287059.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.218217849731445,
|
|
"epoch": 7.268808777429467,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.00014037495646525495,
|
|
"loss": 4.4051,
|
|
"mean_token_accuracy": 0.25291329324245454,
|
|
"num_tokens": 16294880.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 5.269831228256225,
|
|
"epoch": 7.2727272727272725,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001401340690763003,
|
|
"loss": 4.3598,
|
|
"mean_token_accuracy": 0.25545747131109237,
|
|
"num_tokens": 16304101.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 5.3249688148498535,
|
|
"epoch": 7.2766457680250785,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00013989342271463173,
|
|
"loss": 4.5028,
|
|
"mean_token_accuracy": 0.234570150077343,
|
|
"num_tokens": 16312887.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 5.260730934143067,
|
|
"epoch": 7.280564263322884,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001396530178103232,
|
|
"loss": 4.4078,
|
|
"mean_token_accuracy": 0.2483700305223465,
|
|
"num_tokens": 16322256.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 5.290597438812256,
|
|
"epoch": 7.2844827586206895,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001394128547930174,
|
|
"loss": 4.4316,
|
|
"mean_token_accuracy": 0.24239268749952317,
|
|
"num_tokens": 16331810.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 5.356947135925293,
|
|
"epoch": 7.2884012539184955,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001391729340919245,
|
|
"loss": 4.4435,
|
|
"mean_token_accuracy": 0.24665167033672333,
|
|
"num_tokens": 16341385.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 5.2776916980743405,
|
|
"epoch": 7.2923197492163006,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013893325613582172,
|
|
"loss": 4.3885,
|
|
"mean_token_accuracy": 0.2484171584248543,
|
|
"num_tokens": 16350231.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.337796640396118,
|
|
"epoch": 7.2962382445141065,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001386938213530523,
|
|
"loss": 4.4623,
|
|
"mean_token_accuracy": 0.2461619943380356,
|
|
"num_tokens": 16359659.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 5.314506578445434,
|
|
"epoch": 7.3001567398119125,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013845463017152527,
|
|
"loss": 4.4227,
|
|
"mean_token_accuracy": 0.2522300571203232,
|
|
"num_tokens": 16368606.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.3277442932128904,
|
|
"epoch": 7.304075235109718,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013821568301871384,
|
|
"loss": 4.4702,
|
|
"mean_token_accuracy": 0.24263031780719757,
|
|
"num_tokens": 16376428.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 5.277805137634277,
|
|
"epoch": 7.3079937304075235,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00013797698032165534,
|
|
"loss": 4.4143,
|
|
"mean_token_accuracy": 0.2502939999103546,
|
|
"num_tokens": 16384983.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 5.288567733764649,
|
|
"epoch": 7.3119122257053295,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00013773852250695035,
|
|
"loss": 4.4291,
|
|
"mean_token_accuracy": 0.24019632190465928,
|
|
"num_tokens": 16393747.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 5.2937255859375,
|
|
"epoch": 7.315830721003135,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00013750031000076145,
|
|
"loss": 4.434,
|
|
"mean_token_accuracy": 0.24965571165084838,
|
|
"num_tokens": 16401899.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.279248857498169,
|
|
"epoch": 7.3197492163009406,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001372623432288131,
|
|
"loss": 4.4303,
|
|
"mean_token_accuracy": 0.24672488123178482,
|
|
"num_tokens": 16410787.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 5.334970903396607,
|
|
"epoch": 7.3236677115987465,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00013702462261639054,
|
|
"loss": 4.4558,
|
|
"mean_token_accuracy": 0.24444268345832826,
|
|
"num_tokens": 16420023.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 5.309818363189697,
|
|
"epoch": 7.327586206896552,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000136787148588339,
|
|
"loss": 4.3706,
|
|
"mean_token_accuracy": 0.24907097220420837,
|
|
"num_tokens": 16428394.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.32716474533081,
|
|
"epoch": 7.331504702194358,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001365499215690632,
|
|
"loss": 4.3854,
|
|
"mean_token_accuracy": 0.25570245683193205,
|
|
"num_tokens": 16436429.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.243541860580445,
|
|
"epoch": 7.335423197492163,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013631294198252603,
|
|
"loss": 4.3333,
|
|
"mean_token_accuracy": 0.25850107073783873,
|
|
"num_tokens": 16445140.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 5.283485364913941,
|
|
"epoch": 7.339341692789969,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013607621025224864,
|
|
"loss": 4.371,
|
|
"mean_token_accuracy": 0.24377409517765045,
|
|
"num_tokens": 16453932.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 5.2577918529510494,
|
|
"epoch": 7.343260188087775,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000135839726801309,
|
|
"loss": 4.3981,
|
|
"mean_token_accuracy": 0.24727073162794114,
|
|
"num_tokens": 16462539.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 5.2948606491088865,
|
|
"epoch": 7.34717868338558,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013560349205234124,
|
|
"loss": 4.4422,
|
|
"mean_token_accuracy": 0.24491775929927825,
|
|
"num_tokens": 16471979.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.356369209289551,
|
|
"epoch": 7.351097178683386,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013536750642753538,
|
|
"loss": 4.5142,
|
|
"mean_token_accuracy": 0.24074720591306686,
|
|
"num_tokens": 16481266.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 5.300047397613525,
|
|
"epoch": 7.355015673981192,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013513177034863588,
|
|
"loss": 4.4641,
|
|
"mean_token_accuracy": 0.24815348386764527,
|
|
"num_tokens": 16490360.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 5.350519752502441,
|
|
"epoch": 7.358934169278997,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001348962842369413,
|
|
"loss": 4.4897,
|
|
"mean_token_accuracy": 0.23873317688703538,
|
|
"num_tokens": 16498043.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 5.295504236221314,
|
|
"epoch": 7.362852664576803,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013466104851330346,
|
|
"loss": 4.4307,
|
|
"mean_token_accuracy": 0.24975659549236298,
|
|
"num_tokens": 16506343.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 5.281968879699707,
|
|
"epoch": 7.366771159874608,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001344260635981268,
|
|
"loss": 4.4207,
|
|
"mean_token_accuracy": 0.2503260925412178,
|
|
"num_tokens": 16515308.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 5.309352254867553,
|
|
"epoch": 7.370689655172414,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013419132991136762,
|
|
"loss": 4.4276,
|
|
"mean_token_accuracy": 0.2411679059267044,
|
|
"num_tokens": 16524379.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 5.316375875473023,
|
|
"epoch": 7.37460815047022,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00013395684787253278,
|
|
"loss": 4.4376,
|
|
"mean_token_accuracy": 0.2441466733813286,
|
|
"num_tokens": 16532797.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.31067123413086,
|
|
"epoch": 7.378526645768025,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00013372261790068,
|
|
"loss": 4.4456,
|
|
"mean_token_accuracy": 0.24431059062480925,
|
|
"num_tokens": 16540984.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 5.283600950241089,
|
|
"epoch": 7.382445141065831,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013348864041441604,
|
|
"loss": 4.3486,
|
|
"mean_token_accuracy": 0.25452945232391355,
|
|
"num_tokens": 16549386.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 5.254391145706177,
|
|
"epoch": 7.386363636363637,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013325491583189664,
|
|
"loss": 4.4179,
|
|
"mean_token_accuracy": 0.24938802570104598,
|
|
"num_tokens": 16557673.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 5.317046594619751,
|
|
"epoch": 7.390282131661442,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013302144457082553,
|
|
"loss": 4.45,
|
|
"mean_token_accuracy": 0.24535531997680665,
|
|
"num_tokens": 16566383.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 5.315344905853271,
|
|
"epoch": 7.394200626959248,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00013278822704845372,
|
|
"loss": 4.4026,
|
|
"mean_token_accuracy": 0.2485069140791893,
|
|
"num_tokens": 16574819.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 5.316687202453613,
|
|
"epoch": 7.398119122257054,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013255526368157876,
|
|
"loss": 4.4313,
|
|
"mean_token_accuracy": 0.2517440989613533,
|
|
"num_tokens": 16582571.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 5.376802206039429,
|
|
"epoch": 7.402037617554859,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013232255488654387,
|
|
"loss": 4.5033,
|
|
"mean_token_accuracy": 0.2412169963121414,
|
|
"num_tokens": 16592045.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 5.302829837799072,
|
|
"epoch": 7.405956112852665,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013209010107923747,
|
|
"loss": 4.3702,
|
|
"mean_token_accuracy": 0.24961871206760405,
|
|
"num_tokens": 16600864.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 5.33931303024292,
|
|
"epoch": 7.40987460815047,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013185790267509218,
|
|
"loss": 4.5172,
|
|
"mean_token_accuracy": 0.2441935658454895,
|
|
"num_tokens": 16610193.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 5.282327795028687,
|
|
"epoch": 7.413793103448276,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00013162596008908423,
|
|
"loss": 4.4633,
|
|
"mean_token_accuracy": 0.23964929431676865,
|
|
"num_tokens": 16618663.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.295189237594604,
|
|
"epoch": 7.417711598746082,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013139427373573255,
|
|
"loss": 4.3798,
|
|
"mean_token_accuracy": 0.2521142542362213,
|
|
"num_tokens": 16626885.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 5.368727397918701,
|
|
"epoch": 7.421630094043887,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013116284402909834,
|
|
"loss": 4.4954,
|
|
"mean_token_accuracy": 0.24414896965026855,
|
|
"num_tokens": 16636592.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.329002380371094,
|
|
"epoch": 7.425548589341693,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000130931671382784,
|
|
"loss": 4.4892,
|
|
"mean_token_accuracy": 0.2419576331973076,
|
|
"num_tokens": 16645897.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 5.339089202880859,
|
|
"epoch": 7.429467084639499,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00013070075620993233,
|
|
"loss": 4.4228,
|
|
"mean_token_accuracy": 0.2420647457242012,
|
|
"num_tokens": 16654034.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 5.281978273391724,
|
|
"epoch": 7.433385579937304,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013047009892322642,
|
|
"loss": 4.4113,
|
|
"mean_token_accuracy": 0.25225159525871277,
|
|
"num_tokens": 16662510.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.2895246028900145,
|
|
"epoch": 7.43730407523511,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001302396999348882,
|
|
"loss": 4.3959,
|
|
"mean_token_accuracy": 0.25313336253166197,
|
|
"num_tokens": 16671014.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 5.301432085037232,
|
|
"epoch": 7.441222570532915,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00013000955965667788,
|
|
"loss": 4.462,
|
|
"mean_token_accuracy": 0.24404768496751786,
|
|
"num_tokens": 16680068.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 5.3534952163696286,
|
|
"epoch": 7.445141065830721,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001297796784998936,
|
|
"loss": 4.4899,
|
|
"mean_token_accuracy": 0.2528064280748367,
|
|
"num_tokens": 16688478.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 7.445141065830721,
|
|
"eval_entropy": 5.163850973742877,
|
|
"eval_loss": 5.624318599700928,
|
|
"eval_mean_token_accuracy": 0.1758084057490955,
|
|
"eval_num_tokens": 16688478.0,
|
|
"eval_runtime": 2.1693,
|
|
"eval_samples_per_second": 1900.175,
|
|
"eval_steps_per_second": 237.868,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 5.355221843719482,
|
|
"epoch": 7.449059561128527,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00012955005687537025,
|
|
"loss": 4.5383,
|
|
"mean_token_accuracy": 0.2406909555196762,
|
|
"num_tokens": 16698913.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 5.250265789031983,
|
|
"epoch": 7.452978056426332,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00012932069519347873,
|
|
"loss": 4.3705,
|
|
"mean_token_accuracy": 0.24973057359457015,
|
|
"num_tokens": 16708306.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 5.289071130752563,
|
|
"epoch": 7.456896551724138,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012909159386412578,
|
|
"loss": 4.437,
|
|
"mean_token_accuracy": 0.2503813296556473,
|
|
"num_tokens": 16716345.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 5.288973236083985,
|
|
"epoch": 7.460815047021944,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012886275329675246,
|
|
"loss": 4.4253,
|
|
"mean_token_accuracy": 0.2475817769765854,
|
|
"num_tokens": 16725405.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 5.2774420261383055,
|
|
"epoch": 7.464733542319749,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001286341739003341,
|
|
"loss": 4.4342,
|
|
"mean_token_accuracy": 0.24827296435832977,
|
|
"num_tokens": 16734771.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 5.271235990524292,
|
|
"epoch": 7.468652037617555,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00012840585608337907,
|
|
"loss": 4.4777,
|
|
"mean_token_accuracy": 0.24050719738006593,
|
|
"num_tokens": 16742403.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 5.251089572906494,
|
|
"epoch": 7.472570532915361,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012817780025392833,
|
|
"loss": 4.4223,
|
|
"mean_token_accuracy": 0.25051246732473376,
|
|
"num_tokens": 16751450.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 5.276630401611328,
|
|
"epoch": 7.476489028213166,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00012795000681955468,
|
|
"loss": 4.4094,
|
|
"mean_token_accuracy": 0.2449594870209694,
|
|
"num_tokens": 16760216.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 5.284153032302856,
|
|
"epoch": 7.480407523510972,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00012772247618736182,
|
|
"loss": 4.4597,
|
|
"mean_token_accuracy": 0.24596870094537734,
|
|
"num_tokens": 16768648.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 5.2183287143707275,
|
|
"epoch": 7.484326018808777,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012749520876398395,
|
|
"loss": 4.3327,
|
|
"mean_token_accuracy": 0.26086993217468263,
|
|
"num_tokens": 16776587.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 5.299083662033081,
|
|
"epoch": 7.488244514106583,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012726820495558483,
|
|
"loss": 4.4307,
|
|
"mean_token_accuracy": 0.2473188266158104,
|
|
"num_tokens": 16785937.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 5.27997317314148,
|
|
"epoch": 7.492163009404389,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012704146516785705,
|
|
"loss": 4.3864,
|
|
"mean_token_accuracy": 0.24791324585676194,
|
|
"num_tokens": 16794359.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 5.317609882354736,
|
|
"epoch": 7.496081504702194,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001268149898060213,
|
|
"loss": 4.4501,
|
|
"mean_token_accuracy": 0.2508188337087631,
|
|
"num_tokens": 16803685.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 5.215402555465698,
|
|
"epoch": 7.5,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00012658877927482586,
|
|
"loss": 4.3816,
|
|
"mean_token_accuracy": 0.25365270376205445,
|
|
"num_tokens": 16812267.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 5.329932022094726,
|
|
"epoch": 7.503918495297806,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001263628339785456,
|
|
"loss": 4.4957,
|
|
"mean_token_accuracy": 0.2416626423597336,
|
|
"num_tokens": 16820334.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 5.29955677986145,
|
|
"epoch": 7.507836990595611,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012613715432098138,
|
|
"loss": 4.4447,
|
|
"mean_token_accuracy": 0.25171813517808916,
|
|
"num_tokens": 16828908.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 5.355383729934692,
|
|
"epoch": 7.511755485893417,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00012591174070545932,
|
|
"loss": 4.4643,
|
|
"mean_token_accuracy": 0.2444755256175995,
|
|
"num_tokens": 16837772.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 5.344161081314087,
|
|
"epoch": 7.515673981191222,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00012568659353483024,
|
|
"loss": 4.482,
|
|
"mean_token_accuracy": 0.23903487622737885,
|
|
"num_tokens": 16846714.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 5.312340259552002,
|
|
"epoch": 7.519592476489028,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00012546171321146848,
|
|
"loss": 4.3972,
|
|
"mean_token_accuracy": 0.25185195356607437,
|
|
"num_tokens": 16854745.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 5.33575291633606,
|
|
"epoch": 7.523510971786834,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012523710013727173,
|
|
"loss": 4.4347,
|
|
"mean_token_accuracy": 0.25621542781591417,
|
|
"num_tokens": 16864181.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 5.319867038726807,
|
|
"epoch": 7.527429467084639,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00012501275471365998,
|
|
"loss": 4.4695,
|
|
"mean_token_accuracy": 0.24292832612991333,
|
|
"num_tokens": 16873551.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 5.331023454666138,
|
|
"epoch": 7.531347962382445,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012478867734157486,
|
|
"loss": 4.5285,
|
|
"mean_token_accuracy": 0.23911966234445572,
|
|
"num_tokens": 16883108.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 5.268682336807251,
|
|
"epoch": 7.535266457680251,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012456486842147885,
|
|
"loss": 4.3962,
|
|
"mean_token_accuracy": 0.2623940661549568,
|
|
"num_tokens": 16891772.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 5.253333663940429,
|
|
"epoch": 7.539184952978056,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00012434132835335496,
|
|
"loss": 4.3849,
|
|
"mean_token_accuracy": 0.25236410051584246,
|
|
"num_tokens": 16900519.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 5.314716482162476,
|
|
"epoch": 7.543103448275862,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00012411805753670543,
|
|
"loss": 4.4608,
|
|
"mean_token_accuracy": 0.2518236204981804,
|
|
"num_tokens": 16909228.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 5.300777196884155,
|
|
"epoch": 7.547021943573668,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001238950563705514,
|
|
"loss": 4.4642,
|
|
"mean_token_accuracy": 0.24617722481489182,
|
|
"num_tokens": 16918100.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 5.322879695892334,
|
|
"epoch": 7.550940438871473,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001236723252534322,
|
|
"loss": 4.4387,
|
|
"mean_token_accuracy": 0.24768476635217668,
|
|
"num_tokens": 16926911.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 5.323801183700562,
|
|
"epoch": 7.554858934169279,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012344986458340433,
|
|
"loss": 4.4827,
|
|
"mean_token_accuracy": 0.24190146178007127,
|
|
"num_tokens": 16935638.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 5.273169088363647,
|
|
"epoch": 7.558777429467085,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012322767475804122,
|
|
"loss": 4.3199,
|
|
"mean_token_accuracy": 0.2631727740168571,
|
|
"num_tokens": 16944351.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 5.3228480339050295,
|
|
"epoch": 7.56269592476489,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00012300575617443194,
|
|
"loss": 4.4474,
|
|
"mean_token_accuracy": 0.25346849411725997,
|
|
"num_tokens": 16952486.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 5.329959487915039,
|
|
"epoch": 7.566614420062696,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001227841092291812,
|
|
"loss": 4.5061,
|
|
"mean_token_accuracy": 0.23894040584564208,
|
|
"num_tokens": 16960877.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 5.297020769119262,
|
|
"epoch": 7.570532915360501,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00012256273431840792,
|
|
"loss": 4.4798,
|
|
"mean_token_accuracy": 0.24364516884088516,
|
|
"num_tokens": 16970868.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 5.313913345336914,
|
|
"epoch": 7.574451410658307,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00012234163183774482,
|
|
"loss": 4.4554,
|
|
"mean_token_accuracy": 0.2435683935880661,
|
|
"num_tokens": 16980339.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 5.275967836380005,
|
|
"epoch": 7.578369905956113,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012212080218233814,
|
|
"loss": 4.3834,
|
|
"mean_token_accuracy": 0.24716461896896363,
|
|
"num_tokens": 16988786.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 5.285604619979859,
|
|
"epoch": 7.582288401253918,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012190024574684614,
|
|
"loss": 4.4165,
|
|
"mean_token_accuracy": 0.24524554908275603,
|
|
"num_tokens": 16998454.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 5.295511865615845,
|
|
"epoch": 7.586206896551724,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012167996292543889,
|
|
"loss": 4.4382,
|
|
"mean_token_accuracy": 0.24790037125349046,
|
|
"num_tokens": 17006951.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 5.359017467498779,
|
|
"epoch": 7.59012539184953,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001214599541117976,
|
|
"loss": 4.455,
|
|
"mean_token_accuracy": 0.24746038466691972,
|
|
"num_tokens": 17015991.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 5.286804580688477,
|
|
"epoch": 7.594043887147335,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00012124021969911366,
|
|
"loss": 4.447,
|
|
"mean_token_accuracy": 0.24291893541812898,
|
|
"num_tokens": 17024493.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 5.328746509552002,
|
|
"epoch": 7.597962382445141,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00012102076008008804,
|
|
"loss": 4.4376,
|
|
"mean_token_accuracy": 0.24419746547937393,
|
|
"num_tokens": 17033402.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 5.30411205291748,
|
|
"epoch": 7.601880877742946,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00012080157564693073,
|
|
"loss": 4.3935,
|
|
"mean_token_accuracy": 0.2519220754504204,
|
|
"num_tokens": 17042417.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 5.323762226104736,
|
|
"epoch": 7.605799373040752,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00012058266679135987,
|
|
"loss": 4.5446,
|
|
"mean_token_accuracy": 0.23911416828632354,
|
|
"num_tokens": 17050395.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 5.3472789287567135,
|
|
"epoch": 7.609717868338558,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012036403390460107,
|
|
"loss": 4.4947,
|
|
"mean_token_accuracy": 0.24411895126104355,
|
|
"num_tokens": 17060077.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 5.332138109207153,
|
|
"epoch": 7.613636363636363,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012014567737738667,
|
|
"loss": 4.4322,
|
|
"mean_token_accuracy": 0.2398582085967064,
|
|
"num_tokens": 17068099.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 5.2887779712677006,
|
|
"epoch": 7.617554858934169,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011992759759995529,
|
|
"loss": 4.3867,
|
|
"mean_token_accuracy": 0.2438146084547043,
|
|
"num_tokens": 17076846.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 5.295240783691407,
|
|
"epoch": 7.621473354231975,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011970979496205078,
|
|
"loss": 4.4151,
|
|
"mean_token_accuracy": 0.24706131368875503,
|
|
"num_tokens": 17085960.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 5.334226179122925,
|
|
"epoch": 7.62539184952978,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011949226985292183,
|
|
"loss": 4.3945,
|
|
"mean_token_accuracy": 0.24911632090806962,
|
|
"num_tokens": 17095327.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 5.233332729339599,
|
|
"epoch": 7.629310344827586,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00011927502266132098,
|
|
"loss": 4.4056,
|
|
"mean_token_accuracy": 0.2521967738866806,
|
|
"num_tokens": 17103961.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 5.255716228485108,
|
|
"epoch": 7.633228840125392,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011905805377550428,
|
|
"loss": 4.3929,
|
|
"mean_token_accuracy": 0.24839543104171752,
|
|
"num_tokens": 17112427.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 5.246513605117798,
|
|
"epoch": 7.637147335423197,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0001188413635832303,
|
|
"loss": 4.4574,
|
|
"mean_token_accuracy": 0.24096207320690155,
|
|
"num_tokens": 17120890.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 5.268857669830322,
|
|
"epoch": 7.641065830721003,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011862495247175947,
|
|
"loss": 4.3281,
|
|
"mean_token_accuracy": 0.25761065930128096,
|
|
"num_tokens": 17129881.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 5.380213212966919,
|
|
"epoch": 7.644984326018808,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011840882082785357,
|
|
"loss": 4.4889,
|
|
"mean_token_accuracy": 0.25111201256513593,
|
|
"num_tokens": 17138900.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 5.269290256500244,
|
|
"epoch": 7.648902821316614,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000118192969037775,
|
|
"loss": 4.44,
|
|
"mean_token_accuracy": 0.24698970466852188,
|
|
"num_tokens": 17147023.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 5.387117099761963,
|
|
"epoch": 7.65282131661442,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011797739748728575,
|
|
"loss": 4.4899,
|
|
"mean_token_accuracy": 0.23652591109275817,
|
|
"num_tokens": 17156926.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 5.3261676788330075,
|
|
"epoch": 7.656739811912225,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001177621065616473,
|
|
"loss": 4.4616,
|
|
"mean_token_accuracy": 0.24903216809034348,
|
|
"num_tokens": 17165079.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 5.240077018737793,
|
|
"epoch": 7.660658307210031,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011754709664561931,
|
|
"loss": 4.3811,
|
|
"mean_token_accuracy": 0.2534024178981781,
|
|
"num_tokens": 17173509.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 5.342904186248779,
|
|
"epoch": 7.664576802507837,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011733236812345943,
|
|
"loss": 4.4429,
|
|
"mean_token_accuracy": 0.24250003099441528,
|
|
"num_tokens": 17182399.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 5.270763349533081,
|
|
"epoch": 7.668495297805642,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011711792137892221,
|
|
"loss": 4.4269,
|
|
"mean_token_accuracy": 0.24486750215291977,
|
|
"num_tokens": 17190733.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 5.262229251861572,
|
|
"epoch": 7.672413793103448,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011690375679525896,
|
|
"loss": 4.4301,
|
|
"mean_token_accuracy": 0.24390010833740233,
|
|
"num_tokens": 17200078.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 5.287205076217651,
|
|
"epoch": 7.6763322884012535,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011668987475521642,
|
|
"loss": 4.4642,
|
|
"mean_token_accuracy": 0.2456186905503273,
|
|
"num_tokens": 17208430.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 5.306658220291138,
|
|
"epoch": 7.6802507836990594,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00011647627564103636,
|
|
"loss": 4.493,
|
|
"mean_token_accuracy": 0.24013693630695343,
|
|
"num_tokens": 17217309.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 5.281977605819702,
|
|
"epoch": 7.684169278996865,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011626295983445525,
|
|
"loss": 4.3728,
|
|
"mean_token_accuracy": 0.25647018700838087,
|
|
"num_tokens": 17225700.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 5.227178573608398,
|
|
"epoch": 7.6880877742946705,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011604992771670292,
|
|
"loss": 4.396,
|
|
"mean_token_accuracy": 0.24518679231405258,
|
|
"num_tokens": 17234582.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 5.186618280410767,
|
|
"epoch": 7.6920062695924765,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00011583717966850219,
|
|
"loss": 4.4109,
|
|
"mean_token_accuracy": 0.2519668146967888,
|
|
"num_tokens": 17243207.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 5.243455028533935,
|
|
"epoch": 7.695924764890282,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011562471607006845,
|
|
"loss": 4.3852,
|
|
"mean_token_accuracy": 0.2493540972471237,
|
|
"num_tokens": 17251658.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 5.333655881881714,
|
|
"epoch": 7.6998432601880875,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011541253730110866,
|
|
"loss": 4.5262,
|
|
"mean_token_accuracy": 0.24134959429502487,
|
|
"num_tokens": 17261175.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 5.2469769477844235,
|
|
"epoch": 7.7037617554858935,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001152006437408206,
|
|
"loss": 4.323,
|
|
"mean_token_accuracy": 0.25572637319564817,
|
|
"num_tokens": 17269302.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 5.310148048400879,
|
|
"epoch": 7.7076802507836994,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011498903576789233,
|
|
"loss": 4.4944,
|
|
"mean_token_accuracy": 0.242820905148983,
|
|
"num_tokens": 17277911.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 5.315098810195923,
|
|
"epoch": 7.7115987460815045,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011477771376050177,
|
|
"loss": 4.4298,
|
|
"mean_token_accuracy": 0.24483197033405305,
|
|
"num_tokens": 17287219.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 5.266485595703125,
|
|
"epoch": 7.7155172413793105,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011456667809631556,
|
|
"loss": 4.3891,
|
|
"mean_token_accuracy": 0.2505299985408783,
|
|
"num_tokens": 17295716.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 5.282175445556641,
|
|
"epoch": 7.7194357366771165,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011435592915248855,
|
|
"loss": 4.4429,
|
|
"mean_token_accuracy": 0.24854001104831697,
|
|
"num_tokens": 17305870.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 5.2947248935699465,
|
|
"epoch": 7.7233542319749215,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011414546730566342,
|
|
"loss": 4.4221,
|
|
"mean_token_accuracy": 0.2488165646791458,
|
|
"num_tokens": 17314130.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 5.307626724243164,
|
|
"epoch": 7.7272727272727275,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011393529293196955,
|
|
"loss": 4.3396,
|
|
"mean_token_accuracy": 0.25593169778585434,
|
|
"num_tokens": 17322320.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 5.305253744125366,
|
|
"epoch": 7.731191222570533,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011372540640702263,
|
|
"loss": 4.4416,
|
|
"mean_token_accuracy": 0.24936089664697647,
|
|
"num_tokens": 17331162.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 5.23098087310791,
|
|
"epoch": 7.735109717868339,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00011351580810592382,
|
|
"loss": 4.3596,
|
|
"mean_token_accuracy": 0.25074984729290006,
|
|
"num_tokens": 17339568.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 5.316984081268311,
|
|
"epoch": 7.7390282131661445,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00011330649840325938,
|
|
"loss": 4.4565,
|
|
"mean_token_accuracy": 0.24562593251466752,
|
|
"num_tokens": 17349354.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 5.264963912963867,
|
|
"epoch": 7.74294670846395,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001130974776730997,
|
|
"loss": 4.4653,
|
|
"mean_token_accuracy": 0.23963449150323868,
|
|
"num_tokens": 17358668.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 5.2722937107086185,
|
|
"epoch": 7.746865203761756,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011288874628899864,
|
|
"loss": 4.3635,
|
|
"mean_token_accuracy": 0.2484382674098015,
|
|
"num_tokens": 17366792.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 5.308924341201783,
|
|
"epoch": 7.750783699059561,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011268030462399309,
|
|
"loss": 4.4293,
|
|
"mean_token_accuracy": 0.24282660633325576,
|
|
"num_tokens": 17375589.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 5.372103071212768,
|
|
"epoch": 7.754702194357367,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00011247215305060213,
|
|
"loss": 4.5003,
|
|
"mean_token_accuracy": 0.24558619111776353,
|
|
"num_tokens": 17385816.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 5.238821220397949,
|
|
"epoch": 7.758620689655173,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011226429194082627,
|
|
"loss": 4.3707,
|
|
"mean_token_accuracy": 0.2501941457390785,
|
|
"num_tokens": 17395267.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 5.321446180343628,
|
|
"epoch": 7.762539184952978,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000112056721666147,
|
|
"loss": 4.4526,
|
|
"mean_token_accuracy": 0.2521474778652191,
|
|
"num_tokens": 17403544.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 5.305885696411133,
|
|
"epoch": 7.766457680250784,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011184944259752622,
|
|
"loss": 4.4351,
|
|
"mean_token_accuracy": 0.24860943108797073,
|
|
"num_tokens": 17412090.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 5.280656003952027,
|
|
"epoch": 7.77037617554859,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011164245510540512,
|
|
"loss": 4.422,
|
|
"mean_token_accuracy": 0.24793280363082887,
|
|
"num_tokens": 17420558.0,
|
|
"step": 9915
|
|
},
|
|
{
|
|
"entropy": 5.2638993740081785,
|
|
"epoch": 7.774294670846395,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00011143575955970384,
|
|
"loss": 4.442,
|
|
"mean_token_accuracy": 0.2463257536292076,
|
|
"num_tokens": 17429901.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 5.352838706970215,
|
|
"epoch": 7.778213166144201,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00011122935632982101,
|
|
"loss": 4.4691,
|
|
"mean_token_accuracy": 0.24341759979724883,
|
|
"num_tokens": 17438543.0,
|
|
"step": 9925
|
|
},
|
|
{
|
|
"entropy": 5.375067520141601,
|
|
"epoch": 7.782131661442007,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001110232457846325,
|
|
"loss": 4.481,
|
|
"mean_token_accuracy": 0.2451726421713829,
|
|
"num_tokens": 17447234.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 5.220444917678833,
|
|
"epoch": 7.786050156739812,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011081742829249142,
|
|
"loss": 4.3447,
|
|
"mean_token_accuracy": 0.25400242805480955,
|
|
"num_tokens": 17455542.0,
|
|
"step": 9935
|
|
},
|
|
{
|
|
"entropy": 5.302769708633423,
|
|
"epoch": 7.789968652037618,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00011061190422122687,
|
|
"loss": 4.4087,
|
|
"mean_token_accuracy": 0.24850808829069138,
|
|
"num_tokens": 17465035.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 5.291922569274902,
|
|
"epoch": 7.793887147335424,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00011040667393814383,
|
|
"loss": 4.4397,
|
|
"mean_token_accuracy": 0.2470712259411812,
|
|
"num_tokens": 17473748.0,
|
|
"step": 9945
|
|
},
|
|
{
|
|
"entropy": 5.22485499382019,
|
|
"epoch": 7.797805642633229,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011020173781002199,
|
|
"loss": 4.3165,
|
|
"mean_token_accuracy": 0.2589133679866791,
|
|
"num_tokens": 17481525.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 5.274036693572998,
|
|
"epoch": 7.801724137931035,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001099970962031154,
|
|
"loss": 4.4687,
|
|
"mean_token_accuracy": 0.2439611703157425,
|
|
"num_tokens": 17490044.0,
|
|
"step": 9955
|
|
},
|
|
{
|
|
"entropy": 5.366907548904419,
|
|
"epoch": 7.80564263322884,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00010979274948315191,
|
|
"loss": 4.4998,
|
|
"mean_token_accuracy": 0.2417758136987686,
|
|
"num_tokens": 17498094.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 5.325379705429077,
|
|
"epoch": 7.809561128526646,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010958869801533222,
|
|
"loss": 4.4483,
|
|
"mean_token_accuracy": 0.24720122516155243,
|
|
"num_tokens": 17507677.0,
|
|
"step": 9965
|
|
},
|
|
{
|
|
"entropy": 5.292023706436157,
|
|
"epoch": 7.813479623824452,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010938494216432928,
|
|
"loss": 4.3944,
|
|
"mean_token_accuracy": 0.24547045528888703,
|
|
"num_tokens": 17516489.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 5.281532573699951,
|
|
"epoch": 7.817398119122257,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010918148229428796,
|
|
"loss": 4.4317,
|
|
"mean_token_accuracy": 0.24878885298967363,
|
|
"num_tokens": 17524972.0,
|
|
"step": 9975
|
|
},
|
|
{
|
|
"entropy": 5.283537864685059,
|
|
"epoch": 7.821316614420063,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000108978318768824,
|
|
"loss": 4.4317,
|
|
"mean_token_accuracy": 0.2524750351905823,
|
|
"num_tokens": 17533745.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 5.231257486343384,
|
|
"epoch": 7.825235109717868,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010877545195102347,
|
|
"loss": 4.4177,
|
|
"mean_token_accuracy": 0.23633328974246978,
|
|
"num_tokens": 17542812.0,
|
|
"step": 9985
|
|
},
|
|
{
|
|
"entropy": 5.271108198165893,
|
|
"epoch": 7.829153605015674,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00010857288220344242,
|
|
"loss": 4.4157,
|
|
"mean_token_accuracy": 0.25000583529472353,
|
|
"num_tokens": 17551123.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 5.295686149597168,
|
|
"epoch": 7.83307210031348,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010837060988810571,
|
|
"loss": 4.5198,
|
|
"mean_token_accuracy": 0.23068171888589858,
|
|
"num_tokens": 17560729.0,
|
|
"step": 9995
|
|
},
|
|
{
|
|
"entropy": 5.371287488937378,
|
|
"epoch": 7.836990595611285,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010816863536650687,
|
|
"loss": 4.4862,
|
|
"mean_token_accuracy": 0.24116043299436568,
|
|
"num_tokens": 17569842.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"epoch": 7.836990595611285,
|
|
"eval_entropy": 5.147375997646835,
|
|
"eval_loss": 5.620843887329102,
|
|
"eval_mean_token_accuracy": 0.17606263343504694,
|
|
"eval_num_tokens": 17569842.0,
|
|
"eval_runtime": 2.3751,
|
|
"eval_samples_per_second": 1735.495,
|
|
"eval_steps_per_second": 217.253,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"entropy": 5.325961923599243,
|
|
"epoch": 7.840909090909091,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010796695899960696,
|
|
"loss": 4.3862,
|
|
"mean_token_accuracy": 0.2597434654831886,
|
|
"num_tokens": 17578887.0,
|
|
"step": 10005
|
|
},
|
|
{
|
|
"entropy": 5.31726975440979,
|
|
"epoch": 7.844827586206897,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010776558114783455,
|
|
"loss": 4.3985,
|
|
"mean_token_accuracy": 0.24566721320152282,
|
|
"num_tokens": 17587321.0,
|
|
"step": 10010
|
|
},
|
|
{
|
|
"entropy": 5.275127744674682,
|
|
"epoch": 7.848746081504702,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010756450217108441,
|
|
"loss": 4.4012,
|
|
"mean_token_accuracy": 0.24983313232660292,
|
|
"num_tokens": 17596295.0,
|
|
"step": 10015
|
|
},
|
|
{
|
|
"entropy": 5.289866161346436,
|
|
"epoch": 7.852664576802508,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001073637224287172,
|
|
"loss": 4.3926,
|
|
"mean_token_accuracy": 0.25193305164575575,
|
|
"num_tokens": 17605033.0,
|
|
"step": 10020
|
|
},
|
|
{
|
|
"entropy": 5.22728009223938,
|
|
"epoch": 7.856583072100314,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010716324227955904,
|
|
"loss": 4.3998,
|
|
"mean_token_accuracy": 0.24876124858856202,
|
|
"num_tokens": 17613706.0,
|
|
"step": 10025
|
|
},
|
|
{
|
|
"entropy": 5.328704118728638,
|
|
"epoch": 7.860501567398119,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010696306208190034,
|
|
"loss": 4.4799,
|
|
"mean_token_accuracy": 0.24404100626707076,
|
|
"num_tokens": 17622082.0,
|
|
"step": 10030
|
|
},
|
|
{
|
|
"entropy": 5.19866418838501,
|
|
"epoch": 7.864420062695925,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010676318219349564,
|
|
"loss": 4.3453,
|
|
"mean_token_accuracy": 0.2575923725962639,
|
|
"num_tokens": 17630980.0,
|
|
"step": 10035
|
|
},
|
|
{
|
|
"entropy": 5.241474580764771,
|
|
"epoch": 7.868338557993731,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010656360297156262,
|
|
"loss": 4.3882,
|
|
"mean_token_accuracy": 0.24939603954553605,
|
|
"num_tokens": 17639679.0,
|
|
"step": 10040
|
|
},
|
|
{
|
|
"entropy": 5.266773414611817,
|
|
"epoch": 7.872257053291536,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001063643247727818,
|
|
"loss": 4.3515,
|
|
"mean_token_accuracy": 0.2589699819684029,
|
|
"num_tokens": 17648542.0,
|
|
"step": 10045
|
|
},
|
|
{
|
|
"entropy": 5.280829763412475,
|
|
"epoch": 7.876175548589342,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010616534795329553,
|
|
"loss": 4.3371,
|
|
"mean_token_accuracy": 0.2586196094751358,
|
|
"num_tokens": 17657305.0,
|
|
"step": 10050
|
|
},
|
|
{
|
|
"entropy": 5.319469165802002,
|
|
"epoch": 7.880094043887147,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00010596667286870778,
|
|
"loss": 4.4547,
|
|
"mean_token_accuracy": 0.24209896177053453,
|
|
"num_tokens": 17665876.0,
|
|
"step": 10055
|
|
},
|
|
{
|
|
"entropy": 5.265191698074341,
|
|
"epoch": 7.884012539184953,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010576829987408304,
|
|
"loss": 4.3737,
|
|
"mean_token_accuracy": 0.2562585145235062,
|
|
"num_tokens": 17674756.0,
|
|
"step": 10060
|
|
},
|
|
{
|
|
"entropy": 5.268184566497803,
|
|
"epoch": 7.887931034482759,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00010557022932394611,
|
|
"loss": 4.3465,
|
|
"mean_token_accuracy": 0.2691681981086731,
|
|
"num_tokens": 17683989.0,
|
|
"step": 10065
|
|
},
|
|
{
|
|
"entropy": 5.388026237487793,
|
|
"epoch": 7.891849529780564,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010537246157228104,
|
|
"loss": 4.4493,
|
|
"mean_token_accuracy": 0.25004620999097826,
|
|
"num_tokens": 17692776.0,
|
|
"step": 10070
|
|
},
|
|
{
|
|
"entropy": 5.283219957351685,
|
|
"epoch": 7.89576802507837,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010517499697253103,
|
|
"loss": 4.3392,
|
|
"mean_token_accuracy": 0.24918409287929535,
|
|
"num_tokens": 17701428.0,
|
|
"step": 10075
|
|
},
|
|
{
|
|
"entropy": 5.271471452713013,
|
|
"epoch": 7.899686520376176,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010497783587759724,
|
|
"loss": 4.4524,
|
|
"mean_token_accuracy": 0.23854980170726775,
|
|
"num_tokens": 17710343.0,
|
|
"step": 10080
|
|
},
|
|
{
|
|
"entropy": 5.315694093704224,
|
|
"epoch": 7.903605015673981,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00010478097863983857,
|
|
"loss": 4.4397,
|
|
"mean_token_accuracy": 0.2542968034744263,
|
|
"num_tokens": 17719218.0,
|
|
"step": 10085
|
|
},
|
|
{
|
|
"entropy": 5.333209371566772,
|
|
"epoch": 7.907523510971787,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001045844256110707,
|
|
"loss": 4.4846,
|
|
"mean_token_accuracy": 0.2373369202017784,
|
|
"num_tokens": 17727288.0,
|
|
"step": 10090
|
|
},
|
|
{
|
|
"entropy": 5.313266468048096,
|
|
"epoch": 7.911442006269592,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010438817714256589,
|
|
"loss": 4.4846,
|
|
"mean_token_accuracy": 0.24039488583803176,
|
|
"num_tokens": 17736130.0,
|
|
"step": 10095
|
|
},
|
|
{
|
|
"entropy": 5.269415378570557,
|
|
"epoch": 7.915360501567398,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00010419223358505196,
|
|
"loss": 4.3664,
|
|
"mean_token_accuracy": 0.2576622784137726,
|
|
"num_tokens": 17745122.0,
|
|
"step": 10100
|
|
},
|
|
{
|
|
"entropy": 5.196227121353149,
|
|
"epoch": 7.919278996865204,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00010399659528871169,
|
|
"loss": 4.383,
|
|
"mean_token_accuracy": 0.2496015951037407,
|
|
"num_tokens": 17753817.0,
|
|
"step": 10105
|
|
},
|
|
{
|
|
"entropy": 5.294891786575318,
|
|
"epoch": 7.923197492163009,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010380126260318265,
|
|
"loss": 4.4422,
|
|
"mean_token_accuracy": 0.24454985111951827,
|
|
"num_tokens": 17762793.0,
|
|
"step": 10110
|
|
},
|
|
{
|
|
"entropy": 5.231249094009399,
|
|
"epoch": 7.927115987460815,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010360623587755579,
|
|
"loss": 4.402,
|
|
"mean_token_accuracy": 0.2483761116862297,
|
|
"num_tokens": 17772223.0,
|
|
"step": 10115
|
|
},
|
|
{
|
|
"entropy": 5.261945295333862,
|
|
"epoch": 7.931034482758621,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001034115154603757,
|
|
"loss": 4.3842,
|
|
"mean_token_accuracy": 0.2461504727602005,
|
|
"num_tokens": 17781543.0,
|
|
"step": 10120
|
|
},
|
|
{
|
|
"entropy": 5.284945678710938,
|
|
"epoch": 7.934952978056426,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010321710169963919,
|
|
"loss": 4.4076,
|
|
"mean_token_accuracy": 0.2436329111456871,
|
|
"num_tokens": 17790258.0,
|
|
"step": 10125
|
|
},
|
|
{
|
|
"entropy": 5.262395572662354,
|
|
"epoch": 7.938871473354232,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010302299494279527,
|
|
"loss": 4.3908,
|
|
"mean_token_accuracy": 0.24701260179281234,
|
|
"num_tokens": 17798862.0,
|
|
"step": 10130
|
|
},
|
|
{
|
|
"entropy": 5.28599796295166,
|
|
"epoch": 7.942789968652038,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00010282919553674415,
|
|
"loss": 4.4373,
|
|
"mean_token_accuracy": 0.24531031399965286,
|
|
"num_tokens": 17807285.0,
|
|
"step": 10135
|
|
},
|
|
{
|
|
"entropy": 5.285976266860962,
|
|
"epoch": 7.946708463949843,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00010263570382783675,
|
|
"loss": 4.4692,
|
|
"mean_token_accuracy": 0.24353863745927812,
|
|
"num_tokens": 17816891.0,
|
|
"step": 10140
|
|
},
|
|
{
|
|
"entropy": 5.304802465438843,
|
|
"epoch": 7.950626959247649,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00010244252016187419,
|
|
"loss": 4.4408,
|
|
"mean_token_accuracy": 0.24795014709234237,
|
|
"num_tokens": 17825895.0,
|
|
"step": 10145
|
|
},
|
|
{
|
|
"entropy": 5.3282664775848385,
|
|
"epoch": 7.954545454545455,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00010224964488410699,
|
|
"loss": 4.5213,
|
|
"mean_token_accuracy": 0.2436439648270607,
|
|
"num_tokens": 17834921.0,
|
|
"step": 10150
|
|
},
|
|
{
|
|
"entropy": 5.359384155273437,
|
|
"epoch": 7.95846394984326,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001020570783392345,
|
|
"loss": 4.4061,
|
|
"mean_token_accuracy": 0.2576331451535225,
|
|
"num_tokens": 17844310.0,
|
|
"step": 10155
|
|
},
|
|
{
|
|
"entropy": 5.362526178359985,
|
|
"epoch": 7.962382445141066,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010186482087140439,
|
|
"loss": 4.4487,
|
|
"mean_token_accuracy": 0.25078432112932203,
|
|
"num_tokens": 17853561.0,
|
|
"step": 10160
|
|
},
|
|
{
|
|
"entropy": 5.240152072906494,
|
|
"epoch": 7.966300940438871,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000101672872824212,
|
|
"loss": 4.3962,
|
|
"mean_token_accuracy": 0.2438494771718979,
|
|
"num_tokens": 17863077.0,
|
|
"step": 10165
|
|
},
|
|
{
|
|
"entropy": 5.304940509796142,
|
|
"epoch": 7.970219435736677,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010148123454069943,
|
|
"loss": 4.3543,
|
|
"mean_token_accuracy": 0.259673698246479,
|
|
"num_tokens": 17871780.0,
|
|
"step": 10170
|
|
},
|
|
{
|
|
"entropy": 5.252498626708984,
|
|
"epoch": 7.974137931034483,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001012899063633555,
|
|
"loss": 4.3941,
|
|
"mean_token_accuracy": 0.24746048152446748,
|
|
"num_tokens": 17881144.0,
|
|
"step": 10175
|
|
},
|
|
{
|
|
"entropy": 5.28587064743042,
|
|
"epoch": 7.978056426332288,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010109888863411473,
|
|
"loss": 4.4248,
|
|
"mean_token_accuracy": 0.24611652493476868,
|
|
"num_tokens": 17890754.0,
|
|
"step": 10180
|
|
},
|
|
{
|
|
"entropy": 5.318957805633545,
|
|
"epoch": 7.981974921630094,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001009081816943567,
|
|
"loss": 4.4883,
|
|
"mean_token_accuracy": 0.24397450536489487,
|
|
"num_tokens": 17899992.0,
|
|
"step": 10185
|
|
},
|
|
{
|
|
"entropy": 5.227457523345947,
|
|
"epoch": 7.985893416927899,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010071778588490558,
|
|
"loss": 4.3722,
|
|
"mean_token_accuracy": 0.25830663442611695,
|
|
"num_tokens": 17907847.0,
|
|
"step": 10190
|
|
},
|
|
{
|
|
"entropy": 5.310833406448364,
|
|
"epoch": 7.989811912225705,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010052770154602968,
|
|
"loss": 4.4268,
|
|
"mean_token_accuracy": 0.24884800463914872,
|
|
"num_tokens": 17916566.0,
|
|
"step": 10195
|
|
},
|
|
{
|
|
"entropy": 5.272046613693237,
|
|
"epoch": 7.993730407523511,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010033792901744046,
|
|
"loss": 4.3214,
|
|
"mean_token_accuracy": 0.25625889003276825,
|
|
"num_tokens": 17924969.0,
|
|
"step": 10200
|
|
},
|
|
{
|
|
"entropy": 5.28591742515564,
|
|
"epoch": 7.997648902821316,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010014846863829221,
|
|
"loss": 4.3357,
|
|
"mean_token_accuracy": 0.25523446649312975,
|
|
"num_tokens": 17933776.0,
|
|
"step": 10205
|
|
},
|
|
{
|
|
"entropy": 5.230620861053467,
|
|
"epoch": 8.001567398119123,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 9.995932074718131e-05,
|
|
"loss": 4.3241,
|
|
"mean_token_accuracy": 0.2523978725075722,
|
|
"num_tokens": 17941651.0,
|
|
"step": 10210
|
|
},
|
|
{
|
|
"entropy": 5.304376173019409,
|
|
"epoch": 8.005485893416928,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.977048568214578e-05,
|
|
"loss": 4.4661,
|
|
"mean_token_accuracy": 0.24086453169584274,
|
|
"num_tokens": 17951278.0,
|
|
"step": 10215
|
|
},
|
|
{
|
|
"entropy": 5.228555488586426,
|
|
"epoch": 8.009404388714733,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.958196378066444e-05,
|
|
"loss": 4.3723,
|
|
"mean_token_accuracy": 0.25507096499204635,
|
|
"num_tokens": 17960182.0,
|
|
"step": 10220
|
|
},
|
|
{
|
|
"entropy": 5.251254987716675,
|
|
"epoch": 8.013322884012538,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 9.93937553796565e-05,
|
|
"loss": 4.324,
|
|
"mean_token_accuracy": 0.2662073701620102,
|
|
"num_tokens": 17968221.0,
|
|
"step": 10225
|
|
},
|
|
{
|
|
"entropy": 5.239107513427735,
|
|
"epoch": 8.017241379310345,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.920586081548089e-05,
|
|
"loss": 4.27,
|
|
"mean_token_accuracy": 0.256277984380722,
|
|
"num_tokens": 17975724.0,
|
|
"step": 10230
|
|
},
|
|
{
|
|
"entropy": 5.2473687648773195,
|
|
"epoch": 8.02115987460815,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.901828042393573e-05,
|
|
"loss": 4.4404,
|
|
"mean_token_accuracy": 0.24547503888607025,
|
|
"num_tokens": 17984425.0,
|
|
"step": 10235
|
|
},
|
|
{
|
|
"entropy": 5.264162969589234,
|
|
"epoch": 8.025078369905955,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 9.88310145402575e-05,
|
|
"loss": 4.2801,
|
|
"mean_token_accuracy": 0.2587422415614128,
|
|
"num_tokens": 17993319.0,
|
|
"step": 10240
|
|
},
|
|
{
|
|
"entropy": 5.24732346534729,
|
|
"epoch": 8.028996865203762,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.864406349912081e-05,
|
|
"loss": 4.3013,
|
|
"mean_token_accuracy": 0.2514438346028328,
|
|
"num_tokens": 18002239.0,
|
|
"step": 10245
|
|
},
|
|
{
|
|
"entropy": 5.3010694026947025,
|
|
"epoch": 8.032915360501567,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.845742763463739e-05,
|
|
"loss": 4.4062,
|
|
"mean_token_accuracy": 0.25434952676296235,
|
|
"num_tokens": 18010745.0,
|
|
"step": 10250
|
|
},
|
|
{
|
|
"entropy": 5.281768798828125,
|
|
"epoch": 8.036833855799372,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.827110728035584e-05,
|
|
"loss": 4.3763,
|
|
"mean_token_accuracy": 0.25569429248571396,
|
|
"num_tokens": 18019535.0,
|
|
"step": 10255
|
|
},
|
|
{
|
|
"entropy": 5.273960018157959,
|
|
"epoch": 8.04075235109718,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 9.808510276926075e-05,
|
|
"loss": 4.3576,
|
|
"mean_token_accuracy": 0.25073505342006686,
|
|
"num_tokens": 18029128.0,
|
|
"step": 10260
|
|
},
|
|
{
|
|
"entropy": 5.299409675598144,
|
|
"epoch": 8.044670846394984,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.789941443377249e-05,
|
|
"loss": 4.399,
|
|
"mean_token_accuracy": 0.24916187226772307,
|
|
"num_tokens": 18038304.0,
|
|
"step": 10265
|
|
},
|
|
{
|
|
"entropy": 5.279994916915894,
|
|
"epoch": 8.04858934169279,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.771404260574613e-05,
|
|
"loss": 4.3592,
|
|
"mean_token_accuracy": 0.25540266931056976,
|
|
"num_tokens": 18046628.0,
|
|
"step": 10270
|
|
},
|
|
{
|
|
"entropy": 5.325861024856567,
|
|
"epoch": 8.052507836990596,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.752898761647113e-05,
|
|
"loss": 4.4331,
|
|
"mean_token_accuracy": 0.25002172589302063,
|
|
"num_tokens": 18055654.0,
|
|
"step": 10275
|
|
},
|
|
{
|
|
"entropy": 5.245875215530395,
|
|
"epoch": 8.056426332288401,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.734424979667089e-05,
|
|
"loss": 4.3516,
|
|
"mean_token_accuracy": 0.2550953447818756,
|
|
"num_tokens": 18065844.0,
|
|
"step": 10280
|
|
},
|
|
{
|
|
"entropy": 5.297501134872436,
|
|
"epoch": 8.060344827586206,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.715982947650179e-05,
|
|
"loss": 4.4072,
|
|
"mean_token_accuracy": 0.2455964982509613,
|
|
"num_tokens": 18074497.0,
|
|
"step": 10285
|
|
},
|
|
{
|
|
"entropy": 5.34057765007019,
|
|
"epoch": 8.064263322884013,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.697572698555278e-05,
|
|
"loss": 4.395,
|
|
"mean_token_accuracy": 0.25247364938259126,
|
|
"num_tokens": 18083967.0,
|
|
"step": 10290
|
|
},
|
|
{
|
|
"entropy": 5.279856491088867,
|
|
"epoch": 8.068181818181818,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.679194265284492e-05,
|
|
"loss": 4.3675,
|
|
"mean_token_accuracy": 0.25621638894081117,
|
|
"num_tokens": 18092263.0,
|
|
"step": 10295
|
|
},
|
|
{
|
|
"entropy": 5.356545114517212,
|
|
"epoch": 8.072100313479623,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.660847680683069e-05,
|
|
"loss": 4.4229,
|
|
"mean_token_accuracy": 0.25423054546117785,
|
|
"num_tokens": 18101559.0,
|
|
"step": 10300
|
|
},
|
|
{
|
|
"entropy": 5.335407781600952,
|
|
"epoch": 8.07601880877743,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 9.64253297753932e-05,
|
|
"loss": 4.369,
|
|
"mean_token_accuracy": 0.25400051325559614,
|
|
"num_tokens": 18109850.0,
|
|
"step": 10305
|
|
},
|
|
{
|
|
"entropy": 5.317615270614624,
|
|
"epoch": 8.079937304075235,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.62425018858459e-05,
|
|
"loss": 4.3425,
|
|
"mean_token_accuracy": 0.25961905270814895,
|
|
"num_tokens": 18119205.0,
|
|
"step": 10310
|
|
},
|
|
{
|
|
"entropy": 5.224204730987549,
|
|
"epoch": 8.08385579937304,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.605999346493194e-05,
|
|
"loss": 4.304,
|
|
"mean_token_accuracy": 0.25136202573776245,
|
|
"num_tokens": 18128713.0,
|
|
"step": 10315
|
|
},
|
|
{
|
|
"entropy": 5.275329303741455,
|
|
"epoch": 8.087774294670846,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 9.587780483882339e-05,
|
|
"loss": 4.3175,
|
|
"mean_token_accuracy": 0.25439377278089526,
|
|
"num_tokens": 18137080.0,
|
|
"step": 10320
|
|
},
|
|
{
|
|
"entropy": 5.312987518310547,
|
|
"epoch": 8.091692789968652,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.569593633312084e-05,
|
|
"loss": 4.4429,
|
|
"mean_token_accuracy": 0.24573199152946473,
|
|
"num_tokens": 18146449.0,
|
|
"step": 10325
|
|
},
|
|
{
|
|
"entropy": 5.316402530670166,
|
|
"epoch": 8.095611285266457,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.551438827285294e-05,
|
|
"loss": 4.4607,
|
|
"mean_token_accuracy": 0.24105450361967087,
|
|
"num_tokens": 18155394.0,
|
|
"step": 10330
|
|
},
|
|
{
|
|
"entropy": 5.302361488342285,
|
|
"epoch": 8.099529780564263,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.533316098247539e-05,
|
|
"loss": 4.4575,
|
|
"mean_token_accuracy": 0.24103948920965196,
|
|
"num_tokens": 18164538.0,
|
|
"step": 10335
|
|
},
|
|
{
|
|
"entropy": 5.292540550231934,
|
|
"epoch": 8.10344827586207,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.515225478587077e-05,
|
|
"loss": 4.3804,
|
|
"mean_token_accuracy": 0.25642321854829786,
|
|
"num_tokens": 18174396.0,
|
|
"step": 10340
|
|
},
|
|
{
|
|
"entropy": 5.313747596740723,
|
|
"epoch": 8.107366771159874,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.497167000634773e-05,
|
|
"loss": 4.3474,
|
|
"mean_token_accuracy": 0.25371290147304537,
|
|
"num_tokens": 18183030.0,
|
|
"step": 10345
|
|
},
|
|
{
|
|
"entropy": 5.297376012802124,
|
|
"epoch": 8.11128526645768,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.47914069666406e-05,
|
|
"loss": 4.3606,
|
|
"mean_token_accuracy": 0.250477722287178,
|
|
"num_tokens": 18191139.0,
|
|
"step": 10350
|
|
},
|
|
{
|
|
"entropy": 5.227859735488892,
|
|
"epoch": 8.115203761755486,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.461146598890873e-05,
|
|
"loss": 4.3457,
|
|
"mean_token_accuracy": 0.2587070450186729,
|
|
"num_tokens": 18199429.0,
|
|
"step": 10355
|
|
},
|
|
{
|
|
"entropy": 5.266650152206421,
|
|
"epoch": 8.119122257053291,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 9.443184739473573e-05,
|
|
"loss": 4.3889,
|
|
"mean_token_accuracy": 0.2524459555745125,
|
|
"num_tokens": 18207765.0,
|
|
"step": 10360
|
|
},
|
|
{
|
|
"entropy": 5.2160008430480955,
|
|
"epoch": 8.123040752351097,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 9.425255150512921e-05,
|
|
"loss": 4.411,
|
|
"mean_token_accuracy": 0.24840037375688553,
|
|
"num_tokens": 18216524.0,
|
|
"step": 10365
|
|
},
|
|
{
|
|
"entropy": 5.309645128250122,
|
|
"epoch": 8.126959247648903,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 9.407357864051996e-05,
|
|
"loss": 4.3708,
|
|
"mean_token_accuracy": 0.2541784688830376,
|
|
"num_tokens": 18224647.0,
|
|
"step": 10370
|
|
},
|
|
{
|
|
"entropy": 5.249077320098877,
|
|
"epoch": 8.130877742946709,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.389492912076156e-05,
|
|
"loss": 4.3787,
|
|
"mean_token_accuracy": 0.24498774260282516,
|
|
"num_tokens": 18234127.0,
|
|
"step": 10375
|
|
},
|
|
{
|
|
"entropy": 5.281163167953491,
|
|
"epoch": 8.134796238244514,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.371660326512952e-05,
|
|
"loss": 4.3032,
|
|
"mean_token_accuracy": 0.2586949855089188,
|
|
"num_tokens": 18242455.0,
|
|
"step": 10380
|
|
},
|
|
{
|
|
"entropy": 5.3201381206512455,
|
|
"epoch": 8.13871473354232,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 9.353860139232126e-05,
|
|
"loss": 4.4216,
|
|
"mean_token_accuracy": 0.24777960330247878,
|
|
"num_tokens": 18251201.0,
|
|
"step": 10385
|
|
},
|
|
{
|
|
"entropy": 5.281413412094116,
|
|
"epoch": 8.142633228840126,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.336092382045488e-05,
|
|
"loss": 4.3723,
|
|
"mean_token_accuracy": 0.24543503075838088,
|
|
"num_tokens": 18260651.0,
|
|
"step": 10390
|
|
},
|
|
{
|
|
"entropy": 5.339289808273316,
|
|
"epoch": 8.14655172413793,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 9.318357086706894e-05,
|
|
"loss": 4.424,
|
|
"mean_token_accuracy": 0.25103225857019423,
|
|
"num_tokens": 18269218.0,
|
|
"step": 10395
|
|
},
|
|
{
|
|
"entropy": 5.301513576507569,
|
|
"epoch": 8.150470219435737,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.300654284912211e-05,
|
|
"loss": 4.3862,
|
|
"mean_token_accuracy": 0.24179935604333877,
|
|
"num_tokens": 18277371.0,
|
|
"step": 10400
|
|
},
|
|
{
|
|
"entropy": 5.271374177932739,
|
|
"epoch": 8.154388714733543,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.282984008299196e-05,
|
|
"loss": 4.3943,
|
|
"mean_token_accuracy": 0.2514119684696198,
|
|
"num_tokens": 18287503.0,
|
|
"step": 10405
|
|
},
|
|
{
|
|
"entropy": 5.268272542953492,
|
|
"epoch": 8.158307210031348,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 9.265346288447516e-05,
|
|
"loss": 4.3902,
|
|
"mean_token_accuracy": 0.25022612512111664,
|
|
"num_tokens": 18296753.0,
|
|
"step": 10410
|
|
},
|
|
{
|
|
"entropy": 5.347469520568848,
|
|
"epoch": 8.162225705329154,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.24774115687862e-05,
|
|
"loss": 4.4179,
|
|
"mean_token_accuracy": 0.2531977593898773,
|
|
"num_tokens": 18305068.0,
|
|
"step": 10415
|
|
},
|
|
{
|
|
"entropy": 5.314385604858399,
|
|
"epoch": 8.16614420062696,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.23016864505575e-05,
|
|
"loss": 4.4148,
|
|
"mean_token_accuracy": 0.25255493223667147,
|
|
"num_tokens": 18313816.0,
|
|
"step": 10420
|
|
},
|
|
{
|
|
"entropy": 5.256226062774658,
|
|
"epoch": 8.170062695924765,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.212628784383825e-05,
|
|
"loss": 4.3294,
|
|
"mean_token_accuracy": 0.2581601470708847,
|
|
"num_tokens": 18322630.0,
|
|
"step": 10425
|
|
},
|
|
{
|
|
"entropy": 5.225310468673706,
|
|
"epoch": 8.17398119122257,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.195121606209411e-05,
|
|
"loss": 4.3154,
|
|
"mean_token_accuracy": 0.2613031044602394,
|
|
"num_tokens": 18330942.0,
|
|
"step": 10430
|
|
},
|
|
{
|
|
"entropy": 5.225821447372437,
|
|
"epoch": 8.177899686520377,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 9.177647141820693e-05,
|
|
"loss": 4.3811,
|
|
"mean_token_accuracy": 0.25662956684827803,
|
|
"num_tokens": 18339375.0,
|
|
"step": 10435
|
|
},
|
|
{
|
|
"entropy": 5.327909469604492,
|
|
"epoch": 8.181818181818182,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 9.160205422447357e-05,
|
|
"loss": 4.3952,
|
|
"mean_token_accuracy": 0.2502524420619011,
|
|
"num_tokens": 18347884.0,
|
|
"step": 10440
|
|
},
|
|
{
|
|
"entropy": 5.26903600692749,
|
|
"epoch": 8.185736677115987,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.142796479260588e-05,
|
|
"loss": 4.3961,
|
|
"mean_token_accuracy": 0.2517066657543182,
|
|
"num_tokens": 18357201.0,
|
|
"step": 10445
|
|
},
|
|
{
|
|
"entropy": 5.319806671142578,
|
|
"epoch": 8.189655172413794,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.125420343372995e-05,
|
|
"loss": 4.4719,
|
|
"mean_token_accuracy": 0.23923082500696183,
|
|
"num_tokens": 18365716.0,
|
|
"step": 10450
|
|
},
|
|
{
|
|
"entropy": 5.252657890319824,
|
|
"epoch": 8.193573667711599,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.108077045838546e-05,
|
|
"loss": 4.369,
|
|
"mean_token_accuracy": 0.2562560364603996,
|
|
"num_tokens": 18374724.0,
|
|
"step": 10455
|
|
},
|
|
{
|
|
"entropy": 5.284126329421997,
|
|
"epoch": 8.197492163009404,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.090766617652525e-05,
|
|
"loss": 4.4049,
|
|
"mean_token_accuracy": 0.2520502045750618,
|
|
"num_tokens": 18383653.0,
|
|
"step": 10460
|
|
},
|
|
{
|
|
"entropy": 5.320044612884521,
|
|
"epoch": 8.20141065830721,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.073489089751468e-05,
|
|
"loss": 4.391,
|
|
"mean_token_accuracy": 0.25724842995405195,
|
|
"num_tokens": 18392729.0,
|
|
"step": 10465
|
|
},
|
|
{
|
|
"entropy": 5.322581815719604,
|
|
"epoch": 8.205329153605016,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.056244493013129e-05,
|
|
"loss": 4.4294,
|
|
"mean_token_accuracy": 0.24467359781265258,
|
|
"num_tokens": 18402257.0,
|
|
"step": 10470
|
|
},
|
|
{
|
|
"entropy": 5.303873825073242,
|
|
"epoch": 8.20924764890282,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.039032858256401e-05,
|
|
"loss": 4.4279,
|
|
"mean_token_accuracy": 0.24974759817123413,
|
|
"num_tokens": 18410714.0,
|
|
"step": 10475
|
|
},
|
|
{
|
|
"entropy": 5.3025084972381595,
|
|
"epoch": 8.213166144200628,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.021854216241251e-05,
|
|
"loss": 4.4304,
|
|
"mean_token_accuracy": 0.24507664740085602,
|
|
"num_tokens": 18419256.0,
|
|
"step": 10480
|
|
},
|
|
{
|
|
"entropy": 5.261106204986572,
|
|
"epoch": 8.217084639498433,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.004708597668716e-05,
|
|
"loss": 4.3904,
|
|
"mean_token_accuracy": 0.2551694124937057,
|
|
"num_tokens": 18428202.0,
|
|
"step": 10485
|
|
},
|
|
{
|
|
"entropy": 5.297854089736939,
|
|
"epoch": 8.221003134796238,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 8.987596033180784e-05,
|
|
"loss": 4.369,
|
|
"mean_token_accuracy": 0.2512460991740227,
|
|
"num_tokens": 18437339.0,
|
|
"step": 10490
|
|
},
|
|
{
|
|
"entropy": 5.307880020141601,
|
|
"epoch": 8.224921630094045,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.970516553360383e-05,
|
|
"loss": 4.3821,
|
|
"mean_token_accuracy": 0.25112576633691785,
|
|
"num_tokens": 18446667.0,
|
|
"step": 10495
|
|
},
|
|
{
|
|
"entropy": 5.295121383666992,
|
|
"epoch": 8.22884012539185,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 8.953470188731318e-05,
|
|
"loss": 4.3166,
|
|
"mean_token_accuracy": 0.25473653674125674,
|
|
"num_tokens": 18455425.0,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"epoch": 8.22884012539185,
|
|
"eval_entropy": 5.140226233837216,
|
|
"eval_loss": 5.630944728851318,
|
|
"eval_mean_token_accuracy": 0.17593607541837897,
|
|
"eval_num_tokens": 18455425.0,
|
|
"eval_runtime": 2.1762,
|
|
"eval_samples_per_second": 1894.098,
|
|
"eval_steps_per_second": 237.107,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"entropy": 5.272479438781739,
|
|
"epoch": 8.232758620689655,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.936456969758206e-05,
|
|
"loss": 4.3025,
|
|
"mean_token_accuracy": 0.25655621737241746,
|
|
"num_tokens": 18464976.0,
|
|
"step": 10505
|
|
},
|
|
{
|
|
"entropy": 5.2995849609375,
|
|
"epoch": 8.236677115987462,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.919476926846426e-05,
|
|
"loss": 4.3658,
|
|
"mean_token_accuracy": 0.2506093055009842,
|
|
"num_tokens": 18473709.0,
|
|
"step": 10510
|
|
},
|
|
{
|
|
"entropy": 5.2995189189910885,
|
|
"epoch": 8.240595611285267,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.90253009034206e-05,
|
|
"loss": 4.383,
|
|
"mean_token_accuracy": 0.24999577254056932,
|
|
"num_tokens": 18483208.0,
|
|
"step": 10515
|
|
},
|
|
{
|
|
"entropy": 5.329328727722168,
|
|
"epoch": 8.244514106583072,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.88561649053187e-05,
|
|
"loss": 4.4369,
|
|
"mean_token_accuracy": 0.24854538291692735,
|
|
"num_tokens": 18491872.0,
|
|
"step": 10520
|
|
},
|
|
{
|
|
"entropy": 5.347621011734009,
|
|
"epoch": 8.248432601880877,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.868736157643183e-05,
|
|
"loss": 4.458,
|
|
"mean_token_accuracy": 0.24311192333698273,
|
|
"num_tokens": 18500740.0,
|
|
"step": 10525
|
|
},
|
|
{
|
|
"entropy": 5.280243015289306,
|
|
"epoch": 8.252351097178684,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.851889121843904e-05,
|
|
"loss": 4.3586,
|
|
"mean_token_accuracy": 0.25511642545461655,
|
|
"num_tokens": 18509553.0,
|
|
"step": 10530
|
|
},
|
|
{
|
|
"entropy": 5.329476261138916,
|
|
"epoch": 8.256269592476489,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 8.835075413242408e-05,
|
|
"loss": 4.4316,
|
|
"mean_token_accuracy": 0.24932334274053575,
|
|
"num_tokens": 18518549.0,
|
|
"step": 10535
|
|
},
|
|
{
|
|
"entropy": 5.352941465377808,
|
|
"epoch": 8.260188087774294,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.818295061887531e-05,
|
|
"loss": 4.4162,
|
|
"mean_token_accuracy": 0.24284698516130448,
|
|
"num_tokens": 18527264.0,
|
|
"step": 10540
|
|
},
|
|
{
|
|
"entropy": 5.2730189800262455,
|
|
"epoch": 8.2641065830721,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 8.801548097768473e-05,
|
|
"loss": 4.3518,
|
|
"mean_token_accuracy": 0.25624258667230604,
|
|
"num_tokens": 18535193.0,
|
|
"step": 10545
|
|
},
|
|
{
|
|
"entropy": 5.32023024559021,
|
|
"epoch": 8.268025078369906,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.784834550814769e-05,
|
|
"loss": 4.3957,
|
|
"mean_token_accuracy": 0.25001410841941835,
|
|
"num_tokens": 18544386.0,
|
|
"step": 10550
|
|
},
|
|
{
|
|
"entropy": 5.248248481750489,
|
|
"epoch": 8.271943573667711,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.768154450896253e-05,
|
|
"loss": 4.2544,
|
|
"mean_token_accuracy": 0.2704042851924896,
|
|
"num_tokens": 18552745.0,
|
|
"step": 10555
|
|
},
|
|
{
|
|
"entropy": 5.303812074661255,
|
|
"epoch": 8.275862068965518,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.751507827822955e-05,
|
|
"loss": 4.3719,
|
|
"mean_token_accuracy": 0.24957126677036284,
|
|
"num_tokens": 18560953.0,
|
|
"step": 10560
|
|
},
|
|
{
|
|
"entropy": 5.329117012023926,
|
|
"epoch": 8.279780564263323,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.73489471134509e-05,
|
|
"loss": 4.3403,
|
|
"mean_token_accuracy": 0.2676517054438591,
|
|
"num_tokens": 18569324.0,
|
|
"step": 10565
|
|
},
|
|
{
|
|
"entropy": 5.375481605529785,
|
|
"epoch": 8.283699059561128,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.718315131152999e-05,
|
|
"loss": 4.4554,
|
|
"mean_token_accuracy": 0.2506739392876625,
|
|
"num_tokens": 18578432.0,
|
|
"step": 10570
|
|
},
|
|
{
|
|
"entropy": 5.261638069152832,
|
|
"epoch": 8.287617554858935,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.701769116877076e-05,
|
|
"loss": 4.3748,
|
|
"mean_token_accuracy": 0.24919955432415009,
|
|
"num_tokens": 18587147.0,
|
|
"step": 10575
|
|
},
|
|
{
|
|
"entropy": 5.355082893371582,
|
|
"epoch": 8.29153605015674,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.685256698087725e-05,
|
|
"loss": 4.379,
|
|
"mean_token_accuracy": 0.2546279951930046,
|
|
"num_tokens": 18596933.0,
|
|
"step": 10580
|
|
},
|
|
{
|
|
"entropy": 5.396424198150635,
|
|
"epoch": 8.295454545454545,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.668777904295328e-05,
|
|
"loss": 4.4486,
|
|
"mean_token_accuracy": 0.24964260309934616,
|
|
"num_tokens": 18605666.0,
|
|
"step": 10585
|
|
},
|
|
{
|
|
"entropy": 5.338765382766724,
|
|
"epoch": 8.299373040752352,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.652332764950147e-05,
|
|
"loss": 4.4788,
|
|
"mean_token_accuracy": 0.25203475058078767,
|
|
"num_tokens": 18614801.0,
|
|
"step": 10590
|
|
},
|
|
{
|
|
"entropy": 5.30225944519043,
|
|
"epoch": 8.303291536050157,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.635921309442327e-05,
|
|
"loss": 4.36,
|
|
"mean_token_accuracy": 0.24887045472860336,
|
|
"num_tokens": 18623489.0,
|
|
"step": 10595
|
|
},
|
|
{
|
|
"entropy": 5.288321685791016,
|
|
"epoch": 8.307210031347962,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 8.619543567101787e-05,
|
|
"loss": 4.3778,
|
|
"mean_token_accuracy": 0.2502602532505989,
|
|
"num_tokens": 18631496.0,
|
|
"step": 10600
|
|
},
|
|
{
|
|
"entropy": 5.282364082336426,
|
|
"epoch": 8.311128526645769,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 8.603199567198222e-05,
|
|
"loss": 4.3914,
|
|
"mean_token_accuracy": 0.25636345744132993,
|
|
"num_tokens": 18640481.0,
|
|
"step": 10605
|
|
},
|
|
{
|
|
"entropy": 5.360686922073365,
|
|
"epoch": 8.315047021943574,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.586889338941002e-05,
|
|
"loss": 4.4868,
|
|
"mean_token_accuracy": 0.23808801174163818,
|
|
"num_tokens": 18650118.0,
|
|
"step": 10610
|
|
},
|
|
{
|
|
"entropy": 5.223593378067017,
|
|
"epoch": 8.318965517241379,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.570612911479145e-05,
|
|
"loss": 4.3636,
|
|
"mean_token_accuracy": 0.2571857914328575,
|
|
"num_tokens": 18658981.0,
|
|
"step": 10615
|
|
},
|
|
{
|
|
"entropy": 5.333513784408569,
|
|
"epoch": 8.322884012539184,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.554370313901273e-05,
|
|
"loss": 4.4205,
|
|
"mean_token_accuracy": 0.25291977524757386,
|
|
"num_tokens": 18668324.0,
|
|
"step": 10620
|
|
},
|
|
{
|
|
"entropy": 5.263721513748169,
|
|
"epoch": 8.326802507836991,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.538161575235541e-05,
|
|
"loss": 4.328,
|
|
"mean_token_accuracy": 0.2610509663820267,
|
|
"num_tokens": 18676945.0,
|
|
"step": 10625
|
|
},
|
|
{
|
|
"entropy": 5.33682804107666,
|
|
"epoch": 8.330721003134796,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.521986724449593e-05,
|
|
"loss": 4.4258,
|
|
"mean_token_accuracy": 0.25045063346624374,
|
|
"num_tokens": 18685980.0,
|
|
"step": 10630
|
|
},
|
|
{
|
|
"entropy": 5.348587560653686,
|
|
"epoch": 8.334639498432601,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.5058457904505e-05,
|
|
"loss": 4.4125,
|
|
"mean_token_accuracy": 0.2546412229537964,
|
|
"num_tokens": 18694834.0,
|
|
"step": 10635
|
|
},
|
|
{
|
|
"entropy": 5.27523717880249,
|
|
"epoch": 8.338557993730408,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.489738802084742e-05,
|
|
"loss": 4.3804,
|
|
"mean_token_accuracy": 0.24624210745096206,
|
|
"num_tokens": 18703747.0,
|
|
"step": 10640
|
|
},
|
|
{
|
|
"entropy": 5.265012407302857,
|
|
"epoch": 8.342476489028213,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 8.473665788138112e-05,
|
|
"loss": 4.3601,
|
|
"mean_token_accuracy": 0.25592857748270037,
|
|
"num_tokens": 18712202.0,
|
|
"step": 10645
|
|
},
|
|
{
|
|
"entropy": 5.2745460033416744,
|
|
"epoch": 8.346394984326018,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.457626777335691e-05,
|
|
"loss": 4.3921,
|
|
"mean_token_accuracy": 0.25335101038217545,
|
|
"num_tokens": 18721340.0,
|
|
"step": 10650
|
|
},
|
|
{
|
|
"entropy": 5.273308324813843,
|
|
"epoch": 8.350313479623825,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.4416217983418e-05,
|
|
"loss": 4.4253,
|
|
"mean_token_accuracy": 0.2472114533185959,
|
|
"num_tokens": 18730295.0,
|
|
"step": 10655
|
|
},
|
|
{
|
|
"entropy": 5.2570513725280765,
|
|
"epoch": 8.35423197492163,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.425650879759923e-05,
|
|
"loss": 4.3855,
|
|
"mean_token_accuracy": 0.25294836461544035,
|
|
"num_tokens": 18739040.0,
|
|
"step": 10660
|
|
},
|
|
{
|
|
"entropy": 5.195999431610107,
|
|
"epoch": 8.358150470219435,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.409714050132683e-05,
|
|
"loss": 4.2987,
|
|
"mean_token_accuracy": 0.25850880742073057,
|
|
"num_tokens": 18747655.0,
|
|
"step": 10665
|
|
},
|
|
{
|
|
"entropy": 5.292117309570313,
|
|
"epoch": 8.362068965517242,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.393811337941781e-05,
|
|
"loss": 4.3435,
|
|
"mean_token_accuracy": 0.26231554001569746,
|
|
"num_tokens": 18755602.0,
|
|
"step": 10670
|
|
},
|
|
{
|
|
"entropy": 5.261296367645263,
|
|
"epoch": 8.365987460815047,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.377942771607943e-05,
|
|
"loss": 4.3393,
|
|
"mean_token_accuracy": 0.26841824054718016,
|
|
"num_tokens": 18764332.0,
|
|
"step": 10675
|
|
},
|
|
{
|
|
"entropy": 5.23098840713501,
|
|
"epoch": 8.369905956112852,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.362108379490868e-05,
|
|
"loss": 4.3931,
|
|
"mean_token_accuracy": 0.25224552750587464,
|
|
"num_tokens": 18773106.0,
|
|
"step": 10680
|
|
},
|
|
{
|
|
"entropy": 5.270829772949218,
|
|
"epoch": 8.373824451410659,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.346308189889175e-05,
|
|
"loss": 4.4088,
|
|
"mean_token_accuracy": 0.2479039713740349,
|
|
"num_tokens": 18781603.0,
|
|
"step": 10685
|
|
},
|
|
{
|
|
"entropy": 5.222823619842529,
|
|
"epoch": 8.377742946708464,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 8.330542231040379e-05,
|
|
"loss": 4.3408,
|
|
"mean_token_accuracy": 0.2583801791071892,
|
|
"num_tokens": 18790152.0,
|
|
"step": 10690
|
|
},
|
|
{
|
|
"entropy": 5.292102479934693,
|
|
"epoch": 8.38166144200627,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.314810531120795e-05,
|
|
"loss": 4.332,
|
|
"mean_token_accuracy": 0.2617406189441681,
|
|
"num_tokens": 18798397.0,
|
|
"step": 10695
|
|
},
|
|
{
|
|
"entropy": 5.281747913360595,
|
|
"epoch": 8.385579937304076,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.299113118245519e-05,
|
|
"loss": 4.3695,
|
|
"mean_token_accuracy": 0.2566616863012314,
|
|
"num_tokens": 18807523.0,
|
|
"step": 10700
|
|
},
|
|
{
|
|
"entropy": 5.283020544052124,
|
|
"epoch": 8.389498432601881,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.283450020468377e-05,
|
|
"loss": 4.3869,
|
|
"mean_token_accuracy": 0.25127487629652023,
|
|
"num_tokens": 18816469.0,
|
|
"step": 10705
|
|
},
|
|
{
|
|
"entropy": 5.3216142654418945,
|
|
"epoch": 8.393416927899686,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.267821265781875e-05,
|
|
"loss": 4.4089,
|
|
"mean_token_accuracy": 0.2508828520774841,
|
|
"num_tokens": 18823994.0,
|
|
"step": 10710
|
|
},
|
|
{
|
|
"entropy": 5.231358575820923,
|
|
"epoch": 8.397335423197493,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.252226882117118e-05,
|
|
"loss": 4.3127,
|
|
"mean_token_accuracy": 0.2560754418373108,
|
|
"num_tokens": 18833159.0,
|
|
"step": 10715
|
|
},
|
|
{
|
|
"entropy": 5.231699132919312,
|
|
"epoch": 8.401253918495298,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.236666897343797e-05,
|
|
"loss": 4.3243,
|
|
"mean_token_accuracy": 0.25587553828954696,
|
|
"num_tokens": 18841754.0,
|
|
"step": 10720
|
|
},
|
|
{
|
|
"entropy": 5.206751728057862,
|
|
"epoch": 8.405172413793103,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.221141339270138e-05,
|
|
"loss": 4.3504,
|
|
"mean_token_accuracy": 0.2586342439055443,
|
|
"num_tokens": 18850493.0,
|
|
"step": 10725
|
|
},
|
|
{
|
|
"entropy": 5.370335578918457,
|
|
"epoch": 8.409090909090908,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.205650235642828e-05,
|
|
"loss": 4.4535,
|
|
"mean_token_accuracy": 0.2536746859550476,
|
|
"num_tokens": 18859765.0,
|
|
"step": 10730
|
|
},
|
|
{
|
|
"entropy": 5.288048124313354,
|
|
"epoch": 8.413009404388715,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.190193614146971e-05,
|
|
"loss": 4.4192,
|
|
"mean_token_accuracy": 0.25452754348516465,
|
|
"num_tokens": 18868685.0,
|
|
"step": 10735
|
|
},
|
|
{
|
|
"entropy": 5.272276496887207,
|
|
"epoch": 8.41692789968652,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.17477150240607e-05,
|
|
"loss": 4.3446,
|
|
"mean_token_accuracy": 0.25697235614061353,
|
|
"num_tokens": 18877496.0,
|
|
"step": 10740
|
|
},
|
|
{
|
|
"entropy": 5.295509481430054,
|
|
"epoch": 8.420846394984325,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.159383927981931e-05,
|
|
"loss": 4.3606,
|
|
"mean_token_accuracy": 0.2591456785798073,
|
|
"num_tokens": 18885828.0,
|
|
"step": 10745
|
|
},
|
|
{
|
|
"entropy": 5.31722731590271,
|
|
"epoch": 8.424764890282132,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.14403091837464e-05,
|
|
"loss": 4.435,
|
|
"mean_token_accuracy": 0.24441258162260054,
|
|
"num_tokens": 18894498.0,
|
|
"step": 10750
|
|
},
|
|
{
|
|
"entropy": 5.297009992599487,
|
|
"epoch": 8.428683385579937,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.128712501022526e-05,
|
|
"loss": 4.3824,
|
|
"mean_token_accuracy": 0.25322997868061065,
|
|
"num_tokens": 18902996.0,
|
|
"step": 10755
|
|
},
|
|
{
|
|
"entropy": 5.231626415252686,
|
|
"epoch": 8.432601880877742,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.11342870330208e-05,
|
|
"loss": 4.3202,
|
|
"mean_token_accuracy": 0.26427613347768786,
|
|
"num_tokens": 18911731.0,
|
|
"step": 10760
|
|
},
|
|
{
|
|
"entropy": 5.259805822372437,
|
|
"epoch": 8.43652037617555,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.098179552527921e-05,
|
|
"loss": 4.2932,
|
|
"mean_token_accuracy": 0.2621105447411537,
|
|
"num_tokens": 18920912.0,
|
|
"step": 10765
|
|
},
|
|
{
|
|
"entropy": 5.262525320053101,
|
|
"epoch": 8.440438871473354,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 8.082965075952757e-05,
|
|
"loss": 4.3267,
|
|
"mean_token_accuracy": 0.2555623769760132,
|
|
"num_tokens": 18930012.0,
|
|
"step": 10770
|
|
},
|
|
{
|
|
"entropy": 5.28326382637024,
|
|
"epoch": 8.44435736677116,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.067785300767333e-05,
|
|
"loss": 4.4597,
|
|
"mean_token_accuracy": 0.24341157227754592,
|
|
"num_tokens": 18939026.0,
|
|
"step": 10775
|
|
},
|
|
{
|
|
"entropy": 5.2820258140563965,
|
|
"epoch": 8.448275862068966,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.05264025410036e-05,
|
|
"loss": 4.3733,
|
|
"mean_token_accuracy": 0.2493313431739807,
|
|
"num_tokens": 18948387.0,
|
|
"step": 10780
|
|
},
|
|
{
|
|
"entropy": 5.30631103515625,
|
|
"epoch": 8.452194357366771,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.037529963018488e-05,
|
|
"loss": 4.4313,
|
|
"mean_token_accuracy": 0.24923575520515442,
|
|
"num_tokens": 18957234.0,
|
|
"step": 10785
|
|
},
|
|
{
|
|
"entropy": 5.269313049316406,
|
|
"epoch": 8.456112852664576,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 8.022454454526272e-05,
|
|
"loss": 4.3611,
|
|
"mean_token_accuracy": 0.2537208527326584,
|
|
"num_tokens": 18965501.0,
|
|
"step": 10790
|
|
},
|
|
{
|
|
"entropy": 5.277254104614258,
|
|
"epoch": 8.460031347962383,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.00741375556608e-05,
|
|
"loss": 4.3595,
|
|
"mean_token_accuracy": 0.25260152816772463,
|
|
"num_tokens": 18973225.0,
|
|
"step": 10795
|
|
},
|
|
{
|
|
"entropy": 5.328072166442871,
|
|
"epoch": 8.463949843260188,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.99240789301808e-05,
|
|
"loss": 4.3867,
|
|
"mean_token_accuracy": 0.25290377587080004,
|
|
"num_tokens": 18982424.0,
|
|
"step": 10800
|
|
},
|
|
{
|
|
"entropy": 5.2758245944976805,
|
|
"epoch": 8.467868338557993,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 7.977436893700177e-05,
|
|
"loss": 4.4115,
|
|
"mean_token_accuracy": 0.25121662616729734,
|
|
"num_tokens": 18991519.0,
|
|
"step": 10805
|
|
},
|
|
{
|
|
"entropy": 5.240723609924316,
|
|
"epoch": 8.4717868338558,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.962500784367985e-05,
|
|
"loss": 4.3561,
|
|
"mean_token_accuracy": 0.25352340638637544,
|
|
"num_tokens": 19000498.0,
|
|
"step": 10810
|
|
},
|
|
{
|
|
"entropy": 5.327727174758911,
|
|
"epoch": 8.475705329153605,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.947599591714746e-05,
|
|
"loss": 4.4565,
|
|
"mean_token_accuracy": 0.24775171279907227,
|
|
"num_tokens": 19009640.0,
|
|
"step": 10815
|
|
},
|
|
{
|
|
"entropy": 5.257160186767578,
|
|
"epoch": 8.47962382445141,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.932733342371306e-05,
|
|
"loss": 4.4093,
|
|
"mean_token_accuracy": 0.24734651446342468,
|
|
"num_tokens": 19018337.0,
|
|
"step": 10820
|
|
},
|
|
{
|
|
"entropy": 5.360875749588013,
|
|
"epoch": 8.483542319749215,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.917902062906066e-05,
|
|
"loss": 4.4614,
|
|
"mean_token_accuracy": 0.2480304643511772,
|
|
"num_tokens": 19027232.0,
|
|
"step": 10825
|
|
},
|
|
{
|
|
"entropy": 5.206266832351685,
|
|
"epoch": 8.487460815047022,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.90310577982493e-05,
|
|
"loss": 4.3294,
|
|
"mean_token_accuracy": 0.2549117118120193,
|
|
"num_tokens": 19036661.0,
|
|
"step": 10830
|
|
},
|
|
{
|
|
"entropy": 5.26503472328186,
|
|
"epoch": 8.491379310344827,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.888344519571247e-05,
|
|
"loss": 4.3434,
|
|
"mean_token_accuracy": 0.25544548183679583,
|
|
"num_tokens": 19045955.0,
|
|
"step": 10835
|
|
},
|
|
{
|
|
"entropy": 5.361381912231446,
|
|
"epoch": 8.495297805642632,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 7.873618308525797e-05,
|
|
"loss": 4.4486,
|
|
"mean_token_accuracy": 0.24869574457406998,
|
|
"num_tokens": 19054183.0,
|
|
"step": 10840
|
|
},
|
|
{
|
|
"entropy": 5.377130079269409,
|
|
"epoch": 8.49921630094044,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.858927173006694e-05,
|
|
"loss": 4.3997,
|
|
"mean_token_accuracy": 0.2571351394057274,
|
|
"num_tokens": 19063029.0,
|
|
"step": 10845
|
|
},
|
|
{
|
|
"entropy": 5.279720449447632,
|
|
"epoch": 8.503134796238244,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.844271139269384e-05,
|
|
"loss": 4.3578,
|
|
"mean_token_accuracy": 0.2560381978750229,
|
|
"num_tokens": 19071325.0,
|
|
"step": 10850
|
|
},
|
|
{
|
|
"entropy": 5.296114444732666,
|
|
"epoch": 8.50705329153605,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.829650233506569e-05,
|
|
"loss": 4.4445,
|
|
"mean_token_accuracy": 0.24538001418113708,
|
|
"num_tokens": 19080082.0,
|
|
"step": 10855
|
|
},
|
|
{
|
|
"entropy": 5.226892137527466,
|
|
"epoch": 8.510971786833856,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.815064481848188e-05,
|
|
"loss": 4.3301,
|
|
"mean_token_accuracy": 0.2542010396718979,
|
|
"num_tokens": 19089286.0,
|
|
"step": 10860
|
|
},
|
|
{
|
|
"entropy": 5.266781759262085,
|
|
"epoch": 8.514890282131661,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 7.800513910361343e-05,
|
|
"loss": 4.3062,
|
|
"mean_token_accuracy": 0.26650853604078295,
|
|
"num_tokens": 19098397.0,
|
|
"step": 10865
|
|
},
|
|
{
|
|
"entropy": 5.357656335830688,
|
|
"epoch": 8.518808777429467,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.785998545050253e-05,
|
|
"loss": 4.4699,
|
|
"mean_token_accuracy": 0.2350107967853546,
|
|
"num_tokens": 19107939.0,
|
|
"step": 10870
|
|
},
|
|
{
|
|
"entropy": 5.297347164154052,
|
|
"epoch": 8.522727272727273,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.771518411856243e-05,
|
|
"loss": 4.4191,
|
|
"mean_token_accuracy": 0.24820801466703415,
|
|
"num_tokens": 19116541.0,
|
|
"step": 10875
|
|
},
|
|
{
|
|
"entropy": 5.376194810867309,
|
|
"epoch": 8.526645768025078,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.75707353665765e-05,
|
|
"loss": 4.4352,
|
|
"mean_token_accuracy": 0.24832122027873993,
|
|
"num_tokens": 19124830.0,
|
|
"step": 10880
|
|
},
|
|
{
|
|
"entropy": 5.216870880126953,
|
|
"epoch": 8.530564263322884,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.742663945269816e-05,
|
|
"loss": 4.2664,
|
|
"mean_token_accuracy": 0.25994435101747515,
|
|
"num_tokens": 19133518.0,
|
|
"step": 10885
|
|
},
|
|
{
|
|
"entropy": 5.277771282196045,
|
|
"epoch": 8.53448275862069,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.72828966344501e-05,
|
|
"loss": 4.4381,
|
|
"mean_token_accuracy": 0.24971903860569,
|
|
"num_tokens": 19142989.0,
|
|
"step": 10890
|
|
},
|
|
{
|
|
"entropy": 5.373530054092408,
|
|
"epoch": 8.538401253918495,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.713950716872411e-05,
|
|
"loss": 4.5045,
|
|
"mean_token_accuracy": 0.24832094162702562,
|
|
"num_tokens": 19151603.0,
|
|
"step": 10895
|
|
},
|
|
{
|
|
"entropy": 5.275055551528931,
|
|
"epoch": 8.5423197492163,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 7.699647131178038e-05,
|
|
"loss": 4.3055,
|
|
"mean_token_accuracy": 0.260811322927475,
|
|
"num_tokens": 19159949.0,
|
|
"step": 10900
|
|
},
|
|
{
|
|
"entropy": 5.328347778320312,
|
|
"epoch": 8.546238244514107,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.685378931924713e-05,
|
|
"loss": 4.4391,
|
|
"mean_token_accuracy": 0.2489410936832428,
|
|
"num_tokens": 19169594.0,
|
|
"step": 10905
|
|
},
|
|
{
|
|
"entropy": 5.310245752334595,
|
|
"epoch": 8.550156739811912,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 7.67114614461203e-05,
|
|
"loss": 4.3381,
|
|
"mean_token_accuracy": 0.2591031387448311,
|
|
"num_tokens": 19177660.0,
|
|
"step": 10910
|
|
},
|
|
{
|
|
"entropy": 5.2683141231536865,
|
|
"epoch": 8.554075235109718,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.656948794676283e-05,
|
|
"loss": 4.3746,
|
|
"mean_token_accuracy": 0.24653276056051254,
|
|
"num_tokens": 19185890.0,
|
|
"step": 10915
|
|
},
|
|
{
|
|
"entropy": 5.206455564498901,
|
|
"epoch": 8.557993730407524,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.642786907490437e-05,
|
|
"loss": 4.3242,
|
|
"mean_token_accuracy": 0.25133630633354187,
|
|
"num_tokens": 19194820.0,
|
|
"step": 10920
|
|
},
|
|
{
|
|
"entropy": 5.250659322738647,
|
|
"epoch": 8.56191222570533,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.628660508364083e-05,
|
|
"loss": 4.4078,
|
|
"mean_token_accuracy": 0.24584190547466278,
|
|
"num_tokens": 19203887.0,
|
|
"step": 10925
|
|
},
|
|
{
|
|
"entropy": 5.323218297958374,
|
|
"epoch": 8.565830721003135,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.614569622543391e-05,
|
|
"loss": 4.4479,
|
|
"mean_token_accuracy": 0.2405647248029709,
|
|
"num_tokens": 19212540.0,
|
|
"step": 10930
|
|
},
|
|
{
|
|
"entropy": 5.295700883865356,
|
|
"epoch": 8.56974921630094,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.600514275211054e-05,
|
|
"loss": 4.3531,
|
|
"mean_token_accuracy": 0.2557737663388252,
|
|
"num_tokens": 19221838.0,
|
|
"step": 10935
|
|
},
|
|
{
|
|
"entropy": 5.184952878952027,
|
|
"epoch": 8.573667711598747,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 7.58649449148625e-05,
|
|
"loss": 4.2741,
|
|
"mean_token_accuracy": 0.2671185776591301,
|
|
"num_tokens": 19229538.0,
|
|
"step": 10940
|
|
},
|
|
{
|
|
"entropy": 5.207627534866333,
|
|
"epoch": 8.577586206896552,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.572510296424615e-05,
|
|
"loss": 4.2608,
|
|
"mean_token_accuracy": 0.25884290784597397,
|
|
"num_tokens": 19237977.0,
|
|
"step": 10945
|
|
},
|
|
{
|
|
"entropy": 5.258052396774292,
|
|
"epoch": 8.581504702194357,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.558561715018174e-05,
|
|
"loss": 4.3663,
|
|
"mean_token_accuracy": 0.2493900939822197,
|
|
"num_tokens": 19246801.0,
|
|
"step": 10950
|
|
},
|
|
{
|
|
"entropy": 5.25008749961853,
|
|
"epoch": 8.585423197492164,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.544648772195293e-05,
|
|
"loss": 4.3919,
|
|
"mean_token_accuracy": 0.25218057036399844,
|
|
"num_tokens": 19256271.0,
|
|
"step": 10955
|
|
},
|
|
{
|
|
"entropy": 5.297944688796997,
|
|
"epoch": 8.589341692789969,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.530771492820673e-05,
|
|
"loss": 4.3825,
|
|
"mean_token_accuracy": 0.24794732928276061,
|
|
"num_tokens": 19264210.0,
|
|
"step": 10960
|
|
},
|
|
{
|
|
"entropy": 5.217477464675904,
|
|
"epoch": 8.593260188087774,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.516929901695249e-05,
|
|
"loss": 4.2965,
|
|
"mean_token_accuracy": 0.25855760723352433,
|
|
"num_tokens": 19272700.0,
|
|
"step": 10965
|
|
},
|
|
{
|
|
"entropy": 5.300490522384644,
|
|
"epoch": 8.59717868338558,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.503124023556192e-05,
|
|
"loss": 4.4108,
|
|
"mean_token_accuracy": 0.2538084745407104,
|
|
"num_tokens": 19281674.0,
|
|
"step": 10970
|
|
},
|
|
{
|
|
"entropy": 5.296628141403199,
|
|
"epoch": 8.601097178683386,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.48935388307685e-05,
|
|
"loss": 4.341,
|
|
"mean_token_accuracy": 0.25982641726732253,
|
|
"num_tokens": 19290748.0,
|
|
"step": 10975
|
|
},
|
|
{
|
|
"entropy": 5.277150583267212,
|
|
"epoch": 8.60501567398119,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 7.47561950486669e-05,
|
|
"loss": 4.2991,
|
|
"mean_token_accuracy": 0.2613372579216957,
|
|
"num_tokens": 19299018.0,
|
|
"step": 10980
|
|
},
|
|
{
|
|
"entropy": 5.226588916778565,
|
|
"epoch": 8.608934169278998,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.461920913471276e-05,
|
|
"loss": 4.3114,
|
|
"mean_token_accuracy": 0.2584723934531212,
|
|
"num_tokens": 19307915.0,
|
|
"step": 10985
|
|
},
|
|
{
|
|
"entropy": 5.276996326446533,
|
|
"epoch": 8.612852664576803,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.448258133372208e-05,
|
|
"loss": 4.3626,
|
|
"mean_token_accuracy": 0.263133205473423,
|
|
"num_tokens": 19316358.0,
|
|
"step": 10990
|
|
},
|
|
{
|
|
"entropy": 5.273407316207885,
|
|
"epoch": 8.616771159874608,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.434631188987095e-05,
|
|
"loss": 4.4533,
|
|
"mean_token_accuracy": 0.2473789393901825,
|
|
"num_tokens": 19325329.0,
|
|
"step": 10995
|
|
},
|
|
{
|
|
"entropy": 5.289061641693115,
|
|
"epoch": 8.620689655172415,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.42104010466949e-05,
|
|
"loss": 4.3887,
|
|
"mean_token_accuracy": 0.2574890300631523,
|
|
"num_tokens": 19333857.0,
|
|
"step": 11000
|
|
},
|
|
{
|
|
"epoch": 8.620689655172415,
|
|
"eval_entropy": 5.1241024039512455,
|
|
"eval_loss": 5.628636837005615,
|
|
"eval_mean_token_accuracy": 0.17640583462608878,
|
|
"eval_num_tokens": 19333857.0,
|
|
"eval_runtime": 2.3392,
|
|
"eval_samples_per_second": 1762.16,
|
|
"eval_steps_per_second": 220.591,
|
|
"step": 11000
|
|
},
|
|
{
|
|
"entropy": 5.26338210105896,
|
|
"epoch": 8.62460815047022,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.407484904708874e-05,
|
|
"loss": 4.3579,
|
|
"mean_token_accuracy": 0.2524025276303291,
|
|
"num_tokens": 19342386.0,
|
|
"step": 11005
|
|
},
|
|
{
|
|
"entropy": 5.293384456634522,
|
|
"epoch": 8.628526645768025,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.393965613330575e-05,
|
|
"loss": 4.3911,
|
|
"mean_token_accuracy": 0.25173813104629517,
|
|
"num_tokens": 19350727.0,
|
|
"step": 11010
|
|
},
|
|
{
|
|
"entropy": 5.274635362625122,
|
|
"epoch": 8.63244514106583,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.38048225469577e-05,
|
|
"loss": 4.4199,
|
|
"mean_token_accuracy": 0.24573325216770173,
|
|
"num_tokens": 19360078.0,
|
|
"step": 11015
|
|
},
|
|
{
|
|
"entropy": 5.267412900924683,
|
|
"epoch": 8.636363636363637,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 7.3670348529014e-05,
|
|
"loss": 4.3725,
|
|
"mean_token_accuracy": 0.2592716455459595,
|
|
"num_tokens": 19369092.0,
|
|
"step": 11020
|
|
},
|
|
{
|
|
"entropy": 5.299619960784912,
|
|
"epoch": 8.640282131661442,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.353623431980153e-05,
|
|
"loss": 4.4007,
|
|
"mean_token_accuracy": 0.25239683538675306,
|
|
"num_tokens": 19378061.0,
|
|
"step": 11025
|
|
},
|
|
{
|
|
"entropy": 5.30927414894104,
|
|
"epoch": 8.644200626959247,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.340248015900416e-05,
|
|
"loss": 4.3732,
|
|
"mean_token_accuracy": 0.25493209064006805,
|
|
"num_tokens": 19386298.0,
|
|
"step": 11030
|
|
},
|
|
{
|
|
"entropy": 5.2218804359436035,
|
|
"epoch": 8.648119122257054,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.326908628566226e-05,
|
|
"loss": 4.3147,
|
|
"mean_token_accuracy": 0.2634222999215126,
|
|
"num_tokens": 19394001.0,
|
|
"step": 11035
|
|
},
|
|
{
|
|
"entropy": 5.177711629867554,
|
|
"epoch": 8.652037617554859,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.313605293817224e-05,
|
|
"loss": 4.279,
|
|
"mean_token_accuracy": 0.25968458950519563,
|
|
"num_tokens": 19402410.0,
|
|
"step": 11040
|
|
},
|
|
{
|
|
"entropy": 5.31434211730957,
|
|
"epoch": 8.655956112852664,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.300338035428639e-05,
|
|
"loss": 4.4345,
|
|
"mean_token_accuracy": 0.24781117588281631,
|
|
"num_tokens": 19410947.0,
|
|
"step": 11045
|
|
},
|
|
{
|
|
"entropy": 5.266879320144653,
|
|
"epoch": 8.65987460815047,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.287106877111203e-05,
|
|
"loss": 4.4444,
|
|
"mean_token_accuracy": 0.24886106103658676,
|
|
"num_tokens": 19419286.0,
|
|
"step": 11050
|
|
},
|
|
{
|
|
"entropy": 5.259985685348511,
|
|
"epoch": 8.663793103448276,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.273911842511139e-05,
|
|
"loss": 4.3473,
|
|
"mean_token_accuracy": 0.2624583452939987,
|
|
"num_tokens": 19427703.0,
|
|
"step": 11055
|
|
},
|
|
{
|
|
"entropy": 5.3090247631073,
|
|
"epoch": 8.66771159874608,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.260752955210124e-05,
|
|
"loss": 4.396,
|
|
"mean_token_accuracy": 0.2540148630738258,
|
|
"num_tokens": 19436329.0,
|
|
"step": 11060
|
|
},
|
|
{
|
|
"entropy": 5.296910762786865,
|
|
"epoch": 8.671630094043888,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.247630238725212e-05,
|
|
"loss": 4.3412,
|
|
"mean_token_accuracy": 0.2610892325639725,
|
|
"num_tokens": 19444363.0,
|
|
"step": 11065
|
|
},
|
|
{
|
|
"entropy": 5.243938207626343,
|
|
"epoch": 8.675548589341693,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.234543716508836e-05,
|
|
"loss": 4.3499,
|
|
"mean_token_accuracy": 0.25507890731096267,
|
|
"num_tokens": 19452745.0,
|
|
"step": 11070
|
|
},
|
|
{
|
|
"entropy": 5.315601921081543,
|
|
"epoch": 8.679467084639498,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.221493411948723e-05,
|
|
"loss": 4.4015,
|
|
"mean_token_accuracy": 0.24559859484434127,
|
|
"num_tokens": 19462481.0,
|
|
"step": 11075
|
|
},
|
|
{
|
|
"entropy": 5.334863519668579,
|
|
"epoch": 8.683385579937305,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 7.208479348367895e-05,
|
|
"loss": 4.3752,
|
|
"mean_token_accuracy": 0.26423546522855756,
|
|
"num_tokens": 19471276.0,
|
|
"step": 11080
|
|
},
|
|
{
|
|
"entropy": 5.288963842391968,
|
|
"epoch": 8.68730407523511,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.195501549024582e-05,
|
|
"loss": 4.4068,
|
|
"mean_token_accuracy": 0.2489018991589546,
|
|
"num_tokens": 19480101.0,
|
|
"step": 11085
|
|
},
|
|
{
|
|
"entropy": 5.2783746242523195,
|
|
"epoch": 8.691222570532915,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.182560037112217e-05,
|
|
"loss": 4.32,
|
|
"mean_token_accuracy": 0.2565345406532288,
|
|
"num_tokens": 19489027.0,
|
|
"step": 11090
|
|
},
|
|
{
|
|
"entropy": 5.317575931549072,
|
|
"epoch": 8.695141065830722,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 7.169654835759388e-05,
|
|
"loss": 4.3699,
|
|
"mean_token_accuracy": 0.2569080233573914,
|
|
"num_tokens": 19497363.0,
|
|
"step": 11095
|
|
},
|
|
{
|
|
"entropy": 5.281228971481323,
|
|
"epoch": 8.699059561128527,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 7.156785968029778e-05,
|
|
"loss": 4.376,
|
|
"mean_token_accuracy": 0.2517779588699341,
|
|
"num_tokens": 19505362.0,
|
|
"step": 11100
|
|
},
|
|
{
|
|
"entropy": 5.197848701477051,
|
|
"epoch": 8.702978056426332,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.143953456922141e-05,
|
|
"loss": 4.3065,
|
|
"mean_token_accuracy": 0.2621604263782501,
|
|
"num_tokens": 19514536.0,
|
|
"step": 11105
|
|
},
|
|
{
|
|
"entropy": 5.260368680953979,
|
|
"epoch": 8.706896551724139,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.131157325370249e-05,
|
|
"loss": 4.3857,
|
|
"mean_token_accuracy": 0.2518179789185524,
|
|
"num_tokens": 19523276.0,
|
|
"step": 11110
|
|
},
|
|
{
|
|
"entropy": 5.248493432998657,
|
|
"epoch": 8.710815047021944,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.118397596242872e-05,
|
|
"loss": 4.2901,
|
|
"mean_token_accuracy": 0.2562062591314316,
|
|
"num_tokens": 19532020.0,
|
|
"step": 11115
|
|
},
|
|
{
|
|
"entropy": 5.29442949295044,
|
|
"epoch": 8.714733542319749,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.105674292343709e-05,
|
|
"loss": 4.3635,
|
|
"mean_token_accuracy": 0.25278604328632354,
|
|
"num_tokens": 19540175.0,
|
|
"step": 11120
|
|
},
|
|
{
|
|
"entropy": 5.344836139678955,
|
|
"epoch": 8.718652037617554,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.092987436411369e-05,
|
|
"loss": 4.4049,
|
|
"mean_token_accuracy": 0.25461010783910754,
|
|
"num_tokens": 19549728.0,
|
|
"step": 11125
|
|
},
|
|
{
|
|
"entropy": 5.250100469589233,
|
|
"epoch": 8.72257053291536,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.080337051119325e-05,
|
|
"loss": 4.3543,
|
|
"mean_token_accuracy": 0.2511607676744461,
|
|
"num_tokens": 19558859.0,
|
|
"step": 11130
|
|
},
|
|
{
|
|
"entropy": 5.244144296646118,
|
|
"epoch": 8.726489028213166,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 7.067723159075864e-05,
|
|
"loss": 4.3347,
|
|
"mean_token_accuracy": 0.25493472814559937,
|
|
"num_tokens": 19567397.0,
|
|
"step": 11135
|
|
},
|
|
{
|
|
"entropy": 5.239000082015991,
|
|
"epoch": 8.730407523510971,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.055145782824051e-05,
|
|
"loss": 4.3498,
|
|
"mean_token_accuracy": 0.25478562265634536,
|
|
"num_tokens": 19575986.0,
|
|
"step": 11140
|
|
},
|
|
{
|
|
"entropy": 5.264275598526001,
|
|
"epoch": 8.734326018808778,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.042604944841705e-05,
|
|
"loss": 4.3538,
|
|
"mean_token_accuracy": 0.2519843280315399,
|
|
"num_tokens": 19585066.0,
|
|
"step": 11145
|
|
},
|
|
{
|
|
"entropy": 5.3024341583251955,
|
|
"epoch": 8.738244514106583,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.030100667541333e-05,
|
|
"loss": 4.4062,
|
|
"mean_token_accuracy": 0.24596820175647735,
|
|
"num_tokens": 19594044.0,
|
|
"step": 11150
|
|
},
|
|
{
|
|
"entropy": 5.266149282455444,
|
|
"epoch": 8.742163009404388,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.017632973270105e-05,
|
|
"loss": 4.3707,
|
|
"mean_token_accuracy": 0.2527815654873848,
|
|
"num_tokens": 19602416.0,
|
|
"step": 11155
|
|
},
|
|
{
|
|
"entropy": 5.282707595825196,
|
|
"epoch": 8.746081504702195,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.005201884309812e-05,
|
|
"loss": 4.3553,
|
|
"mean_token_accuracy": 0.25651171505451204,
|
|
"num_tokens": 19612435.0,
|
|
"step": 11160
|
|
},
|
|
{
|
|
"entropy": 5.304826831817627,
|
|
"epoch": 8.75,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.992807422876829e-05,
|
|
"loss": 4.3936,
|
|
"mean_token_accuracy": 0.2492402523756027,
|
|
"num_tokens": 19621708.0,
|
|
"step": 11165
|
|
},
|
|
{
|
|
"entropy": 5.328975582122803,
|
|
"epoch": 8.753918495297805,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.980449611122065e-05,
|
|
"loss": 4.386,
|
|
"mean_token_accuracy": 0.25495535880327225,
|
|
"num_tokens": 19630879.0,
|
|
"step": 11170
|
|
},
|
|
{
|
|
"entropy": 5.305544328689575,
|
|
"epoch": 8.757836990595612,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.968128471130932e-05,
|
|
"loss": 4.3528,
|
|
"mean_token_accuracy": 0.2581952720880508,
|
|
"num_tokens": 19640265.0,
|
|
"step": 11175
|
|
},
|
|
{
|
|
"entropy": 5.287809896469116,
|
|
"epoch": 8.761755485893417,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.955844024923308e-05,
|
|
"loss": 4.3952,
|
|
"mean_token_accuracy": 0.25175899267196655,
|
|
"num_tokens": 19648447.0,
|
|
"step": 11180
|
|
},
|
|
{
|
|
"entropy": 5.306576681137085,
|
|
"epoch": 8.765673981191222,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.943596294453492e-05,
|
|
"loss": 4.4133,
|
|
"mean_token_accuracy": 0.25039332658052443,
|
|
"num_tokens": 19657971.0,
|
|
"step": 11185
|
|
},
|
|
{
|
|
"entropy": 5.3105730533599855,
|
|
"epoch": 8.769592476489029,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.931385301610161e-05,
|
|
"loss": 4.3611,
|
|
"mean_token_accuracy": 0.25908557176589964,
|
|
"num_tokens": 19667059.0,
|
|
"step": 11190
|
|
},
|
|
{
|
|
"entropy": 5.270491981506348,
|
|
"epoch": 8.773510971786834,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.919211068216337e-05,
|
|
"loss": 4.3848,
|
|
"mean_token_accuracy": 0.2555903419852257,
|
|
"num_tokens": 19675683.0,
|
|
"step": 11195
|
|
},
|
|
{
|
|
"entropy": 5.318519163131714,
|
|
"epoch": 8.77742946708464,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.907073616029356e-05,
|
|
"loss": 4.4038,
|
|
"mean_token_accuracy": 0.24763167053461074,
|
|
"num_tokens": 19685317.0,
|
|
"step": 11200
|
|
},
|
|
{
|
|
"entropy": 5.343615770339966,
|
|
"epoch": 8.781347962382446,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.894972966740809e-05,
|
|
"loss": 4.4208,
|
|
"mean_token_accuracy": 0.2494013711810112,
|
|
"num_tokens": 19694013.0,
|
|
"step": 11205
|
|
},
|
|
{
|
|
"entropy": 5.233252334594726,
|
|
"epoch": 8.785266457680251,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.882909141976514e-05,
|
|
"loss": 4.3262,
|
|
"mean_token_accuracy": 0.2580722212791443,
|
|
"num_tokens": 19702527.0,
|
|
"step": 11210
|
|
},
|
|
{
|
|
"entropy": 5.313626289367676,
|
|
"epoch": 8.789184952978056,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.87088216329649e-05,
|
|
"loss": 4.3856,
|
|
"mean_token_accuracy": 0.25374105721712115,
|
|
"num_tokens": 19711075.0,
|
|
"step": 11215
|
|
},
|
|
{
|
|
"entropy": 5.275954627990723,
|
|
"epoch": 8.793103448275861,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.858892052194894e-05,
|
|
"loss": 4.3434,
|
|
"mean_token_accuracy": 0.25265001505613327,
|
|
"num_tokens": 19719964.0,
|
|
"step": 11220
|
|
},
|
|
{
|
|
"entropy": 5.235467910766602,
|
|
"epoch": 8.797021943573668,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.846938830099995e-05,
|
|
"loss": 4.3127,
|
|
"mean_token_accuracy": 0.25689061135053637,
|
|
"num_tokens": 19728801.0,
|
|
"step": 11225
|
|
},
|
|
{
|
|
"entropy": 5.287860536575318,
|
|
"epoch": 8.800940438871473,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.835022518374147e-05,
|
|
"loss": 4.404,
|
|
"mean_token_accuracy": 0.25380491763353347,
|
|
"num_tokens": 19737446.0,
|
|
"step": 11230
|
|
},
|
|
{
|
|
"entropy": 5.291885852813721,
|
|
"epoch": 8.804858934169278,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.823143138313725e-05,
|
|
"loss": 4.4085,
|
|
"mean_token_accuracy": 0.2553762584924698,
|
|
"num_tokens": 19745955.0,
|
|
"step": 11235
|
|
},
|
|
{
|
|
"entropy": 5.265475273132324,
|
|
"epoch": 8.808777429467085,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.811300711149109e-05,
|
|
"loss": 4.3399,
|
|
"mean_token_accuracy": 0.2621100589632988,
|
|
"num_tokens": 19754463.0,
|
|
"step": 11240
|
|
},
|
|
{
|
|
"entropy": 5.257623100280762,
|
|
"epoch": 8.81269592476489,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.799495258044633e-05,
|
|
"loss": 4.3728,
|
|
"mean_token_accuracy": 0.25712036788463594,
|
|
"num_tokens": 19762828.0,
|
|
"step": 11245
|
|
},
|
|
{
|
|
"entropy": 5.322431039810181,
|
|
"epoch": 8.816614420062695,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 6.787726800098568e-05,
|
|
"loss": 4.4464,
|
|
"mean_token_accuracy": 0.24742001295089722,
|
|
"num_tokens": 19771159.0,
|
|
"step": 11250
|
|
},
|
|
{
|
|
"entropy": 5.243334245681763,
|
|
"epoch": 8.820532915360502,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.775995358343043e-05,
|
|
"loss": 4.3511,
|
|
"mean_token_accuracy": 0.254022353887558,
|
|
"num_tokens": 19779771.0,
|
|
"step": 11255
|
|
},
|
|
{
|
|
"entropy": 5.286165475845337,
|
|
"epoch": 8.824451410658307,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.764300953744052e-05,
|
|
"loss": 4.3679,
|
|
"mean_token_accuracy": 0.25959489196538926,
|
|
"num_tokens": 19788552.0,
|
|
"step": 11260
|
|
},
|
|
{
|
|
"entropy": 5.264570093154907,
|
|
"epoch": 8.828369905956112,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.752643607201394e-05,
|
|
"loss": 4.4117,
|
|
"mean_token_accuracy": 0.2547892928123474,
|
|
"num_tokens": 19797346.0,
|
|
"step": 11265
|
|
},
|
|
{
|
|
"entropy": 5.340105772018433,
|
|
"epoch": 8.83228840125392,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.741023339548633e-05,
|
|
"loss": 4.4187,
|
|
"mean_token_accuracy": 0.2460606053471565,
|
|
"num_tokens": 19806152.0,
|
|
"step": 11270
|
|
},
|
|
{
|
|
"entropy": 5.285915327072144,
|
|
"epoch": 8.836206896551724,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.729440171553078e-05,
|
|
"loss": 4.36,
|
|
"mean_token_accuracy": 0.2590361192822456,
|
|
"num_tokens": 19814977.0,
|
|
"step": 11275
|
|
},
|
|
{
|
|
"entropy": 5.340801239013672,
|
|
"epoch": 8.84012539184953,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.71789412391572e-05,
|
|
"loss": 4.4576,
|
|
"mean_token_accuracy": 0.24709286838769912,
|
|
"num_tokens": 19824203.0,
|
|
"step": 11280
|
|
},
|
|
{
|
|
"entropy": 5.317534589767456,
|
|
"epoch": 8.844043887147336,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 6.706385217271223e-05,
|
|
"loss": 4.3996,
|
|
"mean_token_accuracy": 0.24965265095233918,
|
|
"num_tokens": 19831974.0,
|
|
"step": 11285
|
|
},
|
|
{
|
|
"entropy": 5.262842798233033,
|
|
"epoch": 8.847962382445141,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.69491347218787e-05,
|
|
"loss": 4.3817,
|
|
"mean_token_accuracy": 0.2455986499786377,
|
|
"num_tokens": 19840709.0,
|
|
"step": 11290
|
|
},
|
|
{
|
|
"entropy": 5.259273338317871,
|
|
"epoch": 8.851880877742946,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.683478909167525e-05,
|
|
"loss": 4.3866,
|
|
"mean_token_accuracy": 0.25065196603536605,
|
|
"num_tokens": 19849409.0,
|
|
"step": 11295
|
|
},
|
|
{
|
|
"entropy": 5.278149223327636,
|
|
"epoch": 8.855799373040753,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.672081548645608e-05,
|
|
"loss": 4.3557,
|
|
"mean_token_accuracy": 0.25910684168338777,
|
|
"num_tokens": 19858331.0,
|
|
"step": 11300
|
|
},
|
|
{
|
|
"entropy": 5.220514965057373,
|
|
"epoch": 8.859717868338558,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.660721410991057e-05,
|
|
"loss": 4.3084,
|
|
"mean_token_accuracy": 0.2578583210706711,
|
|
"num_tokens": 19866469.0,
|
|
"step": 11305
|
|
},
|
|
{
|
|
"entropy": 5.3033795833587645,
|
|
"epoch": 8.863636363636363,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 6.649398516506272e-05,
|
|
"loss": 4.3749,
|
|
"mean_token_accuracy": 0.26280871480703355,
|
|
"num_tokens": 19875183.0,
|
|
"step": 11310
|
|
},
|
|
{
|
|
"entropy": 5.359947919845581,
|
|
"epoch": 8.86755485893417,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.638112885427103e-05,
|
|
"loss": 4.4554,
|
|
"mean_token_accuracy": 0.24318948090076448,
|
|
"num_tokens": 19884440.0,
|
|
"step": 11315
|
|
},
|
|
{
|
|
"entropy": 5.329014873504638,
|
|
"epoch": 8.871473354231975,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.626864537922808e-05,
|
|
"loss": 4.384,
|
|
"mean_token_accuracy": 0.25463264584541323,
|
|
"num_tokens": 19892833.0,
|
|
"step": 11320
|
|
},
|
|
{
|
|
"entropy": 5.320794725418091,
|
|
"epoch": 8.87539184952978,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.615653494096001e-05,
|
|
"loss": 4.4337,
|
|
"mean_token_accuracy": 0.2528904065489769,
|
|
"num_tokens": 19900805.0,
|
|
"step": 11325
|
|
},
|
|
{
|
|
"entropy": 5.33713927268982,
|
|
"epoch": 8.879310344827585,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.604479773982634e-05,
|
|
"loss": 4.3904,
|
|
"mean_token_accuracy": 0.25384185910224916,
|
|
"num_tokens": 19909693.0,
|
|
"step": 11330
|
|
},
|
|
{
|
|
"entropy": 5.3118070602417,
|
|
"epoch": 8.883228840125392,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.593343397551964e-05,
|
|
"loss": 4.4093,
|
|
"mean_token_accuracy": 0.2566418319940567,
|
|
"num_tokens": 19918234.0,
|
|
"step": 11335
|
|
},
|
|
{
|
|
"entropy": 5.26358208656311,
|
|
"epoch": 8.887147335423197,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.582244384706497e-05,
|
|
"loss": 4.3799,
|
|
"mean_token_accuracy": 0.2518415987491608,
|
|
"num_tokens": 19927465.0,
|
|
"step": 11340
|
|
},
|
|
{
|
|
"entropy": 5.236066007614136,
|
|
"epoch": 8.891065830721002,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.571182755281964e-05,
|
|
"loss": 4.3524,
|
|
"mean_token_accuracy": 0.26199161559343337,
|
|
"num_tokens": 19936444.0,
|
|
"step": 11345
|
|
},
|
|
{
|
|
"entropy": 5.22286696434021,
|
|
"epoch": 8.89498432601881,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.5601585290473e-05,
|
|
"loss": 4.3034,
|
|
"mean_token_accuracy": 0.26258273124694825,
|
|
"num_tokens": 19944981.0,
|
|
"step": 11350
|
|
},
|
|
{
|
|
"entropy": 5.324533271789551,
|
|
"epoch": 8.898902821316614,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.549171725704575e-05,
|
|
"loss": 4.4891,
|
|
"mean_token_accuracy": 0.24138792902231215,
|
|
"num_tokens": 19953907.0,
|
|
"step": 11355
|
|
},
|
|
{
|
|
"entropy": 5.253124856948853,
|
|
"epoch": 8.90282131661442,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.538222364889e-05,
|
|
"loss": 4.3413,
|
|
"mean_token_accuracy": 0.2555190876126289,
|
|
"num_tokens": 19962487.0,
|
|
"step": 11360
|
|
},
|
|
{
|
|
"entropy": 5.316452217102051,
|
|
"epoch": 8.906739811912226,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.527310466168842e-05,
|
|
"loss": 4.4218,
|
|
"mean_token_accuracy": 0.2484181597828865,
|
|
"num_tokens": 19971243.0,
|
|
"step": 11365
|
|
},
|
|
{
|
|
"entropy": 5.264421463012695,
|
|
"epoch": 8.910658307210031,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.516436049045452e-05,
|
|
"loss": 4.4363,
|
|
"mean_token_accuracy": 0.24447833597660065,
|
|
"num_tokens": 19980281.0,
|
|
"step": 11370
|
|
},
|
|
{
|
|
"entropy": 5.328791856765747,
|
|
"epoch": 8.914576802507836,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.505599132953165e-05,
|
|
"loss": 4.493,
|
|
"mean_token_accuracy": 0.2358144298195839,
|
|
"num_tokens": 19990340.0,
|
|
"step": 11375
|
|
},
|
|
{
|
|
"entropy": 5.210408401489258,
|
|
"epoch": 8.918495297805643,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 6.494799737259304e-05,
|
|
"loss": 4.2881,
|
|
"mean_token_accuracy": 0.25084335058927537,
|
|
"num_tokens": 19998770.0,
|
|
"step": 11380
|
|
},
|
|
{
|
|
"entropy": 5.225416994094848,
|
|
"epoch": 8.922413793103448,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.48403788126415e-05,
|
|
"loss": 4.2964,
|
|
"mean_token_accuracy": 0.25821626782417295,
|
|
"num_tokens": 20007388.0,
|
|
"step": 11385
|
|
},
|
|
{
|
|
"entropy": 5.233916807174682,
|
|
"epoch": 8.926332288401253,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.473313584200882e-05,
|
|
"loss": 4.2916,
|
|
"mean_token_accuracy": 0.26454881578683853,
|
|
"num_tokens": 20016350.0,
|
|
"step": 11390
|
|
},
|
|
{
|
|
"entropy": 5.2225135326385494,
|
|
"epoch": 8.93025078369906,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.462626865235549e-05,
|
|
"loss": 4.3216,
|
|
"mean_token_accuracy": 0.2560986116528511,
|
|
"num_tokens": 20025777.0,
|
|
"step": 11395
|
|
},
|
|
{
|
|
"entropy": 5.275915622711182,
|
|
"epoch": 8.934169278996865,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.45197774346706e-05,
|
|
"loss": 4.3243,
|
|
"mean_token_accuracy": 0.25879279524087906,
|
|
"num_tokens": 20034396.0,
|
|
"step": 11400
|
|
},
|
|
{
|
|
"entropy": 5.275224685668945,
|
|
"epoch": 8.93808777429467,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.441366237927123e-05,
|
|
"loss": 4.3411,
|
|
"mean_token_accuracy": 0.2569084197282791,
|
|
"num_tokens": 20043190.0,
|
|
"step": 11405
|
|
},
|
|
{
|
|
"entropy": 5.240283012390137,
|
|
"epoch": 8.942006269592476,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.430792367580215e-05,
|
|
"loss": 4.3651,
|
|
"mean_token_accuracy": 0.24973613768815994,
|
|
"num_tokens": 20052108.0,
|
|
"step": 11410
|
|
},
|
|
{
|
|
"entropy": 5.256770849227905,
|
|
"epoch": 8.945924764890282,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.420256151323553e-05,
|
|
"loss": 4.2861,
|
|
"mean_token_accuracy": 0.25861853957176206,
|
|
"num_tokens": 20060540.0,
|
|
"step": 11415
|
|
},
|
|
{
|
|
"entropy": 5.179847955703735,
|
|
"epoch": 8.949843260188088,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 6.409757607987066e-05,
|
|
"loss": 4.2949,
|
|
"mean_token_accuracy": 0.25505208373069765,
|
|
"num_tokens": 20069289.0,
|
|
"step": 11420
|
|
},
|
|
{
|
|
"entropy": 5.308667993545532,
|
|
"epoch": 8.953761755485893,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.399296756333362e-05,
|
|
"loss": 4.3827,
|
|
"mean_token_accuracy": 0.25513807535171507,
|
|
"num_tokens": 20078064.0,
|
|
"step": 11425
|
|
},
|
|
{
|
|
"entropy": 5.293441343307495,
|
|
"epoch": 8.9576802507837,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.388873615057665e-05,
|
|
"loss": 4.3291,
|
|
"mean_token_accuracy": 0.2590105801820755,
|
|
"num_tokens": 20086234.0,
|
|
"step": 11430
|
|
},
|
|
{
|
|
"entropy": 5.244597053527832,
|
|
"epoch": 8.961598746081505,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.378488202787835e-05,
|
|
"loss": 4.3432,
|
|
"mean_token_accuracy": 0.24999353885650635,
|
|
"num_tokens": 20094700.0,
|
|
"step": 11435
|
|
},
|
|
{
|
|
"entropy": 5.263831663131714,
|
|
"epoch": 8.96551724137931,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.368140538084275e-05,
|
|
"loss": 4.4106,
|
|
"mean_token_accuracy": 0.2552211835980415,
|
|
"num_tokens": 20103669.0,
|
|
"step": 11440
|
|
},
|
|
{
|
|
"entropy": 5.287847375869751,
|
|
"epoch": 8.969435736677116,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.357830639439943e-05,
|
|
"loss": 4.3345,
|
|
"mean_token_accuracy": 0.2643582344055176,
|
|
"num_tokens": 20111840.0,
|
|
"step": 11445
|
|
},
|
|
{
|
|
"entropy": 5.255885314941406,
|
|
"epoch": 8.973354231974922,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.347558525280295e-05,
|
|
"loss": 4.3383,
|
|
"mean_token_accuracy": 0.2580055475234985,
|
|
"num_tokens": 20119590.0,
|
|
"step": 11450
|
|
},
|
|
{
|
|
"entropy": 5.2409144878387455,
|
|
"epoch": 8.977272727272727,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.337324213963274e-05,
|
|
"loss": 4.303,
|
|
"mean_token_accuracy": 0.257168111205101,
|
|
"num_tokens": 20128982.0,
|
|
"step": 11455
|
|
},
|
|
{
|
|
"entropy": 5.274861812591553,
|
|
"epoch": 8.981191222570533,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.327127723779248e-05,
|
|
"loss": 4.4685,
|
|
"mean_token_accuracy": 0.2441285952925682,
|
|
"num_tokens": 20138013.0,
|
|
"step": 11460
|
|
},
|
|
{
|
|
"entropy": 5.2557761669158936,
|
|
"epoch": 8.985109717868339,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.316969072950998e-05,
|
|
"loss": 4.35,
|
|
"mean_token_accuracy": 0.25364584773778914,
|
|
"num_tokens": 20146240.0,
|
|
"step": 11465
|
|
},
|
|
{
|
|
"entropy": 5.3469115734100345,
|
|
"epoch": 8.989028213166144,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.306848279633684e-05,
|
|
"loss": 4.474,
|
|
"mean_token_accuracy": 0.2415533408522606,
|
|
"num_tokens": 20155964.0,
|
|
"step": 11470
|
|
},
|
|
{
|
|
"entropy": 5.276971673965454,
|
|
"epoch": 8.99294670846395,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.296765361914798e-05,
|
|
"loss": 4.4168,
|
|
"mean_token_accuracy": 0.2511414512991905,
|
|
"num_tokens": 20165774.0,
|
|
"step": 11475
|
|
},
|
|
{
|
|
"entropy": 5.273919010162354,
|
|
"epoch": 8.996865203761756,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 6.286720337814158e-05,
|
|
"loss": 4.379,
|
|
"mean_token_accuracy": 0.2455829858779907,
|
|
"num_tokens": 20174940.0,
|
|
"step": 11480
|
|
},
|
|
{
|
|
"entropy": 5.2971710681915285,
|
|
"epoch": 9.00078369905956,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.276713225283839e-05,
|
|
"loss": 4.3131,
|
|
"mean_token_accuracy": 0.2647103577852249,
|
|
"num_tokens": 20182706.0,
|
|
"step": 11485
|
|
},
|
|
{
|
|
"entropy": 5.296993970870972,
|
|
"epoch": 9.004702194357368,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.266744042208186e-05,
|
|
"loss": 4.3478,
|
|
"mean_token_accuracy": 0.2566777944564819,
|
|
"num_tokens": 20191457.0,
|
|
"step": 11490
|
|
},
|
|
{
|
|
"entropy": 5.254575920104981,
|
|
"epoch": 9.008620689655173,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.25681280640374e-05,
|
|
"loss": 4.3985,
|
|
"mean_token_accuracy": 0.2412103995680809,
|
|
"num_tokens": 20199331.0,
|
|
"step": 11495
|
|
},
|
|
{
|
|
"entropy": 5.346774387359619,
|
|
"epoch": 9.012539184952978,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.246919535619226e-05,
|
|
"loss": 4.3621,
|
|
"mean_token_accuracy": 0.25093347430229185,
|
|
"num_tokens": 20208386.0,
|
|
"step": 11500
|
|
},
|
|
{
|
|
"epoch": 9.012539184952978,
|
|
"eval_entropy": 5.122093075005583,
|
|
"eval_loss": 5.626368999481201,
|
|
"eval_mean_token_accuracy": 0.17644334288472815,
|
|
"eval_num_tokens": 20208386.0,
|
|
"eval_runtime": 2.1765,
|
|
"eval_samples_per_second": 1893.856,
|
|
"eval_steps_per_second": 237.077,
|
|
"step": 11500
|
|
},
|
|
{
|
|
"entropy": 5.277209663391114,
|
|
"epoch": 9.016457680250785,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.237064247535531e-05,
|
|
"loss": 4.3628,
|
|
"mean_token_accuracy": 0.25131346434354784,
|
|
"num_tokens": 20217649.0,
|
|
"step": 11505
|
|
},
|
|
{
|
|
"entropy": 5.236365890502929,
|
|
"epoch": 9.02037617554859,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.227246959765644e-05,
|
|
"loss": 4.2872,
|
|
"mean_token_accuracy": 0.26100704818964005,
|
|
"num_tokens": 20226219.0,
|
|
"step": 11510
|
|
},
|
|
{
|
|
"entropy": 5.2881745338439945,
|
|
"epoch": 9.024294670846395,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.217467689854654e-05,
|
|
"loss": 4.3377,
|
|
"mean_token_accuracy": 0.258573517203331,
|
|
"num_tokens": 20234788.0,
|
|
"step": 11515
|
|
},
|
|
{
|
|
"entropy": 5.2665056705474855,
|
|
"epoch": 9.0282131661442,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.207726455279705e-05,
|
|
"loss": 4.281,
|
|
"mean_token_accuracy": 0.264527802169323,
|
|
"num_tokens": 20242858.0,
|
|
"step": 11520
|
|
},
|
|
{
|
|
"entropy": 5.229965162277222,
|
|
"epoch": 9.032131661442007,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.198023273449963e-05,
|
|
"loss": 4.3051,
|
|
"mean_token_accuracy": 0.25862052142620084,
|
|
"num_tokens": 20252041.0,
|
|
"step": 11525
|
|
},
|
|
{
|
|
"entropy": 5.27967095375061,
|
|
"epoch": 9.036050156739812,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.188358161706578e-05,
|
|
"loss": 4.3877,
|
|
"mean_token_accuracy": 0.25355862379074096,
|
|
"num_tokens": 20261035.0,
|
|
"step": 11530
|
|
},
|
|
{
|
|
"entropy": 5.3370133399963375,
|
|
"epoch": 9.039968652037617,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.178731137322681e-05,
|
|
"loss": 4.4363,
|
|
"mean_token_accuracy": 0.24429677724838256,
|
|
"num_tokens": 20270091.0,
|
|
"step": 11535
|
|
},
|
|
{
|
|
"entropy": 5.27291054725647,
|
|
"epoch": 9.043887147335424,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.169142217503319e-05,
|
|
"loss": 4.3967,
|
|
"mean_token_accuracy": 0.2503023728728294,
|
|
"num_tokens": 20278890.0,
|
|
"step": 11540
|
|
},
|
|
{
|
|
"entropy": 5.289458608627319,
|
|
"epoch": 9.047805642633229,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.159591419385455e-05,
|
|
"loss": 4.4468,
|
|
"mean_token_accuracy": 0.23924320787191392,
|
|
"num_tokens": 20287692.0,
|
|
"step": 11545
|
|
},
|
|
{
|
|
"entropy": 5.307593441009521,
|
|
"epoch": 9.051724137931034,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.150078760037904e-05,
|
|
"loss": 4.3237,
|
|
"mean_token_accuracy": 0.2654447048902512,
|
|
"num_tokens": 20296357.0,
|
|
"step": 11550
|
|
},
|
|
{
|
|
"entropy": 5.272796535491944,
|
|
"epoch": 9.05564263322884,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 6.140604256461339e-05,
|
|
"loss": 4.3649,
|
|
"mean_token_accuracy": 0.25108554810285566,
|
|
"num_tokens": 20304894.0,
|
|
"step": 11555
|
|
},
|
|
{
|
|
"entropy": 5.247407054901123,
|
|
"epoch": 9.059561128526646,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.131167925588233e-05,
|
|
"loss": 4.3793,
|
|
"mean_token_accuracy": 0.2622376725077629,
|
|
"num_tokens": 20314011.0,
|
|
"step": 11560
|
|
},
|
|
{
|
|
"entropy": 5.214827919006348,
|
|
"epoch": 9.06347962382445,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.121769784282833e-05,
|
|
"loss": 4.2402,
|
|
"mean_token_accuracy": 0.2613084942102432,
|
|
"num_tokens": 20322738.0,
|
|
"step": 11565
|
|
},
|
|
{
|
|
"entropy": 5.259373712539673,
|
|
"epoch": 9.067398119122258,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 6.112409849341147e-05,
|
|
"loss": 4.2902,
|
|
"mean_token_accuracy": 0.25881148129701614,
|
|
"num_tokens": 20331528.0,
|
|
"step": 11570
|
|
},
|
|
{
|
|
"entropy": 5.251845121383667,
|
|
"epoch": 9.071316614420063,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.103088137490895e-05,
|
|
"loss": 4.4047,
|
|
"mean_token_accuracy": 0.2526107668876648,
|
|
"num_tokens": 20340411.0,
|
|
"step": 11575
|
|
},
|
|
{
|
|
"entropy": 5.342277193069458,
|
|
"epoch": 9.075235109717868,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.093804665391488e-05,
|
|
"loss": 4.4171,
|
|
"mean_token_accuracy": 0.25722316950559615,
|
|
"num_tokens": 20349266.0,
|
|
"step": 11580
|
|
},
|
|
{
|
|
"entropy": 5.319170713424683,
|
|
"epoch": 9.079153605015675,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.084559449633992e-05,
|
|
"loss": 4.4106,
|
|
"mean_token_accuracy": 0.252149672806263,
|
|
"num_tokens": 20357849.0,
|
|
"step": 11585
|
|
},
|
|
{
|
|
"entropy": 5.2772050380706785,
|
|
"epoch": 9.08307210031348,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.0753525067411106e-05,
|
|
"loss": 4.3442,
|
|
"mean_token_accuracy": 0.2567505031824112,
|
|
"num_tokens": 20366795.0,
|
|
"step": 11590
|
|
},
|
|
{
|
|
"entropy": 5.215304183959961,
|
|
"epoch": 9.086990595611285,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.06618385316714e-05,
|
|
"loss": 4.2938,
|
|
"mean_token_accuracy": 0.26267911940813066,
|
|
"num_tokens": 20375086.0,
|
|
"step": 11595
|
|
},
|
|
{
|
|
"entropy": 5.23910493850708,
|
|
"epoch": 9.090909090909092,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.0570535052979554e-05,
|
|
"loss": 4.326,
|
|
"mean_token_accuracy": 0.25658717304468154,
|
|
"num_tokens": 20384071.0,
|
|
"step": 11600
|
|
},
|
|
{
|
|
"entropy": 5.299495697021484,
|
|
"epoch": 9.094827586206897,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.047961479450968e-05,
|
|
"loss": 4.3541,
|
|
"mean_token_accuracy": 0.25510676503181456,
|
|
"num_tokens": 20393345.0,
|
|
"step": 11605
|
|
},
|
|
{
|
|
"entropy": 5.259856367111206,
|
|
"epoch": 9.098746081504702,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.038907791875102e-05,
|
|
"loss": 4.3451,
|
|
"mean_token_accuracy": 0.2522570788860321,
|
|
"num_tokens": 20402034.0,
|
|
"step": 11610
|
|
},
|
|
{
|
|
"entropy": 5.290747213363647,
|
|
"epoch": 9.102664576802507,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.0298924587507655e-05,
|
|
"loss": 4.3658,
|
|
"mean_token_accuracy": 0.24741135835647582,
|
|
"num_tokens": 20411681.0,
|
|
"step": 11615
|
|
},
|
|
{
|
|
"entropy": 5.267978239059448,
|
|
"epoch": 9.106583072100314,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.020915496189815e-05,
|
|
"loss": 4.3831,
|
|
"mean_token_accuracy": 0.24931320250034333,
|
|
"num_tokens": 20420084.0,
|
|
"step": 11620
|
|
},
|
|
{
|
|
"entropy": 5.280542087554932,
|
|
"epoch": 9.110501567398119,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.0119769202355455e-05,
|
|
"loss": 4.4313,
|
|
"mean_token_accuracy": 0.24155025482177733,
|
|
"num_tokens": 20429326.0,
|
|
"step": 11625
|
|
},
|
|
{
|
|
"entropy": 5.361446714401245,
|
|
"epoch": 9.114420062695924,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.003076746862637e-05,
|
|
"loss": 4.4536,
|
|
"mean_token_accuracy": 0.24954892843961715,
|
|
"num_tokens": 20438522.0,
|
|
"step": 11630
|
|
},
|
|
{
|
|
"entropy": 5.328919410705566,
|
|
"epoch": 9.11833855799373,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.994214991977143e-05,
|
|
"loss": 4.4267,
|
|
"mean_token_accuracy": 0.2528712138533592,
|
|
"num_tokens": 20447084.0,
|
|
"step": 11635
|
|
},
|
|
{
|
|
"entropy": 5.248061084747315,
|
|
"epoch": 9.122257053291536,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.9853916714164584e-05,
|
|
"loss": 4.2847,
|
|
"mean_token_accuracy": 0.26335230767726897,
|
|
"num_tokens": 20455553.0,
|
|
"step": 11640
|
|
},
|
|
{
|
|
"entropy": 5.281199264526367,
|
|
"epoch": 9.126175548589341,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.976606800949287e-05,
|
|
"loss": 4.3025,
|
|
"mean_token_accuracy": 0.26470163464546204,
|
|
"num_tokens": 20464066.0,
|
|
"step": 11645
|
|
},
|
|
{
|
|
"entropy": 5.264858675003052,
|
|
"epoch": 9.130094043887148,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.9678603962756137e-05,
|
|
"loss": 4.3424,
|
|
"mean_token_accuracy": 0.24944580495357513,
|
|
"num_tokens": 20472706.0,
|
|
"step": 11650
|
|
},
|
|
{
|
|
"entropy": 5.280135440826416,
|
|
"epoch": 9.134012539184953,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.959152473026687e-05,
|
|
"loss": 4.3952,
|
|
"mean_token_accuracy": 0.24861884117126465,
|
|
"num_tokens": 20481666.0,
|
|
"step": 11655
|
|
},
|
|
{
|
|
"entropy": 5.328822469711303,
|
|
"epoch": 9.137931034482758,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.950483046764971e-05,
|
|
"loss": 4.3945,
|
|
"mean_token_accuracy": 0.2510017827153206,
|
|
"num_tokens": 20490157.0,
|
|
"step": 11660
|
|
},
|
|
{
|
|
"entropy": 5.302386045455933,
|
|
"epoch": 9.141849529780565,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.941852132984145e-05,
|
|
"loss": 4.3303,
|
|
"mean_token_accuracy": 0.26086479872465135,
|
|
"num_tokens": 20499180.0,
|
|
"step": 11665
|
|
},
|
|
{
|
|
"entropy": 5.302295351028443,
|
|
"epoch": 9.14576802507837,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.933259747109042e-05,
|
|
"loss": 4.3612,
|
|
"mean_token_accuracy": 0.2591048553586006,
|
|
"num_tokens": 20508291.0,
|
|
"step": 11670
|
|
},
|
|
{
|
|
"entropy": 5.241799402236938,
|
|
"epoch": 9.149686520376175,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.924705904495656e-05,
|
|
"loss": 4.3048,
|
|
"mean_token_accuracy": 0.2596322953701019,
|
|
"num_tokens": 20516500.0,
|
|
"step": 11675
|
|
},
|
|
{
|
|
"entropy": 5.240731859207154,
|
|
"epoch": 9.153605015673982,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.916190620431088e-05,
|
|
"loss": 4.2967,
|
|
"mean_token_accuracy": 0.25969695299863815,
|
|
"num_tokens": 20525408.0,
|
|
"step": 11680
|
|
},
|
|
{
|
|
"entropy": 5.2481859683990475,
|
|
"epoch": 9.157523510971787,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.907713910133529e-05,
|
|
"loss": 4.3122,
|
|
"mean_token_accuracy": 0.2638166084885597,
|
|
"num_tokens": 20534288.0,
|
|
"step": 11685
|
|
},
|
|
{
|
|
"entropy": 5.262470006942749,
|
|
"epoch": 9.161442006269592,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 5.8992757887522394e-05,
|
|
"loss": 4.3516,
|
|
"mean_token_accuracy": 0.2549307614564896,
|
|
"num_tokens": 20543011.0,
|
|
"step": 11690
|
|
},
|
|
{
|
|
"entropy": 5.326808929443359,
|
|
"epoch": 9.165360501567399,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.8908762713675074e-05,
|
|
"loss": 4.3794,
|
|
"mean_token_accuracy": 0.25671765506267546,
|
|
"num_tokens": 20551511.0,
|
|
"step": 11695
|
|
},
|
|
{
|
|
"entropy": 5.201155662536621,
|
|
"epoch": 9.169278996865204,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.882515372990632e-05,
|
|
"loss": 4.3136,
|
|
"mean_token_accuracy": 0.2588320016860962,
|
|
"num_tokens": 20560578.0,
|
|
"step": 11700
|
|
},
|
|
{
|
|
"entropy": 5.294696712493897,
|
|
"epoch": 9.173197492163009,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.874193108563895e-05,
|
|
"loss": 4.366,
|
|
"mean_token_accuracy": 0.2553620502352715,
|
|
"num_tokens": 20569776.0,
|
|
"step": 11705
|
|
},
|
|
{
|
|
"entropy": 5.339007377624512,
|
|
"epoch": 9.177115987460816,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.865909492960536e-05,
|
|
"loss": 4.4559,
|
|
"mean_token_accuracy": 0.247935052216053,
|
|
"num_tokens": 20579791.0,
|
|
"step": 11710
|
|
},
|
|
{
|
|
"entropy": 5.3115931987762455,
|
|
"epoch": 9.181034482758621,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.85766454098471e-05,
|
|
"loss": 4.3408,
|
|
"mean_token_accuracy": 0.25436079651117327,
|
|
"num_tokens": 20588738.0,
|
|
"step": 11715
|
|
},
|
|
{
|
|
"entropy": 5.299120616912842,
|
|
"epoch": 9.184952978056426,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.849458267371489e-05,
|
|
"loss": 4.3305,
|
|
"mean_token_accuracy": 0.26905384510755537,
|
|
"num_tokens": 20597442.0,
|
|
"step": 11720
|
|
},
|
|
{
|
|
"entropy": 5.344450664520264,
|
|
"epoch": 9.188871473354231,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.841290686786818e-05,
|
|
"loss": 4.4667,
|
|
"mean_token_accuracy": 0.23918794542551042,
|
|
"num_tokens": 20607277.0,
|
|
"step": 11725
|
|
},
|
|
{
|
|
"entropy": 5.3287109375,
|
|
"epoch": 9.192789968652038,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.833161813827484e-05,
|
|
"loss": 4.4581,
|
|
"mean_token_accuracy": 0.25214813351631166,
|
|
"num_tokens": 20615721.0,
|
|
"step": 11730
|
|
},
|
|
{
|
|
"entropy": 5.286286306381226,
|
|
"epoch": 9.196708463949843,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.825071663021097e-05,
|
|
"loss": 4.3702,
|
|
"mean_token_accuracy": 0.2541424572467804,
|
|
"num_tokens": 20624192.0,
|
|
"step": 11735
|
|
},
|
|
{
|
|
"entropy": 5.265912389755249,
|
|
"epoch": 9.200626959247648,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.817020248826076e-05,
|
|
"loss": 4.3589,
|
|
"mean_token_accuracy": 0.2587177261710167,
|
|
"num_tokens": 20632947.0,
|
|
"step": 11740
|
|
},
|
|
{
|
|
"entropy": 5.293414306640625,
|
|
"epoch": 9.204545454545455,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.8090075856315994e-05,
|
|
"loss": 4.3611,
|
|
"mean_token_accuracy": 0.2516678377985954,
|
|
"num_tokens": 20643252.0,
|
|
"step": 11745
|
|
},
|
|
{
|
|
"entropy": 5.280563735961914,
|
|
"epoch": 9.20846394984326,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.801033687757598e-05,
|
|
"loss": 4.3185,
|
|
"mean_token_accuracy": 0.25318553298711777,
|
|
"num_tokens": 20651416.0,
|
|
"step": 11750
|
|
},
|
|
{
|
|
"entropy": 5.294426107406617,
|
|
"epoch": 9.212382445141065,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.793098569454713e-05,
|
|
"loss": 4.3691,
|
|
"mean_token_accuracy": 0.2520149424672127,
|
|
"num_tokens": 20660359.0,
|
|
"step": 11755
|
|
},
|
|
{
|
|
"entropy": 5.282356262207031,
|
|
"epoch": 9.216300940438872,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.785202244904296e-05,
|
|
"loss": 4.3994,
|
|
"mean_token_accuracy": 0.2517108142375946,
|
|
"num_tokens": 20669262.0,
|
|
"step": 11760
|
|
},
|
|
{
|
|
"entropy": 5.328759813308716,
|
|
"epoch": 9.220219435736677,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.777344728218356e-05,
|
|
"loss": 4.3875,
|
|
"mean_token_accuracy": 0.25125017315149306,
|
|
"num_tokens": 20677267.0,
|
|
"step": 11765
|
|
},
|
|
{
|
|
"entropy": 5.305834722518921,
|
|
"epoch": 9.224137931034482,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.769526033439547e-05,
|
|
"loss": 4.3736,
|
|
"mean_token_accuracy": 0.2610180452466011,
|
|
"num_tokens": 20685798.0,
|
|
"step": 11770
|
|
},
|
|
{
|
|
"entropy": 5.241535949707031,
|
|
"epoch": 9.228056426332289,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.761746174541148e-05,
|
|
"loss": 4.2914,
|
|
"mean_token_accuracy": 0.2621172174811363,
|
|
"num_tokens": 20694384.0,
|
|
"step": 11775
|
|
},
|
|
{
|
|
"entropy": 5.329678630828857,
|
|
"epoch": 9.231974921630094,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.754005165427027e-05,
|
|
"loss": 4.3969,
|
|
"mean_token_accuracy": 0.2565437272191048,
|
|
"num_tokens": 20703233.0,
|
|
"step": 11780
|
|
},
|
|
{
|
|
"entropy": 5.2877202987670895,
|
|
"epoch": 9.2358934169279,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.746303019931623e-05,
|
|
"loss": 4.3463,
|
|
"mean_token_accuracy": 0.2573419541120529,
|
|
"num_tokens": 20711810.0,
|
|
"step": 11785
|
|
},
|
|
{
|
|
"entropy": 5.288181018829346,
|
|
"epoch": 9.239811912225706,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.738639751819919e-05,
|
|
"loss": 4.3601,
|
|
"mean_token_accuracy": 0.2572316452860832,
|
|
"num_tokens": 20720733.0,
|
|
"step": 11790
|
|
},
|
|
{
|
|
"entropy": 5.214310121536255,
|
|
"epoch": 9.243730407523511,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.7310153747874165e-05,
|
|
"loss": 4.3222,
|
|
"mean_token_accuracy": 0.2610944420099258,
|
|
"num_tokens": 20729409.0,
|
|
"step": 11795
|
|
},
|
|
{
|
|
"entropy": 5.3281553268432615,
|
|
"epoch": 9.247648902821316,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.7234299024601136e-05,
|
|
"loss": 4.3918,
|
|
"mean_token_accuracy": 0.2463616207242012,
|
|
"num_tokens": 20738113.0,
|
|
"step": 11800
|
|
},
|
|
{
|
|
"entropy": 5.2849523544311525,
|
|
"epoch": 9.251567398119123,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.7158833483944796e-05,
|
|
"loss": 4.3256,
|
|
"mean_token_accuracy": 0.26599655002355577,
|
|
"num_tokens": 20746756.0,
|
|
"step": 11805
|
|
},
|
|
{
|
|
"entropy": 5.265203332901001,
|
|
"epoch": 9.255485893416928,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.7083757260774305e-05,
|
|
"loss": 4.2879,
|
|
"mean_token_accuracy": 0.2701498746871948,
|
|
"num_tokens": 20756356.0,
|
|
"step": 11810
|
|
},
|
|
{
|
|
"entropy": 5.2923260688781735,
|
|
"epoch": 9.259404388714733,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.7009070489263036e-05,
|
|
"loss": 4.3883,
|
|
"mean_token_accuracy": 0.2525468200445175,
|
|
"num_tokens": 20766118.0,
|
|
"step": 11815
|
|
},
|
|
{
|
|
"entropy": 5.29071888923645,
|
|
"epoch": 9.263322884012538,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 5.6934773302888365e-05,
|
|
"loss": 4.3735,
|
|
"mean_token_accuracy": 0.25291106998920443,
|
|
"num_tokens": 20775049.0,
|
|
"step": 11820
|
|
},
|
|
{
|
|
"entropy": 5.270327472686768,
|
|
"epoch": 9.267241379310345,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.686086583443143e-05,
|
|
"loss": 4.3576,
|
|
"mean_token_accuracy": 0.25450229793787005,
|
|
"num_tokens": 20783839.0,
|
|
"step": 11825
|
|
},
|
|
{
|
|
"entropy": 5.359158706665039,
|
|
"epoch": 9.27115987460815,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.678734821597681e-05,
|
|
"loss": 4.4202,
|
|
"mean_token_accuracy": 0.2509778827428818,
|
|
"num_tokens": 20792036.0,
|
|
"step": 11830
|
|
},
|
|
{
|
|
"entropy": 5.329336881637573,
|
|
"epoch": 9.275078369905955,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.671422057891248e-05,
|
|
"loss": 4.3838,
|
|
"mean_token_accuracy": 0.2599992945790291,
|
|
"num_tokens": 20800639.0,
|
|
"step": 11835
|
|
},
|
|
{
|
|
"entropy": 5.244553899765014,
|
|
"epoch": 9.278996865203762,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.6641483053929324e-05,
|
|
"loss": 4.3281,
|
|
"mean_token_accuracy": 0.25952979773283,
|
|
"num_tokens": 20810244.0,
|
|
"step": 11840
|
|
},
|
|
{
|
|
"entropy": 5.3582456588745115,
|
|
"epoch": 9.282915360501567,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.656913577102112e-05,
|
|
"loss": 4.4021,
|
|
"mean_token_accuracy": 0.25184016227722167,
|
|
"num_tokens": 20818731.0,
|
|
"step": 11845
|
|
},
|
|
{
|
|
"entropy": 5.222071647644043,
|
|
"epoch": 9.286833855799372,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.649717885948422e-05,
|
|
"loss": 4.3055,
|
|
"mean_token_accuracy": 0.26572169065475465,
|
|
"num_tokens": 20827237.0,
|
|
"step": 11850
|
|
},
|
|
{
|
|
"entropy": 5.219003915786743,
|
|
"epoch": 9.29075235109718,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.642561244791721e-05,
|
|
"loss": 4.3209,
|
|
"mean_token_accuracy": 0.25424577444791796,
|
|
"num_tokens": 20835792.0,
|
|
"step": 11855
|
|
},
|
|
{
|
|
"entropy": 5.231267023086548,
|
|
"epoch": 9.294670846394984,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.635443666422097e-05,
|
|
"loss": 4.2777,
|
|
"mean_token_accuracy": 0.2584951758384705,
|
|
"num_tokens": 20844289.0,
|
|
"step": 11860
|
|
},
|
|
{
|
|
"entropy": 5.284473514556884,
|
|
"epoch": 9.29858934169279,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.6283651635598105e-05,
|
|
"loss": 4.3438,
|
|
"mean_token_accuracy": 0.25316428393125534,
|
|
"num_tokens": 20852683.0,
|
|
"step": 11865
|
|
},
|
|
{
|
|
"entropy": 5.262858486175537,
|
|
"epoch": 9.302507836990596,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.6213257488552946e-05,
|
|
"loss": 4.3899,
|
|
"mean_token_accuracy": 0.2585880845785141,
|
|
"num_tokens": 20861999.0,
|
|
"step": 11870
|
|
},
|
|
{
|
|
"entropy": 5.221003532409668,
|
|
"epoch": 9.306426332288401,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.6143254348891254e-05,
|
|
"loss": 4.3176,
|
|
"mean_token_accuracy": 0.25226514488458635,
|
|
"num_tokens": 20870995.0,
|
|
"step": 11875
|
|
},
|
|
{
|
|
"entropy": 5.2609028816223145,
|
|
"epoch": 9.310344827586206,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.6073642341720006e-05,
|
|
"loss": 4.3444,
|
|
"mean_token_accuracy": 0.25627640932798385,
|
|
"num_tokens": 20880048.0,
|
|
"step": 11880
|
|
},
|
|
{
|
|
"entropy": 5.216706895828247,
|
|
"epoch": 9.314263322884013,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.600442159144712e-05,
|
|
"loss": 4.3543,
|
|
"mean_token_accuracy": 0.253665429353714,
|
|
"num_tokens": 20889268.0,
|
|
"step": 11885
|
|
},
|
|
{
|
|
"entropy": 5.248071384429932,
|
|
"epoch": 9.318181818181818,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.593559222178131e-05,
|
|
"loss": 4.3677,
|
|
"mean_token_accuracy": 0.24769512116909026,
|
|
"num_tokens": 20897978.0,
|
|
"step": 11890
|
|
},
|
|
{
|
|
"entropy": 5.286467361450195,
|
|
"epoch": 9.322100313479623,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.5867154355731795e-05,
|
|
"loss": 4.3683,
|
|
"mean_token_accuracy": 0.2647731348872185,
|
|
"num_tokens": 20906387.0,
|
|
"step": 11895
|
|
},
|
|
{
|
|
"entropy": 5.3155886173248295,
|
|
"epoch": 9.32601880877743,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.579910811560822e-05,
|
|
"loss": 4.3555,
|
|
"mean_token_accuracy": 0.25852142870426176,
|
|
"num_tokens": 20915309.0,
|
|
"step": 11900
|
|
},
|
|
{
|
|
"entropy": 5.277198123931885,
|
|
"epoch": 9.329937304075235,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.573145362302012e-05,
|
|
"loss": 4.4161,
|
|
"mean_token_accuracy": 0.24600287526845932,
|
|
"num_tokens": 20924600.0,
|
|
"step": 11905
|
|
},
|
|
{
|
|
"entropy": 5.321261739730835,
|
|
"epoch": 9.33385579937304,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.566419099887718e-05,
|
|
"loss": 4.428,
|
|
"mean_token_accuracy": 0.2523242339491844,
|
|
"num_tokens": 20933087.0,
|
|
"step": 11910
|
|
},
|
|
{
|
|
"entropy": 5.193663644790649,
|
|
"epoch": 9.337774294670847,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.5597320363388544e-05,
|
|
"loss": 4.2625,
|
|
"mean_token_accuracy": 0.2573970273137093,
|
|
"num_tokens": 20940886.0,
|
|
"step": 11915
|
|
},
|
|
{
|
|
"entropy": 5.2823615074157715,
|
|
"epoch": 9.341692789968652,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.55308418360629e-05,
|
|
"loss": 4.3664,
|
|
"mean_token_accuracy": 0.25412498116493226,
|
|
"num_tokens": 20949591.0,
|
|
"step": 11920
|
|
},
|
|
{
|
|
"entropy": 5.284907627105713,
|
|
"epoch": 9.345611285266457,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.54647555357081e-05,
|
|
"loss": 4.4353,
|
|
"mean_token_accuracy": 0.24123743772506714,
|
|
"num_tokens": 20958001.0,
|
|
"step": 11925
|
|
},
|
|
{
|
|
"entropy": 5.302580213546753,
|
|
"epoch": 9.349529780564263,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.539906158043113e-05,
|
|
"loss": 4.3926,
|
|
"mean_token_accuracy": 0.2510508164763451,
|
|
"num_tokens": 20966671.0,
|
|
"step": 11930
|
|
},
|
|
{
|
|
"entropy": 5.337070941925049,
|
|
"epoch": 9.35344827586207,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.533376008763775e-05,
|
|
"loss": 4.3109,
|
|
"mean_token_accuracy": 0.26137754172086713,
|
|
"num_tokens": 20976184.0,
|
|
"step": 11935
|
|
},
|
|
{
|
|
"entropy": 5.337970972061157,
|
|
"epoch": 9.357366771159874,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.526885117403228e-05,
|
|
"loss": 4.3104,
|
|
"mean_token_accuracy": 0.2650835707783699,
|
|
"num_tokens": 20984960.0,
|
|
"step": 11940
|
|
},
|
|
{
|
|
"entropy": 5.303161764144898,
|
|
"epoch": 9.36128526645768,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.520433495561752e-05,
|
|
"loss": 4.349,
|
|
"mean_token_accuracy": 0.25700899213552475,
|
|
"num_tokens": 20993954.0,
|
|
"step": 11945
|
|
},
|
|
{
|
|
"entropy": 5.296792221069336,
|
|
"epoch": 9.365203761755486,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.5140211547694344e-05,
|
|
"loss": 4.3482,
|
|
"mean_token_accuracy": 0.2619215711951256,
|
|
"num_tokens": 21002391.0,
|
|
"step": 11950
|
|
},
|
|
{
|
|
"entropy": 5.213566303253174,
|
|
"epoch": 9.369122257053291,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.507648106486173e-05,
|
|
"loss": 4.2584,
|
|
"mean_token_accuracy": 0.2629462584853172,
|
|
"num_tokens": 21010996.0,
|
|
"step": 11955
|
|
},
|
|
{
|
|
"entropy": 5.237477350234985,
|
|
"epoch": 9.373040752351097,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.5013143621016376e-05,
|
|
"loss": 4.3049,
|
|
"mean_token_accuracy": 0.26302429735660554,
|
|
"num_tokens": 21020272.0,
|
|
"step": 11960
|
|
},
|
|
{
|
|
"entropy": 5.2119465351104735,
|
|
"epoch": 9.376959247648903,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.495019932935256e-05,
|
|
"loss": 4.297,
|
|
"mean_token_accuracy": 0.25385622680187225,
|
|
"num_tokens": 21029206.0,
|
|
"step": 11965
|
|
},
|
|
{
|
|
"entropy": 5.315959453582764,
|
|
"epoch": 9.380877742946709,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.488764830236194e-05,
|
|
"loss": 4.4531,
|
|
"mean_token_accuracy": 0.2447393625974655,
|
|
"num_tokens": 21038294.0,
|
|
"step": 11970
|
|
},
|
|
{
|
|
"entropy": 5.21376051902771,
|
|
"epoch": 9.384796238244514,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.482549065183333e-05,
|
|
"loss": 4.3182,
|
|
"mean_token_accuracy": 0.25981463938951493,
|
|
"num_tokens": 21047321.0,
|
|
"step": 11975
|
|
},
|
|
{
|
|
"entropy": 5.184910726547241,
|
|
"epoch": 9.38871473354232,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.476372648885255e-05,
|
|
"loss": 4.2592,
|
|
"mean_token_accuracy": 0.27410998940467834,
|
|
"num_tokens": 21055928.0,
|
|
"step": 11980
|
|
},
|
|
{
|
|
"entropy": 5.318206453323365,
|
|
"epoch": 9.392633228840126,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.470235592380219e-05,
|
|
"loss": 4.3592,
|
|
"mean_token_accuracy": 0.2570126712322235,
|
|
"num_tokens": 21064464.0,
|
|
"step": 11985
|
|
},
|
|
{
|
|
"entropy": 5.3305487632751465,
|
|
"epoch": 9.39655172413793,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.464137906636133e-05,
|
|
"loss": 4.4056,
|
|
"mean_token_accuracy": 0.25124004781246184,
|
|
"num_tokens": 21073395.0,
|
|
"step": 11990
|
|
},
|
|
{
|
|
"entropy": 5.243026304244995,
|
|
"epoch": 9.400470219435737,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.458079602550557e-05,
|
|
"loss": 4.3203,
|
|
"mean_token_accuracy": 0.2526607125997543,
|
|
"num_tokens": 21081841.0,
|
|
"step": 11995
|
|
},
|
|
{
|
|
"entropy": 5.312049341201782,
|
|
"epoch": 9.404388714733543,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.452060690950662e-05,
|
|
"loss": 4.3829,
|
|
"mean_token_accuracy": 0.24364778101444245,
|
|
"num_tokens": 21090826.0,
|
|
"step": 12000
|
|
},
|
|
{
|
|
"epoch": 9.404388714733543,
|
|
"eval_entropy": 5.122195672619251,
|
|
"eval_loss": 5.631741523742676,
|
|
"eval_mean_token_accuracy": 0.17631014508678933,
|
|
"eval_num_tokens": 21090826.0,
|
|
"eval_runtime": 2.3456,
|
|
"eval_samples_per_second": 1757.309,
|
|
"eval_steps_per_second": 219.983,
|
|
"step": 12000
|
|
},
|
|
{
|
|
"entropy": 5.31616358757019,
|
|
"epoch": 9.408307210031348,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.4460811825932125e-05,
|
|
"loss": 4.3944,
|
|
"mean_token_accuracy": 0.2541874676942825,
|
|
"num_tokens": 21099167.0,
|
|
"step": 12005
|
|
},
|
|
{
|
|
"entropy": 5.279919147491455,
|
|
"epoch": 9.412225705329153,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.440141088164569e-05,
|
|
"loss": 4.3609,
|
|
"mean_token_accuracy": 0.25574908703565596,
|
|
"num_tokens": 21107416.0,
|
|
"step": 12010
|
|
},
|
|
{
|
|
"entropy": 5.249281930923462,
|
|
"epoch": 9.41614420062696,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.434240418280635e-05,
|
|
"loss": 4.292,
|
|
"mean_token_accuracy": 0.26635205149650576,
|
|
"num_tokens": 21115893.0,
|
|
"step": 12015
|
|
},
|
|
{
|
|
"entropy": 5.268414545059204,
|
|
"epoch": 9.420062695924765,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.4283791834868706e-05,
|
|
"loss": 4.3373,
|
|
"mean_token_accuracy": 0.256488224864006,
|
|
"num_tokens": 21124261.0,
|
|
"step": 12020
|
|
},
|
|
{
|
|
"entropy": 5.303501749038697,
|
|
"epoch": 9.42398119122257,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.422557394258246e-05,
|
|
"loss": 4.3556,
|
|
"mean_token_accuracy": 0.25453159809112547,
|
|
"num_tokens": 21132588.0,
|
|
"step": 12025
|
|
},
|
|
{
|
|
"entropy": 5.19997272491455,
|
|
"epoch": 9.427899686520377,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.416775060999251e-05,
|
|
"loss": 4.2469,
|
|
"mean_token_accuracy": 0.2703492656350136,
|
|
"num_tokens": 21140808.0,
|
|
"step": 12030
|
|
},
|
|
{
|
|
"entropy": 5.246651744842529,
|
|
"epoch": 9.431818181818182,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.411032194043848e-05,
|
|
"loss": 4.313,
|
|
"mean_token_accuracy": 0.26067639589309693,
|
|
"num_tokens": 21149549.0,
|
|
"step": 12035
|
|
},
|
|
{
|
|
"entropy": 5.302351617813111,
|
|
"epoch": 9.435736677115987,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.405328803655471e-05,
|
|
"loss": 4.3774,
|
|
"mean_token_accuracy": 0.2507230833172798,
|
|
"num_tokens": 21159009.0,
|
|
"step": 12040
|
|
},
|
|
{
|
|
"entropy": 5.261591625213623,
|
|
"epoch": 9.439655172413794,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.399664900027003e-05,
|
|
"loss": 4.2679,
|
|
"mean_token_accuracy": 0.25860666632652285,
|
|
"num_tokens": 21166684.0,
|
|
"step": 12045
|
|
},
|
|
{
|
|
"entropy": 5.279745244979859,
|
|
"epoch": 9.443573667711599,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.3940404932807624e-05,
|
|
"loss": 4.393,
|
|
"mean_token_accuracy": 0.2503048777580261,
|
|
"num_tokens": 21175327.0,
|
|
"step": 12050
|
|
},
|
|
{
|
|
"entropy": 5.24882698059082,
|
|
"epoch": 9.447492163009404,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.388455593468471e-05,
|
|
"loss": 4.3818,
|
|
"mean_token_accuracy": 0.25050466060638427,
|
|
"num_tokens": 21184544.0,
|
|
"step": 12055
|
|
},
|
|
{
|
|
"entropy": 5.227666902542114,
|
|
"epoch": 9.45141065830721,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.382910210571249e-05,
|
|
"loss": 4.3313,
|
|
"mean_token_accuracy": 0.25225730538368224,
|
|
"num_tokens": 21193443.0,
|
|
"step": 12060
|
|
},
|
|
{
|
|
"entropy": 5.2899346351623535,
|
|
"epoch": 9.455329153605016,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.377404354499603e-05,
|
|
"loss": 4.3631,
|
|
"mean_token_accuracy": 0.25768481492996215,
|
|
"num_tokens": 21201416.0,
|
|
"step": 12065
|
|
},
|
|
{
|
|
"entropy": 5.282749128341675,
|
|
"epoch": 9.45924764890282,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.3719380350933805e-05,
|
|
"loss": 4.3885,
|
|
"mean_token_accuracy": 0.25512287616729734,
|
|
"num_tokens": 21210861.0,
|
|
"step": 12070
|
|
},
|
|
{
|
|
"entropy": 5.227520084381103,
|
|
"epoch": 9.463166144200628,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.366511262121788e-05,
|
|
"loss": 4.3821,
|
|
"mean_token_accuracy": 0.24735505729913712,
|
|
"num_tokens": 21219783.0,
|
|
"step": 12075
|
|
},
|
|
{
|
|
"entropy": 5.280069971084595,
|
|
"epoch": 9.467084639498433,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.361124045283347e-05,
|
|
"loss": 4.286,
|
|
"mean_token_accuracy": 0.2633885979652405,
|
|
"num_tokens": 21228021.0,
|
|
"step": 12080
|
|
},
|
|
{
|
|
"entropy": 5.3116659164428714,
|
|
"epoch": 9.471003134796238,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.3557763942058866e-05,
|
|
"loss": 4.4721,
|
|
"mean_token_accuracy": 0.246195487678051,
|
|
"num_tokens": 21237235.0,
|
|
"step": 12085
|
|
},
|
|
{
|
|
"entropy": 5.184883403778076,
|
|
"epoch": 9.474921630094045,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.350468318446527e-05,
|
|
"loss": 4.292,
|
|
"mean_token_accuracy": 0.2607369065284729,
|
|
"num_tokens": 21246132.0,
|
|
"step": 12090
|
|
},
|
|
{
|
|
"entropy": 5.187571811676025,
|
|
"epoch": 9.47884012539185,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.345199827491658e-05,
|
|
"loss": 4.272,
|
|
"mean_token_accuracy": 0.27801048904657366,
|
|
"num_tokens": 21255500.0,
|
|
"step": 12095
|
|
},
|
|
{
|
|
"entropy": 5.353012561798096,
|
|
"epoch": 9.482758620689655,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.339970930756929e-05,
|
|
"loss": 4.4093,
|
|
"mean_token_accuracy": 0.25015482008457185,
|
|
"num_tokens": 21264536.0,
|
|
"step": 12100
|
|
},
|
|
{
|
|
"entropy": 5.3580211162567135,
|
|
"epoch": 9.486677115987462,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.334781637587228e-05,
|
|
"loss": 4.4626,
|
|
"mean_token_accuracy": 0.2481685534119606,
|
|
"num_tokens": 21273424.0,
|
|
"step": 12105
|
|
},
|
|
{
|
|
"entropy": 5.2932075500488285,
|
|
"epoch": 9.490595611285267,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.3296319572566603e-05,
|
|
"loss": 4.4266,
|
|
"mean_token_accuracy": 0.24548444002866746,
|
|
"num_tokens": 21281537.0,
|
|
"step": 12110
|
|
},
|
|
{
|
|
"entropy": 5.28658390045166,
|
|
"epoch": 9.494514106583072,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.324521898968542e-05,
|
|
"loss": 4.3335,
|
|
"mean_token_accuracy": 0.25592065304517747,
|
|
"num_tokens": 21289776.0,
|
|
"step": 12115
|
|
},
|
|
{
|
|
"entropy": 5.2551545143127445,
|
|
"epoch": 9.498432601880877,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.319451471855375e-05,
|
|
"loss": 4.2973,
|
|
"mean_token_accuracy": 0.2589580774307251,
|
|
"num_tokens": 21297624.0,
|
|
"step": 12120
|
|
},
|
|
{
|
|
"entropy": 5.250433111190796,
|
|
"epoch": 9.502351097178684,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.314420684978832e-05,
|
|
"loss": 4.3249,
|
|
"mean_token_accuracy": 0.2689234182238579,
|
|
"num_tokens": 21306386.0,
|
|
"step": 12125
|
|
},
|
|
{
|
|
"entropy": 5.2568028450012205,
|
|
"epoch": 9.506269592476489,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.309429547329749e-05,
|
|
"loss": 4.3497,
|
|
"mean_token_accuracy": 0.2537443697452545,
|
|
"num_tokens": 21315139.0,
|
|
"step": 12130
|
|
},
|
|
{
|
|
"entropy": 5.302856731414795,
|
|
"epoch": 9.510188087774294,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.304478067828096e-05,
|
|
"loss": 4.353,
|
|
"mean_token_accuracy": 0.25588456392288206,
|
|
"num_tokens": 21323890.0,
|
|
"step": 12135
|
|
},
|
|
{
|
|
"entropy": 5.280727910995483,
|
|
"epoch": 9.5141065830721,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.29956625532297e-05,
|
|
"loss": 4.4131,
|
|
"mean_token_accuracy": 0.2561857834458351,
|
|
"num_tokens": 21332448.0,
|
|
"step": 12140
|
|
},
|
|
{
|
|
"entropy": 5.375219392776489,
|
|
"epoch": 9.518025078369906,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.2946941185925736e-05,
|
|
"loss": 4.4451,
|
|
"mean_token_accuracy": 0.2523625299334526,
|
|
"num_tokens": 21340869.0,
|
|
"step": 12145
|
|
},
|
|
{
|
|
"entropy": 5.28352952003479,
|
|
"epoch": 9.521943573667711,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.289861666344214e-05,
|
|
"loss": 4.3468,
|
|
"mean_token_accuracy": 0.25556098073720934,
|
|
"num_tokens": 21350335.0,
|
|
"step": 12150
|
|
},
|
|
{
|
|
"entropy": 5.291236019134521,
|
|
"epoch": 9.525862068965518,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.2850689072142584e-05,
|
|
"loss": 4.3422,
|
|
"mean_token_accuracy": 0.25656349062919614,
|
|
"num_tokens": 21359160.0,
|
|
"step": 12155
|
|
},
|
|
{
|
|
"entropy": 5.242954349517822,
|
|
"epoch": 9.529780564263323,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.280315849768146e-05,
|
|
"loss": 4.2563,
|
|
"mean_token_accuracy": 0.2643567770719528,
|
|
"num_tokens": 21367619.0,
|
|
"step": 12160
|
|
},
|
|
{
|
|
"entropy": 5.190964317321777,
|
|
"epoch": 9.533699059561128,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.275602502500364e-05,
|
|
"loss": 4.2886,
|
|
"mean_token_accuracy": 0.2629197061061859,
|
|
"num_tokens": 21376193.0,
|
|
"step": 12165
|
|
},
|
|
{
|
|
"entropy": 5.243483924865723,
|
|
"epoch": 9.537617554858935,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.270928873834425e-05,
|
|
"loss": 4.3241,
|
|
"mean_token_accuracy": 0.25935197472572324,
|
|
"num_tokens": 21385641.0,
|
|
"step": 12170
|
|
},
|
|
{
|
|
"entropy": 5.33450665473938,
|
|
"epoch": 9.54153605015674,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.266294972122864e-05,
|
|
"loss": 4.406,
|
|
"mean_token_accuracy": 0.2563857868313789,
|
|
"num_tokens": 21394239.0,
|
|
"step": 12175
|
|
},
|
|
{
|
|
"entropy": 5.241534233093262,
|
|
"epoch": 9.545454545454545,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.2617008056472096e-05,
|
|
"loss": 4.3447,
|
|
"mean_token_accuracy": 0.25325876772403716,
|
|
"num_tokens": 21403427.0,
|
|
"step": 12180
|
|
},
|
|
{
|
|
"entropy": 5.279167652130127,
|
|
"epoch": 9.549373040752352,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.25714638261799e-05,
|
|
"loss": 4.3809,
|
|
"mean_token_accuracy": 0.247143217921257,
|
|
"num_tokens": 21412062.0,
|
|
"step": 12185
|
|
},
|
|
{
|
|
"entropy": 5.287091207504273,
|
|
"epoch": 9.553291536050157,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.252631711174693e-05,
|
|
"loss": 4.3767,
|
|
"mean_token_accuracy": 0.2478463441133499,
|
|
"num_tokens": 21420906.0,
|
|
"step": 12190
|
|
},
|
|
{
|
|
"entropy": 5.274778699874878,
|
|
"epoch": 9.557210031347962,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.248156799385767e-05,
|
|
"loss": 4.3677,
|
|
"mean_token_accuracy": 0.25143668949604037,
|
|
"num_tokens": 21430796.0,
|
|
"step": 12195
|
|
},
|
|
{
|
|
"entropy": 5.270624208450317,
|
|
"epoch": 9.561128526645769,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.243721655248607e-05,
|
|
"loss": 4.3055,
|
|
"mean_token_accuracy": 0.25949068665504454,
|
|
"num_tokens": 21439835.0,
|
|
"step": 12200
|
|
},
|
|
{
|
|
"entropy": 5.1822590827941895,
|
|
"epoch": 9.565047021943574,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.239326286689537e-05,
|
|
"loss": 4.2456,
|
|
"mean_token_accuracy": 0.26846679896116254,
|
|
"num_tokens": 21448754.0,
|
|
"step": 12205
|
|
},
|
|
{
|
|
"entropy": 5.282749366760254,
|
|
"epoch": 9.568965517241379,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.234970701563789e-05,
|
|
"loss": 4.3864,
|
|
"mean_token_accuracy": 0.25035842657089236,
|
|
"num_tokens": 21457908.0,
|
|
"step": 12210
|
|
},
|
|
{
|
|
"entropy": 5.270887136459351,
|
|
"epoch": 9.572884012539184,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 5.2306549076555047e-05,
|
|
"loss": 4.351,
|
|
"mean_token_accuracy": 0.24660114347934722,
|
|
"num_tokens": 21466915.0,
|
|
"step": 12215
|
|
},
|
|
{
|
|
"entropy": 5.26147780418396,
|
|
"epoch": 9.576802507836991,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.226378912677701e-05,
|
|
"loss": 4.3657,
|
|
"mean_token_accuracy": 0.25292214900255205,
|
|
"num_tokens": 21475641.0,
|
|
"step": 12220
|
|
},
|
|
{
|
|
"entropy": 5.33534688949585,
|
|
"epoch": 9.580721003134796,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.2221427242722815e-05,
|
|
"loss": 4.4004,
|
|
"mean_token_accuracy": 0.2545147567987442,
|
|
"num_tokens": 21483913.0,
|
|
"step": 12225
|
|
},
|
|
{
|
|
"entropy": 5.2458432674407955,
|
|
"epoch": 9.584639498432601,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.2179463500099944e-05,
|
|
"loss": 4.3413,
|
|
"mean_token_accuracy": 0.2615157425403595,
|
|
"num_tokens": 21493105.0,
|
|
"step": 12230
|
|
},
|
|
{
|
|
"entropy": 5.257165956497192,
|
|
"epoch": 9.588557993730408,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.2137897973904476e-05,
|
|
"loss": 4.3934,
|
|
"mean_token_accuracy": 0.25222291350364684,
|
|
"num_tokens": 21501926.0,
|
|
"step": 12235
|
|
},
|
|
{
|
|
"entropy": 5.309366989135742,
|
|
"epoch": 9.592476489028213,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.2096730738420725e-05,
|
|
"loss": 4.3784,
|
|
"mean_token_accuracy": 0.25775621831417084,
|
|
"num_tokens": 21510454.0,
|
|
"step": 12240
|
|
},
|
|
{
|
|
"entropy": 5.2494011402130125,
|
|
"epoch": 9.596394984326018,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.2055961867221176e-05,
|
|
"loss": 4.3632,
|
|
"mean_token_accuracy": 0.2589514046907425,
|
|
"num_tokens": 21519115.0,
|
|
"step": 12245
|
|
},
|
|
{
|
|
"entropy": 5.338603734970093,
|
|
"epoch": 9.600313479623825,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.2015591433166435e-05,
|
|
"loss": 4.4096,
|
|
"mean_token_accuracy": 0.24941024780273438,
|
|
"num_tokens": 21528006.0,
|
|
"step": 12250
|
|
},
|
|
{
|
|
"entropy": 5.27471923828125,
|
|
"epoch": 9.60423197492163,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.1975619508405055e-05,
|
|
"loss": 4.3845,
|
|
"mean_token_accuracy": 0.2498660773038864,
|
|
"num_tokens": 21536208.0,
|
|
"step": 12255
|
|
},
|
|
{
|
|
"entropy": 5.3227519512176515,
|
|
"epoch": 9.608150470219435,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.1936046164373294e-05,
|
|
"loss": 4.4147,
|
|
"mean_token_accuracy": 0.24964919537305832,
|
|
"num_tokens": 21544669.0,
|
|
"step": 12260
|
|
},
|
|
{
|
|
"entropy": 5.243588209152222,
|
|
"epoch": 9.612068965517242,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.1896871471795165e-05,
|
|
"loss": 4.3115,
|
|
"mean_token_accuracy": 0.2604330226778984,
|
|
"num_tokens": 21553963.0,
|
|
"step": 12265
|
|
},
|
|
{
|
|
"entropy": 5.227691316604615,
|
|
"epoch": 9.615987460815047,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.1858095500682195e-05,
|
|
"loss": 4.3443,
|
|
"mean_token_accuracy": 0.26044529527425764,
|
|
"num_tokens": 21563052.0,
|
|
"step": 12270
|
|
},
|
|
{
|
|
"entropy": 5.268552780151367,
|
|
"epoch": 9.619905956112852,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.1819718320333296e-05,
|
|
"loss": 4.3603,
|
|
"mean_token_accuracy": 0.26142162680625913,
|
|
"num_tokens": 21571528.0,
|
|
"step": 12275
|
|
},
|
|
{
|
|
"entropy": 5.322037124633789,
|
|
"epoch": 9.623824451410659,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.178173999933474e-05,
|
|
"loss": 4.3811,
|
|
"mean_token_accuracy": 0.2561865895986557,
|
|
"num_tokens": 21580178.0,
|
|
"step": 12280
|
|
},
|
|
{
|
|
"entropy": 5.288453245162964,
|
|
"epoch": 9.627742946708464,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.1744160605559963e-05,
|
|
"loss": 4.3879,
|
|
"mean_token_accuracy": 0.25142622590065,
|
|
"num_tokens": 21589259.0,
|
|
"step": 12285
|
|
},
|
|
{
|
|
"entropy": 5.305143404006958,
|
|
"epoch": 9.63166144200627,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.170698020616942e-05,
|
|
"loss": 4.387,
|
|
"mean_token_accuracy": 0.24713998883962632,
|
|
"num_tokens": 21598985.0,
|
|
"step": 12290
|
|
},
|
|
{
|
|
"entropy": 5.290097284317016,
|
|
"epoch": 9.635579937304076,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.167019886761049e-05,
|
|
"loss": 4.3552,
|
|
"mean_token_accuracy": 0.2565033361315727,
|
|
"num_tokens": 21607764.0,
|
|
"step": 12295
|
|
},
|
|
{
|
|
"entropy": 5.255919027328491,
|
|
"epoch": 9.639498432601881,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.1633816655617397e-05,
|
|
"loss": 4.2597,
|
|
"mean_token_accuracy": 0.27448892742395403,
|
|
"num_tokens": 21616466.0,
|
|
"step": 12300
|
|
},
|
|
{
|
|
"entropy": 5.396167373657226,
|
|
"epoch": 9.643416927899686,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.159783363521106e-05,
|
|
"loss": 4.3939,
|
|
"mean_token_accuracy": 0.2637874484062195,
|
|
"num_tokens": 21625425.0,
|
|
"step": 12305
|
|
},
|
|
{
|
|
"entropy": 5.295594501495361,
|
|
"epoch": 9.647335423197493,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.156224987069898e-05,
|
|
"loss": 4.3925,
|
|
"mean_token_accuracy": 0.2493973821401596,
|
|
"num_tokens": 21633323.0,
|
|
"step": 12310
|
|
},
|
|
{
|
|
"entropy": 5.2896843433380125,
|
|
"epoch": 9.651253918495298,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.152706542567509e-05,
|
|
"loss": 4.3462,
|
|
"mean_token_accuracy": 0.25214879065752027,
|
|
"num_tokens": 21641486.0,
|
|
"step": 12315
|
|
},
|
|
{
|
|
"entropy": 5.262533807754517,
|
|
"epoch": 9.655172413793103,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.1492280363019734e-05,
|
|
"loss": 4.322,
|
|
"mean_token_accuracy": 0.2647132694721222,
|
|
"num_tokens": 21649902.0,
|
|
"step": 12320
|
|
},
|
|
{
|
|
"entropy": 5.275671863555909,
|
|
"epoch": 9.659090909090908,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.1457894744899454e-05,
|
|
"loss": 4.3673,
|
|
"mean_token_accuracy": 0.2566460132598877,
|
|
"num_tokens": 21658874.0,
|
|
"step": 12325
|
|
},
|
|
{
|
|
"entropy": 5.318418741226196,
|
|
"epoch": 9.663009404388715,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.1423908632766886e-05,
|
|
"loss": 4.3692,
|
|
"mean_token_accuracy": 0.25682917386293413,
|
|
"num_tokens": 21667985.0,
|
|
"step": 12330
|
|
},
|
|
{
|
|
"entropy": 5.3340373039245605,
|
|
"epoch": 9.66692789968652,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 5.1390322087360744e-05,
|
|
"loss": 4.4322,
|
|
"mean_token_accuracy": 0.24797493070364,
|
|
"num_tokens": 21677256.0,
|
|
"step": 12335
|
|
},
|
|
{
|
|
"entropy": 5.313819122314453,
|
|
"epoch": 9.670846394984325,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.135713516870565e-05,
|
|
"loss": 4.3402,
|
|
"mean_token_accuracy": 0.25696066617965696,
|
|
"num_tokens": 21685744.0,
|
|
"step": 12340
|
|
},
|
|
{
|
|
"entropy": 5.27883472442627,
|
|
"epoch": 9.674764890282132,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.132434793611201e-05,
|
|
"loss": 4.3586,
|
|
"mean_token_accuracy": 0.252175372838974,
|
|
"num_tokens": 21694552.0,
|
|
"step": 12345
|
|
},
|
|
{
|
|
"entropy": 5.254234409332275,
|
|
"epoch": 9.678683385579937,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.129196044817586e-05,
|
|
"loss": 4.3544,
|
|
"mean_token_accuracy": 0.2528924822807312,
|
|
"num_tokens": 21703582.0,
|
|
"step": 12350
|
|
},
|
|
{
|
|
"entropy": 5.293454170227051,
|
|
"epoch": 9.682601880877742,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.125997276277899e-05,
|
|
"loss": 4.3288,
|
|
"mean_token_accuracy": 0.2587159425020218,
|
|
"num_tokens": 21712015.0,
|
|
"step": 12355
|
|
},
|
|
{
|
|
"entropy": 5.265674209594726,
|
|
"epoch": 9.68652037617555,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.122838493708849e-05,
|
|
"loss": 4.357,
|
|
"mean_token_accuracy": 0.2574944496154785,
|
|
"num_tokens": 21719957.0,
|
|
"step": 12360
|
|
},
|
|
{
|
|
"entropy": 5.304062557220459,
|
|
"epoch": 9.690438871473354,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.1197197027556974e-05,
|
|
"loss": 4.4433,
|
|
"mean_token_accuracy": 0.24321114271879196,
|
|
"num_tokens": 21728871.0,
|
|
"step": 12365
|
|
},
|
|
{
|
|
"entropy": 5.259101438522339,
|
|
"epoch": 9.69435736677116,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.116640908992229e-05,
|
|
"loss": 4.3935,
|
|
"mean_token_accuracy": 0.2509265050292015,
|
|
"num_tokens": 21737843.0,
|
|
"step": 12370
|
|
},
|
|
{
|
|
"entropy": 5.31389045715332,
|
|
"epoch": 9.698275862068966,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.113602117920747e-05,
|
|
"loss": 4.3894,
|
|
"mean_token_accuracy": 0.2474272757768631,
|
|
"num_tokens": 21747157.0,
|
|
"step": 12375
|
|
},
|
|
{
|
|
"entropy": 5.283699560165405,
|
|
"epoch": 9.702194357366771,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.110603334972064e-05,
|
|
"loss": 4.338,
|
|
"mean_token_accuracy": 0.25600039809942243,
|
|
"num_tokens": 21755398.0,
|
|
"step": 12380
|
|
},
|
|
{
|
|
"entropy": 5.296568059921265,
|
|
"epoch": 9.706112852664576,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.1076445655054916e-05,
|
|
"loss": 4.3727,
|
|
"mean_token_accuracy": 0.25590337812900543,
|
|
"num_tokens": 21764535.0,
|
|
"step": 12385
|
|
},
|
|
{
|
|
"entropy": 5.284930992126465,
|
|
"epoch": 9.710031347962383,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.1047258148088295e-05,
|
|
"loss": 4.3898,
|
|
"mean_token_accuracy": 0.256745257973671,
|
|
"num_tokens": 21773120.0,
|
|
"step": 12390
|
|
},
|
|
{
|
|
"entropy": 5.264096212387085,
|
|
"epoch": 9.713949843260188,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.1018470880983584e-05,
|
|
"loss": 4.3612,
|
|
"mean_token_accuracy": 0.2528813287615776,
|
|
"num_tokens": 21782370.0,
|
|
"step": 12395
|
|
},
|
|
{
|
|
"entropy": 5.258268547058106,
|
|
"epoch": 9.717868338557993,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.099008390518832e-05,
|
|
"loss": 4.3414,
|
|
"mean_token_accuracy": 0.2572950109839439,
|
|
"num_tokens": 21791782.0,
|
|
"step": 12400
|
|
},
|
|
{
|
|
"entropy": 5.3194657325744625,
|
|
"epoch": 9.721786833855798,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.09620972714346e-05,
|
|
"loss": 4.3368,
|
|
"mean_token_accuracy": 0.2607801273465157,
|
|
"num_tokens": 21800449.0,
|
|
"step": 12405
|
|
},
|
|
{
|
|
"entropy": 5.379841899871826,
|
|
"epoch": 9.725705329153605,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.0934511029739105e-05,
|
|
"loss": 4.4791,
|
|
"mean_token_accuracy": 0.24446183294057847,
|
|
"num_tokens": 21810333.0,
|
|
"step": 12410
|
|
},
|
|
{
|
|
"entropy": 5.248914527893066,
|
|
"epoch": 9.72962382445141,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.090732522940288e-05,
|
|
"loss": 4.3526,
|
|
"mean_token_accuracy": 0.24937730729579927,
|
|
"num_tokens": 21818209.0,
|
|
"step": 12415
|
|
},
|
|
{
|
|
"entropy": 5.264568519592285,
|
|
"epoch": 9.733542319749215,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.08805399190114e-05,
|
|
"loss": 4.3452,
|
|
"mean_token_accuracy": 0.26788533478975296,
|
|
"num_tokens": 21827362.0,
|
|
"step": 12420
|
|
},
|
|
{
|
|
"entropy": 5.264806938171387,
|
|
"epoch": 9.737460815047022,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.085415514643435e-05,
|
|
"loss": 4.3031,
|
|
"mean_token_accuracy": 0.26067144274711607,
|
|
"num_tokens": 21836228.0,
|
|
"step": 12425
|
|
},
|
|
{
|
|
"entropy": 5.246355676651001,
|
|
"epoch": 9.741379310344827,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.08281709588256e-05,
|
|
"loss": 4.351,
|
|
"mean_token_accuracy": 0.2543995320796967,
|
|
"num_tokens": 21845121.0,
|
|
"step": 12430
|
|
},
|
|
{
|
|
"entropy": 5.2088298320770265,
|
|
"epoch": 9.745297805642632,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.0802587402623076e-05,
|
|
"loss": 4.2688,
|
|
"mean_token_accuracy": 0.25936016738414763,
|
|
"num_tokens": 21854066.0,
|
|
"step": 12435
|
|
},
|
|
{
|
|
"entropy": 5.257535314559936,
|
|
"epoch": 9.74921630094044,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.077740452354878e-05,
|
|
"loss": 4.3486,
|
|
"mean_token_accuracy": 0.2574338436126709,
|
|
"num_tokens": 21862754.0,
|
|
"step": 12440
|
|
},
|
|
{
|
|
"entropy": 5.277077341079712,
|
|
"epoch": 9.753134796238244,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.0752622366608564e-05,
|
|
"loss": 4.375,
|
|
"mean_token_accuracy": 0.25176107287406924,
|
|
"num_tokens": 21872027.0,
|
|
"step": 12445
|
|
},
|
|
{
|
|
"entropy": 5.387757110595703,
|
|
"epoch": 9.75705329153605,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.072824097609218e-05,
|
|
"loss": 4.4326,
|
|
"mean_token_accuracy": 0.25024505257606505,
|
|
"num_tokens": 21880928.0,
|
|
"step": 12450
|
|
},
|
|
{
|
|
"entropy": 5.331564712524414,
|
|
"epoch": 9.760971786833856,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.070426039557315e-05,
|
|
"loss": 4.3681,
|
|
"mean_token_accuracy": 0.2588734418153763,
|
|
"num_tokens": 21889299.0,
|
|
"step": 12455
|
|
},
|
|
{
|
|
"entropy": 5.248963785171509,
|
|
"epoch": 9.764890282131661,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.068068066790862e-05,
|
|
"loss": 4.316,
|
|
"mean_token_accuracy": 0.2616720750927925,
|
|
"num_tokens": 21897830.0,
|
|
"step": 12460
|
|
},
|
|
{
|
|
"entropy": 5.250711393356323,
|
|
"epoch": 9.768808777429467,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.06575018352394e-05,
|
|
"loss": 4.3344,
|
|
"mean_token_accuracy": 0.25173088610172273,
|
|
"num_tokens": 21906360.0,
|
|
"step": 12465
|
|
},
|
|
{
|
|
"entropy": 5.2406963348388675,
|
|
"epoch": 9.772727272727273,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.063472393898982e-05,
|
|
"loss": 4.3116,
|
|
"mean_token_accuracy": 0.25350545048713685,
|
|
"num_tokens": 21915139.0,
|
|
"step": 12470
|
|
},
|
|
{
|
|
"entropy": 5.249405288696289,
|
|
"epoch": 9.776645768025078,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.061234701986768e-05,
|
|
"loss": 4.3238,
|
|
"mean_token_accuracy": 0.25932156294584274,
|
|
"num_tokens": 21923568.0,
|
|
"step": 12475
|
|
},
|
|
{
|
|
"entropy": 5.29030065536499,
|
|
"epoch": 9.780564263322884,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.059037111786416e-05,
|
|
"loss": 4.3935,
|
|
"mean_token_accuracy": 0.2478021040558815,
|
|
"num_tokens": 21932762.0,
|
|
"step": 12480
|
|
},
|
|
{
|
|
"entropy": 5.313127613067627,
|
|
"epoch": 9.78448275862069,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.056879627225376e-05,
|
|
"loss": 4.3877,
|
|
"mean_token_accuracy": 0.24997245967388154,
|
|
"num_tokens": 21941510.0,
|
|
"step": 12485
|
|
},
|
|
{
|
|
"entropy": 5.29314866065979,
|
|
"epoch": 9.788401253918495,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.054762252159424e-05,
|
|
"loss": 4.4207,
|
|
"mean_token_accuracy": 0.25653643906116486,
|
|
"num_tokens": 21950090.0,
|
|
"step": 12490
|
|
},
|
|
{
|
|
"entropy": 5.269761753082276,
|
|
"epoch": 9.7923197492163,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.0526849903726495e-05,
|
|
"loss": 4.2763,
|
|
"mean_token_accuracy": 0.26627043783664706,
|
|
"num_tokens": 21958297.0,
|
|
"step": 12495
|
|
},
|
|
{
|
|
"entropy": 5.228923511505127,
|
|
"epoch": 9.796238244514107,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.050647845577457e-05,
|
|
"loss": 4.3526,
|
|
"mean_token_accuracy": 0.2556705638766289,
|
|
"num_tokens": 21967143.0,
|
|
"step": 12500
|
|
},
|
|
{
|
|
"epoch": 9.796238244514107,
|
|
"eval_entropy": 5.128561469935631,
|
|
"eval_loss": 5.632709980010986,
|
|
"eval_mean_token_accuracy": 0.17625322103384852,
|
|
"eval_num_tokens": 21967143.0,
|
|
"eval_runtime": 2.1736,
|
|
"eval_samples_per_second": 1896.369,
|
|
"eval_steps_per_second": 237.391,
|
|
"step": 12500
|
|
}
|
|
],
|
|
"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": 4378869340784640.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|