a45fd99152
Model: fpadovani/ind-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.39184952978056425,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.505434942245484,
|
|
"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
|
|
}
|
|
],
|
|
"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": 171439857008640.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|