2057 lines
54 KiB
JSON
2057 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.7836990595611285,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692053604125977,
|
|
"epoch": 0.003918495297805642,
|
|
"grad_norm": 13.875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.821,
|
|
"mean_token_accuracy": 0.00013404826167970896,
|
|
"num_tokens": 9174.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691976261138915,
|
|
"epoch": 0.007836990595611285,
|
|
"grad_norm": 13.25,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7352,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 17790.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.6909686088562,
|
|
"epoch": 0.011755485893416929,
|
|
"grad_norm": 9.5,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4692,
|
|
"mean_token_accuracy": 0.024429761280771345,
|
|
"num_tokens": 25850.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.677701091766357,
|
|
"epoch": 0.01567398119122257,
|
|
"grad_norm": 6.59375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2158,
|
|
"mean_token_accuracy": 0.04213750921189785,
|
|
"num_tokens": 35194.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.598451328277587,
|
|
"epoch": 0.019592476489028215,
|
|
"grad_norm": 4.375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9693,
|
|
"mean_token_accuracy": 0.042495733872056005,
|
|
"num_tokens": 44218.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.572938823699952,
|
|
"epoch": 0.023510971786833857,
|
|
"grad_norm": 3.421875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.8009,
|
|
"mean_token_accuracy": 0.04398789629340172,
|
|
"num_tokens": 53397.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.539127254486084,
|
|
"epoch": 0.0274294670846395,
|
|
"grad_norm": 2.921875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6844,
|
|
"mean_token_accuracy": 0.04925155565142632,
|
|
"num_tokens": 62749.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.50910243988037,
|
|
"epoch": 0.03134796238244514,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.6464,
|
|
"mean_token_accuracy": 0.04322206266224384,
|
|
"num_tokens": 71721.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.538900089263915,
|
|
"epoch": 0.03526645768025078,
|
|
"grad_norm": 2.765625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.5818,
|
|
"mean_token_accuracy": 0.04916957952082157,
|
|
"num_tokens": 79844.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.527848815917968,
|
|
"epoch": 0.03918495297805643,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.567,
|
|
"mean_token_accuracy": 0.05607832483947277,
|
|
"num_tokens": 88866.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.449817848205566,
|
|
"epoch": 0.04310344827586207,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4999,
|
|
"mean_token_accuracy": 0.06071024388074875,
|
|
"num_tokens": 97918.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.289769554138184,
|
|
"epoch": 0.047021943573667714,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.42,
|
|
"mean_token_accuracy": 0.06239944435656071,
|
|
"num_tokens": 107043.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.339974212646485,
|
|
"epoch": 0.050940438871473356,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3525,
|
|
"mean_token_accuracy": 0.06377544328570366,
|
|
"num_tokens": 116000.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.358237171173096,
|
|
"epoch": 0.054858934169279,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2722,
|
|
"mean_token_accuracy": 0.06179721914231777,
|
|
"num_tokens": 124559.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.306263256072999,
|
|
"epoch": 0.05877742946708464,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1558,
|
|
"mean_token_accuracy": 0.0636992421001196,
|
|
"num_tokens": 132868.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.310817909240722,
|
|
"epoch": 0.06269592476489028,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.1437,
|
|
"mean_token_accuracy": 0.06109033823013306,
|
|
"num_tokens": 141286.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.31660270690918,
|
|
"epoch": 0.06661442006269593,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0017,
|
|
"mean_token_accuracy": 0.06191765181720257,
|
|
"num_tokens": 150406.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.19018259048462,
|
|
"epoch": 0.07053291536050156,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9406,
|
|
"mean_token_accuracy": 0.06189325153827667,
|
|
"num_tokens": 158770.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.224696159362793,
|
|
"epoch": 0.07445141065830721,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8329,
|
|
"mean_token_accuracy": 0.06557972468435765,
|
|
"num_tokens": 167763.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.152040767669678,
|
|
"epoch": 0.07836990595611286,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.6853,
|
|
"mean_token_accuracy": 0.06438785269856453,
|
|
"num_tokens": 177306.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.053633880615234,
|
|
"epoch": 0.0822884012539185,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.5993,
|
|
"mean_token_accuracy": 0.06454790569841862,
|
|
"num_tokens": 186014.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.987529850006103,
|
|
"epoch": 0.08620689655172414,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4292,
|
|
"mean_token_accuracy": 0.06873111054301262,
|
|
"num_tokens": 194122.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.858163261413575,
|
|
"epoch": 0.09012539184952978,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.3325,
|
|
"mean_token_accuracy": 0.06681315004825591,
|
|
"num_tokens": 203097.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.74828462600708,
|
|
"epoch": 0.09404388714733543,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.2069,
|
|
"mean_token_accuracy": 0.06950863264501095,
|
|
"num_tokens": 211413.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.593859004974366,
|
|
"epoch": 0.09796238244514106,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.1517,
|
|
"mean_token_accuracy": 0.06460827700793743,
|
|
"num_tokens": 220550.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.457711124420166,
|
|
"epoch": 0.10188087774294671,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.9844,
|
|
"mean_token_accuracy": 0.072696752846241,
|
|
"num_tokens": 229197.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.283110237121582,
|
|
"epoch": 0.10579937304075235,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.9816,
|
|
"mean_token_accuracy": 0.06841139532625676,
|
|
"num_tokens": 238479.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.081107711791992,
|
|
"epoch": 0.109717868338558,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.8324,
|
|
"mean_token_accuracy": 0.07079579271376132,
|
|
"num_tokens": 246318.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.931202793121338,
|
|
"epoch": 0.11363636363636363,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.8567,
|
|
"mean_token_accuracy": 0.06792247369885444,
|
|
"num_tokens": 254783.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.767581272125245,
|
|
"epoch": 0.11755485893416928,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.6935,
|
|
"mean_token_accuracy": 0.0719048798084259,
|
|
"num_tokens": 263416.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.662985897064209,
|
|
"epoch": 0.12147335423197492,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.712,
|
|
"mean_token_accuracy": 0.07060935273766518,
|
|
"num_tokens": 271930.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.506498527526855,
|
|
"epoch": 0.12539184952978055,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.619,
|
|
"mean_token_accuracy": 0.07228946462273597,
|
|
"num_tokens": 280546.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.340063381195069,
|
|
"epoch": 0.12931034482758622,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.6031,
|
|
"mean_token_accuracy": 0.07542856186628341,
|
|
"num_tokens": 288813.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.317859268188476,
|
|
"epoch": 0.13322884012539185,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.5757,
|
|
"mean_token_accuracy": 0.07388538345694542,
|
|
"num_tokens": 297966.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.287030601501465,
|
|
"epoch": 0.1371473354231975,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.6313,
|
|
"mean_token_accuracy": 0.07094714678823948,
|
|
"num_tokens": 307135.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.15151834487915,
|
|
"epoch": 0.14106583072100312,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.543,
|
|
"mean_token_accuracy": 0.07589463070034981,
|
|
"num_tokens": 315546.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.161028385162354,
|
|
"epoch": 0.14498432601880878,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5249,
|
|
"mean_token_accuracy": 0.0759074218571186,
|
|
"num_tokens": 323930.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.058754682540894,
|
|
"epoch": 0.14890282131661442,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.392,
|
|
"mean_token_accuracy": 0.08224981352686882,
|
|
"num_tokens": 332419.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.08242244720459,
|
|
"epoch": 0.15282131661442006,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.4781,
|
|
"mean_token_accuracy": 0.07802417948842048,
|
|
"num_tokens": 341362.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.985074710845947,
|
|
"epoch": 0.15673981191222572,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.3972,
|
|
"mean_token_accuracy": 0.07822780273854732,
|
|
"num_tokens": 349395.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.01458387374878,
|
|
"epoch": 0.16065830721003135,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4184,
|
|
"mean_token_accuracy": 0.07329343855381013,
|
|
"num_tokens": 358413.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.919571828842163,
|
|
"epoch": 0.164576802507837,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.5032,
|
|
"mean_token_accuracy": 0.07889928370714187,
|
|
"num_tokens": 366489.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.988800716400147,
|
|
"epoch": 0.16849529780564262,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3554,
|
|
"mean_token_accuracy": 0.07564577832818031,
|
|
"num_tokens": 375335.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.991494131088257,
|
|
"epoch": 0.1724137931034483,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.407,
|
|
"mean_token_accuracy": 0.08004131242632866,
|
|
"num_tokens": 384276.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.928954362869263,
|
|
"epoch": 0.17633228840125392,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.447,
|
|
"mean_token_accuracy": 0.07608951888978481,
|
|
"num_tokens": 393571.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.913915538787842,
|
|
"epoch": 0.18025078369905956,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.3764,
|
|
"mean_token_accuracy": 0.08197141736745835,
|
|
"num_tokens": 401865.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.777705907821655,
|
|
"epoch": 0.1841692789968652,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.3381,
|
|
"mean_token_accuracy": 0.08282902240753173,
|
|
"num_tokens": 410518.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.857710313796997,
|
|
"epoch": 0.18808777429467086,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.356,
|
|
"mean_token_accuracy": 0.08076885864138603,
|
|
"num_tokens": 419769.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.7723448276519775,
|
|
"epoch": 0.1920062695924765,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2799,
|
|
"mean_token_accuracy": 0.08509492203593254,
|
|
"num_tokens": 428204.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.8825091361999515,
|
|
"epoch": 0.19592476489028213,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.3833,
|
|
"mean_token_accuracy": 0.07587559074163437,
|
|
"num_tokens": 436594.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.804588508605957,
|
|
"epoch": 0.19984326018808776,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1619,
|
|
"mean_token_accuracy": 0.08455468490719795,
|
|
"num_tokens": 444645.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.734504652023316,
|
|
"epoch": 0.20376175548589343,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.2047,
|
|
"mean_token_accuracy": 0.0890711173415184,
|
|
"num_tokens": 453016.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.689095687866211,
|
|
"epoch": 0.20768025078369906,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.2874,
|
|
"mean_token_accuracy": 0.08078921884298325,
|
|
"num_tokens": 462116.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.745668983459472,
|
|
"epoch": 0.2115987460815047,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.2359,
|
|
"mean_token_accuracy": 0.08409795872867107,
|
|
"num_tokens": 470807.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.6965230941772464,
|
|
"epoch": 0.21551724137931033,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.3247,
|
|
"mean_token_accuracy": 0.08189421519637108,
|
|
"num_tokens": 479592.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.784852504730225,
|
|
"epoch": 0.219435736677116,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2423,
|
|
"mean_token_accuracy": 0.08114428222179412,
|
|
"num_tokens": 488107.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.663507890701294,
|
|
"epoch": 0.22335423197492163,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.2547,
|
|
"mean_token_accuracy": 0.08439974337816239,
|
|
"num_tokens": 496775.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.740933275222778,
|
|
"epoch": 0.22727272727272727,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.2816,
|
|
"mean_token_accuracy": 0.08367991298437119,
|
|
"num_tokens": 505415.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.680744886398315,
|
|
"epoch": 0.23119122257053293,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.2656,
|
|
"mean_token_accuracy": 0.08193187937140464,
|
|
"num_tokens": 514447.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.735220909118652,
|
|
"epoch": 0.23510971786833856,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2519,
|
|
"mean_token_accuracy": 0.086246570199728,
|
|
"num_tokens": 522998.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.693396520614624,
|
|
"epoch": 0.2390282131661442,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2173,
|
|
"mean_token_accuracy": 0.08238770589232444,
|
|
"num_tokens": 531542.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.604292917251587,
|
|
"epoch": 0.24294670846394983,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.1457,
|
|
"mean_token_accuracy": 0.08177541941404343,
|
|
"num_tokens": 540143.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.716537094116211,
|
|
"epoch": 0.2468652037617555,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.2444,
|
|
"mean_token_accuracy": 0.08475765213370323,
|
|
"num_tokens": 549156.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.560835886001587,
|
|
"epoch": 0.2507836990595611,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0897,
|
|
"mean_token_accuracy": 0.08665165156126023,
|
|
"num_tokens": 557522.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.5509428024292,
|
|
"epoch": 0.2547021943573668,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1604,
|
|
"mean_token_accuracy": 0.0852648138999939,
|
|
"num_tokens": 566650.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.643481254577637,
|
|
"epoch": 0.25862068965517243,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1157,
|
|
"mean_token_accuracy": 0.08511288538575172,
|
|
"num_tokens": 576219.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.535575675964355,
|
|
"epoch": 0.26253918495297807,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.2172,
|
|
"mean_token_accuracy": 0.08702815398573875,
|
|
"num_tokens": 584304.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.477091073989868,
|
|
"epoch": 0.2664576802507837,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.1379,
|
|
"mean_token_accuracy": 0.08963485434651375,
|
|
"num_tokens": 593163.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.577557039260864,
|
|
"epoch": 0.27037617554858934,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.0509,
|
|
"mean_token_accuracy": 0.08813998103141785,
|
|
"num_tokens": 602077.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.504067659378052,
|
|
"epoch": 0.274294670846395,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1645,
|
|
"mean_token_accuracy": 0.08567056804895401,
|
|
"num_tokens": 610009.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.615056896209717,
|
|
"epoch": 0.2782131661442006,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.1942,
|
|
"mean_token_accuracy": 0.08498035669326783,
|
|
"num_tokens": 619282.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.465681743621826,
|
|
"epoch": 0.28213166144200624,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.1301,
|
|
"mean_token_accuracy": 0.08965396955609321,
|
|
"num_tokens": 628138.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.515298175811767,
|
|
"epoch": 0.28605015673981193,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9813,
|
|
"mean_token_accuracy": 0.0954482302069664,
|
|
"num_tokens": 637021.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.504141473770142,
|
|
"epoch": 0.28996865203761757,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.1534,
|
|
"mean_token_accuracy": 0.0818020910024643,
|
|
"num_tokens": 646703.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.471045875549317,
|
|
"epoch": 0.2938871473354232,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0346,
|
|
"mean_token_accuracy": 0.09074857905507087,
|
|
"num_tokens": 655616.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.431417989730835,
|
|
"epoch": 0.29780564263322884,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0272,
|
|
"mean_token_accuracy": 0.09865856096148491,
|
|
"num_tokens": 663783.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.422271394729615,
|
|
"epoch": 0.3017241379310345,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.0382,
|
|
"mean_token_accuracy": 0.09664822444319725,
|
|
"num_tokens": 671855.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.441711044311523,
|
|
"epoch": 0.3056426332288401,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.0574,
|
|
"mean_token_accuracy": 0.0905517354607582,
|
|
"num_tokens": 680981.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.409027004241944,
|
|
"epoch": 0.30956112852664575,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.9713,
|
|
"mean_token_accuracy": 0.09158209860324859,
|
|
"num_tokens": 689294.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.345873880386352,
|
|
"epoch": 0.31347962382445144,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.936,
|
|
"mean_token_accuracy": 0.09149299785494805,
|
|
"num_tokens": 697703.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.485394239425659,
|
|
"epoch": 0.31739811912225707,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0726,
|
|
"mean_token_accuracy": 0.09204713851213456,
|
|
"num_tokens": 706792.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.409369468688965,
|
|
"epoch": 0.3213166144200627,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.9865,
|
|
"mean_token_accuracy": 0.09500117674469948,
|
|
"num_tokens": 715864.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.2893208980560305,
|
|
"epoch": 0.32523510971786834,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.0287,
|
|
"mean_token_accuracy": 0.0940815657377243,
|
|
"num_tokens": 723921.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.437120723724365,
|
|
"epoch": 0.329153605015674,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0524,
|
|
"mean_token_accuracy": 0.08924909234046936,
|
|
"num_tokens": 732797.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.3198113441467285,
|
|
"epoch": 0.3330721003134796,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8983,
|
|
"mean_token_accuracy": 0.10327747613191604,
|
|
"num_tokens": 741248.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.3485208511352536,
|
|
"epoch": 0.33699059561128525,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0034,
|
|
"mean_token_accuracy": 0.0921932153403759,
|
|
"num_tokens": 749766.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.331899881362915,
|
|
"epoch": 0.3409090909090909,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.9561,
|
|
"mean_token_accuracy": 0.09300489202141762,
|
|
"num_tokens": 758695.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.339229249954224,
|
|
"epoch": 0.3448275862068966,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.9566,
|
|
"mean_token_accuracy": 0.09157029166817665,
|
|
"num_tokens": 767624.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.343198585510254,
|
|
"epoch": 0.3487460815047022,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9819,
|
|
"mean_token_accuracy": 0.09128969609737396,
|
|
"num_tokens": 776616.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.2496236801147464,
|
|
"epoch": 0.35266457680250785,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9895,
|
|
"mean_token_accuracy": 0.0921534925699234,
|
|
"num_tokens": 786147.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.356827306747436,
|
|
"epoch": 0.3565830721003135,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.0808,
|
|
"mean_token_accuracy": 0.08939703032374383,
|
|
"num_tokens": 795213.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.332914733886719,
|
|
"epoch": 0.3605015673981191,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.8956,
|
|
"mean_token_accuracy": 0.09629729315638542,
|
|
"num_tokens": 803118.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.346728849411011,
|
|
"epoch": 0.36442006269592475,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8641,
|
|
"mean_token_accuracy": 0.10322200134396553,
|
|
"num_tokens": 811358.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.147189569473267,
|
|
"epoch": 0.3683385579937304,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.8718,
|
|
"mean_token_accuracy": 0.09597681015729904,
|
|
"num_tokens": 819653.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.195614671707153,
|
|
"epoch": 0.3722570532915361,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.9088,
|
|
"mean_token_accuracy": 0.09375306144356728,
|
|
"num_tokens": 828462.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.361067628860473,
|
|
"epoch": 0.3761755485893417,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.9995,
|
|
"mean_token_accuracy": 0.09758970066905022,
|
|
"num_tokens": 836204.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.163541269302368,
|
|
"epoch": 0.38009404388714735,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8697,
|
|
"mean_token_accuracy": 0.1009294331073761,
|
|
"num_tokens": 845032.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.267932987213134,
|
|
"epoch": 0.384012539184953,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.8906,
|
|
"mean_token_accuracy": 0.10076733604073525,
|
|
"num_tokens": 853324.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.26249213218689,
|
|
"epoch": 0.3879310344827586,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8639,
|
|
"mean_token_accuracy": 0.1036192737519741,
|
|
"num_tokens": 861640.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.264737176895141,
|
|
"epoch": 0.39184952978056425,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.047,
|
|
"mean_token_accuracy": 0.08870344087481499,
|
|
"num_tokens": 870758.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.39184952978056425,
|
|
"eval_entropy": 7.3145306674084924,
|
|
"eval_loss": 7.030722141265869,
|
|
"eval_mean_token_accuracy": 0.09521551475708567,
|
|
"eval_num_tokens": 870758.0,
|
|
"eval_runtime": 2.8925,
|
|
"eval_samples_per_second": 1425.051,
|
|
"eval_steps_per_second": 178.391,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.310623741149902,
|
|
"epoch": 0.3957680250783699,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.9526,
|
|
"mean_token_accuracy": 0.08988267555832863,
|
|
"num_tokens": 879968.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.146830654144287,
|
|
"epoch": 0.3996865203761755,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.9074,
|
|
"mean_token_accuracy": 0.09161595106124878,
|
|
"num_tokens": 888804.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.1906270503997805,
|
|
"epoch": 0.4036050156739812,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.9221,
|
|
"mean_token_accuracy": 0.08995741233229637,
|
|
"num_tokens": 898593.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.173298168182373,
|
|
"epoch": 0.40752351097178685,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.9368,
|
|
"mean_token_accuracy": 0.09437091797590255,
|
|
"num_tokens": 906873.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.218977069854736,
|
|
"epoch": 0.4114420062695925,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.9019,
|
|
"mean_token_accuracy": 0.10439103990793228,
|
|
"num_tokens": 915080.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.28611798286438,
|
|
"epoch": 0.4153605015673981,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.911,
|
|
"mean_token_accuracy": 0.0972377523779869,
|
|
"num_tokens": 923741.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.100715637207031,
|
|
"epoch": 0.41927899686520376,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8103,
|
|
"mean_token_accuracy": 0.10250237360596656,
|
|
"num_tokens": 932455.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.147137594223023,
|
|
"epoch": 0.4231974921630094,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8173,
|
|
"mean_token_accuracy": 0.10221570506691932,
|
|
"num_tokens": 941064.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.074875783920288,
|
|
"epoch": 0.427115987460815,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8391,
|
|
"mean_token_accuracy": 0.10051099881529808,
|
|
"num_tokens": 950261.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.238851642608642,
|
|
"epoch": 0.43103448275862066,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.9112,
|
|
"mean_token_accuracy": 0.09698623940348625,
|
|
"num_tokens": 959151.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.171272468566895,
|
|
"epoch": 0.43495297805642635,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.934,
|
|
"mean_token_accuracy": 0.09741523563861847,
|
|
"num_tokens": 968441.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.075249052047729,
|
|
"epoch": 0.438871473354232,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8133,
|
|
"mean_token_accuracy": 0.09270348027348518,
|
|
"num_tokens": 977058.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.114323377609253,
|
|
"epoch": 0.4427899686520376,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.8275,
|
|
"mean_token_accuracy": 0.10172541886568069,
|
|
"num_tokens": 986531.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.010313701629639,
|
|
"epoch": 0.44670846394984326,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.6613,
|
|
"mean_token_accuracy": 0.10588647425174713,
|
|
"num_tokens": 995081.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.006834888458252,
|
|
"epoch": 0.4506269592476489,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7384,
|
|
"mean_token_accuracy": 0.10787067040801049,
|
|
"num_tokens": 1003459.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.008131837844848,
|
|
"epoch": 0.45454545454545453,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.8013,
|
|
"mean_token_accuracy": 0.09799510389566421,
|
|
"num_tokens": 1012420.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.163480377197265,
|
|
"epoch": 0.45846394984326017,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.8601,
|
|
"mean_token_accuracy": 0.09603445902466774,
|
|
"num_tokens": 1021198.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.066182661056518,
|
|
"epoch": 0.46238244514106586,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8494,
|
|
"mean_token_accuracy": 0.1006757028400898,
|
|
"num_tokens": 1030023.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.019779109954834,
|
|
"epoch": 0.4663009404388715,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7127,
|
|
"mean_token_accuracy": 0.10001757815480232,
|
|
"num_tokens": 1038824.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.121872615814209,
|
|
"epoch": 0.4702194357366771,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.9753,
|
|
"mean_token_accuracy": 0.0965359665453434,
|
|
"num_tokens": 1048207.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.160170698165894,
|
|
"epoch": 0.47413793103448276,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.9391,
|
|
"mean_token_accuracy": 0.10044264048337936,
|
|
"num_tokens": 1057354.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.145117998123169,
|
|
"epoch": 0.4780564263322884,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.8633,
|
|
"mean_token_accuracy": 0.0963246464729309,
|
|
"num_tokens": 1066825.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.0773111343383786,
|
|
"epoch": 0.48197492163009403,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7953,
|
|
"mean_token_accuracy": 0.0973593220114708,
|
|
"num_tokens": 1075752.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.937968826293945,
|
|
"epoch": 0.48589341692789967,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7024,
|
|
"mean_token_accuracy": 0.10333408787846565,
|
|
"num_tokens": 1084826.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.09277081489563,
|
|
"epoch": 0.4898119122257053,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.854,
|
|
"mean_token_accuracy": 0.09841757342219352,
|
|
"num_tokens": 1093425.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.049381494522095,
|
|
"epoch": 0.493730407523511,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.8692,
|
|
"mean_token_accuracy": 0.09519068449735642,
|
|
"num_tokens": 1102619.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.003838348388672,
|
|
"epoch": 0.49764890282131663,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.7563,
|
|
"mean_token_accuracy": 0.10078713521361352,
|
|
"num_tokens": 1111337.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.916393804550171,
|
|
"epoch": 0.5015673981191222,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5811,
|
|
"mean_token_accuracy": 0.11013391688466072,
|
|
"num_tokens": 1119381.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.888113069534302,
|
|
"epoch": 0.5054858934169278,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6508,
|
|
"mean_token_accuracy": 0.10688698589801789,
|
|
"num_tokens": 1127825.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.003930568695068,
|
|
"epoch": 0.5094043887147336,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.6994,
|
|
"mean_token_accuracy": 0.09896195456385612,
|
|
"num_tokens": 1136750.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.882824468612671,
|
|
"epoch": 0.5133228840125392,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.7242,
|
|
"mean_token_accuracy": 0.10310067981481552,
|
|
"num_tokens": 1145739.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.920678424835205,
|
|
"epoch": 0.5172413793103449,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.7827,
|
|
"mean_token_accuracy": 0.10021716952323914,
|
|
"num_tokens": 1154415.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.999461698532104,
|
|
"epoch": 0.5211598746081505,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.6635,
|
|
"mean_token_accuracy": 0.10259130671620369,
|
|
"num_tokens": 1162480.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.945166110992432,
|
|
"epoch": 0.5250783699059561,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.7478,
|
|
"mean_token_accuracy": 0.10068432837724686,
|
|
"num_tokens": 1170774.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.878892850875855,
|
|
"epoch": 0.5289968652037618,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6879,
|
|
"mean_token_accuracy": 0.10098834857344627,
|
|
"num_tokens": 1179629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.955817317962646,
|
|
"epoch": 0.5329153605015674,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.8369,
|
|
"mean_token_accuracy": 0.09804248735308647,
|
|
"num_tokens": 1189111.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.9712419509887695,
|
|
"epoch": 0.536833855799373,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.8281,
|
|
"mean_token_accuracy": 0.0970684178173542,
|
|
"num_tokens": 1198827.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.076535654067993,
|
|
"epoch": 0.5407523510971787,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7525,
|
|
"mean_token_accuracy": 0.10446157231926918,
|
|
"num_tokens": 1207089.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.935127210617066,
|
|
"epoch": 0.5446708463949843,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.736,
|
|
"mean_token_accuracy": 0.09537649154663086,
|
|
"num_tokens": 1216509.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.816225671768189,
|
|
"epoch": 0.54858934169279,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5657,
|
|
"mean_token_accuracy": 0.11179759725928307,
|
|
"num_tokens": 1225684.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.831858444213867,
|
|
"epoch": 0.5525078369905956,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.6121,
|
|
"mean_token_accuracy": 0.10936803743243217,
|
|
"num_tokens": 1233675.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.846443557739258,
|
|
"epoch": 0.5564263322884012,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.621,
|
|
"mean_token_accuracy": 0.1022804893553257,
|
|
"num_tokens": 1242147.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.9075483798980715,
|
|
"epoch": 0.5603448275862069,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.7254,
|
|
"mean_token_accuracy": 0.10935754925012589,
|
|
"num_tokens": 1251127.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.810165548324585,
|
|
"epoch": 0.5642633228840125,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6885,
|
|
"mean_token_accuracy": 0.10543572083115578,
|
|
"num_tokens": 1260589.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.9715752601623535,
|
|
"epoch": 0.5681818181818182,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.7526,
|
|
"mean_token_accuracy": 0.10584053248167039,
|
|
"num_tokens": 1269281.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.892933082580567,
|
|
"epoch": 0.5721003134796239,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6119,
|
|
"mean_token_accuracy": 0.1029752716422081,
|
|
"num_tokens": 1278033.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.8791444301605225,
|
|
"epoch": 0.5760188087774295,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6888,
|
|
"mean_token_accuracy": 0.10946752950549125,
|
|
"num_tokens": 1285847.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.881000947952271,
|
|
"epoch": 0.5799373040752351,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.6685,
|
|
"mean_token_accuracy": 0.1023192547261715,
|
|
"num_tokens": 1294770.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.975464582443237,
|
|
"epoch": 0.5838557993730408,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.8736,
|
|
"mean_token_accuracy": 0.09663443863391877,
|
|
"num_tokens": 1304282.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.9174909591674805,
|
|
"epoch": 0.5877742946708464,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.7056,
|
|
"mean_token_accuracy": 0.10080768540501595,
|
|
"num_tokens": 1313218.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.916672325134277,
|
|
"epoch": 0.591692789968652,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6713,
|
|
"mean_token_accuracy": 0.10844261273741722,
|
|
"num_tokens": 1322093.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.670488548278809,
|
|
"epoch": 0.5956112852664577,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.7051,
|
|
"mean_token_accuracy": 0.10560382381081582,
|
|
"num_tokens": 1330718.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.887430095672608,
|
|
"epoch": 0.5995297805642633,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5598,
|
|
"mean_token_accuracy": 0.1074932336807251,
|
|
"num_tokens": 1339672.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.805772209167481,
|
|
"epoch": 0.603448275862069,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5984,
|
|
"mean_token_accuracy": 0.1087186798453331,
|
|
"num_tokens": 1348587.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.776034593582153,
|
|
"epoch": 0.6073667711598746,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5901,
|
|
"mean_token_accuracy": 0.11292538270354271,
|
|
"num_tokens": 1357597.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.818142938613891,
|
|
"epoch": 0.6112852664576802,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.6842,
|
|
"mean_token_accuracy": 0.10455552414059639,
|
|
"num_tokens": 1366246.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.812788200378418,
|
|
"epoch": 0.6152037617554859,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.596,
|
|
"mean_token_accuracy": 0.10609478428959847,
|
|
"num_tokens": 1374664.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.736437702178955,
|
|
"epoch": 0.6191222570532915,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.6125,
|
|
"mean_token_accuracy": 0.10844489559531212,
|
|
"num_tokens": 1382765.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.722091817855835,
|
|
"epoch": 0.6230407523510971,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5665,
|
|
"mean_token_accuracy": 0.10965784117579461,
|
|
"num_tokens": 1391708.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.765339946746826,
|
|
"epoch": 0.6269592476489029,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.7346,
|
|
"mean_token_accuracy": 0.10098599940538407,
|
|
"num_tokens": 1401373.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.851631355285645,
|
|
"epoch": 0.6308777429467085,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5225,
|
|
"mean_token_accuracy": 0.11173290610313416,
|
|
"num_tokens": 1409702.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.767954397201538,
|
|
"epoch": 0.6347962382445141,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.7358,
|
|
"mean_token_accuracy": 0.10304761379957199,
|
|
"num_tokens": 1419596.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.824428558349609,
|
|
"epoch": 0.6387147335423198,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.6758,
|
|
"mean_token_accuracy": 0.1033481739461422,
|
|
"num_tokens": 1428437.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.820134544372559,
|
|
"epoch": 0.6426332288401254,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5833,
|
|
"mean_token_accuracy": 0.10861420258879662,
|
|
"num_tokens": 1436764.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.72536416053772,
|
|
"epoch": 0.646551724137931,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6309,
|
|
"mean_token_accuracy": 0.11201521828770637,
|
|
"num_tokens": 1445755.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.854031276702881,
|
|
"epoch": 0.6504702194357367,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.6969,
|
|
"mean_token_accuracy": 0.10224815532565117,
|
|
"num_tokens": 1454550.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.7012652397155765,
|
|
"epoch": 0.6543887147335423,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.5611,
|
|
"mean_token_accuracy": 0.11515542790293694,
|
|
"num_tokens": 1463039.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.802173805236817,
|
|
"epoch": 0.658307210031348,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.623,
|
|
"mean_token_accuracy": 0.10527266189455986,
|
|
"num_tokens": 1472029.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.709125900268555,
|
|
"epoch": 0.6622257053291536,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.6453,
|
|
"mean_token_accuracy": 0.10808795914053917,
|
|
"num_tokens": 1481588.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.68847393989563,
|
|
"epoch": 0.6661442006269592,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.545,
|
|
"mean_token_accuracy": 0.10692245215177536,
|
|
"num_tokens": 1489915.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.784275102615356,
|
|
"epoch": 0.6700626959247649,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.6288,
|
|
"mean_token_accuracy": 0.10608619600534439,
|
|
"num_tokens": 1497920.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.68899974822998,
|
|
"epoch": 0.6739811912225705,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4869,
|
|
"mean_token_accuracy": 0.10574968531727791,
|
|
"num_tokens": 1506965.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.722927284240723,
|
|
"epoch": 0.6778996865203761,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.5838,
|
|
"mean_token_accuracy": 0.10984734967350959,
|
|
"num_tokens": 1516028.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.755547428131104,
|
|
"epoch": 0.6818181818181818,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.6197,
|
|
"mean_token_accuracy": 0.10767295137047768,
|
|
"num_tokens": 1524579.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.617767429351806,
|
|
"epoch": 0.6857366771159875,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.5326,
|
|
"mean_token_accuracy": 0.11009506732225419,
|
|
"num_tokens": 1534285.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.716221046447754,
|
|
"epoch": 0.6896551724137931,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5287,
|
|
"mean_token_accuracy": 0.11053434684872628,
|
|
"num_tokens": 1543701.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.79583706855774,
|
|
"epoch": 0.6935736677115988,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.7919,
|
|
"mean_token_accuracy": 0.09309381395578384,
|
|
"num_tokens": 1552715.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.501706457138061,
|
|
"epoch": 0.6974921630094044,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4006,
|
|
"mean_token_accuracy": 0.11031684949994087,
|
|
"num_tokens": 1561523.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.685598754882813,
|
|
"epoch": 0.70141065830721,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.6289,
|
|
"mean_token_accuracy": 0.11112729758024216,
|
|
"num_tokens": 1570057.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.727664756774902,
|
|
"epoch": 0.7053291536050157,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4786,
|
|
"mean_token_accuracy": 0.11088098585605621,
|
|
"num_tokens": 1578795.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.55871696472168,
|
|
"epoch": 0.7092476489028213,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.5632,
|
|
"mean_token_accuracy": 0.10584904998540878,
|
|
"num_tokens": 1588108.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.729115009307861,
|
|
"epoch": 0.713166144200627,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5254,
|
|
"mean_token_accuracy": 0.11291411146521568,
|
|
"num_tokens": 1596545.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.622742557525635,
|
|
"epoch": 0.7170846394984326,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.5094,
|
|
"mean_token_accuracy": 0.11332128643989563,
|
|
"num_tokens": 1605098.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.602113914489746,
|
|
"epoch": 0.7210031347962382,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5232,
|
|
"mean_token_accuracy": 0.1159414529800415,
|
|
"num_tokens": 1613760.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.651577711105347,
|
|
"epoch": 0.7249216300940439,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.5629,
|
|
"mean_token_accuracy": 0.10667054280638695,
|
|
"num_tokens": 1622345.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.7364403247833256,
|
|
"epoch": 0.7288401253918495,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.5521,
|
|
"mean_token_accuracy": 0.11116818189620972,
|
|
"num_tokens": 1630706.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.705845355987549,
|
|
"epoch": 0.7327586206896551,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.624,
|
|
"mean_token_accuracy": 0.10690407380461693,
|
|
"num_tokens": 1639880.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.593637418746948,
|
|
"epoch": 0.7366771159874608,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.4648,
|
|
"mean_token_accuracy": 0.10600230619311332,
|
|
"num_tokens": 1648325.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.491366624832153,
|
|
"epoch": 0.7405956112852664,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4219,
|
|
"mean_token_accuracy": 0.11530695110559464,
|
|
"num_tokens": 1656951.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.594129180908203,
|
|
"epoch": 0.7445141065830722,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.5604,
|
|
"mean_token_accuracy": 0.11204110458493233,
|
|
"num_tokens": 1665832.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.592755603790283,
|
|
"epoch": 0.7484326018808778,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4841,
|
|
"mean_token_accuracy": 0.10641552805900574,
|
|
"num_tokens": 1675116.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.512726259231568,
|
|
"epoch": 0.7523510971786834,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4685,
|
|
"mean_token_accuracy": 0.11560196354985237,
|
|
"num_tokens": 1684195.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.613996267318726,
|
|
"epoch": 0.7562695924764891,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4407,
|
|
"mean_token_accuracy": 0.11013077944517136,
|
|
"num_tokens": 1693025.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.462671899795533,
|
|
"epoch": 0.7601880877742947,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4294,
|
|
"mean_token_accuracy": 0.11215238943696022,
|
|
"num_tokens": 1701210.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.602805709838867,
|
|
"epoch": 0.7641065830721003,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.5362,
|
|
"mean_token_accuracy": 0.10191687792539597,
|
|
"num_tokens": 1711026.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.6010805606842045,
|
|
"epoch": 0.768025078369906,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5314,
|
|
"mean_token_accuracy": 0.11036548092961311,
|
|
"num_tokens": 1719441.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.571003103256226,
|
|
"epoch": 0.7719435736677116,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.5292,
|
|
"mean_token_accuracy": 0.10878988429903984,
|
|
"num_tokens": 1728300.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.600493288040161,
|
|
"epoch": 0.7758620689655172,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4193,
|
|
"mean_token_accuracy": 0.11556374430656433,
|
|
"num_tokens": 1736364.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.384331512451172,
|
|
"epoch": 0.7797805642633229,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.259,
|
|
"mean_token_accuracy": 0.12344810739159584,
|
|
"num_tokens": 1745043.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.534456634521485,
|
|
"epoch": 0.7836990595611285,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.4121,
|
|
"mean_token_accuracy": 0.11444604694843293,
|
|
"num_tokens": 1754008.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.7836990595611285,
|
|
"eval_entropy": 6.353288079417029,
|
|
"eval_loss": 6.573922157287598,
|
|
"eval_mean_token_accuracy": 0.11509622025570666,
|
|
"eval_num_tokens": 1754008.0,
|
|
"eval_runtime": 2.8912,
|
|
"eval_samples_per_second": 1425.689,
|
|
"eval_steps_per_second": 178.471,
|
|
"step": 1000
|
|
}
|
|
],
|
|
"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": 2371902916608000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|