24299 lines
665 KiB
JSON
24299 lines
665 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 9.404388714733543,
|
|
"eval_steps": 500,
|
|
"global_step": 12000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.410930061340332,
|
|
"epoch": 0.7876175548589341,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999998713243189,
|
|
"loss": 6.4673,
|
|
"mean_token_accuracy": 0.10981100052595139,
|
|
"num_tokens": 1763592.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.565805721282959,
|
|
"epoch": 0.7915360501567398,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999993485796164,
|
|
"loss": 6.3337,
|
|
"mean_token_accuracy": 0.11553042531013488,
|
|
"num_tokens": 1773111.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.497714471817017,
|
|
"epoch": 0.7954545454545454,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999984237245962,
|
|
"loss": 6.4084,
|
|
"mean_token_accuracy": 0.12054920345544815,
|
|
"num_tokens": 1781230.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.660019779205323,
|
|
"epoch": 0.799373040752351,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004999970967609109,
|
|
"loss": 6.6371,
|
|
"mean_token_accuracy": 0.11188422217965126,
|
|
"num_tokens": 1790324.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.494213390350342,
|
|
"epoch": 0.8032915360501567,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999953676909322,
|
|
"loss": 6.4729,
|
|
"mean_token_accuracy": 0.10698359534144401,
|
|
"num_tokens": 1798849.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.546792268753052,
|
|
"epoch": 0.8072100313479624,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004999932365177502,
|
|
"loss": 6.3552,
|
|
"mean_token_accuracy": 0.11793587505817413,
|
|
"num_tokens": 1808019.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.554946851730347,
|
|
"epoch": 0.8111285266457681,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999907032451735,
|
|
"loss": 6.467,
|
|
"mean_token_accuracy": 0.11289061605930328,
|
|
"num_tokens": 1816550.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.432303428649902,
|
|
"epoch": 0.8150470219435737,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999877678777296,
|
|
"loss": 6.3661,
|
|
"mean_token_accuracy": 0.11930875778198242,
|
|
"num_tokens": 1825238.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.448621463775635,
|
|
"epoch": 0.8189655172413793,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999844304206645,
|
|
"loss": 6.3297,
|
|
"mean_token_accuracy": 0.11632873341441155,
|
|
"num_tokens": 1833645.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.457649612426758,
|
|
"epoch": 0.822884012539185,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999806908799428,
|
|
"loss": 6.4009,
|
|
"mean_token_accuracy": 0.11582175865769387,
|
|
"num_tokens": 1842892.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.440515804290771,
|
|
"epoch": 0.8268025078369906,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999765492622474,
|
|
"loss": 6.4502,
|
|
"mean_token_accuracy": 0.11736933290958404,
|
|
"num_tokens": 1851400.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.566648626327515,
|
|
"epoch": 0.8307210031347962,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999720055749804,
|
|
"loss": 6.3827,
|
|
"mean_token_accuracy": 0.11811394691467285,
|
|
"num_tokens": 1860088.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.509423923492432,
|
|
"epoch": 0.8346394984326019,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999670598262619,
|
|
"loss": 6.4453,
|
|
"mean_token_accuracy": 0.1164441391825676,
|
|
"num_tokens": 1868558.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.43409571647644,
|
|
"epoch": 0.8385579937304075,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999617120249308,
|
|
"loss": 6.4301,
|
|
"mean_token_accuracy": 0.11259575486183167,
|
|
"num_tokens": 1877624.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.559784078598023,
|
|
"epoch": 0.8424764890282131,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999559621805445,
|
|
"loss": 6.5347,
|
|
"mean_token_accuracy": 0.1132730558514595,
|
|
"num_tokens": 1886505.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.372117948532105,
|
|
"epoch": 0.8463949843260188,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999498103033788,
|
|
"loss": 6.2892,
|
|
"mean_token_accuracy": 0.12123534381389618,
|
|
"num_tokens": 1895701.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.388390398025512,
|
|
"epoch": 0.8503134796238244,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999432564044284,
|
|
"loss": 6.2646,
|
|
"mean_token_accuracy": 0.11819472461938858,
|
|
"num_tokens": 1904640.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.498187494277954,
|
|
"epoch": 0.85423197492163,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999363004954058,
|
|
"loss": 6.4879,
|
|
"mean_token_accuracy": 0.11307620778679847,
|
|
"num_tokens": 1914216.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.445917701721191,
|
|
"epoch": 0.8581504702194357,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.3985,
|
|
"mean_token_accuracy": 0.12216848358511925,
|
|
"num_tokens": 1923280.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.525809192657471,
|
|
"epoch": 0.8620689655172413,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999211826975884,
|
|
"loss": 6.4166,
|
|
"mean_token_accuracy": 0.12132912650704383,
|
|
"num_tokens": 1931403.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.520140981674194,
|
|
"epoch": 0.8659874608150471,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999130208358114,
|
|
"loss": 6.429,
|
|
"mean_token_accuracy": 0.11339464783668518,
|
|
"num_tokens": 1940774.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.503523302078247,
|
|
"epoch": 0.8699059561128527,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999044570179983,
|
|
"loss": 6.5221,
|
|
"mean_token_accuracy": 0.1174374520778656,
|
|
"num_tokens": 1950045.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.526077318191528,
|
|
"epoch": 0.8738244514106583,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998954912594541,
|
|
"loss": 6.3923,
|
|
"mean_token_accuracy": 0.11855587661266327,
|
|
"num_tokens": 1959303.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.462242603302002,
|
|
"epoch": 0.877742946708464,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998861235762018,
|
|
"loss": 6.3603,
|
|
"mean_token_accuracy": 0.11869992539286614,
|
|
"num_tokens": 1968583.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.474194145202636,
|
|
"epoch": 0.8816614420062696,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998763539849832,
|
|
"loss": 6.4832,
|
|
"mean_token_accuracy": 0.11621114015579223,
|
|
"num_tokens": 1977489.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.394860315322876,
|
|
"epoch": 0.8855799373040752,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998661825032579,
|
|
"loss": 6.2376,
|
|
"mean_token_accuracy": 0.12140937894582748,
|
|
"num_tokens": 1985308.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.435638904571533,
|
|
"epoch": 0.8894984326018809,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998556091492043,
|
|
"loss": 6.4123,
|
|
"mean_token_accuracy": 0.10934653729200364,
|
|
"num_tokens": 1993930.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.399746656417847,
|
|
"epoch": 0.8934169278996865,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998446339417184,
|
|
"loss": 6.4064,
|
|
"mean_token_accuracy": 0.11175781339406968,
|
|
"num_tokens": 2003802.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.482501697540283,
|
|
"epoch": 0.8973354231974922,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998332569004149,
|
|
"loss": 6.3189,
|
|
"mean_token_accuracy": 0.12161151990294457,
|
|
"num_tokens": 2011957.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.3887701511383055,
|
|
"epoch": 0.9012539184952978,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998214780456263,
|
|
"loss": 6.314,
|
|
"mean_token_accuracy": 0.12478633299469948,
|
|
"num_tokens": 2020741.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.356020259857178,
|
|
"epoch": 0.9051724137931034,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998092973984033,
|
|
"loss": 6.41,
|
|
"mean_token_accuracy": 0.11564783379435539,
|
|
"num_tokens": 2029804.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.55821795463562,
|
|
"epoch": 0.9090909090909091,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997967149805147,
|
|
"loss": 6.4492,
|
|
"mean_token_accuracy": 0.11650254800915719,
|
|
"num_tokens": 2038732.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.412743139266968,
|
|
"epoch": 0.9130094043887147,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997837308144474,
|
|
"loss": 6.3322,
|
|
"mean_token_accuracy": 0.1288091853260994,
|
|
"num_tokens": 2046362.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.40589394569397,
|
|
"epoch": 0.9169278996865203,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997703449234062,
|
|
"loss": 6.3273,
|
|
"mean_token_accuracy": 0.11811700761318207,
|
|
"num_tokens": 2056018.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.326677036285401,
|
|
"epoch": 0.920846394984326,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997565573313139,
|
|
"loss": 6.3065,
|
|
"mean_token_accuracy": 0.11553404182195663,
|
|
"num_tokens": 2065077.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.469640636444092,
|
|
"epoch": 0.9247648902821317,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997423680628109,
|
|
"loss": 6.295,
|
|
"mean_token_accuracy": 0.11463245674967766,
|
|
"num_tokens": 2073405.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.21384630203247,
|
|
"epoch": 0.9286833855799373,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000499727777143256,
|
|
"loss": 6.2785,
|
|
"mean_token_accuracy": 0.11823253557085991,
|
|
"num_tokens": 2081891.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.534471082687378,
|
|
"epoch": 0.932601880877743,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004997127845987255,
|
|
"loss": 6.3293,
|
|
"mean_token_accuracy": 0.12212807089090347,
|
|
"num_tokens": 2090732.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.3433879852294925,
|
|
"epoch": 0.9365203761755486,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996973904560135,
|
|
"loss": 6.2735,
|
|
"mean_token_accuracy": 0.12710270062088966,
|
|
"num_tokens": 2099352.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.382493019104004,
|
|
"epoch": 0.9404388714733543,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004996815947426317,
|
|
"loss": 6.2946,
|
|
"mean_token_accuracy": 0.1185976468026638,
|
|
"num_tokens": 2107365.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.328860521316528,
|
|
"epoch": 0.9443573667711599,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996653974868097,
|
|
"loss": 6.1709,
|
|
"mean_token_accuracy": 0.12477589845657348,
|
|
"num_tokens": 2116192.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.279233121871949,
|
|
"epoch": 0.9482758620689655,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996487987174946,
|
|
"loss": 6.2726,
|
|
"mean_token_accuracy": 0.11757147014141082,
|
|
"num_tokens": 2124410.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.326535558700561,
|
|
"epoch": 0.9521943573667712,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004996317984643512,
|
|
"loss": 6.2848,
|
|
"mean_token_accuracy": 0.12036795765161515,
|
|
"num_tokens": 2132799.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.467973375320435,
|
|
"epoch": 0.9561128526645768,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996143967577615,
|
|
"loss": 6.4498,
|
|
"mean_token_accuracy": 0.11489988937973976,
|
|
"num_tokens": 2141589.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.386198425292969,
|
|
"epoch": 0.9600313479623824,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004995965936288254,
|
|
"loss": 6.3403,
|
|
"mean_token_accuracy": 0.12025970742106437,
|
|
"num_tokens": 2150704.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.3113349914550785,
|
|
"epoch": 0.9639498432601881,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004995783891093597,
|
|
"loss": 6.2607,
|
|
"mean_token_accuracy": 0.12228266969323158,
|
|
"num_tokens": 2159587.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.234415531158447,
|
|
"epoch": 0.9678683385579937,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499559783231899,
|
|
"loss": 6.2409,
|
|
"mean_token_accuracy": 0.11728857606649398,
|
|
"num_tokens": 2169071.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.3988946914672855,
|
|
"epoch": 0.9717868338557993,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995407760296952,
|
|
"loss": 6.3251,
|
|
"mean_token_accuracy": 0.11511705070734024,
|
|
"num_tokens": 2178248.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.396841096878052,
|
|
"epoch": 0.975705329153605,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995213675367169,
|
|
"loss": 6.229,
|
|
"mean_token_accuracy": 0.11794086545705795,
|
|
"num_tokens": 2187520.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.2435925006866455,
|
|
"epoch": 0.9796238244514106,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995015577876504,
|
|
"loss": 6.3007,
|
|
"mean_token_accuracy": 0.11987854838371277,
|
|
"num_tokens": 2196861.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.314880323410034,
|
|
"epoch": 0.9835423197492164,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499481346817899,
|
|
"loss": 6.2143,
|
|
"mean_token_accuracy": 0.12193886116147042,
|
|
"num_tokens": 2205523.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.318439245223999,
|
|
"epoch": 0.987460815047022,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004994607346635829,
|
|
"loss": 6.1652,
|
|
"mean_token_accuracy": 0.12974872291088105,
|
|
"num_tokens": 2214291.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.366824340820313,
|
|
"epoch": 0.9913793103448276,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004994397213615394,
|
|
"loss": 6.2818,
|
|
"mean_token_accuracy": 0.12176446989178658,
|
|
"num_tokens": 2223093.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.331199264526367,
|
|
"epoch": 0.9952978056426333,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004994183069493225,
|
|
"loss": 6.2364,
|
|
"mean_token_accuracy": 0.11954401284456254,
|
|
"num_tokens": 2231551.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.184268236160278,
|
|
"epoch": 0.9992163009404389,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004993964914652034,
|
|
"loss": 6.2306,
|
|
"mean_token_accuracy": 0.12992196902632713,
|
|
"num_tokens": 2240650.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.461467027664185,
|
|
"epoch": 1.0031347962382444,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004993742749481699,
|
|
"loss": 6.1323,
|
|
"mean_token_accuracy": 0.12565431445837022,
|
|
"num_tokens": 2249392.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.129501914978027,
|
|
"epoch": 1.0070532915360502,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004993516574379265,
|
|
"loss": 5.9558,
|
|
"mean_token_accuracy": 0.13117464035749435,
|
|
"num_tokens": 2258100.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.233500862121582,
|
|
"epoch": 1.0109717868338557,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004993286389748943,
|
|
"loss": 6.0332,
|
|
"mean_token_accuracy": 0.12863276079297065,
|
|
"num_tokens": 2266661.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.204838132858276,
|
|
"epoch": 1.0148902821316614,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004993052196002112,
|
|
"loss": 6.1322,
|
|
"mean_token_accuracy": 0.12399615719914436,
|
|
"num_tokens": 2275475.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.244740676879883,
|
|
"epoch": 1.0188087774294672,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004992813993557312,
|
|
"loss": 6.0349,
|
|
"mean_token_accuracy": 0.13146774247288703,
|
|
"num_tokens": 2284077.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.1790135383605955,
|
|
"epoch": 1.0227272727272727,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499257178284025,
|
|
"loss": 6.0488,
|
|
"mean_token_accuracy": 0.12352109551429749,
|
|
"num_tokens": 2293029.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.212559843063355,
|
|
"epoch": 1.0266457680250785,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004992325564283797,
|
|
"loss": 5.9596,
|
|
"mean_token_accuracy": 0.14028860479593278,
|
|
"num_tokens": 2301574.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.061412286758423,
|
|
"epoch": 1.030564263322884,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004992075338327984,
|
|
"loss": 5.9615,
|
|
"mean_token_accuracy": 0.13310449346899986,
|
|
"num_tokens": 2310366.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.29543924331665,
|
|
"epoch": 1.0344827586206897,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004991821105420006,
|
|
"loss": 6.0925,
|
|
"mean_token_accuracy": 0.12402413934469222,
|
|
"num_tokens": 2318943.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.23583607673645,
|
|
"epoch": 1.0384012539184952,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004991562866014219,
|
|
"loss": 6.0991,
|
|
"mean_token_accuracy": 0.1307850494980812,
|
|
"num_tokens": 2327173.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.260228252410888,
|
|
"epoch": 1.042319749216301,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.0618,
|
|
"mean_token_accuracy": 0.12725646793842316,
|
|
"num_tokens": 2336194.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.191347360610962,
|
|
"epoch": 1.0462382445141065,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004991034369562438,
|
|
"loss": 6.1144,
|
|
"mean_token_accuracy": 0.12965716123580934,
|
|
"num_tokens": 2344721.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.197571229934693,
|
|
"epoch": 1.0501567398119123,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004990764113460953,
|
|
"loss": 6.0465,
|
|
"mean_token_accuracy": 0.12880703136324884,
|
|
"num_tokens": 2353406.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.303739309310913,
|
|
"epoch": 1.0540752351097178,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004990489852750675,
|
|
"loss": 6.2162,
|
|
"mean_token_accuracy": 0.11973949447274208,
|
|
"num_tokens": 2361896.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.172640943527222,
|
|
"epoch": 1.0579937304075235,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004990211587921751,
|
|
"loss": 6.0466,
|
|
"mean_token_accuracy": 0.12820285186171532,
|
|
"num_tokens": 2371861.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.191204833984375,
|
|
"epoch": 1.061912225705329,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004989929319471487,
|
|
"loss": 5.9731,
|
|
"mean_token_accuracy": 0.1299687497317791,
|
|
"num_tokens": 2381555.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.136105632781982,
|
|
"epoch": 1.0658307210031348,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004989643047904341,
|
|
"loss": 5.9618,
|
|
"mean_token_accuracy": 0.13669807985424995,
|
|
"num_tokens": 2389851.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.0609746932983395,
|
|
"epoch": 1.0697492163009406,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004989352773731928,
|
|
"loss": 5.8998,
|
|
"mean_token_accuracy": 0.13646639958024026,
|
|
"num_tokens": 2397764.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.258017063140869,
|
|
"epoch": 1.073667711598746,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004989058497473014,
|
|
"loss": 6.0497,
|
|
"mean_token_accuracy": 0.13653347492218018,
|
|
"num_tokens": 2406688.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.219746732711792,
|
|
"epoch": 1.0775862068965518,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004988760219653518,
|
|
"loss": 6.1178,
|
|
"mean_token_accuracy": 0.12561060190200807,
|
|
"num_tokens": 2415475.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.158693790435791,
|
|
"epoch": 1.0815047021943573,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004988457940806513,
|
|
"loss": 6.0725,
|
|
"mean_token_accuracy": 0.12764996886253357,
|
|
"num_tokens": 2425719.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.134065866470337,
|
|
"epoch": 1.085423197492163,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004988151661472218,
|
|
"loss": 5.9679,
|
|
"mean_token_accuracy": 0.13455401584506035,
|
|
"num_tokens": 2435138.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.214557266235351,
|
|
"epoch": 1.0893416927899686,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004987841382198006,
|
|
"loss": 6.1148,
|
|
"mean_token_accuracy": 0.13025601208209991,
|
|
"num_tokens": 2443976.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.058465433120728,
|
|
"epoch": 1.0932601880877744,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987527103538394,
|
|
"loss": 6.001,
|
|
"mean_token_accuracy": 0.1352849192917347,
|
|
"num_tokens": 2453287.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.232900381088257,
|
|
"epoch": 1.09717868338558,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000498720882605505,
|
|
"loss": 6.0947,
|
|
"mean_token_accuracy": 0.1259259968996048,
|
|
"num_tokens": 2462054.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.098091125488281,
|
|
"epoch": 1.1010971786833856,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004986886550316788,
|
|
"loss": 5.9975,
|
|
"mean_token_accuracy": 0.13868628442287445,
|
|
"num_tokens": 2470333.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.151032543182373,
|
|
"epoch": 1.1050156739811912,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004986560276899565,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.13234539851546287,
|
|
"num_tokens": 2479621.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.181175947189331,
|
|
"epoch": 1.108934169278997,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004986230006386488,
|
|
"loss": 5.9972,
|
|
"mean_token_accuracy": 0.1349916376173496,
|
|
"num_tokens": 2488020.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.213946294784546,
|
|
"epoch": 1.1128526645768024,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004985895739367802,
|
|
"loss": 6.0292,
|
|
"mean_token_accuracy": 0.13072321489453315,
|
|
"num_tokens": 2497779.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.117242479324341,
|
|
"epoch": 1.1167711598746082,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004985557476440895,
|
|
"loss": 5.9767,
|
|
"mean_token_accuracy": 0.13357633352279663,
|
|
"num_tokens": 2506791.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.191507816314697,
|
|
"epoch": 1.1206896551724137,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00049852152182103,
|
|
"loss": 6.0892,
|
|
"mean_token_accuracy": 0.12935334220528602,
|
|
"num_tokens": 2515576.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.180032730102539,
|
|
"epoch": 1.1246081504702194,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004984868965287686,
|
|
"loss": 5.9955,
|
|
"mean_token_accuracy": 0.13097495809197426,
|
|
"num_tokens": 2523933.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.124348115921021,
|
|
"epoch": 1.1285266457680252,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004984518718291864,
|
|
"loss": 5.9525,
|
|
"mean_token_accuracy": 0.12772755995392798,
|
|
"num_tokens": 2532453.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.094198179244995,
|
|
"epoch": 1.1324451410658307,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004984164477848783,
|
|
"loss": 6.0043,
|
|
"mean_token_accuracy": 0.1302856981754303,
|
|
"num_tokens": 2541149.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.125786781311035,
|
|
"epoch": 1.1363636363636362,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004983806244591528,
|
|
"loss": 5.9822,
|
|
"mean_token_accuracy": 0.13764647617936135,
|
|
"num_tokens": 2550302.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.0180809020996096,
|
|
"epoch": 1.140282131661442,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004983444019160318,
|
|
"loss": 5.8351,
|
|
"mean_token_accuracy": 0.1406431756913662,
|
|
"num_tokens": 2558507.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.208771991729736,
|
|
"epoch": 1.1442006269592477,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004983077802202511,
|
|
"loss": 6.1261,
|
|
"mean_token_accuracy": 0.13089546412229539,
|
|
"num_tokens": 2567487.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.145636320114136,
|
|
"epoch": 1.1481191222570533,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004982707594372595,
|
|
"loss": 5.9673,
|
|
"mean_token_accuracy": 0.13152600824832916,
|
|
"num_tokens": 2576232.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.04912371635437,
|
|
"epoch": 1.152037617554859,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000498233339633219,
|
|
"loss": 5.977,
|
|
"mean_token_accuracy": 0.13370415195822716,
|
|
"num_tokens": 2585081.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.064982080459595,
|
|
"epoch": 1.1559561128526645,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000498195520875005,
|
|
"loss": 5.9501,
|
|
"mean_token_accuracy": 0.13194083124399186,
|
|
"num_tokens": 2593890.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.1344743251800535,
|
|
"epoch": 1.1598746081504703,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004981573032302059,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.12466074079275132,
|
|
"num_tokens": 2602011.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.061652421951294,
|
|
"epoch": 1.1637931034482758,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004981186867671225,
|
|
"loss": 5.9258,
|
|
"mean_token_accuracy": 0.13856346756219864,
|
|
"num_tokens": 2610366.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.042865514755249,
|
|
"epoch": 1.1677115987460815,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004980796715547691,
|
|
"loss": 5.946,
|
|
"mean_token_accuracy": 0.13904307335615157,
|
|
"num_tokens": 2619737.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.136406183242798,
|
|
"epoch": 1.171630094043887,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004980402576628717,
|
|
"loss": 5.9699,
|
|
"mean_token_accuracy": 0.1325167238712311,
|
|
"num_tokens": 2628176.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.0512768745422365,
|
|
"epoch": 1.1755485893416928,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004980004451618697,
|
|
"loss": 5.9832,
|
|
"mean_token_accuracy": 0.13477908223867416,
|
|
"num_tokens": 2636195.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.1755485893416928,
|
|
"eval_entropy": 6.076499396516371,
|
|
"eval_loss": 6.271363258361816,
|
|
"eval_mean_token_accuracy": 0.12794691480176393,
|
|
"eval_num_tokens": 2636195.0,
|
|
"eval_runtime": 2.8465,
|
|
"eval_samples_per_second": 1448.09,
|
|
"eval_steps_per_second": 181.275,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.207622623443603,
|
|
"epoch": 1.1794670846394983,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004979602341229144,
|
|
"loss": 6.0515,
|
|
"mean_token_accuracy": 0.1303505390882492,
|
|
"num_tokens": 2645035.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.107669830322266,
|
|
"epoch": 1.183385579937304,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004979196246178694,
|
|
"loss": 6.0111,
|
|
"mean_token_accuracy": 0.1320280283689499,
|
|
"num_tokens": 2653239.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.064078998565674,
|
|
"epoch": 1.1873040752351098,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004978786167193105,
|
|
"loss": 6.0243,
|
|
"mean_token_accuracy": 0.1286700151860714,
|
|
"num_tokens": 2662532.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.17049560546875,
|
|
"epoch": 1.1912225705329154,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004978372105005254,
|
|
"loss": 6.0272,
|
|
"mean_token_accuracy": 0.13095418065786363,
|
|
"num_tokens": 2671867.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.024587631225586,
|
|
"epoch": 1.1951410658307209,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004977954060355139,
|
|
"loss": 5.9538,
|
|
"mean_token_accuracy": 0.13054575994610787,
|
|
"num_tokens": 2680693.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.192902612686157,
|
|
"epoch": 1.1990595611285266,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004977532033989871,
|
|
"loss": 6.0462,
|
|
"mean_token_accuracy": 0.13150823041796683,
|
|
"num_tokens": 2689228.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.0371331691741945,
|
|
"epoch": 1.2029780564263324,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004977106026663681,
|
|
"loss": 5.9103,
|
|
"mean_token_accuracy": 0.13035874292254448,
|
|
"num_tokens": 2698483.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.142786264419556,
|
|
"epoch": 1.206896551724138,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004976676039137914,
|
|
"loss": 6.0491,
|
|
"mean_token_accuracy": 0.13021435514092444,
|
|
"num_tokens": 2706920.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.042557334899902,
|
|
"epoch": 1.2108150470219436,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004976242072181026,
|
|
"loss": 5.9328,
|
|
"mean_token_accuracy": 0.13973603770136833,
|
|
"num_tokens": 2715954.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.062022352218628,
|
|
"epoch": 1.2147335423197492,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004975804126568587,
|
|
"loss": 5.9571,
|
|
"mean_token_accuracy": 0.1364055834710598,
|
|
"num_tokens": 2724890.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.17275595664978,
|
|
"epoch": 1.218652037617555,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004975362203083277,
|
|
"loss": 5.9721,
|
|
"mean_token_accuracy": 0.13264910951256753,
|
|
"num_tokens": 2733688.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.082532548904419,
|
|
"epoch": 1.2225705329153604,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004974916302514886,
|
|
"loss": 5.979,
|
|
"mean_token_accuracy": 0.13056719824671745,
|
|
"num_tokens": 2743364.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.061155843734741,
|
|
"epoch": 1.2264890282131662,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.0256,
|
|
"mean_token_accuracy": 0.13040423914790153,
|
|
"num_tokens": 2754276.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.098849248886109,
|
|
"epoch": 1.2304075235109717,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004974012573323545,
|
|
"loss": 5.8312,
|
|
"mean_token_accuracy": 0.1432400055229664,
|
|
"num_tokens": 2762680.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.957777738571167,
|
|
"epoch": 1.2343260188087775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004973554746315709,
|
|
"loss": 5.9289,
|
|
"mean_token_accuracy": 0.14215950816869735,
|
|
"num_tokens": 2771161.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.126145744323731,
|
|
"epoch": 1.238244514106583,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004973092945455009,
|
|
"loss": 5.9042,
|
|
"mean_token_accuracy": 0.141022390127182,
|
|
"num_tokens": 2779321.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.973110008239746,
|
|
"epoch": 1.2421630094043887,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004972627171566757,
|
|
"loss": 6.0625,
|
|
"mean_token_accuracy": 0.13462768569588662,
|
|
"num_tokens": 2788425.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.077723455429077,
|
|
"epoch": 1.2460815047021945,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004972157425483368,
|
|
"loss": 5.9129,
|
|
"mean_token_accuracy": 0.13926904946565627,
|
|
"num_tokens": 2796747.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.98823504447937,
|
|
"epoch": 1.25,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004971683708044353,
|
|
"loss": 5.9614,
|
|
"mean_token_accuracy": 0.1371827580034733,
|
|
"num_tokens": 2805335.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.1075457572937015,
|
|
"epoch": 1.2539184952978055,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004971206020096323,
|
|
"loss": 5.8454,
|
|
"mean_token_accuracy": 0.14424244835972785,
|
|
"num_tokens": 2813350.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.131724071502686,
|
|
"epoch": 1.2578369905956113,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004970724362492984,
|
|
"loss": 6.0353,
|
|
"mean_token_accuracy": 0.1399139069020748,
|
|
"num_tokens": 2821922.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.015088748931885,
|
|
"epoch": 1.261755485893417,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004970238736095135,
|
|
"loss": 5.952,
|
|
"mean_token_accuracy": 0.13358094021677971,
|
|
"num_tokens": 2830832.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.085702276229858,
|
|
"epoch": 1.2656739811912225,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004969749141770671,
|
|
"loss": 5.9374,
|
|
"mean_token_accuracy": 0.13813378065824508,
|
|
"num_tokens": 2839699.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.056669616699219,
|
|
"epoch": 1.2695924764890283,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004969255580394575,
|
|
"loss": 6.0123,
|
|
"mean_token_accuracy": 0.13840878382325172,
|
|
"num_tokens": 2848293.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.114861154556275,
|
|
"epoch": 1.2735109717868338,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000496875805284892,
|
|
"loss": 5.9607,
|
|
"mean_token_accuracy": 0.13838078081607819,
|
|
"num_tokens": 2856468.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.044304227828979,
|
|
"epoch": 1.2774294670846396,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004968256560022871,
|
|
"loss": 5.9018,
|
|
"mean_token_accuracy": 0.13837780505418779,
|
|
"num_tokens": 2865366.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.9802772998809814,
|
|
"epoch": 1.281347962382445,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004967751102812676,
|
|
"loss": 5.9539,
|
|
"mean_token_accuracy": 0.14084723740816116,
|
|
"num_tokens": 2873520.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.120439767837524,
|
|
"epoch": 1.2852664576802508,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004967241682121669,
|
|
"loss": 6.0169,
|
|
"mean_token_accuracy": 0.13527923449873924,
|
|
"num_tokens": 2883324.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.048051452636718,
|
|
"epoch": 1.2891849529780564,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004966728298860267,
|
|
"loss": 5.948,
|
|
"mean_token_accuracy": 0.13382584750652313,
|
|
"num_tokens": 2892200.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.069919157028198,
|
|
"epoch": 1.293103448275862,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004966210953945969,
|
|
"loss": 5.9192,
|
|
"mean_token_accuracy": 0.13777885362505912,
|
|
"num_tokens": 2900597.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.0597516059875485,
|
|
"epoch": 1.2970219435736676,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004965689648303355,
|
|
"loss": 5.9559,
|
|
"mean_token_accuracy": 0.13197760209441184,
|
|
"num_tokens": 2909869.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.956328582763672,
|
|
"epoch": 1.3009404388714734,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004965164382864083,
|
|
"loss": 5.9761,
|
|
"mean_token_accuracy": 0.13753246441483497,
|
|
"num_tokens": 2919166.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.117870473861695,
|
|
"epoch": 1.3048589341692791,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004964635158566886,
|
|
"loss": 5.9859,
|
|
"mean_token_accuracy": 0.13755642771720886,
|
|
"num_tokens": 2927884.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.016185617446899,
|
|
"epoch": 1.3087774294670846,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004964101976357574,
|
|
"loss": 5.915,
|
|
"mean_token_accuracy": 0.14404550418257714,
|
|
"num_tokens": 2936737.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.071380281448365,
|
|
"epoch": 1.3126959247648902,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000496356483718903,
|
|
"loss": 6.0118,
|
|
"mean_token_accuracy": 0.13691077008843422,
|
|
"num_tokens": 2944203.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.0966150760650635,
|
|
"epoch": 1.316614420062696,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004963023742021211,
|
|
"loss": 6.025,
|
|
"mean_token_accuracy": 0.13278019577264785,
|
|
"num_tokens": 2952639.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.0501056671142575,
|
|
"epoch": 1.3205329153605017,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000496247869182114,
|
|
"loss": 5.9512,
|
|
"mean_token_accuracy": 0.13901304081082344,
|
|
"num_tokens": 2961482.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.046927785873413,
|
|
"epoch": 1.3244514106583072,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004961929687562909,
|
|
"loss": 5.9634,
|
|
"mean_token_accuracy": 0.13337364569306373,
|
|
"num_tokens": 2970927.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.964669847488404,
|
|
"epoch": 1.328369905956113,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004961376730227681,
|
|
"loss": 5.9123,
|
|
"mean_token_accuracy": 0.1409181497991085,
|
|
"num_tokens": 2979329.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.0310773849487305,
|
|
"epoch": 1.3322884012539185,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004960819820803675,
|
|
"loss": 5.8666,
|
|
"mean_token_accuracy": 0.14356955140829086,
|
|
"num_tokens": 2986894.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.004372310638428,
|
|
"epoch": 1.3362068965517242,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004960258960286185,
|
|
"loss": 5.8687,
|
|
"mean_token_accuracy": 0.139854796230793,
|
|
"num_tokens": 2994721.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.086730909347534,
|
|
"epoch": 1.3401253918495297,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004959694149677553,
|
|
"loss": 6.0317,
|
|
"mean_token_accuracy": 0.12823668047785758,
|
|
"num_tokens": 3003165.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.0770214080810545,
|
|
"epoch": 1.3440438871473355,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004959125389987193,
|
|
"loss": 6.008,
|
|
"mean_token_accuracy": 0.13203698322176932,
|
|
"num_tokens": 3012178.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.01055006980896,
|
|
"epoch": 1.347962382445141,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004958552682231567,
|
|
"loss": 5.9027,
|
|
"mean_token_accuracy": 0.14051611423492433,
|
|
"num_tokens": 3021685.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.005949354171753,
|
|
"epoch": 1.3518808777429467,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004957976027434199,
|
|
"loss": 5.956,
|
|
"mean_token_accuracy": 0.12991197034716606,
|
|
"num_tokens": 3030667.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.097392654418945,
|
|
"epoch": 1.3557993730407523,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004957395426625663,
|
|
"loss": 6.0049,
|
|
"mean_token_accuracy": 0.1332964815199375,
|
|
"num_tokens": 3039501.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.981141185760498,
|
|
"epoch": 1.359717868338558,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004956810880843587,
|
|
"loss": 5.9382,
|
|
"mean_token_accuracy": 0.13364407643675805,
|
|
"num_tokens": 3048547.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.120581865310669,
|
|
"epoch": 1.3636363636363638,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000495622239113265,
|
|
"loss": 5.9044,
|
|
"mean_token_accuracy": 0.14041278436779975,
|
|
"num_tokens": 3056521.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.939314699172973,
|
|
"epoch": 1.3675548589341693,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004955629958544577,
|
|
"loss": 5.8949,
|
|
"mean_token_accuracy": 0.13747348114848137,
|
|
"num_tokens": 3066007.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.10064115524292,
|
|
"epoch": 1.3714733542319748,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004955033584138143,
|
|
"loss": 6.0698,
|
|
"mean_token_accuracy": 0.12523617148399352,
|
|
"num_tokens": 3075111.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.022913122177124,
|
|
"epoch": 1.3753918495297806,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004954433268979164,
|
|
"loss": 5.8818,
|
|
"mean_token_accuracy": 0.14991883113980292,
|
|
"num_tokens": 3083766.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.973852968215942,
|
|
"epoch": 1.3793103448275863,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004953829014140502,
|
|
"loss": 5.9353,
|
|
"mean_token_accuracy": 0.13615802749991418,
|
|
"num_tokens": 3092103.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.990107297897339,
|
|
"epoch": 1.3832288401253918,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004953220820702057,
|
|
"loss": 5.9593,
|
|
"mean_token_accuracy": 0.13366186618804932,
|
|
"num_tokens": 3101286.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.151833629608154,
|
|
"epoch": 1.3871473354231976,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004952608689750771,
|
|
"loss": 5.9611,
|
|
"mean_token_accuracy": 0.13463746160268783,
|
|
"num_tokens": 3110554.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.00332202911377,
|
|
"epoch": 1.391065830721003,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004951992622380619,
|
|
"loss": 5.8163,
|
|
"mean_token_accuracy": 0.14524587914347648,
|
|
"num_tokens": 3119265.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.971736478805542,
|
|
"epoch": 1.3949843260188088,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004951372619692615,
|
|
"loss": 5.8515,
|
|
"mean_token_accuracy": 0.14273120015859603,
|
|
"num_tokens": 3127962.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.01099796295166,
|
|
"epoch": 1.3989028213166144,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004950748682794804,
|
|
"loss": 5.9917,
|
|
"mean_token_accuracy": 0.13492952212691306,
|
|
"num_tokens": 3137265.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.9283558368682865,
|
|
"epoch": 1.40282131661442,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004950120812802262,
|
|
"loss": 5.8128,
|
|
"mean_token_accuracy": 0.1431027092039585,
|
|
"num_tokens": 3146458.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.07517991065979,
|
|
"epoch": 1.4067398119122256,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004949489010837095,
|
|
"loss": 5.9588,
|
|
"mean_token_accuracy": 0.13590797632932664,
|
|
"num_tokens": 3155557.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.9258420944213865,
|
|
"epoch": 1.4106583072100314,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 5.8368,
|
|
"mean_token_accuracy": 0.14588673859834672,
|
|
"num_tokens": 3163478.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.979496049880981,
|
|
"epoch": 1.414576802507837,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000494821361551244,
|
|
"loss": 5.9323,
|
|
"mean_token_accuracy": 0.1362486593425274,
|
|
"num_tokens": 3172632.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.013187265396118,
|
|
"epoch": 1.4184952978056427,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004947570024432291,
|
|
"loss": 5.9629,
|
|
"mean_token_accuracy": 0.14079801365733147,
|
|
"num_tokens": 3181325.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.030901575088501,
|
|
"epoch": 1.4224137931034484,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004946922505938189,
|
|
"loss": 6.0188,
|
|
"mean_token_accuracy": 0.14003786072134972,
|
|
"num_tokens": 3189783.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.039680242538452,
|
|
"epoch": 1.426332288401254,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004946271061187354,
|
|
"loss": 5.8608,
|
|
"mean_token_accuracy": 0.13449943587183952,
|
|
"num_tokens": 3198605.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.975720500946045,
|
|
"epoch": 1.4302507836990594,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004945615691344025,
|
|
"loss": 5.9606,
|
|
"mean_token_accuracy": 0.12840349078178406,
|
|
"num_tokens": 3208061.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.996256589889526,
|
|
"epoch": 1.4341692789968652,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004944956397579453,
|
|
"loss": 5.865,
|
|
"mean_token_accuracy": 0.14466958418488501,
|
|
"num_tokens": 3216093.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.9624964714050295,
|
|
"epoch": 1.438087774294671,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004944293181071902,
|
|
"loss": 5.8584,
|
|
"mean_token_accuracy": 0.14238173440098761,
|
|
"num_tokens": 3224446.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.023101758956909,
|
|
"epoch": 1.4420062695924765,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004943626043006649,
|
|
"loss": 5.9836,
|
|
"mean_token_accuracy": 0.1356622613966465,
|
|
"num_tokens": 3233398.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.034554052352905,
|
|
"epoch": 1.4459247648902822,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000494295498457598,
|
|
"loss": 5.9682,
|
|
"mean_token_accuracy": 0.13090459257364273,
|
|
"num_tokens": 3243391.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.992100238800049,
|
|
"epoch": 1.4498432601880877,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004942280006979179,
|
|
"loss": 5.9107,
|
|
"mean_token_accuracy": 0.1431814417243004,
|
|
"num_tokens": 3251954.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.005795526504516,
|
|
"epoch": 1.4537617554858935,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004941601111422546,
|
|
"loss": 5.9356,
|
|
"mean_token_accuracy": 0.13904605209827423,
|
|
"num_tokens": 3261121.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.875429582595825,
|
|
"epoch": 1.457680250783699,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004940918299119375,
|
|
"loss": 5.851,
|
|
"mean_token_accuracy": 0.14492034763097764,
|
|
"num_tokens": 3270179.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.878690958023071,
|
|
"epoch": 1.4615987460815048,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004940231571289962,
|
|
"loss": 5.8587,
|
|
"mean_token_accuracy": 0.13866079300642015,
|
|
"num_tokens": 3279749.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.03740701675415,
|
|
"epoch": 1.4655172413793103,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004939540929161603,
|
|
"loss": 5.8712,
|
|
"mean_token_accuracy": 0.14061601012945174,
|
|
"num_tokens": 3288458.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.949235296249389,
|
|
"epoch": 1.469435736677116,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004938846373968586,
|
|
"loss": 5.8497,
|
|
"mean_token_accuracy": 0.14182733744382858,
|
|
"num_tokens": 3297231.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.013594055175782,
|
|
"epoch": 1.4733542319749215,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004938147906952194,
|
|
"loss": 6.0087,
|
|
"mean_token_accuracy": 0.1348317749798298,
|
|
"num_tokens": 3306915.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.975127410888672,
|
|
"epoch": 1.4772727272727273,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00049374455293607,
|
|
"loss": 5.9349,
|
|
"mean_token_accuracy": 0.13975367248058318,
|
|
"num_tokens": 3315985.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.011495923995971,
|
|
"epoch": 1.481191222570533,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004936739242449369,
|
|
"loss": 5.9395,
|
|
"mean_token_accuracy": 0.13877131789922714,
|
|
"num_tokens": 3324924.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.972328424453735,
|
|
"epoch": 1.4851097178683386,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004936029047480447,
|
|
"loss": 5.923,
|
|
"mean_token_accuracy": 0.1383238084614277,
|
|
"num_tokens": 3333464.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.008746147155762,
|
|
"epoch": 1.489028213166144,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004935314945723171,
|
|
"loss": 5.8358,
|
|
"mean_token_accuracy": 0.14501626715064048,
|
|
"num_tokens": 3342441.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.930044031143188,
|
|
"epoch": 1.4929467084639498,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004934596938453754,
|
|
"loss": 5.8856,
|
|
"mean_token_accuracy": 0.1392125241458416,
|
|
"num_tokens": 3351338.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.987050580978393,
|
|
"epoch": 1.4968652037617556,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004933875026955391,
|
|
"loss": 5.8414,
|
|
"mean_token_accuracy": 0.13805665150284768,
|
|
"num_tokens": 3360356.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.016159772872925,
|
|
"epoch": 1.500783699059561,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004933149212518258,
|
|
"loss": 5.933,
|
|
"mean_token_accuracy": 0.13254097029566764,
|
|
"num_tokens": 3369485.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.981318569183349,
|
|
"epoch": 1.5047021943573666,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004932419496439501,
|
|
"loss": 5.884,
|
|
"mean_token_accuracy": 0.14330490529537201,
|
|
"num_tokens": 3378355.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.914515161514283,
|
|
"epoch": 1.5086206896551724,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000493168588002324,
|
|
"loss": 5.8032,
|
|
"mean_token_accuracy": 0.14702262133359909,
|
|
"num_tokens": 3386729.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.070053672790527,
|
|
"epoch": 1.5125391849529781,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004930948364580569,
|
|
"loss": 6.0245,
|
|
"mean_token_accuracy": 0.13127490803599356,
|
|
"num_tokens": 3395822.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.980561304092407,
|
|
"epoch": 1.5164576802507836,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004930206951429546,
|
|
"loss": 5.9416,
|
|
"mean_token_accuracy": 0.13797413259744645,
|
|
"num_tokens": 3404671.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.985497379302979,
|
|
"epoch": 1.5203761755485894,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004929461641895197,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.14024683982133865,
|
|
"num_tokens": 3413289.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.936700534820557,
|
|
"epoch": 1.5242946708463951,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000492871243730951,
|
|
"loss": 5.917,
|
|
"mean_token_accuracy": 0.13965860977768899,
|
|
"num_tokens": 3421960.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.997755765914917,
|
|
"epoch": 1.5282131661442007,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004927959339011437,
|
|
"loss": 5.9149,
|
|
"mean_token_accuracy": 0.1362414240837097,
|
|
"num_tokens": 3431958.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.822151041030883,
|
|
"epoch": 1.5321316614420062,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004927202348346885,
|
|
"loss": 5.7435,
|
|
"mean_token_accuracy": 0.1449103720486164,
|
|
"num_tokens": 3441438.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.006295967102051,
|
|
"epoch": 1.536050156739812,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000492644146666872,
|
|
"loss": 5.8473,
|
|
"mean_token_accuracy": 0.1476273812353611,
|
|
"num_tokens": 3450176.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.828231477737427,
|
|
"epoch": 1.5399686520376177,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004925676695336761,
|
|
"loss": 5.8462,
|
|
"mean_token_accuracy": 0.14223939031362534,
|
|
"num_tokens": 3458685.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.036013793945313,
|
|
"epoch": 1.5438871473354232,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004924908035717777,
|
|
"loss": 5.9932,
|
|
"mean_token_accuracy": 0.1358640395104885,
|
|
"num_tokens": 3468362.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.99300446510315,
|
|
"epoch": 1.5478056426332287,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000492413548918549,
|
|
"loss": 5.9158,
|
|
"mean_token_accuracy": 0.13908401131629944,
|
|
"num_tokens": 3476791.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.858194875717163,
|
|
"epoch": 1.5517241379310345,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004923359057120563,
|
|
"loss": 5.8306,
|
|
"mean_token_accuracy": 0.1353352040052414,
|
|
"num_tokens": 3486486.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.9135054588317875,
|
|
"epoch": 1.5556426332288402,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004922578740910608,
|
|
"loss": 5.7897,
|
|
"mean_token_accuracy": 0.147303768992424,
|
|
"num_tokens": 3496145.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.895607662200928,
|
|
"epoch": 1.5595611285266457,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004921794541950177,
|
|
"loss": 5.7769,
|
|
"mean_token_accuracy": 0.1445461630821228,
|
|
"num_tokens": 3504644.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.8936514377594,
|
|
"epoch": 1.5634796238244513,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004921006461640758,
|
|
"loss": 5.8104,
|
|
"mean_token_accuracy": 0.13581008464097977,
|
|
"num_tokens": 3513126.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.810939693450928,
|
|
"epoch": 1.567398119122257,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000492021450139078,
|
|
"loss": 5.749,
|
|
"mean_token_accuracy": 0.14536509588360785,
|
|
"num_tokens": 3521505.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.567398119122257,
|
|
"eval_entropy": 5.800363567448402,
|
|
"eval_loss": 6.0877366065979,
|
|
"eval_mean_token_accuracy": 0.13831741765661296,
|
|
"eval_num_tokens": 3521505.0,
|
|
"eval_runtime": 2.8597,
|
|
"eval_samples_per_second": 1441.409,
|
|
"eval_steps_per_second": 180.438,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.01237120628357,
|
|
"epoch": 1.5713166144200628,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004919418662615605,
|
|
"loss": 5.9987,
|
|
"mean_token_accuracy": 0.13824992179870604,
|
|
"num_tokens": 3530008.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.025429964065552,
|
|
"epoch": 1.5752351097178683,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004918618946737525,
|
|
"loss": 5.8827,
|
|
"mean_token_accuracy": 0.1387758158147335,
|
|
"num_tokens": 3538778.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.967436027526856,
|
|
"epoch": 1.579153605015674,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004917815355185762,
|
|
"loss": 5.8638,
|
|
"mean_token_accuracy": 0.13804899901151657,
|
|
"num_tokens": 3547108.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.896082782745362,
|
|
"epoch": 1.5830721003134798,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004917007889396464,
|
|
"loss": 5.9114,
|
|
"mean_token_accuracy": 0.13484615311026574,
|
|
"num_tokens": 3556094.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.933627510070801,
|
|
"epoch": 1.5869905956112853,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004916196550812706,
|
|
"loss": 5.7628,
|
|
"mean_token_accuracy": 0.14669432789087294,
|
|
"num_tokens": 3564222.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.776369333267212,
|
|
"epoch": 1.5909090909090908,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004915381340884477,
|
|
"loss": 5.773,
|
|
"mean_token_accuracy": 0.14765530303120614,
|
|
"num_tokens": 3573123.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.964944887161255,
|
|
"epoch": 1.5948275862068966,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 5.8091,
|
|
"mean_token_accuracy": 0.14593618959188462,
|
|
"num_tokens": 3581462.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.8525710105896,
|
|
"epoch": 1.5987460815047023,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004913739312829181,
|
|
"loss": 5.8172,
|
|
"mean_token_accuracy": 0.14506494328379632,
|
|
"num_tokens": 3590718.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.894586277008057,
|
|
"epoch": 1.6026645768025078,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004912912497636683,
|
|
"loss": 5.8484,
|
|
"mean_token_accuracy": 0.13913847804069518,
|
|
"num_tokens": 3599890.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.029708480834961,
|
|
"epoch": 1.6065830721003134,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004912081816968853,
|
|
"loss": 5.9327,
|
|
"mean_token_accuracy": 0.13791020438075066,
|
|
"num_tokens": 3607927.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.966132402420044,
|
|
"epoch": 1.6105015673981191,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000491124727231025,
|
|
"loss": 5.7923,
|
|
"mean_token_accuracy": 0.147890043258667,
|
|
"num_tokens": 3615935.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.831100559234619,
|
|
"epoch": 1.6144200626959249,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004910408865152343,
|
|
"loss": 5.8251,
|
|
"mean_token_accuracy": 0.13951457142829896,
|
|
"num_tokens": 3624475.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.039763879776001,
|
|
"epoch": 1.6183385579937304,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004909566596993498,
|
|
"loss": 6.0219,
|
|
"mean_token_accuracy": 0.1320653535425663,
|
|
"num_tokens": 3633995.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.985185670852661,
|
|
"epoch": 1.622257053291536,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004908720469338988,
|
|
"loss": 5.8851,
|
|
"mean_token_accuracy": 0.14443439915776252,
|
|
"num_tokens": 3643121.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.913325405120849,
|
|
"epoch": 1.6261755485893417,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004907870483700979,
|
|
"loss": 5.8127,
|
|
"mean_token_accuracy": 0.14168807715177537,
|
|
"num_tokens": 3652285.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.871179533004761,
|
|
"epoch": 1.6300940438871474,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004907016641598534,
|
|
"loss": 5.8916,
|
|
"mean_token_accuracy": 0.13586366698145866,
|
|
"num_tokens": 3661441.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.870393753051758,
|
|
"epoch": 1.634012539184953,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004906158944557607,
|
|
"loss": 5.7592,
|
|
"mean_token_accuracy": 0.1525380626320839,
|
|
"num_tokens": 3669364.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.979267978668213,
|
|
"epoch": 1.6379310344827587,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000490529739411104,
|
|
"loss": 5.8639,
|
|
"mean_token_accuracy": 0.14333550184965133,
|
|
"num_tokens": 3677848.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.851955509185791,
|
|
"epoch": 1.6418495297805644,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004904431991798565,
|
|
"loss": 5.7564,
|
|
"mean_token_accuracy": 0.1478363499045372,
|
|
"num_tokens": 3686617.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.895144653320313,
|
|
"epoch": 1.64576802507837,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004903562739166797,
|
|
"loss": 5.7719,
|
|
"mean_token_accuracy": 0.14103165566921233,
|
|
"num_tokens": 3695106.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.809584045410157,
|
|
"epoch": 1.6496865203761755,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004902689637769229,
|
|
"loss": 5.7045,
|
|
"mean_token_accuracy": 0.1489320859313011,
|
|
"num_tokens": 3703167.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.91784520149231,
|
|
"epoch": 1.6536050156739812,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004901812689166237,
|
|
"loss": 5.8263,
|
|
"mean_token_accuracy": 0.1423416718840599,
|
|
"num_tokens": 3711597.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.876578712463379,
|
|
"epoch": 1.657523510971787,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004900931894925069,
|
|
"loss": 5.9171,
|
|
"mean_token_accuracy": 0.1454581469297409,
|
|
"num_tokens": 3720277.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.875493669509888,
|
|
"epoch": 1.6614420062695925,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004900047256619849,
|
|
"loss": 5.8751,
|
|
"mean_token_accuracy": 0.14267786517739295,
|
|
"num_tokens": 3729832.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.92410454750061,
|
|
"epoch": 1.665360501567398,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004899158775831566,
|
|
"loss": 5.8171,
|
|
"mean_token_accuracy": 0.1460487052798271,
|
|
"num_tokens": 3738390.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.860368490219116,
|
|
"epoch": 1.6692789968652038,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004898266454148081,
|
|
"loss": 5.8842,
|
|
"mean_token_accuracy": 0.14248219579458238,
|
|
"num_tokens": 3747046.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.8945173740386965,
|
|
"epoch": 1.6731974921630095,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004897370293164119,
|
|
"loss": 5.8355,
|
|
"mean_token_accuracy": 0.1434356912970543,
|
|
"num_tokens": 3755818.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.046916913986206,
|
|
"epoch": 1.677115987460815,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004896470294481262,
|
|
"loss": 5.9463,
|
|
"mean_token_accuracy": 0.14069818705320358,
|
|
"num_tokens": 3764123.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.899928379058838,
|
|
"epoch": 1.6810344827586206,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004895566459707954,
|
|
"loss": 6.0044,
|
|
"mean_token_accuracy": 0.13193708062171935,
|
|
"num_tokens": 3773105.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.964847612380981,
|
|
"epoch": 1.6849529780564263,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004894658790459498,
|
|
"loss": 5.8433,
|
|
"mean_token_accuracy": 0.1431967169046402,
|
|
"num_tokens": 3781417.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.018510866165161,
|
|
"epoch": 1.688871473354232,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004893747288358041,
|
|
"loss": 5.9208,
|
|
"mean_token_accuracy": 0.1428132861852646,
|
|
"num_tokens": 3789657.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.967458343505859,
|
|
"epoch": 1.6927899686520376,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000489283195503259,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.13716802671551703,
|
|
"num_tokens": 3798700.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.913800382614136,
|
|
"epoch": 1.6967084639498433,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000489191279211899,
|
|
"loss": 5.7975,
|
|
"mean_token_accuracy": 0.14700483977794648,
|
|
"num_tokens": 3807299.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.914860534667969,
|
|
"epoch": 1.700626959247649,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004890989801259935,
|
|
"loss": 5.8909,
|
|
"mean_token_accuracy": 0.14413824826478958,
|
|
"num_tokens": 3816292.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.9630790710449215,
|
|
"epoch": 1.7045454545454546,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000489006298410496,
|
|
"loss": 5.805,
|
|
"mean_token_accuracy": 0.140317203104496,
|
|
"num_tokens": 3825070.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.899825859069824,
|
|
"epoch": 1.70846394984326,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004889132342310438,
|
|
"loss": 5.7943,
|
|
"mean_token_accuracy": 0.14539300575852393,
|
|
"num_tokens": 3834016.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.92696852684021,
|
|
"epoch": 1.7123824451410659,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004888197877539577,
|
|
"loss": 5.8552,
|
|
"mean_token_accuracy": 0.14206420630216599,
|
|
"num_tokens": 3842172.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.870498418807983,
|
|
"epoch": 1.7163009404388716,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004887259591462417,
|
|
"loss": 5.7546,
|
|
"mean_token_accuracy": 0.14559593349695205,
|
|
"num_tokens": 3851109.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.859376382827759,
|
|
"epoch": 1.7202194357366771,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004886317485755825,
|
|
"loss": 5.8524,
|
|
"mean_token_accuracy": 0.14535145387053489,
|
|
"num_tokens": 3859656.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.879422616958618,
|
|
"epoch": 1.7241379310344827,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004885371562103498,
|
|
"loss": 5.7341,
|
|
"mean_token_accuracy": 0.1510668709874153,
|
|
"num_tokens": 3868698.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.836897468566894,
|
|
"epoch": 1.7280564263322884,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004884421822195957,
|
|
"loss": 5.8164,
|
|
"mean_token_accuracy": 0.14483954310417174,
|
|
"num_tokens": 3877474.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.9123077392578125,
|
|
"epoch": 1.7319749216300941,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004883468267730538,
|
|
"loss": 5.7831,
|
|
"mean_token_accuracy": 0.1473929390311241,
|
|
"num_tokens": 3886328.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.849182271957398,
|
|
"epoch": 1.7358934169278997,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048825109004114006,
|
|
"loss": 5.7698,
|
|
"mean_token_accuracy": 0.1499294877052307,
|
|
"num_tokens": 3894424.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.945364856719971,
|
|
"epoch": 1.7398119122257052,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004881549721949513,
|
|
"loss": 5.8649,
|
|
"mean_token_accuracy": 0.1370491735637188,
|
|
"num_tokens": 3903746.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.7993536472320555,
|
|
"epoch": 1.743730407523511,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004880584734062658,
|
|
"loss": 5.8347,
|
|
"mean_token_accuracy": 0.14640080332756042,
|
|
"num_tokens": 3912033.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.84619402885437,
|
|
"epoch": 1.7476489028213167,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004879615938475425,
|
|
"loss": 5.7356,
|
|
"mean_token_accuracy": 0.14485909938812255,
|
|
"num_tokens": 3920388.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.835551357269287,
|
|
"epoch": 1.7515673981191222,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004878643336919209,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.1401818074285984,
|
|
"num_tokens": 3929926.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.145181274414062,
|
|
"epoch": 1.7554858934169277,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004877666931132206,
|
|
"loss": 5.9229,
|
|
"mean_token_accuracy": 0.1354038953781128,
|
|
"num_tokens": 3938696.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.85701961517334,
|
|
"epoch": 1.7594043887147337,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004876686722859413,
|
|
"loss": 5.7488,
|
|
"mean_token_accuracy": 0.15484919100999833,
|
|
"num_tokens": 3947537.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.902706909179687,
|
|
"epoch": 1.7633228840125392,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000487570271385262,
|
|
"loss": 5.7587,
|
|
"mean_token_accuracy": 0.14329107031226157,
|
|
"num_tokens": 3956539.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.800901651382446,
|
|
"epoch": 1.7672413793103448,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004874714905870411,
|
|
"loss": 5.729,
|
|
"mean_token_accuracy": 0.14343830123543738,
|
|
"num_tokens": 3964927.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.866149950027466,
|
|
"epoch": 1.7711598746081505,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004873723300678159,
|
|
"loss": 5.8463,
|
|
"mean_token_accuracy": 0.1442209616303444,
|
|
"num_tokens": 3973565.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.898542833328247,
|
|
"epoch": 1.7750783699059562,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048727279000480226,
|
|
"loss": 5.893,
|
|
"mean_token_accuracy": 0.1396360144019127,
|
|
"num_tokens": 3982683.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.953893518447876,
|
|
"epoch": 1.7789968652037618,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 5.7974,
|
|
"mean_token_accuracy": 0.14212608486413955,
|
|
"num_tokens": 3991462.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.688703346252441,
|
|
"epoch": 1.7829153605015673,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004870725719596648,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.1619800105690956,
|
|
"num_tokens": 3999821.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.895339012145996,
|
|
"epoch": 1.786833855799373,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004869718943353631,
|
|
"loss": 5.8178,
|
|
"mean_token_accuracy": 0.14287081807851792,
|
|
"num_tokens": 4008655.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.915091705322266,
|
|
"epoch": 1.7907523510971788,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048687083788291656,
|
|
"loss": 5.8518,
|
|
"mean_token_accuracy": 0.14822934269905091,
|
|
"num_tokens": 4017317.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.805630207061768,
|
|
"epoch": 1.7946708463949843,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048676940278292953,
|
|
"loss": 5.6807,
|
|
"mean_token_accuracy": 0.1457840844988823,
|
|
"num_tokens": 4025407.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.82935643196106,
|
|
"epoch": 1.7985893416927898,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048666758921668286,
|
|
"loss": 5.7005,
|
|
"mean_token_accuracy": 0.1469753384590149,
|
|
"num_tokens": 4034310.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.86659049987793,
|
|
"epoch": 1.8025078369905956,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048656539736613403,
|
|
"loss": 5.8567,
|
|
"mean_token_accuracy": 0.14453689455986024,
|
|
"num_tokens": 4042326.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.759845781326294,
|
|
"epoch": 1.8064263322884013,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004864628274139164,
|
|
"loss": 5.7187,
|
|
"mean_token_accuracy": 0.15020606070756912,
|
|
"num_tokens": 4051248.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.875640344619751,
|
|
"epoch": 1.8103448275862069,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004863598795433391,
|
|
"loss": 5.842,
|
|
"mean_token_accuracy": 0.1422995701432228,
|
|
"num_tokens": 4060227.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.000100374221802,
|
|
"epoch": 1.8142633228840124,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048625655393838666,
|
|
"loss": 5.9161,
|
|
"mean_token_accuracy": 0.1404888704419136,
|
|
"num_tokens": 4069700.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.8732929706573485,
|
|
"epoch": 1.8181818181818183,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004861528507837186,
|
|
"loss": 5.7869,
|
|
"mean_token_accuracy": 0.14729702919721605,
|
|
"num_tokens": 4079363.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.989973688125611,
|
|
"epoch": 1.8221003134796239,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004860487702646695,
|
|
"loss": 5.8711,
|
|
"mean_token_accuracy": 0.14336605072021485,
|
|
"num_tokens": 4087944.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.748229169845581,
|
|
"epoch": 1.8260188087774294,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004859443125672479,
|
|
"loss": 5.5825,
|
|
"mean_token_accuracy": 0.16451934427022935,
|
|
"num_tokens": 4096431.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.809984350204468,
|
|
"epoch": 1.8299373040752351,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004858394778781368,
|
|
"loss": 5.7304,
|
|
"mean_token_accuracy": 0.14536977857351302,
|
|
"num_tokens": 4104894.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.85737190246582,
|
|
"epoch": 1.8338557993730409,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004857342663846927,
|
|
"loss": 5.7793,
|
|
"mean_token_accuracy": 0.14765906557440758,
|
|
"num_tokens": 4112878.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.807412147521973,
|
|
"epoch": 1.8377742946708464,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004856286782749456,
|
|
"loss": 5.7972,
|
|
"mean_token_accuracy": 0.15013805031776428,
|
|
"num_tokens": 4122501.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.830691576004028,
|
|
"epoch": 1.841692789968652,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004855227137375986,
|
|
"loss": 5.7605,
|
|
"mean_token_accuracy": 0.14317431300878525,
|
|
"num_tokens": 4131301.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.865339088439941,
|
|
"epoch": 1.8456112852664577,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004854163729620275,
|
|
"loss": 5.8114,
|
|
"mean_token_accuracy": 0.1458986446261406,
|
|
"num_tokens": 4140076.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.822086429595947,
|
|
"epoch": 1.8495297805642634,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004853096561382805,
|
|
"loss": 5.7286,
|
|
"mean_token_accuracy": 0.15770585536956788,
|
|
"num_tokens": 4148728.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.795872402191162,
|
|
"epoch": 1.853448275862069,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004852025634570779,
|
|
"loss": 5.8682,
|
|
"mean_token_accuracy": 0.14420996382832527,
|
|
"num_tokens": 4157520.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.847042655944824,
|
|
"epoch": 1.8573667711598745,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004850950951098116,
|
|
"loss": 5.6724,
|
|
"mean_token_accuracy": 0.15193334817886353,
|
|
"num_tokens": 4165547.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.845370531082153,
|
|
"epoch": 1.8612852664576802,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004849872512885451,
|
|
"loss": 5.8217,
|
|
"mean_token_accuracy": 0.1410464257001877,
|
|
"num_tokens": 4174656.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.869585704803467,
|
|
"epoch": 1.865203761755486,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004848790321860127,
|
|
"loss": 5.8582,
|
|
"mean_token_accuracy": 0.1439307987689972,
|
|
"num_tokens": 4183452.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.851327800750733,
|
|
"epoch": 1.8691222570532915,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048477043799561946,
|
|
"loss": 5.7456,
|
|
"mean_token_accuracy": 0.14440877065062524,
|
|
"num_tokens": 4192306.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.840966320037841,
|
|
"epoch": 1.873040752351097,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004846614689114409,
|
|
"loss": 5.7992,
|
|
"mean_token_accuracy": 0.14650781750679015,
|
|
"num_tokens": 4201201.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.8397688388824465,
|
|
"epoch": 1.876959247648903,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004845521251282223,
|
|
"loss": 5.7082,
|
|
"mean_token_accuracy": 0.15190000385046004,
|
|
"num_tokens": 4209374.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.744638442993164,
|
|
"epoch": 1.8808777429467085,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004844424068413789,
|
|
"loss": 5.6826,
|
|
"mean_token_accuracy": 0.1490471452474594,
|
|
"num_tokens": 4217258.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.77668948173523,
|
|
"epoch": 1.884796238244514,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048433231424699504,
|
|
"loss": 5.7063,
|
|
"mean_token_accuracy": 0.15233246684074403,
|
|
"num_tokens": 4226484.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.866843271255493,
|
|
"epoch": 1.8887147335423198,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048422184754182384,
|
|
"loss": 5.7461,
|
|
"mean_token_accuracy": 0.14515922889113425,
|
|
"num_tokens": 4235546.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.902421998977661,
|
|
"epoch": 1.8926332288401255,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004841110069232875,
|
|
"loss": 5.9077,
|
|
"mean_token_accuracy": 0.14326248839497566,
|
|
"num_tokens": 4244549.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.853962659835815,
|
|
"epoch": 1.896551724137931,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048399979258947597,
|
|
"loss": 5.7864,
|
|
"mean_token_accuracy": 0.15010540187358856,
|
|
"num_tokens": 4252918.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.853643131256104,
|
|
"epoch": 1.9004702194357366,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004838882047391474,
|
|
"loss": 5.7754,
|
|
"mean_token_accuracy": 0.14980721175670625,
|
|
"num_tokens": 4261674.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.83618278503418,
|
|
"epoch": 1.9043887147335423,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048377624357172724,
|
|
"loss": 5.682,
|
|
"mean_token_accuracy": 0.1527015097439289,
|
|
"num_tokens": 4269986.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.860264682769776,
|
|
"epoch": 1.908307210031348,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048366390928730843,
|
|
"loss": 5.7785,
|
|
"mean_token_accuracy": 0.14807360768318176,
|
|
"num_tokens": 4278131.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.86714735031128,
|
|
"epoch": 1.9122257053291536,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004835512020866504,
|
|
"loss": 5.765,
|
|
"mean_token_accuracy": 0.14669661298394204,
|
|
"num_tokens": 4287214.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.832046842575073,
|
|
"epoch": 1.9161442006269591,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048343812217117925,
|
|
"loss": 5.718,
|
|
"mean_token_accuracy": 0.15012799203395844,
|
|
"num_tokens": 4295949.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.764661169052124,
|
|
"epoch": 1.9200626959247649,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048332466974298723,
|
|
"loss": 5.6965,
|
|
"mean_token_accuracy": 0.14930392503738404,
|
|
"num_tokens": 4305014.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.798896789550781,
|
|
"epoch": 1.9239811912225706,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048321084500483203,
|
|
"loss": 5.6813,
|
|
"mean_token_accuracy": 0.15102455914020538,
|
|
"num_tokens": 4313560.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.798026752471924,
|
|
"epoch": 1.9278996865203761,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000483096648160137,
|
|
"loss": 5.7752,
|
|
"mean_token_accuracy": 0.14778882637619972,
|
|
"num_tokens": 4323280.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.883830404281616,
|
|
"epoch": 1.9318181818181817,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048298207941299047,
|
|
"loss": 5.8151,
|
|
"mean_token_accuracy": 0.1466996341943741,
|
|
"num_tokens": 4331757.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.7072854995727536,
|
|
"epoch": 1.9357366771159876,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048286713896814536,
|
|
"loss": 5.7525,
|
|
"mean_token_accuracy": 0.1452985815703869,
|
|
"num_tokens": 4341453.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.942684984207153,
|
|
"epoch": 1.9396551724137931,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048275182703101877,
|
|
"loss": 5.6894,
|
|
"mean_token_accuracy": 0.14936211407184602,
|
|
"num_tokens": 4349095.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.746797847747803,
|
|
"epoch": 1.9435736677115987,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048263614380769193,
|
|
"loss": 5.7959,
|
|
"mean_token_accuracy": 0.1456053450703621,
|
|
"num_tokens": 4357913.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.870775651931763,
|
|
"epoch": 1.9474921630094044,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048252008950490957,
|
|
"loss": 5.7708,
|
|
"mean_token_accuracy": 0.14770165383815764,
|
|
"num_tokens": 4366586.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.838080787658692,
|
|
"epoch": 1.9514106583072102,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048240366433007935,
|
|
"loss": 5.7125,
|
|
"mean_token_accuracy": 0.15084166526794435,
|
|
"num_tokens": 4375393.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.78338623046875,
|
|
"epoch": 1.9553291536050157,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048228686849127213,
|
|
"loss": 5.7729,
|
|
"mean_token_accuracy": 0.15526207983493806,
|
|
"num_tokens": 4383747.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.892977046966553,
|
|
"epoch": 1.9592476489028212,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004821697021972209,
|
|
"loss": 5.7593,
|
|
"mean_token_accuracy": 0.14306001290678977,
|
|
"num_tokens": 4392085.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 1.9592476489028212,
|
|
"eval_entropy": 5.763626485831978,
|
|
"eval_loss": 5.954261302947998,
|
|
"eval_mean_token_accuracy": 0.143982818169707,
|
|
"eval_num_tokens": 4392085.0,
|
|
"eval_runtime": 3.0476,
|
|
"eval_samples_per_second": 1352.559,
|
|
"eval_steps_per_second": 169.316,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.834238481521607,
|
|
"epoch": 1.963166144200627,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 5.7688,
|
|
"mean_token_accuracy": 0.1469443328678608,
|
|
"num_tokens": 4400843.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.706972885131836,
|
|
"epoch": 1.9670846394984327,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004819342590816288,
|
|
"loss": 5.713,
|
|
"mean_token_accuracy": 0.1477135218679905,
|
|
"num_tokens": 4409126.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.846028470993042,
|
|
"epoch": 1.9710031347962382,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004818159826808631,
|
|
"loss": 5.8674,
|
|
"mean_token_accuracy": 0.14325389117002488,
|
|
"num_tokens": 4418401.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.896506214141846,
|
|
"epoch": 1.9749216300940438,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004816973366664026,
|
|
"loss": 5.7801,
|
|
"mean_token_accuracy": 0.14661652594804764,
|
|
"num_tokens": 4428141.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.88172173500061,
|
|
"epoch": 1.9788401253918495,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004815783212502871,
|
|
"loss": 5.7916,
|
|
"mean_token_accuracy": 0.14361262023448945,
|
|
"num_tokens": 4437158.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.880087661743164,
|
|
"epoch": 1.9827586206896552,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00048145893664521645,
|
|
"loss": 5.8604,
|
|
"mean_token_accuracy": 0.13948971331119536,
|
|
"num_tokens": 4445901.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.7677483558654785,
|
|
"epoch": 1.9866771159874608,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048133918306455023,
|
|
"loss": 5.6718,
|
|
"mean_token_accuracy": 0.15518780797719955,
|
|
"num_tokens": 4454200.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.803468894958496,
|
|
"epoch": 1.9905956112852663,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004812190607223075,
|
|
"loss": 5.8031,
|
|
"mean_token_accuracy": 0.1415604218840599,
|
|
"num_tokens": 4463484.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.741824054718018,
|
|
"epoch": 1.9945141065830723,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048109856983316655,
|
|
"loss": 5.6013,
|
|
"mean_token_accuracy": 0.15295095592737198,
|
|
"num_tokens": 4472659.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.794041585922241,
|
|
"epoch": 1.9984326018808778,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000480977710612464,
|
|
"loss": 5.679,
|
|
"mean_token_accuracy": 0.14821849390864372,
|
|
"num_tokens": 4481355.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.810983371734619,
|
|
"epoch": 2.0023510971786833,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004808564832761948,
|
|
"loss": 5.5513,
|
|
"mean_token_accuracy": 0.1496092140674591,
|
|
"num_tokens": 4490005.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.746422338485718,
|
|
"epoch": 2.006269592476489,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004807348880410119,
|
|
"loss": 5.3604,
|
|
"mean_token_accuracy": 0.15983257442712784,
|
|
"num_tokens": 4498645.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.708592414855957,
|
|
"epoch": 2.010188087774295,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004806129251242258,
|
|
"loss": 5.4649,
|
|
"mean_token_accuracy": 0.15696969926357268,
|
|
"num_tokens": 4509675.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.696289110183716,
|
|
"epoch": 2.0141065830721003,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048049059474380393,
|
|
"loss": 5.2851,
|
|
"mean_token_accuracy": 0.16772339046001433,
|
|
"num_tokens": 4518463.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.779043769836425,
|
|
"epoch": 2.018025078369906,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048036789711837047,
|
|
"loss": 5.4456,
|
|
"mean_token_accuracy": 0.15525378733873368,
|
|
"num_tokens": 4527120.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.779226970672608,
|
|
"epoch": 2.0219435736677114,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048024483246720607,
|
|
"loss": 5.4586,
|
|
"mean_token_accuracy": 0.15783494412899018,
|
|
"num_tokens": 4535962.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.769890880584716,
|
|
"epoch": 2.0258620689655173,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004801214010102472,
|
|
"loss": 5.4178,
|
|
"mean_token_accuracy": 0.1544024720788002,
|
|
"num_tokens": 4545233.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.655493211746216,
|
|
"epoch": 2.029780564263323,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004799976029680858,
|
|
"loss": 5.3017,
|
|
"mean_token_accuracy": 0.1586959809064865,
|
|
"num_tokens": 4554658.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.765646028518677,
|
|
"epoch": 2.0336990595611284,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004798734385619691,
|
|
"loss": 5.5054,
|
|
"mean_token_accuracy": 0.15658126920461654,
|
|
"num_tokens": 4563964.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.734431409835816,
|
|
"epoch": 2.0376175548589344,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000479748908013799,
|
|
"loss": 5.3552,
|
|
"mean_token_accuracy": 0.1636643648147583,
|
|
"num_tokens": 4572865.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.65130581855774,
|
|
"epoch": 2.04153605015674,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004796240115461319,
|
|
"loss": 5.3316,
|
|
"mean_token_accuracy": 0.17066773176193237,
|
|
"num_tokens": 4581021.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.677306890487671,
|
|
"epoch": 2.0454545454545454,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004794987493821779,
|
|
"loss": 5.4296,
|
|
"mean_token_accuracy": 0.16218193769454955,
|
|
"num_tokens": 4590779.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.698957872390747,
|
|
"epoch": 2.049373040752351,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047937312174580084,
|
|
"loss": 5.4157,
|
|
"mean_token_accuracy": 0.160083344578743,
|
|
"num_tokens": 4599831.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.789990186691284,
|
|
"epoch": 2.053291536050157,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004792471288615177,
|
|
"loss": 5.5427,
|
|
"mean_token_accuracy": 0.15152812749147415,
|
|
"num_tokens": 4609264.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.760856819152832,
|
|
"epoch": 2.0572100313479624,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004791207709544981,
|
|
"loss": 5.4465,
|
|
"mean_token_accuracy": 0.16058844327926636,
|
|
"num_tokens": 4617885.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.612568521499634,
|
|
"epoch": 2.061128526645768,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047899404825056424,
|
|
"loss": 5.3784,
|
|
"mean_token_accuracy": 0.1592410311102867,
|
|
"num_tokens": 4626536.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.686727571487427,
|
|
"epoch": 2.0650470219435735,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004788669609761901,
|
|
"loss": 5.4694,
|
|
"mean_token_accuracy": 0.15769321471452713,
|
|
"num_tokens": 4635563.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.679330635070801,
|
|
"epoch": 2.0689655172413794,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047873950935850107,
|
|
"loss": 5.383,
|
|
"mean_token_accuracy": 0.15778388381004332,
|
|
"num_tokens": 4644822.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.6939338684082035,
|
|
"epoch": 2.072884012539185,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004786116936252742,
|
|
"loss": 5.3568,
|
|
"mean_token_accuracy": 0.16605196446180343,
|
|
"num_tokens": 4652709.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.627925157546997,
|
|
"epoch": 2.0768025078369905,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004784835140049367,
|
|
"loss": 5.385,
|
|
"mean_token_accuracy": 0.16054873168468475,
|
|
"num_tokens": 4662014.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.69994478225708,
|
|
"epoch": 2.080721003134796,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004783549707265663,
|
|
"loss": 5.3517,
|
|
"mean_token_accuracy": 0.1625790148973465,
|
|
"num_tokens": 4670618.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.681219434738159,
|
|
"epoch": 2.084639498432602,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047822606401989084,
|
|
"loss": 5.4942,
|
|
"mean_token_accuracy": 0.14787036925554276,
|
|
"num_tokens": 4679937.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.704634380340576,
|
|
"epoch": 2.0885579937304075,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004780967941152873,
|
|
"loss": 5.3145,
|
|
"mean_token_accuracy": 0.17359399497509004,
|
|
"num_tokens": 4688006.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.606400442123413,
|
|
"epoch": 2.092476489028213,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004779671612437822,
|
|
"loss": 5.2758,
|
|
"mean_token_accuracy": 0.17059992551803588,
|
|
"num_tokens": 4697106.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.647980356216431,
|
|
"epoch": 2.096394984326019,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047783716563705035,
|
|
"loss": 5.4239,
|
|
"mean_token_accuracy": 0.1599812850356102,
|
|
"num_tokens": 4705582.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.711063385009766,
|
|
"epoch": 2.1003134796238245,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000477706807527415,
|
|
"loss": 5.3471,
|
|
"mean_token_accuracy": 0.1619032919406891,
|
|
"num_tokens": 4713640.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.6495414733886715,
|
|
"epoch": 2.10423197492163,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004775760871478472,
|
|
"loss": 5.462,
|
|
"mean_token_accuracy": 0.15760222524404527,
|
|
"num_tokens": 4722729.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.694121789932251,
|
|
"epoch": 2.1081504702194356,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047744500473196564,
|
|
"loss": 5.3624,
|
|
"mean_token_accuracy": 0.16417956352233887,
|
|
"num_tokens": 4732099.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.713819122314453,
|
|
"epoch": 2.1120689655172415,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047731356051403556,
|
|
"loss": 5.3885,
|
|
"mean_token_accuracy": 0.1580797016620636,
|
|
"num_tokens": 4741766.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.535807085037232,
|
|
"epoch": 2.115987460815047,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004771817547289693,
|
|
"loss": 5.3097,
|
|
"mean_token_accuracy": 0.15867679566144943,
|
|
"num_tokens": 4750890.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.64779224395752,
|
|
"epoch": 2.1199059561128526,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004770495876123251,
|
|
"loss": 5.4227,
|
|
"mean_token_accuracy": 0.1575888440012932,
|
|
"num_tokens": 4760063.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.715673780441284,
|
|
"epoch": 2.123824451410658,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004769170594003071,
|
|
"loss": 5.4254,
|
|
"mean_token_accuracy": 0.162799733877182,
|
|
"num_tokens": 4768928.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.595988321304321,
|
|
"epoch": 2.127742946708464,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047678417032976457,
|
|
"loss": 5.4041,
|
|
"mean_token_accuracy": 0.1568652793765068,
|
|
"num_tokens": 4778129.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.660232877731323,
|
|
"epoch": 2.1316614420062696,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004766509206381919,
|
|
"loss": 5.3938,
|
|
"mean_token_accuracy": 0.1678140342235565,
|
|
"num_tokens": 4786759.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.601139307022095,
|
|
"epoch": 2.135579937304075,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004765173105637279,
|
|
"loss": 5.386,
|
|
"mean_token_accuracy": 0.16851667910814286,
|
|
"num_tokens": 4795355.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.6140483856201175,
|
|
"epoch": 2.139498432601881,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047638334034515547,
|
|
"loss": 5.3485,
|
|
"mean_token_accuracy": 0.16128165572881697,
|
|
"num_tokens": 4804030.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.511086130142212,
|
|
"epoch": 2.1434169278996866,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047624901022190106,
|
|
"loss": 5.3041,
|
|
"mean_token_accuracy": 0.18045085221529006,
|
|
"num_tokens": 4812779.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.670054578781128,
|
|
"epoch": 2.147335423197492,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.4144,
|
|
"mean_token_accuracy": 0.16085021644830705,
|
|
"num_tokens": 4822292.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.751107263565063,
|
|
"epoch": 2.1512539184952977,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004759792712222679,
|
|
"loss": 5.4804,
|
|
"mean_token_accuracy": 0.15966029614210128,
|
|
"num_tokens": 4830524.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.727796506881714,
|
|
"epoch": 2.1551724137931036,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004758438628279565,
|
|
"loss": 5.479,
|
|
"mean_token_accuracy": 0.163248011469841,
|
|
"num_tokens": 4839328.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.61474494934082,
|
|
"epoch": 2.159090909090909,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047570809549309697,
|
|
"loss": 5.3055,
|
|
"mean_token_accuracy": 0.1764591619372368,
|
|
"num_tokens": 4847711.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.754563331604004,
|
|
"epoch": 2.1630094043887147,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004755719694603275,
|
|
"loss": 5.4965,
|
|
"mean_token_accuracy": 0.1527958817780018,
|
|
"num_tokens": 4856706.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.6675133228302,
|
|
"epoch": 2.16692789968652,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004754354849729274,
|
|
"loss": 5.4805,
|
|
"mean_token_accuracy": 0.1656187355518341,
|
|
"num_tokens": 4865250.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.625590085983276,
|
|
"epoch": 2.170846394984326,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047529864227481653,
|
|
"loss": 5.4158,
|
|
"mean_token_accuracy": 0.16449171751737596,
|
|
"num_tokens": 4873826.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.752429628372193,
|
|
"epoch": 2.1747648902821317,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000475161441610555,
|
|
"loss": 5.4052,
|
|
"mean_token_accuracy": 0.16321943402290345,
|
|
"num_tokens": 4882631.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.583358240127564,
|
|
"epoch": 2.1786833855799372,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004750238832253427,
|
|
"loss": 5.4461,
|
|
"mean_token_accuracy": 0.16085778921842575,
|
|
"num_tokens": 4891007.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.633743333816528,
|
|
"epoch": 2.1826018808777428,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004748859673650187,
|
|
"loss": 5.3402,
|
|
"mean_token_accuracy": 0.1616213023662567,
|
|
"num_tokens": 4899692.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.626257991790771,
|
|
"epoch": 2.1865203761755487,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047474769427606115,
|
|
"loss": 5.4681,
|
|
"mean_token_accuracy": 0.16049360036849974,
|
|
"num_tokens": 4909354.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.593847322463989,
|
|
"epoch": 2.1904388714733543,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004746090642055863,
|
|
"loss": 5.4072,
|
|
"mean_token_accuracy": 0.15933744981884956,
|
|
"num_tokens": 4918400.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.7399280071258545,
|
|
"epoch": 2.19435736677116,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047447007740134857,
|
|
"loss": 5.5062,
|
|
"mean_token_accuracy": 0.15213064849376678,
|
|
"num_tokens": 4926852.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.665618944168091,
|
|
"epoch": 2.1982758620689653,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047433073411174,
|
|
"loss": 5.417,
|
|
"mean_token_accuracy": 0.16172380447387696,
|
|
"num_tokens": 4935717.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.553526878356934,
|
|
"epoch": 2.2021943573667713,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004741910345857896,
|
|
"loss": 5.3279,
|
|
"mean_token_accuracy": 0.16639951318502427,
|
|
"num_tokens": 4945389.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.694633340835571,
|
|
"epoch": 2.206112852664577,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047405097907316315,
|
|
"loss": 5.4695,
|
|
"mean_token_accuracy": 0.16526516079902648,
|
|
"num_tokens": 4953670.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.613465595245361,
|
|
"epoch": 2.2100313479623823,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004739105678241625,
|
|
"loss": 5.4892,
|
|
"mean_token_accuracy": 0.16113940477371216,
|
|
"num_tokens": 4962383.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.6064088344573975,
|
|
"epoch": 2.2139498432601883,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004737698010897253,
|
|
"loss": 5.2802,
|
|
"mean_token_accuracy": 0.16448537707328797,
|
|
"num_tokens": 4970672.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.686607408523559,
|
|
"epoch": 2.217868338557994,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004736286791214245,
|
|
"loss": 5.4581,
|
|
"mean_token_accuracy": 0.15645889192819595,
|
|
"num_tokens": 4979303.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.671361589431763,
|
|
"epoch": 2.2217868338557993,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047348720217146807,
|
|
"loss": 5.3387,
|
|
"mean_token_accuracy": 0.1662439689040184,
|
|
"num_tokens": 4988322.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.515157270431518,
|
|
"epoch": 2.225705329153605,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047334537049269806,
|
|
"loss": 5.4164,
|
|
"mean_token_accuracy": 0.1666090041399002,
|
|
"num_tokens": 4997250.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.727092695236206,
|
|
"epoch": 2.229623824451411,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004732031843385909,
|
|
"loss": 5.4815,
|
|
"mean_token_accuracy": 0.15678349286317825,
|
|
"num_tokens": 5005459.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.572847080230713,
|
|
"epoch": 2.2335423197492164,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004730606439632562,
|
|
"loss": 5.2291,
|
|
"mean_token_accuracy": 0.1836922749876976,
|
|
"num_tokens": 5013704.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.526752853393555,
|
|
"epoch": 2.237460815047022,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004729177496214367,
|
|
"loss": 5.3517,
|
|
"mean_token_accuracy": 0.16658095121383668,
|
|
"num_tokens": 5022545.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.614648056030274,
|
|
"epoch": 2.2413793103448274,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047277450156850767,
|
|
"loss": 5.3661,
|
|
"mean_token_accuracy": 0.17083664387464523,
|
|
"num_tokens": 5031133.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.626172876358032,
|
|
"epoch": 2.2452978056426334,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004726309000604768,
|
|
"loss": 5.3934,
|
|
"mean_token_accuracy": 0.16049793362617493,
|
|
"num_tokens": 5039424.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.729151344299316,
|
|
"epoch": 2.249216300940439,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004724869453539832,
|
|
"loss": 5.5865,
|
|
"mean_token_accuracy": 0.15456347092986106,
|
|
"num_tokens": 5048868.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.674163961410523,
|
|
"epoch": 2.2531347962382444,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004723426377062972,
|
|
"loss": 5.409,
|
|
"mean_token_accuracy": 0.16188189685344695,
|
|
"num_tokens": 5057532.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.650389337539673,
|
|
"epoch": 2.2570532915360504,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047219797737532,
|
|
"loss": 5.4448,
|
|
"mean_token_accuracy": 0.15711349695920945,
|
|
"num_tokens": 5066088.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.564129447937011,
|
|
"epoch": 2.260971786833856,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047205296461958314,
|
|
"loss": 5.332,
|
|
"mean_token_accuracy": 0.16345474123954773,
|
|
"num_tokens": 5074968.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.535284471511841,
|
|
"epoch": 2.2648902821316614,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047190759969824785,
|
|
"loss": 5.3864,
|
|
"mean_token_accuracy": 0.16005248427391053,
|
|
"num_tokens": 5083605.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.592043733596801,
|
|
"epoch": 2.268808777429467,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047176188287110485,
|
|
"loss": 5.3728,
|
|
"mean_token_accuracy": 0.16814243644475937,
|
|
"num_tokens": 5092182.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.741699647903443,
|
|
"epoch": 2.2727272727272725,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004716158143985737,
|
|
"loss": 5.5483,
|
|
"mean_token_accuracy": 0.15272048115730286,
|
|
"num_tokens": 5100910.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.6817862033844,
|
|
"epoch": 2.2766457680250785,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047146939454170245,
|
|
"loss": 5.4328,
|
|
"mean_token_accuracy": 0.16204959005117417,
|
|
"num_tokens": 5109760.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.673229742050171,
|
|
"epoch": 2.280564263322884,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004713226235621672,
|
|
"loss": 5.4596,
|
|
"mean_token_accuracy": 0.15853570401668549,
|
|
"num_tokens": 5119193.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.601157140731812,
|
|
"epoch": 2.2844827586206895,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004711755017222714,
|
|
"loss": 5.3559,
|
|
"mean_token_accuracy": 0.16015833467245102,
|
|
"num_tokens": 5127480.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.725550508499145,
|
|
"epoch": 2.2884012539184955,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00047102802928494563,
|
|
"loss": 5.5433,
|
|
"mean_token_accuracy": 0.15101971477270126,
|
|
"num_tokens": 5136648.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.624814748764038,
|
|
"epoch": 2.292319749216301,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004708802065137471,
|
|
"loss": 5.3275,
|
|
"mean_token_accuracy": 0.17106858342885972,
|
|
"num_tokens": 5144708.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.601879692077636,
|
|
"epoch": 2.2962382445141065,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004707320336728591,
|
|
"loss": 5.4544,
|
|
"mean_token_accuracy": 0.15818202570080758,
|
|
"num_tokens": 5153388.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.731164407730103,
|
|
"epoch": 2.300156739811912,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004705835110270904,
|
|
"loss": 5.4263,
|
|
"mean_token_accuracy": 0.15436539202928543,
|
|
"num_tokens": 5161874.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.637136602401734,
|
|
"epoch": 2.304075235109718,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047043463884187517,
|
|
"loss": 5.4185,
|
|
"mean_token_accuracy": 0.1649361878633499,
|
|
"num_tokens": 5170284.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.574511957168579,
|
|
"epoch": 2.3079937304075235,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047028541738327207,
|
|
"loss": 5.4005,
|
|
"mean_token_accuracy": 0.15930936336517335,
|
|
"num_tokens": 5179547.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.655497884750366,
|
|
"epoch": 2.311912225705329,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004701358469179641,
|
|
"loss": 5.4048,
|
|
"mean_token_accuracy": 0.1641696497797966,
|
|
"num_tokens": 5188256.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.663059854507447,
|
|
"epoch": 2.3158307210031346,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004699859277132578,
|
|
"loss": 5.5484,
|
|
"mean_token_accuracy": 0.16418840512633323,
|
|
"num_tokens": 5197464.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.660706186294556,
|
|
"epoch": 2.3197492163009406,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046983566003708336,
|
|
"loss": 5.3437,
|
|
"mean_token_accuracy": 0.16617220491170884,
|
|
"num_tokens": 5205989.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.5186491966247555,
|
|
"epoch": 2.323667711598746,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00046968504415799325,
|
|
"loss": 5.3843,
|
|
"mean_token_accuracy": 0.15985517501831054,
|
|
"num_tokens": 5214490.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.611410856246948,
|
|
"epoch": 2.3275862068965516,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004695340803451625,
|
|
"loss": 5.4376,
|
|
"mean_token_accuracy": 0.16383050829172136,
|
|
"num_tokens": 5223335.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.569272565841675,
|
|
"epoch": 2.3315047021943576,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 5.3592,
|
|
"mean_token_accuracy": 0.16942132711410524,
|
|
"num_tokens": 5231665.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.569160223007202,
|
|
"epoch": 2.335423197492163,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004692311099980878,
|
|
"loss": 5.3637,
|
|
"mean_token_accuracy": 0.16532504111528395,
|
|
"num_tokens": 5239810.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.567423915863037,
|
|
"epoch": 2.3393416927899686,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046907910400530097,
|
|
"loss": 5.3193,
|
|
"mean_token_accuracy": 0.17066740244627,
|
|
"num_tokens": 5247525.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.516646862030029,
|
|
"epoch": 2.343260188087774,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004689267511616868,
|
|
"loss": 5.3818,
|
|
"mean_token_accuracy": 0.16408997029066086,
|
|
"num_tokens": 5255528.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.585805368423462,
|
|
"epoch": 2.34717868338558,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046877405173952465,
|
|
"loss": 5.3266,
|
|
"mean_token_accuracy": 0.174003766477108,
|
|
"num_tokens": 5263944.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.590711212158203,
|
|
"epoch": 2.3510971786833856,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000468621006011713,
|
|
"loss": 5.4438,
|
|
"mean_token_accuracy": 0.1559297725558281,
|
|
"num_tokens": 5271724.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.3510971786833856,
|
|
"eval_entropy": 5.491303259088087,
|
|
"eval_loss": 5.911845684051514,
|
|
"eval_mean_token_accuracy": 0.15017772832697676,
|
|
"eval_num_tokens": 5271724.0,
|
|
"eval_runtime": 2.8537,
|
|
"eval_samples_per_second": 1444.423,
|
|
"eval_steps_per_second": 180.816,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.675783157348633,
|
|
"epoch": 2.355015673981191,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046846761425176943,
|
|
"loss": 5.4643,
|
|
"mean_token_accuracy": 0.16183075308799744,
|
|
"num_tokens": 5281199.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.726640319824218,
|
|
"epoch": 2.3589341692789967,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004683138767338299,
|
|
"loss": 5.4838,
|
|
"mean_token_accuracy": 0.15869480073451997,
|
|
"num_tokens": 5289782.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.470588254928589,
|
|
"epoch": 2.3628526645768027,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004681597937326483,
|
|
"loss": 5.2866,
|
|
"mean_token_accuracy": 0.1662585988640785,
|
|
"num_tokens": 5297922.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.603834104537964,
|
|
"epoch": 2.366771159874608,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004680053655235959,
|
|
"loss": 5.4171,
|
|
"mean_token_accuracy": 0.16951666325330733,
|
|
"num_tokens": 5306178.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.5670802116394045,
|
|
"epoch": 2.3706896551724137,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004678505923826609,
|
|
"loss": 5.3989,
|
|
"mean_token_accuracy": 0.1666540041565895,
|
|
"num_tokens": 5314936.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.604729795455933,
|
|
"epoch": 2.3746081504702197,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046769547458644817,
|
|
"loss": 5.4255,
|
|
"mean_token_accuracy": 0.16180657744407653,
|
|
"num_tokens": 5323549.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.613276863098145,
|
|
"epoch": 2.378526645768025,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004675400124121782,
|
|
"loss": 5.4103,
|
|
"mean_token_accuracy": 0.15842484682798386,
|
|
"num_tokens": 5332310.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.725310039520264,
|
|
"epoch": 2.3824451410658307,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046738420613768716,
|
|
"loss": 5.5382,
|
|
"mean_token_accuracy": 0.15389998108148575,
|
|
"num_tokens": 5340757.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.706901788711548,
|
|
"epoch": 2.3863636363636362,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00046722805604142614,
|
|
"loss": 5.5333,
|
|
"mean_token_accuracy": 0.15607720166444777,
|
|
"num_tokens": 5349433.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.566255140304565,
|
|
"epoch": 2.3902821316614418,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046707156240246076,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.15470652133226395,
|
|
"num_tokens": 5358943.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.619303560256958,
|
|
"epoch": 2.3942006269592477,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046691472550047027,
|
|
"loss": 5.3875,
|
|
"mean_token_accuracy": 0.15985837429761887,
|
|
"num_tokens": 5367782.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.555861377716065,
|
|
"epoch": 2.3981191222570533,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046675754561574783,
|
|
"loss": 5.2596,
|
|
"mean_token_accuracy": 0.17035585790872573,
|
|
"num_tokens": 5376186.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.658821105957031,
|
|
"epoch": 2.402037617554859,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046660002302919933,
|
|
"loss": 5.5285,
|
|
"mean_token_accuracy": 0.15329834818840027,
|
|
"num_tokens": 5384888.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.663078832626343,
|
|
"epoch": 2.4059561128526648,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004664421580223433,
|
|
"loss": 5.4141,
|
|
"mean_token_accuracy": 0.15931179076433183,
|
|
"num_tokens": 5393394.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.643510055541992,
|
|
"epoch": 2.4098746081504703,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046628395087730995,
|
|
"loss": 5.4918,
|
|
"mean_token_accuracy": 0.1582852229475975,
|
|
"num_tokens": 5402901.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.675841808319092,
|
|
"epoch": 2.413793103448276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004661254018768411,
|
|
"loss": 5.4699,
|
|
"mean_token_accuracy": 0.15755298137664794,
|
|
"num_tokens": 5412704.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.633321332931518,
|
|
"epoch": 2.4177115987460813,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004659665113042897,
|
|
"loss": 5.435,
|
|
"mean_token_accuracy": 0.16969642788171768,
|
|
"num_tokens": 5420863.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.5521032333374025,
|
|
"epoch": 2.4216300940438873,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004658072794436187,
|
|
"loss": 5.3852,
|
|
"mean_token_accuracy": 0.166944320499897,
|
|
"num_tokens": 5430002.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.562594604492188,
|
|
"epoch": 2.425548589341693,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00046564770657940146,
|
|
"loss": 5.359,
|
|
"mean_token_accuracy": 0.1669471099972725,
|
|
"num_tokens": 5438424.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.5414379119873045,
|
|
"epoch": 2.4294670846394983,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004654877929968205,
|
|
"loss": 5.3759,
|
|
"mean_token_accuracy": 0.16837773621082305,
|
|
"num_tokens": 5446970.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.6664660453796385,
|
|
"epoch": 2.433385579937304,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004653275389816673,
|
|
"loss": 5.4633,
|
|
"mean_token_accuracy": 0.15225489139556886,
|
|
"num_tokens": 5456689.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.556304597854615,
|
|
"epoch": 2.43730407523511,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046516694482034174,
|
|
"loss": 5.347,
|
|
"mean_token_accuracy": 0.16993902623653412,
|
|
"num_tokens": 5465269.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.574057722091675,
|
|
"epoch": 2.4412225705329154,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046500601079985164,
|
|
"loss": 5.4278,
|
|
"mean_token_accuracy": 0.15998573750257492,
|
|
"num_tokens": 5473930.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.583022117614746,
|
|
"epoch": 2.445141065830721,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004648447372078123,
|
|
"loss": 5.4426,
|
|
"mean_token_accuracy": 0.16931652575731276,
|
|
"num_tokens": 5482885.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.561769342422485,
|
|
"epoch": 2.449059561128527,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004646831243324457,
|
|
"loss": 5.3929,
|
|
"mean_token_accuracy": 0.16230092644691468,
|
|
"num_tokens": 5491489.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.607778263092041,
|
|
"epoch": 2.4529780564263324,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004645211724625802,
|
|
"loss": 5.438,
|
|
"mean_token_accuracy": 0.15634278506040572,
|
|
"num_tokens": 5499694.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.516080284118653,
|
|
"epoch": 2.456896551724138,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046435888188765015,
|
|
"loss": 5.3437,
|
|
"mean_token_accuracy": 0.16546535193920137,
|
|
"num_tokens": 5508037.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.5814862728118895,
|
|
"epoch": 2.4608150470219434,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004641962528976951,
|
|
"loss": 5.3158,
|
|
"mean_token_accuracy": 0.16855327934026718,
|
|
"num_tokens": 5516644.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.5680018901824955,
|
|
"epoch": 2.4647335423197494,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004640332857833593,
|
|
"loss": 5.3807,
|
|
"mean_token_accuracy": 0.1600844755768776,
|
|
"num_tokens": 5524789.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.632185077667236,
|
|
"epoch": 2.468652037617555,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046386998083589156,
|
|
"loss": 5.4096,
|
|
"mean_token_accuracy": 0.1625368997454643,
|
|
"num_tokens": 5533739.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.694439554214478,
|
|
"epoch": 2.4725705329153604,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004637063383471442,
|
|
"loss": 5.4569,
|
|
"mean_token_accuracy": 0.15983943045139312,
|
|
"num_tokens": 5542735.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.507910346984863,
|
|
"epoch": 2.476489028213166,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046354235860957287,
|
|
"loss": 5.2771,
|
|
"mean_token_accuracy": 0.1729465827345848,
|
|
"num_tokens": 5551272.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.603234767913818,
|
|
"epoch": 2.480407523510972,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004633780419162361,
|
|
"loss": 5.4932,
|
|
"mean_token_accuracy": 0.15716639757156373,
|
|
"num_tokens": 5559868.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.567801570892334,
|
|
"epoch": 2.4843260188087775,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046321338856079435,
|
|
"loss": 5.2828,
|
|
"mean_token_accuracy": 0.1731288328766823,
|
|
"num_tokens": 5568276.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.5245575428009035,
|
|
"epoch": 2.488244514106583,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046304839883750987,
|
|
"loss": 5.374,
|
|
"mean_token_accuracy": 0.15898774117231368,
|
|
"num_tokens": 5576462.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.6145881652832035,
|
|
"epoch": 2.492163009404389,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000462883073041246,
|
|
"loss": 5.5271,
|
|
"mean_token_accuracy": 0.15560855269432067,
|
|
"num_tokens": 5586044.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.490754890441894,
|
|
"epoch": 2.4960815047021945,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004627174114674669,
|
|
"loss": 5.2981,
|
|
"mean_token_accuracy": 0.16564093977212907,
|
|
"num_tokens": 5594626.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.593857192993164,
|
|
"epoch": 2.5,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004625514144122365,
|
|
"loss": 5.3428,
|
|
"mean_token_accuracy": 0.16377169042825698,
|
|
"num_tokens": 5603268.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.521280574798584,
|
|
"epoch": 2.5039184952978055,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004623850821722185,
|
|
"loss": 5.3269,
|
|
"mean_token_accuracy": 0.16662093102931977,
|
|
"num_tokens": 5612053.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.552582836151123,
|
|
"epoch": 2.507836990595611,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004622184150446756,
|
|
"loss": 5.6012,
|
|
"mean_token_accuracy": 0.15160997956991196,
|
|
"num_tokens": 5621297.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.662527799606323,
|
|
"epoch": 2.511755485893417,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046205141332746904,
|
|
"loss": 5.4132,
|
|
"mean_token_accuracy": 0.15854287445545195,
|
|
"num_tokens": 5629683.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.500469493865967,
|
|
"epoch": 2.5156739811912225,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 5.339,
|
|
"mean_token_accuracy": 0.16766623109579087,
|
|
"num_tokens": 5638707.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.605044984817505,
|
|
"epoch": 2.519592476489028,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004617164073184987,
|
|
"loss": 5.3992,
|
|
"mean_token_accuracy": 0.16433212906122208,
|
|
"num_tokens": 5648241.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.536972951889038,
|
|
"epoch": 2.523510971786834,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046154840362544496,
|
|
"loss": 5.4382,
|
|
"mean_token_accuracy": 0.1562422513961792,
|
|
"num_tokens": 5656743.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.570316362380981,
|
|
"epoch": 2.5274294670846396,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004613800665401466,
|
|
"loss": 5.4117,
|
|
"mean_token_accuracy": 0.1637290135025978,
|
|
"num_tokens": 5666418.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.535870838165283,
|
|
"epoch": 2.531347962382445,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004612113963634493,
|
|
"loss": 5.3683,
|
|
"mean_token_accuracy": 0.16133923530578614,
|
|
"num_tokens": 5675404.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.551685237884522,
|
|
"epoch": 2.535266457680251,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004610423933967938,
|
|
"loss": 5.46,
|
|
"mean_token_accuracy": 0.1529649630188942,
|
|
"num_tokens": 5684699.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.722001266479492,
|
|
"epoch": 2.5391849529780566,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000460873057942216,
|
|
"loss": 5.4491,
|
|
"mean_token_accuracy": 0.16463801115751267,
|
|
"num_tokens": 5693219.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.606706762313843,
|
|
"epoch": 2.543103448275862,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004607033903023458,
|
|
"loss": 5.4686,
|
|
"mean_token_accuracy": 0.15889834612607956,
|
|
"num_tokens": 5702005.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.662101173400879,
|
|
"epoch": 2.5470219435736676,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046053339078040674,
|
|
"loss": 5.4656,
|
|
"mean_token_accuracy": 0.16294230073690413,
|
|
"num_tokens": 5710715.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.579379224777222,
|
|
"epoch": 2.550940438871473,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004603630596802155,
|
|
"loss": 5.3864,
|
|
"mean_token_accuracy": 0.1617594212293625,
|
|
"num_tokens": 5719444.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.484048748016358,
|
|
"epoch": 2.554858934169279,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004601923973061814,
|
|
"loss": 5.3753,
|
|
"mean_token_accuracy": 0.15965436697006224,
|
|
"num_tokens": 5727755.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.58670973777771,
|
|
"epoch": 2.5587774294670846,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046002140396330575,
|
|
"loss": 5.3151,
|
|
"mean_token_accuracy": 0.16526033729314804,
|
|
"num_tokens": 5736933.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.605556392669678,
|
|
"epoch": 2.56269592476489,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045985007995718154,
|
|
"loss": 5.4788,
|
|
"mean_token_accuracy": 0.1635790839791298,
|
|
"num_tokens": 5745831.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.609490251541137,
|
|
"epoch": 2.566614420062696,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004596784255939923,
|
|
"loss": 5.5104,
|
|
"mean_token_accuracy": 0.16027603894472123,
|
|
"num_tokens": 5755419.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.5760337829589846,
|
|
"epoch": 2.5705329153605017,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004595064411805123,
|
|
"loss": 5.3992,
|
|
"mean_token_accuracy": 0.16867431849241257,
|
|
"num_tokens": 5764008.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.5042609691619875,
|
|
"epoch": 2.574451410658307,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004593341270241057,
|
|
"loss": 5.292,
|
|
"mean_token_accuracy": 0.16733041852712632,
|
|
"num_tokens": 5772129.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.540396451950073,
|
|
"epoch": 2.5783699059561127,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004591614834327257,
|
|
"loss": 5.4202,
|
|
"mean_token_accuracy": 0.1609404519200325,
|
|
"num_tokens": 5781079.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.571345615386963,
|
|
"epoch": 2.5822884012539182,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045898851071491444,
|
|
"loss": 5.3987,
|
|
"mean_token_accuracy": 0.16920142620801926,
|
|
"num_tokens": 5790444.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.472708415985108,
|
|
"epoch": 2.586206896551724,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004588152091798022,
|
|
"loss": 5.2876,
|
|
"mean_token_accuracy": 0.16264791637659073,
|
|
"num_tokens": 5799828.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.700000429153443,
|
|
"epoch": 2.5901253918495297,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004586415791371069,
|
|
"loss": 5.5589,
|
|
"mean_token_accuracy": 0.1546184465289116,
|
|
"num_tokens": 5808765.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.626214504241943,
|
|
"epoch": 2.5940438871473352,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004584676208971336,
|
|
"loss": 5.3271,
|
|
"mean_token_accuracy": 0.16927535235881805,
|
|
"num_tokens": 5817281.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.5023740291595455,
|
|
"epoch": 2.597962382445141,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045829333477077384,
|
|
"loss": 5.3652,
|
|
"mean_token_accuracy": 0.17652563750743866,
|
|
"num_tokens": 5825741.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.629685878753662,
|
|
"epoch": 2.6018808777429467,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004581187210695053,
|
|
"loss": 5.3434,
|
|
"mean_token_accuracy": 0.17185610830783843,
|
|
"num_tokens": 5834466.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.489698600769043,
|
|
"epoch": 2.6057993730407523,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000457943780105391,
|
|
"loss": 5.3849,
|
|
"mean_token_accuracy": 0.1665027320384979,
|
|
"num_tokens": 5843463.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.544461441040039,
|
|
"epoch": 2.6097178683385582,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045776851219107856,
|
|
"loss": 5.4107,
|
|
"mean_token_accuracy": 0.1582068219780922,
|
|
"num_tokens": 5851885.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.617935609817505,
|
|
"epoch": 2.6136363636363638,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045759291763980035,
|
|
"loss": 5.4308,
|
|
"mean_token_accuracy": 0.16288012862205506,
|
|
"num_tokens": 5861537.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.5244855880737305,
|
|
"epoch": 2.6175548589341693,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045741699676537227,
|
|
"loss": 5.3589,
|
|
"mean_token_accuracy": 0.1646078944206238,
|
|
"num_tokens": 5870332.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.563276863098144,
|
|
"epoch": 2.621473354231975,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045724074988219343,
|
|
"loss": 5.5582,
|
|
"mean_token_accuracy": 0.152925768494606,
|
|
"num_tokens": 5879240.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.484467077255249,
|
|
"epoch": 2.6253918495297803,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045706417730524565,
|
|
"loss": 5.304,
|
|
"mean_token_accuracy": 0.16194796562194824,
|
|
"num_tokens": 5887961.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.6248410701751705,
|
|
"epoch": 2.6293103448275863,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004568872793500927,
|
|
"loss": 5.5209,
|
|
"mean_token_accuracy": 0.16444982439279557,
|
|
"num_tokens": 5896551.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.534376049041748,
|
|
"epoch": 2.633228840125392,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045671005633287986,
|
|
"loss": 5.3111,
|
|
"mean_token_accuracy": 0.1659606784582138,
|
|
"num_tokens": 5905485.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.540153169631958,
|
|
"epoch": 2.6371473354231973,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004565325085703336,
|
|
"loss": 5.3428,
|
|
"mean_token_accuracy": 0.16035765260457993,
|
|
"num_tokens": 5915166.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.600683212280273,
|
|
"epoch": 2.6410658307210033,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004563546363797602,
|
|
"loss": 5.3904,
|
|
"mean_token_accuracy": 0.1670147806406021,
|
|
"num_tokens": 5923782.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.589961290359497,
|
|
"epoch": 2.644984326018809,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004561764400790465,
|
|
"loss": 5.5005,
|
|
"mean_token_accuracy": 0.1543971784412861,
|
|
"num_tokens": 5931904.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.6052652359008786,
|
|
"epoch": 2.6489028213166144,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045599791998665796,
|
|
"loss": 5.3691,
|
|
"mean_token_accuracy": 0.1644495204091072,
|
|
"num_tokens": 5940067.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.548397159576416,
|
|
"epoch": 2.6528213166144203,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004558190764216389,
|
|
"loss": 5.3889,
|
|
"mean_token_accuracy": 0.1678234726190567,
|
|
"num_tokens": 5948663.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.630910778045655,
|
|
"epoch": 2.656739811912226,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004556399097036119,
|
|
"loss": 5.3834,
|
|
"mean_token_accuracy": 0.16502148658037186,
|
|
"num_tokens": 5957224.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.590204524993896,
|
|
"epoch": 2.6606583072100314,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004554604201527768,
|
|
"loss": 5.486,
|
|
"mean_token_accuracy": 0.16515010595321655,
|
|
"num_tokens": 5965604.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.530299282073974,
|
|
"epoch": 2.664576802507837,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00045528060808991075,
|
|
"loss": 5.3126,
|
|
"mean_token_accuracy": 0.171157406270504,
|
|
"num_tokens": 5974309.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.555624628067017,
|
|
"epoch": 2.6684952978056424,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004551004738363669,
|
|
"loss": 5.3143,
|
|
"mean_token_accuracy": 0.16621147245168685,
|
|
"num_tokens": 5982315.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.516742372512818,
|
|
"epoch": 2.6724137931034484,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045492001771407434,
|
|
"loss": 5.3561,
|
|
"mean_token_accuracy": 0.17239132523536682,
|
|
"num_tokens": 5990515.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.569012784957886,
|
|
"epoch": 2.676332288401254,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004547392400455374,
|
|
"loss": 5.5178,
|
|
"mean_token_accuracy": 0.16187777370214462,
|
|
"num_tokens": 6000431.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.597844839096069,
|
|
"epoch": 2.6802507836990594,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004545581411538353,
|
|
"loss": 5.4443,
|
|
"mean_token_accuracy": 0.15897563546895982,
|
|
"num_tokens": 6009022.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.594429779052734,
|
|
"epoch": 2.6841692789968654,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045437672136262083,
|
|
"loss": 5.3694,
|
|
"mean_token_accuracy": 0.16563894748687744,
|
|
"num_tokens": 6019330.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.486822032928467,
|
|
"epoch": 2.688087774294671,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004541949809961208,
|
|
"loss": 5.3737,
|
|
"mean_token_accuracy": 0.1706068456172943,
|
|
"num_tokens": 6027492.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.587631607055664,
|
|
"epoch": 2.6920062695924765,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004540129203791346,
|
|
"loss": 5.3991,
|
|
"mean_token_accuracy": 0.16987380385398865,
|
|
"num_tokens": 6036467.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.54700608253479,
|
|
"epoch": 2.695924764890282,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045383053983703413,
|
|
"loss": 5.3714,
|
|
"mean_token_accuracy": 0.16302234828472137,
|
|
"num_tokens": 6045711.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.588875722885132,
|
|
"epoch": 2.6998432601880875,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 5.43,
|
|
"mean_token_accuracy": 0.15987755954265595,
|
|
"num_tokens": 6054626.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.5814910411834715,
|
|
"epoch": 2.7037617554858935,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045346482028183583,
|
|
"loss": 5.4934,
|
|
"mean_token_accuracy": 0.16142257750034333,
|
|
"num_tokens": 6064118.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.575299453735352,
|
|
"epoch": 2.707680250783699,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045328148192233823,
|
|
"loss": 5.4026,
|
|
"mean_token_accuracy": 0.1676523730158806,
|
|
"num_tokens": 6072723.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.561681270599365,
|
|
"epoch": 2.7115987460815045,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004530978249449254,
|
|
"loss": 5.3919,
|
|
"mean_token_accuracy": 0.16447827816009522,
|
|
"num_tokens": 6081535.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.527007961273194,
|
|
"epoch": 2.7155172413793105,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004529138496778222,
|
|
"loss": 5.3236,
|
|
"mean_token_accuracy": 0.16587052047252654,
|
|
"num_tokens": 6090524.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.573948192596435,
|
|
"epoch": 2.719435736677116,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004527295564498222,
|
|
"loss": 5.4102,
|
|
"mean_token_accuracy": 0.16189964413642882,
|
|
"num_tokens": 6099870.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.568688821792603,
|
|
"epoch": 2.7233542319749215,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004525449455902874,
|
|
"loss": 5.3849,
|
|
"mean_token_accuracy": 0.16267680674791335,
|
|
"num_tokens": 6107760.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.5809601783752445,
|
|
"epoch": 2.7272727272727275,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004523600174291473,
|
|
"loss": 5.4656,
|
|
"mean_token_accuracy": 0.16025887057185173,
|
|
"num_tokens": 6117135.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.544527196884156,
|
|
"epoch": 2.731191222570533,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004521747722968985,
|
|
"loss": 5.3902,
|
|
"mean_token_accuracy": 0.1645651176571846,
|
|
"num_tokens": 6125711.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.55906310081482,
|
|
"epoch": 2.7351097178683386,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045198921052460396,
|
|
"loss": 5.4568,
|
|
"mean_token_accuracy": 0.15940805822610854,
|
|
"num_tokens": 6134623.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.598245477676391,
|
|
"epoch": 2.739028213166144,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004518033324438928,
|
|
"loss": 5.4533,
|
|
"mean_token_accuracy": 0.1645047664642334,
|
|
"num_tokens": 6143228.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.548443794250488,
|
|
"epoch": 2.7429467084639496,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004516171383869593,
|
|
"loss": 5.4337,
|
|
"mean_token_accuracy": 0.16507748067378997,
|
|
"num_tokens": 6152021.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.7429467084639496,
|
|
"eval_entropy": 5.416865946710572,
|
|
"eval_loss": 5.858214378356934,
|
|
"eval_mean_token_accuracy": 0.15347288648576238,
|
|
"eval_num_tokens": 6152021.0,
|
|
"eval_runtime": 2.8653,
|
|
"eval_samples_per_second": 1438.589,
|
|
"eval_steps_per_second": 180.085,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.621888637542725,
|
|
"epoch": 2.7468652037617556,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045143062868656234,
|
|
"loss": 5.4853,
|
|
"mean_token_accuracy": 0.16335086822509765,
|
|
"num_tokens": 6161389.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.6750434875488285,
|
|
"epoch": 2.750783699059561,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000451243803676025,
|
|
"loss": 5.4182,
|
|
"mean_token_accuracy": 0.16270672380924225,
|
|
"num_tokens": 6170381.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.5134519100189205,
|
|
"epoch": 2.7547021943573666,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045105666368923397,
|
|
"loss": 5.365,
|
|
"mean_token_accuracy": 0.16540434807538987,
|
|
"num_tokens": 6179071.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.488037967681885,
|
|
"epoch": 2.7586206896551726,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045086920906063866,
|
|
"loss": 5.3753,
|
|
"mean_token_accuracy": 0.16212851256132127,
|
|
"num_tokens": 6187896.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.5593883991241455,
|
|
"epoch": 2.762539184952978,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045068144012525095,
|
|
"loss": 5.3431,
|
|
"mean_token_accuracy": 0.17042842507362366,
|
|
"num_tokens": 6196328.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.472257471084594,
|
|
"epoch": 2.7664576802507836,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045049335721864426,
|
|
"loss": 5.3834,
|
|
"mean_token_accuracy": 0.16024037301540375,
|
|
"num_tokens": 6204703.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.513653898239136,
|
|
"epoch": 2.7703761755485896,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045030496067695336,
|
|
"loss": 5.3696,
|
|
"mean_token_accuracy": 0.16477195769548417,
|
|
"num_tokens": 6213496.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.569023132324219,
|
|
"epoch": 2.774294670846395,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004501162508368733,
|
|
"loss": 5.3278,
|
|
"mean_token_accuracy": 0.17372318655252456,
|
|
"num_tokens": 6222112.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.556356048583984,
|
|
"epoch": 2.7782131661442007,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004499272280356594,
|
|
"loss": 5.4361,
|
|
"mean_token_accuracy": 0.15649492889642716,
|
|
"num_tokens": 6230288.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.499981451034546,
|
|
"epoch": 2.782131661442006,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004497378926111257,
|
|
"loss": 5.3108,
|
|
"mean_token_accuracy": 0.17359569519758225,
|
|
"num_tokens": 6238608.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.534588098526001,
|
|
"epoch": 2.7860501567398117,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004495482449016456,
|
|
"loss": 5.4406,
|
|
"mean_token_accuracy": 0.1648330047726631,
|
|
"num_tokens": 6247902.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.5432658195495605,
|
|
"epoch": 2.7899686520376177,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004493582852461501,
|
|
"loss": 5.4038,
|
|
"mean_token_accuracy": 0.16332560032606125,
|
|
"num_tokens": 6257433.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.543182420730591,
|
|
"epoch": 2.793887147335423,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004491680139841281,
|
|
"loss": 5.3766,
|
|
"mean_token_accuracy": 0.1634378582239151,
|
|
"num_tokens": 6266017.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.605280590057373,
|
|
"epoch": 2.7978056426332287,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000448977431455625,
|
|
"loss": 5.471,
|
|
"mean_token_accuracy": 0.15485770851373673,
|
|
"num_tokens": 6274759.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.60925555229187,
|
|
"epoch": 2.8017241379310347,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044878653800124285,
|
|
"loss": 5.3302,
|
|
"mean_token_accuracy": 0.16837369352579118,
|
|
"num_tokens": 6283699.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.5365588665008545,
|
|
"epoch": 2.80564263322884,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004485953339621391,
|
|
"loss": 5.3709,
|
|
"mean_token_accuracy": 0.16740549057722093,
|
|
"num_tokens": 6292562.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.51601734161377,
|
|
"epoch": 2.8095611285266457,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004484038196800265,
|
|
"loss": 5.3509,
|
|
"mean_token_accuracy": 0.1703738674521446,
|
|
"num_tokens": 6300928.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.493614912033081,
|
|
"epoch": 2.8134796238244513,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004482119954971719,
|
|
"loss": 5.2886,
|
|
"mean_token_accuracy": 0.1659258261322975,
|
|
"num_tokens": 6309994.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.549010610580444,
|
|
"epoch": 2.817398119122257,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00044801986175639635,
|
|
"loss": 5.3782,
|
|
"mean_token_accuracy": 0.16762072443962098,
|
|
"num_tokens": 6318856.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.505040168762207,
|
|
"epoch": 2.8213166144200628,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004478274188010741,
|
|
"loss": 5.3889,
|
|
"mean_token_accuracy": 0.16300486773252487,
|
|
"num_tokens": 6327274.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.617763519287109,
|
|
"epoch": 2.8252351097178683,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00044763466697513173,
|
|
"loss": 5.4573,
|
|
"mean_token_accuracy": 0.17388699054718018,
|
|
"num_tokens": 6335957.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.585629558563232,
|
|
"epoch": 2.829153605015674,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00044744160662304805,
|
|
"loss": 5.4605,
|
|
"mean_token_accuracy": 0.1734134614467621,
|
|
"num_tokens": 6344673.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.4783607006073,
|
|
"epoch": 2.83307210031348,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044724823808985325,
|
|
"loss": 5.3171,
|
|
"mean_token_accuracy": 0.1722720220685005,
|
|
"num_tokens": 6353085.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.553834056854248,
|
|
"epoch": 2.8369905956112853,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004470545617211283,
|
|
"loss": 5.373,
|
|
"mean_token_accuracy": 0.16508337557315828,
|
|
"num_tokens": 6362107.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.57023286819458,
|
|
"epoch": 2.840909090909091,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044686057786300423,
|
|
"loss": 5.3586,
|
|
"mean_token_accuracy": 0.1676985129714012,
|
|
"num_tokens": 6370381.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.514395332336425,
|
|
"epoch": 2.844827586206897,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044666628686216154,
|
|
"loss": 5.3961,
|
|
"mean_token_accuracy": 0.16086816638708115,
|
|
"num_tokens": 6378790.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.505329370498657,
|
|
"epoch": 2.8487460815047023,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044647168906583,
|
|
"loss": 5.3777,
|
|
"mean_token_accuracy": 0.16837207078933716,
|
|
"num_tokens": 6387997.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.62245397567749,
|
|
"epoch": 2.852664576802508,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044627678482178716,
|
|
"loss": 5.3878,
|
|
"mean_token_accuracy": 0.16976817399263383,
|
|
"num_tokens": 6397458.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.551675081253052,
|
|
"epoch": 2.8565830721003134,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004460815744783587,
|
|
"loss": 5.3445,
|
|
"mean_token_accuracy": 0.16652641594409942,
|
|
"num_tokens": 6405481.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.540624332427979,
|
|
"epoch": 2.860501567398119,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000445886058384417,
|
|
"loss": 5.455,
|
|
"mean_token_accuracy": 0.1594948023557663,
|
|
"num_tokens": 6414087.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.594117116928101,
|
|
"epoch": 2.864420062695925,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004456902368893811,
|
|
"loss": 5.2854,
|
|
"mean_token_accuracy": 0.17743499875068663,
|
|
"num_tokens": 6421934.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.4886205196380615,
|
|
"epoch": 2.8683385579937304,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004454941103432158,
|
|
"loss": 5.2909,
|
|
"mean_token_accuracy": 0.1713653177022934,
|
|
"num_tokens": 6431028.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.430035018920899,
|
|
"epoch": 2.872257053291536,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044529767909643093,
|
|
"loss": 5.2644,
|
|
"mean_token_accuracy": 0.1731930896639824,
|
|
"num_tokens": 6439913.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.572323274612427,
|
|
"epoch": 2.876175548589342,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004451009435000811,
|
|
"loss": 5.3442,
|
|
"mean_token_accuracy": 0.1642979145050049,
|
|
"num_tokens": 6449161.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.4994124412536625,
|
|
"epoch": 2.8800940438871474,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004449039039057647,
|
|
"loss": 5.4629,
|
|
"mean_token_accuracy": 0.167383174598217,
|
|
"num_tokens": 6457834.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.481597995758056,
|
|
"epoch": 2.884012539184953,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 5.2848,
|
|
"mean_token_accuracy": 0.1744975820183754,
|
|
"num_tokens": 6465879.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.4826271533966064,
|
|
"epoch": 2.887931034482759,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004445089141323415,
|
|
"loss": 5.2887,
|
|
"mean_token_accuracy": 0.1704733982682228,
|
|
"num_tokens": 6474801.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.492102003097534,
|
|
"epoch": 2.8918495297805644,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00044431096465914554,
|
|
"loss": 5.3379,
|
|
"mean_token_accuracy": 0.17320799082517624,
|
|
"num_tokens": 6483532.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.512108945846558,
|
|
"epoch": 2.89576802507837,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044411271259980315,
|
|
"loss": 5.3198,
|
|
"mean_token_accuracy": 0.1696050226688385,
|
|
"num_tokens": 6491372.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.45632038116455,
|
|
"epoch": 2.8996865203761755,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004439141583086231,
|
|
"loss": 5.3132,
|
|
"mean_token_accuracy": 0.1632131263613701,
|
|
"num_tokens": 6500001.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.577138662338257,
|
|
"epoch": 2.903605015673981,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044371530214045395,
|
|
"loss": 5.4162,
|
|
"mean_token_accuracy": 0.1663136973977089,
|
|
"num_tokens": 6507750.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.488527536392212,
|
|
"epoch": 2.907523510971787,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044351614445068413,
|
|
"loss": 5.3099,
|
|
"mean_token_accuracy": 0.1647246077656746,
|
|
"num_tokens": 6517888.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.556266832351684,
|
|
"epoch": 2.9114420062695925,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044331668559524043,
|
|
"loss": 5.4259,
|
|
"mean_token_accuracy": 0.16335552632808686,
|
|
"num_tokens": 6526573.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.525750923156738,
|
|
"epoch": 2.915360501567398,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004431169259305883,
|
|
"loss": 5.3663,
|
|
"mean_token_accuracy": 0.16534322053194045,
|
|
"num_tokens": 6536544.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.561939907073975,
|
|
"epoch": 2.919278996865204,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004429168658137306,
|
|
"loss": 5.3852,
|
|
"mean_token_accuracy": 0.15881865173578263,
|
|
"num_tokens": 6544982.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.509143447875976,
|
|
"epoch": 2.9231974921630095,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044271650560220746,
|
|
"loss": 5.3933,
|
|
"mean_token_accuracy": 0.16829295456409454,
|
|
"num_tokens": 6553664.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.5341473579406735,
|
|
"epoch": 2.927115987460815,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044251584565409464,
|
|
"loss": 5.3926,
|
|
"mean_token_accuracy": 0.16476417034864427,
|
|
"num_tokens": 6562234.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.508896017074585,
|
|
"epoch": 2.9310344827586206,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004423148863280044,
|
|
"loss": 5.3451,
|
|
"mean_token_accuracy": 0.16751128137111665,
|
|
"num_tokens": 6570707.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.582256126403808,
|
|
"epoch": 2.934952978056426,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044211362798308334,
|
|
"loss": 5.3953,
|
|
"mean_token_accuracy": 0.16314985156059264,
|
|
"num_tokens": 6580054.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.5249100685119625,
|
|
"epoch": 2.938871473354232,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004419120709790129,
|
|
"loss": 5.437,
|
|
"mean_token_accuracy": 0.1647406041622162,
|
|
"num_tokens": 6588585.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.583944368362427,
|
|
"epoch": 2.9427899686520376,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044171021567600814,
|
|
"loss": 5.4412,
|
|
"mean_token_accuracy": 0.16572290360927583,
|
|
"num_tokens": 6597290.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.575886011123657,
|
|
"epoch": 2.946708463949843,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044150806243481715,
|
|
"loss": 5.3613,
|
|
"mean_token_accuracy": 0.1625833809375763,
|
|
"num_tokens": 6604966.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.518132829666138,
|
|
"epoch": 2.950626959247649,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004413056116167206,
|
|
"loss": 5.3261,
|
|
"mean_token_accuracy": 0.17080285102128984,
|
|
"num_tokens": 6613588.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.5981645584106445,
|
|
"epoch": 2.9545454545454546,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004411028635835309,
|
|
"loss": 5.5663,
|
|
"mean_token_accuracy": 0.15079731792211531,
|
|
"num_tokens": 6622907.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.522223806381225,
|
|
"epoch": 2.95846394984326,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004408998186975917,
|
|
"loss": 5.3584,
|
|
"mean_token_accuracy": 0.16596754342317582,
|
|
"num_tokens": 6632074.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.575008773803711,
|
|
"epoch": 2.962382445141066,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00044069647732177696,
|
|
"loss": 5.3715,
|
|
"mean_token_accuracy": 0.16168408691883088,
|
|
"num_tokens": 6640595.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.595236015319824,
|
|
"epoch": 2.9663009404388716,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044049283981949103,
|
|
"loss": 5.4159,
|
|
"mean_token_accuracy": 0.16023389548063277,
|
|
"num_tokens": 6649861.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.566992712020874,
|
|
"epoch": 2.970219435736677,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004402889065546667,
|
|
"loss": 5.4592,
|
|
"mean_token_accuracy": 0.1607225641608238,
|
|
"num_tokens": 6659157.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.548617935180664,
|
|
"epoch": 2.9741379310344827,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044008467789176625,
|
|
"loss": 5.42,
|
|
"mean_token_accuracy": 0.16288147270679473,
|
|
"num_tokens": 6667979.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.5352644443511965,
|
|
"epoch": 2.978056426332288,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004398801541957791,
|
|
"loss": 5.3702,
|
|
"mean_token_accuracy": 0.16329635232686995,
|
|
"num_tokens": 6677239.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.594531679153443,
|
|
"epoch": 2.981974921630094,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004396753358322223,
|
|
"loss": 5.4912,
|
|
"mean_token_accuracy": 0.1588613659143448,
|
|
"num_tokens": 6687317.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.4902747631072994,
|
|
"epoch": 2.9858934169278997,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004394702231671396,
|
|
"loss": 5.361,
|
|
"mean_token_accuracy": 0.16307796984910966,
|
|
"num_tokens": 6695732.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.516343259811402,
|
|
"epoch": 2.989811912225705,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004392648165671004,
|
|
"loss": 5.2819,
|
|
"mean_token_accuracy": 0.17569045573472977,
|
|
"num_tokens": 6704272.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.494945240020752,
|
|
"epoch": 2.993730407523511,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004390591163991998,
|
|
"loss": 5.3404,
|
|
"mean_token_accuracy": 0.16861142218112946,
|
|
"num_tokens": 6713355.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.574056911468506,
|
|
"epoch": 2.9976489028213167,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043885312303105725,
|
|
"loss": 5.4232,
|
|
"mean_token_accuracy": 0.1638496369123459,
|
|
"num_tokens": 6721952.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.500212478637695,
|
|
"epoch": 3.001567398119122,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004386468368308163,
|
|
"loss": 5.2236,
|
|
"mean_token_accuracy": 0.17236360609531404,
|
|
"num_tokens": 6730582.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.510211610794068,
|
|
"epoch": 3.0054858934169277,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004384402581671438,
|
|
"loss": 5.0182,
|
|
"mean_token_accuracy": 0.18019506633281707,
|
|
"num_tokens": 6739370.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.498809146881103,
|
|
"epoch": 3.0094043887147337,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004382333874092295,
|
|
"loss": 5.0093,
|
|
"mean_token_accuracy": 0.1874949023127556,
|
|
"num_tokens": 6748282.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.55729079246521,
|
|
"epoch": 3.0133228840125392,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043802622492678466,
|
|
"loss": 5.1104,
|
|
"mean_token_accuracy": 0.17181859612464906,
|
|
"num_tokens": 6757622.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.540243196487427,
|
|
"epoch": 3.0172413793103448,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004378187710900426,
|
|
"loss": 5.0617,
|
|
"mean_token_accuracy": 0.17972142547369002,
|
|
"num_tokens": 6765780.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.506430625915527,
|
|
"epoch": 3.0211598746081503,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043761102626975674,
|
|
"loss": 5.0646,
|
|
"mean_token_accuracy": 0.18475002646446229,
|
|
"num_tokens": 6774343.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.449203777313232,
|
|
"epoch": 3.0250783699059562,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004374029908372007,
|
|
"loss": 5.0548,
|
|
"mean_token_accuracy": 0.18161061704158782,
|
|
"num_tokens": 6783765.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.581590366363526,
|
|
"epoch": 3.0289968652037618,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043719466516416774,
|
|
"loss": 5.1186,
|
|
"mean_token_accuracy": 0.17392480820417405,
|
|
"num_tokens": 6792102.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.5454864501953125,
|
|
"epoch": 3.0329153605015673,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043698604962296946,
|
|
"loss": 5.0389,
|
|
"mean_token_accuracy": 0.1776745066046715,
|
|
"num_tokens": 6801435.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.405967092514038,
|
|
"epoch": 3.0368338557993733,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043677714458643566,
|
|
"loss": 5.0834,
|
|
"mean_token_accuracy": 0.1805256873369217,
|
|
"num_tokens": 6810409.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.38959321975708,
|
|
"epoch": 3.040752351097179,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043656795042791357,
|
|
"loss": 4.9489,
|
|
"mean_token_accuracy": 0.19217908829450608,
|
|
"num_tokens": 6818752.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.4814122200012205,
|
|
"epoch": 3.0446708463949843,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004363584675212671,
|
|
"loss": 5.0562,
|
|
"mean_token_accuracy": 0.1762182220816612,
|
|
"num_tokens": 6827880.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.424868869781494,
|
|
"epoch": 3.04858934169279,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004361486962408761,
|
|
"loss": 5.042,
|
|
"mean_token_accuracy": 0.18900462985038757,
|
|
"num_tokens": 6835865.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.433160018920899,
|
|
"epoch": 3.052507836990596,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004359386369616359,
|
|
"loss": 5.0042,
|
|
"mean_token_accuracy": 0.18738968521356583,
|
|
"num_tokens": 6843851.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.4276283264160154,
|
|
"epoch": 3.0564263322884013,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004357282900589565,
|
|
"loss": 5.0398,
|
|
"mean_token_accuracy": 0.18537967801094055,
|
|
"num_tokens": 6852649.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.448413610458374,
|
|
"epoch": 3.060344827586207,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043551765590876183,
|
|
"loss": 5.0023,
|
|
"mean_token_accuracy": 0.18639151155948638,
|
|
"num_tokens": 6861607.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.4296623229980465,
|
|
"epoch": 3.0642633228840124,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004353067348874894,
|
|
"loss": 5.0492,
|
|
"mean_token_accuracy": 0.17867960333824157,
|
|
"num_tokens": 6870650.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.475912618637085,
|
|
"epoch": 3.0681818181818183,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 5.0914,
|
|
"mean_token_accuracy": 0.1736911416053772,
|
|
"num_tokens": 6879820.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.374644184112549,
|
|
"epoch": 3.072100313479624,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004348840337400233,
|
|
"loss": 5.0022,
|
|
"mean_token_accuracy": 0.18161854892969131,
|
|
"num_tokens": 6888329.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.419334554672242,
|
|
"epoch": 3.0760188087774294,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043467225436926517,
|
|
"loss": 5.0629,
|
|
"mean_token_accuracy": 0.17774185985326768,
|
|
"num_tokens": 6897635.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.478842735290527,
|
|
"epoch": 3.079937304075235,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004344601896382988,
|
|
"loss": 5.1145,
|
|
"mean_token_accuracy": 0.18426743000745774,
|
|
"num_tokens": 6907275.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.451950645446777,
|
|
"epoch": 3.083855799373041,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043424783992611837,
|
|
"loss": 5.0555,
|
|
"mean_token_accuracy": 0.17988815158605576,
|
|
"num_tokens": 6916377.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.4796899318695065,
|
|
"epoch": 3.0877742946708464,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043403520561222705,
|
|
"loss": 4.9582,
|
|
"mean_token_accuracy": 0.19082407802343368,
|
|
"num_tokens": 6924355.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.400261449813843,
|
|
"epoch": 3.091692789968652,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004338222870766371,
|
|
"loss": 5.0629,
|
|
"mean_token_accuracy": 0.18493644297122955,
|
|
"num_tokens": 6933328.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.465732622146606,
|
|
"epoch": 3.0956112852664575,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043360908469986827,
|
|
"loss": 5.0599,
|
|
"mean_token_accuracy": 0.18203602880239486,
|
|
"num_tokens": 6942189.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.416319704055786,
|
|
"epoch": 3.0995297805642634,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004333955988629478,
|
|
"loss": 5.0397,
|
|
"mean_token_accuracy": 0.18563192188739777,
|
|
"num_tokens": 6951188.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.408259057998658,
|
|
"epoch": 3.103448275862069,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043318182994740945,
|
|
"loss": 4.9923,
|
|
"mean_token_accuracy": 0.19361870884895324,
|
|
"num_tokens": 6958718.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.420421409606933,
|
|
"epoch": 3.1073667711598745,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004329677783352931,
|
|
"loss": 5.065,
|
|
"mean_token_accuracy": 0.1767393723130226,
|
|
"num_tokens": 6968040.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.443540143966675,
|
|
"epoch": 3.1112852664576804,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004327534444091436,
|
|
"loss": 5.0618,
|
|
"mean_token_accuracy": 0.1819816365838051,
|
|
"num_tokens": 6976977.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.451086139678955,
|
|
"epoch": 3.115203761755486,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00043253882855201037,
|
|
"loss": 5.0475,
|
|
"mean_token_accuracy": 0.18677449077367783,
|
|
"num_tokens": 6985859.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.316517972946167,
|
|
"epoch": 3.1191222570532915,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043232393114744683,
|
|
"loss": 4.9638,
|
|
"mean_token_accuracy": 0.19123670011758803,
|
|
"num_tokens": 6994031.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.365807771682739,
|
|
"epoch": 3.123040752351097,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004321087525795095,
|
|
"loss": 5.0323,
|
|
"mean_token_accuracy": 0.18311454802751542,
|
|
"num_tokens": 7002527.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.451259613037109,
|
|
"epoch": 3.126959247648903,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004318932932327573,
|
|
"loss": 5.135,
|
|
"mean_token_accuracy": 0.1824721798300743,
|
|
"num_tokens": 7011983.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.460281324386597,
|
|
"epoch": 3.1308777429467085,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000431677553492251,
|
|
"loss": 5.0778,
|
|
"mean_token_accuracy": 0.17979770451784133,
|
|
"num_tokens": 7020819.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.404204511642456,
|
|
"epoch": 3.134796238244514,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043146153374355256,
|
|
"loss": 4.9475,
|
|
"mean_token_accuracy": 0.18701418936252595,
|
|
"num_tokens": 7029257.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.134796238244514,
|
|
"eval_entropy": 5.3317289518755535,
|
|
"eval_loss": 5.822052955627441,
|
|
"eval_mean_token_accuracy": 0.15691759988896606,
|
|
"eval_num_tokens": 7029257.0,
|
|
"eval_runtime": 2.8485,
|
|
"eval_samples_per_second": 1447.088,
|
|
"eval_steps_per_second": 181.149,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.4291181564331055,
|
|
"epoch": 3.1387147335423196,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043124523437272427,
|
|
"loss": 5.1376,
|
|
"mean_token_accuracy": 0.17223237901926042,
|
|
"num_tokens": 7038942.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.427195024490357,
|
|
"epoch": 3.1426332288401255,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004310286557663282,
|
|
"loss": 5.0969,
|
|
"mean_token_accuracy": 0.1853697270154953,
|
|
"num_tokens": 7048074.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.459545612335205,
|
|
"epoch": 3.146551724137931,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004308117983114254,
|
|
"loss": 5.0182,
|
|
"mean_token_accuracy": 0.18311821520328522,
|
|
"num_tokens": 7056887.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.4102888107299805,
|
|
"epoch": 3.1504702194357366,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004305946623955754,
|
|
"loss": 4.989,
|
|
"mean_token_accuracy": 0.19069620966911316,
|
|
"num_tokens": 7065462.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.422263050079346,
|
|
"epoch": 3.1543887147335425,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00043037724840683516,
|
|
"loss": 5.0992,
|
|
"mean_token_accuracy": 0.17892909795045853,
|
|
"num_tokens": 7074188.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.393792390823364,
|
|
"epoch": 3.158307210031348,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00043015955673375876,
|
|
"loss": 5.0792,
|
|
"mean_token_accuracy": 0.18168304413557052,
|
|
"num_tokens": 7082282.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.467434215545654,
|
|
"epoch": 3.1622257053291536,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004299415877653966,
|
|
"loss": 5.1135,
|
|
"mean_token_accuracy": 0.17735314518213272,
|
|
"num_tokens": 7090631.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.447216415405274,
|
|
"epoch": 3.166144200626959,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004297233418912945,
|
|
"loss": 5.0586,
|
|
"mean_token_accuracy": 0.17843555361032487,
|
|
"num_tokens": 7099546.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.415751600265503,
|
|
"epoch": 3.170062695924765,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004295048195014932,
|
|
"loss": 5.1124,
|
|
"mean_token_accuracy": 0.1767762377858162,
|
|
"num_tokens": 7109423.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.361036872863769,
|
|
"epoch": 3.1739811912225706,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004292860209865277,
|
|
"loss": 4.9716,
|
|
"mean_token_accuracy": 0.18291346430778505,
|
|
"num_tokens": 7117717.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.42275185585022,
|
|
"epoch": 3.177899686520376,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004290669467374263,
|
|
"loss": 5.2065,
|
|
"mean_token_accuracy": 0.16665955036878585,
|
|
"num_tokens": 7126573.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.434532737731933,
|
|
"epoch": 3.1818181818181817,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00042884759714571037,
|
|
"loss": 5.0977,
|
|
"mean_token_accuracy": 0.18158412277698516,
|
|
"num_tokens": 7134471.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.35801944732666,
|
|
"epoch": 3.1857366771159876,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004286279726033932,
|
|
"loss": 5.0233,
|
|
"mean_token_accuracy": 0.18104632943868637,
|
|
"num_tokens": 7143092.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.481039714813233,
|
|
"epoch": 3.189655172413793,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00042840807350297933,
|
|
"loss": 5.1079,
|
|
"mean_token_accuracy": 0.17933820486068724,
|
|
"num_tokens": 7152137.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.424618101119995,
|
|
"epoch": 3.1935736677115987,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00042818790023746407,
|
|
"loss": 5.0628,
|
|
"mean_token_accuracy": 0.17451367080211638,
|
|
"num_tokens": 7160037.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.3562406539917,
|
|
"epoch": 3.197492163009404,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042796745320033296,
|
|
"loss": 5.0451,
|
|
"mean_token_accuracy": 0.1772144466638565,
|
|
"num_tokens": 7168354.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.420835638046265,
|
|
"epoch": 3.20141065830721,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00042774673278556043,
|
|
"loss": 5.1368,
|
|
"mean_token_accuracy": 0.17543467730283738,
|
|
"num_tokens": 7176757.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.473498201370239,
|
|
"epoch": 3.2053291536050157,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004275257393876097,
|
|
"loss": 5.1117,
|
|
"mean_token_accuracy": 0.17528067976236344,
|
|
"num_tokens": 7185489.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.4419642925262455,
|
|
"epoch": 3.209247648902821,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004273044734014318,
|
|
"loss": 5.0455,
|
|
"mean_token_accuracy": 0.18430220037698747,
|
|
"num_tokens": 7194431.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.3320338249206545,
|
|
"epoch": 3.2131661442006267,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00042708293522246486,
|
|
"loss": 5.0235,
|
|
"mean_token_accuracy": 0.18735198825597763,
|
|
"num_tokens": 7203074.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.444218158721924,
|
|
"epoch": 3.2170846394984327,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004268611252466337,
|
|
"loss": 5.0679,
|
|
"mean_token_accuracy": 0.18504944145679475,
|
|
"num_tokens": 7211692.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.351670360565185,
|
|
"epoch": 3.2210031347962382,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004266390438703486,
|
|
"loss": 5.0555,
|
|
"mean_token_accuracy": 0.18174369037151336,
|
|
"num_tokens": 7220285.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.485449695587159,
|
|
"epoch": 3.2249216300940438,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042641669149050493,
|
|
"loss": 5.1643,
|
|
"mean_token_accuracy": 0.17582350373268127,
|
|
"num_tokens": 7230007.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.431193876266479,
|
|
"epoch": 3.2288401253918497,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004261940685044825,
|
|
"loss": 5.0458,
|
|
"mean_token_accuracy": 0.1844669461250305,
|
|
"num_tokens": 7238520.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.380627584457398,
|
|
"epoch": 3.2327586206896552,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042597117531014474,
|
|
"loss": 4.9787,
|
|
"mean_token_accuracy": 0.18394738882780076,
|
|
"num_tokens": 7247228.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.475986003875732,
|
|
"epoch": 3.2366771159874608,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004257480123058378,
|
|
"loss": 5.2206,
|
|
"mean_token_accuracy": 0.17249116897583008,
|
|
"num_tokens": 7255782.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.481035137176514,
|
|
"epoch": 3.2405956112852663,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042552457989039036,
|
|
"loss": 5.0917,
|
|
"mean_token_accuracy": 0.17183260768651962,
|
|
"num_tokens": 7264946.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.279617786407471,
|
|
"epoch": 3.2445141065830723,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042530087846311213,
|
|
"loss": 4.9343,
|
|
"mean_token_accuracy": 0.19145012497901917,
|
|
"num_tokens": 7273647.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.404915523529053,
|
|
"epoch": 3.248432601880878,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00042507690842379396,
|
|
"loss": 5.0686,
|
|
"mean_token_accuracy": 0.18024197965860367,
|
|
"num_tokens": 7283393.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.501522541046143,
|
|
"epoch": 3.2523510971786833,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 5.1691,
|
|
"mean_token_accuracy": 0.1766907885670662,
|
|
"num_tokens": 7292368.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.373268365859985,
|
|
"epoch": 3.256269592476489,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042462816411060025,
|
|
"loss": 5.0944,
|
|
"mean_token_accuracy": 0.1837262213230133,
|
|
"num_tokens": 7301505.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.413186407089233,
|
|
"epoch": 3.260188087774295,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004244033906387035,
|
|
"loss": 5.0505,
|
|
"mean_token_accuracy": 0.1898556739091873,
|
|
"num_tokens": 7310011.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.458008861541748,
|
|
"epoch": 3.2641065830721003,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004241783501587231,
|
|
"loss": 5.0878,
|
|
"mean_token_accuracy": 0.185248726606369,
|
|
"num_tokens": 7319295.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.448028945922852,
|
|
"epoch": 3.268025078369906,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00042395304307284284,
|
|
"loss": 5.1108,
|
|
"mean_token_accuracy": 0.18034909218549727,
|
|
"num_tokens": 7328390.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.476062536239624,
|
|
"epoch": 3.271943573667712,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004237274697837229,
|
|
"loss": 5.1413,
|
|
"mean_token_accuracy": 0.17422386407852172,
|
|
"num_tokens": 7337062.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.385330677032471,
|
|
"epoch": 3.2758620689655173,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004235016306944996,
|
|
"loss": 5.1225,
|
|
"mean_token_accuracy": 0.1811202198266983,
|
|
"num_tokens": 7346364.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.42358512878418,
|
|
"epoch": 3.279780564263323,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004232755262087837,
|
|
"loss": 5.0718,
|
|
"mean_token_accuracy": 0.17982283979654312,
|
|
"num_tokens": 7354896.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.436400270462036,
|
|
"epoch": 3.2836990595611284,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00042304915673066083,
|
|
"loss": 5.1204,
|
|
"mean_token_accuracy": 0.18572596311569214,
|
|
"num_tokens": 7363591.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.431533670425415,
|
|
"epoch": 3.287617554858934,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00042282252266468985,
|
|
"loss": 5.1244,
|
|
"mean_token_accuracy": 0.17862443774938583,
|
|
"num_tokens": 7372891.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.42274022102356,
|
|
"epoch": 3.29153605015674,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004225956244159025,
|
|
"loss": 5.0386,
|
|
"mean_token_accuracy": 0.18299809098243713,
|
|
"num_tokens": 7381945.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.501552629470825,
|
|
"epoch": 3.2954545454545454,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004223684623898029,
|
|
"loss": 5.174,
|
|
"mean_token_accuracy": 0.18130065351724625,
|
|
"num_tokens": 7391449.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.408674716949463,
|
|
"epoch": 3.299373040752351,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004221410369923662,
|
|
"loss": 5.0175,
|
|
"mean_token_accuracy": 0.18889281302690505,
|
|
"num_tokens": 7399762.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.3221982479095455,
|
|
"epoch": 3.303291536050157,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00042191334863003873,
|
|
"loss": 4.9413,
|
|
"mean_token_accuracy": 0.18665122389793395,
|
|
"num_tokens": 7408711.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.3030026912689205,
|
|
"epoch": 3.3072100313479624,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004216853977097363,
|
|
"loss": 5.0447,
|
|
"mean_token_accuracy": 0.1859362781047821,
|
|
"num_tokens": 7416868.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.454968357086182,
|
|
"epoch": 3.311128526645768,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004214571846388442,
|
|
"loss": 5.1428,
|
|
"mean_token_accuracy": 0.1734541654586792,
|
|
"num_tokens": 7425687.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.425799322128296,
|
|
"epoch": 3.3150470219435735,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004212287098252164,
|
|
"loss": 5.0524,
|
|
"mean_token_accuracy": 0.17986010611057282,
|
|
"num_tokens": 7433714.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.372937440872192,
|
|
"epoch": 3.3189655172413794,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004209999736771742,
|
|
"loss": 5.1021,
|
|
"mean_token_accuracy": 0.18723660707473755,
|
|
"num_tokens": 7442765.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.413239336013794,
|
|
"epoch": 3.322884012539185,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004207709766035063,
|
|
"loss": 5.0897,
|
|
"mean_token_accuracy": 0.1808013901114464,
|
|
"num_tokens": 7450672.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.41263747215271,
|
|
"epoch": 3.3268025078369905,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004205417190134674,
|
|
"loss": 5.1479,
|
|
"mean_token_accuracy": 0.1773154303431511,
|
|
"num_tokens": 7458758.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.431203556060791,
|
|
"epoch": 3.330721003134796,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004203122013167783,
|
|
"loss": 5.113,
|
|
"mean_token_accuracy": 0.17936882078647615,
|
|
"num_tokens": 7467527.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.4187102794647215,
|
|
"epoch": 3.334639498432602,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00042008242392362413,
|
|
"loss": 5.1008,
|
|
"mean_token_accuracy": 0.18423034101724625,
|
|
"num_tokens": 7476659.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.3434404850006105,
|
|
"epoch": 3.3385579937304075,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041985238724465433,
|
|
"loss": 5.0159,
|
|
"mean_token_accuracy": 0.19122233092784882,
|
|
"num_tokens": 7485813.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.428614282608033,
|
|
"epoch": 3.342476489028213,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041962209169098193,
|
|
"loss": 5.1351,
|
|
"mean_token_accuracy": 0.1730467364192009,
|
|
"num_tokens": 7494842.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.468219375610351,
|
|
"epoch": 3.346394984326019,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004193915376741823,
|
|
"loss": 5.1443,
|
|
"mean_token_accuracy": 0.17279646843671798,
|
|
"num_tokens": 7503633.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.401078128814698,
|
|
"epoch": 3.3503134796238245,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004191607256062928,
|
|
"loss": 5.0683,
|
|
"mean_token_accuracy": 0.18162844628095626,
|
|
"num_tokens": 7512759.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.411725044250488,
|
|
"epoch": 3.35423197492163,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004189296558998121,
|
|
"loss": 5.1652,
|
|
"mean_token_accuracy": 0.18083991706371308,
|
|
"num_tokens": 7522284.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.4328460693359375,
|
|
"epoch": 3.3581504702194356,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004186983289676992,
|
|
"loss": 5.0846,
|
|
"mean_token_accuracy": 0.18146735727787017,
|
|
"num_tokens": 7531286.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.4291825771331785,
|
|
"epoch": 3.3620689655172415,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041846674522337296,
|
|
"loss": 5.0784,
|
|
"mean_token_accuracy": 0.17839567959308625,
|
|
"num_tokens": 7540450.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.32267165184021,
|
|
"epoch": 3.365987460815047,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00041823490508071076,
|
|
"loss": 5.0019,
|
|
"mean_token_accuracy": 0.18825284838676454,
|
|
"num_tokens": 7549386.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.443134832382202,
|
|
"epoch": 3.3699059561128526,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000418002808954049,
|
|
"loss": 5.0827,
|
|
"mean_token_accuracy": 0.1806928724050522,
|
|
"num_tokens": 7558236.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.341430521011352,
|
|
"epoch": 3.373824451410658,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041777045725818057,
|
|
"loss": 4.9678,
|
|
"mean_token_accuracy": 0.18331473767757417,
|
|
"num_tokens": 7566984.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.361105108261109,
|
|
"epoch": 3.377742946708464,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000417537850408356,
|
|
"loss": 5.11,
|
|
"mean_token_accuracy": 0.17382620126008988,
|
|
"num_tokens": 7576018.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.438305187225342,
|
|
"epoch": 3.3816614420062696,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004173049888202814,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.18349119424819946,
|
|
"num_tokens": 7585294.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.380229711532593,
|
|
"epoch": 3.385579937304075,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004170718729101179,
|
|
"loss": 5.0453,
|
|
"mean_token_accuracy": 0.18017769306898118,
|
|
"num_tokens": 7594634.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.378760528564453,
|
|
"epoch": 3.389498432601881,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00041683850309448187,
|
|
"loss": 5.1084,
|
|
"mean_token_accuracy": 0.1723770394921303,
|
|
"num_tokens": 7603153.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.4110680103302,
|
|
"epoch": 3.3934169278996866,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041660487979044264,
|
|
"loss": 5.0467,
|
|
"mean_token_accuracy": 0.183244688808918,
|
|
"num_tokens": 7612027.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.37716588973999,
|
|
"epoch": 3.397335423197492,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004163710034155231,
|
|
"loss": 5.0642,
|
|
"mean_token_accuracy": 0.1793832615017891,
|
|
"num_tokens": 7620725.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.31814751625061,
|
|
"epoch": 3.4012539184952977,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004161368743876982,
|
|
"loss": 5.0295,
|
|
"mean_token_accuracy": 0.18604897558689118,
|
|
"num_tokens": 7628847.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.423111152648926,
|
|
"epoch": 3.405172413793103,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004159024931253945,
|
|
"loss": 5.1185,
|
|
"mean_token_accuracy": 0.1817140057682991,
|
|
"num_tokens": 7637668.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.436103343963623,
|
|
"epoch": 3.409090909090909,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041566786004748943,
|
|
"loss": 5.1478,
|
|
"mean_token_accuracy": 0.1800548553466797,
|
|
"num_tokens": 7646545.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.45118818283081,
|
|
"epoch": 3.4130094043887147,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00041543297557331015,
|
|
"loss": 5.1471,
|
|
"mean_token_accuracy": 0.17297317385673522,
|
|
"num_tokens": 7654967.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.386139583587647,
|
|
"epoch": 3.41692789968652,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004151978401226335,
|
|
"loss": 5.0773,
|
|
"mean_token_accuracy": 0.1843633234500885,
|
|
"num_tokens": 7663718.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.434453105926513,
|
|
"epoch": 3.420846394984326,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00041496245411568435,
|
|
"loss": 5.1107,
|
|
"mean_token_accuracy": 0.1814684808254242,
|
|
"num_tokens": 7672430.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.432828378677368,
|
|
"epoch": 3.4247648902821317,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004147268179731359,
|
|
"loss": 5.0985,
|
|
"mean_token_accuracy": 0.1738584592938423,
|
|
"num_tokens": 7681270.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.393769693374634,
|
|
"epoch": 3.4286833855799372,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00041449093211610815,
|
|
"loss": 5.1002,
|
|
"mean_token_accuracy": 0.17736238688230516,
|
|
"num_tokens": 7690124.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.434223556518555,
|
|
"epoch": 3.4326018808777428,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041425479696616734,
|
|
"loss": 5.1264,
|
|
"mean_token_accuracy": 0.1761149376630783,
|
|
"num_tokens": 7699511.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.376140117645264,
|
|
"epoch": 3.4365203761755487,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 5.0706,
|
|
"mean_token_accuracy": 0.1834137573838234,
|
|
"num_tokens": 7708796.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.414622211456299,
|
|
"epoch": 3.4404388714733543,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00041378178047603845,
|
|
"loss": 5.0763,
|
|
"mean_token_accuracy": 0.185263991355896,
|
|
"num_tokens": 7717446.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.412569570541382,
|
|
"epoch": 3.44435736677116,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004135448999812074,
|
|
"loss": 5.1178,
|
|
"mean_token_accuracy": 0.17992717474699021,
|
|
"num_tokens": 7726814.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.370216321945191,
|
|
"epoch": 3.4482758620689653,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004133077718841763,
|
|
"loss": 5.0172,
|
|
"mean_token_accuracy": 0.18345807045698165,
|
|
"num_tokens": 7735895.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.377612638473511,
|
|
"epoch": 3.4521943573667713,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00041307039660873113,
|
|
"loss": 5.0777,
|
|
"mean_token_accuracy": 0.17952295243740082,
|
|
"num_tokens": 7744160.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.450047969818115,
|
|
"epoch": 3.456112852664577,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004128327745791002,
|
|
"loss": 5.1726,
|
|
"mean_token_accuracy": 0.17732140719890593,
|
|
"num_tokens": 7752844.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.405539798736572,
|
|
"epoch": 3.4600313479623823,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004125949062199526,
|
|
"loss": 5.1307,
|
|
"mean_token_accuracy": 0.1762731269001961,
|
|
"num_tokens": 7762327.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.385516357421875,
|
|
"epoch": 3.4639498432601883,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00041235679195639766,
|
|
"loss": 5.0673,
|
|
"mean_token_accuracy": 0.1847527414560318,
|
|
"num_tokens": 7770813.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.337443876266479,
|
|
"epoch": 3.467868338557994,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041211843221398406,
|
|
"loss": 5.0993,
|
|
"mean_token_accuracy": 0.17929710298776627,
|
|
"num_tokens": 7779357.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.433915281295777,
|
|
"epoch": 3.4717868338557993,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004118798274186994,
|
|
"loss": 5.1087,
|
|
"mean_token_accuracy": 0.17941119521856308,
|
|
"num_tokens": 7787302.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.3698526382446286,
|
|
"epoch": 3.475705329153605,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004116409779969691,
|
|
"loss": 5.1214,
|
|
"mean_token_accuracy": 0.1788015678524971,
|
|
"num_tokens": 7795651.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.374856567382812,
|
|
"epoch": 3.479623824451411,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00041140188437565586,
|
|
"loss": 5.0945,
|
|
"mean_token_accuracy": 0.18368430584669113,
|
|
"num_tokens": 7804644.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.455676507949829,
|
|
"epoch": 3.4835423197492164,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00041116254698205873,
|
|
"loss": 5.0653,
|
|
"mean_token_accuracy": 0.19262382239103318,
|
|
"num_tokens": 7812479.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.351228189468384,
|
|
"epoch": 3.487460815047022,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00041092296624391244,
|
|
"loss": 5.0713,
|
|
"mean_token_accuracy": 0.18133934289216996,
|
|
"num_tokens": 7820678.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.362814950942993,
|
|
"epoch": 3.4913793103448274,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004106831425893865,
|
|
"loss": 5.1157,
|
|
"mean_token_accuracy": 0.1836672082543373,
|
|
"num_tokens": 7829063.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.4492950439453125,
|
|
"epoch": 3.4952978056426334,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004104430764470849,
|
|
"loss": 5.1295,
|
|
"mean_token_accuracy": 0.18300817161798477,
|
|
"num_tokens": 7837274.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.430435609817505,
|
|
"epoch": 3.499216300940439,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004102027682460445,
|
|
"loss": 5.0388,
|
|
"mean_token_accuracy": 0.190749391913414,
|
|
"num_tokens": 7845379.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.383829975128174,
|
|
"epoch": 3.5031347962382444,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004099622184157353,
|
|
"loss": 5.0816,
|
|
"mean_token_accuracy": 0.1794472575187683,
|
|
"num_tokens": 7853760.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.4970954895019535,
|
|
"epoch": 3.5070532915360504,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004097214273860586,
|
|
"loss": 5.1367,
|
|
"mean_token_accuracy": 0.17909106910228728,
|
|
"num_tokens": 7862026.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.363224363327026,
|
|
"epoch": 3.510971786833856,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004094803955873471,
|
|
"loss": 5.1104,
|
|
"mean_token_accuracy": 0.18607434630393982,
|
|
"num_tokens": 7870369.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.394609355926514,
|
|
"epoch": 3.5148902821316614,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004092391234503638,
|
|
"loss": 5.0786,
|
|
"mean_token_accuracy": 0.18537619560956956,
|
|
"num_tokens": 7878524.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.473204469680786,
|
|
"epoch": 3.518808777429467,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00040899761140630094,
|
|
"loss": 5.1562,
|
|
"mean_token_accuracy": 0.16975836157798768,
|
|
"num_tokens": 7887369.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.464102363586425,
|
|
"epoch": 3.5227272727272725,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040875585988677985,
|
|
"loss": 5.1277,
|
|
"mean_token_accuracy": 0.17074015885591506,
|
|
"num_tokens": 7896112.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.339993190765381,
|
|
"epoch": 3.5266457680250785,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004085138693238497,
|
|
"loss": 4.9817,
|
|
"mean_token_accuracy": 0.18741547614336013,
|
|
"num_tokens": 7904580.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.5266457680250785,
|
|
"eval_entropy": 5.314358168794203,
|
|
"eval_loss": 5.79091739654541,
|
|
"eval_mean_token_accuracy": 0.15934031100784854,
|
|
"eval_num_tokens": 7904580.0,
|
|
"eval_runtime": 2.856,
|
|
"eval_samples_per_second": 1443.261,
|
|
"eval_steps_per_second": 180.67,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.443260574340821,
|
|
"epoch": 3.530564263322884,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004082716401499867,
|
|
"loss": 5.1291,
|
|
"mean_token_accuracy": 0.18045676350593567,
|
|
"num_tokens": 7913755.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.392041540145874,
|
|
"epoch": 3.5344827586206895,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00040802917279809383,
|
|
"loss": 5.102,
|
|
"mean_token_accuracy": 0.18171899616718293,
|
|
"num_tokens": 7922385.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.373143243789673,
|
|
"epoch": 3.5384012539184955,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040778646770149953,
|
|
"loss": 5.1244,
|
|
"mean_token_accuracy": 0.17722288221120835,
|
|
"num_tokens": 7931634.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.376545810699463,
|
|
"epoch": 3.542319749216301,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040754352529395716,
|
|
"loss": 5.046,
|
|
"mean_token_accuracy": 0.18162104338407517,
|
|
"num_tokens": 7939866.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.409450387954712,
|
|
"epoch": 3.5462382445141065,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040730034600964415,
|
|
"loss": 5.1634,
|
|
"mean_token_accuracy": 0.16843382120132447,
|
|
"num_tokens": 7948961.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.249195337295532,
|
|
"epoch": 3.5501567398119125,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004070569302831613,
|
|
"loss": 5.0098,
|
|
"mean_token_accuracy": 0.18459998071193695,
|
|
"num_tokens": 7958526.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.399295330047607,
|
|
"epoch": 3.554075235109718,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000406813278549532,
|
|
"loss": 5.1222,
|
|
"mean_token_accuracy": 0.18191463947296144,
|
|
"num_tokens": 7966887.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.434154796600342,
|
|
"epoch": 3.5579937304075235,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00040656939124420144,
|
|
"loss": 5.0857,
|
|
"mean_token_accuracy": 0.18093272149562836,
|
|
"num_tokens": 7974929.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.344184732437133,
|
|
"epoch": 3.561912225705329,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004063252688030358,
|
|
"loss": 5.0562,
|
|
"mean_token_accuracy": 0.18410776257514955,
|
|
"num_tokens": 7983821.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.347458362579346,
|
|
"epoch": 3.5658307210031346,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004060809116623213,
|
|
"loss": 5.0648,
|
|
"mean_token_accuracy": 0.1814362123608589,
|
|
"num_tokens": 7992109.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.339373064041138,
|
|
"epoch": 3.5697492163009406,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000405836320258764,
|
|
"loss": 5.0911,
|
|
"mean_token_accuracy": 0.18736453950405121,
|
|
"num_tokens": 8001098.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.457185077667236,
|
|
"epoch": 3.573667711598746,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004055914950294882,
|
|
"loss": 5.0862,
|
|
"mean_token_accuracy": 0.1816763922572136,
|
|
"num_tokens": 8009559.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.346025466918945,
|
|
"epoch": 3.5775862068965516,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00040534643641203645,
|
|
"loss": 5.0297,
|
|
"mean_token_accuracy": 0.18479364216327668,
|
|
"num_tokens": 8018021.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.3208576202392575,
|
|
"epoch": 3.5815047021943576,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004051011448443681,
|
|
"loss": 5.0591,
|
|
"mean_token_accuracy": 0.19013548046350479,
|
|
"num_tokens": 8026258.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.38658332824707,
|
|
"epoch": 3.585423197492163,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000404855620764859,
|
|
"loss": 5.125,
|
|
"mean_token_accuracy": 0.18151704519987105,
|
|
"num_tokens": 8036761.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.48537974357605,
|
|
"epoch": 3.5893416927899686,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004046098646123006,
|
|
"loss": 5.1505,
|
|
"mean_token_accuracy": 0.18237053155899047,
|
|
"num_tokens": 8045115.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.474244928359985,
|
|
"epoch": 3.593260188087774,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040436387682589876,
|
|
"loss": 5.1369,
|
|
"mean_token_accuracy": 0.17690041214227675,
|
|
"num_tokens": 8054334.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.340980768203735,
|
|
"epoch": 3.5971786833855797,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004041176578452737,
|
|
"loss": 5.11,
|
|
"mean_token_accuracy": 0.1812202140688896,
|
|
"num_tokens": 8063165.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.378905439376831,
|
|
"epoch": 3.6010971786833856,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004038712081104587,
|
|
"loss": 5.0903,
|
|
"mean_token_accuracy": 0.1769264370203018,
|
|
"num_tokens": 8072498.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.372360944747925,
|
|
"epoch": 3.605015673981191,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00040362452806189927,
|
|
"loss": 5.1094,
|
|
"mean_token_accuracy": 0.1775502309203148,
|
|
"num_tokens": 8081455.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.344131708145142,
|
|
"epoch": 3.6089341692789967,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004033776181404527,
|
|
"loss": 5.1276,
|
|
"mean_token_accuracy": 0.17704904228448867,
|
|
"num_tokens": 8090992.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.4004453182220455,
|
|
"epoch": 3.6128526645768027,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040313047878738704,
|
|
"loss": 5.1036,
|
|
"mean_token_accuracy": 0.17663901448249816,
|
|
"num_tokens": 8099976.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.410310220718384,
|
|
"epoch": 3.616771159874608,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004028831104443805,
|
|
"loss": 5.0115,
|
|
"mean_token_accuracy": 0.19510315209627152,
|
|
"num_tokens": 8108275.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.348009967803955,
|
|
"epoch": 3.6206896551724137,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 5.1012,
|
|
"mean_token_accuracy": 0.18073388785123826,
|
|
"num_tokens": 8117441.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.359053277969361,
|
|
"epoch": 3.6246081504702197,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00040238768855730214,
|
|
"loss": 5.0662,
|
|
"mean_token_accuracy": 0.1826333299279213,
|
|
"num_tokens": 8125709.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.367337989807129,
|
|
"epoch": 3.628526645768025,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00040213963589862963,
|
|
"loss": 5.0882,
|
|
"mean_token_accuracy": 0.17644327729940415,
|
|
"num_tokens": 8134418.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.4221583843231205,
|
|
"epoch": 3.6324451410658307,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004018913560208131,
|
|
"loss": 5.1482,
|
|
"mean_token_accuracy": 0.1751092329621315,
|
|
"num_tokens": 8143729.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.367877531051636,
|
|
"epoch": 3.6363636363636362,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004016428493675689,
|
|
"loss": 5.0277,
|
|
"mean_token_accuracy": 0.18457772582769394,
|
|
"num_tokens": 8152613.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.321438980102539,
|
|
"epoch": 3.6402821316614418,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004013941163830187,
|
|
"loss": 5.041,
|
|
"mean_token_accuracy": 0.18404249250888824,
|
|
"num_tokens": 8160862.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.415549468994141,
|
|
"epoch": 3.6442006269592477,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004011451575116887,
|
|
"loss": 5.1091,
|
|
"mean_token_accuracy": 0.1787499725818634,
|
|
"num_tokens": 8169369.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.375511407852173,
|
|
"epoch": 3.6481191222570533,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004008959731985087,
|
|
"loss": 5.0317,
|
|
"mean_token_accuracy": 0.18156421333551406,
|
|
"num_tokens": 8178152.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.3223206996917725,
|
|
"epoch": 3.652037617554859,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00040064656388881157,
|
|
"loss": 4.9985,
|
|
"mean_token_accuracy": 0.1848632201552391,
|
|
"num_tokens": 8186690.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.301701831817627,
|
|
"epoch": 3.6559561128526648,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004003969300283321,
|
|
"loss": 4.98,
|
|
"mean_token_accuracy": 0.19033852070569993,
|
|
"num_tokens": 8194976.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.316612434387207,
|
|
"epoch": 3.6598746081504703,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00040014707206320653,
|
|
"loss": 5.0762,
|
|
"mean_token_accuracy": 0.18378447145223617,
|
|
"num_tokens": 8202964.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.35722246170044,
|
|
"epoch": 3.663793103448276,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039989699043997153,
|
|
"loss": 5.0938,
|
|
"mean_token_accuracy": 0.18213022798299788,
|
|
"num_tokens": 8213084.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.36103401184082,
|
|
"epoch": 3.6677115987460818,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039964668560556356,
|
|
"loss": 4.9954,
|
|
"mean_token_accuracy": 0.1894006162881851,
|
|
"num_tokens": 8221645.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.3249406814575195,
|
|
"epoch": 3.6716300940438873,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00039939615800731784,
|
|
"loss": 4.9984,
|
|
"mean_token_accuracy": 0.19127800315618515,
|
|
"num_tokens": 8230195.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.385104703903198,
|
|
"epoch": 3.675548589341693,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039914540809296795,
|
|
"loss": 5.0775,
|
|
"mean_token_accuracy": 0.18279951959848403,
|
|
"num_tokens": 8239608.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.41096715927124,
|
|
"epoch": 3.6794670846394983,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003988944363106445,
|
|
"loss": 5.1272,
|
|
"mean_token_accuracy": 0.18332620859146118,
|
|
"num_tokens": 8247993.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.335915231704712,
|
|
"epoch": 3.683385579937304,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003986432431088749,
|
|
"loss": 5.099,
|
|
"mean_token_accuracy": 0.1738138824701309,
|
|
"num_tokens": 8257279.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.45527491569519,
|
|
"epoch": 3.68730407523511,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000398391828936582,
|
|
"loss": 5.1472,
|
|
"mean_token_accuracy": 0.17864662855863572,
|
|
"num_tokens": 8265166.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.290485668182373,
|
|
"epoch": 3.6912225705329154,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003981401942430838,
|
|
"loss": 4.9278,
|
|
"mean_token_accuracy": 0.19530342817306517,
|
|
"num_tokens": 8273281.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.326362800598145,
|
|
"epoch": 3.695141065830721,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00039788833947809217,
|
|
"loss": 5.1453,
|
|
"mean_token_accuracy": 0.1722833603620529,
|
|
"num_tokens": 8282432.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.301543045043945,
|
|
"epoch": 3.699059561128527,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003976362650917125,
|
|
"loss": 4.9493,
|
|
"mean_token_accuracy": 0.19138956516981126,
|
|
"num_tokens": 8290909.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.324641847610474,
|
|
"epoch": 3.7029780564263324,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00039738397153444264,
|
|
"loss": 5.0796,
|
|
"mean_token_accuracy": 0.1851937383413315,
|
|
"num_tokens": 8299581.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.376465511322022,
|
|
"epoch": 3.706896551724138,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003971314592571719,
|
|
"loss": 5.0671,
|
|
"mean_token_accuracy": 0.18401720523834228,
|
|
"num_tokens": 8308184.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.355161046981811,
|
|
"epoch": 3.7108150470219434,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003968787287111809,
|
|
"loss": 5.076,
|
|
"mean_token_accuracy": 0.1832886129617691,
|
|
"num_tokens": 8317054.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.3086090087890625,
|
|
"epoch": 3.714733542319749,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00039662578034814,
|
|
"loss": 5.0884,
|
|
"mean_token_accuracy": 0.18718038648366928,
|
|
"num_tokens": 8325961.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.490419816970825,
|
|
"epoch": 3.718652037617555,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039637261462010886,
|
|
"loss": 5.183,
|
|
"mean_token_accuracy": 0.17251200079917908,
|
|
"num_tokens": 8334401.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.4552045345306395,
|
|
"epoch": 3.7225705329153604,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003961192319795358,
|
|
"loss": 5.1665,
|
|
"mean_token_accuracy": 0.1736621856689453,
|
|
"num_tokens": 8343786.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.413653755187989,
|
|
"epoch": 3.726489028213166,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003958656328792565,
|
|
"loss": 5.1396,
|
|
"mean_token_accuracy": 0.18157389611005784,
|
|
"num_tokens": 8352286.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.314235210418701,
|
|
"epoch": 3.730407523510972,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00039561181777249396,
|
|
"loss": 5.0603,
|
|
"mean_token_accuracy": 0.18286719918251038,
|
|
"num_tokens": 8361274.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.368144369125366,
|
|
"epoch": 3.7343260188087775,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00039535778711285676,
|
|
"loss": 5.0975,
|
|
"mean_token_accuracy": 0.18291845619678498,
|
|
"num_tokens": 8369881.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.4078105926513675,
|
|
"epoch": 3.738244514106583,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003951035413543388,
|
|
"loss": 5.1693,
|
|
"mean_token_accuracy": 0.17080339640378953,
|
|
"num_tokens": 8378234.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.370164012908935,
|
|
"epoch": 3.742163009404389,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039484908095131874,
|
|
"loss": 5.0771,
|
|
"mean_token_accuracy": 0.18567045778036118,
|
|
"num_tokens": 8387023.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.321150493621826,
|
|
"epoch": 3.7460815047021945,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003945944063585582,
|
|
"loss": 5.0396,
|
|
"mean_token_accuracy": 0.1883723333477974,
|
|
"num_tokens": 8396110.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.34753828048706,
|
|
"epoch": 3.75,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039433951803120225,
|
|
"loss": 5.0094,
|
|
"mean_token_accuracy": 0.19095030725002288,
|
|
"num_tokens": 8405219.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.445649290084839,
|
|
"epoch": 3.7539184952978055,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00039408441642477764,
|
|
"loss": 5.2044,
|
|
"mean_token_accuracy": 0.1734558403491974,
|
|
"num_tokens": 8413996.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.3355896949768065,
|
|
"epoch": 3.757836990595611,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003938291019951922,
|
|
"loss": 4.9486,
|
|
"mean_token_accuracy": 0.19205181300640106,
|
|
"num_tokens": 8421931.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.350619602203369,
|
|
"epoch": 3.761755485893417,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003935735751987344,
|
|
"loss": 5.0635,
|
|
"mean_token_accuracy": 0.17809719145298003,
|
|
"num_tokens": 8430090.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.268694448471069,
|
|
"epoch": 3.7656739811912225,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039331783649207175,
|
|
"loss": 5.0339,
|
|
"mean_token_accuracy": 0.18174760043621063,
|
|
"num_tokens": 8439311.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.407308912277221,
|
|
"epoch": 3.769592476489028,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039306188633225097,
|
|
"loss": 5.1,
|
|
"mean_token_accuracy": 0.1792532056570053,
|
|
"num_tokens": 8448952.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.32148814201355,
|
|
"epoch": 3.773510971786834,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003928057251766965,
|
|
"loss": 5.0404,
|
|
"mean_token_accuracy": 0.18206110149621962,
|
|
"num_tokens": 8457177.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.320786428451538,
|
|
"epoch": 3.7774294670846396,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00039254935348320984,
|
|
"loss": 5.0858,
|
|
"mean_token_accuracy": 0.18461327850818635,
|
|
"num_tokens": 8465844.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.478414678573609,
|
|
"epoch": 3.781347962382445,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00039229277170996885,
|
|
"loss": 5.192,
|
|
"mean_token_accuracy": 0.17193579524755478,
|
|
"num_tokens": 8475198.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.4253462791442875,
|
|
"epoch": 3.785266457680251,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003920359803155266,
|
|
"loss": 5.0559,
|
|
"mean_token_accuracy": 0.18259523659944535,
|
|
"num_tokens": 8484050.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.2889374732971195,
|
|
"epoch": 3.7891849529780566,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00039177897975881115,
|
|
"loss": 5.0425,
|
|
"mean_token_accuracy": 0.1863778352737427,
|
|
"num_tokens": 8491825.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.339927005767822,
|
|
"epoch": 3.793103448275862,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003915217704991239,
|
|
"loss": 5.0543,
|
|
"mean_token_accuracy": 0.1847598910331726,
|
|
"num_tokens": 8500651.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.420097494125367,
|
|
"epoch": 3.7970219435736676,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003912643529961397,
|
|
"loss": 5.0682,
|
|
"mean_token_accuracy": 0.18167142421007157,
|
|
"num_tokens": 8508464.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.354662084579468,
|
|
"epoch": 3.800940438871473,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003910067277099053,
|
|
"loss": 5.0916,
|
|
"mean_token_accuracy": 0.19040142893791198,
|
|
"num_tokens": 8517676.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.441104936599731,
|
|
"epoch": 3.804858934169279,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 5.1433,
|
|
"mean_token_accuracy": 0.17872390449047088,
|
|
"num_tokens": 8526926.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.370850419998169,
|
|
"epoch": 3.8087774294670846,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003904908556297293,
|
|
"loss": 5.0359,
|
|
"mean_token_accuracy": 0.18077736645936965,
|
|
"num_tokens": 8535814.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.401374053955078,
|
|
"epoch": 3.81269592476489,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003902326097577345,
|
|
"loss": 5.1391,
|
|
"mean_token_accuracy": 0.18166149258613587,
|
|
"num_tokens": 8544814.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.399645709991455,
|
|
"epoch": 3.816614420062696,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038997415794638206,
|
|
"loss": 5.129,
|
|
"mean_token_accuracy": 0.1790841594338417,
|
|
"num_tokens": 8553455.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.317953109741211,
|
|
"epoch": 3.8205329153605017,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003897155006575672,
|
|
"loss": 5.0444,
|
|
"mean_token_accuracy": 0.1792813926935196,
|
|
"num_tokens": 8562048.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.3612377643585205,
|
|
"epoch": 3.824451410658307,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038945663835355247,
|
|
"loss": 5.0921,
|
|
"mean_token_accuracy": 0.18234582394361495,
|
|
"num_tokens": 8571855.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.346320152282715,
|
|
"epoch": 3.8283699059561127,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00038919757149696665,
|
|
"loss": 5.0867,
|
|
"mean_token_accuracy": 0.17777179181575775,
|
|
"num_tokens": 8580425.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.3517656326293945,
|
|
"epoch": 3.8322884012539182,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038893830055080437,
|
|
"loss": 5.1118,
|
|
"mean_token_accuracy": 0.1857918068766594,
|
|
"num_tokens": 8589751.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.377425050735473,
|
|
"epoch": 3.836206896551724,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003886788259784248,
|
|
"loss": 5.1114,
|
|
"mean_token_accuracy": 0.1744757518172264,
|
|
"num_tokens": 8598077.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.443184471130371,
|
|
"epoch": 3.8401253918495297,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00038841914824355093,
|
|
"loss": 5.1278,
|
|
"mean_token_accuracy": 0.18244832903146743,
|
|
"num_tokens": 8606691.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.33152494430542,
|
|
"epoch": 3.8440438871473352,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038815926781026914,
|
|
"loss": 5.0787,
|
|
"mean_token_accuracy": 0.17986321300268174,
|
|
"num_tokens": 8615477.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.33702335357666,
|
|
"epoch": 3.847962382445141,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003878991851430279,
|
|
"loss": 5.0837,
|
|
"mean_token_accuracy": 0.1862553298473358,
|
|
"num_tokens": 8623982.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.365121841430664,
|
|
"epoch": 3.8518808777429467,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003876389007066371,
|
|
"loss": 5.1242,
|
|
"mean_token_accuracy": 0.18310100734233856,
|
|
"num_tokens": 8633017.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.360111474990845,
|
|
"epoch": 3.8557993730407523,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003873784149662672,
|
|
"loss": 5.077,
|
|
"mean_token_accuracy": 0.1806945398449898,
|
|
"num_tokens": 8642299.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.353548812866211,
|
|
"epoch": 3.8597178683385582,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003871177283874484,
|
|
"loss": 4.9915,
|
|
"mean_token_accuracy": 0.1794583886861801,
|
|
"num_tokens": 8651008.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.3401891708374025,
|
|
"epoch": 3.8636363636363638,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038685684143606995,
|
|
"loss": 5.0899,
|
|
"mean_token_accuracy": 0.1844473212957382,
|
|
"num_tokens": 8660103.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.2554792881011965,
|
|
"epoch": 3.8675548589341693,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003865957545783791,
|
|
"loss": 5.0268,
|
|
"mean_token_accuracy": 0.18795547038316726,
|
|
"num_tokens": 8668646.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.384418392181397,
|
|
"epoch": 3.871473354231975,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00038633446828098046,
|
|
"loss": 5.1922,
|
|
"mean_token_accuracy": 0.1762129470705986,
|
|
"num_tokens": 8678190.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.511695098876953,
|
|
"epoch": 3.8753918495297803,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000386072983010835,
|
|
"loss": 5.2008,
|
|
"mean_token_accuracy": 0.17127986401319503,
|
|
"num_tokens": 8687336.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.44140739440918,
|
|
"epoch": 3.8793103448275863,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00038581129923525914,
|
|
"loss": 5.0966,
|
|
"mean_token_accuracy": 0.17991429269313813,
|
|
"num_tokens": 8695939.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.376219797134399,
|
|
"epoch": 3.883228840125392,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00038554941742192445,
|
|
"loss": 5.1235,
|
|
"mean_token_accuracy": 0.17844197005033494,
|
|
"num_tokens": 8704828.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.375438070297241,
|
|
"epoch": 3.8871473354231973,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003852873380388561,
|
|
"loss": 5.0483,
|
|
"mean_token_accuracy": 0.19140560925006866,
|
|
"num_tokens": 8714427.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.420164775848389,
|
|
"epoch": 3.8910658307210033,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003850250615544323,
|
|
"loss": 5.1504,
|
|
"mean_token_accuracy": 0.18523310422897338,
|
|
"num_tokens": 8723737.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.2913489818573,
|
|
"epoch": 3.894984326018809,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00038476258843738386,
|
|
"loss": 5.0829,
|
|
"mean_token_accuracy": 0.18507641702890396,
|
|
"num_tokens": 8733149.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.292956113815308,
|
|
"epoch": 3.8989028213166144,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038449991915679273,
|
|
"loss": 5.0462,
|
|
"mean_token_accuracy": 0.19215177744627,
|
|
"num_tokens": 8742000.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.374302625656128,
|
|
"epoch": 3.9028213166144203,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003842370541820915,
|
|
"loss": 5.1125,
|
|
"mean_token_accuracy": 0.18079639077186585,
|
|
"num_tokens": 8751069.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.37040376663208,
|
|
"epoch": 3.906739811912226,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038397399398306243,
|
|
"loss": 5.0388,
|
|
"mean_token_accuracy": 0.18093398511409758,
|
|
"num_tokens": 8760092.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.337316226959229,
|
|
"epoch": 3.9106583072100314,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00038371073902983684,
|
|
"loss": 5.0446,
|
|
"mean_token_accuracy": 0.18595774471759796,
|
|
"num_tokens": 8768039.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.3745136737823485,
|
|
"epoch": 3.914576802507837,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000383447289792894,
|
|
"loss": 5.0998,
|
|
"mean_token_accuracy": 0.1770160123705864,
|
|
"num_tokens": 8776630.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.422601556777954,
|
|
"epoch": 3.9184952978056424,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038318364674306036,
|
|
"loss": 5.0505,
|
|
"mean_token_accuracy": 0.1845804810523987,
|
|
"num_tokens": 8785272.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 3.9184952978056424,
|
|
"eval_entropy": 5.288773534833923,
|
|
"eval_loss": 5.7375288009643555,
|
|
"eval_mean_token_accuracy": 0.1628826753229134,
|
|
"eval_num_tokens": 8785272.0,
|
|
"eval_runtime": 3.0474,
|
|
"eval_samples_per_second": 1352.622,
|
|
"eval_steps_per_second": 169.324,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.384165525436401,
|
|
"epoch": 3.9224137931034484,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038291981035150883,
|
|
"loss": 5.0643,
|
|
"mean_token_accuracy": 0.18399680107831956,
|
|
"num_tokens": 8794197.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.347819757461548,
|
|
"epoch": 3.926332288401254,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003826557810897579,
|
|
"loss": 5.1248,
|
|
"mean_token_accuracy": 0.17706111073493958,
|
|
"num_tokens": 8803195.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.352082395553589,
|
|
"epoch": 3.9302507836990594,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003823915594296705,
|
|
"loss": 5.0611,
|
|
"mean_token_accuracy": 0.1865521937608719,
|
|
"num_tokens": 8811772.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.217619132995606,
|
|
"epoch": 3.9341692789968654,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003821271458434538,
|
|
"loss": 5.0405,
|
|
"mean_token_accuracy": 0.18679626286029816,
|
|
"num_tokens": 8820482.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.34322943687439,
|
|
"epoch": 3.938087774294671,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00038186254080365756,
|
|
"loss": 5.1259,
|
|
"mean_token_accuracy": 0.18093776553869248,
|
|
"num_tokens": 8828761.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.333748912811279,
|
|
"epoch": 3.9420062695924765,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000381597744783174,
|
|
"loss": 4.9886,
|
|
"mean_token_accuracy": 0.18413858711719513,
|
|
"num_tokens": 8837165.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.348248863220215,
|
|
"epoch": 3.945924764890282,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038133275825523645,
|
|
"loss": 5.0788,
|
|
"mean_token_accuracy": 0.18073509335517884,
|
|
"num_tokens": 8845472.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.339105653762817,
|
|
"epoch": 3.9498432601880875,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038106758169341887,
|
|
"loss": 5.0713,
|
|
"mean_token_accuracy": 0.18627960234880447,
|
|
"num_tokens": 8855246.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.452056121826172,
|
|
"epoch": 3.9537617554858935,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003808022155716348,
|
|
"loss": 5.1479,
|
|
"mean_token_accuracy": 0.17604552060365677,
|
|
"num_tokens": 8863667.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.357966709136963,
|
|
"epoch": 3.957680250783699,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003805366603641364,
|
|
"loss": 4.9789,
|
|
"mean_token_accuracy": 0.19253498315811157,
|
|
"num_tokens": 8872306.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.251607275009155,
|
|
"epoch": 3.9615987460815045,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038027091654551406,
|
|
"loss": 5.0385,
|
|
"mean_token_accuracy": 0.18627630025148392,
|
|
"num_tokens": 8882037.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.29943060874939,
|
|
"epoch": 3.9655172413793105,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00038000498459069487,
|
|
"loss": 5.0611,
|
|
"mean_token_accuracy": 0.18815120458602905,
|
|
"num_tokens": 8890685.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.35674524307251,
|
|
"epoch": 3.969435736677116,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003797388649749426,
|
|
"loss": 5.0515,
|
|
"mean_token_accuracy": 0.1858385905623436,
|
|
"num_tokens": 8899927.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.349644517898559,
|
|
"epoch": 3.9733542319749215,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003794725581738559,
|
|
"loss": 5.095,
|
|
"mean_token_accuracy": 0.18175891786813736,
|
|
"num_tokens": 8908571.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.304893636703492,
|
|
"epoch": 3.9772727272727275,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037920606466336834,
|
|
"loss": 5.003,
|
|
"mean_token_accuracy": 0.18737710863351822,
|
|
"num_tokens": 8917246.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.349433660507202,
|
|
"epoch": 3.981191222570533,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000378939384919747,
|
|
"loss": 5.1006,
|
|
"mean_token_accuracy": 0.18233946561813355,
|
|
"num_tokens": 8926116.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.414449262619018,
|
|
"epoch": 3.9851097178683386,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003786725194195918,
|
|
"loss": 5.1315,
|
|
"mean_token_accuracy": 0.1743440270423889,
|
|
"num_tokens": 8934914.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.364781761169434,
|
|
"epoch": 3.989028213166144,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 5.0294,
|
|
"mean_token_accuracy": 0.18088567554950713,
|
|
"num_tokens": 8943321.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.31170506477356,
|
|
"epoch": 3.9929467084639496,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00037813823305773883,
|
|
"loss": 5.0881,
|
|
"mean_token_accuracy": 0.18345759063959122,
|
|
"num_tokens": 8951848.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.3401905536651615,
|
|
"epoch": 3.9968652037617556,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003778708131508974,
|
|
"loss": 5.075,
|
|
"mean_token_accuracy": 0.1792066738009453,
|
|
"num_tokens": 8961291.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.4452203750610355,
|
|
"epoch": 4.0007836990595615,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037760320939723307,
|
|
"loss": 5.1063,
|
|
"mean_token_accuracy": 0.18336477428674697,
|
|
"num_tokens": 8971116.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.329280948638916,
|
|
"epoch": 4.004702194357367,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037733542227499727,
|
|
"loss": 4.7187,
|
|
"mean_token_accuracy": 0.20930335372686387,
|
|
"num_tokens": 8979362.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.4096739292144775,
|
|
"epoch": 4.008620689655173,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00037706745226276893,
|
|
"loss": 4.9132,
|
|
"mean_token_accuracy": 0.19055889099836348,
|
|
"num_tokens": 8987728.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.378854036331177,
|
|
"epoch": 4.012539184952978,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003767992998394539,
|
|
"loss": 4.7461,
|
|
"mean_token_accuracy": 0.20209902673959732,
|
|
"num_tokens": 8995817.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.314801120758057,
|
|
"epoch": 4.016457680250784,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003765309654842842,
|
|
"loss": 4.8341,
|
|
"mean_token_accuracy": 0.19608426541090013,
|
|
"num_tokens": 9005683.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.249972867965698,
|
|
"epoch": 4.02037617554859,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003762624496768166,
|
|
"loss": 4.7318,
|
|
"mean_token_accuracy": 0.20734037905931474,
|
|
"num_tokens": 9015408.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.441287899017334,
|
|
"epoch": 4.024294670846395,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003759937528969325,
|
|
"loss": 4.8712,
|
|
"mean_token_accuracy": 0.19440686106681823,
|
|
"num_tokens": 9024381.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.325917053222656,
|
|
"epoch": 4.028213166144201,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037572487562483666,
|
|
"loss": 4.7649,
|
|
"mean_token_accuracy": 0.20230175852775573,
|
|
"num_tokens": 9033092.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.288177108764648,
|
|
"epoch": 4.032131661442007,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037545581834105623,
|
|
"loss": 4.8842,
|
|
"mean_token_accuracy": 0.19988031387329103,
|
|
"num_tokens": 9042251.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.247293424606323,
|
|
"epoch": 4.036050156739812,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00037518658152644035,
|
|
"loss": 4.7707,
|
|
"mean_token_accuracy": 0.2000918999314308,
|
|
"num_tokens": 9050628.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.313902378082275,
|
|
"epoch": 4.039968652037618,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003749171656621588,
|
|
"loss": 4.7916,
|
|
"mean_token_accuracy": 0.2125375136733055,
|
|
"num_tokens": 9060080.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.323680734634399,
|
|
"epoch": 4.043887147335423,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003746475712297014,
|
|
"loss": 4.7976,
|
|
"mean_token_accuracy": 0.20550839006900787,
|
|
"num_tokens": 9069456.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.302190971374512,
|
|
"epoch": 4.047805642633229,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003743777987108771,
|
|
"loss": 4.8174,
|
|
"mean_token_accuracy": 0.20415401756763457,
|
|
"num_tokens": 9078790.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.268424129486084,
|
|
"epoch": 4.051724137931035,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003741078485878132,
|
|
"loss": 4.8038,
|
|
"mean_token_accuracy": 0.20454448759555816,
|
|
"num_tokens": 9086888.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.333013820648193,
|
|
"epoch": 4.05564263322884,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003738377213429542,
|
|
"loss": 4.8454,
|
|
"mean_token_accuracy": 0.1926748350262642,
|
|
"num_tokens": 9095139.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.313342332839966,
|
|
"epoch": 4.059561128526646,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003735674174590614,
|
|
"loss": 4.7535,
|
|
"mean_token_accuracy": 0.20612601041793824,
|
|
"num_tokens": 9103464.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.331502103805542,
|
|
"epoch": 4.063479623824452,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00037329693741921166,
|
|
"loss": 4.7755,
|
|
"mean_token_accuracy": 0.1971474438905716,
|
|
"num_tokens": 9111290.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.194469881057739,
|
|
"epoch": 4.067398119122257,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003730262817067967,
|
|
"loss": 4.6513,
|
|
"mean_token_accuracy": 0.20742494910955428,
|
|
"num_tokens": 9120224.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.281076240539551,
|
|
"epoch": 4.071316614420063,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003727554508055221,
|
|
"loss": 4.8043,
|
|
"mean_token_accuracy": 0.2017815366387367,
|
|
"num_tokens": 9128524.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.228507328033447,
|
|
"epoch": 4.075235109717869,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003724844451994066,
|
|
"loss": 4.6775,
|
|
"mean_token_accuracy": 0.20459066331386566,
|
|
"num_tokens": 9137026.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.284112358093262,
|
|
"epoch": 4.079153605015674,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037221326537278113,
|
|
"loss": 4.7735,
|
|
"mean_token_accuracy": 0.19970867335796355,
|
|
"num_tokens": 9146312.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.269183921813965,
|
|
"epoch": 4.08307210031348,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037194191181028806,
|
|
"loss": 4.794,
|
|
"mean_token_accuracy": 0.19847405552864075,
|
|
"num_tokens": 9155226.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.381943702697754,
|
|
"epoch": 4.086990595611285,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00037167038499688024,
|
|
"loss": 4.9393,
|
|
"mean_token_accuracy": 0.19385314881801605,
|
|
"num_tokens": 9164090.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.3243756771087645,
|
|
"epoch": 4.090909090909091,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037139868541782,
|
|
"loss": 4.777,
|
|
"mean_token_accuracy": 0.2073064014315605,
|
|
"num_tokens": 9173002.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.33658618927002,
|
|
"epoch": 4.094827586206897,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003711268135586787,
|
|
"loss": 4.7933,
|
|
"mean_token_accuracy": 0.19626413881778718,
|
|
"num_tokens": 9182020.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.270061302185058,
|
|
"epoch": 4.098746081504702,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037085476990533523,
|
|
"loss": 4.7997,
|
|
"mean_token_accuracy": 0.1949327051639557,
|
|
"num_tokens": 9190848.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.227729082107544,
|
|
"epoch": 4.102664576802508,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003705825549439759,
|
|
"loss": 4.7397,
|
|
"mean_token_accuracy": 0.2057284951210022,
|
|
"num_tokens": 9198808.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.372107934951782,
|
|
"epoch": 4.106583072100314,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00037031016916109295,
|
|
"loss": 4.8755,
|
|
"mean_token_accuracy": 0.19922751039266587,
|
|
"num_tokens": 9207923.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.319027233123779,
|
|
"epoch": 4.110501567398119,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037003761304348374,
|
|
"loss": 4.8246,
|
|
"mean_token_accuracy": 0.19773749262094498,
|
|
"num_tokens": 9216759.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.223453426361084,
|
|
"epoch": 4.114420062695925,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00036976488707825056,
|
|
"loss": 4.7706,
|
|
"mean_token_accuracy": 0.1890528604388237,
|
|
"num_tokens": 9225951.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.23214635848999,
|
|
"epoch": 4.118338557993731,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036949199175279876,
|
|
"loss": 4.8383,
|
|
"mean_token_accuracy": 0.1947033777832985,
|
|
"num_tokens": 9234413.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.376208305358887,
|
|
"epoch": 4.122257053291536,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003692189275548364,
|
|
"loss": 4.9646,
|
|
"mean_token_accuracy": 0.18719761818647385,
|
|
"num_tokens": 9244002.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.350819444656372,
|
|
"epoch": 4.126175548589342,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003689456949723737,
|
|
"loss": 4.79,
|
|
"mean_token_accuracy": 0.20028213560581207,
|
|
"num_tokens": 9252689.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.342959260940551,
|
|
"epoch": 4.130094043887147,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003686722944937215,
|
|
"loss": 4.7721,
|
|
"mean_token_accuracy": 0.20335770547389984,
|
|
"num_tokens": 9261118.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.262115812301635,
|
|
"epoch": 4.134012539184953,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036839872660749084,
|
|
"loss": 4.7793,
|
|
"mean_token_accuracy": 0.19953766018152236,
|
|
"num_tokens": 9269205.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.246290063858032,
|
|
"epoch": 4.137931034482759,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003681249918025918,
|
|
"loss": 4.7892,
|
|
"mean_token_accuracy": 0.20386335849761963,
|
|
"num_tokens": 9277770.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.240208864212036,
|
|
"epoch": 4.141849529780564,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00036785109056823297,
|
|
"loss": 4.8373,
|
|
"mean_token_accuracy": 0.19764283001422883,
|
|
"num_tokens": 9286589.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.311579799652099,
|
|
"epoch": 4.14576802507837,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00036757702339392,
|
|
"loss": 4.799,
|
|
"mean_token_accuracy": 0.19371658563613892,
|
|
"num_tokens": 9295264.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.264638423919678,
|
|
"epoch": 4.149686520376176,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036730279076945574,
|
|
"loss": 4.8107,
|
|
"mean_token_accuracy": 0.20059947520494462,
|
|
"num_tokens": 9303525.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.260356140136719,
|
|
"epoch": 4.153605015673981,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003670283931849382,
|
|
"loss": 4.8644,
|
|
"mean_token_accuracy": 0.18903878480195999,
|
|
"num_tokens": 9312839.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.286557722091675,
|
|
"epoch": 4.157523510971787,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00036675383113076023,
|
|
"loss": 4.819,
|
|
"mean_token_accuracy": 0.1983775407075882,
|
|
"num_tokens": 9321462.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.247763395309448,
|
|
"epoch": 4.161442006269592,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00036647910509760896,
|
|
"loss": 4.841,
|
|
"mean_token_accuracy": 0.19876728355884551,
|
|
"num_tokens": 9329315.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.283821105957031,
|
|
"epoch": 4.165360501567398,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00036620421557646414,
|
|
"loss": 4.786,
|
|
"mean_token_accuracy": 0.20168010145425797,
|
|
"num_tokens": 9336997.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.251159143447876,
|
|
"epoch": 4.169278996865204,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000365929163058598,
|
|
"loss": 4.775,
|
|
"mean_token_accuracy": 0.2019730567932129,
|
|
"num_tokens": 9345364.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.279034566879273,
|
|
"epoch": 4.173197492163009,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 4.7725,
|
|
"mean_token_accuracy": 0.20545876026153564,
|
|
"num_tokens": 9353961.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.396962261199951,
|
|
"epoch": 4.177115987460815,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003653785709992462,
|
|
"loss": 4.9335,
|
|
"mean_token_accuracy": 0.188700170814991,
|
|
"num_tokens": 9362901.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.292047214508057,
|
|
"epoch": 4.181034482758621,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036510303244175775,
|
|
"loss": 4.8338,
|
|
"mean_token_accuracy": 0.1959599420428276,
|
|
"num_tokens": 9371800.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.27794885635376,
|
|
"epoch": 4.184952978056426,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00036482733285554073,
|
|
"loss": 4.8274,
|
|
"mean_token_accuracy": 0.20195936858654023,
|
|
"num_tokens": 9380033.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.400831937789917,
|
|
"epoch": 4.188871473354232,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00036455147273331507,
|
|
"loss": 4.8951,
|
|
"mean_token_accuracy": 0.19887401163578033,
|
|
"num_tokens": 9389264.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.277568387985229,
|
|
"epoch": 4.192789968652038,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003642754525680876,
|
|
"loss": 4.8092,
|
|
"mean_token_accuracy": 0.2042933940887451,
|
|
"num_tokens": 9398112.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.314821195602417,
|
|
"epoch": 4.196708463949843,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036399927285315104,
|
|
"loss": 4.9188,
|
|
"mean_token_accuracy": 0.18789517730474473,
|
|
"num_tokens": 9407400.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.306748771667481,
|
|
"epoch": 4.200626959247649,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003637229340820834,
|
|
"loss": 4.811,
|
|
"mean_token_accuracy": 0.19472367465496063,
|
|
"num_tokens": 9416351.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.3867582321167,
|
|
"epoch": 4.204545454545454,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003634464367487467,
|
|
"loss": 4.8415,
|
|
"mean_token_accuracy": 0.1919476330280304,
|
|
"num_tokens": 9424577.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.293810176849365,
|
|
"epoch": 4.20846394984326,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003631697813472867,
|
|
"loss": 4.8366,
|
|
"mean_token_accuracy": 0.2024887129664421,
|
|
"num_tokens": 9434242.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.179585790634155,
|
|
"epoch": 4.212382445141066,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036289296837213136,
|
|
"loss": 4.7925,
|
|
"mean_token_accuracy": 0.2028203636407852,
|
|
"num_tokens": 9443247.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.230278253555298,
|
|
"epoch": 4.216300940438871,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036261599831799034,
|
|
"loss": 4.7693,
|
|
"mean_token_accuracy": 0.2032170847058296,
|
|
"num_tokens": 9451998.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.283214044570923,
|
|
"epoch": 4.220219435736677,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003623388716798541,
|
|
"loss": 4.7247,
|
|
"mean_token_accuracy": 0.20366513431072236,
|
|
"num_tokens": 9459993.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.233270788192749,
|
|
"epoch": 4.224137931034483,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00036206158895299277,
|
|
"loss": 4.7732,
|
|
"mean_token_accuracy": 0.19897193908691407,
|
|
"num_tokens": 9468036.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.23637900352478,
|
|
"epoch": 4.228056426332288,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036178415063295574,
|
|
"loss": 4.844,
|
|
"mean_token_accuracy": 0.20170646607875825,
|
|
"num_tokens": 9477031.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.291466951370239,
|
|
"epoch": 4.231974921630094,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00036150655721557,
|
|
"loss": 4.7687,
|
|
"mean_token_accuracy": 0.204982291162014,
|
|
"num_tokens": 9485687.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.2552018642425535,
|
|
"epoch": 4.2358934169279,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003612288091969402,
|
|
"loss": 4.81,
|
|
"mean_token_accuracy": 0.20634790658950805,
|
|
"num_tokens": 9494613.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.285842657089233,
|
|
"epoch": 4.239811912225705,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00036095090707344697,
|
|
"loss": 4.8157,
|
|
"mean_token_accuracy": 0.1996786192059517,
|
|
"num_tokens": 9503522.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.248120927810669,
|
|
"epoch": 4.243730407523511,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036067285134174654,
|
|
"loss": 4.8342,
|
|
"mean_token_accuracy": 0.19701357185840607,
|
|
"num_tokens": 9512293.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.3384205341339115,
|
|
"epoch": 4.247648902821316,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003603946424987696,
|
|
"loss": 4.9405,
|
|
"mean_token_accuracy": 0.1904897838830948,
|
|
"num_tokens": 9520999.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.309277725219727,
|
|
"epoch": 4.251567398119122,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00036011628104172045,
|
|
"loss": 4.8726,
|
|
"mean_token_accuracy": 0.19428194910287858,
|
|
"num_tokens": 9529798.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.367800235748291,
|
|
"epoch": 4.255485893416928,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035983776746807613,
|
|
"loss": 4.8081,
|
|
"mean_token_accuracy": 0.19572492688894272,
|
|
"num_tokens": 9538668.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.325085163116455,
|
|
"epoch": 4.259404388714733,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003595591022755855,
|
|
"loss": 4.8431,
|
|
"mean_token_accuracy": 0.20056636333465577,
|
|
"num_tokens": 9546973.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.20649676322937,
|
|
"epoch": 4.263322884012539,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003592802859622684,
|
|
"loss": 4.7847,
|
|
"mean_token_accuracy": 0.20024689733982087,
|
|
"num_tokens": 9555947.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.27953724861145,
|
|
"epoch": 4.267241379310345,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000359001319026415,
|
|
"loss": 4.8723,
|
|
"mean_token_accuracy": 0.196561498939991,
|
|
"num_tokens": 9564730.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.226915597915649,
|
|
"epoch": 4.27115987460815,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035872220196658426,
|
|
"loss": 4.8182,
|
|
"mean_token_accuracy": 0.20109005719423295,
|
|
"num_tokens": 9573470.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.323717164993286,
|
|
"epoch": 4.275078369905956,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00035844293528160384,
|
|
"loss": 4.8581,
|
|
"mean_token_accuracy": 0.19655824154615403,
|
|
"num_tokens": 9581607.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.288580369949341,
|
|
"epoch": 4.278996865203762,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003581635194705682,
|
|
"loss": 4.8018,
|
|
"mean_token_accuracy": 0.20255643427371978,
|
|
"num_tokens": 9589801.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.289205074310303,
|
|
"epoch": 4.282915360501567,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035788395503283914,
|
|
"loss": 4.8855,
|
|
"mean_token_accuracy": 0.18393632769584656,
|
|
"num_tokens": 9599983.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.302262639999389,
|
|
"epoch": 4.286833855799373,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035760424246804334,
|
|
"loss": 4.7716,
|
|
"mean_token_accuracy": 0.20446890145540236,
|
|
"num_tokens": 9608681.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.274475574493408,
|
|
"epoch": 4.290752351097178,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003573243822760726,
|
|
"loss": 4.7736,
|
|
"mean_token_accuracy": 0.20583966970443726,
|
|
"num_tokens": 9616461.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.302184915542602,
|
|
"epoch": 4.294670846394984,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003570443749570825,
|
|
"loss": 4.8791,
|
|
"mean_token_accuracy": 0.1958416685461998,
|
|
"num_tokens": 9625480.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.287403964996338,
|
|
"epoch": 4.29858934169279,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003567642210114914,
|
|
"loss": 4.8564,
|
|
"mean_token_accuracy": 0.1961355224251747,
|
|
"num_tokens": 9634172.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.252014875411987,
|
|
"epoch": 4.302507836990595,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035648392093998,
|
|
"loss": 4.8272,
|
|
"mean_token_accuracy": 0.19759142100811006,
|
|
"num_tokens": 9643196.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.216412973403931,
|
|
"epoch": 4.306426332288401,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003562034752434899,
|
|
"loss": 4.748,
|
|
"mean_token_accuracy": 0.2056722715497017,
|
|
"num_tokens": 9652440.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.263983535766601,
|
|
"epoch": 4.310344827586207,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003559228844232229,
|
|
"loss": 4.8242,
|
|
"mean_token_accuracy": 0.19313945174217223,
|
|
"num_tokens": 9661248.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.310344827586207,
|
|
"eval_entropy": 5.201999014662218,
|
|
"eval_loss": 5.744182109832764,
|
|
"eval_mean_token_accuracy": 0.16393170540415963,
|
|
"eval_num_tokens": 9661248.0,
|
|
"eval_runtime": 2.8444,
|
|
"eval_samples_per_second": 1449.173,
|
|
"eval_steps_per_second": 181.41,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.320192575454712,
|
|
"epoch": 4.314263322884012,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003556421489806406,
|
|
"loss": 4.8571,
|
|
"mean_token_accuracy": 0.20088933855295182,
|
|
"num_tokens": 9670544.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.286400890350341,
|
|
"epoch": 4.318181818181818,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003553612694174624,
|
|
"loss": 4.8365,
|
|
"mean_token_accuracy": 0.20016209185123443,
|
|
"num_tokens": 9679331.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.382794952392578,
|
|
"epoch": 4.322100313479623,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003550802462356659,
|
|
"loss": 4.8945,
|
|
"mean_token_accuracy": 0.19355344474315644,
|
|
"num_tokens": 9688211.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.157554197311401,
|
|
"epoch": 4.326018808777429,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00035479907993748484,
|
|
"loss": 4.7544,
|
|
"mean_token_accuracy": 0.20544749796390532,
|
|
"num_tokens": 9696231.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.254225826263427,
|
|
"epoch": 4.329937304075235,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035451777102540915,
|
|
"loss": 4.8346,
|
|
"mean_token_accuracy": 0.19444499909877777,
|
|
"num_tokens": 9705577.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.424444675445557,
|
|
"epoch": 4.33385579937304,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003542363200021835,
|
|
"loss": 4.8785,
|
|
"mean_token_accuracy": 0.19774623960256577,
|
|
"num_tokens": 9715003.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.220383977890014,
|
|
"epoch": 4.337774294670846,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003539547273708064,
|
|
"loss": 4.722,
|
|
"mean_token_accuracy": 0.20750671178102492,
|
|
"num_tokens": 9723513.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.2017497539520265,
|
|
"epoch": 4.341692789968652,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00035367299363452966,
|
|
"loss": 4.8433,
|
|
"mean_token_accuracy": 0.19815308451652527,
|
|
"num_tokens": 9731800.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.257679605484009,
|
|
"epoch": 4.3456112852664575,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035339111929685694,
|
|
"loss": 4.7882,
|
|
"mean_token_accuracy": 0.20065640211105346,
|
|
"num_tokens": 9740356.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.304519367218018,
|
|
"epoch": 4.349529780564263,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00035310910486154365,
|
|
"loss": 4.8422,
|
|
"mean_token_accuracy": 0.1902172312140465,
|
|
"num_tokens": 9749529.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.286208820343018,
|
|
"epoch": 4.353448275862069,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003528269508325951,
|
|
"loss": 4.8074,
|
|
"mean_token_accuracy": 0.19902238994836807,
|
|
"num_tokens": 9758750.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.2709071159362795,
|
|
"epoch": 4.3573667711598745,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 4.8985,
|
|
"mean_token_accuracy": 0.2050875261425972,
|
|
"num_tokens": 9766992.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.330464839935303,
|
|
"epoch": 4.36128526645768,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00035226222601106105,
|
|
"loss": 4.9114,
|
|
"mean_token_accuracy": 0.1947867602109909,
|
|
"num_tokens": 9777022.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.222631978988647,
|
|
"epoch": 4.3652037617554855,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003519796562277303,
|
|
"loss": 4.6674,
|
|
"mean_token_accuracy": 0.2095522850751877,
|
|
"num_tokens": 9785863.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.237377643585205,
|
|
"epoch": 4.3691222570532915,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003516969488692723,
|
|
"loss": 4.8297,
|
|
"mean_token_accuracy": 0.19563706815242768,
|
|
"num_tokens": 9794357.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.21888165473938,
|
|
"epoch": 4.3730407523510975,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003514141044409308,
|
|
"loss": 4.8006,
|
|
"mean_token_accuracy": 0.19991396367549896,
|
|
"num_tokens": 9803128.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.298209953308105,
|
|
"epoch": 4.3769592476489025,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003511311234481947,
|
|
"loss": 4.9026,
|
|
"mean_token_accuracy": 0.19144590944051743,
|
|
"num_tokens": 9812046.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.274967288970947,
|
|
"epoch": 4.3808777429467085,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00035084800639679695,
|
|
"loss": 4.7756,
|
|
"mean_token_accuracy": 0.20507780462503433,
|
|
"num_tokens": 9820935.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.296483039855957,
|
|
"epoch": 4.3847962382445145,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035056475379271356,
|
|
"loss": 4.8986,
|
|
"mean_token_accuracy": 0.19356585144996644,
|
|
"num_tokens": 9829585.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.3491191387176515,
|
|
"epoch": 4.38871473354232,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003502813661421629,
|
|
"loss": 4.9567,
|
|
"mean_token_accuracy": 0.19294051826000214,
|
|
"num_tokens": 9839302.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.297818851470947,
|
|
"epoch": 4.3926332288401255,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003499978439516047,
|
|
"loss": 4.85,
|
|
"mean_token_accuracy": 0.20032887309789657,
|
|
"num_tokens": 9848330.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.224711990356445,
|
|
"epoch": 4.396551724137931,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034971418772773895,
|
|
"loss": 4.8116,
|
|
"mean_token_accuracy": 0.20705717355012893,
|
|
"num_tokens": 9856928.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.31253342628479,
|
|
"epoch": 4.400470219435737,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003494303979775054,
|
|
"loss": 4.8676,
|
|
"mean_token_accuracy": 0.19687178134918212,
|
|
"num_tokens": 9865464.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.285409545898437,
|
|
"epoch": 4.4043887147335425,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034914647520808235,
|
|
"loss": 4.8323,
|
|
"mean_token_accuracy": 0.1991928666830063,
|
|
"num_tokens": 9874345.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.22224006652832,
|
|
"epoch": 4.408307210031348,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003488624199268859,
|
|
"loss": 4.7992,
|
|
"mean_token_accuracy": 0.20444443821907043,
|
|
"num_tokens": 9883135.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.313176155090332,
|
|
"epoch": 4.412225705329154,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003485782326415687,
|
|
"loss": 4.8902,
|
|
"mean_token_accuracy": 0.19405024349689484,
|
|
"num_tokens": 9892229.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.325419616699219,
|
|
"epoch": 4.41614420062696,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003482939138600197,
|
|
"loss": 4.8739,
|
|
"mean_token_accuracy": 0.19669069200754166,
|
|
"num_tokens": 9901044.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.282191324234009,
|
|
"epoch": 4.420062695924765,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003480094640903626,
|
|
"loss": 4.8926,
|
|
"mean_token_accuracy": 0.19395661503076553,
|
|
"num_tokens": 9909818.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.243837642669678,
|
|
"epoch": 4.423981191222571,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00034772488384095517,
|
|
"loss": 4.7609,
|
|
"mean_token_accuracy": 0.2052598312497139,
|
|
"num_tokens": 9919295.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.243894004821778,
|
|
"epoch": 4.427899686520377,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00034744017362038854,
|
|
"loss": 4.8547,
|
|
"mean_token_accuracy": 0.19718240350484847,
|
|
"num_tokens": 9927539.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.275289058685303,
|
|
"epoch": 4.431818181818182,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00034715533393748604,
|
|
"loss": 4.9028,
|
|
"mean_token_accuracy": 0.19591504633426665,
|
|
"num_tokens": 9936540.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.248186826705933,
|
|
"epoch": 4.435736677115988,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00034687036530130244,
|
|
"loss": 4.7609,
|
|
"mean_token_accuracy": 0.21108741462230682,
|
|
"num_tokens": 9945065.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.258121728897095,
|
|
"epoch": 4.439655172413793,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003465852682211227,
|
|
"loss": 4.7776,
|
|
"mean_token_accuracy": 0.19737946689128877,
|
|
"num_tokens": 9954155.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.249235391616821,
|
|
"epoch": 4.443573667711599,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003463000432064617,
|
|
"loss": 4.8562,
|
|
"mean_token_accuracy": 0.19955746382474898,
|
|
"num_tokens": 9963269.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.321204757690429,
|
|
"epoch": 4.447492163009405,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00034601469076706276,
|
|
"loss": 4.9038,
|
|
"mean_token_accuracy": 0.20010546147823333,
|
|
"num_tokens": 9972688.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.3547741889953615,
|
|
"epoch": 4.45141065830721,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003457292114128968,
|
|
"loss": 4.8479,
|
|
"mean_token_accuracy": 0.194638255238533,
|
|
"num_tokens": 9981283.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.214753341674805,
|
|
"epoch": 4.455329153605016,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000345443605654162,
|
|
"loss": 4.8692,
|
|
"mean_token_accuracy": 0.19639173448085784,
|
|
"num_tokens": 9989870.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.284179925918579,
|
|
"epoch": 4.459247648902822,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00034515787400128195,
|
|
"loss": 4.8667,
|
|
"mean_token_accuracy": 0.1953931286931038,
|
|
"num_tokens": 9998136.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.290607023239136,
|
|
"epoch": 4.463166144200627,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003448720169649054,
|
|
"loss": 4.8171,
|
|
"mean_token_accuracy": 0.19769317060709,
|
|
"num_tokens": 10006730.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.265090751647949,
|
|
"epoch": 4.467084639498433,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003445860350559055,
|
|
"loss": 4.8268,
|
|
"mean_token_accuracy": 0.20683852583169937,
|
|
"num_tokens": 10015517.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.218235778808594,
|
|
"epoch": 4.471003134796238,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000344299928785378,
|
|
"loss": 4.8521,
|
|
"mean_token_accuracy": 0.20003726333379745,
|
|
"num_tokens": 10024552.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.244536399841309,
|
|
"epoch": 4.474921630094044,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00034401369866464133,
|
|
"loss": 4.8351,
|
|
"mean_token_accuracy": 0.19964713901281356,
|
|
"num_tokens": 10033523.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.244927072525025,
|
|
"epoch": 4.47884012539185,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034372734520523505,
|
|
"loss": 4.7255,
|
|
"mean_token_accuracy": 0.20214163810014724,
|
|
"num_tokens": 10041988.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.2023186683654785,
|
|
"epoch": 4.482758620689655,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003434408689189193,
|
|
"loss": 4.8115,
|
|
"mean_token_accuracy": 0.20148774087429047,
|
|
"num_tokens": 10050543.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.278959083557129,
|
|
"epoch": 4.486677115987461,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003431542703176737,
|
|
"loss": 4.8436,
|
|
"mean_token_accuracy": 0.19792723804712295,
|
|
"num_tokens": 10059300.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.235785913467407,
|
|
"epoch": 4.490595611285267,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003428675499136963,
|
|
"loss": 4.8622,
|
|
"mean_token_accuracy": 0.20373777896165848,
|
|
"num_tokens": 10068535.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.329932403564453,
|
|
"epoch": 4.494514106583072,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000342580708219403,
|
|
"loss": 4.926,
|
|
"mean_token_accuracy": 0.19719588309526442,
|
|
"num_tokens": 10077731.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.3053487777709964,
|
|
"epoch": 4.498432601880878,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003422937457474264,
|
|
"loss": 4.8255,
|
|
"mean_token_accuracy": 0.192769393324852,
|
|
"num_tokens": 10087146.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.262833070755005,
|
|
"epoch": 4.502351097178684,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034200666301061495,
|
|
"loss": 4.813,
|
|
"mean_token_accuracy": 0.20547911822795867,
|
|
"num_tokens": 10096659.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.285718297958374,
|
|
"epoch": 4.506269592476489,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00034171946052203217,
|
|
"loss": 4.9096,
|
|
"mean_token_accuracy": 0.1946066826581955,
|
|
"num_tokens": 10105403.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.381711101531982,
|
|
"epoch": 4.510188087774295,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00034143213879495526,
|
|
"loss": 4.8599,
|
|
"mean_token_accuracy": 0.195270374417305,
|
|
"num_tokens": 10115410.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.247734069824219,
|
|
"epoch": 4.514106583072101,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000341144698342875,
|
|
"loss": 4.8506,
|
|
"mean_token_accuracy": 0.19717508554458618,
|
|
"num_tokens": 10123800.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.277109813690186,
|
|
"epoch": 4.518025078369906,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003408571396794939,
|
|
"loss": 4.9055,
|
|
"mean_token_accuracy": 0.19607753455638885,
|
|
"num_tokens": 10132775.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.282625675201416,
|
|
"epoch": 4.521943573667712,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003405694633187259,
|
|
"loss": 4.7915,
|
|
"mean_token_accuracy": 0.20083234012126921,
|
|
"num_tokens": 10141811.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.1926469802856445,
|
|
"epoch": 4.525862068965517,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00034028166977469545,
|
|
"loss": 4.7941,
|
|
"mean_token_accuracy": 0.20052414238452912,
|
|
"num_tokens": 10150823.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.282280826568604,
|
|
"epoch": 4.529780564263323,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033999375956173607,
|
|
"loss": 4.8167,
|
|
"mean_token_accuracy": 0.19880249798297883,
|
|
"num_tokens": 10158983.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.327436447143555,
|
|
"epoch": 4.533699059561129,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003397057331943902,
|
|
"loss": 4.8159,
|
|
"mean_token_accuracy": 0.20252034962177276,
|
|
"num_tokens": 10167596.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.249426031112671,
|
|
"epoch": 4.537617554858934,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003394175911874076,
|
|
"loss": 4.8867,
|
|
"mean_token_accuracy": 0.20285695642232895,
|
|
"num_tokens": 10176216.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.2126476764678955,
|
|
"epoch": 4.54153605015674,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 4.7947,
|
|
"mean_token_accuracy": 0.20142941027879716,
|
|
"num_tokens": 10185685.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.260278749465942,
|
|
"epoch": 4.545454545454545,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003388409623145637,
|
|
"loss": 4.796,
|
|
"mean_token_accuracy": 0.20086440742015838,
|
|
"num_tokens": 10193679.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.2498561382293705,
|
|
"epoch": 4.549373040752351,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00033855247647923177,
|
|
"loss": 4.8488,
|
|
"mean_token_accuracy": 0.20274218320846557,
|
|
"num_tokens": 10201154.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.239627981185913,
|
|
"epoch": 4.553291536050157,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003382638770653198,
|
|
"loss": 4.7615,
|
|
"mean_token_accuracy": 0.20632180869579314,
|
|
"num_tokens": 10209180.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.2880054950714115,
|
|
"epoch": 4.557210031347962,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003379751645886017,
|
|
"loss": 4.8686,
|
|
"mean_token_accuracy": 0.19341773837804793,
|
|
"num_tokens": 10217430.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.2620607852935795,
|
|
"epoch": 4.561128526645768,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003376863395650537,
|
|
"loss": 4.915,
|
|
"mean_token_accuracy": 0.20187284350395202,
|
|
"num_tokens": 10226453.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.3547447681427,
|
|
"epoch": 4.565047021943574,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00033739740251085263,
|
|
"loss": 4.9666,
|
|
"mean_token_accuracy": 0.19466182291507722,
|
|
"num_tokens": 10235005.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.320493268966675,
|
|
"epoch": 4.568965517241379,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00033710835394237603,
|
|
"loss": 4.8748,
|
|
"mean_token_accuracy": 0.1945085495710373,
|
|
"num_tokens": 10243800.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.245484638214111,
|
|
"epoch": 4.572884012539185,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00033681919437620066,
|
|
"loss": 4.8401,
|
|
"mean_token_accuracy": 0.19890396296977997,
|
|
"num_tokens": 10252370.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.310286045074463,
|
|
"epoch": 4.576802507836991,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003365299243291014,
|
|
"loss": 4.9013,
|
|
"mean_token_accuracy": 0.19252691566944122,
|
|
"num_tokens": 10260777.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.276307582855225,
|
|
"epoch": 4.580721003134796,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033624054431805067,
|
|
"loss": 4.9112,
|
|
"mean_token_accuracy": 0.18616441041231155,
|
|
"num_tokens": 10270571.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.315344619750976,
|
|
"epoch": 4.584639498432602,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00033595105486021763,
|
|
"loss": 4.8757,
|
|
"mean_token_accuracy": 0.1976679965853691,
|
|
"num_tokens": 10279284.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.239494800567627,
|
|
"epoch": 4.588557993730408,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033566145647296675,
|
|
"loss": 4.7773,
|
|
"mean_token_accuracy": 0.19849830567836763,
|
|
"num_tokens": 10288610.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.244922685623169,
|
|
"epoch": 4.592476489028213,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00033537174967385744,
|
|
"loss": 4.7643,
|
|
"mean_token_accuracy": 0.20348892211914063,
|
|
"num_tokens": 10296603.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.205677318572998,
|
|
"epoch": 4.596394984326019,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033508193498064254,
|
|
"loss": 4.7737,
|
|
"mean_token_accuracy": 0.20668493062257767,
|
|
"num_tokens": 10305199.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.173105192184448,
|
|
"epoch": 4.600313479623824,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033479201291126807,
|
|
"loss": 4.7184,
|
|
"mean_token_accuracy": 0.20839987993240355,
|
|
"num_tokens": 10313967.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.270901107788086,
|
|
"epoch": 4.60423197492163,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003345019839838717,
|
|
"loss": 4.8379,
|
|
"mean_token_accuracy": 0.20386506915092467,
|
|
"num_tokens": 10322532.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.240105772018433,
|
|
"epoch": 4.608150470219436,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003342118487167822,
|
|
"loss": 4.7186,
|
|
"mean_token_accuracy": 0.2093840181827545,
|
|
"num_tokens": 10330631.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.156818819046021,
|
|
"epoch": 4.612068965517241,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033392160762851813,
|
|
"loss": 4.7249,
|
|
"mean_token_accuracy": 0.20375409573316575,
|
|
"num_tokens": 10339229.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.267489004135132,
|
|
"epoch": 4.615987460815047,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003336312612377877,
|
|
"loss": 4.8576,
|
|
"mean_token_accuracy": 0.1860215499997139,
|
|
"num_tokens": 10348882.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.366767263412475,
|
|
"epoch": 4.619905956112853,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00033334081006348683,
|
|
"loss": 4.9732,
|
|
"mean_token_accuracy": 0.18846775442361832,
|
|
"num_tokens": 10358154.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.32326226234436,
|
|
"epoch": 4.623824451410658,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003330502546246988,
|
|
"loss": 4.8905,
|
|
"mean_token_accuracy": 0.19722211956977845,
|
|
"num_tokens": 10367100.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.163663291931153,
|
|
"epoch": 4.627742946708464,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033275959544069335,
|
|
"loss": 4.69,
|
|
"mean_token_accuracy": 0.21335452795028687,
|
|
"num_tokens": 10375711.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.243529605865478,
|
|
"epoch": 4.631661442006269,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033246883303092564,
|
|
"loss": 4.7943,
|
|
"mean_token_accuracy": 0.20526561439037322,
|
|
"num_tokens": 10384513.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.3067769527435305,
|
|
"epoch": 4.635579937304075,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00033217796791503503,
|
|
"loss": 4.8753,
|
|
"mean_token_accuracy": 0.1975191280245781,
|
|
"num_tokens": 10393564.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.315864896774292,
|
|
"epoch": 4.639498432601881,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003318870006128449,
|
|
"loss": 4.8846,
|
|
"mean_token_accuracy": 0.19700464308261872,
|
|
"num_tokens": 10402042.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.224531650543213,
|
|
"epoch": 4.643416927899686,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003315959316443608,
|
|
"loss": 4.7667,
|
|
"mean_token_accuracy": 0.1970340132713318,
|
|
"num_tokens": 10410847.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.2861260890960695,
|
|
"epoch": 4.647335423197492,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003313047615297703,
|
|
"loss": 4.8823,
|
|
"mean_token_accuracy": 0.19239578694105147,
|
|
"num_tokens": 10420366.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.259707880020142,
|
|
"epoch": 4.651253918495298,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00033101349078944165,
|
|
"loss": 4.7974,
|
|
"mean_token_accuracy": 0.2046971380710602,
|
|
"num_tokens": 10428870.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.221111631393432,
|
|
"epoch": 4.655172413793103,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003307221199439227,
|
|
"loss": 4.8389,
|
|
"mean_token_accuracy": 0.19255790561437608,
|
|
"num_tokens": 10437714.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.195745611190796,
|
|
"epoch": 4.659090909090909,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033043064951394045,
|
|
"loss": 4.7592,
|
|
"mean_token_accuracy": 0.20461451709270478,
|
|
"num_tokens": 10447198.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.263057851791382,
|
|
"epoch": 4.663009404388715,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00033013908002039994,
|
|
"loss": 4.8034,
|
|
"mean_token_accuracy": 0.20022739470005035,
|
|
"num_tokens": 10455790.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.289310693740845,
|
|
"epoch": 4.66692789968652,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032984741198438305,
|
|
"loss": 4.9036,
|
|
"mean_token_accuracy": 0.2009410321712494,
|
|
"num_tokens": 10464094.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.232710838317871,
|
|
"epoch": 4.670846394984326,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003295556459271478,
|
|
"loss": 4.8727,
|
|
"mean_token_accuracy": 0.19428551942110062,
|
|
"num_tokens": 10473201.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.286605310440064,
|
|
"epoch": 4.674764890282132,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003292637823701276,
|
|
"loss": 4.8982,
|
|
"mean_token_accuracy": 0.19300175309181214,
|
|
"num_tokens": 10482617.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.2989133358001705,
|
|
"epoch": 4.678683385579937,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003289718218349298,
|
|
"loss": 4.7605,
|
|
"mean_token_accuracy": 0.20499369502067566,
|
|
"num_tokens": 10491702.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.350911617279053,
|
|
"epoch": 4.682601880877743,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003286797648433351,
|
|
"loss": 4.9487,
|
|
"mean_token_accuracy": 0.1910773277282715,
|
|
"num_tokens": 10501189.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.26896071434021,
|
|
"epoch": 4.686520376175548,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000328387611917297,
|
|
"loss": 4.8243,
|
|
"mean_token_accuracy": 0.195672906935215,
|
|
"num_tokens": 10510272.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.314244747161865,
|
|
"epoch": 4.690438871473354,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003280953635789401,
|
|
"loss": 4.9084,
|
|
"mean_token_accuracy": 0.19853879809379577,
|
|
"num_tokens": 10518782.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.227982807159424,
|
|
"epoch": 4.69435736677116,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003278030203505594,
|
|
"loss": 4.8398,
|
|
"mean_token_accuracy": 0.19611885398626328,
|
|
"num_tokens": 10528262.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.194601678848267,
|
|
"epoch": 4.698275862068965,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003275105827546197,
|
|
"loss": 4.8136,
|
|
"mean_token_accuracy": 0.19929687082767486,
|
|
"num_tokens": 10536399.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.29431939125061,
|
|
"epoch": 4.702194357366771,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032721805131375446,
|
|
"loss": 4.8729,
|
|
"mean_token_accuracy": 0.19718658179044724,
|
|
"num_tokens": 10544842.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 4.702194357366771,
|
|
"eval_entropy": 5.222095316694689,
|
|
"eval_loss": 5.711119174957275,
|
|
"eval_mean_token_accuracy": 0.16745245306528816,
|
|
"eval_num_tokens": 10544842.0,
|
|
"eval_runtime": 2.8473,
|
|
"eval_samples_per_second": 1447.664,
|
|
"eval_steps_per_second": 181.221,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.280520009994507,
|
|
"epoch": 4.706112852664576,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003269254265507647,
|
|
"loss": 4.801,
|
|
"mean_token_accuracy": 0.20380959361791612,
|
|
"num_tokens": 10553630.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.215929222106934,
|
|
"epoch": 4.710031347962382,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00032663270898861845,
|
|
"loss": 4.8253,
|
|
"mean_token_accuracy": 0.1986591786146164,
|
|
"num_tokens": 10561714.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.216139459609986,
|
|
"epoch": 4.713949843260188,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00032633989915044944,
|
|
"loss": 4.891,
|
|
"mean_token_accuracy": 0.18992457538843155,
|
|
"num_tokens": 10571075.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.283609914779663,
|
|
"epoch": 4.717868338557993,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00032604699755955615,
|
|
"loss": 4.7841,
|
|
"mean_token_accuracy": 0.20595482289791106,
|
|
"num_tokens": 10579698.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.22645411491394,
|
|
"epoch": 4.721786833855799,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00032575400473940135,
|
|
"loss": 4.8134,
|
|
"mean_token_accuracy": 0.20072342008352279,
|
|
"num_tokens": 10588060.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.189909410476685,
|
|
"epoch": 4.725705329153605,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 4.839,
|
|
"mean_token_accuracy": 0.20117655098438264,
|
|
"num_tokens": 10596482.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.226874828338623,
|
|
"epoch": 4.72962382445141,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003251677475059721,
|
|
"loss": 4.8777,
|
|
"mean_token_accuracy": 0.19612164795398712,
|
|
"num_tokens": 10605630.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.305343961715698,
|
|
"epoch": 4.733542319749216,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032487448414043435,
|
|
"loss": 4.8629,
|
|
"mean_token_accuracy": 0.2001108482480049,
|
|
"num_tokens": 10614772.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.269188356399536,
|
|
"epoch": 4.737460815047022,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032458113164110665,
|
|
"loss": 4.8269,
|
|
"mean_token_accuracy": 0.2039674624800682,
|
|
"num_tokens": 10623583.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.266746664047242,
|
|
"epoch": 4.741379310344827,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003242876905322575,
|
|
"loss": 4.8176,
|
|
"mean_token_accuracy": 0.20840690433979034,
|
|
"num_tokens": 10632136.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.24975996017456,
|
|
"epoch": 4.745297805642633,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003239941613383138,
|
|
"loss": 4.9171,
|
|
"mean_token_accuracy": 0.19654276371002197,
|
|
"num_tokens": 10641306.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.270605182647705,
|
|
"epoch": 4.749216300940439,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032370054458385973,
|
|
"loss": 4.8549,
|
|
"mean_token_accuracy": 0.20089448988437653,
|
|
"num_tokens": 10650866.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.309944105148316,
|
|
"epoch": 4.753134796238244,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000323406840793636,
|
|
"loss": 4.8736,
|
|
"mean_token_accuracy": 0.20075544118881225,
|
|
"num_tokens": 10659252.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.207181596755982,
|
|
"epoch": 4.75705329153605,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000323113050492539,
|
|
"loss": 4.7543,
|
|
"mean_token_accuracy": 0.19956870526075363,
|
|
"num_tokens": 10667987.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.267623233795166,
|
|
"epoch": 4.7609717868338555,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00032281917420561967,
|
|
"loss": 4.938,
|
|
"mean_token_accuracy": 0.19012248814105986,
|
|
"num_tokens": 10676524.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.27276577949524,
|
|
"epoch": 4.764890282131661,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003225252124580825,
|
|
"loss": 4.8549,
|
|
"mean_token_accuracy": 0.20154722929000854,
|
|
"num_tokens": 10685940.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.249235391616821,
|
|
"epoch": 4.768808777429467,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00032223116577528475,
|
|
"loss": 4.8587,
|
|
"mean_token_accuracy": 0.1966584861278534,
|
|
"num_tokens": 10694522.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.246975040435791,
|
|
"epoch": 4.7727272727272725,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003219370346827356,
|
|
"loss": 4.7687,
|
|
"mean_token_accuracy": 0.20040610134601594,
|
|
"num_tokens": 10703868.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.235792589187622,
|
|
"epoch": 4.7766457680250785,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000321642819706095,
|
|
"loss": 4.809,
|
|
"mean_token_accuracy": 0.2003234624862671,
|
|
"num_tokens": 10712044.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.3406147956848145,
|
|
"epoch": 4.7805642633228835,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003213485213711729,
|
|
"loss": 4.9836,
|
|
"mean_token_accuracy": 0.19017882347106935,
|
|
"num_tokens": 10720806.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.205180692672729,
|
|
"epoch": 4.7844827586206895,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000321054140203928,
|
|
"loss": 4.7072,
|
|
"mean_token_accuracy": 0.21185464709997176,
|
|
"num_tokens": 10729238.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.208652400970459,
|
|
"epoch": 4.7884012539184955,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003207596767304672,
|
|
"loss": 4.7991,
|
|
"mean_token_accuracy": 0.20003243833780288,
|
|
"num_tokens": 10738175.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.282381582260132,
|
|
"epoch": 4.7923197492163006,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032046513147704456,
|
|
"loss": 4.8698,
|
|
"mean_token_accuracy": 0.19474064111709594,
|
|
"num_tokens": 10747475.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.267691230773925,
|
|
"epoch": 4.7962382445141065,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00032017050497006,
|
|
"loss": 4.8146,
|
|
"mean_token_accuracy": 0.2019771322607994,
|
|
"num_tokens": 10756739.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.2428955078125,
|
|
"epoch": 4.8001567398119125,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00031987579773605913,
|
|
"loss": 4.8245,
|
|
"mean_token_accuracy": 0.20127459317445756,
|
|
"num_tokens": 10765141.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.280072164535523,
|
|
"epoch": 4.804075235109718,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003195810103017314,
|
|
"loss": 4.9192,
|
|
"mean_token_accuracy": 0.19173355400562286,
|
|
"num_tokens": 10774759.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.294880819320679,
|
|
"epoch": 4.8079937304075235,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003192861431939098,
|
|
"loss": 4.8799,
|
|
"mean_token_accuracy": 0.1963331177830696,
|
|
"num_tokens": 10783717.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.2743908882141115,
|
|
"epoch": 4.8119122257053295,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003189911969395697,
|
|
"loss": 4.8758,
|
|
"mean_token_accuracy": 0.19668658822774887,
|
|
"num_tokens": 10792797.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.178160667419434,
|
|
"epoch": 4.815830721003135,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00031869617206582777,
|
|
"loss": 4.747,
|
|
"mean_token_accuracy": 0.20796505957841874,
|
|
"num_tokens": 10802249.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.2319474697113035,
|
|
"epoch": 4.8197492163009406,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003184010690999413,
|
|
"loss": 4.8411,
|
|
"mean_token_accuracy": 0.19480388313531877,
|
|
"num_tokens": 10810722.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.345753765106201,
|
|
"epoch": 4.8236677115987465,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003181058885693073,
|
|
"loss": 4.9097,
|
|
"mean_token_accuracy": 0.19312914609909057,
|
|
"num_tokens": 10819115.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.307505941390991,
|
|
"epoch": 4.827586206896552,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003178106310014612,
|
|
"loss": 4.8974,
|
|
"mean_token_accuracy": 0.19501842558383942,
|
|
"num_tokens": 10827419.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.169783973693848,
|
|
"epoch": 4.831504702194358,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003175152969240759,
|
|
"loss": 4.7761,
|
|
"mean_token_accuracy": 0.20091764479875565,
|
|
"num_tokens": 10836610.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.2425792694091795,
|
|
"epoch": 4.835423197492163,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031721988686496156,
|
|
"loss": 4.862,
|
|
"mean_token_accuracy": 0.20159070789813996,
|
|
"num_tokens": 10845127.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.276719808578491,
|
|
"epoch": 4.839341692789969,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003169244013520638,
|
|
"loss": 4.8581,
|
|
"mean_token_accuracy": 0.1988273099064827,
|
|
"num_tokens": 10853458.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.3454773902893065,
|
|
"epoch": 4.843260188087775,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000316628840913463,
|
|
"loss": 4.9695,
|
|
"mean_token_accuracy": 0.19331085681915283,
|
|
"num_tokens": 10862134.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.299845361709595,
|
|
"epoch": 4.84717868338558,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003163332060773738,
|
|
"loss": 4.9721,
|
|
"mean_token_accuracy": 0.19477631598711015,
|
|
"num_tokens": 10872127.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.280215883255005,
|
|
"epoch": 4.851097178683386,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003160374973721434,
|
|
"loss": 4.8835,
|
|
"mean_token_accuracy": 0.19357476532459258,
|
|
"num_tokens": 10880617.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.300616979598999,
|
|
"epoch": 4.855015673981192,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003157417153262513,
|
|
"loss": 4.8505,
|
|
"mean_token_accuracy": 0.19842603355646132,
|
|
"num_tokens": 10888769.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.270349836349487,
|
|
"epoch": 4.858934169278997,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031544586046830796,
|
|
"loss": 4.827,
|
|
"mean_token_accuracy": 0.19484897553920746,
|
|
"num_tokens": 10897638.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.236645603179932,
|
|
"epoch": 4.862852664576803,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000315149933327054,
|
|
"loss": 4.7815,
|
|
"mean_token_accuracy": 0.19957851022481918,
|
|
"num_tokens": 10906224.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.195483064651489,
|
|
"epoch": 4.866771159874608,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031485393443135914,
|
|
"loss": 4.7863,
|
|
"mean_token_accuracy": 0.2075406566262245,
|
|
"num_tokens": 10915390.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.308579158782959,
|
|
"epoch": 4.870689655172414,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003145578643102214,
|
|
"loss": 4.846,
|
|
"mean_token_accuracy": 0.1972092717885971,
|
|
"num_tokens": 10923719.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.239797592163086,
|
|
"epoch": 4.87460815047022,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003142617234927662,
|
|
"loss": 4.8103,
|
|
"mean_token_accuracy": 0.19712402373552323,
|
|
"num_tokens": 10932610.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.263326168060303,
|
|
"epoch": 4.878526645768025,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031396551250824513,
|
|
"loss": 4.8086,
|
|
"mean_token_accuracy": 0.2004881337285042,
|
|
"num_tokens": 10941315.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.123945045471191,
|
|
"epoch": 4.882445141065831,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003136692318860352,
|
|
"loss": 4.7575,
|
|
"mean_token_accuracy": 0.2049916386604309,
|
|
"num_tokens": 10949781.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.304575204849243,
|
|
"epoch": 4.886363636363637,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003133728821556381,
|
|
"loss": 4.9792,
|
|
"mean_token_accuracy": 0.19220007061958314,
|
|
"num_tokens": 10958207.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.30261197090149,
|
|
"epoch": 4.890282131661442,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00031307646384667854,
|
|
"loss": 4.8635,
|
|
"mean_token_accuracy": 0.19421962946653365,
|
|
"num_tokens": 10967162.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.301410055160522,
|
|
"epoch": 4.894200626959248,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00031277997748890433,
|
|
"loss": 4.924,
|
|
"mean_token_accuracy": 0.19225797355175017,
|
|
"num_tokens": 10975517.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.229112386703491,
|
|
"epoch": 4.898119122257054,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003124834236121847,
|
|
"loss": 4.806,
|
|
"mean_token_accuracy": 0.19817917793989182,
|
|
"num_tokens": 10984668.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.207870864868164,
|
|
"epoch": 4.902037617554859,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031218680274650934,
|
|
"loss": 4.7903,
|
|
"mean_token_accuracy": 0.19721684157848357,
|
|
"num_tokens": 10993475.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.181294250488281,
|
|
"epoch": 4.905956112852665,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00031189011542198794,
|
|
"loss": 4.7585,
|
|
"mean_token_accuracy": 0.2150219902396202,
|
|
"num_tokens": 11002419.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.192610359191894,
|
|
"epoch": 4.90987460815047,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 4.8233,
|
|
"mean_token_accuracy": 0.20186050683259965,
|
|
"num_tokens": 11011271.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.3181178092956545,
|
|
"epoch": 4.913793103448276,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031129654351743816,
|
|
"loss": 4.8656,
|
|
"mean_token_accuracy": 0.1954418435692787,
|
|
"num_tokens": 11020461.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.299691677093506,
|
|
"epoch": 4.917711598746082,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00031099965999821906,
|
|
"loss": 4.7897,
|
|
"mean_token_accuracy": 0.2035943776369095,
|
|
"num_tokens": 11029251.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.272819471359253,
|
|
"epoch": 4.921630094043887,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003107027121417704,
|
|
"loss": 4.8072,
|
|
"mean_token_accuracy": 0.20285213142633438,
|
|
"num_tokens": 11038149.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.325166797637939,
|
|
"epoch": 4.925548589341693,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031040570047878634,
|
|
"loss": 4.8764,
|
|
"mean_token_accuracy": 0.19991490244865417,
|
|
"num_tokens": 11046262.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.307986116409301,
|
|
"epoch": 4.929467084639499,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031010862554007473,
|
|
"loss": 4.922,
|
|
"mean_token_accuracy": 0.19148829579353333,
|
|
"num_tokens": 11055755.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.193844413757324,
|
|
"epoch": 4.933385579937304,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003098114878565567,
|
|
"loss": 4.8287,
|
|
"mean_token_accuracy": 0.1977865993976593,
|
|
"num_tokens": 11064534.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.238815689086914,
|
|
"epoch": 4.93730407523511,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00030951428795926536,
|
|
"loss": 4.8963,
|
|
"mean_token_accuracy": 0.19446095824241638,
|
|
"num_tokens": 11073768.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.2553685188293455,
|
|
"epoch": 4.941222570532915,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003092170263793452,
|
|
"loss": 4.8608,
|
|
"mean_token_accuracy": 0.1992950364947319,
|
|
"num_tokens": 11083683.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.313916015625,
|
|
"epoch": 4.945141065830721,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00030891970364805096,
|
|
"loss": 4.9109,
|
|
"mean_token_accuracy": 0.19331224709749223,
|
|
"num_tokens": 11092368.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.205358505249023,
|
|
"epoch": 4.949059561128527,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003086223202967464,
|
|
"loss": 4.808,
|
|
"mean_token_accuracy": 0.20062625855207444,
|
|
"num_tokens": 11100544.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.253615856170654,
|
|
"epoch": 4.952978056426332,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000308324876856904,
|
|
"loss": 4.8495,
|
|
"mean_token_accuracy": 0.2002647861838341,
|
|
"num_tokens": 11108962.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.320260143280029,
|
|
"epoch": 4.956896551724138,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00030802737386010314,
|
|
"loss": 4.9154,
|
|
"mean_token_accuracy": 0.1996607005596161,
|
|
"num_tokens": 11117011.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.263175678253174,
|
|
"epoch": 4.960815047021944,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003077298118380301,
|
|
"loss": 4.8607,
|
|
"mean_token_accuracy": 0.19937022626399994,
|
|
"num_tokens": 11126009.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.210135173797608,
|
|
"epoch": 4.964733542319749,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003074321913224765,
|
|
"loss": 4.7548,
|
|
"mean_token_accuracy": 0.200472129881382,
|
|
"num_tokens": 11134404.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.290211486816406,
|
|
"epoch": 4.968652037617555,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003071345128453382,
|
|
"loss": 4.8859,
|
|
"mean_token_accuracy": 0.1966423511505127,
|
|
"num_tokens": 11142245.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.2199828147888185,
|
|
"epoch": 4.972570532915361,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003068367769386153,
|
|
"loss": 4.8705,
|
|
"mean_token_accuracy": 0.20033932775259017,
|
|
"num_tokens": 11150881.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.190885925292969,
|
|
"epoch": 4.976489028213166,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00030653898413440973,
|
|
"loss": 4.7988,
|
|
"mean_token_accuracy": 0.20155466347932816,
|
|
"num_tokens": 11159421.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.2747221946716305,
|
|
"epoch": 4.980407523510972,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003062411349649258,
|
|
"loss": 4.8564,
|
|
"mean_token_accuracy": 0.19786821603775023,
|
|
"num_tokens": 11168633.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.2202331066131595,
|
|
"epoch": 4.984326018808778,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00030594322996246816,
|
|
"loss": 4.7206,
|
|
"mean_token_accuracy": 0.20651595145463944,
|
|
"num_tokens": 11177435.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.29007134437561,
|
|
"epoch": 4.988244514106583,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00030564526965944127,
|
|
"loss": 4.9318,
|
|
"mean_token_accuracy": 0.1955521136522293,
|
|
"num_tokens": 11186195.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.240896463394165,
|
|
"epoch": 4.992163009404389,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003053472545883488,
|
|
"loss": 4.7824,
|
|
"mean_token_accuracy": 0.20579567402601243,
|
|
"num_tokens": 11194087.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.258622646331787,
|
|
"epoch": 4.996081504702194,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00030504918528179164,
|
|
"loss": 4.872,
|
|
"mean_token_accuracy": 0.20134468823671342,
|
|
"num_tokens": 11202677.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.266303253173828,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00030475106227246824,
|
|
"loss": 4.9042,
|
|
"mean_token_accuracy": 0.19408685117959976,
|
|
"num_tokens": 11211560.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.2364280223846436,
|
|
"epoch": 5.003918495297806,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003044528860931726,
|
|
"loss": 4.6179,
|
|
"mean_token_accuracy": 0.21948612928390504,
|
|
"num_tokens": 11220186.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.26678581237793,
|
|
"epoch": 5.007836990595611,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003041546572767939,
|
|
"loss": 4.6318,
|
|
"mean_token_accuracy": 0.2137708380818367,
|
|
"num_tokens": 11229148.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.245816469192505,
|
|
"epoch": 5.011755485893417,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003038563763563156,
|
|
"loss": 4.5761,
|
|
"mean_token_accuracy": 0.2217778295278549,
|
|
"num_tokens": 11237585.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.192490291595459,
|
|
"epoch": 5.015673981191223,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00030355804386481374,
|
|
"loss": 4.5241,
|
|
"mean_token_accuracy": 0.21872923225164415,
|
|
"num_tokens": 11246203.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.2333899974823,
|
|
"epoch": 5.019592476489028,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00030325966033545716,
|
|
"loss": 4.65,
|
|
"mean_token_accuracy": 0.21849207133054732,
|
|
"num_tokens": 11254999.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.2397199153900145,
|
|
"epoch": 5.023510971786834,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00030296122630150534,
|
|
"loss": 4.6013,
|
|
"mean_token_accuracy": 0.21860271990299224,
|
|
"num_tokens": 11264110.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.2301438331604,
|
|
"epoch": 5.027429467084639,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00030266274229630844,
|
|
"loss": 4.5346,
|
|
"mean_token_accuracy": 0.22812028229236603,
|
|
"num_tokens": 11273452.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.145605850219726,
|
|
"epoch": 5.031347962382445,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00030236420885330586,
|
|
"loss": 4.5435,
|
|
"mean_token_accuracy": 0.21982075721025468,
|
|
"num_tokens": 11281926.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.229217481613159,
|
|
"epoch": 5.035266457680251,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003020656265060251,
|
|
"loss": 4.6114,
|
|
"mean_token_accuracy": 0.22499076873064042,
|
|
"num_tokens": 11290630.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.268545150756836,
|
|
"epoch": 5.039184952978056,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003017669957880813,
|
|
"loss": 4.6211,
|
|
"mean_token_accuracy": 0.22055379301309586,
|
|
"num_tokens": 11299126.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.283909559249878,
|
|
"epoch": 5.043103448275862,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000301468317233176,
|
|
"loss": 4.7254,
|
|
"mean_token_accuracy": 0.20617055296897888,
|
|
"num_tokens": 11308331.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.246973705291748,
|
|
"epoch": 5.047021943573668,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00030116959137509614,
|
|
"loss": 4.5996,
|
|
"mean_token_accuracy": 0.2162681445479393,
|
|
"num_tokens": 11316245.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.2497552871704105,
|
|
"epoch": 5.050940438871473,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00030087081874771325,
|
|
"loss": 4.607,
|
|
"mean_token_accuracy": 0.21317148506641387,
|
|
"num_tokens": 11324994.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.158684206008911,
|
|
"epoch": 5.054858934169279,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00030057199988498245,
|
|
"loss": 4.5632,
|
|
"mean_token_accuracy": 0.21644362658262253,
|
|
"num_tokens": 11333255.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.200306129455567,
|
|
"epoch": 5.058777429467085,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003002731353209416,
|
|
"loss": 4.5924,
|
|
"mean_token_accuracy": 0.21721154153347016,
|
|
"num_tokens": 11341857.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.228753614425659,
|
|
"epoch": 5.06269592476489,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002999742255897102,
|
|
"loss": 4.7003,
|
|
"mean_token_accuracy": 0.21000982224941253,
|
|
"num_tokens": 11350933.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.170748949050903,
|
|
"epoch": 5.066614420062696,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00029967527122548826,
|
|
"loss": 4.5985,
|
|
"mean_token_accuracy": 0.2100742131471634,
|
|
"num_tokens": 11360346.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.242378234863281,
|
|
"epoch": 5.070532915360501,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002993762727625556,
|
|
"loss": 4.6479,
|
|
"mean_token_accuracy": 0.21341129243373871,
|
|
"num_tokens": 11369401.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.227012014389038,
|
|
"epoch": 5.074451410658307,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002990772307352712,
|
|
"loss": 4.5985,
|
|
"mean_token_accuracy": 0.21415295898914338,
|
|
"num_tokens": 11377882.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.2070471286773685,
|
|
"epoch": 5.078369905956113,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002987781456780715,
|
|
"loss": 4.5947,
|
|
"mean_token_accuracy": 0.22146014422178267,
|
|
"num_tokens": 11386934.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.224644804000855,
|
|
"epoch": 5.082288401253918,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00029847901812547,
|
|
"loss": 4.6433,
|
|
"mean_token_accuracy": 0.21103544533252716,
|
|
"num_tokens": 11395929.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.248774766921997,
|
|
"epoch": 5.086206896551724,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00029817984861205626,
|
|
"loss": 4.583,
|
|
"mean_token_accuracy": 0.22032165974378587,
|
|
"num_tokens": 11405664.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.211741542816162,
|
|
"epoch": 5.09012539184953,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002978806376724945,
|
|
"loss": 4.6729,
|
|
"mean_token_accuracy": 0.2116482824087143,
|
|
"num_tokens": 11414667.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.257224035263062,
|
|
"epoch": 5.094043887147335,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 4.6456,
|
|
"mean_token_accuracy": 0.21135239601135253,
|
|
"num_tokens": 11423907.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.094043887147335,
|
|
"eval_entropy": 5.129956393278847,
|
|
"eval_loss": 5.700186729431152,
|
|
"eval_mean_token_accuracy": 0.16855117677920317,
|
|
"eval_num_tokens": 11423907.0,
|
|
"eval_runtime": 2.8504,
|
|
"eval_samples_per_second": 1446.107,
|
|
"eval_steps_per_second": 181.027,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.22438383102417,
|
|
"epoch": 5.097962382445141,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00029728209365395433,
|
|
"loss": 4.6703,
|
|
"mean_token_accuracy": 0.21903328895568847,
|
|
"num_tokens": 11433824.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.226818180084228,
|
|
"epoch": 5.101880877742946,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00029698276164467105,
|
|
"loss": 4.5388,
|
|
"mean_token_accuracy": 0.21981107890605928,
|
|
"num_tokens": 11442371.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.253150987625122,
|
|
"epoch": 5.105799373040752,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002966833903486284,
|
|
"loss": 4.6102,
|
|
"mean_token_accuracy": 0.21206783056259154,
|
|
"num_tokens": 11451657.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.166251945495605,
|
|
"epoch": 5.109717868338558,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00029638398030085134,
|
|
"loss": 4.5357,
|
|
"mean_token_accuracy": 0.22079613506793977,
|
|
"num_tokens": 11460064.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.213920545578003,
|
|
"epoch": 5.113636363636363,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029608453203643434,
|
|
"loss": 4.629,
|
|
"mean_token_accuracy": 0.21178915053606034,
|
|
"num_tokens": 11469184.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.272148084640503,
|
|
"epoch": 5.117554858934169,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002957850460905398,
|
|
"loss": 4.5742,
|
|
"mean_token_accuracy": 0.2239845335483551,
|
|
"num_tokens": 11478069.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.243617534637451,
|
|
"epoch": 5.121473354231975,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00029548552299839774,
|
|
"loss": 4.6194,
|
|
"mean_token_accuracy": 0.21812971979379653,
|
|
"num_tokens": 11486572.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.182248830795288,
|
|
"epoch": 5.12539184952978,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002951859632953046,
|
|
"loss": 4.6528,
|
|
"mean_token_accuracy": 0.21227784007787703,
|
|
"num_tokens": 11495857.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.194695663452149,
|
|
"epoch": 5.129310344827586,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00029488636751662196,
|
|
"loss": 4.6325,
|
|
"mean_token_accuracy": 0.20998696982860565,
|
|
"num_tokens": 11504875.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.238571500778198,
|
|
"epoch": 5.133228840125392,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002945867361977762,
|
|
"loss": 4.6395,
|
|
"mean_token_accuracy": 0.21048633754253387,
|
|
"num_tokens": 11514509.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.261797952651977,
|
|
"epoch": 5.137147335423197,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00029428706987425705,
|
|
"loss": 4.6687,
|
|
"mean_token_accuracy": 0.20905876457691192,
|
|
"num_tokens": 11523046.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.21034688949585,
|
|
"epoch": 5.141065830721003,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002939873690816167,
|
|
"loss": 4.5689,
|
|
"mean_token_accuracy": 0.2204560175538063,
|
|
"num_tokens": 11531270.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.196085786819458,
|
|
"epoch": 5.144984326018808,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002936876343554692,
|
|
"loss": 4.6311,
|
|
"mean_token_accuracy": 0.21046918332576753,
|
|
"num_tokens": 11539722.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.246710920333863,
|
|
"epoch": 5.148902821316614,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000293387866231489,
|
|
"loss": 4.6502,
|
|
"mean_token_accuracy": 0.21711070239543914,
|
|
"num_tokens": 11549614.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.175629138946533,
|
|
"epoch": 5.15282131661442,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00029308806524541047,
|
|
"loss": 4.5029,
|
|
"mean_token_accuracy": 0.21426898688077928,
|
|
"num_tokens": 11557674.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.16873722076416,
|
|
"epoch": 5.156739811912225,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00029278823193302643,
|
|
"loss": 4.5738,
|
|
"mean_token_accuracy": 0.21743806302547455,
|
|
"num_tokens": 11566691.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.228893041610718,
|
|
"epoch": 5.160658307210031,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00029248836683018774,
|
|
"loss": 4.5558,
|
|
"mean_token_accuracy": 0.22316784709692,
|
|
"num_tokens": 11574824.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.17508544921875,
|
|
"epoch": 5.164576802507837,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00029218847047280197,
|
|
"loss": 4.5743,
|
|
"mean_token_accuracy": 0.21441404074430465,
|
|
"num_tokens": 11582988.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.195289134979248,
|
|
"epoch": 5.168495297805642,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002918885433968323,
|
|
"loss": 4.6503,
|
|
"mean_token_accuracy": 0.21421522051095962,
|
|
"num_tokens": 11591553.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.159967470169067,
|
|
"epoch": 5.172413793103448,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002915885861382974,
|
|
"loss": 4.6668,
|
|
"mean_token_accuracy": 0.2085229054093361,
|
|
"num_tokens": 11600347.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.2218280792236325,
|
|
"epoch": 5.1763322884012535,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00029128859923326935,
|
|
"loss": 4.6675,
|
|
"mean_token_accuracy": 0.2137725681066513,
|
|
"num_tokens": 11609217.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.279037952423096,
|
|
"epoch": 5.1802507836990594,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00029098858321787336,
|
|
"loss": 4.6249,
|
|
"mean_token_accuracy": 0.2146102726459503,
|
|
"num_tokens": 11617355.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.23271803855896,
|
|
"epoch": 5.184169278996865,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00029068853862828677,
|
|
"loss": 4.6127,
|
|
"mean_token_accuracy": 0.21522096991539003,
|
|
"num_tokens": 11626152.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.183769845962525,
|
|
"epoch": 5.1880877742946705,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002903884660007378,
|
|
"loss": 4.586,
|
|
"mean_token_accuracy": 0.215636745095253,
|
|
"num_tokens": 11634868.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.196994590759277,
|
|
"epoch": 5.1920062695924765,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00029008836587150505,
|
|
"loss": 4.6888,
|
|
"mean_token_accuracy": 0.21035524010658263,
|
|
"num_tokens": 11644199.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.298375463485717,
|
|
"epoch": 5.195924764890282,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002897882387769159,
|
|
"loss": 4.7185,
|
|
"mean_token_accuracy": 0.2126232549548149,
|
|
"num_tokens": 11653227.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.143515157699585,
|
|
"epoch": 5.1998432601880875,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002894880852533464,
|
|
"loss": 4.5348,
|
|
"mean_token_accuracy": 0.2287617117166519,
|
|
"num_tokens": 11661628.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.174734497070313,
|
|
"epoch": 5.2037617554858935,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028918790583721936,
|
|
"loss": 4.5896,
|
|
"mean_token_accuracy": 0.21662066876888275,
|
|
"num_tokens": 11670616.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.154978561401367,
|
|
"epoch": 5.2076802507836994,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002888877010650041,
|
|
"loss": 4.6035,
|
|
"mean_token_accuracy": 0.21982019990682602,
|
|
"num_tokens": 11679069.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.254749298095703,
|
|
"epoch": 5.2115987460815045,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00028858747147321527,
|
|
"loss": 4.5921,
|
|
"mean_token_accuracy": 0.22407314628362657,
|
|
"num_tokens": 11688184.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.211394596099853,
|
|
"epoch": 5.2155172413793105,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002882872175984118,
|
|
"loss": 4.6472,
|
|
"mean_token_accuracy": 0.20936754047870637,
|
|
"num_tokens": 11696389.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.18823733329773,
|
|
"epoch": 5.219435736677116,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002879869399771962,
|
|
"loss": 4.6641,
|
|
"mean_token_accuracy": 0.20836714208126067,
|
|
"num_tokens": 11704881.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.266515731811523,
|
|
"epoch": 5.2233542319749215,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00028768663914621295,
|
|
"loss": 4.6991,
|
|
"mean_token_accuracy": 0.21385935842990875,
|
|
"num_tokens": 11713566.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.23788275718689,
|
|
"epoch": 5.2272727272727275,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002873863156421486,
|
|
"loss": 4.6552,
|
|
"mean_token_accuracy": 0.21374809741973877,
|
|
"num_tokens": 11722427.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.212584352493286,
|
|
"epoch": 5.231191222570533,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00028708597000172984,
|
|
"loss": 4.6105,
|
|
"mean_token_accuracy": 0.21856350749731063,
|
|
"num_tokens": 11731010.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.159174060821533,
|
|
"epoch": 5.235109717868339,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002867856027617229,
|
|
"loss": 4.6594,
|
|
"mean_token_accuracy": 0.21367558538913728,
|
|
"num_tokens": 11739569.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.230992269515991,
|
|
"epoch": 5.2390282131661445,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00028648521445893304,
|
|
"loss": 4.6063,
|
|
"mean_token_accuracy": 0.2139237105846405,
|
|
"num_tokens": 11747687.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.2080339908599855,
|
|
"epoch": 5.24294670846395,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002861848056302026,
|
|
"loss": 4.6788,
|
|
"mean_token_accuracy": 0.21584125310182573,
|
|
"num_tokens": 11756713.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.195843315124511,
|
|
"epoch": 5.246865203761756,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00028588437681241106,
|
|
"loss": 4.5875,
|
|
"mean_token_accuracy": 0.2166181594133377,
|
|
"num_tokens": 11765625.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.24586501121521,
|
|
"epoch": 5.250783699059561,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028558392854247333,
|
|
"loss": 4.6715,
|
|
"mean_token_accuracy": 0.21293792724609376,
|
|
"num_tokens": 11774594.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.228031063079834,
|
|
"epoch": 5.254702194357367,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002852834613573391,
|
|
"loss": 4.6192,
|
|
"mean_token_accuracy": 0.21586490720510482,
|
|
"num_tokens": 11782983.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.186589908599854,
|
|
"epoch": 5.258620689655173,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002849829757939922,
|
|
"loss": 4.6091,
|
|
"mean_token_accuracy": 0.22042226791381836,
|
|
"num_tokens": 11791940.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.22114930152893,
|
|
"epoch": 5.262539184952978,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002846824723894487,
|
|
"loss": 4.6307,
|
|
"mean_token_accuracy": 0.21385848373174668,
|
|
"num_tokens": 11800357.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.261589479446411,
|
|
"epoch": 5.266457680250784,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002843819516807572,
|
|
"loss": 4.6691,
|
|
"mean_token_accuracy": 0.20724345296621322,
|
|
"num_tokens": 11809005.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.247757768630981,
|
|
"epoch": 5.27037617554859,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028408141420499685,
|
|
"loss": 4.6378,
|
|
"mean_token_accuracy": 0.20396198332309723,
|
|
"num_tokens": 11818021.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.250593566894532,
|
|
"epoch": 5.274294670846395,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002837808604992768,
|
|
"loss": 4.6575,
|
|
"mean_token_accuracy": 0.21005174070596694,
|
|
"num_tokens": 11826791.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.194195222854614,
|
|
"epoch": 5.278213166144201,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 4.5489,
|
|
"mean_token_accuracy": 0.2235215499997139,
|
|
"num_tokens": 11835110.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.161451387405395,
|
|
"epoch": 5.282131661442007,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002831797065465386,
|
|
"loss": 4.6708,
|
|
"mean_token_accuracy": 0.2132233500480652,
|
|
"num_tokens": 11842658.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.172523736953735,
|
|
"epoch": 5.286050156739812,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00028287910737388,
|
|
"loss": 4.6498,
|
|
"mean_token_accuracy": 0.21333245635032655,
|
|
"num_tokens": 11851028.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.221333599090576,
|
|
"epoch": 5.289968652037618,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000282578494119979,
|
|
"loss": 4.5807,
|
|
"mean_token_accuracy": 0.2207292228937149,
|
|
"num_tokens": 11859471.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.183758878707886,
|
|
"epoch": 5.293887147335424,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028227786732208014,
|
|
"loss": 4.5585,
|
|
"mean_token_accuracy": 0.21526792347431184,
|
|
"num_tokens": 11868625.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.143281030654907,
|
|
"epoch": 5.297805642633229,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00028197722751745246,
|
|
"loss": 4.5973,
|
|
"mean_token_accuracy": 0.21724660694599152,
|
|
"num_tokens": 11877744.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.199517011642456,
|
|
"epoch": 5.301724137931035,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002816765752433879,
|
|
"loss": 4.5838,
|
|
"mean_token_accuracy": 0.22125312238931655,
|
|
"num_tokens": 11886445.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.156913185119629,
|
|
"epoch": 5.30564263322884,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00028137591103720075,
|
|
"loss": 4.6086,
|
|
"mean_token_accuracy": 0.21808119118213654,
|
|
"num_tokens": 11896040.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.129882192611694,
|
|
"epoch": 5.309561128526646,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002810752354362267,
|
|
"loss": 4.6342,
|
|
"mean_token_accuracy": 0.2029130220413208,
|
|
"num_tokens": 11904996.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.2222145080566404,
|
|
"epoch": 5.313479623824452,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000280774548977822,
|
|
"loss": 4.6348,
|
|
"mean_token_accuracy": 0.21673110574483873,
|
|
"num_tokens": 11913357.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.220171165466309,
|
|
"epoch": 5.317398119122257,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028047385219936196,
|
|
"loss": 4.6721,
|
|
"mean_token_accuracy": 0.21784114241600036,
|
|
"num_tokens": 11922196.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.158899259567261,
|
|
"epoch": 5.321316614420063,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028017314563824044,
|
|
"loss": 4.52,
|
|
"mean_token_accuracy": 0.22218949645757674,
|
|
"num_tokens": 11930682.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.109750413894654,
|
|
"epoch": 5.325235109717869,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00027987242983186896,
|
|
"loss": 4.6386,
|
|
"mean_token_accuracy": 0.21207127422094346,
|
|
"num_tokens": 11939470.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.206501483917236,
|
|
"epoch": 5.329153605015674,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00027957170531767525,
|
|
"loss": 4.6272,
|
|
"mean_token_accuracy": 0.21173324137926103,
|
|
"num_tokens": 11948183.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.269264030456543,
|
|
"epoch": 5.33307210031348,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027927097263310294,
|
|
"loss": 4.6626,
|
|
"mean_token_accuracy": 0.21051669865846634,
|
|
"num_tokens": 11957027.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.168219995498657,
|
|
"epoch": 5.336990595611285,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002789702323156099,
|
|
"loss": 4.627,
|
|
"mean_token_accuracy": 0.21025221347808837,
|
|
"num_tokens": 11965871.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.23391170501709,
|
|
"epoch": 5.340909090909091,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00027866948490266815,
|
|
"loss": 4.7194,
|
|
"mean_token_accuracy": 0.19903713315725327,
|
|
"num_tokens": 11974384.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.259866237640381,
|
|
"epoch": 5.344827586206897,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002783687309317618,
|
|
"loss": 4.643,
|
|
"mean_token_accuracy": 0.21134224981069566,
|
|
"num_tokens": 11983892.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.2947350025177,
|
|
"epoch": 5.348746081504702,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002780679709403871,
|
|
"loss": 4.6738,
|
|
"mean_token_accuracy": 0.2144150421023369,
|
|
"num_tokens": 11992990.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.084823656082153,
|
|
"epoch": 5.352664576802508,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027776720546605086,
|
|
"loss": 4.4947,
|
|
"mean_token_accuracy": 0.2264516606926918,
|
|
"num_tokens": 12001679.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.192262363433838,
|
|
"epoch": 5.356583072100314,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002774664350462697,
|
|
"loss": 4.6625,
|
|
"mean_token_accuracy": 0.2140783354640007,
|
|
"num_tokens": 12011300.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.1648256301879885,
|
|
"epoch": 5.360501567398119,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00027716566021856933,
|
|
"loss": 4.6215,
|
|
"mean_token_accuracy": 0.21201424449682235,
|
|
"num_tokens": 12019776.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.178238439559936,
|
|
"epoch": 5.364420062695925,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000276864881520483,
|
|
"loss": 4.6356,
|
|
"mean_token_accuracy": 0.21334771811962128,
|
|
"num_tokens": 12028838.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.23083643913269,
|
|
"epoch": 5.368338557993731,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002765640994895509,
|
|
"loss": 4.5989,
|
|
"mean_token_accuracy": 0.21693123877048492,
|
|
"num_tokens": 12036903.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.189289140701294,
|
|
"epoch": 5.372257053291536,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00027626331466331955,
|
|
"loss": 4.5939,
|
|
"mean_token_accuracy": 0.21167089641094208,
|
|
"num_tokens": 12045838.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.261508846282959,
|
|
"epoch": 5.376175548589342,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00027596252757933995,
|
|
"loss": 4.7633,
|
|
"mean_token_accuracy": 0.20112868696451186,
|
|
"num_tokens": 12055791.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.180715942382813,
|
|
"epoch": 5.380094043887147,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002756617387751674,
|
|
"loss": 4.5515,
|
|
"mean_token_accuracy": 0.21855847239494325,
|
|
"num_tokens": 12064842.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.220953941345215,
|
|
"epoch": 5.384012539184953,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002753609487883606,
|
|
"loss": 4.6242,
|
|
"mean_token_accuracy": 0.21866626739501954,
|
|
"num_tokens": 12073487.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.1472776412963865,
|
|
"epoch": 5.387931034482759,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00027506015815647965,
|
|
"loss": 4.5806,
|
|
"mean_token_accuracy": 0.21335016340017318,
|
|
"num_tokens": 12082000.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.206481075286865,
|
|
"epoch": 5.391849529780564,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00027475936741708636,
|
|
"loss": 4.6395,
|
|
"mean_token_accuracy": 0.2065572366118431,
|
|
"num_tokens": 12090347.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.171456718444825,
|
|
"epoch": 5.39576802507837,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002744585771077425,
|
|
"loss": 4.6005,
|
|
"mean_token_accuracy": 0.22072034925222397,
|
|
"num_tokens": 12098340.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.192743921279908,
|
|
"epoch": 5.399686520376176,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00027415778776600913,
|
|
"loss": 4.636,
|
|
"mean_token_accuracy": 0.21654542535543442,
|
|
"num_tokens": 12106771.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.228982973098755,
|
|
"epoch": 5.403605015673981,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002738569999294456,
|
|
"loss": 4.6473,
|
|
"mean_token_accuracy": 0.21133457720279694,
|
|
"num_tokens": 12115436.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.169658994674682,
|
|
"epoch": 5.407523510971787,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002735562141356085,
|
|
"loss": 4.5371,
|
|
"mean_token_accuracy": 0.22196700274944306,
|
|
"num_tokens": 12123939.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.296599388122559,
|
|
"epoch": 5.411442006269592,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002732554309220509,
|
|
"loss": 4.7664,
|
|
"mean_token_accuracy": 0.20740340948104857,
|
|
"num_tokens": 12132803.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.18677716255188,
|
|
"epoch": 5.415360501567398,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002729546508263211,
|
|
"loss": 4.6967,
|
|
"mean_token_accuracy": 0.21406860798597335,
|
|
"num_tokens": 12141875.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.212797355651856,
|
|
"epoch": 5.419278996865204,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002726538743859618,
|
|
"loss": 4.6123,
|
|
"mean_token_accuracy": 0.22063491195440293,
|
|
"num_tokens": 12150685.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.1996142864227295,
|
|
"epoch": 5.423197492163009,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002723531021385095,
|
|
"loss": 4.5965,
|
|
"mean_token_accuracy": 0.21483431309461593,
|
|
"num_tokens": 12159070.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.193313980102539,
|
|
"epoch": 5.427115987460815,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002720523346214928,
|
|
"loss": 4.6327,
|
|
"mean_token_accuracy": 0.2149344339966774,
|
|
"num_tokens": 12167676.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.149933815002441,
|
|
"epoch": 5.431034482758621,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00027175157237243204,
|
|
"loss": 4.6088,
|
|
"mean_token_accuracy": 0.21479639559984207,
|
|
"num_tokens": 12176222.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.215902614593506,
|
|
"epoch": 5.434952978056426,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002714508159288382,
|
|
"loss": 4.6219,
|
|
"mean_token_accuracy": 0.22018057554960252,
|
|
"num_tokens": 12185171.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.255028009414673,
|
|
"epoch": 5.438871473354232,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002711500658282118,
|
|
"loss": 4.68,
|
|
"mean_token_accuracy": 0.2202504798769951,
|
|
"num_tokens": 12193875.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.144360685348511,
|
|
"epoch": 5.442789968652038,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00027084932260804193,
|
|
"loss": 4.6034,
|
|
"mean_token_accuracy": 0.21905288845300674,
|
|
"num_tokens": 12202681.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.167861270904541,
|
|
"epoch": 5.446708463949843,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002705485868058056,
|
|
"loss": 4.5993,
|
|
"mean_token_accuracy": 0.2227531775832176,
|
|
"num_tokens": 12210854.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.268987607955933,
|
|
"epoch": 5.450626959247649,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00027024785895896644,
|
|
"loss": 4.6355,
|
|
"mean_token_accuracy": 0.212411367893219,
|
|
"num_tokens": 12219643.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.22519154548645,
|
|
"epoch": 5.454545454545454,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00026994713960497383,
|
|
"loss": 4.6313,
|
|
"mean_token_accuracy": 0.22014989107847213,
|
|
"num_tokens": 12228767.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.206828022003174,
|
|
"epoch": 5.45846394984326,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00026964642928126197,
|
|
"loss": 4.7114,
|
|
"mean_token_accuracy": 0.21292246282100677,
|
|
"num_tokens": 12237468.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.165378379821777,
|
|
"epoch": 5.462382445141066,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 4.6778,
|
|
"mean_token_accuracy": 0.21189580857753754,
|
|
"num_tokens": 12246040.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.251343154907227,
|
|
"epoch": 5.466300940438871,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026904503787433614,
|
|
"loss": 4.7497,
|
|
"mean_token_accuracy": 0.20434968024492264,
|
|
"num_tokens": 12254646.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.211121416091919,
|
|
"epoch": 5.470219435736677,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000268744357865906,
|
|
"loss": 4.6316,
|
|
"mean_token_accuracy": 0.21168104857206343,
|
|
"num_tokens": 12263306.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.181400871276855,
|
|
"epoch": 5.474137931034483,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026844368903732263,
|
|
"loss": 4.6342,
|
|
"mean_token_accuracy": 0.21740086674690245,
|
|
"num_tokens": 12272498.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.130239534378052,
|
|
"epoch": 5.478056426332288,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026814303192593015,
|
|
"loss": 4.52,
|
|
"mean_token_accuracy": 0.22091476023197174,
|
|
"num_tokens": 12281592.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.184746170043946,
|
|
"epoch": 5.481974921630094,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026784238706905136,
|
|
"loss": 4.632,
|
|
"mean_token_accuracy": 0.21792078912258148,
|
|
"num_tokens": 12290112.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.2326037883758545,
|
|
"epoch": 5.485893416927899,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002675417550039874,
|
|
"loss": 4.682,
|
|
"mean_token_accuracy": 0.21183982044458388,
|
|
"num_tokens": 12299868.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 5.485893416927899,
|
|
"eval_entropy": 5.1245528182318045,
|
|
"eval_loss": 5.686053276062012,
|
|
"eval_mean_token_accuracy": 0.16947007936591557,
|
|
"eval_num_tokens": 12299868.0,
|
|
"eval_runtime": 2.8502,
|
|
"eval_samples_per_second": 1446.233,
|
|
"eval_steps_per_second": 181.042,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.222954845428466,
|
|
"epoch": 5.489811912225705,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026724113626801655,
|
|
"loss": 4.6121,
|
|
"mean_token_accuracy": 0.21416960805654525,
|
|
"num_tokens": 12307903.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.2783678531646725,
|
|
"epoch": 5.493730407523511,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002669405313983932,
|
|
"loss": 4.7249,
|
|
"mean_token_accuracy": 0.21144020259380342,
|
|
"num_tokens": 12317287.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.288879156112671,
|
|
"epoch": 5.497648902821316,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000266639940932347,
|
|
"loss": 4.7237,
|
|
"mean_token_accuracy": 0.21198034137487412,
|
|
"num_tokens": 12325770.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.191423559188843,
|
|
"epoch": 5.501567398119122,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002663393654070819,
|
|
"loss": 4.6378,
|
|
"mean_token_accuracy": 0.20911073982715606,
|
|
"num_tokens": 12334800.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.2338494777679445,
|
|
"epoch": 5.505485893416928,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002660388053597752,
|
|
"loss": 4.6669,
|
|
"mean_token_accuracy": 0.2108100339770317,
|
|
"num_tokens": 12343362.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.235997581481934,
|
|
"epoch": 5.509404388714733,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00026573826132757623,
|
|
"loss": 4.6796,
|
|
"mean_token_accuracy": 0.2062300980091095,
|
|
"num_tokens": 12352277.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.193193149566651,
|
|
"epoch": 5.513322884012539,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00026543773384760593,
|
|
"loss": 4.5273,
|
|
"mean_token_accuracy": 0.22276844531297685,
|
|
"num_tokens": 12361244.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.184199094772339,
|
|
"epoch": 5.517241379310345,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002651372234569557,
|
|
"loss": 4.6317,
|
|
"mean_token_accuracy": 0.2114301323890686,
|
|
"num_tokens": 12370228.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.171536159515381,
|
|
"epoch": 5.52115987460815,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002648367306926864,
|
|
"loss": 4.5973,
|
|
"mean_token_accuracy": 0.221302630007267,
|
|
"num_tokens": 12379392.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.176364278793335,
|
|
"epoch": 5.525078369905956,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002645362560918271,
|
|
"loss": 4.6723,
|
|
"mean_token_accuracy": 0.2092619866132736,
|
|
"num_tokens": 12387334.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.223438119888305,
|
|
"epoch": 5.528996865203762,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000264235800191375,
|
|
"loss": 4.5899,
|
|
"mean_token_accuracy": 0.2184743896126747,
|
|
"num_tokens": 12396875.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.247631692886353,
|
|
"epoch": 5.532915360501567,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002639353635282932,
|
|
"loss": 4.5963,
|
|
"mean_token_accuracy": 0.22194595485925675,
|
|
"num_tokens": 12405123.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.28490104675293,
|
|
"epoch": 5.536833855799373,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026363494663951076,
|
|
"loss": 4.7179,
|
|
"mean_token_accuracy": 0.2074203833937645,
|
|
"num_tokens": 12413208.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.22181978225708,
|
|
"epoch": 5.540752351097178,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026333455006192146,
|
|
"loss": 4.7227,
|
|
"mean_token_accuracy": 0.206602481007576,
|
|
"num_tokens": 12421704.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.226918697357178,
|
|
"epoch": 5.544670846394984,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00026303417433238267,
|
|
"loss": 4.6652,
|
|
"mean_token_accuracy": 0.2067098543047905,
|
|
"num_tokens": 12430610.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.184394121170044,
|
|
"epoch": 5.54858934169279,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002627338199877146,
|
|
"loss": 4.5752,
|
|
"mean_token_accuracy": 0.21671690046787262,
|
|
"num_tokens": 12439867.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.192955398559571,
|
|
"epoch": 5.552507836990595,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002624334875646989,
|
|
"loss": 4.6464,
|
|
"mean_token_accuracy": 0.213504059612751,
|
|
"num_tokens": 12449540.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.19539966583252,
|
|
"epoch": 5.556426332288401,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026213317760007864,
|
|
"loss": 4.6609,
|
|
"mean_token_accuracy": 0.21679029017686843,
|
|
"num_tokens": 12458138.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.287289094924927,
|
|
"epoch": 5.560344827586206,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002618328906305562,
|
|
"loss": 4.7659,
|
|
"mean_token_accuracy": 0.20251403450965882,
|
|
"num_tokens": 12467921.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.28335919380188,
|
|
"epoch": 5.564263322884012,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002615326271927931,
|
|
"loss": 4.6209,
|
|
"mean_token_accuracy": 0.21365261524915696,
|
|
"num_tokens": 12476873.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.260688781738281,
|
|
"epoch": 5.568181818181818,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026123238782340893,
|
|
"loss": 4.6603,
|
|
"mean_token_accuracy": 0.21147153973579408,
|
|
"num_tokens": 12485924.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.1946587562561035,
|
|
"epoch": 5.572100313479623,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026093217305898003,
|
|
"loss": 4.6015,
|
|
"mean_token_accuracy": 0.21444240510463713,
|
|
"num_tokens": 12494533.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.116658115386963,
|
|
"epoch": 5.576018808777429,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026063198343603887,
|
|
"loss": 4.5337,
|
|
"mean_token_accuracy": 0.22517998665571212,
|
|
"num_tokens": 12503137.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.1550921440124515,
|
|
"epoch": 5.579937304075235,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00026033181949107286,
|
|
"loss": 4.6688,
|
|
"mean_token_accuracy": 0.2161506548523903,
|
|
"num_tokens": 12511701.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.158056926727295,
|
|
"epoch": 5.58385579937304,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002600316817605239,
|
|
"loss": 4.6456,
|
|
"mean_token_accuracy": 0.22071924954652786,
|
|
"num_tokens": 12520097.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.195622444152832,
|
|
"epoch": 5.587774294670846,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002597315707807866,
|
|
"loss": 4.6524,
|
|
"mean_token_accuracy": 0.21899636685848237,
|
|
"num_tokens": 12528526.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.191586303710937,
|
|
"epoch": 5.591692789968652,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025943148708820784,
|
|
"loss": 4.659,
|
|
"mean_token_accuracy": 0.21588651835918427,
|
|
"num_tokens": 12537888.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.174389362335205,
|
|
"epoch": 5.5956112852664575,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002591314312190861,
|
|
"loss": 4.638,
|
|
"mean_token_accuracy": 0.21484518945217132,
|
|
"num_tokens": 12546396.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.171988105773925,
|
|
"epoch": 5.599529780564263,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025883140370966975,
|
|
"loss": 4.578,
|
|
"mean_token_accuracy": 0.21792082637548446,
|
|
"num_tokens": 12554985.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.210405731201172,
|
|
"epoch": 5.603448275862069,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002585314050961565,
|
|
"loss": 4.6011,
|
|
"mean_token_accuracy": 0.23491737246513367,
|
|
"num_tokens": 12564379.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.210981035232544,
|
|
"epoch": 5.6073667711598745,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002582314359146927,
|
|
"loss": 4.6468,
|
|
"mean_token_accuracy": 0.20984706431627273,
|
|
"num_tokens": 12572655.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.212760877609253,
|
|
"epoch": 5.61128526645768,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00025793149670137175,
|
|
"loss": 4.6651,
|
|
"mean_token_accuracy": 0.21172828525304793,
|
|
"num_tokens": 12581778.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.152795648574829,
|
|
"epoch": 5.6152037617554855,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002576315879922337,
|
|
"loss": 4.5601,
|
|
"mean_token_accuracy": 0.21682798862457275,
|
|
"num_tokens": 12590324.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.189353799819946,
|
|
"epoch": 5.6191222570532915,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025733171032326404,
|
|
"loss": 4.6886,
|
|
"mean_token_accuracy": 0.20380861461162567,
|
|
"num_tokens": 12599761.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.228964996337891,
|
|
"epoch": 5.6230407523510975,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002570318642303929,
|
|
"loss": 4.6875,
|
|
"mean_token_accuracy": 0.21000344902276993,
|
|
"num_tokens": 12608853.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.187328100204468,
|
|
"epoch": 5.6269592476489025,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00025673205024949377,
|
|
"loss": 4.5953,
|
|
"mean_token_accuracy": 0.21664550453424453,
|
|
"num_tokens": 12617434.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.194105529785157,
|
|
"epoch": 5.6308777429467085,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002564322689163828,
|
|
"loss": 4.6454,
|
|
"mean_token_accuracy": 0.22051969766616822,
|
|
"num_tokens": 12626021.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.214135694503784,
|
|
"epoch": 5.6347962382445145,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00025613252076681786,
|
|
"loss": 4.6175,
|
|
"mean_token_accuracy": 0.21851568967103957,
|
|
"num_tokens": 12634952.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.176747035980225,
|
|
"epoch": 5.63871473354232,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002558328063364976,
|
|
"loss": 4.6335,
|
|
"mean_token_accuracy": 0.21343381851911544,
|
|
"num_tokens": 12644094.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.198800420761108,
|
|
"epoch": 5.6426332288401255,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002555331261610603,
|
|
"loss": 4.662,
|
|
"mean_token_accuracy": 0.20895349383354186,
|
|
"num_tokens": 12652687.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.191704130172729,
|
|
"epoch": 5.646551724137931,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 4.6176,
|
|
"mean_token_accuracy": 0.22008429914712907,
|
|
"num_tokens": 12661437.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.242120981216431,
|
|
"epoch": 5.650470219435737,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025493387071708024,
|
|
"loss": 4.6693,
|
|
"mean_token_accuracy": 0.21642181277275085,
|
|
"num_tokens": 12669893.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.191336011886596,
|
|
"epoch": 5.6543887147335425,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00025463429651950426,
|
|
"loss": 4.6865,
|
|
"mean_token_accuracy": 0.21414238065481186,
|
|
"num_tokens": 12679349.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.213102388381958,
|
|
"epoch": 5.658307210031348,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002543347587187425,
|
|
"loss": 4.6339,
|
|
"mean_token_accuracy": 0.20388923734426498,
|
|
"num_tokens": 12687163.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.218425703048706,
|
|
"epoch": 5.662225705329154,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00025403525785011767,
|
|
"loss": 4.6378,
|
|
"mean_token_accuracy": 0.21101368367671966,
|
|
"num_tokens": 12695646.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.220873165130615,
|
|
"epoch": 5.66614420062696,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025373579444888634,
|
|
"loss": 4.5952,
|
|
"mean_token_accuracy": 0.2185002788901329,
|
|
"num_tokens": 12705144.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.196035432815552,
|
|
"epoch": 5.670062695924765,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000253436369050238,
|
|
"loss": 4.6174,
|
|
"mean_token_accuracy": 0.21428689509630203,
|
|
"num_tokens": 12713881.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.18876085281372,
|
|
"epoch": 5.673981191222571,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025313698218929474,
|
|
"loss": 4.589,
|
|
"mean_token_accuracy": 0.224277301132679,
|
|
"num_tokens": 12722452.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.17009539604187,
|
|
"epoch": 5.677899686520377,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025283763440110923,
|
|
"loss": 4.605,
|
|
"mean_token_accuracy": 0.2155233696103096,
|
|
"num_tokens": 12731105.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.153394174575806,
|
|
"epoch": 5.681818181818182,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002525383262206645,
|
|
"loss": 4.63,
|
|
"mean_token_accuracy": 0.2251857191324234,
|
|
"num_tokens": 12739096.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.191949272155762,
|
|
"epoch": 5.685736677115988,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002522390581828726,
|
|
"loss": 4.6162,
|
|
"mean_token_accuracy": 0.21707140356302262,
|
|
"num_tokens": 12747910.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.134769105911255,
|
|
"epoch": 5.689655172413794,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025193983082257444,
|
|
"loss": 4.5267,
|
|
"mean_token_accuracy": 0.22594287246465683,
|
|
"num_tokens": 12757387.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.15739893913269,
|
|
"epoch": 5.693573667711599,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002516406446745376,
|
|
"loss": 4.6402,
|
|
"mean_token_accuracy": 0.2178972691297531,
|
|
"num_tokens": 12766678.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.157175922393799,
|
|
"epoch": 5.697492163009405,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002513415002734562,
|
|
"loss": 4.601,
|
|
"mean_token_accuracy": 0.21595963537693025,
|
|
"num_tokens": 12775499.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.274277925491333,
|
|
"epoch": 5.70141065830721,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00025104239815394995,
|
|
"loss": 4.6955,
|
|
"mean_token_accuracy": 0.21228510439395903,
|
|
"num_tokens": 12784173.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.256857585906983,
|
|
"epoch": 5.705329153605016,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002507433388505627,
|
|
"loss": 4.7101,
|
|
"mean_token_accuracy": 0.21198654919862747,
|
|
"num_tokens": 12792991.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.153625726699829,
|
|
"epoch": 5.709247648902822,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00025044432289776197,
|
|
"loss": 4.6306,
|
|
"mean_token_accuracy": 0.21308987885713576,
|
|
"num_tokens": 12801868.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.2092146396636965,
|
|
"epoch": 5.713166144200627,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00025014535082993765,
|
|
"loss": 4.6525,
|
|
"mean_token_accuracy": 0.22001767307519912,
|
|
"num_tokens": 12810467.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.200398397445679,
|
|
"epoch": 5.717084639498433,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002498464231814015,
|
|
"loss": 4.6444,
|
|
"mean_token_accuracy": 0.21809316873550416,
|
|
"num_tokens": 12818874.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.169918727874756,
|
|
"epoch": 5.721003134796238,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024954754048638555,
|
|
"loss": 4.5497,
|
|
"mean_token_accuracy": 0.22096756845712662,
|
|
"num_tokens": 12828029.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.18907527923584,
|
|
"epoch": 5.724921630094044,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00024924870327904165,
|
|
"loss": 4.6115,
|
|
"mean_token_accuracy": 0.21728481501340866,
|
|
"num_tokens": 12836185.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.2753080368042,
|
|
"epoch": 5.72884012539185,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002489499120934403,
|
|
"loss": 4.7465,
|
|
"mean_token_accuracy": 0.20540612190961838,
|
|
"num_tokens": 12845810.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.165975522994995,
|
|
"epoch": 5.732758620689655,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024865116746357004,
|
|
"loss": 4.5153,
|
|
"mean_token_accuracy": 0.2299176871776581,
|
|
"num_tokens": 12854567.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.123243904113769,
|
|
"epoch": 5.736677115987461,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002483524699233356,
|
|
"loss": 4.6046,
|
|
"mean_token_accuracy": 0.219584883749485,
|
|
"num_tokens": 12863172.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.191251802444458,
|
|
"epoch": 5.740595611285267,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024805382000655824,
|
|
"loss": 4.7065,
|
|
"mean_token_accuracy": 0.21064341068267822,
|
|
"num_tokens": 12873335.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.258235502243042,
|
|
"epoch": 5.744514106583072,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002477552182469736,
|
|
"loss": 4.6554,
|
|
"mean_token_accuracy": 0.21603169292211533,
|
|
"num_tokens": 12882695.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.146210479736328,
|
|
"epoch": 5.748432601880878,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002474566651782315,
|
|
"loss": 4.6784,
|
|
"mean_token_accuracy": 0.20781407207250596,
|
|
"num_tokens": 12890739.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.120088768005371,
|
|
"epoch": 5.752351097178684,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002471581613338946,
|
|
"loss": 4.6395,
|
|
"mean_token_accuracy": 0.21514222472906114,
|
|
"num_tokens": 12899503.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.240623044967651,
|
|
"epoch": 5.756269592476489,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002468597072474378,
|
|
"loss": 4.6715,
|
|
"mean_token_accuracy": 0.21621336787939072,
|
|
"num_tokens": 12908405.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.242657709121704,
|
|
"epoch": 5.760188087774295,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00024656130345224685,
|
|
"loss": 4.7337,
|
|
"mean_token_accuracy": 0.2059269890189171,
|
|
"num_tokens": 12917529.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.239910984039307,
|
|
"epoch": 5.764106583072101,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00024626295048161775,
|
|
"loss": 4.671,
|
|
"mean_token_accuracy": 0.21257943511009217,
|
|
"num_tokens": 12926183.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.297617721557617,
|
|
"epoch": 5.768025078369906,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002459646488687557,
|
|
"loss": 4.7479,
|
|
"mean_token_accuracy": 0.20610129833221436,
|
|
"num_tokens": 12934598.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.187813949584961,
|
|
"epoch": 5.771943573667712,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002456663991467739,
|
|
"loss": 4.6307,
|
|
"mean_token_accuracy": 0.20830924212932586,
|
|
"num_tokens": 12943101.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.23620285987854,
|
|
"epoch": 5.775862068965517,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00024536820184869313,
|
|
"loss": 4.686,
|
|
"mean_token_accuracy": 0.2076297789812088,
|
|
"num_tokens": 12951717.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.2588213920593265,
|
|
"epoch": 5.779780564263323,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002450700575074401,
|
|
"loss": 4.7511,
|
|
"mean_token_accuracy": 0.20891714245080947,
|
|
"num_tokens": 12960761.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.186924076080322,
|
|
"epoch": 5.783699059561129,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024477196665584735,
|
|
"loss": 4.5424,
|
|
"mean_token_accuracy": 0.2324463203549385,
|
|
"num_tokens": 12969216.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.208795213699341,
|
|
"epoch": 5.787617554858934,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002444739298266513,
|
|
"loss": 4.6337,
|
|
"mean_token_accuracy": 0.21086293011903762,
|
|
"num_tokens": 12977260.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.1615358829498295,
|
|
"epoch": 5.79153605015674,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00024417594755249227,
|
|
"loss": 4.6209,
|
|
"mean_token_accuracy": 0.21702422797679902,
|
|
"num_tokens": 12985383.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.214687299728394,
|
|
"epoch": 5.795454545454545,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024387802036591295,
|
|
"loss": 4.6321,
|
|
"mean_token_accuracy": 0.21540583074092864,
|
|
"num_tokens": 12994675.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.231916618347168,
|
|
"epoch": 5.799373040752351,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002435801487993573,
|
|
"loss": 4.6259,
|
|
"mean_token_accuracy": 0.21895947605371474,
|
|
"num_tokens": 13002979.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.184161949157715,
|
|
"epoch": 5.803291536050157,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00024328233338517024,
|
|
"loss": 4.6247,
|
|
"mean_token_accuracy": 0.2098724588751793,
|
|
"num_tokens": 13011343.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.171076393127441,
|
|
"epoch": 5.807210031347962,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024298457465559626,
|
|
"loss": 4.7562,
|
|
"mean_token_accuracy": 0.2046315610408783,
|
|
"num_tokens": 13019684.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.196204757690429,
|
|
"epoch": 5.811128526645768,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024268687314277844,
|
|
"loss": 4.6194,
|
|
"mean_token_accuracy": 0.21375007480382918,
|
|
"num_tokens": 13028358.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.133615779876709,
|
|
"epoch": 5.815047021943574,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024238922937875758,
|
|
"loss": 4.5857,
|
|
"mean_token_accuracy": 0.2163202002644539,
|
|
"num_tokens": 13037092.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.234962463378906,
|
|
"epoch": 5.818965517241379,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002420916438954714,
|
|
"loss": 4.7186,
|
|
"mean_token_accuracy": 0.21346863359212875,
|
|
"num_tokens": 13045624.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.212952995300293,
|
|
"epoch": 5.822884012539185,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002417941172247535,
|
|
"loss": 4.6699,
|
|
"mean_token_accuracy": 0.2139253318309784,
|
|
"num_tokens": 13054384.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.183205509185791,
|
|
"epoch": 5.826802507836991,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00024149664989833215,
|
|
"loss": 4.6985,
|
|
"mean_token_accuracy": 0.21559885144233704,
|
|
"num_tokens": 13063786.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.256461191177368,
|
|
"epoch": 5.830721003134796,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 4.7126,
|
|
"mean_token_accuracy": 0.21031047850847245,
|
|
"num_tokens": 13073446.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.171071338653564,
|
|
"epoch": 5.834639498432602,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024090189540476153,
|
|
"loss": 4.6471,
|
|
"mean_token_accuracy": 0.21195615082979202,
|
|
"num_tokens": 13081780.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.28557014465332,
|
|
"epoch": 5.838557993730408,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00024060460930053513,
|
|
"loss": 4.6647,
|
|
"mean_token_accuracy": 0.21461522281169892,
|
|
"num_tokens": 13089995.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.190529489517212,
|
|
"epoch": 5.842476489028213,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00024030738466644875,
|
|
"loss": 4.6331,
|
|
"mean_token_accuracy": 0.21480135023593902,
|
|
"num_tokens": 13098767.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.152538633346557,
|
|
"epoch": 5.846394984326019,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002400102220336911,
|
|
"loss": 4.692,
|
|
"mean_token_accuracy": 0.21186856627464296,
|
|
"num_tokens": 13107744.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.16839747428894,
|
|
"epoch": 5.850313479623824,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002397131219333401,
|
|
"loss": 4.5915,
|
|
"mean_token_accuracy": 0.22014763653278352,
|
|
"num_tokens": 13116400.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.17002305984497,
|
|
"epoch": 5.85423197492163,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002394160848963617,
|
|
"loss": 4.5998,
|
|
"mean_token_accuracy": 0.22167249023914337,
|
|
"num_tokens": 13124973.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.267059373855591,
|
|
"epoch": 5.858150470219436,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023911911145360916,
|
|
"loss": 4.6551,
|
|
"mean_token_accuracy": 0.21477824151515962,
|
|
"num_tokens": 13133061.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.252498197555542,
|
|
"epoch": 5.862068965517241,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023882220213582235,
|
|
"loss": 4.7098,
|
|
"mean_token_accuracy": 0.20931968986988067,
|
|
"num_tokens": 13141322.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.180319023132324,
|
|
"epoch": 5.865987460815047,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002385253574736262,
|
|
"loss": 4.6624,
|
|
"mean_token_accuracy": 0.2172859087586403,
|
|
"num_tokens": 13150553.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.314070415496826,
|
|
"epoch": 5.869905956112853,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023822857799753034,
|
|
"loss": 4.7455,
|
|
"mean_token_accuracy": 0.20848397612571717,
|
|
"num_tokens": 13160308.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.2595438957214355,
|
|
"epoch": 5.873824451410658,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023793186423792784,
|
|
"loss": 4.7071,
|
|
"mean_token_accuracy": 0.21092278957366944,
|
|
"num_tokens": 13168219.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.232971715927124,
|
|
"epoch": 5.877742946708464,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023763521672509432,
|
|
"loss": 4.6861,
|
|
"mean_token_accuracy": 0.21097994297742845,
|
|
"num_tokens": 13176980.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 5.877742946708464,
|
|
"eval_entropy": 5.120905934378158,
|
|
"eval_loss": 5.669462203979492,
|
|
"eval_mean_token_accuracy": 0.17142084718125966,
|
|
"eval_num_tokens": 13176980.0,
|
|
"eval_runtime": 3.0239,
|
|
"eval_samples_per_second": 1363.151,
|
|
"eval_steps_per_second": 170.642,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.263127660751342,
|
|
"epoch": 5.881661442006269,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023733863598918688,
|
|
"loss": 4.7313,
|
|
"mean_token_accuracy": 0.2059478595852852,
|
|
"num_tokens": 13186267.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.251651239395142,
|
|
"epoch": 5.885579937304075,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002370421225602433,
|
|
"loss": 4.7373,
|
|
"mean_token_accuracy": 0.2102463498711586,
|
|
"num_tokens": 13195457.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.258962488174438,
|
|
"epoch": 5.889498432601881,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023674567696818145,
|
|
"loss": 4.7129,
|
|
"mean_token_accuracy": 0.2111515983939171,
|
|
"num_tokens": 13203707.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.242599439620972,
|
|
"epoch": 5.893416927899686,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002364492997427975,
|
|
"loss": 4.6257,
|
|
"mean_token_accuracy": 0.21226882934570312,
|
|
"num_tokens": 13213013.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.182432508468628,
|
|
"epoch": 5.897335423197492,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023615299141376545,
|
|
"loss": 4.6352,
|
|
"mean_token_accuracy": 0.2174764335155487,
|
|
"num_tokens": 13222356.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.182303476333618,
|
|
"epoch": 5.901253918495298,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023585675251063665,
|
|
"loss": 4.6593,
|
|
"mean_token_accuracy": 0.21198585033416747,
|
|
"num_tokens": 13231791.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.171900987625122,
|
|
"epoch": 5.905172413793103,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002355605835628379,
|
|
"loss": 4.587,
|
|
"mean_token_accuracy": 0.2223715901374817,
|
|
"num_tokens": 13240445.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.218419599533081,
|
|
"epoch": 5.909090909090909,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023526448509967097,
|
|
"loss": 4.6857,
|
|
"mean_token_accuracy": 0.20360318571329117,
|
|
"num_tokens": 13249048.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.175121688842774,
|
|
"epoch": 5.913009404388715,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023496845765031215,
|
|
"loss": 4.5596,
|
|
"mean_token_accuracy": 0.22203529626131058,
|
|
"num_tokens": 13258206.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.242780303955078,
|
|
"epoch": 5.91692789968652,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023467250174381021,
|
|
"loss": 4.6502,
|
|
"mean_token_accuracy": 0.21152700930833818,
|
|
"num_tokens": 13266756.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.194359159469604,
|
|
"epoch": 5.920846394984326,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00023437661790908648,
|
|
"loss": 4.7145,
|
|
"mean_token_accuracy": 0.2064350426197052,
|
|
"num_tokens": 13276033.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.251874828338623,
|
|
"epoch": 5.924764890282132,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023408080667493332,
|
|
"loss": 4.661,
|
|
"mean_token_accuracy": 0.2151060238480568,
|
|
"num_tokens": 13285151.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.147282314300537,
|
|
"epoch": 5.928683385579937,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00023378506857001335,
|
|
"loss": 4.5937,
|
|
"mean_token_accuracy": 0.22084648609161378,
|
|
"num_tokens": 13293412.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.19869647026062,
|
|
"epoch": 5.932601880877743,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023348940412285858,
|
|
"loss": 4.6804,
|
|
"mean_token_accuracy": 0.2102070078253746,
|
|
"num_tokens": 13301982.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.240865325927734,
|
|
"epoch": 5.936520376175548,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00023319381386186923,
|
|
"loss": 4.6995,
|
|
"mean_token_accuracy": 0.21361820101737977,
|
|
"num_tokens": 13309766.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.212366390228271,
|
|
"epoch": 5.940438871473354,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000232898298315313,
|
|
"loss": 4.6272,
|
|
"mean_token_accuracy": 0.21989676207304001,
|
|
"num_tokens": 13317960.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.279466390609741,
|
|
"epoch": 5.94435736677116,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002326028580113243,
|
|
"loss": 4.714,
|
|
"mean_token_accuracy": 0.2081019699573517,
|
|
"num_tokens": 13326909.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.299083232879639,
|
|
"epoch": 5.948275862068965,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00023230749347790265,
|
|
"loss": 4.7234,
|
|
"mean_token_accuracy": 0.2039751812815666,
|
|
"num_tokens": 13335861.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.247184705734253,
|
|
"epoch": 5.952194357366771,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023201220524291232,
|
|
"loss": 4.6878,
|
|
"mean_token_accuracy": 0.20759655982255937,
|
|
"num_tokens": 13345253.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.204712629318237,
|
|
"epoch": 5.956112852664576,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00023171699383408152,
|
|
"loss": 4.6571,
|
|
"mean_token_accuracy": 0.21724827140569686,
|
|
"num_tokens": 13354353.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.238126134872436,
|
|
"epoch": 5.960031347962382,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023142185977900065,
|
|
"loss": 4.6955,
|
|
"mean_token_accuracy": 0.21011000275611877,
|
|
"num_tokens": 13363030.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.208169174194336,
|
|
"epoch": 5.963949843260188,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023112680360512218,
|
|
"loss": 4.6638,
|
|
"mean_token_accuracy": 0.21716339141130447,
|
|
"num_tokens": 13372797.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.202608442306518,
|
|
"epoch": 5.967868338557993,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023083182583975942,
|
|
"loss": 4.61,
|
|
"mean_token_accuracy": 0.21355608701705933,
|
|
"num_tokens": 13381425.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.1318872451782225,
|
|
"epoch": 5.971786833855799,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002305369270100855,
|
|
"loss": 4.574,
|
|
"mean_token_accuracy": 0.22273569107055663,
|
|
"num_tokens": 13390246.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.174476099014282,
|
|
"epoch": 5.975705329153605,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002302421076431322,
|
|
"loss": 4.6191,
|
|
"mean_token_accuracy": 0.22174610197544098,
|
|
"num_tokens": 13398851.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.225561475753784,
|
|
"epoch": 5.97962382445141,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022994736826578976,
|
|
"loss": 4.7028,
|
|
"mean_token_accuracy": 0.207380510866642,
|
|
"num_tokens": 13407870.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.241159439086914,
|
|
"epoch": 5.983542319749216,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002296527094048052,
|
|
"loss": 4.6793,
|
|
"mean_token_accuracy": 0.21136046648025514,
|
|
"num_tokens": 13416311.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.218026494979858,
|
|
"epoch": 5.987460815047022,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002293581315867817,
|
|
"loss": 4.7433,
|
|
"mean_token_accuracy": 0.2047022119164467,
|
|
"num_tokens": 13425669.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.195754909515381,
|
|
"epoch": 5.991379310344827,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002290636353381774,
|
|
"loss": 4.6175,
|
|
"mean_token_accuracy": 0.2245378464460373,
|
|
"num_tokens": 13434448.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.255562877655029,
|
|
"epoch": 5.995297805642633,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022876922118530508,
|
|
"loss": 4.6767,
|
|
"mean_token_accuracy": 0.2073802798986435,
|
|
"num_tokens": 13443712.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.130512857437134,
|
|
"epoch": 5.999216300940439,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002284748896543305,
|
|
"loss": 4.4826,
|
|
"mean_token_accuracy": 0.22456343472003937,
|
|
"num_tokens": 13452511.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.109282445907593,
|
|
"epoch": 6.003134796238244,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022818064127127163,
|
|
"loss": 4.4519,
|
|
"mean_token_accuracy": 0.2308727666735649,
|
|
"num_tokens": 13460990.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.190958404541016,
|
|
"epoch": 6.00705329153605,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022788647656199834,
|
|
"loss": 4.5716,
|
|
"mean_token_accuracy": 0.22674012333154678,
|
|
"num_tokens": 13470293.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.249566888809204,
|
|
"epoch": 6.0109717868338555,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00022759239605223052,
|
|
"loss": 4.4869,
|
|
"mean_token_accuracy": 0.2273533284664154,
|
|
"num_tokens": 13479508.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.162486171722412,
|
|
"epoch": 6.014890282131661,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 4.4125,
|
|
"mean_token_accuracy": 0.2333389237523079,
|
|
"num_tokens": 13488163.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.26130518913269,
|
|
"epoch": 6.018808777429467,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022700448973333808,
|
|
"loss": 4.5405,
|
|
"mean_token_accuracy": 0.22693584263324737,
|
|
"num_tokens": 13497404.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.2248235702514645,
|
|
"epoch": 6.0227272727272725,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022671066497489744,
|
|
"loss": 4.4864,
|
|
"mean_token_accuracy": 0.2284241423010826,
|
|
"num_tokens": 13506142.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.1909058570861815,
|
|
"epoch": 6.0266457680250785,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002264169265173282,
|
|
"loss": 4.3938,
|
|
"mean_token_accuracy": 0.23894092738628386,
|
|
"num_tokens": 13514525.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.123483085632325,
|
|
"epoch": 6.030564263322884,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022612327488558866,
|
|
"loss": 4.4106,
|
|
"mean_token_accuracy": 0.2375443994998932,
|
|
"num_tokens": 13523732.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.135363435745239,
|
|
"epoch": 6.0344827586206895,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022582971060448187,
|
|
"loss": 4.4207,
|
|
"mean_token_accuracy": 0.23679692298173904,
|
|
"num_tokens": 13532904.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.182462072372436,
|
|
"epoch": 6.0384012539184955,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022553623419865488,
|
|
"loss": 4.4533,
|
|
"mean_token_accuracy": 0.2263141468167305,
|
|
"num_tokens": 13542174.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.1852696418762205,
|
|
"epoch": 6.0423197492163006,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022524284619259745,
|
|
"loss": 4.4865,
|
|
"mean_token_accuracy": 0.22531844973564147,
|
|
"num_tokens": 13550591.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.150124406814575,
|
|
"epoch": 6.0462382445141065,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002249495471106417,
|
|
"loss": 4.4564,
|
|
"mean_token_accuracy": 0.23042653053998946,
|
|
"num_tokens": 13559088.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.143749475479126,
|
|
"epoch": 6.0501567398119125,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022465633747696052,
|
|
"loss": 4.3709,
|
|
"mean_token_accuracy": 0.23328056037425995,
|
|
"num_tokens": 13568511.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.183082866668701,
|
|
"epoch": 6.054075235109718,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022436321781556712,
|
|
"loss": 4.5375,
|
|
"mean_token_accuracy": 0.2159865140914917,
|
|
"num_tokens": 13577750.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.214763450622558,
|
|
"epoch": 6.0579937304075235,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002240701886503138,
|
|
"loss": 4.4617,
|
|
"mean_token_accuracy": 0.22958761155605317,
|
|
"num_tokens": 13587138.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.215879821777344,
|
|
"epoch": 6.0619122257053295,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022377725050489128,
|
|
"loss": 4.4869,
|
|
"mean_token_accuracy": 0.23262526988983154,
|
|
"num_tokens": 13595459.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.229042911529541,
|
|
"epoch": 6.065830721003135,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022348440390282753,
|
|
"loss": 4.5421,
|
|
"mean_token_accuracy": 0.229093299806118,
|
|
"num_tokens": 13605227.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.249359178543091,
|
|
"epoch": 6.0697492163009406,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022319164936748676,
|
|
"loss": 4.5452,
|
|
"mean_token_accuracy": 0.2224561259150505,
|
|
"num_tokens": 13614139.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.206703615188599,
|
|
"epoch": 6.0736677115987465,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022289898742206904,
|
|
"loss": 4.4401,
|
|
"mean_token_accuracy": 0.23292553573846816,
|
|
"num_tokens": 13623420.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.19505205154419,
|
|
"epoch": 6.077586206896552,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002226064185896085,
|
|
"loss": 4.4918,
|
|
"mean_token_accuracy": 0.23566789776086808,
|
|
"num_tokens": 13632655.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.190142488479614,
|
|
"epoch": 6.081504702194358,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022231394339297322,
|
|
"loss": 4.4336,
|
|
"mean_token_accuracy": 0.23156808465719222,
|
|
"num_tokens": 13641782.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.1697358131408695,
|
|
"epoch": 6.085423197492163,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022202156235486358,
|
|
"loss": 4.4271,
|
|
"mean_token_accuracy": 0.2366559088230133,
|
|
"num_tokens": 13650023.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.146405172348023,
|
|
"epoch": 6.089341692789969,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022172927599781224,
|
|
"loss": 4.4445,
|
|
"mean_token_accuracy": 0.23115811944007875,
|
|
"num_tokens": 13659025.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.224780797958374,
|
|
"epoch": 6.093260188087775,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022143708484418206,
|
|
"loss": 4.4919,
|
|
"mean_token_accuracy": 0.2270088642835617,
|
|
"num_tokens": 13668133.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.194550609588623,
|
|
"epoch": 6.09717868338558,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022114498941616607,
|
|
"loss": 4.4819,
|
|
"mean_token_accuracy": 0.22373466640710832,
|
|
"num_tokens": 13677661.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.1507758617401125,
|
|
"epoch": 6.101097178683386,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022085299023578614,
|
|
"loss": 4.4707,
|
|
"mean_token_accuracy": 0.22817153632640838,
|
|
"num_tokens": 13686117.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.160686111450195,
|
|
"epoch": 6.105015673981192,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00022056108782489225,
|
|
"loss": 4.383,
|
|
"mean_token_accuracy": 0.2362288549542427,
|
|
"num_tokens": 13694792.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.1920067310333256,
|
|
"epoch": 6.108934169278997,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022026928270516122,
|
|
"loss": 4.5063,
|
|
"mean_token_accuracy": 0.22947430312633516,
|
|
"num_tokens": 13704038.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.230186128616333,
|
|
"epoch": 6.112852664576803,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021997757539809622,
|
|
"loss": 4.4844,
|
|
"mean_token_accuracy": 0.22281495481729507,
|
|
"num_tokens": 13713266.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.14278359413147,
|
|
"epoch": 6.116771159874608,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021968596642502546,
|
|
"loss": 4.4149,
|
|
"mean_token_accuracy": 0.24159783273935317,
|
|
"num_tokens": 13721847.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.110155963897705,
|
|
"epoch": 6.120689655172414,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00021939445630710138,
|
|
"loss": 4.4338,
|
|
"mean_token_accuracy": 0.234576815366745,
|
|
"num_tokens": 13730925.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.162999629974365,
|
|
"epoch": 6.12460815047022,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021910304556529986,
|
|
"loss": 4.5451,
|
|
"mean_token_accuracy": 0.2260875716805458,
|
|
"num_tokens": 13739814.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.223678302764893,
|
|
"epoch": 6.128526645768025,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021881173472041926,
|
|
"loss": 4.4161,
|
|
"mean_token_accuracy": 0.24100612848997116,
|
|
"num_tokens": 13748859.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.184223079681397,
|
|
"epoch": 6.132445141065831,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021852052429307924,
|
|
"loss": 4.4952,
|
|
"mean_token_accuracy": 0.23219623416662216,
|
|
"num_tokens": 13757249.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.167526578903198,
|
|
"epoch": 6.136363636363637,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00021822941480371995,
|
|
"loss": 4.4842,
|
|
"mean_token_accuracy": 0.22832140773534776,
|
|
"num_tokens": 13765838.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.186400127410889,
|
|
"epoch": 6.140282131661442,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00021793840677260135,
|
|
"loss": 4.4239,
|
|
"mean_token_accuracy": 0.2421524554491043,
|
|
"num_tokens": 13775465.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.271755790710449,
|
|
"epoch": 6.144200626959248,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021764750071980195,
|
|
"loss": 4.5551,
|
|
"mean_token_accuracy": 0.21566301882266997,
|
|
"num_tokens": 13785373.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.202267217636108,
|
|
"epoch": 6.148119122257054,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021735669716521806,
|
|
"loss": 4.4766,
|
|
"mean_token_accuracy": 0.23547795861959459,
|
|
"num_tokens": 13794812.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.152792692184448,
|
|
"epoch": 6.152037617554859,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00021706599662856265,
|
|
"loss": 4.443,
|
|
"mean_token_accuracy": 0.2279117450118065,
|
|
"num_tokens": 13803681.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.181588506698608,
|
|
"epoch": 6.155956112852665,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021677539962936494,
|
|
"loss": 4.5331,
|
|
"mean_token_accuracy": 0.21740331649780273,
|
|
"num_tokens": 13812772.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.193535041809082,
|
|
"epoch": 6.15987460815047,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00021648490668696875,
|
|
"loss": 4.4734,
|
|
"mean_token_accuracy": 0.2270330384373665,
|
|
"num_tokens": 13821280.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.152463102340699,
|
|
"epoch": 6.163793103448276,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021619451832053199,
|
|
"loss": 4.4346,
|
|
"mean_token_accuracy": 0.23219755589962005,
|
|
"num_tokens": 13829717.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.2274477005004885,
|
|
"epoch": 6.167711598746082,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000215904235049026,
|
|
"loss": 4.5469,
|
|
"mean_token_accuracy": 0.22532479017972945,
|
|
"num_tokens": 13838723.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.187258863449097,
|
|
"epoch": 6.171630094043887,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002156140573912338,
|
|
"loss": 4.522,
|
|
"mean_token_accuracy": 0.22969917356967925,
|
|
"num_tokens": 13847081.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.20615029335022,
|
|
"epoch": 6.175548589341693,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021532398586574997,
|
|
"loss": 4.4552,
|
|
"mean_token_accuracy": 0.2343477800488472,
|
|
"num_tokens": 13855142.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.252102708816528,
|
|
"epoch": 6.179467084639499,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021503402099097959,
|
|
"loss": 4.5399,
|
|
"mean_token_accuracy": 0.22619110345840454,
|
|
"num_tokens": 13863586.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.194668340682983,
|
|
"epoch": 6.183385579937304,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00021474416328513664,
|
|
"loss": 4.5362,
|
|
"mean_token_accuracy": 0.23049330711364746,
|
|
"num_tokens": 13872286.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.175752210617065,
|
|
"epoch": 6.18730407523511,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002144544132662438,
|
|
"loss": 4.4881,
|
|
"mean_token_accuracy": 0.22709787636995316,
|
|
"num_tokens": 13880866.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.144612884521484,
|
|
"epoch": 6.191222570532915,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002141647714521316,
|
|
"loss": 4.4365,
|
|
"mean_token_accuracy": 0.23182642459869385,
|
|
"num_tokens": 13889015.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.201743459701538,
|
|
"epoch": 6.195141065830721,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021387523836043676,
|
|
"loss": 4.5168,
|
|
"mean_token_accuracy": 0.22543469071388245,
|
|
"num_tokens": 13898457.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.205221843719483,
|
|
"epoch": 6.199059561128527,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 4.4577,
|
|
"mean_token_accuracy": 0.23470936864614486,
|
|
"num_tokens": 13906722.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.1669141292572025,
|
|
"epoch": 6.202978056426332,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002132965004138742,
|
|
"loss": 4.4151,
|
|
"mean_token_accuracy": 0.24419744163751603,
|
|
"num_tokens": 13914714.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.1972206115722654,
|
|
"epoch": 6.206896551724138,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002130072965933052,
|
|
"loss": 4.4857,
|
|
"mean_token_accuracy": 0.22751836776733397,
|
|
"num_tokens": 13923378.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.168292140960693,
|
|
"epoch": 6.210815047021944,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021271820356374876,
|
|
"loss": 4.4885,
|
|
"mean_token_accuracy": 0.23144884705543517,
|
|
"num_tokens": 13931947.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.15594220161438,
|
|
"epoch": 6.214733542319749,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021242922184186108,
|
|
"loss": 4.4094,
|
|
"mean_token_accuracy": 0.22749640345573424,
|
|
"num_tokens": 13940949.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.179297971725464,
|
|
"epoch": 6.218652037617555,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002121403519440993,
|
|
"loss": 4.5177,
|
|
"mean_token_accuracy": 0.22423951774835588,
|
|
"num_tokens": 13949944.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.201075077056885,
|
|
"epoch": 6.222570532915361,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021185159438672085,
|
|
"loss": 4.4679,
|
|
"mean_token_accuracy": 0.2243833675980568,
|
|
"num_tokens": 13958938.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.162245607376098,
|
|
"epoch": 6.226489028213166,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021156294968578232,
|
|
"loss": 4.4206,
|
|
"mean_token_accuracy": 0.2424091100692749,
|
|
"num_tokens": 13967875.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.1394754409790036,
|
|
"epoch": 6.230407523510972,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021127441835713838,
|
|
"loss": 4.409,
|
|
"mean_token_accuracy": 0.2328205242753029,
|
|
"num_tokens": 13975972.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.141781234741211,
|
|
"epoch": 6.234326018808777,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002109860009164416,
|
|
"loss": 4.3948,
|
|
"mean_token_accuracy": 0.23651588559150696,
|
|
"num_tokens": 13985081.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.146054029464722,
|
|
"epoch": 6.238244514106583,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002106976978791405,
|
|
"loss": 4.5215,
|
|
"mean_token_accuracy": 0.2249189630150795,
|
|
"num_tokens": 13992426.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.1393210887908936,
|
|
"epoch": 6.242163009404389,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00021040950976047935,
|
|
"loss": 4.4371,
|
|
"mean_token_accuracy": 0.23310969173908233,
|
|
"num_tokens": 14001328.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.1402599811553955,
|
|
"epoch": 6.246081504702194,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021012143707549726,
|
|
"loss": 4.4023,
|
|
"mean_token_accuracy": 0.23658570200204848,
|
|
"num_tokens": 14010340.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.216256332397461,
|
|
"epoch": 6.25,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020983348033902662,
|
|
"loss": 4.5841,
|
|
"mean_token_accuracy": 0.22180630564689635,
|
|
"num_tokens": 14019213.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.193468618392944,
|
|
"epoch": 6.253918495297806,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020954564006569294,
|
|
"loss": 4.5272,
|
|
"mean_token_accuracy": 0.22164179831743241,
|
|
"num_tokens": 14028501.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.125868511199951,
|
|
"epoch": 6.257836990595611,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020925791676991318,
|
|
"loss": 4.4097,
|
|
"mean_token_accuracy": 0.2290080264210701,
|
|
"num_tokens": 14037969.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.205960893630982,
|
|
"epoch": 6.261755485893417,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00020897031096589586,
|
|
"loss": 4.455,
|
|
"mean_token_accuracy": 0.2404683753848076,
|
|
"num_tokens": 14046987.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.187693309783936,
|
|
"epoch": 6.265673981191223,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020868282316763902,
|
|
"loss": 4.491,
|
|
"mean_token_accuracy": 0.22226281613111495,
|
|
"num_tokens": 14055894.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.1222827434539795,
|
|
"epoch": 6.269592476489028,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020839545388893,
|
|
"loss": 4.4676,
|
|
"mean_token_accuracy": 0.2310195505619049,
|
|
"num_tokens": 14065119.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 6.269592476489028,
|
|
"eval_entropy": 5.143553155337194,
|
|
"eval_loss": 5.674208164215088,
|
|
"eval_mean_token_accuracy": 0.170985809390166,
|
|
"eval_num_tokens": 14065119.0,
|
|
"eval_runtime": 2.8478,
|
|
"eval_samples_per_second": 1447.431,
|
|
"eval_steps_per_second": 181.192,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.2170250415802,
|
|
"epoch": 6.273510971786834,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020810820364334438,
|
|
"loss": 4.4381,
|
|
"mean_token_accuracy": 0.23909281194210052,
|
|
"num_tokens": 14073320.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.177502775192261,
|
|
"epoch": 6.277429467084639,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020782107294424484,
|
|
"loss": 4.4662,
|
|
"mean_token_accuracy": 0.22817270010709761,
|
|
"num_tokens": 14081730.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.107434034347534,
|
|
"epoch": 6.281347962382445,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002075340623047804,
|
|
"loss": 4.4477,
|
|
"mean_token_accuracy": 0.2312992349267006,
|
|
"num_tokens": 14090738.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.2022912979125975,
|
|
"epoch": 6.285266457680251,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020724717223788574,
|
|
"loss": 4.5154,
|
|
"mean_token_accuracy": 0.2252133533358574,
|
|
"num_tokens": 14099412.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.172847938537598,
|
|
"epoch": 6.289184952978056,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00020696040325627991,
|
|
"loss": 4.5137,
|
|
"mean_token_accuracy": 0.22777172327041625,
|
|
"num_tokens": 14108334.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.159052753448487,
|
|
"epoch": 6.293103448275862,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00020667375587246563,
|
|
"loss": 4.523,
|
|
"mean_token_accuracy": 0.22679134756326674,
|
|
"num_tokens": 14116853.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.242621040344238,
|
|
"epoch": 6.297021943573668,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020638723059872806,
|
|
"loss": 4.5983,
|
|
"mean_token_accuracy": 0.2171285107731819,
|
|
"num_tokens": 14126735.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.206567430496216,
|
|
"epoch": 6.300940438871473,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020610082794713454,
|
|
"loss": 4.5607,
|
|
"mean_token_accuracy": 0.23040833175182343,
|
|
"num_tokens": 14136749.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.183790016174316,
|
|
"epoch": 6.304858934169279,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002058145484295329,
|
|
"loss": 4.5163,
|
|
"mean_token_accuracy": 0.22837133407592775,
|
|
"num_tokens": 14145204.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.270503997802734,
|
|
"epoch": 6.308777429467085,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020552839255755107,
|
|
"loss": 4.5946,
|
|
"mean_token_accuracy": 0.22222377359867096,
|
|
"num_tokens": 14154347.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.202564430236817,
|
|
"epoch": 6.31269592476489,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020524236084259613,
|
|
"loss": 4.4876,
|
|
"mean_token_accuracy": 0.23444598615169526,
|
|
"num_tokens": 14163039.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.1180419921875,
|
|
"epoch": 6.316614420062696,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020495645379585303,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.2414604365825653,
|
|
"num_tokens": 14172228.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.19399938583374,
|
|
"epoch": 6.320532915360501,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00020467067192828404,
|
|
"loss": 4.5682,
|
|
"mean_token_accuracy": 0.22553053498268127,
|
|
"num_tokens": 14181514.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.218965482711792,
|
|
"epoch": 6.324451410658307,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020438501575062752,
|
|
"loss": 4.5686,
|
|
"mean_token_accuracy": 0.2185903251171112,
|
|
"num_tokens": 14190654.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.175355577468872,
|
|
"epoch": 6.328369905956113,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020409948577339765,
|
|
"loss": 4.4105,
|
|
"mean_token_accuracy": 0.23519833981990815,
|
|
"num_tokens": 14199177.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.1564277648925785,
|
|
"epoch": 6.332288401253918,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020381408250688267,
|
|
"loss": 4.4609,
|
|
"mean_token_accuracy": 0.23914071619510652,
|
|
"num_tokens": 14207643.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.1970601081848145,
|
|
"epoch": 6.336206896551724,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002035288064611444,
|
|
"loss": 4.5314,
|
|
"mean_token_accuracy": 0.22607545852661132,
|
|
"num_tokens": 14216884.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.175978326797486,
|
|
"epoch": 6.34012539184953,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002032436581460177,
|
|
"loss": 4.506,
|
|
"mean_token_accuracy": 0.22346396148204803,
|
|
"num_tokens": 14225324.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.206655216217041,
|
|
"epoch": 6.344043887147335,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020295863807110853,
|
|
"loss": 4.4474,
|
|
"mean_token_accuracy": 0.23401571065187454,
|
|
"num_tokens": 14233732.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.286617231369019,
|
|
"epoch": 6.347962382445141,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00020267374674579408,
|
|
"loss": 4.5256,
|
|
"mean_token_accuracy": 0.22527532130479813,
|
|
"num_tokens": 14242080.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.14070315361023,
|
|
"epoch": 6.351880877742946,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020238898467922129,
|
|
"loss": 4.5227,
|
|
"mean_token_accuracy": 0.22860418558120726,
|
|
"num_tokens": 14251140.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.1925571918487545,
|
|
"epoch": 6.355799373040752,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020210435238030626,
|
|
"loss": 4.5756,
|
|
"mean_token_accuracy": 0.2187289535999298,
|
|
"num_tokens": 14259900.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.189757394790649,
|
|
"epoch": 6.359717868338558,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00020181985035773304,
|
|
"loss": 4.5036,
|
|
"mean_token_accuracy": 0.2279673457145691,
|
|
"num_tokens": 14268414.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.197866344451905,
|
|
"epoch": 6.363636363636363,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020153547911995272,
|
|
"loss": 4.5288,
|
|
"mean_token_accuracy": 0.22256577163934707,
|
|
"num_tokens": 14277339.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.1864417552947994,
|
|
"epoch": 6.367554858934169,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020125123917518308,
|
|
"loss": 4.4623,
|
|
"mean_token_accuracy": 0.22905606627464295,
|
|
"num_tokens": 14285710.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.187597799301147,
|
|
"epoch": 6.371473354231975,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020096713103140673,
|
|
"loss": 4.4956,
|
|
"mean_token_accuracy": 0.22795476466417314,
|
|
"num_tokens": 14294625.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.117865705490113,
|
|
"epoch": 6.37539184952978,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020068315519637103,
|
|
"loss": 4.3409,
|
|
"mean_token_accuracy": 0.23920684009790422,
|
|
"num_tokens": 14303030.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.108934020996093,
|
|
"epoch": 6.379310344827586,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000200399312177587,
|
|
"loss": 4.4524,
|
|
"mean_token_accuracy": 0.23085338324308396,
|
|
"num_tokens": 14311826.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.219910955429077,
|
|
"epoch": 6.383228840125392,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 4.5275,
|
|
"mean_token_accuracy": 0.23371956944465638,
|
|
"num_tokens": 14319600.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.131806325912476,
|
|
"epoch": 6.387147335423197,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001998320266176294,
|
|
"loss": 4.4417,
|
|
"mean_token_accuracy": 0.23326896578073503,
|
|
"num_tokens": 14328768.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.17605619430542,
|
|
"epoch": 6.391065830721003,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000199548585090287,
|
|
"loss": 4.5252,
|
|
"mean_token_accuracy": 0.22236035019159317,
|
|
"num_tokens": 14338236.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.241018438339234,
|
|
"epoch": 6.394984326018808,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000199265278406857,
|
|
"loss": 4.5833,
|
|
"mean_token_accuracy": 0.22161969393491746,
|
|
"num_tokens": 14347190.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.212325382232666,
|
|
"epoch": 6.398902821316614,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00019898210707365428,
|
|
"loss": 4.5205,
|
|
"mean_token_accuracy": 0.2274225249886513,
|
|
"num_tokens": 14355847.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.142733860015869,
|
|
"epoch": 6.40282131661442,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019869907159675187,
|
|
"loss": 4.445,
|
|
"mean_token_accuracy": 0.2253383755683899,
|
|
"num_tokens": 14364280.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.202603197097778,
|
|
"epoch": 6.406739811912225,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019841617248198017,
|
|
"loss": 4.4949,
|
|
"mean_token_accuracy": 0.22453133165836334,
|
|
"num_tokens": 14372857.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.174392461776733,
|
|
"epoch": 6.410658307210031,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001981334102349258,
|
|
"loss": 4.4485,
|
|
"mean_token_accuracy": 0.22787954807281494,
|
|
"num_tokens": 14381157.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.21751914024353,
|
|
"epoch": 6.414576802507837,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00019785078536093065,
|
|
"loss": 4.5365,
|
|
"mean_token_accuracy": 0.21949482709169388,
|
|
"num_tokens": 14390404.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.190119075775146,
|
|
"epoch": 6.418495297805642,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019756829836509117,
|
|
"loss": 4.5284,
|
|
"mean_token_accuracy": 0.2255229949951172,
|
|
"num_tokens": 14399426.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.167094087600708,
|
|
"epoch": 6.422413793103448,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001972859497522575,
|
|
"loss": 4.4194,
|
|
"mean_token_accuracy": 0.22723287343978882,
|
|
"num_tokens": 14408233.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.190805292129516,
|
|
"epoch": 6.4263322884012535,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001970037400270323,
|
|
"loss": 4.4956,
|
|
"mean_token_accuracy": 0.22578304708004,
|
|
"num_tokens": 14416828.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.225488901138306,
|
|
"epoch": 6.4302507836990594,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001967216696937702,
|
|
"loss": 4.5151,
|
|
"mean_token_accuracy": 0.23103769570589067,
|
|
"num_tokens": 14425590.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.250370216369629,
|
|
"epoch": 6.434169278996865,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019643973925657666,
|
|
"loss": 4.6459,
|
|
"mean_token_accuracy": 0.21187071800231932,
|
|
"num_tokens": 14434646.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.23318395614624,
|
|
"epoch": 6.4380877742946705,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019615794921930697,
|
|
"loss": 4.5382,
|
|
"mean_token_accuracy": 0.2229375883936882,
|
|
"num_tokens": 14443862.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.155437850952149,
|
|
"epoch": 6.4420062695924765,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00019587630008556563,
|
|
"loss": 4.4393,
|
|
"mean_token_accuracy": 0.2312757134437561,
|
|
"num_tokens": 14452813.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.181336545944214,
|
|
"epoch": 6.445924764890282,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019559479235870537,
|
|
"loss": 4.5344,
|
|
"mean_token_accuracy": 0.22492219060659407,
|
|
"num_tokens": 14461912.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.229310083389282,
|
|
"epoch": 6.4498432601880875,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001953134265418261,
|
|
"loss": 4.4665,
|
|
"mean_token_accuracy": 0.22883182615041733,
|
|
"num_tokens": 14470460.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.204417324066162,
|
|
"epoch": 6.4537617554858935,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001950322031377742,
|
|
"loss": 4.4841,
|
|
"mean_token_accuracy": 0.2294685050845146,
|
|
"num_tokens": 14479258.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.206909990310669,
|
|
"epoch": 6.4576802507836994,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019475112264914142,
|
|
"loss": 4.5034,
|
|
"mean_token_accuracy": 0.2330746904015541,
|
|
"num_tokens": 14488209.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.089283037185669,
|
|
"epoch": 6.4615987460815045,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019447018557826424,
|
|
"loss": 4.4543,
|
|
"mean_token_accuracy": 0.2263825058937073,
|
|
"num_tokens": 14497234.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.082912158966065,
|
|
"epoch": 6.4655172413793105,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001941893924272227,
|
|
"loss": 4.3756,
|
|
"mean_token_accuracy": 0.23621445894241333,
|
|
"num_tokens": 14506374.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.199085712432861,
|
|
"epoch": 6.469435736677116,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019390874369783974,
|
|
"loss": 4.5276,
|
|
"mean_token_accuracy": 0.22609951049089433,
|
|
"num_tokens": 14514796.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.2290452003479,
|
|
"epoch": 6.4733542319749215,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019362823989168006,
|
|
"loss": 4.4927,
|
|
"mean_token_accuracy": 0.2290377914905548,
|
|
"num_tokens": 14523771.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.1413627624511715,
|
|
"epoch": 6.4772727272727275,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019334788151004962,
|
|
"loss": 4.4037,
|
|
"mean_token_accuracy": 0.23957442343235016,
|
|
"num_tokens": 14532130.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.066401481628418,
|
|
"epoch": 6.481191222570533,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019306766905399415,
|
|
"loss": 4.4158,
|
|
"mean_token_accuracy": 0.23049562573432922,
|
|
"num_tokens": 14540278.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.1511236190795895,
|
|
"epoch": 6.485109717868339,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019278760302429872,
|
|
"loss": 4.5229,
|
|
"mean_token_accuracy": 0.21745371967554092,
|
|
"num_tokens": 14549519.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.176907920837403,
|
|
"epoch": 6.4890282131661445,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019250768392148682,
|
|
"loss": 4.5558,
|
|
"mean_token_accuracy": 0.2274658888578415,
|
|
"num_tokens": 14558576.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.166570854187012,
|
|
"epoch": 6.49294670846395,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019222791224581925,
|
|
"loss": 4.4606,
|
|
"mean_token_accuracy": 0.2293439030647278,
|
|
"num_tokens": 14567544.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.146781587600708,
|
|
"epoch": 6.496865203761756,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019194828849729336,
|
|
"loss": 4.4558,
|
|
"mean_token_accuracy": 0.22305788546800615,
|
|
"num_tokens": 14576662.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.111741399765014,
|
|
"epoch": 6.500783699059561,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019166881317564217,
|
|
"loss": 4.4388,
|
|
"mean_token_accuracy": 0.228020741045475,
|
|
"num_tokens": 14585163.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.107633209228515,
|
|
"epoch": 6.504702194357367,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00019138948678033336,
|
|
"loss": 4.3857,
|
|
"mean_token_accuracy": 0.23553906977176667,
|
|
"num_tokens": 14593213.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.156840419769287,
|
|
"epoch": 6.508620689655173,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019111030981056832,
|
|
"loss": 4.4827,
|
|
"mean_token_accuracy": 0.23495438396930696,
|
|
"num_tokens": 14601479.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.215953731536866,
|
|
"epoch": 6.512539184952978,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001908312827652819,
|
|
"loss": 4.4866,
|
|
"mean_token_accuracy": 0.22685063183307647,
|
|
"num_tokens": 14610418.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.171939992904663,
|
|
"epoch": 6.516457680250784,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019055240614314047,
|
|
"loss": 4.498,
|
|
"mean_token_accuracy": 0.23028390258550643,
|
|
"num_tokens": 14618387.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.158885955810547,
|
|
"epoch": 6.52037617554859,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019027368044254177,
|
|
"loss": 4.4383,
|
|
"mean_token_accuracy": 0.23545875102281572,
|
|
"num_tokens": 14627003.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.191231727600098,
|
|
"epoch": 6.524294670846395,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001899951061616138,
|
|
"loss": 4.5117,
|
|
"mean_token_accuracy": 0.2251594290137291,
|
|
"num_tokens": 14635825.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.1522469997406,
|
|
"epoch": 6.528213166144201,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018971668379821405,
|
|
"loss": 4.4358,
|
|
"mean_token_accuracy": 0.2358198881149292,
|
|
"num_tokens": 14644830.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.135208415985107,
|
|
"epoch": 6.532131661442007,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001894384138499284,
|
|
"loss": 4.492,
|
|
"mean_token_accuracy": 0.2240090176463127,
|
|
"num_tokens": 14653633.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.115923738479614,
|
|
"epoch": 6.536050156739812,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00018916029681407032,
|
|
"loss": 4.3418,
|
|
"mean_token_accuracy": 0.24994510859251023,
|
|
"num_tokens": 14662212.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.175853538513183,
|
|
"epoch": 6.539968652037618,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001888823331876801,
|
|
"loss": 4.4871,
|
|
"mean_token_accuracy": 0.2311457008123398,
|
|
"num_tokens": 14670942.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.221028995513916,
|
|
"epoch": 6.543887147335424,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018860452346752382,
|
|
"loss": 4.4396,
|
|
"mean_token_accuracy": 0.22948154509067537,
|
|
"num_tokens": 14679167.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.125361824035645,
|
|
"epoch": 6.547805642633229,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00018832686815009242,
|
|
"loss": 4.481,
|
|
"mean_token_accuracy": 0.22732346057891845,
|
|
"num_tokens": 14688188.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.1897797107696535,
|
|
"epoch": 6.551724137931035,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018804936773160108,
|
|
"loss": 4.5584,
|
|
"mean_token_accuracy": 0.2209342435002327,
|
|
"num_tokens": 14696958.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.13513822555542,
|
|
"epoch": 6.55564263322884,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000187772022707988,
|
|
"loss": 4.4915,
|
|
"mean_token_accuracy": 0.22181811332702636,
|
|
"num_tokens": 14705524.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.151407623291016,
|
|
"epoch": 6.559561128526646,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018749483357491365,
|
|
"loss": 4.5311,
|
|
"mean_token_accuracy": 0.22398922145366668,
|
|
"num_tokens": 14714500.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.1847740650177006,
|
|
"epoch": 6.563479623824452,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018721780082775997,
|
|
"loss": 4.4966,
|
|
"mean_token_accuracy": 0.2229219689965248,
|
|
"num_tokens": 14722816.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.182420825958252,
|
|
"epoch": 6.567398119122257,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 4.502,
|
|
"mean_token_accuracy": 0.2317389503121376,
|
|
"num_tokens": 14731453.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.225334787368775,
|
|
"epoch": 6.571316614420063,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018666420647134419,
|
|
"loss": 4.5021,
|
|
"mean_token_accuracy": 0.2366289898753166,
|
|
"num_tokens": 14739924.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.071468544006348,
|
|
"epoch": 6.575235109717868,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001863876458514448,
|
|
"loss": 4.3726,
|
|
"mean_token_accuracy": 0.24622797220945358,
|
|
"num_tokens": 14748356.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.236114645004273,
|
|
"epoch": 6.579153605015674,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00018611124359619024,
|
|
"loss": 4.5684,
|
|
"mean_token_accuracy": 0.22372687309980394,
|
|
"num_tokens": 14758846.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.092509508132935,
|
|
"epoch": 6.58307210031348,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018583500019955595,
|
|
"loss": 4.3637,
|
|
"mean_token_accuracy": 0.24048162251710892,
|
|
"num_tokens": 14767030.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.169730710983276,
|
|
"epoch": 6.586990595611285,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018555891615523374,
|
|
"loss": 4.4554,
|
|
"mean_token_accuracy": 0.2304292783141136,
|
|
"num_tokens": 14776774.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.138438129425049,
|
|
"epoch": 6.590909090909091,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018528299195663046,
|
|
"loss": 4.4396,
|
|
"mean_token_accuracy": 0.23334923535585403,
|
|
"num_tokens": 14785662.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.216218423843384,
|
|
"epoch": 6.594827586206897,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001850072280968676,
|
|
"loss": 4.5863,
|
|
"mean_token_accuracy": 0.2125610515475273,
|
|
"num_tokens": 14795070.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.196914100646973,
|
|
"epoch": 6.598746081504702,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018473162506877977,
|
|
"loss": 4.5103,
|
|
"mean_token_accuracy": 0.2260113850235939,
|
|
"num_tokens": 14805452.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.174579811096192,
|
|
"epoch": 6.602664576802508,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00018445618336491415,
|
|
"loss": 4.4706,
|
|
"mean_token_accuracy": 0.23543512672185898,
|
|
"num_tokens": 14815632.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.1355678081512455,
|
|
"epoch": 6.606583072100314,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00018418090347752997,
|
|
"loss": 4.4064,
|
|
"mean_token_accuracy": 0.23239424973726272,
|
|
"num_tokens": 14824228.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.177458620071411,
|
|
"epoch": 6.610501567398119,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001839057858985968,
|
|
"loss": 4.5429,
|
|
"mean_token_accuracy": 0.22129137814044952,
|
|
"num_tokens": 14833609.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.178340482711792,
|
|
"epoch": 6.614420062695925,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00018363083111979452,
|
|
"loss": 4.4848,
|
|
"mean_token_accuracy": 0.2314935326576233,
|
|
"num_tokens": 14841967.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.171598958969116,
|
|
"epoch": 6.618338557993731,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018335603963251184,
|
|
"loss": 4.5301,
|
|
"mean_token_accuracy": 0.21894141882658005,
|
|
"num_tokens": 14851472.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.243009328842163,
|
|
"epoch": 6.622257053291536,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018308141192784573,
|
|
"loss": 4.4972,
|
|
"mean_token_accuracy": 0.22838148027658461,
|
|
"num_tokens": 14859672.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.194585466384888,
|
|
"epoch": 6.626175548589342,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00018280694849660034,
|
|
"loss": 4.4143,
|
|
"mean_token_accuracy": 0.23910920172929764,
|
|
"num_tokens": 14867919.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.129553651809692,
|
|
"epoch": 6.630094043887147,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018253264982928636,
|
|
"loss": 4.4211,
|
|
"mean_token_accuracy": 0.23089367896318436,
|
|
"num_tokens": 14875670.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.159038829803467,
|
|
"epoch": 6.634012539184953,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00018225851641612007,
|
|
"loss": 4.453,
|
|
"mean_token_accuracy": 0.2313536986708641,
|
|
"num_tokens": 14883835.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.115158128738403,
|
|
"epoch": 6.637931034482759,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001819845487470223,
|
|
"loss": 4.4018,
|
|
"mean_token_accuracy": 0.23669884353876114,
|
|
"num_tokens": 14892296.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.107250642776489,
|
|
"epoch": 6.641849529780564,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018171074731161758,
|
|
"loss": 4.4552,
|
|
"mean_token_accuracy": 0.24038095474243165,
|
|
"num_tokens": 14901334.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.1280131340026855,
|
|
"epoch": 6.64576802507837,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018143711259923363,
|
|
"loss": 4.4671,
|
|
"mean_token_accuracy": 0.22917289286851883,
|
|
"num_tokens": 14910327.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.225454616546631,
|
|
"epoch": 6.649686520376176,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001811636450989,
|
|
"loss": 4.5781,
|
|
"mean_token_accuracy": 0.2231076180934906,
|
|
"num_tokens": 14919513.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.099105882644653,
|
|
"epoch": 6.653605015673981,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001808903452993474,
|
|
"loss": 4.3452,
|
|
"mean_token_accuracy": 0.23841775804758072,
|
|
"num_tokens": 14928127.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.157456874847412,
|
|
"epoch": 6.657523510971787,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001806172136890068,
|
|
"loss": 4.4439,
|
|
"mean_token_accuracy": 0.2313700169324875,
|
|
"num_tokens": 14936771.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.173252534866333,
|
|
"epoch": 6.661442006269592,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001803442507560089,
|
|
"loss": 4.4335,
|
|
"mean_token_accuracy": 0.2290744349360466,
|
|
"num_tokens": 14945423.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 6.661442006269592,
|
|
"eval_entropy": 5.074613044428271,
|
|
"eval_loss": 5.661206245422363,
|
|
"eval_mean_token_accuracy": 0.17226281056050644,
|
|
"eval_num_tokens": 14945423.0,
|
|
"eval_runtime": 2.8395,
|
|
"eval_samples_per_second": 1451.674,
|
|
"eval_steps_per_second": 181.723,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.114336824417114,
|
|
"epoch": 6.665360501567398,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001800714569881825,
|
|
"loss": 4.4576,
|
|
"mean_token_accuracy": 0.23190360218286515,
|
|
"num_tokens": 14953859.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.126488161087036,
|
|
"epoch": 6.669278996865204,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017979883287305454,
|
|
"loss": 4.4508,
|
|
"mean_token_accuracy": 0.23488958925008774,
|
|
"num_tokens": 14962598.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.186865377426147,
|
|
"epoch": 6.673197492163009,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017952637889784824,
|
|
"loss": 4.4513,
|
|
"mean_token_accuracy": 0.23311954587697983,
|
|
"num_tokens": 14971606.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.129375696182251,
|
|
"epoch": 6.677115987460815,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00017925409554948329,
|
|
"loss": 4.4732,
|
|
"mean_token_accuracy": 0.23375730812549592,
|
|
"num_tokens": 14979686.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.18281569480896,
|
|
"epoch": 6.681034482758621,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00017898198331457393,
|
|
"loss": 4.5659,
|
|
"mean_token_accuracy": 0.2199096143245697,
|
|
"num_tokens": 14988977.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.167682123184204,
|
|
"epoch": 6.684952978056426,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017871004267942908,
|
|
"loss": 4.5596,
|
|
"mean_token_accuracy": 0.22058653831481934,
|
|
"num_tokens": 14996871.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.198287582397461,
|
|
"epoch": 6.688871473354232,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00017843827413005064,
|
|
"loss": 4.4968,
|
|
"mean_token_accuracy": 0.2268854111433029,
|
|
"num_tokens": 15006022.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.128358268737793,
|
|
"epoch": 6.692789968652038,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001781666781521331,
|
|
"loss": 4.5219,
|
|
"mean_token_accuracy": 0.23203974664211274,
|
|
"num_tokens": 15014827.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.137425708770752,
|
|
"epoch": 6.696708463949843,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00017789525523106243,
|
|
"loss": 4.422,
|
|
"mean_token_accuracy": 0.23400594890117646,
|
|
"num_tokens": 15023157.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.193604516983032,
|
|
"epoch": 6.700626959247649,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017762400585191545,
|
|
"loss": 4.4877,
|
|
"mean_token_accuracy": 0.22220598608255387,
|
|
"num_tokens": 15032539.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.231438875198364,
|
|
"epoch": 6.704545454545455,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017735293049945888,
|
|
"loss": 4.6451,
|
|
"mean_token_accuracy": 0.21557974815368652,
|
|
"num_tokens": 15041527.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.165105247497559,
|
|
"epoch": 6.70846394984326,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017708202965814807,
|
|
"loss": 4.4831,
|
|
"mean_token_accuracy": 0.2259528636932373,
|
|
"num_tokens": 15050794.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.143991374969483,
|
|
"epoch": 6.712382445141066,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00017681130381212707,
|
|
"loss": 4.4713,
|
|
"mean_token_accuracy": 0.2367523729801178,
|
|
"num_tokens": 15059357.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.229987478256225,
|
|
"epoch": 6.716300940438871,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00017654075344522652,
|
|
"loss": 4.5655,
|
|
"mean_token_accuracy": 0.22231560200452805,
|
|
"num_tokens": 15067668.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.201261568069458,
|
|
"epoch": 6.720219435736677,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017627037904096416,
|
|
"loss": 4.4607,
|
|
"mean_token_accuracy": 0.22466526180505753,
|
|
"num_tokens": 15076463.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.0934442520141605,
|
|
"epoch": 6.724137931034483,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017600018108254257,
|
|
"loss": 4.4575,
|
|
"mean_token_accuracy": 0.223613041639328,
|
|
"num_tokens": 15085024.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.120546007156372,
|
|
"epoch": 6.728056426332288,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017573016005284958,
|
|
"loss": 4.4528,
|
|
"mean_token_accuracy": 0.2332012489438057,
|
|
"num_tokens": 15093735.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.175649499893188,
|
|
"epoch": 6.731974921630094,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017546031643445636,
|
|
"loss": 4.4676,
|
|
"mean_token_accuracy": 0.2259786084294319,
|
|
"num_tokens": 15102274.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.193916177749633,
|
|
"epoch": 6.735893416927899,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001751906507096172,
|
|
"loss": 4.4724,
|
|
"mean_token_accuracy": 0.23314101248979568,
|
|
"num_tokens": 15110762.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.168850708007812,
|
|
"epoch": 6.739811912225705,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001749211633602687,
|
|
"loss": 4.4547,
|
|
"mean_token_accuracy": 0.23063663244247437,
|
|
"num_tokens": 15119104.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.142283201217651,
|
|
"epoch": 6.743730407523511,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017465185486802826,
|
|
"loss": 4.4467,
|
|
"mean_token_accuracy": 0.23175692111253737,
|
|
"num_tokens": 15127351.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.231444454193115,
|
|
"epoch": 6.747648902821316,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017438272571419373,
|
|
"loss": 4.5915,
|
|
"mean_token_accuracy": 0.2246356040239334,
|
|
"num_tokens": 15136246.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.1746166229248045,
|
|
"epoch": 6.751567398119122,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 4.5167,
|
|
"mean_token_accuracy": 0.22524366825819014,
|
|
"num_tokens": 15144633.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.165651798248291,
|
|
"epoch": 6.755485893416928,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017384500734533137,
|
|
"loss": 4.4939,
|
|
"mean_token_accuracy": 0.22862782925367356,
|
|
"num_tokens": 15153496.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.146964406967163,
|
|
"epoch": 6.759404388714733,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017357641909129324,
|
|
"loss": 4.4713,
|
|
"mean_token_accuracy": 0.23192338049411773,
|
|
"num_tokens": 15162242.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.207448101043701,
|
|
"epoch": 6.763322884012539,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017330801209763938,
|
|
"loss": 4.5307,
|
|
"mean_token_accuracy": 0.22685355693101883,
|
|
"num_tokens": 15170001.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.166784429550171,
|
|
"epoch": 6.767241379310345,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017303978684405635,
|
|
"loss": 4.4552,
|
|
"mean_token_accuracy": 0.2308524191379547,
|
|
"num_tokens": 15178086.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.180467987060547,
|
|
"epoch": 6.77115987460815,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017277174380990647,
|
|
"loss": 4.5335,
|
|
"mean_token_accuracy": 0.22246409356594085,
|
|
"num_tokens": 15187087.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.204708051681519,
|
|
"epoch": 6.775078369905956,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017250388347422578,
|
|
"loss": 4.4824,
|
|
"mean_token_accuracy": 0.2350981429219246,
|
|
"num_tokens": 15195850.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.189690828323364,
|
|
"epoch": 6.778996865203762,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017223620631572433,
|
|
"loss": 4.5176,
|
|
"mean_token_accuracy": 0.2315893977880478,
|
|
"num_tokens": 15204174.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.191708040237427,
|
|
"epoch": 6.782915360501567,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017196871281278465,
|
|
"loss": 4.3967,
|
|
"mean_token_accuracy": 0.24332556873559952,
|
|
"num_tokens": 15212640.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.207776927947998,
|
|
"epoch": 6.786833855799373,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017170140344346086,
|
|
"loss": 4.4481,
|
|
"mean_token_accuracy": 0.22964880615472794,
|
|
"num_tokens": 15220996.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.167762422561646,
|
|
"epoch": 6.790752351097178,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001714342786854781,
|
|
"loss": 4.492,
|
|
"mean_token_accuracy": 0.23046358972787856,
|
|
"num_tokens": 15229511.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.168343782424927,
|
|
"epoch": 6.794670846394984,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017116733901623182,
|
|
"loss": 4.5534,
|
|
"mean_token_accuracy": 0.22356040477752687,
|
|
"num_tokens": 15238673.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.214832210540772,
|
|
"epoch": 6.79858934169279,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017090058491278637,
|
|
"loss": 4.4954,
|
|
"mean_token_accuracy": 0.22163631319999694,
|
|
"num_tokens": 15246717.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.173599433898926,
|
|
"epoch": 6.802507836990595,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017063401685187457,
|
|
"loss": 4.4776,
|
|
"mean_token_accuracy": 0.22634219974279404,
|
|
"num_tokens": 15255543.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.102309846878052,
|
|
"epoch": 6.806426332288401,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017036763530989688,
|
|
"loss": 4.3854,
|
|
"mean_token_accuracy": 0.23870061337947845,
|
|
"num_tokens": 15264275.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.181909656524658,
|
|
"epoch": 6.810344827586206,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017010144076292012,
|
|
"loss": 4.4912,
|
|
"mean_token_accuracy": 0.23729965686798096,
|
|
"num_tokens": 15272376.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.223774576187134,
|
|
"epoch": 6.814263322884012,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016983543368667722,
|
|
"loss": 4.5359,
|
|
"mean_token_accuracy": 0.22106768786907197,
|
|
"num_tokens": 15282362.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.172113275527954,
|
|
"epoch": 6.818181818181818,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016956961455656607,
|
|
"loss": 4.5138,
|
|
"mean_token_accuracy": 0.22095201909542084,
|
|
"num_tokens": 15290934.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.142864179611206,
|
|
"epoch": 6.822100313479623,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016930398384764836,
|
|
"loss": 4.4183,
|
|
"mean_token_accuracy": 0.24238865971565246,
|
|
"num_tokens": 15299743.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.174276351928711,
|
|
"epoch": 6.826018808777429,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00016903854203464939,
|
|
"loss": 4.4979,
|
|
"mean_token_accuracy": 0.22837163805961608,
|
|
"num_tokens": 15307858.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.129810047149658,
|
|
"epoch": 6.829937304075235,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016877328959195653,
|
|
"loss": 4.469,
|
|
"mean_token_accuracy": 0.22808943539857865,
|
|
"num_tokens": 15316829.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.210107469558716,
|
|
"epoch": 6.83385579937304,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016850822699361913,
|
|
"loss": 4.5891,
|
|
"mean_token_accuracy": 0.21930038630962373,
|
|
"num_tokens": 15325394.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.147527122497559,
|
|
"epoch": 6.837774294670846,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001682433547133469,
|
|
"loss": 4.4668,
|
|
"mean_token_accuracy": 0.22663547992706298,
|
|
"num_tokens": 15333988.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.218999814987183,
|
|
"epoch": 6.841692789968652,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016797867322450948,
|
|
"loss": 4.5314,
|
|
"mean_token_accuracy": 0.22266142070293427,
|
|
"num_tokens": 15343107.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.224789953231811,
|
|
"epoch": 6.8456112852664575,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001677141830001358,
|
|
"loss": 4.4611,
|
|
"mean_token_accuracy": 0.2314899280667305,
|
|
"num_tokens": 15351440.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.120916652679443,
|
|
"epoch": 6.849529780564263,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016744988451291266,
|
|
"loss": 4.4237,
|
|
"mean_token_accuracy": 0.22634343653917313,
|
|
"num_tokens": 15360426.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.1538286209106445,
|
|
"epoch": 6.853448275862069,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016718577823518426,
|
|
"loss": 4.4913,
|
|
"mean_token_accuracy": 0.2270601898431778,
|
|
"num_tokens": 15369252.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.192946147918701,
|
|
"epoch": 6.8573667711598745,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016692186463895136,
|
|
"loss": 4.5001,
|
|
"mean_token_accuracy": 0.22213947772979736,
|
|
"num_tokens": 15377672.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.131526136398316,
|
|
"epoch": 6.86128526645768,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016665814419587055,
|
|
"loss": 4.4075,
|
|
"mean_token_accuracy": 0.24084848165512085,
|
|
"num_tokens": 15385585.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.189429187774659,
|
|
"epoch": 6.8652037617554855,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001663946173772528,
|
|
"loss": 4.4879,
|
|
"mean_token_accuracy": 0.22492904365062713,
|
|
"num_tokens": 15394193.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.214086055755615,
|
|
"epoch": 6.8691222570532915,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016613128465406329,
|
|
"loss": 4.514,
|
|
"mean_token_accuracy": 0.22208772450685502,
|
|
"num_tokens": 15402498.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.0962437152862545,
|
|
"epoch": 6.8730407523510975,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001658681464969204,
|
|
"loss": 4.3881,
|
|
"mean_token_accuracy": 0.2373163402080536,
|
|
"num_tokens": 15410893.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.157757425308228,
|
|
"epoch": 6.8769592476489025,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016560520337609465,
|
|
"loss": 4.5599,
|
|
"mean_token_accuracy": 0.22278459072113038,
|
|
"num_tokens": 15420024.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.066093778610229,
|
|
"epoch": 6.8808777429467085,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016534245576150787,
|
|
"loss": 4.4804,
|
|
"mean_token_accuracy": 0.22778714299201966,
|
|
"num_tokens": 15428474.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.127930450439453,
|
|
"epoch": 6.8847962382445145,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001650799041227329,
|
|
"loss": 4.4097,
|
|
"mean_token_accuracy": 0.23413589000701904,
|
|
"num_tokens": 15437019.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.2130287170410154,
|
|
"epoch": 6.88871473354232,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00016481754892899193,
|
|
"loss": 4.5213,
|
|
"mean_token_accuracy": 0.23412419706583024,
|
|
"num_tokens": 15445838.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.19812912940979,
|
|
"epoch": 6.8926332288401255,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001645553906491563,
|
|
"loss": 4.5269,
|
|
"mean_token_accuracy": 0.23130232244729995,
|
|
"num_tokens": 15454378.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.172721433639526,
|
|
"epoch": 6.896551724137931,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001642934297517452,
|
|
"loss": 4.5016,
|
|
"mean_token_accuracy": 0.22626632004976271,
|
|
"num_tokens": 15462447.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.119561767578125,
|
|
"epoch": 6.900470219435737,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001640316667049254,
|
|
"loss": 4.4848,
|
|
"mean_token_accuracy": 0.22574831992387773,
|
|
"num_tokens": 15471646.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.162883567810058,
|
|
"epoch": 6.9043887147335425,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016377010197650993,
|
|
"loss": 4.4805,
|
|
"mean_token_accuracy": 0.22928102016448976,
|
|
"num_tokens": 15481945.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.19529275894165,
|
|
"epoch": 6.908307210031348,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001635087360339572,
|
|
"loss": 4.4415,
|
|
"mean_token_accuracy": 0.2418562352657318,
|
|
"num_tokens": 15490558.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.183184957504272,
|
|
"epoch": 6.912225705329154,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016324756934437074,
|
|
"loss": 4.4564,
|
|
"mean_token_accuracy": 0.22945396453142167,
|
|
"num_tokens": 15499150.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.146198368072509,
|
|
"epoch": 6.91614420062696,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001629866023744977,
|
|
"loss": 4.4417,
|
|
"mean_token_accuracy": 0.2253652274608612,
|
|
"num_tokens": 15508736.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.231228637695312,
|
|
"epoch": 6.920062695924765,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00016272583559072832,
|
|
"loss": 4.5052,
|
|
"mean_token_accuracy": 0.2370871752500534,
|
|
"num_tokens": 15517445.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.16759033203125,
|
|
"epoch": 6.923981191222571,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00016246526945909505,
|
|
"loss": 4.5143,
|
|
"mean_token_accuracy": 0.23803826719522475,
|
|
"num_tokens": 15526282.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.200987148284912,
|
|
"epoch": 6.927899686520377,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016220490444527198,
|
|
"loss": 4.4941,
|
|
"mean_token_accuracy": 0.2303578481078148,
|
|
"num_tokens": 15534776.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.1729247093200685,
|
|
"epoch": 6.931818181818182,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001619447410145735,
|
|
"loss": 4.4671,
|
|
"mean_token_accuracy": 0.23443028032779695,
|
|
"num_tokens": 15543173.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.121985530853271,
|
|
"epoch": 6.935736677115988,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016168477963195382,
|
|
"loss": 4.4067,
|
|
"mean_token_accuracy": 0.23026833385229112,
|
|
"num_tokens": 15551603.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.142138147354126,
|
|
"epoch": 6.939655172413794,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016142502076200622,
|
|
"loss": 4.4537,
|
|
"mean_token_accuracy": 0.22532009780406953,
|
|
"num_tokens": 15560412.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.059034442901611,
|
|
"epoch": 6.943573667711599,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016116546486896183,
|
|
"loss": 4.4447,
|
|
"mean_token_accuracy": 0.22283483147621155,
|
|
"num_tokens": 15569781.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.158576011657715,
|
|
"epoch": 6.947492163009405,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016090611241668904,
|
|
"loss": 4.4383,
|
|
"mean_token_accuracy": 0.23619789779186248,
|
|
"num_tokens": 15577640.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.188737916946411,
|
|
"epoch": 6.95141065830721,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001606469638686928,
|
|
"loss": 4.5557,
|
|
"mean_token_accuracy": 0.22551365792751313,
|
|
"num_tokens": 15586950.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.162472724914551,
|
|
"epoch": 6.955329153605016,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001603880196881137,
|
|
"loss": 4.5081,
|
|
"mean_token_accuracy": 0.2299259051680565,
|
|
"num_tokens": 15595554.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.1813316822052,
|
|
"epoch": 6.959247648902822,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001601292803377269,
|
|
"loss": 4.4667,
|
|
"mean_token_accuracy": 0.22852424830198287,
|
|
"num_tokens": 15603998.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.256630897521973,
|
|
"epoch": 6.963166144200627,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001598707462799414,
|
|
"loss": 4.5452,
|
|
"mean_token_accuracy": 0.225949726998806,
|
|
"num_tokens": 15612977.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.2053117752075195,
|
|
"epoch": 6.967084639498433,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00015961241797679977,
|
|
"loss": 4.5284,
|
|
"mean_token_accuracy": 0.2287838414311409,
|
|
"num_tokens": 15622914.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.181555366516113,
|
|
"epoch": 6.971003134796238,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015935429588997637,
|
|
"loss": 4.4938,
|
|
"mean_token_accuracy": 0.2240627184510231,
|
|
"num_tokens": 15631929.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.110169553756714,
|
|
"epoch": 6.974921630094044,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00015909638048077724,
|
|
"loss": 4.3655,
|
|
"mean_token_accuracy": 0.2349713146686554,
|
|
"num_tokens": 15640400.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.2417977809906,
|
|
"epoch": 6.97884012539185,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015883867221013915,
|
|
"loss": 4.5532,
|
|
"mean_token_accuracy": 0.2287902683019638,
|
|
"num_tokens": 15648934.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.209074354171753,
|
|
"epoch": 6.982758620689655,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001585811715386285,
|
|
"loss": 4.4916,
|
|
"mean_token_accuracy": 0.21758565604686736,
|
|
"num_tokens": 15658461.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.127295398712159,
|
|
"epoch": 6.986677115987461,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015832387892644072,
|
|
"loss": 4.4114,
|
|
"mean_token_accuracy": 0.23567891269922256,
|
|
"num_tokens": 15667182.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.161234951019287,
|
|
"epoch": 6.990595611285267,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001580667948333994,
|
|
"loss": 4.5335,
|
|
"mean_token_accuracy": 0.22770303338766099,
|
|
"num_tokens": 15675951.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.204468870162964,
|
|
"epoch": 6.994514106583072,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001578099197189555,
|
|
"loss": 4.487,
|
|
"mean_token_accuracy": 0.2292037844657898,
|
|
"num_tokens": 15684171.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.115168285369873,
|
|
"epoch": 6.998432601880878,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00015755325404218662,
|
|
"loss": 4.4673,
|
|
"mean_token_accuracy": 0.23573379516601561,
|
|
"num_tokens": 15692986.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.211828947067261,
|
|
"epoch": 7.002351097178684,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015729679826179577,
|
|
"loss": 4.4528,
|
|
"mean_token_accuracy": 0.23337782472372054,
|
|
"num_tokens": 15702591.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.153503561019898,
|
|
"epoch": 7.006269592476489,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015704055283611123,
|
|
"loss": 4.3717,
|
|
"mean_token_accuracy": 0.23074493408203126,
|
|
"num_tokens": 15711186.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.13337516784668,
|
|
"epoch": 7.010188087774295,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015678451822308495,
|
|
"loss": 4.392,
|
|
"mean_token_accuracy": 0.24478850215673448,
|
|
"num_tokens": 15719831.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.243912267684936,
|
|
"epoch": 7.0141065830721,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001565286948802923,
|
|
"loss": 4.4337,
|
|
"mean_token_accuracy": 0.23278718441724777,
|
|
"num_tokens": 15727926.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.168224859237671,
|
|
"epoch": 7.018025078369906,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00015627308326493118,
|
|
"loss": 4.2923,
|
|
"mean_token_accuracy": 0.24760921597480773,
|
|
"num_tokens": 15735934.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.110566234588623,
|
|
"epoch": 7.021943573667712,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001560176838338209,
|
|
"loss": 4.3679,
|
|
"mean_token_accuracy": 0.2367020606994629,
|
|
"num_tokens": 15745014.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.110121059417724,
|
|
"epoch": 7.025862068965517,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015576249704340173,
|
|
"loss": 4.2897,
|
|
"mean_token_accuracy": 0.24610390961170198,
|
|
"num_tokens": 15753360.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.174406671524048,
|
|
"epoch": 7.029780564263323,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00015550752334973368,
|
|
"loss": 4.3345,
|
|
"mean_token_accuracy": 0.24363444298505782,
|
|
"num_tokens": 15762395.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.143373012542725,
|
|
"epoch": 7.033699059561129,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015525276320849628,
|
|
"loss": 4.2731,
|
|
"mean_token_accuracy": 0.25160236060619356,
|
|
"num_tokens": 15771292.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.211108541488647,
|
|
"epoch": 7.037617554858934,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015499821707498702,
|
|
"loss": 4.3644,
|
|
"mean_token_accuracy": 0.23597102761268615,
|
|
"num_tokens": 15779879.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.173543167114258,
|
|
"epoch": 7.04153605015674,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015474388540412138,
|
|
"loss": 4.3634,
|
|
"mean_token_accuracy": 0.24144828021526338,
|
|
"num_tokens": 15788960.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.181360340118408,
|
|
"epoch": 7.045454545454546,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000154489768650431,
|
|
"loss": 4.3851,
|
|
"mean_token_accuracy": 0.2341240555047989,
|
|
"num_tokens": 15798260.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.156524229049682,
|
|
"epoch": 7.049373040752351,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00015423586726806394,
|
|
"loss": 4.3779,
|
|
"mean_token_accuracy": 0.24158457517623902,
|
|
"num_tokens": 15807480.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.257797193527222,
|
|
"epoch": 7.053291536050157,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015398218171078305,
|
|
"loss": 4.468,
|
|
"mean_token_accuracy": 0.23388794362545012,
|
|
"num_tokens": 15816330.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 7.053291536050157,
|
|
"eval_entropy": 5.116944043211235,
|
|
"eval_loss": 5.659308910369873,
|
|
"eval_mean_token_accuracy": 0.17265465416649514,
|
|
"eval_num_tokens": 15816330.0,
|
|
"eval_runtime": 3.0251,
|
|
"eval_samples_per_second": 1362.591,
|
|
"eval_steps_per_second": 170.572,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 5.172320127487183,
|
|
"epoch": 7.057210031347962,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015372871243196545,
|
|
"loss": 4.3551,
|
|
"mean_token_accuracy": 0.24317787885665892,
|
|
"num_tokens": 15824600.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 5.171639680862427,
|
|
"epoch": 7.061128526645768,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001534754598846021,
|
|
"loss": 4.3803,
|
|
"mean_token_accuracy": 0.2391081526875496,
|
|
"num_tokens": 15834170.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 5.20056734085083,
|
|
"epoch": 7.065047021943574,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015322242452129616,
|
|
"loss": 4.4131,
|
|
"mean_token_accuracy": 0.23915647715330124,
|
|
"num_tokens": 15843061.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 5.188879156112671,
|
|
"epoch": 7.068965517241379,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015296960679426285,
|
|
"loss": 4.3617,
|
|
"mean_token_accuracy": 0.24120075553655623,
|
|
"num_tokens": 15851788.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 5.1940234184265135,
|
|
"epoch": 7.072884012539185,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00015271700715532854,
|
|
"loss": 4.4052,
|
|
"mean_token_accuracy": 0.23646428883075715,
|
|
"num_tokens": 15860216.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 5.141410970687867,
|
|
"epoch": 7.076802507836991,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015246462605592972,
|
|
"loss": 4.2447,
|
|
"mean_token_accuracy": 0.25318610966205596,
|
|
"num_tokens": 15869498.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 5.159812116622925,
|
|
"epoch": 7.080721003134796,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015221246394711224,
|
|
"loss": 4.3125,
|
|
"mean_token_accuracy": 0.24497749507427216,
|
|
"num_tokens": 15877981.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 5.171072864532471,
|
|
"epoch": 7.084639498432602,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00015196052127953085,
|
|
"loss": 4.3672,
|
|
"mean_token_accuracy": 0.23169391304254533,
|
|
"num_tokens": 15886454.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 5.135480833053589,
|
|
"epoch": 7.088557993730408,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001517087985034477,
|
|
"loss": 4.3145,
|
|
"mean_token_accuracy": 0.24771325737237931,
|
|
"num_tokens": 15895025.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 5.182568883895874,
|
|
"epoch": 7.092476489028213,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001514572960687325,
|
|
"loss": 4.437,
|
|
"mean_token_accuracy": 0.23548483103513718,
|
|
"num_tokens": 15903611.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.229039049148559,
|
|
"epoch": 7.096394984326019,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015120601442486061,
|
|
"loss": 4.4046,
|
|
"mean_token_accuracy": 0.23894298523664476,
|
|
"num_tokens": 15912078.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.106166696548462,
|
|
"epoch": 7.100313479623824,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00015095495402091329,
|
|
"loss": 4.3174,
|
|
"mean_token_accuracy": 0.24649193733930588,
|
|
"num_tokens": 15920914.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 5.2056702136993405,
|
|
"epoch": 7.10423197492163,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001507041153055761,
|
|
"loss": 4.3221,
|
|
"mean_token_accuracy": 0.239074245095253,
|
|
"num_tokens": 15929199.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.24063925743103,
|
|
"epoch": 7.108150470219436,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015045349872713838,
|
|
"loss": 4.4417,
|
|
"mean_token_accuracy": 0.24018698036670685,
|
|
"num_tokens": 15937817.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 5.137736892700195,
|
|
"epoch": 7.112068965517241,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015020310473349275,
|
|
"loss": 4.3521,
|
|
"mean_token_accuracy": 0.24369862973690032,
|
|
"num_tokens": 15945997.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 5.1303243160247805,
|
|
"epoch": 7.115987460815047,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014995293377213388,
|
|
"loss": 4.3307,
|
|
"mean_token_accuracy": 0.23796301186084748,
|
|
"num_tokens": 15954215.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 5.195749950408936,
|
|
"epoch": 7.119905956112853,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001497029862901578,
|
|
"loss": 4.3571,
|
|
"mean_token_accuracy": 0.247829869389534,
|
|
"num_tokens": 15962222.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 5.20902795791626,
|
|
"epoch": 7.123824451410658,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00014945326273426102,
|
|
"loss": 4.311,
|
|
"mean_token_accuracy": 0.24291178584098816,
|
|
"num_tokens": 15970712.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 5.201204681396485,
|
|
"epoch": 7.127742946708464,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00014920376355074034,
|
|
"loss": 4.3863,
|
|
"mean_token_accuracy": 0.23635998368263245,
|
|
"num_tokens": 15979049.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 5.14085431098938,
|
|
"epoch": 7.131661442006269,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014895448918549097,
|
|
"loss": 4.3542,
|
|
"mean_token_accuracy": 0.2391251102089882,
|
|
"num_tokens": 15988221.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 5.13284912109375,
|
|
"epoch": 7.135579937304075,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00014870544008400681,
|
|
"loss": 4.3394,
|
|
"mean_token_accuracy": 0.23945988565683365,
|
|
"num_tokens": 15996579.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 5.175402688980102,
|
|
"epoch": 7.139498432601881,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001484566166913789,
|
|
"loss": 4.3549,
|
|
"mean_token_accuracy": 0.24497151821851731,
|
|
"num_tokens": 16005489.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 5.121687316894532,
|
|
"epoch": 7.143416927899686,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014820801945229504,
|
|
"loss": 4.3128,
|
|
"mean_token_accuracy": 0.24248952716588973,
|
|
"num_tokens": 16014167.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 5.142640781402588,
|
|
"epoch": 7.147335423197492,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001479596488110387,
|
|
"loss": 4.2617,
|
|
"mean_token_accuracy": 0.2520608529448509,
|
|
"num_tokens": 16022184.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 5.164277172088623,
|
|
"epoch": 7.151253918495298,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00014771150521148863,
|
|
"loss": 4.3964,
|
|
"mean_token_accuracy": 0.2363363280892372,
|
|
"num_tokens": 16031704.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 5.155966567993164,
|
|
"epoch": 7.155172413793103,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001474635890971176,
|
|
"loss": 4.3996,
|
|
"mean_token_accuracy": 0.24038975685834885,
|
|
"num_tokens": 16040913.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 5.147045373916626,
|
|
"epoch": 7.159090909090909,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001472159009109919,
|
|
"loss": 4.3253,
|
|
"mean_token_accuracy": 0.23955851495265962,
|
|
"num_tokens": 16048943.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 5.158711624145508,
|
|
"epoch": 7.163009404388715,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00014696844109577043,
|
|
"loss": 4.2941,
|
|
"mean_token_accuracy": 0.2441721960902214,
|
|
"num_tokens": 16057999.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 5.1926085472106935,
|
|
"epoch": 7.16692789968652,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014672121009370415,
|
|
"loss": 4.4458,
|
|
"mean_token_accuracy": 0.22476742267608643,
|
|
"num_tokens": 16066920.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 5.121954154968262,
|
|
"epoch": 7.170846394984326,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014647420834663482,
|
|
"loss": 4.3976,
|
|
"mean_token_accuracy": 0.23367435932159425,
|
|
"num_tokens": 16075128.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 5.131617259979248,
|
|
"epoch": 7.174764890282131,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001462274362959946,
|
|
"loss": 4.3845,
|
|
"mean_token_accuracy": 0.2422056719660759,
|
|
"num_tokens": 16083949.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 5.180741643905639,
|
|
"epoch": 7.178683385579937,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001459808943828053,
|
|
"loss": 4.3299,
|
|
"mean_token_accuracy": 0.2455672726035118,
|
|
"num_tokens": 16092891.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 5.208114910125732,
|
|
"epoch": 7.182601880877743,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014573458304767707,
|
|
"loss": 4.4372,
|
|
"mean_token_accuracy": 0.22611262351274491,
|
|
"num_tokens": 16101602.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 5.181302499771118,
|
|
"epoch": 7.186520376175548,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014548850273080839,
|
|
"loss": 4.4135,
|
|
"mean_token_accuracy": 0.22906814217567445,
|
|
"num_tokens": 16110430.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 5.1032572269439695,
|
|
"epoch": 7.190438871473354,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00014524265387198465,
|
|
"loss": 4.3258,
|
|
"mean_token_accuracy": 0.2457436889410019,
|
|
"num_tokens": 16119444.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 5.1285487651824955,
|
|
"epoch": 7.19435736677116,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014499703691057766,
|
|
"loss": 4.3228,
|
|
"mean_token_accuracy": 0.2485710322856903,
|
|
"num_tokens": 16128721.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 5.236220455169677,
|
|
"epoch": 7.198275862068965,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001447516522855447,
|
|
"loss": 4.4171,
|
|
"mean_token_accuracy": 0.23363594561815262,
|
|
"num_tokens": 16137508.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 5.212601280212402,
|
|
"epoch": 7.202194357366771,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00014450650043542775,
|
|
"loss": 4.352,
|
|
"mean_token_accuracy": 0.245132614672184,
|
|
"num_tokens": 16146239.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.176015615463257,
|
|
"epoch": 7.206112852664576,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014426158179835317,
|
|
"loss": 4.3392,
|
|
"mean_token_accuracy": 0.24269825965166092,
|
|
"num_tokens": 16154769.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 5.0692079067230225,
|
|
"epoch": 7.210031347962382,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001440168968120301,
|
|
"loss": 4.2619,
|
|
"mean_token_accuracy": 0.25400163531303405,
|
|
"num_tokens": 16163307.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.141335725784302,
|
|
"epoch": 7.213949843260188,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00014377244591375033,
|
|
"loss": 4.3441,
|
|
"mean_token_accuracy": 0.2402351140975952,
|
|
"num_tokens": 16171533.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 5.09602952003479,
|
|
"epoch": 7.217868338557993,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001435282295403873,
|
|
"loss": 4.3355,
|
|
"mean_token_accuracy": 0.24809646159410476,
|
|
"num_tokens": 16180163.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 5.14062614440918,
|
|
"epoch": 7.221786833855799,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014328424812839525,
|
|
"loss": 4.2898,
|
|
"mean_token_accuracy": 0.24811505675315856,
|
|
"num_tokens": 16188060.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 5.127449083328247,
|
|
"epoch": 7.225705329153605,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014304050211380836,
|
|
"loss": 4.3999,
|
|
"mean_token_accuracy": 0.23575907349586486,
|
|
"num_tokens": 16197597.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 5.171171617507935,
|
|
"epoch": 7.22962382445141,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00014279699193224047,
|
|
"loss": 4.4211,
|
|
"mean_token_accuracy": 0.2271368160843849,
|
|
"num_tokens": 16206288.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 5.148954105377197,
|
|
"epoch": 7.233542319749216,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014255371801888379,
|
|
"loss": 4.3712,
|
|
"mean_token_accuracy": 0.23841252326965331,
|
|
"num_tokens": 16215321.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 5.162658262252807,
|
|
"epoch": 7.237460815047022,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00014231068080850813,
|
|
"loss": 4.2925,
|
|
"mean_token_accuracy": 0.25578602850437165,
|
|
"num_tokens": 16224445.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 5.126517868041992,
|
|
"epoch": 7.241379310344827,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00014206788073546035,
|
|
"loss": 4.3584,
|
|
"mean_token_accuracy": 0.24329910725355147,
|
|
"num_tokens": 16233189.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 5.107948064804077,
|
|
"epoch": 7.245297805642633,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00014182531823366372,
|
|
"loss": 4.3208,
|
|
"mean_token_accuracy": 0.24786730110645294,
|
|
"num_tokens": 16241664.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 5.129499340057373,
|
|
"epoch": 7.2492163009404385,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014158299373661666,
|
|
"loss": 4.3829,
|
|
"mean_token_accuracy": 0.2374998614192009,
|
|
"num_tokens": 16251293.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 5.211177730560303,
|
|
"epoch": 7.253134796238244,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014134090767739223,
|
|
"loss": 4.3498,
|
|
"mean_token_accuracy": 0.24167472422122954,
|
|
"num_tokens": 16260134.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 5.175754737854004,
|
|
"epoch": 7.25705329153605,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014109906048863772,
|
|
"loss": 4.3694,
|
|
"mean_token_accuracy": 0.23533758670091628,
|
|
"num_tokens": 16269027.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 5.167252063751221,
|
|
"epoch": 7.2609717868338555,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001408574526025731,
|
|
"loss": 4.3687,
|
|
"mean_token_accuracy": 0.24147286415100097,
|
|
"num_tokens": 16277692.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 5.1410637378692625,
|
|
"epoch": 7.264890282131661,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014061608445099083,
|
|
"loss": 4.3885,
|
|
"mean_token_accuracy": 0.23590022921562195,
|
|
"num_tokens": 16287059.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.103167200088501,
|
|
"epoch": 7.268808777429467,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00014037495646525495,
|
|
"loss": 4.339,
|
|
"mean_token_accuracy": 0.24257039427757263,
|
|
"num_tokens": 16294880.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 5.144848203659057,
|
|
"epoch": 7.2727272727272725,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001401340690763003,
|
|
"loss": 4.307,
|
|
"mean_token_accuracy": 0.24800651520490646,
|
|
"num_tokens": 16304101.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 5.192517471313477,
|
|
"epoch": 7.2766457680250785,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013989342271463173,
|
|
"loss": 4.4545,
|
|
"mean_token_accuracy": 0.22900593131780625,
|
|
"num_tokens": 16312887.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 5.137587928771973,
|
|
"epoch": 7.280564263322884,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001396530178103232,
|
|
"loss": 4.342,
|
|
"mean_token_accuracy": 0.24301686882972717,
|
|
"num_tokens": 16322256.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 5.177801942825317,
|
|
"epoch": 7.2844827586206895,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001394128547930174,
|
|
"loss": 4.3814,
|
|
"mean_token_accuracy": 0.2377953752875328,
|
|
"num_tokens": 16331810.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 5.212534761428833,
|
|
"epoch": 7.2884012539184955,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001391729340919245,
|
|
"loss": 4.3886,
|
|
"mean_token_accuracy": 0.23496057838201523,
|
|
"num_tokens": 16341385.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 5.146743392944336,
|
|
"epoch": 7.2923197492163006,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013893325613582172,
|
|
"loss": 4.3535,
|
|
"mean_token_accuracy": 0.24320681691169738,
|
|
"num_tokens": 16350231.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.15025782585144,
|
|
"epoch": 7.2962382445141065,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001386938213530523,
|
|
"loss": 4.4133,
|
|
"mean_token_accuracy": 0.2374377056956291,
|
|
"num_tokens": 16359659.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 5.138523912429809,
|
|
"epoch": 7.3001567398119125,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013845463017152527,
|
|
"loss": 4.3455,
|
|
"mean_token_accuracy": 0.2445626825094223,
|
|
"num_tokens": 16368606.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.167492866516113,
|
|
"epoch": 7.304075235109718,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013821568301871384,
|
|
"loss": 4.4183,
|
|
"mean_token_accuracy": 0.23451228439807892,
|
|
"num_tokens": 16376428.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 5.123425817489624,
|
|
"epoch": 7.3079937304075235,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00013797698032165534,
|
|
"loss": 4.3646,
|
|
"mean_token_accuracy": 0.23807963877916336,
|
|
"num_tokens": 16384983.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 5.175596475601196,
|
|
"epoch": 7.3119122257053295,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013773852250695035,
|
|
"loss": 4.3773,
|
|
"mean_token_accuracy": 0.23751893341541291,
|
|
"num_tokens": 16393747.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 5.162880277633667,
|
|
"epoch": 7.315830721003135,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00013750031000076145,
|
|
"loss": 4.3649,
|
|
"mean_token_accuracy": 0.24169625490903854,
|
|
"num_tokens": 16401899.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.17450361251831,
|
|
"epoch": 7.3197492163009406,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001372623432288131,
|
|
"loss": 4.3666,
|
|
"mean_token_accuracy": 0.23831921964883804,
|
|
"num_tokens": 16410787.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 5.197457313537598,
|
|
"epoch": 7.3236677115987465,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013702462261639054,
|
|
"loss": 4.4044,
|
|
"mean_token_accuracy": 0.23608262836933136,
|
|
"num_tokens": 16420023.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 5.143267440795898,
|
|
"epoch": 7.327586206896552,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000136787148588339,
|
|
"loss": 4.3166,
|
|
"mean_token_accuracy": 0.24315857738256455,
|
|
"num_tokens": 16428394.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.164436626434326,
|
|
"epoch": 7.331504702194358,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001365499215690632,
|
|
"loss": 4.3502,
|
|
"mean_token_accuracy": 0.24195693731307982,
|
|
"num_tokens": 16436429.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.091635227203369,
|
|
"epoch": 7.335423197492163,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013631294198252603,
|
|
"loss": 4.283,
|
|
"mean_token_accuracy": 0.2498973399400711,
|
|
"num_tokens": 16445140.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 5.1505317211151125,
|
|
"epoch": 7.339341692789969,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013607621025224864,
|
|
"loss": 4.3308,
|
|
"mean_token_accuracy": 0.24184153527021407,
|
|
"num_tokens": 16453932.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 5.129813528060913,
|
|
"epoch": 7.343260188087775,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000135839726801309,
|
|
"loss": 4.3296,
|
|
"mean_token_accuracy": 0.23761074393987655,
|
|
"num_tokens": 16462539.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 5.151110410690308,
|
|
"epoch": 7.34717868338558,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013560349205234124,
|
|
"loss": 4.3989,
|
|
"mean_token_accuracy": 0.2339349642395973,
|
|
"num_tokens": 16471979.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.196987056732178,
|
|
"epoch": 7.351097178683386,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00013536750642753538,
|
|
"loss": 4.4718,
|
|
"mean_token_accuracy": 0.22756578773260117,
|
|
"num_tokens": 16481266.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 5.160649299621582,
|
|
"epoch": 7.355015673981192,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013513177034863588,
|
|
"loss": 4.416,
|
|
"mean_token_accuracy": 0.23570969253778457,
|
|
"num_tokens": 16490360.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 5.19037094116211,
|
|
"epoch": 7.358934169278997,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001348962842369413,
|
|
"loss": 4.4553,
|
|
"mean_token_accuracy": 0.22769348323345184,
|
|
"num_tokens": 16498043.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 5.143081426620483,
|
|
"epoch": 7.362852664576803,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00013466104851330346,
|
|
"loss": 4.389,
|
|
"mean_token_accuracy": 0.2438863217830658,
|
|
"num_tokens": 16506343.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 5.161760568618774,
|
|
"epoch": 7.366771159874608,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001344260635981268,
|
|
"loss": 4.3614,
|
|
"mean_token_accuracy": 0.24074523895978928,
|
|
"num_tokens": 16515308.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 5.1723614692687985,
|
|
"epoch": 7.370689655172414,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013419132991136762,
|
|
"loss": 4.3654,
|
|
"mean_token_accuracy": 0.23662531077861787,
|
|
"num_tokens": 16524379.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 5.192597150802612,
|
|
"epoch": 7.37460815047022,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013395684787253278,
|
|
"loss": 4.3802,
|
|
"mean_token_accuracy": 0.244297493994236,
|
|
"num_tokens": 16532797.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.1614905834198,
|
|
"epoch": 7.378526645768025,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00013372261790068,
|
|
"loss": 4.3888,
|
|
"mean_token_accuracy": 0.23768314719200134,
|
|
"num_tokens": 16540984.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 5.126627588272095,
|
|
"epoch": 7.382445141065831,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013348864041441604,
|
|
"loss": 4.2698,
|
|
"mean_token_accuracy": 0.2483128383755684,
|
|
"num_tokens": 16549386.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 5.115985345840454,
|
|
"epoch": 7.386363636363637,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013325491583189664,
|
|
"loss": 4.3655,
|
|
"mean_token_accuracy": 0.23724571317434312,
|
|
"num_tokens": 16557673.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 5.180406665802002,
|
|
"epoch": 7.390282131661442,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013302144457082553,
|
|
"loss": 4.4117,
|
|
"mean_token_accuracy": 0.2350928530097008,
|
|
"num_tokens": 16566383.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 5.201940393447876,
|
|
"epoch": 7.394200626959248,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013278822704845372,
|
|
"loss": 4.3504,
|
|
"mean_token_accuracy": 0.24445975720882415,
|
|
"num_tokens": 16574819.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 5.175523853302002,
|
|
"epoch": 7.398119122257054,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013255526368157876,
|
|
"loss": 4.366,
|
|
"mean_token_accuracy": 0.2422635242342949,
|
|
"num_tokens": 16582571.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 5.193794584274292,
|
|
"epoch": 7.402037617554859,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013232255488654387,
|
|
"loss": 4.4246,
|
|
"mean_token_accuracy": 0.2433891236782074,
|
|
"num_tokens": 16592045.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 5.140433883666992,
|
|
"epoch": 7.405956112852665,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013209010107923747,
|
|
"loss": 4.2923,
|
|
"mean_token_accuracy": 0.24770380258560182,
|
|
"num_tokens": 16600864.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 5.196175384521484,
|
|
"epoch": 7.40987460815047,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013185790267509218,
|
|
"loss": 4.4693,
|
|
"mean_token_accuracy": 0.23893872946500777,
|
|
"num_tokens": 16610193.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 5.1561229705810545,
|
|
"epoch": 7.413793103448276,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00013162596008908423,
|
|
"loss": 4.4058,
|
|
"mean_token_accuracy": 0.2320484235882759,
|
|
"num_tokens": 16618663.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.125619220733642,
|
|
"epoch": 7.417711598746082,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013139427373573255,
|
|
"loss": 4.3249,
|
|
"mean_token_accuracy": 0.2468751385807991,
|
|
"num_tokens": 16626885.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 5.201581621170044,
|
|
"epoch": 7.421630094043887,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00013116284402909834,
|
|
"loss": 4.4272,
|
|
"mean_token_accuracy": 0.2348688945174217,
|
|
"num_tokens": 16636592.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.204153537750244,
|
|
"epoch": 7.425548589341693,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000130931671382784,
|
|
"loss": 4.4465,
|
|
"mean_token_accuracy": 0.23496971577405928,
|
|
"num_tokens": 16645897.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 5.207467842102051,
|
|
"epoch": 7.429467084639499,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013070075620993233,
|
|
"loss": 4.3509,
|
|
"mean_token_accuracy": 0.24144454300403595,
|
|
"num_tokens": 16654034.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 5.1261414051055905,
|
|
"epoch": 7.433385579937304,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013047009892322642,
|
|
"loss": 4.3717,
|
|
"mean_token_accuracy": 0.24160025715827943,
|
|
"num_tokens": 16662510.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.160886478424072,
|
|
"epoch": 7.43730407523511,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001302396999348882,
|
|
"loss": 4.3383,
|
|
"mean_token_accuracy": 0.24568344801664352,
|
|
"num_tokens": 16671014.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 5.1490833282470705,
|
|
"epoch": 7.441222570532915,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00013000955965667788,
|
|
"loss": 4.4146,
|
|
"mean_token_accuracy": 0.23760631084442138,
|
|
"num_tokens": 16680068.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 5.167875289916992,
|
|
"epoch": 7.445141065830721,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001297796784998936,
|
|
"loss": 4.4399,
|
|
"mean_token_accuracy": 0.23991014957427978,
|
|
"num_tokens": 16688478.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 7.445141065830721,
|
|
"eval_entropy": 5.070259285527606,
|
|
"eval_loss": 5.6523284912109375,
|
|
"eval_mean_token_accuracy": 0.17335379995809969,
|
|
"eval_num_tokens": 16688478.0,
|
|
"eval_runtime": 2.8428,
|
|
"eval_samples_per_second": 1449.985,
|
|
"eval_steps_per_second": 181.512,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 5.172882127761841,
|
|
"epoch": 7.449059561128527,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00012955005687537025,
|
|
"loss": 4.4755,
|
|
"mean_token_accuracy": 0.23101314008235932,
|
|
"num_tokens": 16698913.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 5.117330074310303,
|
|
"epoch": 7.452978056426332,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012932069519347873,
|
|
"loss": 4.3149,
|
|
"mean_token_accuracy": 0.24778521209955215,
|
|
"num_tokens": 16708306.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 5.207182455062866,
|
|
"epoch": 7.456896551724138,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012909159386412578,
|
|
"loss": 4.3765,
|
|
"mean_token_accuracy": 0.2492469221353531,
|
|
"num_tokens": 16716345.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 5.145506381988525,
|
|
"epoch": 7.460815047021944,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012886275329675246,
|
|
"loss": 4.3604,
|
|
"mean_token_accuracy": 0.24152330905199051,
|
|
"num_tokens": 16725405.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 5.135979175567627,
|
|
"epoch": 7.464733542319749,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001286341739003341,
|
|
"loss": 4.3907,
|
|
"mean_token_accuracy": 0.2358911395072937,
|
|
"num_tokens": 16734771.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 5.1282106876373295,
|
|
"epoch": 7.468652037617555,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012840585608337907,
|
|
"loss": 4.4005,
|
|
"mean_token_accuracy": 0.2330407604575157,
|
|
"num_tokens": 16742403.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 5.12530255317688,
|
|
"epoch": 7.472570532915361,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012817780025392833,
|
|
"loss": 4.3618,
|
|
"mean_token_accuracy": 0.24328250735998153,
|
|
"num_tokens": 16751450.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 5.130151081085205,
|
|
"epoch": 7.476489028213166,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012795000681955468,
|
|
"loss": 4.3435,
|
|
"mean_token_accuracy": 0.23610852509737015,
|
|
"num_tokens": 16760216.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 5.1511969566345215,
|
|
"epoch": 7.480407523510972,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012772247618736182,
|
|
"loss": 4.41,
|
|
"mean_token_accuracy": 0.23723017573356628,
|
|
"num_tokens": 16768648.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 5.08496413230896,
|
|
"epoch": 7.484326018808777,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012749520876398395,
|
|
"loss": 4.3077,
|
|
"mean_token_accuracy": 0.2501813217997551,
|
|
"num_tokens": 16776587.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 5.13529372215271,
|
|
"epoch": 7.488244514106583,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012726820495558483,
|
|
"loss": 4.3575,
|
|
"mean_token_accuracy": 0.24500802159309387,
|
|
"num_tokens": 16785937.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 5.139859628677368,
|
|
"epoch": 7.492163009404389,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012704146516785705,
|
|
"loss": 4.3383,
|
|
"mean_token_accuracy": 0.2441197469830513,
|
|
"num_tokens": 16794359.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 5.190577173233033,
|
|
"epoch": 7.496081504702194,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001268149898060213,
|
|
"loss": 4.4011,
|
|
"mean_token_accuracy": 0.23856491595506668,
|
|
"num_tokens": 16803685.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 5.093593502044678,
|
|
"epoch": 7.5,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00012658877927482586,
|
|
"loss": 4.3306,
|
|
"mean_token_accuracy": 0.2402963399887085,
|
|
"num_tokens": 16812267.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 5.194755601882934,
|
|
"epoch": 7.503918495297806,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001263628339785456,
|
|
"loss": 4.4235,
|
|
"mean_token_accuracy": 0.2380077838897705,
|
|
"num_tokens": 16820334.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 5.158464241027832,
|
|
"epoch": 7.507836990595611,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00012613715432098138,
|
|
"loss": 4.3872,
|
|
"mean_token_accuracy": 0.24125765413045883,
|
|
"num_tokens": 16828908.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 5.205153179168701,
|
|
"epoch": 7.511755485893417,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012591174070545932,
|
|
"loss": 4.4081,
|
|
"mean_token_accuracy": 0.23722308725118638,
|
|
"num_tokens": 16837772.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 5.178397941589355,
|
|
"epoch": 7.515673981191222,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012568659353483024,
|
|
"loss": 4.4004,
|
|
"mean_token_accuracy": 0.2387392282485962,
|
|
"num_tokens": 16846714.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 5.162786626815796,
|
|
"epoch": 7.519592476489028,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012546171321146848,
|
|
"loss": 4.3464,
|
|
"mean_token_accuracy": 0.24733814746141433,
|
|
"num_tokens": 16854745.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 5.18314208984375,
|
|
"epoch": 7.523510971786834,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00012523710013727173,
|
|
"loss": 4.3833,
|
|
"mean_token_accuracy": 0.24683560132980348,
|
|
"num_tokens": 16864181.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 5.159389638900757,
|
|
"epoch": 7.527429467084639,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012501275471365998,
|
|
"loss": 4.4092,
|
|
"mean_token_accuracy": 0.2388117715716362,
|
|
"num_tokens": 16873551.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 5.198206615447998,
|
|
"epoch": 7.531347962382445,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012478867734157486,
|
|
"loss": 4.4704,
|
|
"mean_token_accuracy": 0.23625562489032745,
|
|
"num_tokens": 16883108.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 5.107951593399048,
|
|
"epoch": 7.535266457680251,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012456486842147885,
|
|
"loss": 4.3668,
|
|
"mean_token_accuracy": 0.24997108429670334,
|
|
"num_tokens": 16891772.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 5.147489738464356,
|
|
"epoch": 7.539184952978056,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012434132835335496,
|
|
"loss": 4.3535,
|
|
"mean_token_accuracy": 0.2393592491745949,
|
|
"num_tokens": 16900519.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 5.177788114547729,
|
|
"epoch": 7.543103448275862,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012411805753670543,
|
|
"loss": 4.4024,
|
|
"mean_token_accuracy": 0.2480081334710121,
|
|
"num_tokens": 16909228.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 5.162473201751709,
|
|
"epoch": 7.547021943573668,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001238950563705514,
|
|
"loss": 4.417,
|
|
"mean_token_accuracy": 0.23712728470563887,
|
|
"num_tokens": 16918100.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 5.178057098388672,
|
|
"epoch": 7.550940438871473,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001236723252534322,
|
|
"loss": 4.3661,
|
|
"mean_token_accuracy": 0.2371207520365715,
|
|
"num_tokens": 16926911.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 5.184252309799194,
|
|
"epoch": 7.554858934169279,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012344986458340433,
|
|
"loss": 4.4153,
|
|
"mean_token_accuracy": 0.23689059913158417,
|
|
"num_tokens": 16935638.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 5.117220878601074,
|
|
"epoch": 7.558777429467085,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012322767475804122,
|
|
"loss": 4.2661,
|
|
"mean_token_accuracy": 0.25316938012838364,
|
|
"num_tokens": 16944351.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 5.17743673324585,
|
|
"epoch": 7.56269592476489,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00012300575617443194,
|
|
"loss": 4.3761,
|
|
"mean_token_accuracy": 0.2410977840423584,
|
|
"num_tokens": 16952486.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 5.193868017196655,
|
|
"epoch": 7.566614420062696,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001227841092291812,
|
|
"loss": 4.4471,
|
|
"mean_token_accuracy": 0.2309522047638893,
|
|
"num_tokens": 16960877.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 5.147039270401001,
|
|
"epoch": 7.570532915360501,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00012256273431840792,
|
|
"loss": 4.413,
|
|
"mean_token_accuracy": 0.24519958198070527,
|
|
"num_tokens": 16970868.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 5.159555292129516,
|
|
"epoch": 7.574451410658307,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00012234163183774482,
|
|
"loss": 4.4018,
|
|
"mean_token_accuracy": 0.23584541380405427,
|
|
"num_tokens": 16980339.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 5.150513935089111,
|
|
"epoch": 7.578369905956113,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00012212080218233814,
|
|
"loss": 4.3425,
|
|
"mean_token_accuracy": 0.2422019734978676,
|
|
"num_tokens": 16988786.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 5.146929931640625,
|
|
"epoch": 7.582288401253918,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00012190024574684614,
|
|
"loss": 4.3863,
|
|
"mean_token_accuracy": 0.23603826761245728,
|
|
"num_tokens": 16998454.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 5.187962293624878,
|
|
"epoch": 7.586206896551724,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012167996292543889,
|
|
"loss": 4.3602,
|
|
"mean_token_accuracy": 0.24268772602081298,
|
|
"num_tokens": 17006951.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 5.225212240219117,
|
|
"epoch": 7.59012539184953,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001214599541117976,
|
|
"loss": 4.364,
|
|
"mean_token_accuracy": 0.24474995583295822,
|
|
"num_tokens": 17015991.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 5.161370086669922,
|
|
"epoch": 7.594043887147335,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012124021969911366,
|
|
"loss": 4.4063,
|
|
"mean_token_accuracy": 0.2354728415608406,
|
|
"num_tokens": 17024493.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 5.1819908142089846,
|
|
"epoch": 7.597962382445141,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00012102076008008804,
|
|
"loss": 4.3889,
|
|
"mean_token_accuracy": 0.2394329398870468,
|
|
"num_tokens": 17033402.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 5.112108993530273,
|
|
"epoch": 7.601880877742946,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00012080157564693073,
|
|
"loss": 4.3333,
|
|
"mean_token_accuracy": 0.24755663573741912,
|
|
"num_tokens": 17042417.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 5.198781585693359,
|
|
"epoch": 7.605799373040752,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012058266679135987,
|
|
"loss": 4.5113,
|
|
"mean_token_accuracy": 0.2357491135597229,
|
|
"num_tokens": 17050395.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 5.201287794113159,
|
|
"epoch": 7.609717868338558,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012036403390460107,
|
|
"loss": 4.4352,
|
|
"mean_token_accuracy": 0.23476205319166182,
|
|
"num_tokens": 17060077.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 5.211333322525024,
|
|
"epoch": 7.613636363636363,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00012014567737738667,
|
|
"loss": 4.3733,
|
|
"mean_token_accuracy": 0.23190424591302872,
|
|
"num_tokens": 17068099.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 5.1647758960723875,
|
|
"epoch": 7.617554858934169,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011992759759995529,
|
|
"loss": 4.3381,
|
|
"mean_token_accuracy": 0.23662445098161697,
|
|
"num_tokens": 17076846.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 5.153008317947387,
|
|
"epoch": 7.621473354231975,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011970979496205078,
|
|
"loss": 4.3509,
|
|
"mean_token_accuracy": 0.23953068256378174,
|
|
"num_tokens": 17085960.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 5.128280687332153,
|
|
"epoch": 7.62539184952978,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011949226985292183,
|
|
"loss": 4.3326,
|
|
"mean_token_accuracy": 0.24813373386859894,
|
|
"num_tokens": 17095327.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 5.117192268371582,
|
|
"epoch": 7.629310344827586,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011927502266132098,
|
|
"loss": 4.354,
|
|
"mean_token_accuracy": 0.24284843653440474,
|
|
"num_tokens": 17103961.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 5.113250398635865,
|
|
"epoch": 7.633228840125392,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011905805377550428,
|
|
"loss": 4.3486,
|
|
"mean_token_accuracy": 0.24174848943948746,
|
|
"num_tokens": 17112427.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 5.159414052963257,
|
|
"epoch": 7.637147335423197,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001188413635832303,
|
|
"loss": 4.4157,
|
|
"mean_token_accuracy": 0.23921623826026917,
|
|
"num_tokens": 17120890.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 5.133922386169433,
|
|
"epoch": 7.641065830721003,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011862495247175947,
|
|
"loss": 4.278,
|
|
"mean_token_accuracy": 0.24748827666044235,
|
|
"num_tokens": 17129881.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 5.200897121429444,
|
|
"epoch": 7.644984326018808,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011840882082785357,
|
|
"loss": 4.4397,
|
|
"mean_token_accuracy": 0.2382202237844467,
|
|
"num_tokens": 17138900.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 5.1409711837768555,
|
|
"epoch": 7.648902821316614,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000118192969037775,
|
|
"loss": 4.3737,
|
|
"mean_token_accuracy": 0.2367101490497589,
|
|
"num_tokens": 17147023.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 5.207751131057739,
|
|
"epoch": 7.65282131661442,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011797739748728575,
|
|
"loss": 4.4475,
|
|
"mean_token_accuracy": 0.22819736748933792,
|
|
"num_tokens": 17156926.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 5.180087423324585,
|
|
"epoch": 7.656739811912225,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001177621065616473,
|
|
"loss": 4.4378,
|
|
"mean_token_accuracy": 0.2275232568383217,
|
|
"num_tokens": 17165079.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 5.112714052200317,
|
|
"epoch": 7.660658307210031,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011754709664561931,
|
|
"loss": 4.3215,
|
|
"mean_token_accuracy": 0.2504021406173706,
|
|
"num_tokens": 17173509.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 5.2129803657531735,
|
|
"epoch": 7.664576802507837,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011733236812345943,
|
|
"loss": 4.385,
|
|
"mean_token_accuracy": 0.23691839277744292,
|
|
"num_tokens": 17182399.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 5.15443754196167,
|
|
"epoch": 7.668495297805642,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011711792137892221,
|
|
"loss": 4.3614,
|
|
"mean_token_accuracy": 0.2444360598921776,
|
|
"num_tokens": 17190733.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 5.128833484649658,
|
|
"epoch": 7.672413793103448,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011690375679525896,
|
|
"loss": 4.3826,
|
|
"mean_token_accuracy": 0.2353457823395729,
|
|
"num_tokens": 17200078.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 5.158110189437866,
|
|
"epoch": 7.6763322884012535,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011668987475521642,
|
|
"loss": 4.4075,
|
|
"mean_token_accuracy": 0.24146130234003066,
|
|
"num_tokens": 17208430.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 5.174743032455444,
|
|
"epoch": 7.6802507836990594,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011647627564103636,
|
|
"loss": 4.4549,
|
|
"mean_token_accuracy": 0.23484891802072524,
|
|
"num_tokens": 17217309.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 5.149191904067993,
|
|
"epoch": 7.684169278996865,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011626295983445525,
|
|
"loss": 4.3054,
|
|
"mean_token_accuracy": 0.24357050508260727,
|
|
"num_tokens": 17225700.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 5.117889404296875,
|
|
"epoch": 7.6880877742946705,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011604992771670292,
|
|
"loss": 4.3598,
|
|
"mean_token_accuracy": 0.23986896872520447,
|
|
"num_tokens": 17234582.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 5.092018270492554,
|
|
"epoch": 7.6920062695924765,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011583717966850219,
|
|
"loss": 4.3511,
|
|
"mean_token_accuracy": 0.248282228410244,
|
|
"num_tokens": 17243207.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 5.1117864608764645,
|
|
"epoch": 7.695924764890282,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011562471607006845,
|
|
"loss": 4.339,
|
|
"mean_token_accuracy": 0.2406468152999878,
|
|
"num_tokens": 17251658.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 5.184154748916626,
|
|
"epoch": 7.6998432601880875,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011541253730110866,
|
|
"loss": 4.4797,
|
|
"mean_token_accuracy": 0.23856445401906967,
|
|
"num_tokens": 17261175.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 5.099665355682373,
|
|
"epoch": 7.7037617554858935,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001152006437408206,
|
|
"loss": 4.2785,
|
|
"mean_token_accuracy": 0.2493626743555069,
|
|
"num_tokens": 17269302.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 5.162882614135742,
|
|
"epoch": 7.7076802507836994,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011498903576789233,
|
|
"loss": 4.4531,
|
|
"mean_token_accuracy": 0.23471538722515106,
|
|
"num_tokens": 17277911.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 5.1605311870574955,
|
|
"epoch": 7.7115987460815045,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00011477771376050177,
|
|
"loss": 4.3694,
|
|
"mean_token_accuracy": 0.24458009004592896,
|
|
"num_tokens": 17287219.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 5.147277402877807,
|
|
"epoch": 7.7155172413793105,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011456667809631556,
|
|
"loss": 4.3292,
|
|
"mean_token_accuracy": 0.24855772256851197,
|
|
"num_tokens": 17295716.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 5.143343114852906,
|
|
"epoch": 7.7194357366771165,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00011435592915248855,
|
|
"loss": 4.3914,
|
|
"mean_token_accuracy": 0.2383372351527214,
|
|
"num_tokens": 17305870.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 5.139862537384033,
|
|
"epoch": 7.7233542319749215,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011414546730566342,
|
|
"loss": 4.3692,
|
|
"mean_token_accuracy": 0.2396491587162018,
|
|
"num_tokens": 17314130.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 5.1659938335418705,
|
|
"epoch": 7.7272727272727275,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011393529293196955,
|
|
"loss": 4.2725,
|
|
"mean_token_accuracy": 0.2542906180024147,
|
|
"num_tokens": 17322320.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 5.164686965942383,
|
|
"epoch": 7.731191222570533,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011372540640702263,
|
|
"loss": 4.3831,
|
|
"mean_token_accuracy": 0.24489184617996215,
|
|
"num_tokens": 17331162.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 5.10954008102417,
|
|
"epoch": 7.735109717868339,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00011351580810592382,
|
|
"loss": 4.3019,
|
|
"mean_token_accuracy": 0.24972813427448273,
|
|
"num_tokens": 17339568.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 5.168736600875855,
|
|
"epoch": 7.7390282131661445,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00011330649840325938,
|
|
"loss": 4.3832,
|
|
"mean_token_accuracy": 0.2424519345164299,
|
|
"num_tokens": 17349354.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 5.114012432098389,
|
|
"epoch": 7.74294670846395,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001130974776730997,
|
|
"loss": 4.4052,
|
|
"mean_token_accuracy": 0.23417973816394805,
|
|
"num_tokens": 17358668.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 5.1022755146026615,
|
|
"epoch": 7.746865203761756,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011288874628899864,
|
|
"loss": 4.2959,
|
|
"mean_token_accuracy": 0.24230333864688874,
|
|
"num_tokens": 17366792.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 5.163437986373902,
|
|
"epoch": 7.750783699059561,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011268030462399309,
|
|
"loss": 4.3581,
|
|
"mean_token_accuracy": 0.23704227507114412,
|
|
"num_tokens": 17375589.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 5.238901519775391,
|
|
"epoch": 7.754702194357367,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011247215305060213,
|
|
"loss": 4.4256,
|
|
"mean_token_accuracy": 0.24069685786962508,
|
|
"num_tokens": 17385816.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 5.10908784866333,
|
|
"epoch": 7.758620689655173,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011226429194082627,
|
|
"loss": 4.3301,
|
|
"mean_token_accuracy": 0.24753255397081375,
|
|
"num_tokens": 17395267.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 5.203282451629638,
|
|
"epoch": 7.762539184952978,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000112056721666147,
|
|
"loss": 4.3858,
|
|
"mean_token_accuracy": 0.24585672318935395,
|
|
"num_tokens": 17403544.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 5.172358417510987,
|
|
"epoch": 7.766457680250784,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011184944259752622,
|
|
"loss": 4.3662,
|
|
"mean_token_accuracy": 0.24221378564834595,
|
|
"num_tokens": 17412090.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 5.139830017089844,
|
|
"epoch": 7.77037617554859,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011164245510540512,
|
|
"loss": 4.3702,
|
|
"mean_token_accuracy": 0.24007170647382736,
|
|
"num_tokens": 17420558.0,
|
|
"step": 9915
|
|
},
|
|
{
|
|
"entropy": 5.111873626708984,
|
|
"epoch": 7.774294670846395,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011143575955970384,
|
|
"loss": 4.3829,
|
|
"mean_token_accuracy": 0.2386606067419052,
|
|
"num_tokens": 17429901.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 5.196046829223633,
|
|
"epoch": 7.778213166144201,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011122935632982101,
|
|
"loss": 4.4234,
|
|
"mean_token_accuracy": 0.2316632553935051,
|
|
"num_tokens": 17438543.0,
|
|
"step": 9925
|
|
},
|
|
{
|
|
"entropy": 5.249775171279907,
|
|
"epoch": 7.782131661442007,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001110232457846325,
|
|
"loss": 4.4305,
|
|
"mean_token_accuracy": 0.23833946734666825,
|
|
"num_tokens": 17447234.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 5.107551765441895,
|
|
"epoch": 7.786050156739812,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011081742829249142,
|
|
"loss": 4.2972,
|
|
"mean_token_accuracy": 0.2457659959793091,
|
|
"num_tokens": 17455542.0,
|
|
"step": 9935
|
|
},
|
|
{
|
|
"entropy": 5.169161367416382,
|
|
"epoch": 7.789968652037618,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011061190422122687,
|
|
"loss": 4.3457,
|
|
"mean_token_accuracy": 0.25333226323127744,
|
|
"num_tokens": 17465035.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 5.15905990600586,
|
|
"epoch": 7.793887147335424,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011040667393814383,
|
|
"loss": 4.3661,
|
|
"mean_token_accuracy": 0.24173829406499864,
|
|
"num_tokens": 17473748.0,
|
|
"step": 9945
|
|
},
|
|
{
|
|
"entropy": 5.081380128860474,
|
|
"epoch": 7.797805642633229,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011020173781002199,
|
|
"loss": 4.241,
|
|
"mean_token_accuracy": 0.2525812774896622,
|
|
"num_tokens": 17481525.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 5.15574049949646,
|
|
"epoch": 7.801724137931035,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001099970962031154,
|
|
"loss": 4.4066,
|
|
"mean_token_accuracy": 0.24207190424203873,
|
|
"num_tokens": 17490044.0,
|
|
"step": 9955
|
|
},
|
|
{
|
|
"entropy": 5.215651178359986,
|
|
"epoch": 7.80564263322884,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010979274948315191,
|
|
"loss": 4.4479,
|
|
"mean_token_accuracy": 0.23504060059785842,
|
|
"num_tokens": 17498094.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 5.162240362167358,
|
|
"epoch": 7.809561128526646,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010958869801533222,
|
|
"loss": 4.3912,
|
|
"mean_token_accuracy": 0.24124593138694764,
|
|
"num_tokens": 17507677.0,
|
|
"step": 9965
|
|
},
|
|
{
|
|
"entropy": 5.143397092819214,
|
|
"epoch": 7.813479623824452,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010938494216432928,
|
|
"loss": 4.3564,
|
|
"mean_token_accuracy": 0.24242679476737977,
|
|
"num_tokens": 17516489.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 5.154837226867675,
|
|
"epoch": 7.817398119122257,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010918148229428796,
|
|
"loss": 4.3852,
|
|
"mean_token_accuracy": 0.24105837643146516,
|
|
"num_tokens": 17524972.0,
|
|
"step": 9975
|
|
},
|
|
{
|
|
"entropy": 5.142380142211914,
|
|
"epoch": 7.821316614420063,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000108978318768824,
|
|
"loss": 4.3868,
|
|
"mean_token_accuracy": 0.23907871246337892,
|
|
"num_tokens": 17533745.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 5.1022858142852785,
|
|
"epoch": 7.825235109717868,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00010877545195102347,
|
|
"loss": 4.3604,
|
|
"mean_token_accuracy": 0.23333367109298705,
|
|
"num_tokens": 17542812.0,
|
|
"step": 9985
|
|
},
|
|
{
|
|
"entropy": 5.138338994979859,
|
|
"epoch": 7.829153605015674,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00010857288220344242,
|
|
"loss": 4.3443,
|
|
"mean_token_accuracy": 0.24474188983440398,
|
|
"num_tokens": 17551123.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 5.180613422393799,
|
|
"epoch": 7.83307210031348,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010837060988810571,
|
|
"loss": 4.4707,
|
|
"mean_token_accuracy": 0.232795050740242,
|
|
"num_tokens": 17560729.0,
|
|
"step": 9995
|
|
},
|
|
{
|
|
"entropy": 5.20838737487793,
|
|
"epoch": 7.836990595611285,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010816863536650687,
|
|
"loss": 4.4166,
|
|
"mean_token_accuracy": 0.23585203438997268,
|
|
"num_tokens": 17569842.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"epoch": 7.836990595611285,
|
|
"eval_entropy": 5.092260955840118,
|
|
"eval_loss": 5.646599769592285,
|
|
"eval_mean_token_accuracy": 0.17366565442680157,
|
|
"eval_num_tokens": 17569842.0,
|
|
"eval_runtime": 2.8443,
|
|
"eval_samples_per_second": 1449.202,
|
|
"eval_steps_per_second": 181.414,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"entropy": 5.159105587005615,
|
|
"epoch": 7.840909090909091,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010796695899960696,
|
|
"loss": 4.3367,
|
|
"mean_token_accuracy": 0.24487050324678422,
|
|
"num_tokens": 17578887.0,
|
|
"step": 10005
|
|
},
|
|
{
|
|
"entropy": 5.175011110305786,
|
|
"epoch": 7.844827586206897,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010776558114783455,
|
|
"loss": 4.354,
|
|
"mean_token_accuracy": 0.23506431579589843,
|
|
"num_tokens": 17587321.0,
|
|
"step": 10010
|
|
},
|
|
{
|
|
"entropy": 5.130767011642456,
|
|
"epoch": 7.848746081504702,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010756450217108441,
|
|
"loss": 4.3489,
|
|
"mean_token_accuracy": 0.23910555690526963,
|
|
"num_tokens": 17596295.0,
|
|
"step": 10015
|
|
},
|
|
{
|
|
"entropy": 5.1350462436676025,
|
|
"epoch": 7.852664576802508,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001073637224287172,
|
|
"loss": 4.3379,
|
|
"mean_token_accuracy": 0.24048364162445068,
|
|
"num_tokens": 17605033.0,
|
|
"step": 10020
|
|
},
|
|
{
|
|
"entropy": 5.100909566879272,
|
|
"epoch": 7.856583072100314,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010716324227955904,
|
|
"loss": 4.3401,
|
|
"mean_token_accuracy": 0.2406132325530052,
|
|
"num_tokens": 17613706.0,
|
|
"step": 10025
|
|
},
|
|
{
|
|
"entropy": 5.209266471862793,
|
|
"epoch": 7.860501567398119,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010696306208190034,
|
|
"loss": 4.4261,
|
|
"mean_token_accuracy": 0.23783658295869828,
|
|
"num_tokens": 17622082.0,
|
|
"step": 10030
|
|
},
|
|
{
|
|
"entropy": 5.074306535720825,
|
|
"epoch": 7.864420062695925,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010676318219349564,
|
|
"loss": 4.3029,
|
|
"mean_token_accuracy": 0.245320625603199,
|
|
"num_tokens": 17630980.0,
|
|
"step": 10035
|
|
},
|
|
{
|
|
"entropy": 5.080348539352417,
|
|
"epoch": 7.868338557993731,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010656360297156262,
|
|
"loss": 4.3285,
|
|
"mean_token_accuracy": 0.2459738716483116,
|
|
"num_tokens": 17639679.0,
|
|
"step": 10040
|
|
},
|
|
{
|
|
"entropy": 5.07864818572998,
|
|
"epoch": 7.872257053291536,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001063643247727818,
|
|
"loss": 4.2771,
|
|
"mean_token_accuracy": 0.2497877672314644,
|
|
"num_tokens": 17648542.0,
|
|
"step": 10045
|
|
},
|
|
{
|
|
"entropy": 5.111378145217896,
|
|
"epoch": 7.876175548589342,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010616534795329553,
|
|
"loss": 4.287,
|
|
"mean_token_accuracy": 0.24321618527173997,
|
|
"num_tokens": 17657305.0,
|
|
"step": 10050
|
|
},
|
|
{
|
|
"entropy": 5.17450704574585,
|
|
"epoch": 7.880094043887147,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010596667286870778,
|
|
"loss": 4.3956,
|
|
"mean_token_accuracy": 0.236778324842453,
|
|
"num_tokens": 17665876.0,
|
|
"step": 10055
|
|
},
|
|
{
|
|
"entropy": 5.144435119628906,
|
|
"epoch": 7.884012539184953,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010576829987408304,
|
|
"loss": 4.3213,
|
|
"mean_token_accuracy": 0.2526321724057198,
|
|
"num_tokens": 17674756.0,
|
|
"step": 10060
|
|
},
|
|
{
|
|
"entropy": 5.119069051742554,
|
|
"epoch": 7.887931034482759,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010557022932394611,
|
|
"loss": 4.3,
|
|
"mean_token_accuracy": 0.25992358922958375,
|
|
"num_tokens": 17683989.0,
|
|
"step": 10065
|
|
},
|
|
{
|
|
"entropy": 5.2391551494598385,
|
|
"epoch": 7.891849529780564,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010537246157228104,
|
|
"loss": 4.402,
|
|
"mean_token_accuracy": 0.2436270996928215,
|
|
"num_tokens": 17692776.0,
|
|
"step": 10070
|
|
},
|
|
{
|
|
"entropy": 5.144537019729614,
|
|
"epoch": 7.89576802507837,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010517499697253103,
|
|
"loss": 4.2624,
|
|
"mean_token_accuracy": 0.24970337450504304,
|
|
"num_tokens": 17701428.0,
|
|
"step": 10075
|
|
},
|
|
{
|
|
"entropy": 5.125144147872925,
|
|
"epoch": 7.899686520376176,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010497783587759724,
|
|
"loss": 4.3931,
|
|
"mean_token_accuracy": 0.23424016684293747,
|
|
"num_tokens": 17710343.0,
|
|
"step": 10080
|
|
},
|
|
{
|
|
"entropy": 5.151565742492676,
|
|
"epoch": 7.903605015673981,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010478097863983857,
|
|
"loss": 4.3895,
|
|
"mean_token_accuracy": 0.24054247587919236,
|
|
"num_tokens": 17719218.0,
|
|
"step": 10085
|
|
},
|
|
{
|
|
"entropy": 5.1868781566619875,
|
|
"epoch": 7.907523510971787,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001045844256110707,
|
|
"loss": 4.4444,
|
|
"mean_token_accuracy": 0.2267446607351303,
|
|
"num_tokens": 17727288.0,
|
|
"step": 10090
|
|
},
|
|
{
|
|
"entropy": 5.184895038604736,
|
|
"epoch": 7.911442006269592,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010438817714256589,
|
|
"loss": 4.4168,
|
|
"mean_token_accuracy": 0.23658441454172136,
|
|
"num_tokens": 17736130.0,
|
|
"step": 10095
|
|
},
|
|
{
|
|
"entropy": 5.112253618240357,
|
|
"epoch": 7.915360501567398,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010419223358505196,
|
|
"loss": 4.3009,
|
|
"mean_token_accuracy": 0.25216116309165953,
|
|
"num_tokens": 17745122.0,
|
|
"step": 10100
|
|
},
|
|
{
|
|
"entropy": 5.113187551498413,
|
|
"epoch": 7.919278996865204,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010399659528871169,
|
|
"loss": 4.3187,
|
|
"mean_token_accuracy": 0.23868532180786134,
|
|
"num_tokens": 17753817.0,
|
|
"step": 10105
|
|
},
|
|
{
|
|
"entropy": 5.190814399719239,
|
|
"epoch": 7.923197492163009,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010380126260318265,
|
|
"loss": 4.3942,
|
|
"mean_token_accuracy": 0.2371252715587616,
|
|
"num_tokens": 17762793.0,
|
|
"step": 10110
|
|
},
|
|
{
|
|
"entropy": 5.095741081237793,
|
|
"epoch": 7.927115987460815,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00010360623587755579,
|
|
"loss": 4.3463,
|
|
"mean_token_accuracy": 0.24920580983161927,
|
|
"num_tokens": 17772223.0,
|
|
"step": 10115
|
|
},
|
|
{
|
|
"entropy": 5.109301424026489,
|
|
"epoch": 7.931034482758621,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001034115154603757,
|
|
"loss": 4.3316,
|
|
"mean_token_accuracy": 0.23779338151216506,
|
|
"num_tokens": 17781543.0,
|
|
"step": 10120
|
|
},
|
|
{
|
|
"entropy": 5.117827844619751,
|
|
"epoch": 7.934952978056426,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010321710169963919,
|
|
"loss": 4.3515,
|
|
"mean_token_accuracy": 0.23903332501649857,
|
|
"num_tokens": 17790258.0,
|
|
"step": 10125
|
|
},
|
|
{
|
|
"entropy": 5.102209663391113,
|
|
"epoch": 7.938871473354232,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010302299494279527,
|
|
"loss": 4.3333,
|
|
"mean_token_accuracy": 0.2405731424689293,
|
|
"num_tokens": 17798862.0,
|
|
"step": 10130
|
|
},
|
|
{
|
|
"entropy": 5.134487438201904,
|
|
"epoch": 7.942789968652038,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010282919553674415,
|
|
"loss": 4.3869,
|
|
"mean_token_accuracy": 0.235950331389904,
|
|
"num_tokens": 17807285.0,
|
|
"step": 10135
|
|
},
|
|
{
|
|
"entropy": 5.146253824234009,
|
|
"epoch": 7.946708463949843,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010263570382783675,
|
|
"loss": 4.4192,
|
|
"mean_token_accuracy": 0.2321263551712036,
|
|
"num_tokens": 17816891.0,
|
|
"step": 10140
|
|
},
|
|
{
|
|
"entropy": 5.1723565578460695,
|
|
"epoch": 7.950626959247649,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00010244252016187419,
|
|
"loss": 4.3778,
|
|
"mean_token_accuracy": 0.24342295676469802,
|
|
"num_tokens": 17825895.0,
|
|
"step": 10145
|
|
},
|
|
{
|
|
"entropy": 5.197049760818482,
|
|
"epoch": 7.954545454545455,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00010224964488410699,
|
|
"loss": 4.4462,
|
|
"mean_token_accuracy": 0.23937027752399445,
|
|
"num_tokens": 17834921.0,
|
|
"step": 10150
|
|
},
|
|
{
|
|
"entropy": 5.199371242523194,
|
|
"epoch": 7.95846394984326,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001020570783392345,
|
|
"loss": 4.3391,
|
|
"mean_token_accuracy": 0.2457537516951561,
|
|
"num_tokens": 17844310.0,
|
|
"step": 10155
|
|
},
|
|
{
|
|
"entropy": 5.232431268692016,
|
|
"epoch": 7.962382445141066,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010186482087140439,
|
|
"loss": 4.3878,
|
|
"mean_token_accuracy": 0.24423097521066667,
|
|
"num_tokens": 17853561.0,
|
|
"step": 10160
|
|
},
|
|
{
|
|
"entropy": 5.104188060760498,
|
|
"epoch": 7.966300940438871,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000101672872824212,
|
|
"loss": 4.3295,
|
|
"mean_token_accuracy": 0.23699247539043428,
|
|
"num_tokens": 17863077.0,
|
|
"step": 10165
|
|
},
|
|
{
|
|
"entropy": 5.158885145187378,
|
|
"epoch": 7.970219435736677,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010148123454069943,
|
|
"loss": 4.2756,
|
|
"mean_token_accuracy": 0.25387916415929795,
|
|
"num_tokens": 17871780.0,
|
|
"step": 10170
|
|
},
|
|
{
|
|
"entropy": 5.1009547233581545,
|
|
"epoch": 7.974137931034483,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001012899063633555,
|
|
"loss": 4.3337,
|
|
"mean_token_accuracy": 0.2455006092786789,
|
|
"num_tokens": 17881144.0,
|
|
"step": 10175
|
|
},
|
|
{
|
|
"entropy": 5.11955304145813,
|
|
"epoch": 7.978056426332288,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010109888863411473,
|
|
"loss": 4.3809,
|
|
"mean_token_accuracy": 0.23911157995462418,
|
|
"num_tokens": 17890754.0,
|
|
"step": 10180
|
|
},
|
|
{
|
|
"entropy": 5.164124488830566,
|
|
"epoch": 7.981974921630094,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001009081816943567,
|
|
"loss": 4.4411,
|
|
"mean_token_accuracy": 0.23462337702512742,
|
|
"num_tokens": 17899992.0,
|
|
"step": 10185
|
|
},
|
|
{
|
|
"entropy": 5.086103010177612,
|
|
"epoch": 7.985893416927899,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010071778588490558,
|
|
"loss": 4.3243,
|
|
"mean_token_accuracy": 0.24716224819421767,
|
|
"num_tokens": 17907847.0,
|
|
"step": 10190
|
|
},
|
|
{
|
|
"entropy": 5.164031648635865,
|
|
"epoch": 7.989811912225705,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010052770154602968,
|
|
"loss": 4.3689,
|
|
"mean_token_accuracy": 0.24059428721666337,
|
|
"num_tokens": 17916566.0,
|
|
"step": 10195
|
|
},
|
|
{
|
|
"entropy": 5.123573303222656,
|
|
"epoch": 7.993730407523511,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010033792901744046,
|
|
"loss": 4.2589,
|
|
"mean_token_accuracy": 0.25082471668720246,
|
|
"num_tokens": 17924969.0,
|
|
"step": 10200
|
|
},
|
|
{
|
|
"entropy": 5.138540172576905,
|
|
"epoch": 7.997648902821316,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010014846863829221,
|
|
"loss": 4.2805,
|
|
"mean_token_accuracy": 0.24164951592683792,
|
|
"num_tokens": 17933776.0,
|
|
"step": 10205
|
|
},
|
|
{
|
|
"entropy": 5.086595916748047,
|
|
"epoch": 8.001567398119123,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.995932074718131e-05,
|
|
"loss": 4.2582,
|
|
"mean_token_accuracy": 0.24661437273025513,
|
|
"num_tokens": 17941651.0,
|
|
"step": 10210
|
|
},
|
|
{
|
|
"entropy": 5.158220529556274,
|
|
"epoch": 8.005485893416928,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.977048568214578e-05,
|
|
"loss": 4.388,
|
|
"mean_token_accuracy": 0.2424011066555977,
|
|
"num_tokens": 17951278.0,
|
|
"step": 10215
|
|
},
|
|
{
|
|
"entropy": 5.107753324508667,
|
|
"epoch": 8.009404388714733,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.958196378066444e-05,
|
|
"loss": 4.2977,
|
|
"mean_token_accuracy": 0.2439126566052437,
|
|
"num_tokens": 17960182.0,
|
|
"step": 10220
|
|
},
|
|
{
|
|
"entropy": 5.119613170623779,
|
|
"epoch": 8.013322884012538,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.93937553796565e-05,
|
|
"loss": 4.2592,
|
|
"mean_token_accuracy": 0.25174637585878373,
|
|
"num_tokens": 17968221.0,
|
|
"step": 10225
|
|
},
|
|
{
|
|
"entropy": 5.11217885017395,
|
|
"epoch": 8.017241379310345,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.920586081548089e-05,
|
|
"loss": 4.1798,
|
|
"mean_token_accuracy": 0.2544050425291061,
|
|
"num_tokens": 17975724.0,
|
|
"step": 10230
|
|
},
|
|
{
|
|
"entropy": 5.137327337265015,
|
|
"epoch": 8.02115987460815,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.901828042393573e-05,
|
|
"loss": 4.3747,
|
|
"mean_token_accuracy": 0.24488297700881959,
|
|
"num_tokens": 17984425.0,
|
|
"step": 10235
|
|
},
|
|
{
|
|
"entropy": 5.113169431686401,
|
|
"epoch": 8.025078369905955,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.88310145402575e-05,
|
|
"loss": 4.1955,
|
|
"mean_token_accuracy": 0.2508566856384277,
|
|
"num_tokens": 17993319.0,
|
|
"step": 10240
|
|
},
|
|
{
|
|
"entropy": 5.109821796417236,
|
|
"epoch": 8.028996865203762,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.864406349912081e-05,
|
|
"loss": 4.2338,
|
|
"mean_token_accuracy": 0.25495136231184007,
|
|
"num_tokens": 18002239.0,
|
|
"step": 10245
|
|
},
|
|
{
|
|
"entropy": 5.170929050445556,
|
|
"epoch": 8.032915360501567,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.845742763463739e-05,
|
|
"loss": 4.3286,
|
|
"mean_token_accuracy": 0.2484916016459465,
|
|
"num_tokens": 18010745.0,
|
|
"step": 10250
|
|
},
|
|
{
|
|
"entropy": 5.1247718334198,
|
|
"epoch": 8.036833855799372,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.827110728035584e-05,
|
|
"loss": 4.2961,
|
|
"mean_token_accuracy": 0.24692442566156386,
|
|
"num_tokens": 18019535.0,
|
|
"step": 10255
|
|
},
|
|
{
|
|
"entropy": 5.15582857131958,
|
|
"epoch": 8.04075235109718,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.808510276926075e-05,
|
|
"loss": 4.2889,
|
|
"mean_token_accuracy": 0.24930407702922822,
|
|
"num_tokens": 18029128.0,
|
|
"step": 10260
|
|
},
|
|
{
|
|
"entropy": 5.158373689651489,
|
|
"epoch": 8.044670846394984,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.789941443377249e-05,
|
|
"loss": 4.321,
|
|
"mean_token_accuracy": 0.2462007537484169,
|
|
"num_tokens": 18038304.0,
|
|
"step": 10265
|
|
},
|
|
{
|
|
"entropy": 5.1370209693908695,
|
|
"epoch": 8.04858934169279,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.771404260574613e-05,
|
|
"loss": 4.2686,
|
|
"mean_token_accuracy": 0.25358373671770096,
|
|
"num_tokens": 18046628.0,
|
|
"step": 10270
|
|
},
|
|
{
|
|
"entropy": 5.201504516601562,
|
|
"epoch": 8.052507836990596,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.752898761647113e-05,
|
|
"loss": 4.3527,
|
|
"mean_token_accuracy": 0.2406791180372238,
|
|
"num_tokens": 18055654.0,
|
|
"step": 10275
|
|
},
|
|
{
|
|
"entropy": 5.13857946395874,
|
|
"epoch": 8.056426332288401,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.734424979667089e-05,
|
|
"loss": 4.2898,
|
|
"mean_token_accuracy": 0.2481779634952545,
|
|
"num_tokens": 18065844.0,
|
|
"step": 10280
|
|
},
|
|
{
|
|
"entropy": 5.162217426300049,
|
|
"epoch": 8.060344827586206,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.715982947650179e-05,
|
|
"loss": 4.3221,
|
|
"mean_token_accuracy": 0.23832941353321074,
|
|
"num_tokens": 18074497.0,
|
|
"step": 10285
|
|
},
|
|
{
|
|
"entropy": 5.2129807472229,
|
|
"epoch": 8.064263322884013,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.697572698555278e-05,
|
|
"loss": 4.3231,
|
|
"mean_token_accuracy": 0.24593500196933746,
|
|
"num_tokens": 18083967.0,
|
|
"step": 10290
|
|
},
|
|
{
|
|
"entropy": 5.14147310256958,
|
|
"epoch": 8.068181818181818,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.679194265284492e-05,
|
|
"loss": 4.3109,
|
|
"mean_token_accuracy": 0.2532788813114166,
|
|
"num_tokens": 18092263.0,
|
|
"step": 10295
|
|
},
|
|
{
|
|
"entropy": 5.241281414031983,
|
|
"epoch": 8.072100313479623,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.660847680683069e-05,
|
|
"loss": 4.3716,
|
|
"mean_token_accuracy": 0.24825457036495208,
|
|
"num_tokens": 18101559.0,
|
|
"step": 10300
|
|
},
|
|
{
|
|
"entropy": 5.209451913833618,
|
|
"epoch": 8.07601880877743,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.64253297753932e-05,
|
|
"loss": 4.2911,
|
|
"mean_token_accuracy": 0.2498155266046524,
|
|
"num_tokens": 18109850.0,
|
|
"step": 10305
|
|
},
|
|
{
|
|
"entropy": 5.157417106628418,
|
|
"epoch": 8.079937304075235,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.62425018858459e-05,
|
|
"loss": 4.2707,
|
|
"mean_token_accuracy": 0.24812538474798201,
|
|
"num_tokens": 18119205.0,
|
|
"step": 10310
|
|
},
|
|
{
|
|
"entropy": 5.132119941711426,
|
|
"epoch": 8.08385579937304,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.605999346493194e-05,
|
|
"loss": 4.2519,
|
|
"mean_token_accuracy": 0.248558409512043,
|
|
"num_tokens": 18128713.0,
|
|
"step": 10315
|
|
},
|
|
{
|
|
"entropy": 5.138487958908081,
|
|
"epoch": 8.087774294670846,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 9.587780483882339e-05,
|
|
"loss": 4.2341,
|
|
"mean_token_accuracy": 0.2532330170273781,
|
|
"num_tokens": 18137080.0,
|
|
"step": 10320
|
|
},
|
|
{
|
|
"entropy": 5.184178113937378,
|
|
"epoch": 8.091692789968652,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.569593633312084e-05,
|
|
"loss": 4.367,
|
|
"mean_token_accuracy": 0.24271986931562423,
|
|
"num_tokens": 18146449.0,
|
|
"step": 10325
|
|
},
|
|
{
|
|
"entropy": 5.183739280700683,
|
|
"epoch": 8.095611285266457,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.551438827285294e-05,
|
|
"loss": 4.3807,
|
|
"mean_token_accuracy": 0.23422101736068726,
|
|
"num_tokens": 18155394.0,
|
|
"step": 10330
|
|
},
|
|
{
|
|
"entropy": 5.188652229309082,
|
|
"epoch": 8.099529780564263,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.533316098247539e-05,
|
|
"loss": 4.3817,
|
|
"mean_token_accuracy": 0.23140805810689927,
|
|
"num_tokens": 18164538.0,
|
|
"step": 10335
|
|
},
|
|
{
|
|
"entropy": 5.168311643600464,
|
|
"epoch": 8.10344827586207,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.515225478587077e-05,
|
|
"loss": 4.3051,
|
|
"mean_token_accuracy": 0.2514050602912903,
|
|
"num_tokens": 18174396.0,
|
|
"step": 10340
|
|
},
|
|
{
|
|
"entropy": 5.184533643722534,
|
|
"epoch": 8.107366771159874,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.497167000634773e-05,
|
|
"loss": 4.2735,
|
|
"mean_token_accuracy": 0.24658851325511932,
|
|
"num_tokens": 18183030.0,
|
|
"step": 10345
|
|
},
|
|
{
|
|
"entropy": 5.139373683929444,
|
|
"epoch": 8.11128526645768,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.47914069666406e-05,
|
|
"loss": 4.305,
|
|
"mean_token_accuracy": 0.24609907120466232,
|
|
"num_tokens": 18191139.0,
|
|
"step": 10350
|
|
},
|
|
{
|
|
"entropy": 5.097940349578858,
|
|
"epoch": 8.115203761755486,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.461146598890873e-05,
|
|
"loss": 4.2447,
|
|
"mean_token_accuracy": 0.2530363991856575,
|
|
"num_tokens": 18199429.0,
|
|
"step": 10355
|
|
},
|
|
{
|
|
"entropy": 5.154330778121948,
|
|
"epoch": 8.119122257053291,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.443184739473573e-05,
|
|
"loss": 4.3135,
|
|
"mean_token_accuracy": 0.2474463552236557,
|
|
"num_tokens": 18207765.0,
|
|
"step": 10360
|
|
},
|
|
{
|
|
"entropy": 5.11970329284668,
|
|
"epoch": 8.123040752351097,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.425255150512921e-05,
|
|
"loss": 4.3384,
|
|
"mean_token_accuracy": 0.24743917286396028,
|
|
"num_tokens": 18216524.0,
|
|
"step": 10365
|
|
},
|
|
{
|
|
"entropy": 5.174101829528809,
|
|
"epoch": 8.126959247648903,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.407357864051996e-05,
|
|
"loss": 4.3058,
|
|
"mean_token_accuracy": 0.24716053903102875,
|
|
"num_tokens": 18224647.0,
|
|
"step": 10370
|
|
},
|
|
{
|
|
"entropy": 5.123722696304322,
|
|
"epoch": 8.130877742946709,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.389492912076156e-05,
|
|
"loss": 4.3145,
|
|
"mean_token_accuracy": 0.24552964568138122,
|
|
"num_tokens": 18234127.0,
|
|
"step": 10375
|
|
},
|
|
{
|
|
"entropy": 5.135565519332886,
|
|
"epoch": 8.134796238244514,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.371660326512952e-05,
|
|
"loss": 4.2401,
|
|
"mean_token_accuracy": 0.25146869122982024,
|
|
"num_tokens": 18242455.0,
|
|
"step": 10380
|
|
},
|
|
{
|
|
"entropy": 5.1871881008148195,
|
|
"epoch": 8.13871473354232,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 9.353860139232126e-05,
|
|
"loss": 4.3461,
|
|
"mean_token_accuracy": 0.24084988236427307,
|
|
"num_tokens": 18251201.0,
|
|
"step": 10385
|
|
},
|
|
{
|
|
"entropy": 5.156402349472046,
|
|
"epoch": 8.142633228840126,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.336092382045488e-05,
|
|
"loss": 4.28,
|
|
"mean_token_accuracy": 0.24270834326744078,
|
|
"num_tokens": 18260651.0,
|
|
"step": 10390
|
|
},
|
|
{
|
|
"entropy": 5.2094522476196286,
|
|
"epoch": 8.14655172413793,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.318357086706894e-05,
|
|
"loss": 4.3628,
|
|
"mean_token_accuracy": 0.24292825162410736,
|
|
"num_tokens": 18269218.0,
|
|
"step": 10395
|
|
},
|
|
{
|
|
"entropy": 5.166447734832763,
|
|
"epoch": 8.150470219435737,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.300654284912211e-05,
|
|
"loss": 4.2899,
|
|
"mean_token_accuracy": 0.24695176184177398,
|
|
"num_tokens": 18277371.0,
|
|
"step": 10400
|
|
},
|
|
{
|
|
"entropy": 5.107787132263184,
|
|
"epoch": 8.154388714733543,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.282984008299196e-05,
|
|
"loss": 4.3314,
|
|
"mean_token_accuracy": 0.24110763072967528,
|
|
"num_tokens": 18287503.0,
|
|
"step": 10405
|
|
},
|
|
{
|
|
"entropy": 5.134560489654541,
|
|
"epoch": 8.158307210031348,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 9.265346288447516e-05,
|
|
"loss": 4.3199,
|
|
"mean_token_accuracy": 0.24310127943754195,
|
|
"num_tokens": 18296753.0,
|
|
"step": 10410
|
|
},
|
|
{
|
|
"entropy": 5.169224405288697,
|
|
"epoch": 8.162225705329154,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.24774115687862e-05,
|
|
"loss": 4.3396,
|
|
"mean_token_accuracy": 0.2462587609887123,
|
|
"num_tokens": 18305068.0,
|
|
"step": 10415
|
|
},
|
|
{
|
|
"entropy": 5.17865219116211,
|
|
"epoch": 8.16614420062696,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.23016864505575e-05,
|
|
"loss": 4.3376,
|
|
"mean_token_accuracy": 0.24545031636953354,
|
|
"num_tokens": 18313816.0,
|
|
"step": 10420
|
|
},
|
|
{
|
|
"entropy": 5.153687858581543,
|
|
"epoch": 8.170062695924765,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.212628784383825e-05,
|
|
"loss": 4.2601,
|
|
"mean_token_accuracy": 0.2548660829663277,
|
|
"num_tokens": 18322630.0,
|
|
"step": 10425
|
|
},
|
|
{
|
|
"entropy": 5.11886191368103,
|
|
"epoch": 8.17398119122257,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.195121606209411e-05,
|
|
"loss": 4.2561,
|
|
"mean_token_accuracy": 0.24716718196868898,
|
|
"num_tokens": 18330942.0,
|
|
"step": 10430
|
|
},
|
|
{
|
|
"entropy": 5.105663061141968,
|
|
"epoch": 8.177899686520377,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.177647141820693e-05,
|
|
"loss": 4.3172,
|
|
"mean_token_accuracy": 0.2458041191101074,
|
|
"num_tokens": 18339375.0,
|
|
"step": 10435
|
|
},
|
|
{
|
|
"entropy": 5.2007852554321286,
|
|
"epoch": 8.181818181818182,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.160205422447357e-05,
|
|
"loss": 4.3229,
|
|
"mean_token_accuracy": 0.24289220720529556,
|
|
"num_tokens": 18347884.0,
|
|
"step": 10440
|
|
},
|
|
{
|
|
"entropy": 5.124487638473511,
|
|
"epoch": 8.185736677115987,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.142796479260588e-05,
|
|
"loss": 4.3169,
|
|
"mean_token_accuracy": 0.24610556066036224,
|
|
"num_tokens": 18357201.0,
|
|
"step": 10445
|
|
},
|
|
{
|
|
"entropy": 5.173965930938721,
|
|
"epoch": 8.189655172413794,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.125420343372995e-05,
|
|
"loss": 4.3694,
|
|
"mean_token_accuracy": 0.2421983227133751,
|
|
"num_tokens": 18365716.0,
|
|
"step": 10450
|
|
},
|
|
{
|
|
"entropy": 5.124088001251221,
|
|
"epoch": 8.193573667711599,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.108077045838546e-05,
|
|
"loss": 4.2886,
|
|
"mean_token_accuracy": 0.25823282152414323,
|
|
"num_tokens": 18374724.0,
|
|
"step": 10455
|
|
},
|
|
{
|
|
"entropy": 5.151387071609497,
|
|
"epoch": 8.197492163009404,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.090766617652525e-05,
|
|
"loss": 4.3231,
|
|
"mean_token_accuracy": 0.24793402552604676,
|
|
"num_tokens": 18383653.0,
|
|
"step": 10460
|
|
},
|
|
{
|
|
"entropy": 5.171075487136841,
|
|
"epoch": 8.20141065830721,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.073489089751468e-05,
|
|
"loss": 4.2996,
|
|
"mean_token_accuracy": 0.25093998163938525,
|
|
"num_tokens": 18392729.0,
|
|
"step": 10465
|
|
},
|
|
{
|
|
"entropy": 5.173304033279419,
|
|
"epoch": 8.205329153605016,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.056244493013129e-05,
|
|
"loss": 4.3408,
|
|
"mean_token_accuracy": 0.23575303107500076,
|
|
"num_tokens": 18402257.0,
|
|
"step": 10470
|
|
},
|
|
{
|
|
"entropy": 5.154960870742798,
|
|
"epoch": 8.20924764890282,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.039032858256401e-05,
|
|
"loss": 4.3479,
|
|
"mean_token_accuracy": 0.2458760157227516,
|
|
"num_tokens": 18410714.0,
|
|
"step": 10475
|
|
},
|
|
{
|
|
"entropy": 5.162651014328003,
|
|
"epoch": 8.213166144200628,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.021854216241251e-05,
|
|
"loss": 4.3616,
|
|
"mean_token_accuracy": 0.2427298590540886,
|
|
"num_tokens": 18419256.0,
|
|
"step": 10480
|
|
},
|
|
{
|
|
"entropy": 5.175302505493164,
|
|
"epoch": 8.217084639498433,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.004708597668716e-05,
|
|
"loss": 4.32,
|
|
"mean_token_accuracy": 0.24969494491815566,
|
|
"num_tokens": 18428202.0,
|
|
"step": 10485
|
|
},
|
|
{
|
|
"entropy": 5.1682408332824705,
|
|
"epoch": 8.221003134796238,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.987596033180784e-05,
|
|
"loss": 4.2768,
|
|
"mean_token_accuracy": 0.24557400792837142,
|
|
"num_tokens": 18437339.0,
|
|
"step": 10490
|
|
},
|
|
{
|
|
"entropy": 5.1663188457489015,
|
|
"epoch": 8.224921630094045,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.970516553360383e-05,
|
|
"loss": 4.3099,
|
|
"mean_token_accuracy": 0.24497049450874328,
|
|
"num_tokens": 18446667.0,
|
|
"step": 10495
|
|
},
|
|
{
|
|
"entropy": 5.145668888092041,
|
|
"epoch": 8.22884012539185,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.953470188731318e-05,
|
|
"loss": 4.244,
|
|
"mean_token_accuracy": 0.2506525546312332,
|
|
"num_tokens": 18455425.0,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"epoch": 8.22884012539185,
|
|
"eval_entropy": 5.073418850122496,
|
|
"eval_loss": 5.650219440460205,
|
|
"eval_mean_token_accuracy": 0.17383484948108824,
|
|
"eval_num_tokens": 18455425.0,
|
|
"eval_runtime": 3.0342,
|
|
"eval_samples_per_second": 1358.51,
|
|
"eval_steps_per_second": 170.061,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"entropy": 5.1344342708587645,
|
|
"epoch": 8.232758620689655,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.936456969758206e-05,
|
|
"loss": 4.2292,
|
|
"mean_token_accuracy": 0.25337707847356794,
|
|
"num_tokens": 18464976.0,
|
|
"step": 10505
|
|
},
|
|
{
|
|
"entropy": 5.1826269149780275,
|
|
"epoch": 8.236677115987462,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.919476926846426e-05,
|
|
"loss": 4.2758,
|
|
"mean_token_accuracy": 0.2515713036060333,
|
|
"num_tokens": 18473709.0,
|
|
"step": 10510
|
|
},
|
|
{
|
|
"entropy": 5.166347360610962,
|
|
"epoch": 8.240595611285267,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.90253009034206e-05,
|
|
"loss": 4.2999,
|
|
"mean_token_accuracy": 0.24776636809110641,
|
|
"num_tokens": 18483208.0,
|
|
"step": 10515
|
|
},
|
|
{
|
|
"entropy": 5.212204027175903,
|
|
"epoch": 8.244514106583072,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.88561649053187e-05,
|
|
"loss": 4.3641,
|
|
"mean_token_accuracy": 0.24391481280326843,
|
|
"num_tokens": 18491872.0,
|
|
"step": 10520
|
|
},
|
|
{
|
|
"entropy": 5.2012434005737305,
|
|
"epoch": 8.248432601880877,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.868736157643183e-05,
|
|
"loss": 4.3898,
|
|
"mean_token_accuracy": 0.2355497196316719,
|
|
"num_tokens": 18500740.0,
|
|
"step": 10525
|
|
},
|
|
{
|
|
"entropy": 5.15441312789917,
|
|
"epoch": 8.252351097178684,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.851889121843904e-05,
|
|
"loss": 4.2924,
|
|
"mean_token_accuracy": 0.24854207187891006,
|
|
"num_tokens": 18509553.0,
|
|
"step": 10530
|
|
},
|
|
{
|
|
"entropy": 5.214558839797974,
|
|
"epoch": 8.256269592476489,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 8.835075413242408e-05,
|
|
"loss": 4.3664,
|
|
"mean_token_accuracy": 0.24509138166904448,
|
|
"num_tokens": 18518549.0,
|
|
"step": 10535
|
|
},
|
|
{
|
|
"entropy": 5.19227933883667,
|
|
"epoch": 8.260188087774294,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.818295061887531e-05,
|
|
"loss": 4.3391,
|
|
"mean_token_accuracy": 0.23471049666404725,
|
|
"num_tokens": 18527264.0,
|
|
"step": 10540
|
|
},
|
|
{
|
|
"entropy": 5.131226682662964,
|
|
"epoch": 8.2641065830721,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.801548097768473e-05,
|
|
"loss": 4.2926,
|
|
"mean_token_accuracy": 0.24918805062770844,
|
|
"num_tokens": 18535193.0,
|
|
"step": 10545
|
|
},
|
|
{
|
|
"entropy": 5.161487293243408,
|
|
"epoch": 8.268025078369906,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.784834550814769e-05,
|
|
"loss": 4.3215,
|
|
"mean_token_accuracy": 0.24534369260072708,
|
|
"num_tokens": 18544386.0,
|
|
"step": 10550
|
|
},
|
|
{
|
|
"entropy": 5.125006866455078,
|
|
"epoch": 8.271943573667711,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.768154450896253e-05,
|
|
"loss": 4.1822,
|
|
"mean_token_accuracy": 0.2571590572595596,
|
|
"num_tokens": 18552745.0,
|
|
"step": 10555
|
|
},
|
|
{
|
|
"entropy": 5.184374523162842,
|
|
"epoch": 8.275862068965518,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.751507827822955e-05,
|
|
"loss": 4.2879,
|
|
"mean_token_accuracy": 0.24628626555204391,
|
|
"num_tokens": 18560953.0,
|
|
"step": 10560
|
|
},
|
|
{
|
|
"entropy": 5.183347415924072,
|
|
"epoch": 8.279780564263323,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.73489471134509e-05,
|
|
"loss": 4.2253,
|
|
"mean_token_accuracy": 0.2634794443845749,
|
|
"num_tokens": 18569324.0,
|
|
"step": 10565
|
|
},
|
|
{
|
|
"entropy": 5.221614646911621,
|
|
"epoch": 8.283699059561128,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.718315131152999e-05,
|
|
"loss": 4.3818,
|
|
"mean_token_accuracy": 0.24350043386220932,
|
|
"num_tokens": 18578432.0,
|
|
"step": 10570
|
|
},
|
|
{
|
|
"entropy": 5.1335474967956545,
|
|
"epoch": 8.287617554858935,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.701769116877076e-05,
|
|
"loss": 4.2989,
|
|
"mean_token_accuracy": 0.2472938597202301,
|
|
"num_tokens": 18587147.0,
|
|
"step": 10575
|
|
},
|
|
{
|
|
"entropy": 5.209079265594482,
|
|
"epoch": 8.29153605015674,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 8.685256698087725e-05,
|
|
"loss": 4.3083,
|
|
"mean_token_accuracy": 0.24362783879041672,
|
|
"num_tokens": 18596933.0,
|
|
"step": 10580
|
|
},
|
|
{
|
|
"entropy": 5.227485275268554,
|
|
"epoch": 8.295454545454545,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.668777904295328e-05,
|
|
"loss": 4.3693,
|
|
"mean_token_accuracy": 0.24524417221546174,
|
|
"num_tokens": 18605666.0,
|
|
"step": 10585
|
|
},
|
|
{
|
|
"entropy": 5.196883201599121,
|
|
"epoch": 8.299373040752352,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.652332764950147e-05,
|
|
"loss": 4.443,
|
|
"mean_token_accuracy": 0.2375338762998581,
|
|
"num_tokens": 18614801.0,
|
|
"step": 10590
|
|
},
|
|
{
|
|
"entropy": 5.172837924957276,
|
|
"epoch": 8.303291536050157,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.635921309442327e-05,
|
|
"loss": 4.2945,
|
|
"mean_token_accuracy": 0.24749386459589004,
|
|
"num_tokens": 18623489.0,
|
|
"step": 10595
|
|
},
|
|
{
|
|
"entropy": 5.17074499130249,
|
|
"epoch": 8.307210031347962,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.619543567101787e-05,
|
|
"loss": 4.2965,
|
|
"mean_token_accuracy": 0.2478623941540718,
|
|
"num_tokens": 18631496.0,
|
|
"step": 10600
|
|
},
|
|
{
|
|
"entropy": 5.1863258361816404,
|
|
"epoch": 8.311128526645769,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.603199567198222e-05,
|
|
"loss": 4.3406,
|
|
"mean_token_accuracy": 0.24548270404338837,
|
|
"num_tokens": 18640481.0,
|
|
"step": 10605
|
|
},
|
|
{
|
|
"entropy": 5.202200126647949,
|
|
"epoch": 8.315047021943574,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.586889338941002e-05,
|
|
"loss": 4.4046,
|
|
"mean_token_accuracy": 0.23673862665891648,
|
|
"num_tokens": 18650118.0,
|
|
"step": 10610
|
|
},
|
|
{
|
|
"entropy": 5.119663381576538,
|
|
"epoch": 8.318965517241379,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.570612911479145e-05,
|
|
"loss": 4.3026,
|
|
"mean_token_accuracy": 0.24832355976104736,
|
|
"num_tokens": 18658981.0,
|
|
"step": 10615
|
|
},
|
|
{
|
|
"entropy": 5.1835321426391605,
|
|
"epoch": 8.322884012539184,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.554370313901273e-05,
|
|
"loss": 4.3626,
|
|
"mean_token_accuracy": 0.24960713237524032,
|
|
"num_tokens": 18668324.0,
|
|
"step": 10620
|
|
},
|
|
{
|
|
"entropy": 5.139247941970825,
|
|
"epoch": 8.326802507836991,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.538161575235541e-05,
|
|
"loss": 4.2498,
|
|
"mean_token_accuracy": 0.25600864440202714,
|
|
"num_tokens": 18676945.0,
|
|
"step": 10625
|
|
},
|
|
{
|
|
"entropy": 5.220499753952026,
|
|
"epoch": 8.330721003134796,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.521986724449593e-05,
|
|
"loss": 4.355,
|
|
"mean_token_accuracy": 0.24304592311382295,
|
|
"num_tokens": 18685980.0,
|
|
"step": 10630
|
|
},
|
|
{
|
|
"entropy": 5.182883787155151,
|
|
"epoch": 8.334639498432601,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.5058457904505e-05,
|
|
"loss": 4.3488,
|
|
"mean_token_accuracy": 0.2412552371621132,
|
|
"num_tokens": 18694834.0,
|
|
"step": 10635
|
|
},
|
|
{
|
|
"entropy": 5.114372730255127,
|
|
"epoch": 8.338557993730408,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.489738802084742e-05,
|
|
"loss": 4.3005,
|
|
"mean_token_accuracy": 0.2426683634519577,
|
|
"num_tokens": 18703747.0,
|
|
"step": 10640
|
|
},
|
|
{
|
|
"entropy": 5.132203149795532,
|
|
"epoch": 8.342476489028213,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.473665788138112e-05,
|
|
"loss": 4.278,
|
|
"mean_token_accuracy": 0.2501032218337059,
|
|
"num_tokens": 18712202.0,
|
|
"step": 10645
|
|
},
|
|
{
|
|
"entropy": 5.132572603225708,
|
|
"epoch": 8.346394984326018,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.457626777335691e-05,
|
|
"loss": 4.3077,
|
|
"mean_token_accuracy": 0.24737530946731567,
|
|
"num_tokens": 18721340.0,
|
|
"step": 10650
|
|
},
|
|
{
|
|
"entropy": 5.13887848854065,
|
|
"epoch": 8.350313479623825,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.4416217983418e-05,
|
|
"loss": 4.3465,
|
|
"mean_token_accuracy": 0.2420864522457123,
|
|
"num_tokens": 18730295.0,
|
|
"step": 10655
|
|
},
|
|
{
|
|
"entropy": 5.138090991973877,
|
|
"epoch": 8.35423197492163,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.425650879759923e-05,
|
|
"loss": 4.2832,
|
|
"mean_token_accuracy": 0.24162434935569763,
|
|
"num_tokens": 18739040.0,
|
|
"step": 10660
|
|
},
|
|
{
|
|
"entropy": 5.076887226104736,
|
|
"epoch": 8.358150470219435,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.409714050132683e-05,
|
|
"loss": 4.2217,
|
|
"mean_token_accuracy": 0.26241556107997893,
|
|
"num_tokens": 18747655.0,
|
|
"step": 10665
|
|
},
|
|
{
|
|
"entropy": 5.155777359008789,
|
|
"epoch": 8.362068965517242,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.393811337941781e-05,
|
|
"loss": 4.2672,
|
|
"mean_token_accuracy": 0.2527438119053841,
|
|
"num_tokens": 18755602.0,
|
|
"step": 10670
|
|
},
|
|
{
|
|
"entropy": 5.122378492355347,
|
|
"epoch": 8.365987460815047,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.377942771607943e-05,
|
|
"loss": 4.2757,
|
|
"mean_token_accuracy": 0.25667230039834976,
|
|
"num_tokens": 18764332.0,
|
|
"step": 10675
|
|
},
|
|
{
|
|
"entropy": 5.120349645614624,
|
|
"epoch": 8.369905956112852,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.362108379490868e-05,
|
|
"loss": 4.3287,
|
|
"mean_token_accuracy": 0.24209575355052948,
|
|
"num_tokens": 18773106.0,
|
|
"step": 10680
|
|
},
|
|
{
|
|
"entropy": 5.148451280593872,
|
|
"epoch": 8.373824451410659,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.346308189889175e-05,
|
|
"loss": 4.3463,
|
|
"mean_token_accuracy": 0.24066369831562043,
|
|
"num_tokens": 18781603.0,
|
|
"step": 10685
|
|
},
|
|
{
|
|
"entropy": 5.125776290893555,
|
|
"epoch": 8.377742946708464,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.330542231040379e-05,
|
|
"loss": 4.2555,
|
|
"mean_token_accuracy": 0.25770496428012846,
|
|
"num_tokens": 18790152.0,
|
|
"step": 10690
|
|
},
|
|
{
|
|
"entropy": 5.175959348678589,
|
|
"epoch": 8.38166144200627,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.314810531120795e-05,
|
|
"loss": 4.243,
|
|
"mean_token_accuracy": 0.258194737136364,
|
|
"num_tokens": 18798397.0,
|
|
"step": 10695
|
|
},
|
|
{
|
|
"entropy": 5.131862354278565,
|
|
"epoch": 8.385579937304076,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.299113118245519e-05,
|
|
"loss": 4.2941,
|
|
"mean_token_accuracy": 0.24984417110681534,
|
|
"num_tokens": 18807523.0,
|
|
"step": 10700
|
|
},
|
|
{
|
|
"entropy": 5.1447326183319095,
|
|
"epoch": 8.389498432601881,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.283450020468377e-05,
|
|
"loss": 4.3057,
|
|
"mean_token_accuracy": 0.24783380180597306,
|
|
"num_tokens": 18816469.0,
|
|
"step": 10705
|
|
},
|
|
{
|
|
"entropy": 5.21796178817749,
|
|
"epoch": 8.393416927899686,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.267821265781875e-05,
|
|
"loss": 4.3273,
|
|
"mean_token_accuracy": 0.24533143341541291,
|
|
"num_tokens": 18823994.0,
|
|
"step": 10710
|
|
},
|
|
{
|
|
"entropy": 5.117038583755493,
|
|
"epoch": 8.397335423197493,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.252226882117118e-05,
|
|
"loss": 4.2391,
|
|
"mean_token_accuracy": 0.2543315142393112,
|
|
"num_tokens": 18833159.0,
|
|
"step": 10715
|
|
},
|
|
{
|
|
"entropy": 5.100088024139405,
|
|
"epoch": 8.401253918495298,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.236666897343797e-05,
|
|
"loss": 4.2288,
|
|
"mean_token_accuracy": 0.2513937935233116,
|
|
"num_tokens": 18841754.0,
|
|
"step": 10720
|
|
},
|
|
{
|
|
"entropy": 5.11743392944336,
|
|
"epoch": 8.405172413793103,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.221141339270138e-05,
|
|
"loss": 4.28,
|
|
"mean_token_accuracy": 0.2580964684486389,
|
|
"num_tokens": 18850493.0,
|
|
"step": 10725
|
|
},
|
|
{
|
|
"entropy": 5.190040206909179,
|
|
"epoch": 8.409090909090908,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.205650235642828e-05,
|
|
"loss": 4.3572,
|
|
"mean_token_accuracy": 0.2435516208410263,
|
|
"num_tokens": 18859765.0,
|
|
"step": 10730
|
|
},
|
|
{
|
|
"entropy": 5.147920608520508,
|
|
"epoch": 8.413009404388715,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.190193614146971e-05,
|
|
"loss": 4.3401,
|
|
"mean_token_accuracy": 0.2443802237510681,
|
|
"num_tokens": 18868685.0,
|
|
"step": 10735
|
|
},
|
|
{
|
|
"entropy": 5.125345945358276,
|
|
"epoch": 8.41692789968652,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.17477150240607e-05,
|
|
"loss": 4.278,
|
|
"mean_token_accuracy": 0.25237028747797013,
|
|
"num_tokens": 18877496.0,
|
|
"step": 10740
|
|
},
|
|
{
|
|
"entropy": 5.144855737686157,
|
|
"epoch": 8.420846394984325,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.159383927981931e-05,
|
|
"loss": 4.2904,
|
|
"mean_token_accuracy": 0.25125064700841904,
|
|
"num_tokens": 18885828.0,
|
|
"step": 10745
|
|
},
|
|
{
|
|
"entropy": 5.1991126537323,
|
|
"epoch": 8.424764890282132,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.14403091837464e-05,
|
|
"loss": 4.3745,
|
|
"mean_token_accuracy": 0.23551496863365173,
|
|
"num_tokens": 18894498.0,
|
|
"step": 10750
|
|
},
|
|
{
|
|
"entropy": 5.170447778701782,
|
|
"epoch": 8.428683385579937,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.128712501022526e-05,
|
|
"loss": 4.3222,
|
|
"mean_token_accuracy": 0.2466443046927452,
|
|
"num_tokens": 18902996.0,
|
|
"step": 10755
|
|
},
|
|
{
|
|
"entropy": 5.115683841705322,
|
|
"epoch": 8.432601880877742,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.11342870330208e-05,
|
|
"loss": 4.2489,
|
|
"mean_token_accuracy": 0.2581145793199539,
|
|
"num_tokens": 18911731.0,
|
|
"step": 10760
|
|
},
|
|
{
|
|
"entropy": 5.096177053451538,
|
|
"epoch": 8.43652037617555,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.098179552527921e-05,
|
|
"loss": 4.2163,
|
|
"mean_token_accuracy": 0.2612070143222809,
|
|
"num_tokens": 18920912.0,
|
|
"step": 10765
|
|
},
|
|
{
|
|
"entropy": 5.111154317855835,
|
|
"epoch": 8.440438871473354,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 8.082965075952757e-05,
|
|
"loss": 4.2631,
|
|
"mean_token_accuracy": 0.25049475729465487,
|
|
"num_tokens": 18930012.0,
|
|
"step": 10770
|
|
},
|
|
{
|
|
"entropy": 5.149368190765381,
|
|
"epoch": 8.44435736677116,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.067785300767333e-05,
|
|
"loss": 4.3905,
|
|
"mean_token_accuracy": 0.23546600937843323,
|
|
"num_tokens": 18939026.0,
|
|
"step": 10775
|
|
},
|
|
{
|
|
"entropy": 5.177121353149414,
|
|
"epoch": 8.448275862068966,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.05264025410036e-05,
|
|
"loss": 4.3189,
|
|
"mean_token_accuracy": 0.2359721392393112,
|
|
"num_tokens": 18948387.0,
|
|
"step": 10780
|
|
},
|
|
{
|
|
"entropy": 5.174796056747437,
|
|
"epoch": 8.452194357366771,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.037529963018488e-05,
|
|
"loss": 4.36,
|
|
"mean_token_accuracy": 0.2367745503783226,
|
|
"num_tokens": 18957234.0,
|
|
"step": 10785
|
|
},
|
|
{
|
|
"entropy": 5.13149585723877,
|
|
"epoch": 8.456112852664576,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 8.022454454526272e-05,
|
|
"loss": 4.2879,
|
|
"mean_token_accuracy": 0.25240783840417863,
|
|
"num_tokens": 18965501.0,
|
|
"step": 10790
|
|
},
|
|
{
|
|
"entropy": 5.126470518112183,
|
|
"epoch": 8.460031347962383,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.00741375556608e-05,
|
|
"loss": 4.2896,
|
|
"mean_token_accuracy": 0.24570735394954682,
|
|
"num_tokens": 18973225.0,
|
|
"step": 10795
|
|
},
|
|
{
|
|
"entropy": 5.165433549880982,
|
|
"epoch": 8.463949843260188,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.99240789301808e-05,
|
|
"loss": 4.3051,
|
|
"mean_token_accuracy": 0.24470762461423873,
|
|
"num_tokens": 18982424.0,
|
|
"step": 10800
|
|
},
|
|
{
|
|
"entropy": 5.140762233734131,
|
|
"epoch": 8.467868338557993,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.977436893700177e-05,
|
|
"loss": 4.3546,
|
|
"mean_token_accuracy": 0.23737767338752747,
|
|
"num_tokens": 18991519.0,
|
|
"step": 10805
|
|
},
|
|
{
|
|
"entropy": 5.119337844848633,
|
|
"epoch": 8.4717868338558,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.962500784367985e-05,
|
|
"loss": 4.2963,
|
|
"mean_token_accuracy": 0.24609725922346115,
|
|
"num_tokens": 19000498.0,
|
|
"step": 10810
|
|
},
|
|
{
|
|
"entropy": 5.199625682830811,
|
|
"epoch": 8.475705329153605,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.947599591714746e-05,
|
|
"loss": 4.3871,
|
|
"mean_token_accuracy": 0.23921813666820527,
|
|
"num_tokens": 19009640.0,
|
|
"step": 10815
|
|
},
|
|
{
|
|
"entropy": 5.108003044128418,
|
|
"epoch": 8.47962382445141,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.932733342371306e-05,
|
|
"loss": 4.3424,
|
|
"mean_token_accuracy": 0.24270165115594863,
|
|
"num_tokens": 19018337.0,
|
|
"step": 10820
|
|
},
|
|
{
|
|
"entropy": 5.233252811431885,
|
|
"epoch": 8.483542319749215,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.917902062906066e-05,
|
|
"loss": 4.4178,
|
|
"mean_token_accuracy": 0.24028658717870713,
|
|
"num_tokens": 19027232.0,
|
|
"step": 10825
|
|
},
|
|
{
|
|
"entropy": 5.0702110767364506,
|
|
"epoch": 8.487460815047022,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.90310577982493e-05,
|
|
"loss": 4.2649,
|
|
"mean_token_accuracy": 0.25677852481603625,
|
|
"num_tokens": 19036661.0,
|
|
"step": 10830
|
|
},
|
|
{
|
|
"entropy": 5.122553730010987,
|
|
"epoch": 8.491379310344827,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.888344519571247e-05,
|
|
"loss": 4.2653,
|
|
"mean_token_accuracy": 0.2524389877915382,
|
|
"num_tokens": 19045955.0,
|
|
"step": 10835
|
|
},
|
|
{
|
|
"entropy": 5.202718257904053,
|
|
"epoch": 8.495297805642632,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.873618308525797e-05,
|
|
"loss": 4.3827,
|
|
"mean_token_accuracy": 0.23742946982383728,
|
|
"num_tokens": 19054183.0,
|
|
"step": 10840
|
|
},
|
|
{
|
|
"entropy": 5.243192434310913,
|
|
"epoch": 8.49921630094044,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.858927173006694e-05,
|
|
"loss": 4.3261,
|
|
"mean_token_accuracy": 0.25030425637960435,
|
|
"num_tokens": 19063029.0,
|
|
"step": 10845
|
|
},
|
|
{
|
|
"entropy": 5.137408638000489,
|
|
"epoch": 8.503134796238244,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.844271139269384e-05,
|
|
"loss": 4.2516,
|
|
"mean_token_accuracy": 0.2570894494652748,
|
|
"num_tokens": 19071325.0,
|
|
"step": 10850
|
|
},
|
|
{
|
|
"entropy": 5.166427803039551,
|
|
"epoch": 8.50705329153605,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.829650233506569e-05,
|
|
"loss": 4.369,
|
|
"mean_token_accuracy": 0.23896535933017732,
|
|
"num_tokens": 19080082.0,
|
|
"step": 10855
|
|
},
|
|
{
|
|
"entropy": 5.123944234848023,
|
|
"epoch": 8.510971786833856,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.815064481848188e-05,
|
|
"loss": 4.2618,
|
|
"mean_token_accuracy": 0.2480817660689354,
|
|
"num_tokens": 19089286.0,
|
|
"step": 10860
|
|
},
|
|
{
|
|
"entropy": 5.119918489456177,
|
|
"epoch": 8.514890282131661,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.800513910361343e-05,
|
|
"loss": 4.2078,
|
|
"mean_token_accuracy": 0.26679884642362595,
|
|
"num_tokens": 19098397.0,
|
|
"step": 10865
|
|
},
|
|
{
|
|
"entropy": 5.210055828094482,
|
|
"epoch": 8.518808777429467,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.785998545050253e-05,
|
|
"loss": 4.3917,
|
|
"mean_token_accuracy": 0.23431249409914018,
|
|
"num_tokens": 19107939.0,
|
|
"step": 10870
|
|
},
|
|
{
|
|
"entropy": 5.14661717414856,
|
|
"epoch": 8.522727272727273,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.771518411856243e-05,
|
|
"loss": 4.3586,
|
|
"mean_token_accuracy": 0.23951740562915802,
|
|
"num_tokens": 19116541.0,
|
|
"step": 10875
|
|
},
|
|
{
|
|
"entropy": 5.218864154815674,
|
|
"epoch": 8.526645768025078,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.75707353665765e-05,
|
|
"loss": 4.3607,
|
|
"mean_token_accuracy": 0.23700881153345107,
|
|
"num_tokens": 19124830.0,
|
|
"step": 10880
|
|
},
|
|
{
|
|
"entropy": 5.085940933227539,
|
|
"epoch": 8.530564263322884,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.742663945269816e-05,
|
|
"loss": 4.2074,
|
|
"mean_token_accuracy": 0.2563341915607452,
|
|
"num_tokens": 19133518.0,
|
|
"step": 10885
|
|
},
|
|
{
|
|
"entropy": 5.161911296844482,
|
|
"epoch": 8.53448275862069,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.72828966344501e-05,
|
|
"loss": 4.3782,
|
|
"mean_token_accuracy": 0.2425289884209633,
|
|
"num_tokens": 19142989.0,
|
|
"step": 10890
|
|
},
|
|
{
|
|
"entropy": 5.261181879043579,
|
|
"epoch": 8.538401253918495,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.713950716872411e-05,
|
|
"loss": 4.4155,
|
|
"mean_token_accuracy": 0.24546622186899186,
|
|
"num_tokens": 19151603.0,
|
|
"step": 10895
|
|
},
|
|
{
|
|
"entropy": 5.12997350692749,
|
|
"epoch": 8.5423197492163,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.699647131178038e-05,
|
|
"loss": 4.245,
|
|
"mean_token_accuracy": 0.24870358258485795,
|
|
"num_tokens": 19159949.0,
|
|
"step": 10900
|
|
},
|
|
{
|
|
"entropy": 5.133790493011475,
|
|
"epoch": 8.546238244514107,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.685378931924713e-05,
|
|
"loss": 4.3583,
|
|
"mean_token_accuracy": 0.24696255922317506,
|
|
"num_tokens": 19169594.0,
|
|
"step": 10905
|
|
},
|
|
{
|
|
"entropy": 5.146625661849976,
|
|
"epoch": 8.550156739811912,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.67114614461203e-05,
|
|
"loss": 4.2731,
|
|
"mean_token_accuracy": 0.2488436833024025,
|
|
"num_tokens": 19177660.0,
|
|
"step": 10910
|
|
},
|
|
{
|
|
"entropy": 5.129921388626099,
|
|
"epoch": 8.554075235109718,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.656948794676283e-05,
|
|
"loss": 4.3183,
|
|
"mean_token_accuracy": 0.24421289265155793,
|
|
"num_tokens": 19185890.0,
|
|
"step": 10915
|
|
},
|
|
{
|
|
"entropy": 5.085262632369995,
|
|
"epoch": 8.557993730407524,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.642786907490437e-05,
|
|
"loss": 4.2458,
|
|
"mean_token_accuracy": 0.25329305976629257,
|
|
"num_tokens": 19194820.0,
|
|
"step": 10920
|
|
},
|
|
{
|
|
"entropy": 5.130264616012573,
|
|
"epoch": 8.56191222570533,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.628660508364083e-05,
|
|
"loss": 4.3275,
|
|
"mean_token_accuracy": 0.2401234433054924,
|
|
"num_tokens": 19203887.0,
|
|
"step": 10925
|
|
},
|
|
{
|
|
"entropy": 5.1927924156188965,
|
|
"epoch": 8.565830721003135,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.614569622543391e-05,
|
|
"loss": 4.3972,
|
|
"mean_token_accuracy": 0.2397266745567322,
|
|
"num_tokens": 19212540.0,
|
|
"step": 10930
|
|
},
|
|
{
|
|
"entropy": 5.146560430526733,
|
|
"epoch": 8.56974921630094,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.600514275211054e-05,
|
|
"loss": 4.2765,
|
|
"mean_token_accuracy": 0.25184258818626404,
|
|
"num_tokens": 19221838.0,
|
|
"step": 10935
|
|
},
|
|
{
|
|
"entropy": 5.065685224533081,
|
|
"epoch": 8.573667711598747,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.58649449148625e-05,
|
|
"loss": 4.2031,
|
|
"mean_token_accuracy": 0.2566603526473045,
|
|
"num_tokens": 19229538.0,
|
|
"step": 10940
|
|
},
|
|
{
|
|
"entropy": 5.0783693313598635,
|
|
"epoch": 8.577586206896552,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.572510296424615e-05,
|
|
"loss": 4.1853,
|
|
"mean_token_accuracy": 0.2563704803586006,
|
|
"num_tokens": 19237977.0,
|
|
"step": 10945
|
|
},
|
|
{
|
|
"entropy": 5.121308326721191,
|
|
"epoch": 8.581504702194357,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.558561715018174e-05,
|
|
"loss": 4.3019,
|
|
"mean_token_accuracy": 0.23797413408756257,
|
|
"num_tokens": 19246801.0,
|
|
"step": 10950
|
|
},
|
|
{
|
|
"entropy": 5.1258703708648685,
|
|
"epoch": 8.585423197492164,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.544648772195293e-05,
|
|
"loss": 4.3246,
|
|
"mean_token_accuracy": 0.24596971720457078,
|
|
"num_tokens": 19256271.0,
|
|
"step": 10955
|
|
},
|
|
{
|
|
"entropy": 5.178742074966431,
|
|
"epoch": 8.589341692789969,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.530771492820673e-05,
|
|
"loss": 4.3003,
|
|
"mean_token_accuracy": 0.24155728667974471,
|
|
"num_tokens": 19264210.0,
|
|
"step": 10960
|
|
},
|
|
{
|
|
"entropy": 5.110103273391724,
|
|
"epoch": 8.593260188087774,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.516929901695249e-05,
|
|
"loss": 4.2155,
|
|
"mean_token_accuracy": 0.24910247772932054,
|
|
"num_tokens": 19272700.0,
|
|
"step": 10965
|
|
},
|
|
{
|
|
"entropy": 5.146948432922363,
|
|
"epoch": 8.59717868338558,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.503124023556192e-05,
|
|
"loss": 4.341,
|
|
"mean_token_accuracy": 0.24717204719781877,
|
|
"num_tokens": 19281674.0,
|
|
"step": 10970
|
|
},
|
|
{
|
|
"entropy": 5.16120719909668,
|
|
"epoch": 8.601097178683386,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.48935388307685e-05,
|
|
"loss": 4.269,
|
|
"mean_token_accuracy": 0.24772965908050537,
|
|
"num_tokens": 19290748.0,
|
|
"step": 10975
|
|
},
|
|
{
|
|
"entropy": 5.115499639511109,
|
|
"epoch": 8.60501567398119,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.47561950486669e-05,
|
|
"loss": 4.1968,
|
|
"mean_token_accuracy": 0.25816443413496015,
|
|
"num_tokens": 19299018.0,
|
|
"step": 10980
|
|
},
|
|
{
|
|
"entropy": 5.118164587020874,
|
|
"epoch": 8.608934169278998,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.461920913471276e-05,
|
|
"loss": 4.225,
|
|
"mean_token_accuracy": 0.25119697153568266,
|
|
"num_tokens": 19307915.0,
|
|
"step": 10985
|
|
},
|
|
{
|
|
"entropy": 5.163539457321167,
|
|
"epoch": 8.612852664576803,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.448258133372208e-05,
|
|
"loss": 4.2811,
|
|
"mean_token_accuracy": 0.2604338049888611,
|
|
"num_tokens": 19316358.0,
|
|
"step": 10990
|
|
},
|
|
{
|
|
"entropy": 5.1456409931182865,
|
|
"epoch": 8.616771159874608,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.434631188987095e-05,
|
|
"loss": 4.3712,
|
|
"mean_token_accuracy": 0.24101352244615554,
|
|
"num_tokens": 19325329.0,
|
|
"step": 10995
|
|
},
|
|
{
|
|
"entropy": 5.150396251678467,
|
|
"epoch": 8.620689655172415,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.42104010466949e-05,
|
|
"loss": 4.3381,
|
|
"mean_token_accuracy": 0.24429805725812911,
|
|
"num_tokens": 19333857.0,
|
|
"step": 11000
|
|
},
|
|
{
|
|
"epoch": 8.620689655172415,
|
|
"eval_entropy": 5.057033445483954,
|
|
"eval_loss": 5.646059513092041,
|
|
"eval_mean_token_accuracy": 0.17429619923580525,
|
|
"eval_num_tokens": 19333857.0,
|
|
"eval_runtime": 2.8616,
|
|
"eval_samples_per_second": 1440.455,
|
|
"eval_steps_per_second": 180.319,
|
|
"step": 11000
|
|
},
|
|
{
|
|
"entropy": 5.121606349945068,
|
|
"epoch": 8.62460815047022,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.407484904708874e-05,
|
|
"loss": 4.2808,
|
|
"mean_token_accuracy": 0.2505472257733345,
|
|
"num_tokens": 19342386.0,
|
|
"step": 11005
|
|
},
|
|
{
|
|
"entropy": 5.143646717071533,
|
|
"epoch": 8.628526645768025,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.393965613330575e-05,
|
|
"loss": 4.3297,
|
|
"mean_token_accuracy": 0.24426771700382233,
|
|
"num_tokens": 19350727.0,
|
|
"step": 11010
|
|
},
|
|
{
|
|
"entropy": 5.151013612747192,
|
|
"epoch": 8.63244514106583,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.38048225469577e-05,
|
|
"loss": 4.3293,
|
|
"mean_token_accuracy": 0.24359753578901291,
|
|
"num_tokens": 19360078.0,
|
|
"step": 11015
|
|
},
|
|
{
|
|
"entropy": 5.1212060928344725,
|
|
"epoch": 8.636363636363637,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 7.3670348529014e-05,
|
|
"loss": 4.3111,
|
|
"mean_token_accuracy": 0.2522352486848831,
|
|
"num_tokens": 19369092.0,
|
|
"step": 11020
|
|
},
|
|
{
|
|
"entropy": 5.153883838653565,
|
|
"epoch": 8.640282131661442,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.353623431980153e-05,
|
|
"loss": 4.3119,
|
|
"mean_token_accuracy": 0.2447877675294876,
|
|
"num_tokens": 19378061.0,
|
|
"step": 11025
|
|
},
|
|
{
|
|
"entropy": 5.1507118225097654,
|
|
"epoch": 8.644200626959247,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.340248015900416e-05,
|
|
"loss": 4.2924,
|
|
"mean_token_accuracy": 0.25570725053548815,
|
|
"num_tokens": 19386298.0,
|
|
"step": 11030
|
|
},
|
|
{
|
|
"entropy": 5.1119640350341795,
|
|
"epoch": 8.648119122257054,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.326908628566226e-05,
|
|
"loss": 4.2546,
|
|
"mean_token_accuracy": 0.2559199154376984,
|
|
"num_tokens": 19394001.0,
|
|
"step": 11035
|
|
},
|
|
{
|
|
"entropy": 5.041086673736572,
|
|
"epoch": 8.652037617554859,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.313605293817224e-05,
|
|
"loss": 4.1904,
|
|
"mean_token_accuracy": 0.2570702224969864,
|
|
"num_tokens": 19402410.0,
|
|
"step": 11040
|
|
},
|
|
{
|
|
"entropy": 5.157697772979736,
|
|
"epoch": 8.655956112852664,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.300338035428639e-05,
|
|
"loss": 4.3465,
|
|
"mean_token_accuracy": 0.2431817501783371,
|
|
"num_tokens": 19410947.0,
|
|
"step": 11045
|
|
},
|
|
{
|
|
"entropy": 5.146329689025879,
|
|
"epoch": 8.65987460815047,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.287106877111203e-05,
|
|
"loss": 4.3796,
|
|
"mean_token_accuracy": 0.24223367422819136,
|
|
"num_tokens": 19419286.0,
|
|
"step": 11050
|
|
},
|
|
{
|
|
"entropy": 5.121708679199219,
|
|
"epoch": 8.663793103448276,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.273911842511139e-05,
|
|
"loss": 4.2769,
|
|
"mean_token_accuracy": 0.2510464146733284,
|
|
"num_tokens": 19427703.0,
|
|
"step": 11055
|
|
},
|
|
{
|
|
"entropy": 5.150105237960815,
|
|
"epoch": 8.66771159874608,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.260752955210124e-05,
|
|
"loss": 4.3039,
|
|
"mean_token_accuracy": 0.2453116610646248,
|
|
"num_tokens": 19436329.0,
|
|
"step": 11060
|
|
},
|
|
{
|
|
"entropy": 5.157839250564575,
|
|
"epoch": 8.671630094043888,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.247630238725212e-05,
|
|
"loss": 4.2757,
|
|
"mean_token_accuracy": 0.2538643941283226,
|
|
"num_tokens": 19444363.0,
|
|
"step": 11065
|
|
},
|
|
{
|
|
"entropy": 5.157150459289551,
|
|
"epoch": 8.675548589341693,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.234543716508836e-05,
|
|
"loss": 4.2742,
|
|
"mean_token_accuracy": 0.24959492683410645,
|
|
"num_tokens": 19452745.0,
|
|
"step": 11070
|
|
},
|
|
{
|
|
"entropy": 5.192605638504029,
|
|
"epoch": 8.679467084639498,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.221493411948723e-05,
|
|
"loss": 4.3439,
|
|
"mean_token_accuracy": 0.23977122753858565,
|
|
"num_tokens": 19462481.0,
|
|
"step": 11075
|
|
},
|
|
{
|
|
"entropy": 5.2010979652404785,
|
|
"epoch": 8.683385579937305,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.208479348367895e-05,
|
|
"loss": 4.3144,
|
|
"mean_token_accuracy": 0.24763442873954772,
|
|
"num_tokens": 19471276.0,
|
|
"step": 11080
|
|
},
|
|
{
|
|
"entropy": 5.141250705718994,
|
|
"epoch": 8.68730407523511,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.195501549024582e-05,
|
|
"loss": 4.3305,
|
|
"mean_token_accuracy": 0.23991815596818925,
|
|
"num_tokens": 19480101.0,
|
|
"step": 11085
|
|
},
|
|
{
|
|
"entropy": 5.119322061538696,
|
|
"epoch": 8.691222570532915,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.182560037112217e-05,
|
|
"loss": 4.257,
|
|
"mean_token_accuracy": 0.2552376210689545,
|
|
"num_tokens": 19489027.0,
|
|
"step": 11090
|
|
},
|
|
{
|
|
"entropy": 5.152660512924195,
|
|
"epoch": 8.695141065830722,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.169654835759388e-05,
|
|
"loss": 4.2987,
|
|
"mean_token_accuracy": 0.24635542035102845,
|
|
"num_tokens": 19497363.0,
|
|
"step": 11095
|
|
},
|
|
{
|
|
"entropy": 5.143252229690551,
|
|
"epoch": 8.699059561128527,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.156785968029778e-05,
|
|
"loss": 4.2952,
|
|
"mean_token_accuracy": 0.24618889540433883,
|
|
"num_tokens": 19505362.0,
|
|
"step": 11100
|
|
},
|
|
{
|
|
"entropy": 5.087399911880493,
|
|
"epoch": 8.702978056426332,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 7.143953456922141e-05,
|
|
"loss": 4.2278,
|
|
"mean_token_accuracy": 0.25209988206624984,
|
|
"num_tokens": 19514536.0,
|
|
"step": 11105
|
|
},
|
|
{
|
|
"entropy": 5.124710369110107,
|
|
"epoch": 8.706896551724139,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.131157325370249e-05,
|
|
"loss": 4.3094,
|
|
"mean_token_accuracy": 0.24189272075891494,
|
|
"num_tokens": 19523276.0,
|
|
"step": 11110
|
|
},
|
|
{
|
|
"entropy": 5.095740079879761,
|
|
"epoch": 8.710815047021944,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.118397596242872e-05,
|
|
"loss": 4.1972,
|
|
"mean_token_accuracy": 0.2514286831021309,
|
|
"num_tokens": 19532020.0,
|
|
"step": 11115
|
|
},
|
|
{
|
|
"entropy": 5.15518798828125,
|
|
"epoch": 8.714733542319749,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.105674292343709e-05,
|
|
"loss": 4.2846,
|
|
"mean_token_accuracy": 0.24525388479232788,
|
|
"num_tokens": 19540175.0,
|
|
"step": 11120
|
|
},
|
|
{
|
|
"entropy": 5.207593965530395,
|
|
"epoch": 8.718652037617554,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.092987436411369e-05,
|
|
"loss": 4.3199,
|
|
"mean_token_accuracy": 0.25077468901872635,
|
|
"num_tokens": 19549728.0,
|
|
"step": 11125
|
|
},
|
|
{
|
|
"entropy": 5.101954984664917,
|
|
"epoch": 8.72257053291536,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.080337051119325e-05,
|
|
"loss": 4.2895,
|
|
"mean_token_accuracy": 0.2483597457408905,
|
|
"num_tokens": 19558859.0,
|
|
"step": 11130
|
|
},
|
|
{
|
|
"entropy": 5.106394481658936,
|
|
"epoch": 8.726489028213166,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.067723159075864e-05,
|
|
"loss": 4.2368,
|
|
"mean_token_accuracy": 0.25619224458932877,
|
|
"num_tokens": 19567397.0,
|
|
"step": 11135
|
|
},
|
|
{
|
|
"entropy": 5.110416460037231,
|
|
"epoch": 8.730407523510971,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.055145782824051e-05,
|
|
"loss": 4.2599,
|
|
"mean_token_accuracy": 0.24583133757114412,
|
|
"num_tokens": 19575986.0,
|
|
"step": 11140
|
|
},
|
|
{
|
|
"entropy": 5.136254739761353,
|
|
"epoch": 8.734326018808778,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.042604944841705e-05,
|
|
"loss": 4.2627,
|
|
"mean_token_accuracy": 0.25209995806217195,
|
|
"num_tokens": 19585066.0,
|
|
"step": 11145
|
|
},
|
|
{
|
|
"entropy": 5.157420635223389,
|
|
"epoch": 8.738244514106583,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.030100667541333e-05,
|
|
"loss": 4.2873,
|
|
"mean_token_accuracy": 0.24294230341911316,
|
|
"num_tokens": 19594044.0,
|
|
"step": 11150
|
|
},
|
|
{
|
|
"entropy": 5.1300678730010985,
|
|
"epoch": 8.742163009404388,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.017632973270105e-05,
|
|
"loss": 4.2907,
|
|
"mean_token_accuracy": 0.24794731438159942,
|
|
"num_tokens": 19602416.0,
|
|
"step": 11155
|
|
},
|
|
{
|
|
"entropy": 5.1288305759429935,
|
|
"epoch": 8.746081504702195,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.005201884309812e-05,
|
|
"loss": 4.2926,
|
|
"mean_token_accuracy": 0.24962397813796997,
|
|
"num_tokens": 19612435.0,
|
|
"step": 11160
|
|
},
|
|
{
|
|
"entropy": 5.164767837524414,
|
|
"epoch": 8.75,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.992807422876829e-05,
|
|
"loss": 4.2955,
|
|
"mean_token_accuracy": 0.24952344596385956,
|
|
"num_tokens": 19621708.0,
|
|
"step": 11165
|
|
},
|
|
{
|
|
"entropy": 5.182446432113648,
|
|
"epoch": 8.753918495297805,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.980449611122065e-05,
|
|
"loss": 4.3182,
|
|
"mean_token_accuracy": 0.2506682172417641,
|
|
"num_tokens": 19630879.0,
|
|
"step": 11170
|
|
},
|
|
{
|
|
"entropy": 5.145661878585815,
|
|
"epoch": 8.757836990595612,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.968128471130932e-05,
|
|
"loss": 4.2784,
|
|
"mean_token_accuracy": 0.2520853653550148,
|
|
"num_tokens": 19640265.0,
|
|
"step": 11175
|
|
},
|
|
{
|
|
"entropy": 5.132181835174561,
|
|
"epoch": 8.761755485893417,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.955844024923308e-05,
|
|
"loss": 4.3148,
|
|
"mean_token_accuracy": 0.25204054117202757,
|
|
"num_tokens": 19648447.0,
|
|
"step": 11180
|
|
},
|
|
{
|
|
"entropy": 5.166639232635498,
|
|
"epoch": 8.765673981191222,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.943596294453492e-05,
|
|
"loss": 4.3224,
|
|
"mean_token_accuracy": 0.24289816617965698,
|
|
"num_tokens": 19657971.0,
|
|
"step": 11185
|
|
},
|
|
{
|
|
"entropy": 5.178141021728516,
|
|
"epoch": 8.769592476489029,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.931385301610161e-05,
|
|
"loss": 4.2784,
|
|
"mean_token_accuracy": 0.2542914435267448,
|
|
"num_tokens": 19667059.0,
|
|
"step": 11190
|
|
},
|
|
{
|
|
"entropy": 5.140672492980957,
|
|
"epoch": 8.773510971786834,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.919211068216337e-05,
|
|
"loss": 4.3087,
|
|
"mean_token_accuracy": 0.24780102372169494,
|
|
"num_tokens": 19675683.0,
|
|
"step": 11195
|
|
},
|
|
{
|
|
"entropy": 5.176463794708252,
|
|
"epoch": 8.77742946708464,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.907073616029356e-05,
|
|
"loss": 4.3156,
|
|
"mean_token_accuracy": 0.2437306672334671,
|
|
"num_tokens": 19685317.0,
|
|
"step": 11200
|
|
},
|
|
{
|
|
"entropy": 5.1706531047821045,
|
|
"epoch": 8.781347962382446,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.894972966740809e-05,
|
|
"loss": 4.3397,
|
|
"mean_token_accuracy": 0.24703810811042787,
|
|
"num_tokens": 19694013.0,
|
|
"step": 11205
|
|
},
|
|
{
|
|
"entropy": 5.089597749710083,
|
|
"epoch": 8.785266457680251,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.882909141976514e-05,
|
|
"loss": 4.2595,
|
|
"mean_token_accuracy": 0.2505308255553246,
|
|
"num_tokens": 19702527.0,
|
|
"step": 11210
|
|
},
|
|
{
|
|
"entropy": 5.173648500442505,
|
|
"epoch": 8.789184952978056,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.87088216329649e-05,
|
|
"loss": 4.3271,
|
|
"mean_token_accuracy": 0.24254980385303498,
|
|
"num_tokens": 19711075.0,
|
|
"step": 11215
|
|
},
|
|
{
|
|
"entropy": 5.126655387878418,
|
|
"epoch": 8.793103448275861,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.858892052194894e-05,
|
|
"loss": 4.2651,
|
|
"mean_token_accuracy": 0.24814883023500442,
|
|
"num_tokens": 19719964.0,
|
|
"step": 11220
|
|
},
|
|
{
|
|
"entropy": 5.081950712203979,
|
|
"epoch": 8.797021943573668,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.846938830099995e-05,
|
|
"loss": 4.2495,
|
|
"mean_token_accuracy": 0.24885703176259993,
|
|
"num_tokens": 19728801.0,
|
|
"step": 11225
|
|
},
|
|
{
|
|
"entropy": 5.149282836914063,
|
|
"epoch": 8.800940438871473,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.835022518374147e-05,
|
|
"loss": 4.3197,
|
|
"mean_token_accuracy": 0.24789837300777434,
|
|
"num_tokens": 19737446.0,
|
|
"step": 11230
|
|
},
|
|
{
|
|
"entropy": 5.156679296493531,
|
|
"epoch": 8.804858934169278,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.823143138313725e-05,
|
|
"loss": 4.3157,
|
|
"mean_token_accuracy": 0.25159947723150256,
|
|
"num_tokens": 19745955.0,
|
|
"step": 11235
|
|
},
|
|
{
|
|
"entropy": 5.126048469543457,
|
|
"epoch": 8.808777429467085,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.811300711149109e-05,
|
|
"loss": 4.2564,
|
|
"mean_token_accuracy": 0.259194079041481,
|
|
"num_tokens": 19754463.0,
|
|
"step": 11240
|
|
},
|
|
{
|
|
"entropy": 5.1331006526947025,
|
|
"epoch": 8.81269592476489,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.799495258044633e-05,
|
|
"loss": 4.3174,
|
|
"mean_token_accuracy": 0.2500792399048805,
|
|
"num_tokens": 19762828.0,
|
|
"step": 11245
|
|
},
|
|
{
|
|
"entropy": 5.194873523712158,
|
|
"epoch": 8.816614420062695,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.787726800098568e-05,
|
|
"loss": 4.3532,
|
|
"mean_token_accuracy": 0.24257101863622665,
|
|
"num_tokens": 19771159.0,
|
|
"step": 11250
|
|
},
|
|
{
|
|
"entropy": 5.105468797683716,
|
|
"epoch": 8.820532915360502,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.775995358343043e-05,
|
|
"loss": 4.2892,
|
|
"mean_token_accuracy": 0.24844644963741302,
|
|
"num_tokens": 19779771.0,
|
|
"step": 11255
|
|
},
|
|
{
|
|
"entropy": 5.151581239700318,
|
|
"epoch": 8.824451410658307,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.764300953744052e-05,
|
|
"loss": 4.3014,
|
|
"mean_token_accuracy": 0.24857334196567535,
|
|
"num_tokens": 19788552.0,
|
|
"step": 11260
|
|
},
|
|
{
|
|
"entropy": 5.135622262954712,
|
|
"epoch": 8.828369905956112,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.752643607201394e-05,
|
|
"loss": 4.335,
|
|
"mean_token_accuracy": 0.24956384152173997,
|
|
"num_tokens": 19797346.0,
|
|
"step": 11265
|
|
},
|
|
{
|
|
"entropy": 5.183726406097412,
|
|
"epoch": 8.83228840125392,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.741023339548633e-05,
|
|
"loss": 4.3347,
|
|
"mean_token_accuracy": 0.24237982630729676,
|
|
"num_tokens": 19806152.0,
|
|
"step": 11270
|
|
},
|
|
{
|
|
"entropy": 5.158955478668213,
|
|
"epoch": 8.836206896551724,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.729440171553078e-05,
|
|
"loss": 4.3023,
|
|
"mean_token_accuracy": 0.24944265335798263,
|
|
"num_tokens": 19814977.0,
|
|
"step": 11275
|
|
},
|
|
{
|
|
"entropy": 5.173917055130005,
|
|
"epoch": 8.84012539184953,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.71789412391572e-05,
|
|
"loss": 4.373,
|
|
"mean_token_accuracy": 0.2402506336569786,
|
|
"num_tokens": 19824203.0,
|
|
"step": 11280
|
|
},
|
|
{
|
|
"entropy": 5.194516038894653,
|
|
"epoch": 8.844043887147336,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.706385217271223e-05,
|
|
"loss": 4.3232,
|
|
"mean_token_accuracy": 0.24385690540075303,
|
|
"num_tokens": 19831974.0,
|
|
"step": 11285
|
|
},
|
|
{
|
|
"entropy": 5.1278034210205075,
|
|
"epoch": 8.847962382445141,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.69491347218787e-05,
|
|
"loss": 4.2961,
|
|
"mean_token_accuracy": 0.24713827967643737,
|
|
"num_tokens": 19840709.0,
|
|
"step": 11290
|
|
},
|
|
{
|
|
"entropy": 5.118567276000976,
|
|
"epoch": 8.851880877742946,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.683478909167525e-05,
|
|
"loss": 4.311,
|
|
"mean_token_accuracy": 0.25123821049928663,
|
|
"num_tokens": 19849409.0,
|
|
"step": 11295
|
|
},
|
|
{
|
|
"entropy": 5.1362607955932615,
|
|
"epoch": 8.855799373040753,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.672081548645608e-05,
|
|
"loss": 4.2648,
|
|
"mean_token_accuracy": 0.25464241951704025,
|
|
"num_tokens": 19858331.0,
|
|
"step": 11300
|
|
},
|
|
{
|
|
"entropy": 5.107956171035767,
|
|
"epoch": 8.859717868338558,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.660721410991057e-05,
|
|
"loss": 4.2567,
|
|
"mean_token_accuracy": 0.25162831097841265,
|
|
"num_tokens": 19866469.0,
|
|
"step": 11305
|
|
},
|
|
{
|
|
"entropy": 5.167484188079834,
|
|
"epoch": 8.863636363636363,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.649398516506272e-05,
|
|
"loss": 4.2997,
|
|
"mean_token_accuracy": 0.25251161903142927,
|
|
"num_tokens": 19875183.0,
|
|
"step": 11310
|
|
},
|
|
{
|
|
"entropy": 5.19882230758667,
|
|
"epoch": 8.86755485893417,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.638112885427103e-05,
|
|
"loss": 4.3618,
|
|
"mean_token_accuracy": 0.24218684434890747,
|
|
"num_tokens": 19884440.0,
|
|
"step": 11315
|
|
},
|
|
{
|
|
"entropy": 5.156722927093506,
|
|
"epoch": 8.871473354231975,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.626864537922808e-05,
|
|
"loss": 4.2956,
|
|
"mean_token_accuracy": 0.2481937065720558,
|
|
"num_tokens": 19892833.0,
|
|
"step": 11320
|
|
},
|
|
{
|
|
"entropy": 5.1739531517028805,
|
|
"epoch": 8.87539184952978,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.615653494096001e-05,
|
|
"loss": 4.3502,
|
|
"mean_token_accuracy": 0.24811737537384032,
|
|
"num_tokens": 19900805.0,
|
|
"step": 11325
|
|
},
|
|
{
|
|
"entropy": 5.172651624679565,
|
|
"epoch": 8.879310344827585,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.604479773982634e-05,
|
|
"loss": 4.3055,
|
|
"mean_token_accuracy": 0.24768787175416945,
|
|
"num_tokens": 19909693.0,
|
|
"step": 11330
|
|
},
|
|
{
|
|
"entropy": 5.177678632736206,
|
|
"epoch": 8.883228840125392,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.593343397551964e-05,
|
|
"loss": 4.3282,
|
|
"mean_token_accuracy": 0.2535400092601776,
|
|
"num_tokens": 19918234.0,
|
|
"step": 11335
|
|
},
|
|
{
|
|
"entropy": 5.134753227233887,
|
|
"epoch": 8.887147335423197,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.582244384706497e-05,
|
|
"loss": 4.3048,
|
|
"mean_token_accuracy": 0.24834633618593216,
|
|
"num_tokens": 19927465.0,
|
|
"step": 11340
|
|
},
|
|
{
|
|
"entropy": 5.103481292724609,
|
|
"epoch": 8.891065830721002,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.571182755281964e-05,
|
|
"loss": 4.2794,
|
|
"mean_token_accuracy": 0.2605300396680832,
|
|
"num_tokens": 19936444.0,
|
|
"step": 11345
|
|
},
|
|
{
|
|
"entropy": 5.074260520935058,
|
|
"epoch": 8.89498432601881,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.5601585290473e-05,
|
|
"loss": 4.2261,
|
|
"mean_token_accuracy": 0.25564686954021454,
|
|
"num_tokens": 19944981.0,
|
|
"step": 11350
|
|
},
|
|
{
|
|
"entropy": 5.1847292423248295,
|
|
"epoch": 8.898902821316614,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.549171725704575e-05,
|
|
"loss": 4.4057,
|
|
"mean_token_accuracy": 0.23672655671834947,
|
|
"num_tokens": 19953907.0,
|
|
"step": 11355
|
|
},
|
|
{
|
|
"entropy": 5.095714378356933,
|
|
"epoch": 8.90282131661442,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.538222364889e-05,
|
|
"loss": 4.263,
|
|
"mean_token_accuracy": 0.24764550775289534,
|
|
"num_tokens": 19962487.0,
|
|
"step": 11360
|
|
},
|
|
{
|
|
"entropy": 5.179597330093384,
|
|
"epoch": 8.906739811912226,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.527310466168842e-05,
|
|
"loss": 4.3337,
|
|
"mean_token_accuracy": 0.24254990816116334,
|
|
"num_tokens": 19971243.0,
|
|
"step": 11365
|
|
},
|
|
{
|
|
"entropy": 5.13467493057251,
|
|
"epoch": 8.910658307210031,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.516436049045452e-05,
|
|
"loss": 4.3533,
|
|
"mean_token_accuracy": 0.2389046236872673,
|
|
"num_tokens": 19980281.0,
|
|
"step": 11370
|
|
},
|
|
{
|
|
"entropy": 5.177463006973267,
|
|
"epoch": 8.914576802507836,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.505599132953165e-05,
|
|
"loss": 4.433,
|
|
"mean_token_accuracy": 0.22970882207155227,
|
|
"num_tokens": 19990340.0,
|
|
"step": 11375
|
|
},
|
|
{
|
|
"entropy": 5.092184686660767,
|
|
"epoch": 8.918495297805643,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.494799737259304e-05,
|
|
"loss": 4.2199,
|
|
"mean_token_accuracy": 0.250371678173542,
|
|
"num_tokens": 19998770.0,
|
|
"step": 11380
|
|
},
|
|
{
|
|
"entropy": 5.084789133071899,
|
|
"epoch": 8.922413793103448,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.48403788126415e-05,
|
|
"loss": 4.2325,
|
|
"mean_token_accuracy": 0.25736636072397234,
|
|
"num_tokens": 20007388.0,
|
|
"step": 11385
|
|
},
|
|
{
|
|
"entropy": 5.10077018737793,
|
|
"epoch": 8.926332288401253,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.473313584200882e-05,
|
|
"loss": 4.2122,
|
|
"mean_token_accuracy": 0.2578940108418465,
|
|
"num_tokens": 20016350.0,
|
|
"step": 11390
|
|
},
|
|
{
|
|
"entropy": 5.080121326446533,
|
|
"epoch": 8.93025078369906,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.462626865235549e-05,
|
|
"loss": 4.2285,
|
|
"mean_token_accuracy": 0.2500581920146942,
|
|
"num_tokens": 20025777.0,
|
|
"step": 11395
|
|
},
|
|
{
|
|
"entropy": 5.138501977920532,
|
|
"epoch": 8.934169278996865,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.45197774346706e-05,
|
|
"loss": 4.2739,
|
|
"mean_token_accuracy": 0.2531504452228546,
|
|
"num_tokens": 20034396.0,
|
|
"step": 11400
|
|
},
|
|
{
|
|
"entropy": 5.135597896575928,
|
|
"epoch": 8.93808777429467,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.441366237927123e-05,
|
|
"loss": 4.2757,
|
|
"mean_token_accuracy": 0.24710915088653565,
|
|
"num_tokens": 20043190.0,
|
|
"step": 11405
|
|
},
|
|
{
|
|
"entropy": 5.113222932815551,
|
|
"epoch": 8.942006269592476,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.430792367580215e-05,
|
|
"loss": 4.3095,
|
|
"mean_token_accuracy": 0.24314149618148803,
|
|
"num_tokens": 20052108.0,
|
|
"step": 11410
|
|
},
|
|
{
|
|
"entropy": 5.1220457553863525,
|
|
"epoch": 8.945924764890282,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.420256151323553e-05,
|
|
"loss": 4.2194,
|
|
"mean_token_accuracy": 0.25585377812385557,
|
|
"num_tokens": 20060540.0,
|
|
"step": 11415
|
|
},
|
|
{
|
|
"entropy": 5.071147489547729,
|
|
"epoch": 8.949843260188088,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.409757607987066e-05,
|
|
"loss": 4.2209,
|
|
"mean_token_accuracy": 0.24800773411989213,
|
|
"num_tokens": 20069289.0,
|
|
"step": 11420
|
|
},
|
|
{
|
|
"entropy": 5.160710763931275,
|
|
"epoch": 8.953761755485893,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.399296756333362e-05,
|
|
"loss": 4.2987,
|
|
"mean_token_accuracy": 0.2484084278345108,
|
|
"num_tokens": 20078064.0,
|
|
"step": 11425
|
|
},
|
|
{
|
|
"entropy": 5.145409679412841,
|
|
"epoch": 8.9576802507837,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.388873615057665e-05,
|
|
"loss": 4.2644,
|
|
"mean_token_accuracy": 0.2544803857803345,
|
|
"num_tokens": 20086234.0,
|
|
"step": 11430
|
|
},
|
|
{
|
|
"entropy": 5.128638792037964,
|
|
"epoch": 8.961598746081505,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.378488202787835e-05,
|
|
"loss": 4.2504,
|
|
"mean_token_accuracy": 0.2479208901524544,
|
|
"num_tokens": 20094700.0,
|
|
"step": 11435
|
|
},
|
|
{
|
|
"entropy": 5.155966186523438,
|
|
"epoch": 8.96551724137931,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.368140538084275e-05,
|
|
"loss": 4.3407,
|
|
"mean_token_accuracy": 0.24506728649139403,
|
|
"num_tokens": 20103669.0,
|
|
"step": 11440
|
|
},
|
|
{
|
|
"entropy": 5.142144060134887,
|
|
"epoch": 8.969435736677116,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.357830639439943e-05,
|
|
"loss": 4.2734,
|
|
"mean_token_accuracy": 0.25109430551528933,
|
|
"num_tokens": 20111840.0,
|
|
"step": 11445
|
|
},
|
|
{
|
|
"entropy": 5.131678199768066,
|
|
"epoch": 8.973354231974922,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.347558525280295e-05,
|
|
"loss": 4.2621,
|
|
"mean_token_accuracy": 0.2531795099377632,
|
|
"num_tokens": 20119590.0,
|
|
"step": 11450
|
|
},
|
|
{
|
|
"entropy": 5.121889114379883,
|
|
"epoch": 8.977272727272727,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.337324213963274e-05,
|
|
"loss": 4.234,
|
|
"mean_token_accuracy": 0.25344163328409197,
|
|
"num_tokens": 20128982.0,
|
|
"step": 11455
|
|
},
|
|
{
|
|
"entropy": 5.160292387008667,
|
|
"epoch": 8.981191222570533,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.327127723779248e-05,
|
|
"loss": 4.394,
|
|
"mean_token_accuracy": 0.2408914864063263,
|
|
"num_tokens": 20138013.0,
|
|
"step": 11460
|
|
},
|
|
{
|
|
"entropy": 5.130927515029907,
|
|
"epoch": 8.985109717868339,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.316969072950998e-05,
|
|
"loss": 4.2745,
|
|
"mean_token_accuracy": 0.24781085103750228,
|
|
"num_tokens": 20146240.0,
|
|
"step": 11465
|
|
},
|
|
{
|
|
"entropy": 5.215606355667115,
|
|
"epoch": 8.989028213166144,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.306848279633684e-05,
|
|
"loss": 4.3839,
|
|
"mean_token_accuracy": 0.2411717042326927,
|
|
"num_tokens": 20155964.0,
|
|
"step": 11470
|
|
},
|
|
{
|
|
"entropy": 5.136001062393189,
|
|
"epoch": 8.99294670846395,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.296765361914798e-05,
|
|
"loss": 4.3454,
|
|
"mean_token_accuracy": 0.24890373796224594,
|
|
"num_tokens": 20165774.0,
|
|
"step": 11475
|
|
},
|
|
{
|
|
"entropy": 5.131713151931763,
|
|
"epoch": 8.996865203761756,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 6.286720337814158e-05,
|
|
"loss": 4.3262,
|
|
"mean_token_accuracy": 0.24242897629737853,
|
|
"num_tokens": 20174940.0,
|
|
"step": 11480
|
|
},
|
|
{
|
|
"entropy": 5.167542409896851,
|
|
"epoch": 9.00078369905956,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.276713225283839e-05,
|
|
"loss": 4.2444,
|
|
"mean_token_accuracy": 0.25624677538871765,
|
|
"num_tokens": 20182706.0,
|
|
"step": 11485
|
|
},
|
|
{
|
|
"entropy": 5.154136085510254,
|
|
"epoch": 9.004702194357368,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.266744042208186e-05,
|
|
"loss": 4.2652,
|
|
"mean_token_accuracy": 0.24923783391714097,
|
|
"num_tokens": 20191457.0,
|
|
"step": 11490
|
|
},
|
|
{
|
|
"entropy": 5.15484585762024,
|
|
"epoch": 9.008620689655173,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.25681280640374e-05,
|
|
"loss": 4.3314,
|
|
"mean_token_accuracy": 0.2359151840209961,
|
|
"num_tokens": 20199331.0,
|
|
"step": 11495
|
|
},
|
|
{
|
|
"entropy": 5.174219799041748,
|
|
"epoch": 9.012539184952978,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.246919535619226e-05,
|
|
"loss": 4.2637,
|
|
"mean_token_accuracy": 0.24946436136960984,
|
|
"num_tokens": 20208386.0,
|
|
"step": 11500
|
|
},
|
|
{
|
|
"epoch": 9.012539184952978,
|
|
"eval_entropy": 5.067491701407026,
|
|
"eval_loss": 5.643849849700928,
|
|
"eval_mean_token_accuracy": 0.17467680609561215,
|
|
"eval_num_tokens": 20208386.0,
|
|
"eval_runtime": 2.8634,
|
|
"eval_samples_per_second": 1439.57,
|
|
"eval_steps_per_second": 180.208,
|
|
"step": 11500
|
|
},
|
|
{
|
|
"entropy": 5.156183815002441,
|
|
"epoch": 9.016457680250785,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.237064247535531e-05,
|
|
"loss": 4.3044,
|
|
"mean_token_accuracy": 0.24605339914560317,
|
|
"num_tokens": 20217649.0,
|
|
"step": 11505
|
|
},
|
|
{
|
|
"entropy": 5.100423336029053,
|
|
"epoch": 9.02037617554859,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.227246959765644e-05,
|
|
"loss": 4.2231,
|
|
"mean_token_accuracy": 0.2574188619852066,
|
|
"num_tokens": 20226219.0,
|
|
"step": 11510
|
|
},
|
|
{
|
|
"entropy": 5.1473511219024655,
|
|
"epoch": 9.024294670846395,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.217467689854654e-05,
|
|
"loss": 4.2641,
|
|
"mean_token_accuracy": 0.2518458738923073,
|
|
"num_tokens": 20234788.0,
|
|
"step": 11515
|
|
},
|
|
{
|
|
"entropy": 5.137230682373047,
|
|
"epoch": 9.0282131661442,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.207726455279705e-05,
|
|
"loss": 4.2044,
|
|
"mean_token_accuracy": 0.26205248981714246,
|
|
"num_tokens": 20242858.0,
|
|
"step": 11520
|
|
},
|
|
{
|
|
"entropy": 5.0954601764678955,
|
|
"epoch": 9.032131661442007,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.198023273449963e-05,
|
|
"loss": 4.2449,
|
|
"mean_token_accuracy": 0.25264994204044344,
|
|
"num_tokens": 20252041.0,
|
|
"step": 11525
|
|
},
|
|
{
|
|
"entropy": 5.1624492645263675,
|
|
"epoch": 9.036050156739812,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.188358161706578e-05,
|
|
"loss": 4.3298,
|
|
"mean_token_accuracy": 0.24375341683626175,
|
|
"num_tokens": 20261035.0,
|
|
"step": 11530
|
|
},
|
|
{
|
|
"entropy": 5.188075160980224,
|
|
"epoch": 9.039968652037617,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.178731137322681e-05,
|
|
"loss": 4.3565,
|
|
"mean_token_accuracy": 0.24159966111183168,
|
|
"num_tokens": 20270091.0,
|
|
"step": 11535
|
|
},
|
|
{
|
|
"entropy": 5.1567711353302,
|
|
"epoch": 9.043887147335424,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.169142217503319e-05,
|
|
"loss": 4.3213,
|
|
"mean_token_accuracy": 0.249404676258564,
|
|
"num_tokens": 20278890.0,
|
|
"step": 11540
|
|
},
|
|
{
|
|
"entropy": 5.146059799194336,
|
|
"epoch": 9.047805642633229,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.159591419385455e-05,
|
|
"loss": 4.3594,
|
|
"mean_token_accuracy": 0.24170825034379959,
|
|
"num_tokens": 20287692.0,
|
|
"step": 11545
|
|
},
|
|
{
|
|
"entropy": 5.176144456863403,
|
|
"epoch": 9.051724137931034,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.150078760037904e-05,
|
|
"loss": 4.2437,
|
|
"mean_token_accuracy": 0.2588124230504036,
|
|
"num_tokens": 20296357.0,
|
|
"step": 11550
|
|
},
|
|
{
|
|
"entropy": 5.158664608001709,
|
|
"epoch": 9.05564263322884,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.140604256461339e-05,
|
|
"loss": 4.2862,
|
|
"mean_token_accuracy": 0.2556474328041077,
|
|
"num_tokens": 20304894.0,
|
|
"step": 11555
|
|
},
|
|
{
|
|
"entropy": 5.120600318908691,
|
|
"epoch": 9.059561128526646,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.131167925588233e-05,
|
|
"loss": 4.2985,
|
|
"mean_token_accuracy": 0.2468557208776474,
|
|
"num_tokens": 20314011.0,
|
|
"step": 11560
|
|
},
|
|
{
|
|
"entropy": 5.08376898765564,
|
|
"epoch": 9.06347962382445,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.121769784282833e-05,
|
|
"loss": 4.1472,
|
|
"mean_token_accuracy": 0.26083437651395797,
|
|
"num_tokens": 20322738.0,
|
|
"step": 11565
|
|
},
|
|
{
|
|
"entropy": 5.136650848388672,
|
|
"epoch": 9.067398119122258,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.112409849341147e-05,
|
|
"loss": 4.1987,
|
|
"mean_token_accuracy": 0.25827079862356184,
|
|
"num_tokens": 20331528.0,
|
|
"step": 11570
|
|
},
|
|
{
|
|
"entropy": 5.135408163070679,
|
|
"epoch": 9.071316614420063,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.103088137490895e-05,
|
|
"loss": 4.3033,
|
|
"mean_token_accuracy": 0.2460256814956665,
|
|
"num_tokens": 20340411.0,
|
|
"step": 11575
|
|
},
|
|
{
|
|
"entropy": 5.191963005065918,
|
|
"epoch": 9.075235109717868,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.093804665391488e-05,
|
|
"loss": 4.3391,
|
|
"mean_token_accuracy": 0.24553284347057341,
|
|
"num_tokens": 20349266.0,
|
|
"step": 11580
|
|
},
|
|
{
|
|
"entropy": 5.198410606384277,
|
|
"epoch": 9.079153605015675,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.084559449633992e-05,
|
|
"loss": 4.3282,
|
|
"mean_token_accuracy": 0.24727063477039338,
|
|
"num_tokens": 20357849.0,
|
|
"step": 11585
|
|
},
|
|
{
|
|
"entropy": 5.146946716308594,
|
|
"epoch": 9.08307210031348,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.0753525067411106e-05,
|
|
"loss": 4.2622,
|
|
"mean_token_accuracy": 0.2526800185441971,
|
|
"num_tokens": 20366795.0,
|
|
"step": 11590
|
|
},
|
|
{
|
|
"entropy": 5.099748373031616,
|
|
"epoch": 9.086990595611285,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.06618385316714e-05,
|
|
"loss": 4.2037,
|
|
"mean_token_accuracy": 0.26288470178842543,
|
|
"num_tokens": 20375086.0,
|
|
"step": 11595
|
|
},
|
|
{
|
|
"entropy": 5.115091037750244,
|
|
"epoch": 9.090909090909092,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.0570535052979554e-05,
|
|
"loss": 4.2414,
|
|
"mean_token_accuracy": 0.2533798336982727,
|
|
"num_tokens": 20384071.0,
|
|
"step": 11600
|
|
},
|
|
{
|
|
"entropy": 5.1568115711212155,
|
|
"epoch": 9.094827586206897,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.047961479450968e-05,
|
|
"loss": 4.2543,
|
|
"mean_token_accuracy": 0.2545537084341049,
|
|
"num_tokens": 20393345.0,
|
|
"step": 11605
|
|
},
|
|
{
|
|
"entropy": 5.13493800163269,
|
|
"epoch": 9.098746081504702,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.038907791875102e-05,
|
|
"loss": 4.2517,
|
|
"mean_token_accuracy": 0.24425923079252243,
|
|
"num_tokens": 20402034.0,
|
|
"step": 11610
|
|
},
|
|
{
|
|
"entropy": 5.155802869796753,
|
|
"epoch": 9.102664576802507,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.0298924587507655e-05,
|
|
"loss": 4.2989,
|
|
"mean_token_accuracy": 0.24162483662366868,
|
|
"num_tokens": 20411681.0,
|
|
"step": 11615
|
|
},
|
|
{
|
|
"entropy": 5.125881862640381,
|
|
"epoch": 9.106583072100314,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.020915496189815e-05,
|
|
"loss": 4.3211,
|
|
"mean_token_accuracy": 0.24362093359231948,
|
|
"num_tokens": 20420084.0,
|
|
"step": 11620
|
|
},
|
|
{
|
|
"entropy": 5.153845119476318,
|
|
"epoch": 9.110501567398119,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.0119769202355455e-05,
|
|
"loss": 4.3628,
|
|
"mean_token_accuracy": 0.23611244261264802,
|
|
"num_tokens": 20429326.0,
|
|
"step": 11625
|
|
},
|
|
{
|
|
"entropy": 5.216349506378174,
|
|
"epoch": 9.114420062695924,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.003076746862637e-05,
|
|
"loss": 4.3688,
|
|
"mean_token_accuracy": 0.24864142537117004,
|
|
"num_tokens": 20438522.0,
|
|
"step": 11630
|
|
},
|
|
{
|
|
"entropy": 5.215970420837403,
|
|
"epoch": 9.11833855799373,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.994214991977143e-05,
|
|
"loss": 4.3566,
|
|
"mean_token_accuracy": 0.24914633631706237,
|
|
"num_tokens": 20447084.0,
|
|
"step": 11635
|
|
},
|
|
{
|
|
"entropy": 5.104770231246948,
|
|
"epoch": 9.122257053291536,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.9853916714164584e-05,
|
|
"loss": 4.1876,
|
|
"mean_token_accuracy": 0.26067315191030505,
|
|
"num_tokens": 20455553.0,
|
|
"step": 11640
|
|
},
|
|
{
|
|
"entropy": 5.148068284988403,
|
|
"epoch": 9.126175548589341,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.976606800949287e-05,
|
|
"loss": 4.228,
|
|
"mean_token_accuracy": 0.2580898553133011,
|
|
"num_tokens": 20464066.0,
|
|
"step": 11645
|
|
},
|
|
{
|
|
"entropy": 5.122917509078979,
|
|
"epoch": 9.130094043887148,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.9678603962756137e-05,
|
|
"loss": 4.2749,
|
|
"mean_token_accuracy": 0.24986519962549208,
|
|
"num_tokens": 20472706.0,
|
|
"step": 11650
|
|
},
|
|
{
|
|
"entropy": 5.151169681549073,
|
|
"epoch": 9.134012539184953,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.959152473026687e-05,
|
|
"loss": 4.3162,
|
|
"mean_token_accuracy": 0.24555399417877197,
|
|
"num_tokens": 20481666.0,
|
|
"step": 11655
|
|
},
|
|
{
|
|
"entropy": 5.202339506149292,
|
|
"epoch": 9.137931034482758,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.950483046764971e-05,
|
|
"loss": 4.3083,
|
|
"mean_token_accuracy": 0.24815445095300676,
|
|
"num_tokens": 20490157.0,
|
|
"step": 11660
|
|
},
|
|
{
|
|
"entropy": 5.145409440994262,
|
|
"epoch": 9.141849529780565,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.941852132984145e-05,
|
|
"loss": 4.2371,
|
|
"mean_token_accuracy": 0.2607190445065498,
|
|
"num_tokens": 20499180.0,
|
|
"step": 11665
|
|
},
|
|
{
|
|
"entropy": 5.171636152267456,
|
|
"epoch": 9.14576802507837,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.933259747109042e-05,
|
|
"loss": 4.2647,
|
|
"mean_token_accuracy": 0.25156770646572113,
|
|
"num_tokens": 20508291.0,
|
|
"step": 11670
|
|
},
|
|
{
|
|
"entropy": 5.110851001739502,
|
|
"epoch": 9.149686520376175,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.924705904495656e-05,
|
|
"loss": 4.2252,
|
|
"mean_token_accuracy": 0.2501804083585739,
|
|
"num_tokens": 20516500.0,
|
|
"step": 11675
|
|
},
|
|
{
|
|
"entropy": 5.1351643085479735,
|
|
"epoch": 9.153605015673982,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.916190620431088e-05,
|
|
"loss": 4.2094,
|
|
"mean_token_accuracy": 0.2549794435501099,
|
|
"num_tokens": 20525408.0,
|
|
"step": 11680
|
|
},
|
|
{
|
|
"entropy": 5.139088678359985,
|
|
"epoch": 9.157523510971787,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.907713910133529e-05,
|
|
"loss": 4.2528,
|
|
"mean_token_accuracy": 0.2536822408437729,
|
|
"num_tokens": 20534288.0,
|
|
"step": 11685
|
|
},
|
|
{
|
|
"entropy": 5.13928484916687,
|
|
"epoch": 9.161442006269592,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.8992757887522394e-05,
|
|
"loss": 4.2796,
|
|
"mean_token_accuracy": 0.2484561637043953,
|
|
"num_tokens": 20543011.0,
|
|
"step": 11690
|
|
},
|
|
{
|
|
"entropy": 5.20034704208374,
|
|
"epoch": 9.165360501567399,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.8908762713675074e-05,
|
|
"loss": 4.3027,
|
|
"mean_token_accuracy": 0.24296649694442748,
|
|
"num_tokens": 20551511.0,
|
|
"step": 11695
|
|
},
|
|
{
|
|
"entropy": 5.062475252151489,
|
|
"epoch": 9.169278996865204,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.882515372990632e-05,
|
|
"loss": 4.2238,
|
|
"mean_token_accuracy": 0.2546987310051918,
|
|
"num_tokens": 20560578.0,
|
|
"step": 11700
|
|
},
|
|
{
|
|
"entropy": 5.158094596862793,
|
|
"epoch": 9.173197492163009,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.874193108563895e-05,
|
|
"loss": 4.2707,
|
|
"mean_token_accuracy": 0.2474416509270668,
|
|
"num_tokens": 20569776.0,
|
|
"step": 11705
|
|
},
|
|
{
|
|
"entropy": 5.194279909133911,
|
|
"epoch": 9.177115987460816,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.865909492960536e-05,
|
|
"loss": 4.3754,
|
|
"mean_token_accuracy": 0.24572616815567017,
|
|
"num_tokens": 20579791.0,
|
|
"step": 11710
|
|
},
|
|
{
|
|
"entropy": 5.156844568252564,
|
|
"epoch": 9.181034482758621,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.85766454098471e-05,
|
|
"loss": 4.2651,
|
|
"mean_token_accuracy": 0.24781606942415238,
|
|
"num_tokens": 20588738.0,
|
|
"step": 11715
|
|
},
|
|
{
|
|
"entropy": 5.1566356182098385,
|
|
"epoch": 9.184952978056426,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.849458267371489e-05,
|
|
"loss": 4.2359,
|
|
"mean_token_accuracy": 0.2582111984491348,
|
|
"num_tokens": 20597442.0,
|
|
"step": 11720
|
|
},
|
|
{
|
|
"entropy": 5.21008791923523,
|
|
"epoch": 9.188871473354231,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.841290686786818e-05,
|
|
"loss": 4.3642,
|
|
"mean_token_accuracy": 0.2414606422185898,
|
|
"num_tokens": 20607277.0,
|
|
"step": 11725
|
|
},
|
|
{
|
|
"entropy": 5.205646610260009,
|
|
"epoch": 9.192789968652038,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.833161813827484e-05,
|
|
"loss": 4.3896,
|
|
"mean_token_accuracy": 0.2437393382191658,
|
|
"num_tokens": 20615721.0,
|
|
"step": 11730
|
|
},
|
|
{
|
|
"entropy": 5.174949884414673,
|
|
"epoch": 9.196708463949843,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.825071663021097e-05,
|
|
"loss": 4.2862,
|
|
"mean_token_accuracy": 0.2458395019173622,
|
|
"num_tokens": 20624192.0,
|
|
"step": 11735
|
|
},
|
|
{
|
|
"entropy": 5.139639520645142,
|
|
"epoch": 9.200626959247648,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.817020248826076e-05,
|
|
"loss": 4.2756,
|
|
"mean_token_accuracy": 0.25669341534376144,
|
|
"num_tokens": 20632947.0,
|
|
"step": 11740
|
|
},
|
|
{
|
|
"entropy": 5.157860326766968,
|
|
"epoch": 9.204545454545455,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.8090075856315994e-05,
|
|
"loss": 4.2785,
|
|
"mean_token_accuracy": 0.2491871938109398,
|
|
"num_tokens": 20643252.0,
|
|
"step": 11745
|
|
},
|
|
{
|
|
"entropy": 5.147110891342163,
|
|
"epoch": 9.20846394984326,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.801033687757598e-05,
|
|
"loss": 4.2335,
|
|
"mean_token_accuracy": 0.24949477165937423,
|
|
"num_tokens": 20651416.0,
|
|
"step": 11750
|
|
},
|
|
{
|
|
"entropy": 5.151375436782837,
|
|
"epoch": 9.212382445141065,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 5.793098569454713e-05,
|
|
"loss": 4.279,
|
|
"mean_token_accuracy": 0.2488558605313301,
|
|
"num_tokens": 20660359.0,
|
|
"step": 11755
|
|
},
|
|
{
|
|
"entropy": 5.145324468612671,
|
|
"epoch": 9.216300940438872,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.785202244904296e-05,
|
|
"loss": 4.3282,
|
|
"mean_token_accuracy": 0.2432573616504669,
|
|
"num_tokens": 20669262.0,
|
|
"step": 11760
|
|
},
|
|
{
|
|
"entropy": 5.1958402633667,
|
|
"epoch": 9.220219435736677,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.777344728218356e-05,
|
|
"loss": 4.3048,
|
|
"mean_token_accuracy": 0.2460179314017296,
|
|
"num_tokens": 20677267.0,
|
|
"step": 11765
|
|
},
|
|
{
|
|
"entropy": 5.156361293792725,
|
|
"epoch": 9.224137931034482,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.769526033439547e-05,
|
|
"loss": 4.3053,
|
|
"mean_token_accuracy": 0.2549519255757332,
|
|
"num_tokens": 20685798.0,
|
|
"step": 11770
|
|
},
|
|
{
|
|
"entropy": 5.105309391021729,
|
|
"epoch": 9.228056426332289,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.761746174541148e-05,
|
|
"loss": 4.2181,
|
|
"mean_token_accuracy": 0.2560352966189384,
|
|
"num_tokens": 20694384.0,
|
|
"step": 11775
|
|
},
|
|
{
|
|
"entropy": 5.177390003204346,
|
|
"epoch": 9.231974921630094,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.754005165427027e-05,
|
|
"loss": 4.3068,
|
|
"mean_token_accuracy": 0.25264636278152464,
|
|
"num_tokens": 20703233.0,
|
|
"step": 11780
|
|
},
|
|
{
|
|
"entropy": 5.145479297637939,
|
|
"epoch": 9.2358934169279,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.746303019931623e-05,
|
|
"loss": 4.2592,
|
|
"mean_token_accuracy": 0.25591143369674685,
|
|
"num_tokens": 20711810.0,
|
|
"step": 11785
|
|
},
|
|
{
|
|
"entropy": 5.142142629623413,
|
|
"epoch": 9.239811912225706,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.738639751819919e-05,
|
|
"loss": 4.2743,
|
|
"mean_token_accuracy": 0.25458563715219495,
|
|
"num_tokens": 20720733.0,
|
|
"step": 11790
|
|
},
|
|
{
|
|
"entropy": 5.102780866622925,
|
|
"epoch": 9.243730407523511,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.7310153747874165e-05,
|
|
"loss": 4.2283,
|
|
"mean_token_accuracy": 0.2595305874943733,
|
|
"num_tokens": 20729409.0,
|
|
"step": 11795
|
|
},
|
|
{
|
|
"entropy": 5.18357629776001,
|
|
"epoch": 9.247648902821316,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.7234299024601136e-05,
|
|
"loss": 4.3197,
|
|
"mean_token_accuracy": 0.2472957819700241,
|
|
"num_tokens": 20738113.0,
|
|
"step": 11800
|
|
},
|
|
{
|
|
"entropy": 5.13516321182251,
|
|
"epoch": 9.251567398119123,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.7158833483944796e-05,
|
|
"loss": 4.2476,
|
|
"mean_token_accuracy": 0.26126396358013154,
|
|
"num_tokens": 20746756.0,
|
|
"step": 11805
|
|
},
|
|
{
|
|
"entropy": 5.108553266525268,
|
|
"epoch": 9.255485893416928,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.7083757260774305e-05,
|
|
"loss": 4.2076,
|
|
"mean_token_accuracy": 0.26153937429189683,
|
|
"num_tokens": 20756356.0,
|
|
"step": 11810
|
|
},
|
|
{
|
|
"entropy": 5.141072463989258,
|
|
"epoch": 9.259404388714733,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.7009070489263036e-05,
|
|
"loss": 4.2909,
|
|
"mean_token_accuracy": 0.25231483280658723,
|
|
"num_tokens": 20766118.0,
|
|
"step": 11815
|
|
},
|
|
{
|
|
"entropy": 5.150651597976685,
|
|
"epoch": 9.263322884012538,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.6934773302888365e-05,
|
|
"loss": 4.2743,
|
|
"mean_token_accuracy": 0.2526905477046967,
|
|
"num_tokens": 20775049.0,
|
|
"step": 11820
|
|
},
|
|
{
|
|
"entropy": 5.138689565658569,
|
|
"epoch": 9.267241379310345,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.686086583443143e-05,
|
|
"loss": 4.2624,
|
|
"mean_token_accuracy": 0.2515874445438385,
|
|
"num_tokens": 20783839.0,
|
|
"step": 11825
|
|
},
|
|
{
|
|
"entropy": 5.2018678188323975,
|
|
"epoch": 9.27115987460815,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.678734821597681e-05,
|
|
"loss": 4.3525,
|
|
"mean_token_accuracy": 0.24425503611564636,
|
|
"num_tokens": 20792036.0,
|
|
"step": 11830
|
|
},
|
|
{
|
|
"entropy": 5.186731672286987,
|
|
"epoch": 9.275078369905955,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.671422057891248e-05,
|
|
"loss": 4.2942,
|
|
"mean_token_accuracy": 0.24676901400089263,
|
|
"num_tokens": 20800639.0,
|
|
"step": 11835
|
|
},
|
|
{
|
|
"entropy": 5.114787483215332,
|
|
"epoch": 9.278996865203762,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.6641483053929324e-05,
|
|
"loss": 4.25,
|
|
"mean_token_accuracy": 0.2554177179932594,
|
|
"num_tokens": 20810244.0,
|
|
"step": 11840
|
|
},
|
|
{
|
|
"entropy": 5.18644609451294,
|
|
"epoch": 9.282915360501567,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.656913577102112e-05,
|
|
"loss": 4.2841,
|
|
"mean_token_accuracy": 0.25389973521232606,
|
|
"num_tokens": 20818731.0,
|
|
"step": 11845
|
|
},
|
|
{
|
|
"entropy": 5.092796421051025,
|
|
"epoch": 9.286833855799372,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.649717885948422e-05,
|
|
"loss": 4.2275,
|
|
"mean_token_accuracy": 0.25722129046916964,
|
|
"num_tokens": 20827237.0,
|
|
"step": 11850
|
|
},
|
|
{
|
|
"entropy": 5.115517663955688,
|
|
"epoch": 9.29075235109718,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.642561244791721e-05,
|
|
"loss": 4.2385,
|
|
"mean_token_accuracy": 0.24651909917593,
|
|
"num_tokens": 20835792.0,
|
|
"step": 11855
|
|
},
|
|
{
|
|
"entropy": 5.088610506057739,
|
|
"epoch": 9.294670846394984,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.635443666422097e-05,
|
|
"loss": 4.1836,
|
|
"mean_token_accuracy": 0.25471382290124894,
|
|
"num_tokens": 20844289.0,
|
|
"step": 11860
|
|
},
|
|
{
|
|
"entropy": 5.138458299636841,
|
|
"epoch": 9.29858934169279,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.6283651635598105e-05,
|
|
"loss": 4.2609,
|
|
"mean_token_accuracy": 0.2507146045565605,
|
|
"num_tokens": 20852683.0,
|
|
"step": 11865
|
|
},
|
|
{
|
|
"entropy": 5.144859552383423,
|
|
"epoch": 9.302507836990596,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.6213257488552946e-05,
|
|
"loss": 4.3236,
|
|
"mean_token_accuracy": 0.25336811542510984,
|
|
"num_tokens": 20861999.0,
|
|
"step": 11870
|
|
},
|
|
{
|
|
"entropy": 5.065114879608155,
|
|
"epoch": 9.306426332288401,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.6143254348891254e-05,
|
|
"loss": 4.2192,
|
|
"mean_token_accuracy": 0.2556204229593277,
|
|
"num_tokens": 20870995.0,
|
|
"step": 11875
|
|
},
|
|
{
|
|
"entropy": 5.131003713607788,
|
|
"epoch": 9.310344827586206,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.6073642341720006e-05,
|
|
"loss": 4.2799,
|
|
"mean_token_accuracy": 0.24982597678899765,
|
|
"num_tokens": 20880048.0,
|
|
"step": 11880
|
|
},
|
|
{
|
|
"entropy": 5.079294490814209,
|
|
"epoch": 9.314263322884013,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.600442159144712e-05,
|
|
"loss": 4.2837,
|
|
"mean_token_accuracy": 0.25290794372558595,
|
|
"num_tokens": 20889268.0,
|
|
"step": 11885
|
|
},
|
|
{
|
|
"entropy": 5.127438402175903,
|
|
"epoch": 9.318181818181818,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.593559222178131e-05,
|
|
"loss": 4.2871,
|
|
"mean_token_accuracy": 0.24227110296487808,
|
|
"num_tokens": 20897978.0,
|
|
"step": 11890
|
|
},
|
|
{
|
|
"entropy": 5.137479400634765,
|
|
"epoch": 9.322100313479623,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.5867154355731795e-05,
|
|
"loss": 4.2774,
|
|
"mean_token_accuracy": 0.26173880994319915,
|
|
"num_tokens": 20906387.0,
|
|
"step": 11895
|
|
},
|
|
{
|
|
"entropy": 5.177455043792724,
|
|
"epoch": 9.32601880877743,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.579910811560822e-05,
|
|
"loss": 4.2473,
|
|
"mean_token_accuracy": 0.2566817417740822,
|
|
"num_tokens": 20915309.0,
|
|
"step": 11900
|
|
},
|
|
{
|
|
"entropy": 5.173292112350464,
|
|
"epoch": 9.329937304075235,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.573145362302012e-05,
|
|
"loss": 4.3165,
|
|
"mean_token_accuracy": 0.246855491399765,
|
|
"num_tokens": 20924600.0,
|
|
"step": 11905
|
|
},
|
|
{
|
|
"entropy": 5.201668834686279,
|
|
"epoch": 9.33385579937304,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.566419099887718e-05,
|
|
"loss": 4.3371,
|
|
"mean_token_accuracy": 0.24452054053544997,
|
|
"num_tokens": 20933087.0,
|
|
"step": 11910
|
|
},
|
|
{
|
|
"entropy": 5.0714398384094235,
|
|
"epoch": 9.337774294670847,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.5597320363388544e-05,
|
|
"loss": 4.1747,
|
|
"mean_token_accuracy": 0.2554696649312973,
|
|
"num_tokens": 20940886.0,
|
|
"step": 11915
|
|
},
|
|
{
|
|
"entropy": 5.1693682193756105,
|
|
"epoch": 9.341692789968652,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.55308418360629e-05,
|
|
"loss": 4.2942,
|
|
"mean_token_accuracy": 0.25061956495046617,
|
|
"num_tokens": 20949591.0,
|
|
"step": 11920
|
|
},
|
|
{
|
|
"entropy": 5.1567182540893555,
|
|
"epoch": 9.345611285266457,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.54647555357081e-05,
|
|
"loss": 4.3379,
|
|
"mean_token_accuracy": 0.24172937273979186,
|
|
"num_tokens": 20958001.0,
|
|
"step": 11925
|
|
},
|
|
{
|
|
"entropy": 5.170311975479126,
|
|
"epoch": 9.349529780564263,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.539906158043113e-05,
|
|
"loss": 4.3141,
|
|
"mean_token_accuracy": 0.24481735229492188,
|
|
"num_tokens": 20966671.0,
|
|
"step": 11930
|
|
},
|
|
{
|
|
"entropy": 5.164891576766967,
|
|
"epoch": 9.35344827586207,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.533376008763775e-05,
|
|
"loss": 4.209,
|
|
"mean_token_accuracy": 0.2572582334280014,
|
|
"num_tokens": 20976184.0,
|
|
"step": 11935
|
|
},
|
|
{
|
|
"entropy": 5.151515531539917,
|
|
"epoch": 9.357366771159874,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.526885117403228e-05,
|
|
"loss": 4.1931,
|
|
"mean_token_accuracy": 0.2600631803274155,
|
|
"num_tokens": 20984960.0,
|
|
"step": 11940
|
|
},
|
|
{
|
|
"entropy": 5.167630434036255,
|
|
"epoch": 9.36128526645768,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.520433495561752e-05,
|
|
"loss": 4.2616,
|
|
"mean_token_accuracy": 0.25031389147043226,
|
|
"num_tokens": 20993954.0,
|
|
"step": 11945
|
|
},
|
|
{
|
|
"entropy": 5.149802160263062,
|
|
"epoch": 9.365203761755486,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.5140211547694344e-05,
|
|
"loss": 4.2751,
|
|
"mean_token_accuracy": 0.25358356684446337,
|
|
"num_tokens": 21002391.0,
|
|
"step": 11950
|
|
},
|
|
{
|
|
"entropy": 5.089098119735718,
|
|
"epoch": 9.369122257053291,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.507648106486173e-05,
|
|
"loss": 4.1705,
|
|
"mean_token_accuracy": 0.26195476204156876,
|
|
"num_tokens": 21010996.0,
|
|
"step": 11955
|
|
},
|
|
{
|
|
"entropy": 5.126409387588501,
|
|
"epoch": 9.373040752351097,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.5013143621016376e-05,
|
|
"loss": 4.2165,
|
|
"mean_token_accuracy": 0.2580819472670555,
|
|
"num_tokens": 21020272.0,
|
|
"step": 11960
|
|
},
|
|
{
|
|
"entropy": 5.086160659790039,
|
|
"epoch": 9.376959247648903,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.495019932935256e-05,
|
|
"loss": 4.2167,
|
|
"mean_token_accuracy": 0.25032933950424197,
|
|
"num_tokens": 21029206.0,
|
|
"step": 11965
|
|
},
|
|
{
|
|
"entropy": 5.179266309738159,
|
|
"epoch": 9.380877742946709,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.488764830236194e-05,
|
|
"loss": 4.3563,
|
|
"mean_token_accuracy": 0.23920056074857712,
|
|
"num_tokens": 21038294.0,
|
|
"step": 11970
|
|
},
|
|
{
|
|
"entropy": 5.111535310745239,
|
|
"epoch": 9.384796238244514,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.482549065183333e-05,
|
|
"loss": 4.2347,
|
|
"mean_token_accuracy": 0.25430411845445633,
|
|
"num_tokens": 21047321.0,
|
|
"step": 11975
|
|
},
|
|
{
|
|
"entropy": 5.049402141571045,
|
|
"epoch": 9.38871473354232,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.476372648885255e-05,
|
|
"loss": 4.1613,
|
|
"mean_token_accuracy": 0.2704687789082527,
|
|
"num_tokens": 21055928.0,
|
|
"step": 11980
|
|
},
|
|
{
|
|
"entropy": 5.183910703659057,
|
|
"epoch": 9.392633228840126,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.470235592380219e-05,
|
|
"loss": 4.2984,
|
|
"mean_token_accuracy": 0.24703067243099214,
|
|
"num_tokens": 21064464.0,
|
|
"step": 11985
|
|
},
|
|
{
|
|
"entropy": 5.156952428817749,
|
|
"epoch": 9.39655172413793,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.464137906636133e-05,
|
|
"loss": 4.3143,
|
|
"mean_token_accuracy": 0.24968716353178025,
|
|
"num_tokens": 21073395.0,
|
|
"step": 11990
|
|
},
|
|
{
|
|
"entropy": 5.121283960342407,
|
|
"epoch": 9.400470219435737,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.458079602550557e-05,
|
|
"loss": 4.252,
|
|
"mean_token_accuracy": 0.25148714929819105,
|
|
"num_tokens": 21081841.0,
|
|
"step": 11995
|
|
},
|
|
{
|
|
"entropy": 5.162518882751465,
|
|
"epoch": 9.404388714733543,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.452060690950662e-05,
|
|
"loss": 4.3022,
|
|
"mean_token_accuracy": 0.24145271331071855,
|
|
"num_tokens": 21090826.0,
|
|
"step": 12000
|
|
},
|
|
{
|
|
"epoch": 9.404388714733543,
|
|
"eval_entropy": 5.068024970764338,
|
|
"eval_loss": 5.646355628967285,
|
|
"eval_mean_token_accuracy": 0.17484631499116735,
|
|
"eval_num_tokens": 21090826.0,
|
|
"eval_runtime": 3.0264,
|
|
"eval_samples_per_second": 1362.015,
|
|
"eval_steps_per_second": 170.5,
|
|
"step": 12000
|
|
}
|
|
],
|
|
"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": 2.837220391872e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|