c867a9a1f1
Model: fpadovani/ind-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
6101 lines
165 KiB
JSON
6101 lines
165 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.3510971786833856,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.789229106903076,
|
|
"epoch": 0.003918495297805642,
|
|
"grad_norm": 17.125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.3537,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9174.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.8115012645721436,
|
|
"epoch": 0.007836990595611285,
|
|
"grad_norm": 19.625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.2452,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 17790.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.899150800704956,
|
|
"epoch": 0.011755485893416929,
|
|
"grad_norm": 24.25,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.9044,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 25850.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.367580604553223,
|
|
"epoch": 0.01567398119122257,
|
|
"grad_norm": 32.5,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.1444,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 35194.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 8.583788537979126,
|
|
"epoch": 0.019592476489028215,
|
|
"grad_norm": 7.5,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.3911,
|
|
"mean_token_accuracy": 0.00023256096756085753,
|
|
"num_tokens": 44218.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.630141735076904,
|
|
"epoch": 0.023510971786833857,
|
|
"grad_norm": 3.234375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.7102,
|
|
"mean_token_accuracy": 0.014597209030762314,
|
|
"num_tokens": 53397.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.695956897735595,
|
|
"epoch": 0.0274294670846395,
|
|
"grad_norm": 3.0,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.4664,
|
|
"mean_token_accuracy": 0.01781447734683752,
|
|
"num_tokens": 62749.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.673796558380127,
|
|
"epoch": 0.03134796238244514,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1632,
|
|
"mean_token_accuracy": 0.0182854525744915,
|
|
"num_tokens": 71721.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.631663513183593,
|
|
"epoch": 0.03526645768025078,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8792,
|
|
"mean_token_accuracy": 0.03653257880359888,
|
|
"num_tokens": 79844.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.549837970733643,
|
|
"epoch": 0.03918495297805643,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.7665,
|
|
"mean_token_accuracy": 0.04605128690600395,
|
|
"num_tokens": 88866.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.509830379486084,
|
|
"epoch": 0.04310344827586207,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.6605,
|
|
"mean_token_accuracy": 0.044031094387173654,
|
|
"num_tokens": 97918.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.524475193023681,
|
|
"epoch": 0.047021943573667714,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5619,
|
|
"mean_token_accuracy": 0.04601282589137554,
|
|
"num_tokens": 107043.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.512676334381103,
|
|
"epoch": 0.050940438871473356,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4987,
|
|
"mean_token_accuracy": 0.04643600396811962,
|
|
"num_tokens": 116000.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.477371215820312,
|
|
"epoch": 0.054858934169279,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.4179,
|
|
"mean_token_accuracy": 0.04148977212607861,
|
|
"num_tokens": 124559.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.470399188995362,
|
|
"epoch": 0.05877742946708464,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2945,
|
|
"mean_token_accuracy": 0.04952896051108837,
|
|
"num_tokens": 132868.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.453096103668212,
|
|
"epoch": 0.06269592476489028,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.2848,
|
|
"mean_token_accuracy": 0.05274602882564068,
|
|
"num_tokens": 141286.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.431593227386475,
|
|
"epoch": 0.06661442006269593,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1405,
|
|
"mean_token_accuracy": 0.05872356928884983,
|
|
"num_tokens": 150406.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.36865291595459,
|
|
"epoch": 0.07053291536050156,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.0678,
|
|
"mean_token_accuracy": 0.059385529905557635,
|
|
"num_tokens": 158770.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.264866065979003,
|
|
"epoch": 0.07445141065830721,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.9633,
|
|
"mean_token_accuracy": 0.06288341507315635,
|
|
"num_tokens": 167763.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.183263969421386,
|
|
"epoch": 0.07836990595611286,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.819,
|
|
"mean_token_accuracy": 0.0607046652585268,
|
|
"num_tokens": 177306.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.144334697723389,
|
|
"epoch": 0.0822884012539185,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.7349,
|
|
"mean_token_accuracy": 0.06028640605509281,
|
|
"num_tokens": 186014.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.06214361190796,
|
|
"epoch": 0.08620689655172414,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.5758,
|
|
"mean_token_accuracy": 0.06410923898220063,
|
|
"num_tokens": 194122.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.952830028533935,
|
|
"epoch": 0.09012539184952978,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.4698,
|
|
"mean_token_accuracy": 0.05936008468270302,
|
|
"num_tokens": 203097.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.820581531524658,
|
|
"epoch": 0.09404388714733543,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.3405,
|
|
"mean_token_accuracy": 0.06221077479422092,
|
|
"num_tokens": 211413.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.732498264312744,
|
|
"epoch": 0.09796238244514106,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.2718,
|
|
"mean_token_accuracy": 0.061625415459275246,
|
|
"num_tokens": 220550.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.504752349853515,
|
|
"epoch": 0.10188087774294671,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.0995,
|
|
"mean_token_accuracy": 0.0649514563381672,
|
|
"num_tokens": 229197.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.307702922821045,
|
|
"epoch": 0.10579937304075235,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.0979,
|
|
"mean_token_accuracy": 0.05685936994850636,
|
|
"num_tokens": 238479.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.162922954559326,
|
|
"epoch": 0.109717868338558,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.9442,
|
|
"mean_token_accuracy": 0.059861503541469574,
|
|
"num_tokens": 246318.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.96123743057251,
|
|
"epoch": 0.11363636363636363,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.9513,
|
|
"mean_token_accuracy": 0.05959276556968689,
|
|
"num_tokens": 254783.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.760778617858886,
|
|
"epoch": 0.11755485893416928,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.7945,
|
|
"mean_token_accuracy": 0.06369670145213605,
|
|
"num_tokens": 263416.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.68117027282715,
|
|
"epoch": 0.12147335423197492,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.8147,
|
|
"mean_token_accuracy": 0.0631796333938837,
|
|
"num_tokens": 271930.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.476777839660645,
|
|
"epoch": 0.12539184952978055,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.7159,
|
|
"mean_token_accuracy": 0.06549291461706161,
|
|
"num_tokens": 280546.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.308262157440186,
|
|
"epoch": 0.12931034482758622,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7078,
|
|
"mean_token_accuracy": 0.06602046675980092,
|
|
"num_tokens": 288813.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.279962158203125,
|
|
"epoch": 0.13322884012539185,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.6847,
|
|
"mean_token_accuracy": 0.06443305909633637,
|
|
"num_tokens": 297966.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.152728843688966,
|
|
"epoch": 0.1371473354231975,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.7467,
|
|
"mean_token_accuracy": 0.06189337000250816,
|
|
"num_tokens": 307135.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.145699501037598,
|
|
"epoch": 0.14106583072100312,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.6581,
|
|
"mean_token_accuracy": 0.06488074697554111,
|
|
"num_tokens": 315546.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.149786376953125,
|
|
"epoch": 0.14498432601880878,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.6538,
|
|
"mean_token_accuracy": 0.06405953019857406,
|
|
"num_tokens": 323930.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.983444690704346,
|
|
"epoch": 0.14890282131661442,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.5166,
|
|
"mean_token_accuracy": 0.07129846066236496,
|
|
"num_tokens": 332419.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.974339866638184,
|
|
"epoch": 0.15282131661442006,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.6157,
|
|
"mean_token_accuracy": 0.06940566822886467,
|
|
"num_tokens": 341362.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.973450374603272,
|
|
"epoch": 0.15673981191222572,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.5198,
|
|
"mean_token_accuracy": 0.07214542552828788,
|
|
"num_tokens": 349395.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.946638202667236,
|
|
"epoch": 0.16065830721003135,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.545,
|
|
"mean_token_accuracy": 0.06696730926632881,
|
|
"num_tokens": 358413.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.863241577148438,
|
|
"epoch": 0.164576802507837,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.6316,
|
|
"mean_token_accuracy": 0.06982938721776008,
|
|
"num_tokens": 366489.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.940403842926026,
|
|
"epoch": 0.16849529780564262,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.5023,
|
|
"mean_token_accuracy": 0.06740010716021061,
|
|
"num_tokens": 375335.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.874079847335816,
|
|
"epoch": 0.1724137931034483,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.5555,
|
|
"mean_token_accuracy": 0.07188675999641418,
|
|
"num_tokens": 384276.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.911562585830689,
|
|
"epoch": 0.17633228840125392,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.5929,
|
|
"mean_token_accuracy": 0.06714313849806786,
|
|
"num_tokens": 393571.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.920306205749512,
|
|
"epoch": 0.18025078369905956,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.5193,
|
|
"mean_token_accuracy": 0.07089398205280303,
|
|
"num_tokens": 401865.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.848536252975464,
|
|
"epoch": 0.1841692789968652,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.4905,
|
|
"mean_token_accuracy": 0.07226377129554748,
|
|
"num_tokens": 410518.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.869985485076905,
|
|
"epoch": 0.18808777429467086,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.4997,
|
|
"mean_token_accuracy": 0.07303371652960777,
|
|
"num_tokens": 419769.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.837644481658936,
|
|
"epoch": 0.1920062695924765,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.437,
|
|
"mean_token_accuracy": 0.0742616519331932,
|
|
"num_tokens": 428204.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.897941255569458,
|
|
"epoch": 0.19592476489028213,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.5267,
|
|
"mean_token_accuracy": 0.06978406608104706,
|
|
"num_tokens": 436594.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.80897855758667,
|
|
"epoch": 0.19984326018808776,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.3256,
|
|
"mean_token_accuracy": 0.07486266531050205,
|
|
"num_tokens": 444645.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.8133704662323,
|
|
"epoch": 0.20376175548589343,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3726,
|
|
"mean_token_accuracy": 0.07738698273897171,
|
|
"num_tokens": 453016.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.736161422729492,
|
|
"epoch": 0.20768025078369906,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.4507,
|
|
"mean_token_accuracy": 0.06978621035814285,
|
|
"num_tokens": 462116.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.664476203918457,
|
|
"epoch": 0.2115987460815047,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3961,
|
|
"mean_token_accuracy": 0.07115238644182682,
|
|
"num_tokens": 470807.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.7677568912506105,
|
|
"epoch": 0.21551724137931033,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.4818,
|
|
"mean_token_accuracy": 0.07023664973676205,
|
|
"num_tokens": 479592.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.912611389160157,
|
|
"epoch": 0.219435736677116,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.4068,
|
|
"mean_token_accuracy": 0.07160313390195369,
|
|
"num_tokens": 488107.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.755217599868774,
|
|
"epoch": 0.22335423197492163,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.4212,
|
|
"mean_token_accuracy": 0.07538670524954796,
|
|
"num_tokens": 496775.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.762103033065796,
|
|
"epoch": 0.22727272727272727,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.4447,
|
|
"mean_token_accuracy": 0.07036296911537647,
|
|
"num_tokens": 505415.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.757038116455078,
|
|
"epoch": 0.23119122257053293,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.4344,
|
|
"mean_token_accuracy": 0.074312524497509,
|
|
"num_tokens": 514447.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.7855620861053465,
|
|
"epoch": 0.23510971786833856,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.4189,
|
|
"mean_token_accuracy": 0.07684484757483005,
|
|
"num_tokens": 522998.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.774819326400757,
|
|
"epoch": 0.2390282131661442,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.3794,
|
|
"mean_token_accuracy": 0.07512850686907768,
|
|
"num_tokens": 531542.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.7300177097320555,
|
|
"epoch": 0.24294670846394983,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.3247,
|
|
"mean_token_accuracy": 0.07261879369616508,
|
|
"num_tokens": 540143.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.773348236083985,
|
|
"epoch": 0.2468652037617555,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.4017,
|
|
"mean_token_accuracy": 0.07950926274061203,
|
|
"num_tokens": 549156.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.7285737037658695,
|
|
"epoch": 0.2507836990595611,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.2557,
|
|
"mean_token_accuracy": 0.07481630519032478,
|
|
"num_tokens": 557522.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.654256391525268,
|
|
"epoch": 0.2547021943573668,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.331,
|
|
"mean_token_accuracy": 0.07940150126814842,
|
|
"num_tokens": 566650.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.672131299972534,
|
|
"epoch": 0.25862068965517243,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.2864,
|
|
"mean_token_accuracy": 0.07869702018797398,
|
|
"num_tokens": 576219.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.713160848617553,
|
|
"epoch": 0.26253918495297807,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.374,
|
|
"mean_token_accuracy": 0.07567794360220433,
|
|
"num_tokens": 584304.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.634349060058594,
|
|
"epoch": 0.2664576802507837,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.3003,
|
|
"mean_token_accuracy": 0.07832697704434395,
|
|
"num_tokens": 593163.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.6303457736969,
|
|
"epoch": 0.27037617554858934,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.2164,
|
|
"mean_token_accuracy": 0.07754571028053761,
|
|
"num_tokens": 602077.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.6355628490448,
|
|
"epoch": 0.274294670846395,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.3284,
|
|
"mean_token_accuracy": 0.08122679404914379,
|
|
"num_tokens": 610009.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.685596513748169,
|
|
"epoch": 0.2782131661442006,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.362,
|
|
"mean_token_accuracy": 0.07597106769680977,
|
|
"num_tokens": 619282.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.619720935821533,
|
|
"epoch": 0.28213166144200624,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.2893,
|
|
"mean_token_accuracy": 0.08018167689442635,
|
|
"num_tokens": 628138.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.6796112060546875,
|
|
"epoch": 0.28605015673981193,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.1745,
|
|
"mean_token_accuracy": 0.08669476807117463,
|
|
"num_tokens": 637021.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.619413709640503,
|
|
"epoch": 0.28996865203761757,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.3161,
|
|
"mean_token_accuracy": 0.07477690353989601,
|
|
"num_tokens": 646703.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.623689222335815,
|
|
"epoch": 0.2938871473354232,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.1981,
|
|
"mean_token_accuracy": 0.08060777708888053,
|
|
"num_tokens": 655616.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.556978368759156,
|
|
"epoch": 0.29780564263322884,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1994,
|
|
"mean_token_accuracy": 0.08719867020845413,
|
|
"num_tokens": 663783.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.5845866203308105,
|
|
"epoch": 0.3017241379310345,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.2094,
|
|
"mean_token_accuracy": 0.08289245739579201,
|
|
"num_tokens": 671855.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.527840566635132,
|
|
"epoch": 0.3056426332288401,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.2219,
|
|
"mean_token_accuracy": 0.07747755497694016,
|
|
"num_tokens": 680981.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.6136561870574955,
|
|
"epoch": 0.30956112852664575,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.1491,
|
|
"mean_token_accuracy": 0.08336339518427849,
|
|
"num_tokens": 689294.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.469813442230224,
|
|
"epoch": 0.31347962382445144,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1035,
|
|
"mean_token_accuracy": 0.08146922513842583,
|
|
"num_tokens": 697703.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.550826740264893,
|
|
"epoch": 0.31739811912225707,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.2372,
|
|
"mean_token_accuracy": 0.08058681413531303,
|
|
"num_tokens": 706792.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.606830406188965,
|
|
"epoch": 0.3213166144200627,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.1473,
|
|
"mean_token_accuracy": 0.08346155509352685,
|
|
"num_tokens": 715864.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.3859583854675295,
|
|
"epoch": 0.32523510971786834,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.1975,
|
|
"mean_token_accuracy": 0.0853593334555626,
|
|
"num_tokens": 723921.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.5406107902526855,
|
|
"epoch": 0.329153605015674,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.2071,
|
|
"mean_token_accuracy": 0.08046000376343727,
|
|
"num_tokens": 732797.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.510403490066528,
|
|
"epoch": 0.3330721003134796,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.0654,
|
|
"mean_token_accuracy": 0.0873202033340931,
|
|
"num_tokens": 741248.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.501159954071045,
|
|
"epoch": 0.33699059561128525,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.1615,
|
|
"mean_token_accuracy": 0.08190247714519501,
|
|
"num_tokens": 749766.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.408373832702637,
|
|
"epoch": 0.3409090909090909,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.1268,
|
|
"mean_token_accuracy": 0.08113668784499169,
|
|
"num_tokens": 758695.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.44956521987915,
|
|
"epoch": 0.3448275862068966,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.1248,
|
|
"mean_token_accuracy": 0.08192591443657875,
|
|
"num_tokens": 767624.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.422909212112427,
|
|
"epoch": 0.3487460815047022,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.117,
|
|
"mean_token_accuracy": 0.0853099413216114,
|
|
"num_tokens": 776616.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.365292644500732,
|
|
"epoch": 0.35266457680250785,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.1317,
|
|
"mean_token_accuracy": 0.08413158729672432,
|
|
"num_tokens": 786147.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.536469745635986,
|
|
"epoch": 0.3565830721003135,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.2317,
|
|
"mean_token_accuracy": 0.08228531405329705,
|
|
"num_tokens": 795213.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.461417722702026,
|
|
"epoch": 0.3605015673981191,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.0349,
|
|
"mean_token_accuracy": 0.09094136133790016,
|
|
"num_tokens": 803118.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.444038438796997,
|
|
"epoch": 0.36442006269592475,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0219,
|
|
"mean_token_accuracy": 0.09442275986075402,
|
|
"num_tokens": 811358.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.324700260162354,
|
|
"epoch": 0.3683385579937304,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.0256,
|
|
"mean_token_accuracy": 0.08778790757060051,
|
|
"num_tokens": 819653.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.2960240840911865,
|
|
"epoch": 0.3722570532915361,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.0511,
|
|
"mean_token_accuracy": 0.08624262139201164,
|
|
"num_tokens": 828462.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.437795686721802,
|
|
"epoch": 0.3761755485893417,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.1429,
|
|
"mean_token_accuracy": 0.0912679947912693,
|
|
"num_tokens": 836204.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.2959794998168945,
|
|
"epoch": 0.38009404388714735,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.0169,
|
|
"mean_token_accuracy": 0.09246607050299645,
|
|
"num_tokens": 845032.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.4119359970092775,
|
|
"epoch": 0.384012539184953,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.0308,
|
|
"mean_token_accuracy": 0.08805579245090485,
|
|
"num_tokens": 853324.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.404975366592407,
|
|
"epoch": 0.3879310344827586,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.992,
|
|
"mean_token_accuracy": 0.1035026639699936,
|
|
"num_tokens": 861640.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.385119247436523,
|
|
"epoch": 0.39184952978056425,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.1744,
|
|
"mean_token_accuracy": 0.082430200278759,
|
|
"num_tokens": 870758.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.39184952978056425,
|
|
"eval_entropy": 7.167502074278603,
|
|
"eval_loss": 7.156619548797607,
|
|
"eval_mean_token_accuracy": 0.08891707594152684,
|
|
"eval_num_tokens": 870758.0,
|
|
"eval_runtime": 2.8546,
|
|
"eval_samples_per_second": 1444.004,
|
|
"eval_steps_per_second": 180.763,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.3540655136108395,
|
|
"epoch": 0.3957680250783699,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.0906,
|
|
"mean_token_accuracy": 0.0840302512049675,
|
|
"num_tokens": 879968.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.318256664276123,
|
|
"epoch": 0.3996865203761755,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.0471,
|
|
"mean_token_accuracy": 0.08631112575531005,
|
|
"num_tokens": 888804.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.259663057327271,
|
|
"epoch": 0.4036050156739812,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.0624,
|
|
"mean_token_accuracy": 0.08012468516826629,
|
|
"num_tokens": 898593.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.2833295345306395,
|
|
"epoch": 0.40752351097178685,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.0799,
|
|
"mean_token_accuracy": 0.08342506065964699,
|
|
"num_tokens": 906873.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.3285657405853275,
|
|
"epoch": 0.4114420062695925,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.0359,
|
|
"mean_token_accuracy": 0.0934828281402588,
|
|
"num_tokens": 915080.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.422465038299561,
|
|
"epoch": 0.4153605015673981,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.0396,
|
|
"mean_token_accuracy": 0.08701496720314025,
|
|
"num_tokens": 923741.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.216967153549194,
|
|
"epoch": 0.41927899686520376,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.9459,
|
|
"mean_token_accuracy": 0.0904500350356102,
|
|
"num_tokens": 932455.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.235203742980957,
|
|
"epoch": 0.4231974921630094,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.9492,
|
|
"mean_token_accuracy": 0.09165697544813156,
|
|
"num_tokens": 941064.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.23210301399231,
|
|
"epoch": 0.427115987460815,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.9795,
|
|
"mean_token_accuracy": 0.0916426420211792,
|
|
"num_tokens": 950261.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.305574369430542,
|
|
"epoch": 0.43103448275862066,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.0461,
|
|
"mean_token_accuracy": 0.0905070275068283,
|
|
"num_tokens": 959151.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.299721527099609,
|
|
"epoch": 0.43495297805642635,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.0677,
|
|
"mean_token_accuracy": 0.09062978066504002,
|
|
"num_tokens": 968441.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.134230327606201,
|
|
"epoch": 0.438871473354232,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.9371,
|
|
"mean_token_accuracy": 0.09018276557326317,
|
|
"num_tokens": 977058.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.334470558166504,
|
|
"epoch": 0.4427899686520376,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.9519,
|
|
"mean_token_accuracy": 0.08950636759400368,
|
|
"num_tokens": 986531.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.123916816711426,
|
|
"epoch": 0.44670846394984326,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.8108,
|
|
"mean_token_accuracy": 0.09824811816215515,
|
|
"num_tokens": 995081.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.103513240814209,
|
|
"epoch": 0.4506269592476489,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.8802,
|
|
"mean_token_accuracy": 0.09345417022705078,
|
|
"num_tokens": 1003459.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.101778936386109,
|
|
"epoch": 0.45454545454545453,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.9239,
|
|
"mean_token_accuracy": 0.09018066227436065,
|
|
"num_tokens": 1012420.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.261321687698365,
|
|
"epoch": 0.45846394984326017,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.9954,
|
|
"mean_token_accuracy": 0.09160361662507058,
|
|
"num_tokens": 1021198.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.202180290222168,
|
|
"epoch": 0.46238244514106586,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.9694,
|
|
"mean_token_accuracy": 0.09484207406640052,
|
|
"num_tokens": 1030023.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.164184045791626,
|
|
"epoch": 0.4663009404388715,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.8295,
|
|
"mean_token_accuracy": 0.0953464850783348,
|
|
"num_tokens": 1038824.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.216883039474487,
|
|
"epoch": 0.4702194357366771,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.0882,
|
|
"mean_token_accuracy": 0.09343515783548355,
|
|
"num_tokens": 1048207.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.304975414276123,
|
|
"epoch": 0.47413793103448276,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.0559,
|
|
"mean_token_accuracy": 0.08962106555700303,
|
|
"num_tokens": 1057354.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.198044776916504,
|
|
"epoch": 0.4780564263322884,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.9893,
|
|
"mean_token_accuracy": 0.09330410435795784,
|
|
"num_tokens": 1066825.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.137618494033814,
|
|
"epoch": 0.48197492163009403,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.9155,
|
|
"mean_token_accuracy": 0.08996602892875671,
|
|
"num_tokens": 1075752.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.007559776306152,
|
|
"epoch": 0.48589341692789967,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.8323,
|
|
"mean_token_accuracy": 0.09399376884102821,
|
|
"num_tokens": 1084826.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.223559427261352,
|
|
"epoch": 0.4898119122257053,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.9666,
|
|
"mean_token_accuracy": 0.09359999522566795,
|
|
"num_tokens": 1093425.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.108446407318115,
|
|
"epoch": 0.493730407523511,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.9828,
|
|
"mean_token_accuracy": 0.089838757365942,
|
|
"num_tokens": 1102619.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.145089435577392,
|
|
"epoch": 0.49764890282131663,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.8601,
|
|
"mean_token_accuracy": 0.09595804288983345,
|
|
"num_tokens": 1111337.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.005008172988892,
|
|
"epoch": 0.5015673981191222,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7035,
|
|
"mean_token_accuracy": 0.0990886114537716,
|
|
"num_tokens": 1119381.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.96934700012207,
|
|
"epoch": 0.5054858934169278,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.7597,
|
|
"mean_token_accuracy": 0.09665322229266167,
|
|
"num_tokens": 1127825.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.11965799331665,
|
|
"epoch": 0.5094043887147336,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.8203,
|
|
"mean_token_accuracy": 0.09057366773486138,
|
|
"num_tokens": 1136750.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.017684459686279,
|
|
"epoch": 0.5133228840125392,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.8499,
|
|
"mean_token_accuracy": 0.09456580057740212,
|
|
"num_tokens": 1145739.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.034306764602661,
|
|
"epoch": 0.5172413793103449,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.8973,
|
|
"mean_token_accuracy": 0.09757498279213905,
|
|
"num_tokens": 1154415.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.052440977096557,
|
|
"epoch": 0.5211598746081505,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7751,
|
|
"mean_token_accuracy": 0.09611302688717842,
|
|
"num_tokens": 1162480.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.977913856506348,
|
|
"epoch": 0.5250783699059561,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.8559,
|
|
"mean_token_accuracy": 0.09223495721817017,
|
|
"num_tokens": 1170774.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.030881881713867,
|
|
"epoch": 0.5289968652037618,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.7924,
|
|
"mean_token_accuracy": 0.09683787003159523,
|
|
"num_tokens": 1179629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.053485012054443,
|
|
"epoch": 0.5329153605015674,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.938,
|
|
"mean_token_accuracy": 0.09229604452848435,
|
|
"num_tokens": 1189111.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.108199834823608,
|
|
"epoch": 0.536833855799373,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.9495,
|
|
"mean_token_accuracy": 0.09128881692886352,
|
|
"num_tokens": 1198827.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.092297840118408,
|
|
"epoch": 0.5407523510971787,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.8482,
|
|
"mean_token_accuracy": 0.09970205947756768,
|
|
"num_tokens": 1207089.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.057426071166992,
|
|
"epoch": 0.5446708463949843,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.8435,
|
|
"mean_token_accuracy": 0.08852889537811279,
|
|
"num_tokens": 1216509.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.900876426696778,
|
|
"epoch": 0.54858934169279,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6756,
|
|
"mean_token_accuracy": 0.10320580378174782,
|
|
"num_tokens": 1225684.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.005167055130005,
|
|
"epoch": 0.5525078369905956,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.7057,
|
|
"mean_token_accuracy": 0.10480915755033493,
|
|
"num_tokens": 1233675.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.892497873306274,
|
|
"epoch": 0.5564263322884012,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.7091,
|
|
"mean_token_accuracy": 0.10067695155739784,
|
|
"num_tokens": 1242147.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.934285736083984,
|
|
"epoch": 0.5603448275862069,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.8249,
|
|
"mean_token_accuracy": 0.09867035746574401,
|
|
"num_tokens": 1251127.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.978139781951905,
|
|
"epoch": 0.5642633228840125,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.7775,
|
|
"mean_token_accuracy": 0.09740801975131035,
|
|
"num_tokens": 1260589.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.05702314376831,
|
|
"epoch": 0.5681818181818182,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.8522,
|
|
"mean_token_accuracy": 0.09754758477210998,
|
|
"num_tokens": 1269281.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.951777076721191,
|
|
"epoch": 0.5721003134796239,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.7117,
|
|
"mean_token_accuracy": 0.09818840324878693,
|
|
"num_tokens": 1278033.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.001934242248535,
|
|
"epoch": 0.5760188087774295,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.7808,
|
|
"mean_token_accuracy": 0.10354246944189072,
|
|
"num_tokens": 1285847.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.932463216781616,
|
|
"epoch": 0.5799373040752351,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.7726,
|
|
"mean_token_accuracy": 0.09513568431138993,
|
|
"num_tokens": 1294770.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.036710739135742,
|
|
"epoch": 0.5838557993730408,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.9512,
|
|
"mean_token_accuracy": 0.0940048098564148,
|
|
"num_tokens": 1304282.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.997774505615235,
|
|
"epoch": 0.5877742946708464,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.7923,
|
|
"mean_token_accuracy": 0.09595935121178627,
|
|
"num_tokens": 1313218.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.966546869277954,
|
|
"epoch": 0.591692789968652,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.754,
|
|
"mean_token_accuracy": 0.0985157236456871,
|
|
"num_tokens": 1322093.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.83478045463562,
|
|
"epoch": 0.5956112852664577,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.7857,
|
|
"mean_token_accuracy": 0.09971616193652152,
|
|
"num_tokens": 1330718.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.954968690872192,
|
|
"epoch": 0.5995297805642633,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.6457,
|
|
"mean_token_accuracy": 0.10431547313928605,
|
|
"num_tokens": 1339672.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.861670970916748,
|
|
"epoch": 0.603448275862069,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.7001,
|
|
"mean_token_accuracy": 0.10293109342455864,
|
|
"num_tokens": 1348587.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.8885609149932865,
|
|
"epoch": 0.6073667711598746,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.6826,
|
|
"mean_token_accuracy": 0.10689334198832512,
|
|
"num_tokens": 1357597.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.883357572555542,
|
|
"epoch": 0.6112852664576802,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.7744,
|
|
"mean_token_accuracy": 0.101459039747715,
|
|
"num_tokens": 1366246.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.89829683303833,
|
|
"epoch": 0.6152037617554859,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.6695,
|
|
"mean_token_accuracy": 0.10532263070344924,
|
|
"num_tokens": 1374664.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.8325498580932615,
|
|
"epoch": 0.6191222570532915,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.6889,
|
|
"mean_token_accuracy": 0.10108358785510063,
|
|
"num_tokens": 1382765.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.7815876483917235,
|
|
"epoch": 0.6230407523510971,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.6442,
|
|
"mean_token_accuracy": 0.10871021300554276,
|
|
"num_tokens": 1391708.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.786308908462525,
|
|
"epoch": 0.6269592476489029,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.814,
|
|
"mean_token_accuracy": 0.09444232732057571,
|
|
"num_tokens": 1401373.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.926651859283448,
|
|
"epoch": 0.6308777429467085,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.6124,
|
|
"mean_token_accuracy": 0.10876154825091362,
|
|
"num_tokens": 1409702.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.87204418182373,
|
|
"epoch": 0.6347962382445141,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.8084,
|
|
"mean_token_accuracy": 0.10008606985211373,
|
|
"num_tokens": 1419596.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.807423734664917,
|
|
"epoch": 0.6387147335423198,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.7461,
|
|
"mean_token_accuracy": 0.10076582729816437,
|
|
"num_tokens": 1428437.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.9204614639282225,
|
|
"epoch": 0.6426332288401254,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.6547,
|
|
"mean_token_accuracy": 0.10060450211167335,
|
|
"num_tokens": 1436764.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.812792015075684,
|
|
"epoch": 0.646551724137931,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.7007,
|
|
"mean_token_accuracy": 0.10875345095992088,
|
|
"num_tokens": 1445755.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.902109909057617,
|
|
"epoch": 0.6504702194357367,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.7627,
|
|
"mean_token_accuracy": 0.10212339907884598,
|
|
"num_tokens": 1454550.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.757165718078613,
|
|
"epoch": 0.6543887147335423,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.6529,
|
|
"mean_token_accuracy": 0.1076908752322197,
|
|
"num_tokens": 1463039.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.920673799514771,
|
|
"epoch": 0.658307210031348,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.7125,
|
|
"mean_token_accuracy": 0.10059169679880142,
|
|
"num_tokens": 1472029.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.742834091186523,
|
|
"epoch": 0.6622257053291536,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.723,
|
|
"mean_token_accuracy": 0.10695556625723839,
|
|
"num_tokens": 1481588.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.7755883693695065,
|
|
"epoch": 0.6661442006269592,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.6131,
|
|
"mean_token_accuracy": 0.1035246841609478,
|
|
"num_tokens": 1489915.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.87349271774292,
|
|
"epoch": 0.6700626959247649,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.7211,
|
|
"mean_token_accuracy": 0.10070185288786888,
|
|
"num_tokens": 1497920.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.742019605636597,
|
|
"epoch": 0.6739811912225705,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5571,
|
|
"mean_token_accuracy": 0.10514650270342826,
|
|
"num_tokens": 1506965.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.755401372909546,
|
|
"epoch": 0.6778996865203761,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.6626,
|
|
"mean_token_accuracy": 0.10886923670768738,
|
|
"num_tokens": 1516028.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.824488735198974,
|
|
"epoch": 0.6818181818181818,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.6802,
|
|
"mean_token_accuracy": 0.10832962691783905,
|
|
"num_tokens": 1524579.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.729337167739868,
|
|
"epoch": 0.6857366771159875,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.605,
|
|
"mean_token_accuracy": 0.1062053769826889,
|
|
"num_tokens": 1534285.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.7189655780792235,
|
|
"epoch": 0.6896551724137931,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5898,
|
|
"mean_token_accuracy": 0.10706395953893662,
|
|
"num_tokens": 1543701.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.891758966445923,
|
|
"epoch": 0.6935736677115988,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.8576,
|
|
"mean_token_accuracy": 0.09462928622961045,
|
|
"num_tokens": 1552715.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.6106942653656,
|
|
"epoch": 0.6974921630094044,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4728,
|
|
"mean_token_accuracy": 0.1057778999209404,
|
|
"num_tokens": 1561523.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.685323572158813,
|
|
"epoch": 0.70141065830721,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.6995,
|
|
"mean_token_accuracy": 0.10680059865117073,
|
|
"num_tokens": 1570057.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.810699081420898,
|
|
"epoch": 0.7053291536050157,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5487,
|
|
"mean_token_accuracy": 0.10731169655919075,
|
|
"num_tokens": 1578795.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.618034505844117,
|
|
"epoch": 0.7092476489028213,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.627,
|
|
"mean_token_accuracy": 0.10445040464401245,
|
|
"num_tokens": 1588108.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.834760522842407,
|
|
"epoch": 0.713166144200627,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5897,
|
|
"mean_token_accuracy": 0.10842868015170097,
|
|
"num_tokens": 1596545.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.617375135421753,
|
|
"epoch": 0.7170846394984326,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.5774,
|
|
"mean_token_accuracy": 0.11101481318473816,
|
|
"num_tokens": 1605098.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.7286604881286625,
|
|
"epoch": 0.7210031347962382,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5928,
|
|
"mean_token_accuracy": 0.11244359910488129,
|
|
"num_tokens": 1613760.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.685993957519531,
|
|
"epoch": 0.7249216300940439,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.6311,
|
|
"mean_token_accuracy": 0.10528192594647408,
|
|
"num_tokens": 1622345.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.825484371185302,
|
|
"epoch": 0.7288401253918495,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.6198,
|
|
"mean_token_accuracy": 0.10753775164484977,
|
|
"num_tokens": 1630706.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.736838388442993,
|
|
"epoch": 0.7327586206896551,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.6851,
|
|
"mean_token_accuracy": 0.10634701699018478,
|
|
"num_tokens": 1639880.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.709425592422486,
|
|
"epoch": 0.7366771159874608,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.5189,
|
|
"mean_token_accuracy": 0.10119672417640686,
|
|
"num_tokens": 1648325.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.547716999053955,
|
|
"epoch": 0.7405956112852664,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4928,
|
|
"mean_token_accuracy": 0.11216101795434952,
|
|
"num_tokens": 1656951.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.6757955074310305,
|
|
"epoch": 0.7445141065830722,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.6264,
|
|
"mean_token_accuracy": 0.1065959431231022,
|
|
"num_tokens": 1665832.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.653597450256347,
|
|
"epoch": 0.7484326018808778,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.5458,
|
|
"mean_token_accuracy": 0.10681739151477813,
|
|
"num_tokens": 1675116.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.55257477760315,
|
|
"epoch": 0.7523510971786834,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5317,
|
|
"mean_token_accuracy": 0.11514699757099152,
|
|
"num_tokens": 1684195.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.689774370193481,
|
|
"epoch": 0.7562695924764891,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4895,
|
|
"mean_token_accuracy": 0.11039396822452545,
|
|
"num_tokens": 1693025.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.481554937362671,
|
|
"epoch": 0.7601880877742947,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4836,
|
|
"mean_token_accuracy": 0.10764000788331032,
|
|
"num_tokens": 1701210.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.6965916633605955,
|
|
"epoch": 0.7641065830721003,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.5847,
|
|
"mean_token_accuracy": 0.10125251486897469,
|
|
"num_tokens": 1711026.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.65680046081543,
|
|
"epoch": 0.768025078369906,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5874,
|
|
"mean_token_accuracy": 0.10824618190526962,
|
|
"num_tokens": 1719441.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.599896430969238,
|
|
"epoch": 0.7719435736677116,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.5849,
|
|
"mean_token_accuracy": 0.10633337944746017,
|
|
"num_tokens": 1728300.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.620725393295288,
|
|
"epoch": 0.7758620689655172,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4769,
|
|
"mean_token_accuracy": 0.11418513432145119,
|
|
"num_tokens": 1736364.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.466502714157104,
|
|
"epoch": 0.7797805642633229,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3224,
|
|
"mean_token_accuracy": 0.1263233445584774,
|
|
"num_tokens": 1745043.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.654029464721679,
|
|
"epoch": 0.7836990595611285,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.4667,
|
|
"mean_token_accuracy": 0.11487501338124276,
|
|
"num_tokens": 1754008.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.7836990595611285,
|
|
"eval_entropy": 6.3669652088668,
|
|
"eval_loss": 6.630230903625488,
|
|
"eval_mean_token_accuracy": 0.11402813254227472,
|
|
"eval_num_tokens": 1754008.0,
|
|
"eval_runtime": 2.8378,
|
|
"eval_samples_per_second": 1452.509,
|
|
"eval_steps_per_second": 181.828,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.444351482391357,
|
|
"epoch": 0.7876175548589341,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999998713243189,
|
|
"loss": 6.5346,
|
|
"mean_token_accuracy": 0.11141253411769866,
|
|
"num_tokens": 1763592.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.584927940368653,
|
|
"epoch": 0.7915360501567398,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999993485796164,
|
|
"loss": 6.403,
|
|
"mean_token_accuracy": 0.1155676744878292,
|
|
"num_tokens": 1773111.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.580813217163086,
|
|
"epoch": 0.7954545454545454,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004999984237245962,
|
|
"loss": 6.4624,
|
|
"mean_token_accuracy": 0.11948382258415222,
|
|
"num_tokens": 1781230.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.740645933151245,
|
|
"epoch": 0.799373040752351,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004999970967609109,
|
|
"loss": 6.6906,
|
|
"mean_token_accuracy": 0.1132524773478508,
|
|
"num_tokens": 1790324.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.513845157623291,
|
|
"epoch": 0.8032915360501567,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004999953676909322,
|
|
"loss": 6.5348,
|
|
"mean_token_accuracy": 0.1044821061193943,
|
|
"num_tokens": 1798849.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.588596630096435,
|
|
"epoch": 0.8072100313479624,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004999932365177502,
|
|
"loss": 6.4097,
|
|
"mean_token_accuracy": 0.11618589833378792,
|
|
"num_tokens": 1808019.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.624749708175659,
|
|
"epoch": 0.8111285266457681,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999907032451735,
|
|
"loss": 6.5302,
|
|
"mean_token_accuracy": 0.11330178454518318,
|
|
"num_tokens": 1816550.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.507697629928589,
|
|
"epoch": 0.8150470219435737,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999877678777296,
|
|
"loss": 6.4044,
|
|
"mean_token_accuracy": 0.12338297590613365,
|
|
"num_tokens": 1825238.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.4800478458404545,
|
|
"epoch": 0.8189655172413793,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999844304206645,
|
|
"loss": 6.3892,
|
|
"mean_token_accuracy": 0.11694491282105446,
|
|
"num_tokens": 1833645.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.564522838592529,
|
|
"epoch": 0.822884012539185,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999806908799428,
|
|
"loss": 6.4637,
|
|
"mean_token_accuracy": 0.11064697802066803,
|
|
"num_tokens": 1842892.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.528485631942749,
|
|
"epoch": 0.8268025078369906,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999765492622474,
|
|
"loss": 6.5046,
|
|
"mean_token_accuracy": 0.11715267226099968,
|
|
"num_tokens": 1851400.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.596149349212647,
|
|
"epoch": 0.8307210031347962,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999720055749804,
|
|
"loss": 6.4464,
|
|
"mean_token_accuracy": 0.11808413341641426,
|
|
"num_tokens": 1860088.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.574128770828247,
|
|
"epoch": 0.8346394984326019,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999670598262619,
|
|
"loss": 6.4971,
|
|
"mean_token_accuracy": 0.11734354719519616,
|
|
"num_tokens": 1868558.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.427276229858398,
|
|
"epoch": 0.8385579937304075,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999617120249308,
|
|
"loss": 6.4897,
|
|
"mean_token_accuracy": 0.11183837950229644,
|
|
"num_tokens": 1877624.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.671892261505127,
|
|
"epoch": 0.8424764890282131,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999559621805445,
|
|
"loss": 6.5975,
|
|
"mean_token_accuracy": 0.11244333311915397,
|
|
"num_tokens": 1886505.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.452223205566407,
|
|
"epoch": 0.8463949843260188,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999498103033788,
|
|
"loss": 6.3456,
|
|
"mean_token_accuracy": 0.12142532989382744,
|
|
"num_tokens": 1895701.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.390986537933349,
|
|
"epoch": 0.8503134796238244,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999432564044284,
|
|
"loss": 6.3008,
|
|
"mean_token_accuracy": 0.12294157966971397,
|
|
"num_tokens": 1904640.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.614175415039062,
|
|
"epoch": 0.85423197492163,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999363004954058,
|
|
"loss": 6.5357,
|
|
"mean_token_accuracy": 0.11326182112097741,
|
|
"num_tokens": 1914216.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.48394021987915,
|
|
"epoch": 0.8581504702194357,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.4573,
|
|
"mean_token_accuracy": 0.11912907212972641,
|
|
"num_tokens": 1923280.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.611083984375,
|
|
"epoch": 0.8620689655172413,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999211826975884,
|
|
"loss": 6.4764,
|
|
"mean_token_accuracy": 0.11999231576919556,
|
|
"num_tokens": 1931403.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.5545590877532955,
|
|
"epoch": 0.8659874608150471,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999130208358114,
|
|
"loss": 6.4882,
|
|
"mean_token_accuracy": 0.11854184418916702,
|
|
"num_tokens": 1940774.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.561131238937378,
|
|
"epoch": 0.8699059561128527,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999044570179983,
|
|
"loss": 6.5687,
|
|
"mean_token_accuracy": 0.11162992268800735,
|
|
"num_tokens": 1950045.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.606174612045288,
|
|
"epoch": 0.8738244514106583,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998954912594541,
|
|
"loss": 6.4405,
|
|
"mean_token_accuracy": 0.11669689863920212,
|
|
"num_tokens": 1959303.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.492517614364624,
|
|
"epoch": 0.877742946708464,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998861235762018,
|
|
"loss": 6.424,
|
|
"mean_token_accuracy": 0.1151531957089901,
|
|
"num_tokens": 1968583.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.590066766738891,
|
|
"epoch": 0.8816614420062696,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998763539849832,
|
|
"loss": 6.5317,
|
|
"mean_token_accuracy": 0.11314806714653969,
|
|
"num_tokens": 1977489.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.413249540328979,
|
|
"epoch": 0.8855799373040752,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998661825032579,
|
|
"loss": 6.306,
|
|
"mean_token_accuracy": 0.1186840571463108,
|
|
"num_tokens": 1985308.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.476549482345581,
|
|
"epoch": 0.8894984326018809,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998556091492043,
|
|
"loss": 6.4658,
|
|
"mean_token_accuracy": 0.11185984387993812,
|
|
"num_tokens": 1993930.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.523940944671631,
|
|
"epoch": 0.8934169278996865,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998446339417184,
|
|
"loss": 6.4676,
|
|
"mean_token_accuracy": 0.10710446015000344,
|
|
"num_tokens": 2003802.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.495583724975586,
|
|
"epoch": 0.8973354231974922,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998332569004149,
|
|
"loss": 6.3809,
|
|
"mean_token_accuracy": 0.11923198625445366,
|
|
"num_tokens": 2011957.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.457823705673218,
|
|
"epoch": 0.9012539184952978,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998214780456263,
|
|
"loss": 6.3813,
|
|
"mean_token_accuracy": 0.12053523734211921,
|
|
"num_tokens": 2020741.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.47773175239563,
|
|
"epoch": 0.9051724137931034,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998092973984033,
|
|
"loss": 6.4479,
|
|
"mean_token_accuracy": 0.11438319608569145,
|
|
"num_tokens": 2029804.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.543193197250366,
|
|
"epoch": 0.9090909090909091,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997967149805147,
|
|
"loss": 6.5013,
|
|
"mean_token_accuracy": 0.116908498108387,
|
|
"num_tokens": 2038732.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.49971113204956,
|
|
"epoch": 0.9130094043887147,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997837308144474,
|
|
"loss": 6.4002,
|
|
"mean_token_accuracy": 0.12262723073363305,
|
|
"num_tokens": 2046362.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.484364652633667,
|
|
"epoch": 0.9169278996865203,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997703449234062,
|
|
"loss": 6.3721,
|
|
"mean_token_accuracy": 0.11172138154506683,
|
|
"num_tokens": 2056018.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.384702682495117,
|
|
"epoch": 0.920846394984326,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997565573313139,
|
|
"loss": 6.3547,
|
|
"mean_token_accuracy": 0.11563765183091164,
|
|
"num_tokens": 2065077.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.540685033798217,
|
|
"epoch": 0.9247648902821317,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997423680628109,
|
|
"loss": 6.3473,
|
|
"mean_token_accuracy": 0.1140012837946415,
|
|
"num_tokens": 2073405.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.3114136219024655,
|
|
"epoch": 0.9286833855799373,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000499727777143256,
|
|
"loss": 6.3174,
|
|
"mean_token_accuracy": 0.11831804886460304,
|
|
"num_tokens": 2081891.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.565793228149414,
|
|
"epoch": 0.932601880877743,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997127845987255,
|
|
"loss": 6.3823,
|
|
"mean_token_accuracy": 0.11906470507383346,
|
|
"num_tokens": 2090732.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.347681045532227,
|
|
"epoch": 0.9365203761755486,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996973904560135,
|
|
"loss": 6.306,
|
|
"mean_token_accuracy": 0.12649127021431922,
|
|
"num_tokens": 2099352.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.5376229763031,
|
|
"epoch": 0.9404388714733543,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996815947426317,
|
|
"loss": 6.346,
|
|
"mean_token_accuracy": 0.1162750244140625,
|
|
"num_tokens": 2107365.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.353217506408692,
|
|
"epoch": 0.9443573667711599,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996653974868097,
|
|
"loss": 6.218,
|
|
"mean_token_accuracy": 0.1250108003616333,
|
|
"num_tokens": 2116192.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.376781034469604,
|
|
"epoch": 0.9482758620689655,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996487987174946,
|
|
"loss": 6.3167,
|
|
"mean_token_accuracy": 0.12126539349555969,
|
|
"num_tokens": 2124410.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.452775669097901,
|
|
"epoch": 0.9521943573667712,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996317984643512,
|
|
"loss": 6.3483,
|
|
"mean_token_accuracy": 0.11980568394064903,
|
|
"num_tokens": 2132799.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.459938097000122,
|
|
"epoch": 0.9561128526645768,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996143967577615,
|
|
"loss": 6.4935,
|
|
"mean_token_accuracy": 0.11375040411949158,
|
|
"num_tokens": 2141589.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.466125345230102,
|
|
"epoch": 0.9600313479623824,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004995965936288254,
|
|
"loss": 6.3714,
|
|
"mean_token_accuracy": 0.1200237862765789,
|
|
"num_tokens": 2150704.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.399578809738159,
|
|
"epoch": 0.9639498432601881,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995783891093597,
|
|
"loss": 6.2958,
|
|
"mean_token_accuracy": 0.12307887598872184,
|
|
"num_tokens": 2159587.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.330127573013305,
|
|
"epoch": 0.9678683385579937,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499559783231899,
|
|
"loss": 6.2928,
|
|
"mean_token_accuracy": 0.11578266024589538,
|
|
"num_tokens": 2169071.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.421966409683227,
|
|
"epoch": 0.9717868338557993,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995407760296952,
|
|
"loss": 6.3688,
|
|
"mean_token_accuracy": 0.11582615077495576,
|
|
"num_tokens": 2178248.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.425446796417236,
|
|
"epoch": 0.975705329153605,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995213675367169,
|
|
"loss": 6.2732,
|
|
"mean_token_accuracy": 0.11606336981058121,
|
|
"num_tokens": 2187520.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.3828057765960695,
|
|
"epoch": 0.9796238244514106,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004995015577876504,
|
|
"loss": 6.3415,
|
|
"mean_token_accuracy": 0.11500431299209594,
|
|
"num_tokens": 2196861.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.361384725570678,
|
|
"epoch": 0.9835423197492164,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499481346817899,
|
|
"loss": 6.2502,
|
|
"mean_token_accuracy": 0.12506693974137306,
|
|
"num_tokens": 2205523.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.345028877258301,
|
|
"epoch": 0.987460815047022,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004994607346635829,
|
|
"loss": 6.215,
|
|
"mean_token_accuracy": 0.12786355316638948,
|
|
"num_tokens": 2214291.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.440300559997558,
|
|
"epoch": 0.9913793103448276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004994397213615394,
|
|
"loss": 6.3295,
|
|
"mean_token_accuracy": 0.11995508670806884,
|
|
"num_tokens": 2223093.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.406629800796509,
|
|
"epoch": 0.9952978056426333,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004994183069493225,
|
|
"loss": 6.2832,
|
|
"mean_token_accuracy": 0.1210085429251194,
|
|
"num_tokens": 2231551.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.222655868530273,
|
|
"epoch": 0.9992163009404389,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993964914652034,
|
|
"loss": 6.2829,
|
|
"mean_token_accuracy": 0.1268168218433857,
|
|
"num_tokens": 2240650.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.511307573318481,
|
|
"epoch": 1.0031347962382444,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004993742749481699,
|
|
"loss": 6.2046,
|
|
"mean_token_accuracy": 0.12171700075268746,
|
|
"num_tokens": 2249392.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.226445293426513,
|
|
"epoch": 1.0070532915360502,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004993516574379265,
|
|
"loss": 6.0221,
|
|
"mean_token_accuracy": 0.12889871895313262,
|
|
"num_tokens": 2258100.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.256803798675537,
|
|
"epoch": 1.0109717868338557,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004993286389748943,
|
|
"loss": 6.0869,
|
|
"mean_token_accuracy": 0.13081810474395753,
|
|
"num_tokens": 2266661.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.348307514190674,
|
|
"epoch": 1.0148902821316614,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004993052196002112,
|
|
"loss": 6.1897,
|
|
"mean_token_accuracy": 0.12489164769649505,
|
|
"num_tokens": 2275475.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.266294240951538,
|
|
"epoch": 1.0188087774294672,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004992813993557312,
|
|
"loss": 6.1038,
|
|
"mean_token_accuracy": 0.12707522958517076,
|
|
"num_tokens": 2284077.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.261200428009033,
|
|
"epoch": 1.0227272727272727,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499257178284025,
|
|
"loss": 6.1127,
|
|
"mean_token_accuracy": 0.12442725002765656,
|
|
"num_tokens": 2293029.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.262345981597901,
|
|
"epoch": 1.0266457680250785,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004992325564283797,
|
|
"loss": 6.0351,
|
|
"mean_token_accuracy": 0.13578465953469276,
|
|
"num_tokens": 2301574.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.176527118682861,
|
|
"epoch": 1.030564263322884,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004992075338327984,
|
|
"loss": 6.0354,
|
|
"mean_token_accuracy": 0.13138693422079087,
|
|
"num_tokens": 2310366.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.3460509300231935,
|
|
"epoch": 1.0344827586206897,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004991821105420006,
|
|
"loss": 6.1592,
|
|
"mean_token_accuracy": 0.1265966959297657,
|
|
"num_tokens": 2318943.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.320439672470092,
|
|
"epoch": 1.0384012539184952,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004991562866014219,
|
|
"loss": 6.1486,
|
|
"mean_token_accuracy": 0.12730259895324708,
|
|
"num_tokens": 2327173.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.314588499069214,
|
|
"epoch": 1.042319749216301,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.1095,
|
|
"mean_token_accuracy": 0.1259642593562603,
|
|
"num_tokens": 2336194.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.226988649368286,
|
|
"epoch": 1.0462382445141065,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991034369562438,
|
|
"loss": 6.1824,
|
|
"mean_token_accuracy": 0.13008867129683493,
|
|
"num_tokens": 2344721.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.249088525772095,
|
|
"epoch": 1.0501567398119123,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004990764113460953,
|
|
"loss": 6.1052,
|
|
"mean_token_accuracy": 0.1238970473408699,
|
|
"num_tokens": 2353406.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.400324296951294,
|
|
"epoch": 1.0540752351097178,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004990489852750675,
|
|
"loss": 6.2738,
|
|
"mean_token_accuracy": 0.11888742819428444,
|
|
"num_tokens": 2361896.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.248968696594238,
|
|
"epoch": 1.0579937304075235,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004990211587921751,
|
|
"loss": 6.1014,
|
|
"mean_token_accuracy": 0.1269259825348854,
|
|
"num_tokens": 2371861.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.221920919418335,
|
|
"epoch": 1.061912225705329,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004989929319471487,
|
|
"loss": 6.0119,
|
|
"mean_token_accuracy": 0.13410131856799126,
|
|
"num_tokens": 2381555.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.21340970993042,
|
|
"epoch": 1.0658307210031348,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004989643047904341,
|
|
"loss": 6.019,
|
|
"mean_token_accuracy": 0.13659467175602913,
|
|
"num_tokens": 2389851.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.126182651519775,
|
|
"epoch": 1.0697492163009406,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004989352773731928,
|
|
"loss": 5.9411,
|
|
"mean_token_accuracy": 0.13557685017585755,
|
|
"num_tokens": 2397764.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.2651125431060795,
|
|
"epoch": 1.073667711598746,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004989058497473014,
|
|
"loss": 6.1303,
|
|
"mean_token_accuracy": 0.13204967901110648,
|
|
"num_tokens": 2406688.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.271961975097656,
|
|
"epoch": 1.0775862068965518,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004988760219653518,
|
|
"loss": 6.1871,
|
|
"mean_token_accuracy": 0.12600790411233903,
|
|
"num_tokens": 2415475.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.261941576004029,
|
|
"epoch": 1.0815047021943573,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004988457940806513,
|
|
"loss": 6.1287,
|
|
"mean_token_accuracy": 0.1264335960149765,
|
|
"num_tokens": 2425719.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.207798719406128,
|
|
"epoch": 1.085423197492163,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004988151661472218,
|
|
"loss": 6.0311,
|
|
"mean_token_accuracy": 0.13111473768949508,
|
|
"num_tokens": 2435138.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.299766397476196,
|
|
"epoch": 1.0893416927899686,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004987841382198006,
|
|
"loss": 6.1646,
|
|
"mean_token_accuracy": 0.1282002493739128,
|
|
"num_tokens": 2443976.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.123603677749633,
|
|
"epoch": 1.0932601880877744,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004987527103538394,
|
|
"loss": 6.0608,
|
|
"mean_token_accuracy": 0.1322481580078602,
|
|
"num_tokens": 2453287.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.353779220581055,
|
|
"epoch": 1.09717868338558,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000498720882605505,
|
|
"loss": 6.1542,
|
|
"mean_token_accuracy": 0.12652794942259787,
|
|
"num_tokens": 2462054.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.106056785583496,
|
|
"epoch": 1.1010971786833856,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004986886550316788,
|
|
"loss": 6.0653,
|
|
"mean_token_accuracy": 0.13403623849153518,
|
|
"num_tokens": 2470333.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.214111185073852,
|
|
"epoch": 1.1050156739811912,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004986560276899565,
|
|
"loss": 6.0972,
|
|
"mean_token_accuracy": 0.1270233377814293,
|
|
"num_tokens": 2479621.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.246433115005493,
|
|
"epoch": 1.108934169278997,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004986230006386488,
|
|
"loss": 6.0655,
|
|
"mean_token_accuracy": 0.12891049459576606,
|
|
"num_tokens": 2488020.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.274629163742065,
|
|
"epoch": 1.1128526645768024,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004985895739367802,
|
|
"loss": 6.1085,
|
|
"mean_token_accuracy": 0.12846623882651328,
|
|
"num_tokens": 2497779.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.158310031890869,
|
|
"epoch": 1.1167711598746082,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004985557476440895,
|
|
"loss": 6.0415,
|
|
"mean_token_accuracy": 0.12804254293441772,
|
|
"num_tokens": 2506791.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.2934564590454105,
|
|
"epoch": 1.1206896551724137,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00049852152182103,
|
|
"loss": 6.1559,
|
|
"mean_token_accuracy": 0.1281921535730362,
|
|
"num_tokens": 2515576.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.240061712265015,
|
|
"epoch": 1.1246081504702194,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004984868965287686,
|
|
"loss": 6.0674,
|
|
"mean_token_accuracy": 0.1283339627087116,
|
|
"num_tokens": 2523933.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.1820862770080565,
|
|
"epoch": 1.1285266457680252,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004984518718291864,
|
|
"loss": 6.0056,
|
|
"mean_token_accuracy": 0.12331884428858757,
|
|
"num_tokens": 2532453.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.157241773605347,
|
|
"epoch": 1.1324451410658307,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004984164477848783,
|
|
"loss": 6.064,
|
|
"mean_token_accuracy": 0.12599845975637436,
|
|
"num_tokens": 2541149.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.237585258483887,
|
|
"epoch": 1.1363636363636362,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004983806244591528,
|
|
"loss": 6.0312,
|
|
"mean_token_accuracy": 0.13384631648659706,
|
|
"num_tokens": 2550302.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.050166320800781,
|
|
"epoch": 1.140282131661442,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004983444019160318,
|
|
"loss": 5.8966,
|
|
"mean_token_accuracy": 0.13734075129032136,
|
|
"num_tokens": 2558507.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.258242225646972,
|
|
"epoch": 1.1442006269592477,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004983077802202511,
|
|
"loss": 6.1788,
|
|
"mean_token_accuracy": 0.1290736488997936,
|
|
"num_tokens": 2567487.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.212359237670898,
|
|
"epoch": 1.1481191222570533,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004982707594372595,
|
|
"loss": 6.0304,
|
|
"mean_token_accuracy": 0.12699214443564416,
|
|
"num_tokens": 2576232.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.1208006858825685,
|
|
"epoch": 1.152037617554859,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000498233339633219,
|
|
"loss": 6.0329,
|
|
"mean_token_accuracy": 0.13209858015179635,
|
|
"num_tokens": 2585081.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.143232536315918,
|
|
"epoch": 1.1559561128526645,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000498195520875005,
|
|
"loss": 6.0115,
|
|
"mean_token_accuracy": 0.12682496458292009,
|
|
"num_tokens": 2593890.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.200338554382324,
|
|
"epoch": 1.1598746081504703,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004981573032302059,
|
|
"loss": 6.0958,
|
|
"mean_token_accuracy": 0.1289630651473999,
|
|
"num_tokens": 2602011.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.149346542358399,
|
|
"epoch": 1.1637931034482758,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004981186867671225,
|
|
"loss": 5.9939,
|
|
"mean_token_accuracy": 0.1361051805317402,
|
|
"num_tokens": 2610366.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.113060522079468,
|
|
"epoch": 1.1677115987460815,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004980796715547691,
|
|
"loss": 6.0004,
|
|
"mean_token_accuracy": 0.13451212123036385,
|
|
"num_tokens": 2619737.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.229964303970337,
|
|
"epoch": 1.171630094043887,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004980402576628717,
|
|
"loss": 6.0233,
|
|
"mean_token_accuracy": 0.13166360333561897,
|
|
"num_tokens": 2628176.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.0950154781341555,
|
|
"epoch": 1.1755485893416928,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004980004451618697,
|
|
"loss": 6.0384,
|
|
"mean_token_accuracy": 0.13299553468823433,
|
|
"num_tokens": 2636195.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.1755485893416928,
|
|
"eval_entropy": 6.165283305700435,
|
|
"eval_loss": 6.313858985900879,
|
|
"eval_mean_token_accuracy": 0.1281498552105108,
|
|
"eval_num_tokens": 2636195.0,
|
|
"eval_runtime": 2.8434,
|
|
"eval_samples_per_second": 1449.67,
|
|
"eval_steps_per_second": 181.472,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.294966459274292,
|
|
"epoch": 1.1794670846394983,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004979602341229144,
|
|
"loss": 6.1167,
|
|
"mean_token_accuracy": 0.13073519617319107,
|
|
"num_tokens": 2645035.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.177581834793091,
|
|
"epoch": 1.183385579937304,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004979196246178694,
|
|
"loss": 6.0716,
|
|
"mean_token_accuracy": 0.13222582265734673,
|
|
"num_tokens": 2653239.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.1044244289398195,
|
|
"epoch": 1.1873040752351098,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004978786167193105,
|
|
"loss": 6.0998,
|
|
"mean_token_accuracy": 0.13036856576800346,
|
|
"num_tokens": 2662532.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.214647531509399,
|
|
"epoch": 1.1912225705329154,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004978372105005254,
|
|
"loss": 6.0841,
|
|
"mean_token_accuracy": 0.12801975160837173,
|
|
"num_tokens": 2671867.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.091090679168701,
|
|
"epoch": 1.1951410658307209,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004977954060355139,
|
|
"loss": 6.0188,
|
|
"mean_token_accuracy": 0.1304849162697792,
|
|
"num_tokens": 2680693.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.281388568878174,
|
|
"epoch": 1.1990595611285266,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004977532033989871,
|
|
"loss": 6.0883,
|
|
"mean_token_accuracy": 0.1320122368633747,
|
|
"num_tokens": 2689228.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.080573844909668,
|
|
"epoch": 1.2029780564263324,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004977106026663681,
|
|
"loss": 5.9666,
|
|
"mean_token_accuracy": 0.12833520472049714,
|
|
"num_tokens": 2698483.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.26170859336853,
|
|
"epoch": 1.206896551724138,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004976676039137914,
|
|
"loss": 6.1056,
|
|
"mean_token_accuracy": 0.126655612885952,
|
|
"num_tokens": 2706920.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.0963341236114506,
|
|
"epoch": 1.2108150470219436,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004976242072181026,
|
|
"loss": 5.9939,
|
|
"mean_token_accuracy": 0.13806139603257178,
|
|
"num_tokens": 2715954.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.142190265655517,
|
|
"epoch": 1.2147335423197492,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004975804126568587,
|
|
"loss": 6.0112,
|
|
"mean_token_accuracy": 0.13544101044535636,
|
|
"num_tokens": 2724890.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.178206443786621,
|
|
"epoch": 1.218652037617555,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004975362203083277,
|
|
"loss": 6.037,
|
|
"mean_token_accuracy": 0.12968128994107248,
|
|
"num_tokens": 2733688.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.206477451324463,
|
|
"epoch": 1.2225705329153604,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004974916302514886,
|
|
"loss": 6.0428,
|
|
"mean_token_accuracy": 0.12749146521091462,
|
|
"num_tokens": 2743364.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.040554809570312,
|
|
"epoch": 1.2264890282131662,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.1001,
|
|
"mean_token_accuracy": 0.1249497301876545,
|
|
"num_tokens": 2754276.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.225854349136353,
|
|
"epoch": 1.2304075235109717,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004974012573323545,
|
|
"loss": 5.883,
|
|
"mean_token_accuracy": 0.14270371645689012,
|
|
"num_tokens": 2762680.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.993971490859986,
|
|
"epoch": 1.2343260188087775,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004973554746315709,
|
|
"loss": 5.9893,
|
|
"mean_token_accuracy": 0.14108432978391647,
|
|
"num_tokens": 2771161.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.177520227432251,
|
|
"epoch": 1.238244514106583,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004973092945455009,
|
|
"loss": 5.9578,
|
|
"mean_token_accuracy": 0.14068967550992967,
|
|
"num_tokens": 2779321.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.1567606925964355,
|
|
"epoch": 1.2421630094043887,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004972627171566757,
|
|
"loss": 6.1195,
|
|
"mean_token_accuracy": 0.13284459933638573,
|
|
"num_tokens": 2788425.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.097169017791748,
|
|
"epoch": 1.2460815047021945,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004972157425483368,
|
|
"loss": 5.9674,
|
|
"mean_token_accuracy": 0.1364126294851303,
|
|
"num_tokens": 2796747.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.055204153060913,
|
|
"epoch": 1.25,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004971683708044353,
|
|
"loss": 6.0172,
|
|
"mean_token_accuracy": 0.1358514852821827,
|
|
"num_tokens": 2805335.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.133439636230468,
|
|
"epoch": 1.2539184952978055,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004971206020096323,
|
|
"loss": 5.9043,
|
|
"mean_token_accuracy": 0.13950337022542952,
|
|
"num_tokens": 2813350.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.165687799453735,
|
|
"epoch": 1.2578369905956113,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004970724362492984,
|
|
"loss": 6.0934,
|
|
"mean_token_accuracy": 0.13864696100354196,
|
|
"num_tokens": 2821922.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.101494073867798,
|
|
"epoch": 1.261755485893417,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004970238736095135,
|
|
"loss": 5.9931,
|
|
"mean_token_accuracy": 0.1329360119998455,
|
|
"num_tokens": 2830832.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.160137414932251,
|
|
"epoch": 1.2656739811912225,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004969749141770671,
|
|
"loss": 6.0053,
|
|
"mean_token_accuracy": 0.13711865544319152,
|
|
"num_tokens": 2839699.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.093396949768066,
|
|
"epoch": 1.2695924764890283,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004969255580394575,
|
|
"loss": 6.0811,
|
|
"mean_token_accuracy": 0.13654499500989914,
|
|
"num_tokens": 2848293.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.201265144348144,
|
|
"epoch": 1.2735109717868338,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000496875805284892,
|
|
"loss": 6.0327,
|
|
"mean_token_accuracy": 0.13467409685254098,
|
|
"num_tokens": 2856468.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.086582183837891,
|
|
"epoch": 1.2774294670846396,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004968256560022871,
|
|
"loss": 5.9502,
|
|
"mean_token_accuracy": 0.1351684033870697,
|
|
"num_tokens": 2865366.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.079522609710693,
|
|
"epoch": 1.281347962382445,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004967751102812676,
|
|
"loss": 6.0118,
|
|
"mean_token_accuracy": 0.13846911042928695,
|
|
"num_tokens": 2873520.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.133967542648316,
|
|
"epoch": 1.2852664576802508,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004967241682121669,
|
|
"loss": 6.0771,
|
|
"mean_token_accuracy": 0.13166362345218657,
|
|
"num_tokens": 2883324.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.170493459701538,
|
|
"epoch": 1.2891849529780564,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004966728298860267,
|
|
"loss": 6.0178,
|
|
"mean_token_accuracy": 0.13498894423246383,
|
|
"num_tokens": 2892200.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.1070939064025875,
|
|
"epoch": 1.293103448275862,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004966210953945969,
|
|
"loss": 5.9858,
|
|
"mean_token_accuracy": 0.13817497938871384,
|
|
"num_tokens": 2900597.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.192414903640747,
|
|
"epoch": 1.2970219435736676,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004965689648303355,
|
|
"loss": 6.0118,
|
|
"mean_token_accuracy": 0.13011416494846345,
|
|
"num_tokens": 2909869.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.000342845916748,
|
|
"epoch": 1.3009404388714734,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004965164382864083,
|
|
"loss": 6.0278,
|
|
"mean_token_accuracy": 0.1339656464755535,
|
|
"num_tokens": 2919166.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.232674837112427,
|
|
"epoch": 1.3048589341692791,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004964635158566886,
|
|
"loss": 6.0383,
|
|
"mean_token_accuracy": 0.13142292499542235,
|
|
"num_tokens": 2927884.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.099806451797486,
|
|
"epoch": 1.3087774294670846,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004964101976357574,
|
|
"loss": 5.9802,
|
|
"mean_token_accuracy": 0.14010540917515754,
|
|
"num_tokens": 2936737.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.117660140991211,
|
|
"epoch": 1.3126959247648902,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000496356483718903,
|
|
"loss": 6.0705,
|
|
"mean_token_accuracy": 0.13366047590970992,
|
|
"num_tokens": 2944203.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.167582654953003,
|
|
"epoch": 1.316614420062696,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004963023742021211,
|
|
"loss": 6.0739,
|
|
"mean_token_accuracy": 0.1311576023697853,
|
|
"num_tokens": 2952639.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.16412582397461,
|
|
"epoch": 1.3205329153605017,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000496247869182114,
|
|
"loss": 6.0037,
|
|
"mean_token_accuracy": 0.1342972233891487,
|
|
"num_tokens": 2961482.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.074027061462402,
|
|
"epoch": 1.3244514106583072,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004961929687562909,
|
|
"loss": 6.0286,
|
|
"mean_token_accuracy": 0.1309918761253357,
|
|
"num_tokens": 2970927.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.046236276626587,
|
|
"epoch": 1.328369905956113,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004961376730227681,
|
|
"loss": 5.9461,
|
|
"mean_token_accuracy": 0.1402522951364517,
|
|
"num_tokens": 2979329.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.147798299789429,
|
|
"epoch": 1.3322884012539185,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004960819820803675,
|
|
"loss": 5.9328,
|
|
"mean_token_accuracy": 0.14073051139712334,
|
|
"num_tokens": 2986894.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.0277563571929935,
|
|
"epoch": 1.3362068965517242,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004960258960286185,
|
|
"loss": 5.9188,
|
|
"mean_token_accuracy": 0.1406124599277973,
|
|
"num_tokens": 2994721.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.188552713394165,
|
|
"epoch": 1.3401253918495297,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004959694149677553,
|
|
"loss": 6.0716,
|
|
"mean_token_accuracy": 0.12887891680002211,
|
|
"num_tokens": 3003165.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.130622291564942,
|
|
"epoch": 1.3440438871473355,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004959125389987193,
|
|
"loss": 6.0519,
|
|
"mean_token_accuracy": 0.12692190557718278,
|
|
"num_tokens": 3012178.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.024265241622925,
|
|
"epoch": 1.347962382445141,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004958552682231567,
|
|
"loss": 5.957,
|
|
"mean_token_accuracy": 0.1394257813692093,
|
|
"num_tokens": 3021685.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.13150839805603,
|
|
"epoch": 1.3518808777429467,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004957976027434199,
|
|
"loss": 6.0119,
|
|
"mean_token_accuracy": 0.12856094017624856,
|
|
"num_tokens": 3030667.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.137048673629761,
|
|
"epoch": 1.3557993730407523,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004957395426625663,
|
|
"loss": 6.064,
|
|
"mean_token_accuracy": 0.13227416425943375,
|
|
"num_tokens": 3039501.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.055467891693115,
|
|
"epoch": 1.359717868338558,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004956810880843587,
|
|
"loss": 5.9946,
|
|
"mean_token_accuracy": 0.13389588594436647,
|
|
"num_tokens": 3048547.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.150904512405395,
|
|
"epoch": 1.3636363636363638,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000495622239113265,
|
|
"loss": 5.9482,
|
|
"mean_token_accuracy": 0.13820023238658904,
|
|
"num_tokens": 3056521.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.032876205444336,
|
|
"epoch": 1.3675548589341693,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004955629958544577,
|
|
"loss": 5.9428,
|
|
"mean_token_accuracy": 0.1336909145116806,
|
|
"num_tokens": 3066007.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.174657106399536,
|
|
"epoch": 1.3714733542319748,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004955033584138143,
|
|
"loss": 6.1364,
|
|
"mean_token_accuracy": 0.12483339086174965,
|
|
"num_tokens": 3075111.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.001473474502563,
|
|
"epoch": 1.3753918495297806,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004954433268979164,
|
|
"loss": 5.93,
|
|
"mean_token_accuracy": 0.15043148621916771,
|
|
"num_tokens": 3083766.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.056762075424194,
|
|
"epoch": 1.3793103448275863,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004953829014140502,
|
|
"loss": 6.0034,
|
|
"mean_token_accuracy": 0.13454223051667213,
|
|
"num_tokens": 3092103.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.0849854946136475,
|
|
"epoch": 1.3832288401253918,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004953220820702057,
|
|
"loss": 6.0065,
|
|
"mean_token_accuracy": 0.13198206946253777,
|
|
"num_tokens": 3101286.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.222050523757934,
|
|
"epoch": 1.3871473354231976,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004952608689750771,
|
|
"loss": 6.024,
|
|
"mean_token_accuracy": 0.13246920630335807,
|
|
"num_tokens": 3110554.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.981888675689698,
|
|
"epoch": 1.391065830721003,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004951992622380619,
|
|
"loss": 5.8669,
|
|
"mean_token_accuracy": 0.14516761153936386,
|
|
"num_tokens": 3119265.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.022592401504516,
|
|
"epoch": 1.3949843260188088,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004951372619692615,
|
|
"loss": 5.9176,
|
|
"mean_token_accuracy": 0.13864269405603408,
|
|
"num_tokens": 3127962.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.138263750076294,
|
|
"epoch": 1.3989028213166144,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004950748682794804,
|
|
"loss": 6.0343,
|
|
"mean_token_accuracy": 0.13279442861676216,
|
|
"num_tokens": 3137265.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.982170152664184,
|
|
"epoch": 1.40282131661442,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004950120812802262,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.13771192952990532,
|
|
"num_tokens": 3146458.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.121512508392334,
|
|
"epoch": 1.4067398119122256,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004949489010837095,
|
|
"loss": 6.0051,
|
|
"mean_token_accuracy": 0.13202869817614554,
|
|
"num_tokens": 3155557.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.031117582321167,
|
|
"epoch": 1.4106583072100314,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 5.8921,
|
|
"mean_token_accuracy": 0.1407647594809532,
|
|
"num_tokens": 3163478.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.06480302810669,
|
|
"epoch": 1.414576802507837,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000494821361551244,
|
|
"loss": 5.9734,
|
|
"mean_token_accuracy": 0.13325524404644967,
|
|
"num_tokens": 3172632.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.034041261672973,
|
|
"epoch": 1.4184952978056427,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004947570024432291,
|
|
"loss": 6.0159,
|
|
"mean_token_accuracy": 0.14027554914355278,
|
|
"num_tokens": 3181325.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.120739078521728,
|
|
"epoch": 1.4224137931034484,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004946922505938189,
|
|
"loss": 6.058,
|
|
"mean_token_accuracy": 0.140015921741724,
|
|
"num_tokens": 3189783.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.078346109390258,
|
|
"epoch": 1.426332288401254,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004946271061187354,
|
|
"loss": 5.9115,
|
|
"mean_token_accuracy": 0.1306297406554222,
|
|
"num_tokens": 3198605.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.07373309135437,
|
|
"epoch": 1.4302507836990594,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004945615691344025,
|
|
"loss": 6.0025,
|
|
"mean_token_accuracy": 0.12800363823771477,
|
|
"num_tokens": 3208061.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.044371461868286,
|
|
"epoch": 1.4341692789968652,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004944956397579453,
|
|
"loss": 5.9286,
|
|
"mean_token_accuracy": 0.1459761567413807,
|
|
"num_tokens": 3216093.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.0173381805419925,
|
|
"epoch": 1.438087774294671,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004944293181071902,
|
|
"loss": 5.9027,
|
|
"mean_token_accuracy": 0.14357266277074815,
|
|
"num_tokens": 3224446.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.0938514232635494,
|
|
"epoch": 1.4420062695924765,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004943626043006649,
|
|
"loss": 6.0177,
|
|
"mean_token_accuracy": 0.135706390440464,
|
|
"num_tokens": 3233398.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.068123865127563,
|
|
"epoch": 1.4459247648902822,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000494295498457598,
|
|
"loss": 6.0156,
|
|
"mean_token_accuracy": 0.13109127432107925,
|
|
"num_tokens": 3243391.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.0765832424163815,
|
|
"epoch": 1.4498432601880877,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004942280006979179,
|
|
"loss": 5.96,
|
|
"mean_token_accuracy": 0.13749199137091636,
|
|
"num_tokens": 3251954.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.0438025951385494,
|
|
"epoch": 1.4537617554858935,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004941601111422546,
|
|
"loss": 5.9851,
|
|
"mean_token_accuracy": 0.1373910054564476,
|
|
"num_tokens": 3261121.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.976307153701782,
|
|
"epoch": 1.457680250783699,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004940918299119375,
|
|
"loss": 5.9028,
|
|
"mean_token_accuracy": 0.1375894144177437,
|
|
"num_tokens": 3270179.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.898191833496094,
|
|
"epoch": 1.4615987460815048,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004940231571289962,
|
|
"loss": 5.8893,
|
|
"mean_token_accuracy": 0.14039459228515624,
|
|
"num_tokens": 3279749.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.1549577713012695,
|
|
"epoch": 1.4655172413793103,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004939540929161603,
|
|
"loss": 5.9246,
|
|
"mean_token_accuracy": 0.13927078545093535,
|
|
"num_tokens": 3288458.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.00097017288208,
|
|
"epoch": 1.469435736677116,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004938846373968586,
|
|
"loss": 5.8991,
|
|
"mean_token_accuracy": 0.14272075444459914,
|
|
"num_tokens": 3297231.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.032529544830322,
|
|
"epoch": 1.4733542319749215,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004938147906952194,
|
|
"loss": 6.0329,
|
|
"mean_token_accuracy": 0.13757839426398277,
|
|
"num_tokens": 3306915.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.079828405380249,
|
|
"epoch": 1.4772727272727273,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00049374455293607,
|
|
"loss": 5.9958,
|
|
"mean_token_accuracy": 0.13423861265182496,
|
|
"num_tokens": 3315985.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.122959566116333,
|
|
"epoch": 1.481191222570533,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004936739242449369,
|
|
"loss": 5.9895,
|
|
"mean_token_accuracy": 0.13724515214562416,
|
|
"num_tokens": 3324924.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.991518545150757,
|
|
"epoch": 1.4851097178683386,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004936029047480447,
|
|
"loss": 5.9801,
|
|
"mean_token_accuracy": 0.13726715967059136,
|
|
"num_tokens": 3333464.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.0822971820831295,
|
|
"epoch": 1.489028213166144,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004935314945723171,
|
|
"loss": 5.8806,
|
|
"mean_token_accuracy": 0.14093400090932845,
|
|
"num_tokens": 3342441.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.940945720672607,
|
|
"epoch": 1.4929467084639498,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004934596938453754,
|
|
"loss": 5.9352,
|
|
"mean_token_accuracy": 0.1342185415327549,
|
|
"num_tokens": 3351338.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.092441558837891,
|
|
"epoch": 1.4968652037617556,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004933875026955391,
|
|
"loss": 5.9069,
|
|
"mean_token_accuracy": 0.1339087277650833,
|
|
"num_tokens": 3360356.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.088913869857788,
|
|
"epoch": 1.500783699059561,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004933149212518258,
|
|
"loss": 5.988,
|
|
"mean_token_accuracy": 0.13323872461915015,
|
|
"num_tokens": 3369485.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.02648138999939,
|
|
"epoch": 1.5047021943573666,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004932419496439501,
|
|
"loss": 5.9519,
|
|
"mean_token_accuracy": 0.14375862777233123,
|
|
"num_tokens": 3378355.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.99138445854187,
|
|
"epoch": 1.5086206896551724,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000493168588002324,
|
|
"loss": 5.8649,
|
|
"mean_token_accuracy": 0.14118586033582686,
|
|
"num_tokens": 3386729.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.119981384277343,
|
|
"epoch": 1.5125391849529781,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004930948364580569,
|
|
"loss": 6.088,
|
|
"mean_token_accuracy": 0.1284665696322918,
|
|
"num_tokens": 3395822.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.04303822517395,
|
|
"epoch": 1.5164576802507836,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004930206951429546,
|
|
"loss": 5.99,
|
|
"mean_token_accuracy": 0.14219039529561997,
|
|
"num_tokens": 3404671.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.036664438247681,
|
|
"epoch": 1.5203761755485894,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004929461641895197,
|
|
"loss": 5.9301,
|
|
"mean_token_accuracy": 0.13862462490797042,
|
|
"num_tokens": 3413289.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.010977125167846,
|
|
"epoch": 1.5242946708463951,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000492871243730951,
|
|
"loss": 5.9675,
|
|
"mean_token_accuracy": 0.1358363598585129,
|
|
"num_tokens": 3421960.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.0582923412323,
|
|
"epoch": 1.5282131661442007,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004927959339011437,
|
|
"loss": 5.9564,
|
|
"mean_token_accuracy": 0.13902525305747987,
|
|
"num_tokens": 3431958.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.846984767913819,
|
|
"epoch": 1.5321316614420062,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004927202348346885,
|
|
"loss": 5.7801,
|
|
"mean_token_accuracy": 0.14725423008203506,
|
|
"num_tokens": 3441438.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.118220949172974,
|
|
"epoch": 1.536050156739812,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492644146666872,
|
|
"loss": 5.9105,
|
|
"mean_token_accuracy": 0.14317999482154847,
|
|
"num_tokens": 3450176.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.923649597167969,
|
|
"epoch": 1.5399686520376177,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004925676695336761,
|
|
"loss": 5.8946,
|
|
"mean_token_accuracy": 0.14172052592039108,
|
|
"num_tokens": 3458685.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.046487474441529,
|
|
"epoch": 1.5438871473354232,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004924908035717777,
|
|
"loss": 6.0445,
|
|
"mean_token_accuracy": 0.13102332279086112,
|
|
"num_tokens": 3468362.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.12939486503601,
|
|
"epoch": 1.5478056426332287,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492413548918549,
|
|
"loss": 5.9607,
|
|
"mean_token_accuracy": 0.13970244973897933,
|
|
"num_tokens": 3476791.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.925748825073242,
|
|
"epoch": 1.5517241379310345,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004923359057120563,
|
|
"loss": 5.8781,
|
|
"mean_token_accuracy": 0.13300700336694718,
|
|
"num_tokens": 3486486.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.951593446731567,
|
|
"epoch": 1.5556426332288402,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004922578740910608,
|
|
"loss": 5.8334,
|
|
"mean_token_accuracy": 0.14717195332050323,
|
|
"num_tokens": 3496145.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.967483854293823,
|
|
"epoch": 1.5595611285266457,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004921794541950177,
|
|
"loss": 5.8214,
|
|
"mean_token_accuracy": 0.14525392055511474,
|
|
"num_tokens": 3504644.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.994989633560181,
|
|
"epoch": 1.5634796238244513,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004921006461640758,
|
|
"loss": 5.8562,
|
|
"mean_token_accuracy": 0.13699238896369934,
|
|
"num_tokens": 3513126.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.833573675155639,
|
|
"epoch": 1.567398119122257,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000492021450139078,
|
|
"loss": 5.7981,
|
|
"mean_token_accuracy": 0.14626943692564964,
|
|
"num_tokens": 3521505.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.567398119122257,
|
|
"eval_entropy": 5.888265516406806,
|
|
"eval_loss": 6.116030216217041,
|
|
"eval_mean_token_accuracy": 0.13753249434855327,
|
|
"eval_num_tokens": 3521505.0,
|
|
"eval_runtime": 2.8299,
|
|
"eval_samples_per_second": 1456.593,
|
|
"eval_steps_per_second": 182.339,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.132147789001465,
|
|
"epoch": 1.5713166144200628,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004919418662615605,
|
|
"loss": 6.0583,
|
|
"mean_token_accuracy": 0.1335877738893032,
|
|
"num_tokens": 3530008.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.083332824707031,
|
|
"epoch": 1.5752351097178683,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004918618946737525,
|
|
"loss": 5.9278,
|
|
"mean_token_accuracy": 0.13504896759986879,
|
|
"num_tokens": 3538778.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.979771709442138,
|
|
"epoch": 1.579153605015674,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004917815355185762,
|
|
"loss": 5.9164,
|
|
"mean_token_accuracy": 0.13633493483066558,
|
|
"num_tokens": 3547108.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.988805103302002,
|
|
"epoch": 1.5830721003134798,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004917007889396464,
|
|
"loss": 5.9513,
|
|
"mean_token_accuracy": 0.13188610523939132,
|
|
"num_tokens": 3556094.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.985982990264892,
|
|
"epoch": 1.5869905956112853,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004916196550812706,
|
|
"loss": 5.8151,
|
|
"mean_token_accuracy": 0.14568774104118348,
|
|
"num_tokens": 3564222.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.818160581588745,
|
|
"epoch": 1.5909090909090908,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004915381340884477,
|
|
"loss": 5.827,
|
|
"mean_token_accuracy": 0.14547126069664956,
|
|
"num_tokens": 3573123.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.050127172470093,
|
|
"epoch": 1.5948275862068966,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 5.8571,
|
|
"mean_token_accuracy": 0.14239953309297562,
|
|
"num_tokens": 3581462.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.88056173324585,
|
|
"epoch": 1.5987460815047023,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004913739312829181,
|
|
"loss": 5.8744,
|
|
"mean_token_accuracy": 0.14017535969614983,
|
|
"num_tokens": 3590718.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.000182294845581,
|
|
"epoch": 1.6026645768025078,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004912912497636683,
|
|
"loss": 5.8899,
|
|
"mean_token_accuracy": 0.13969212546944618,
|
|
"num_tokens": 3599890.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.039797258377075,
|
|
"epoch": 1.6065830721003134,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004912081816968853,
|
|
"loss": 5.9815,
|
|
"mean_token_accuracy": 0.14007024392485617,
|
|
"num_tokens": 3607927.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.0202032089233395,
|
|
"epoch": 1.6105015673981191,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000491124727231025,
|
|
"loss": 5.838,
|
|
"mean_token_accuracy": 0.14422516226768495,
|
|
"num_tokens": 3615935.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.8747539043426515,
|
|
"epoch": 1.6144200626959249,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004910408865152343,
|
|
"loss": 5.8654,
|
|
"mean_token_accuracy": 0.1397896021604538,
|
|
"num_tokens": 3624475.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.097391796112061,
|
|
"epoch": 1.6183385579937304,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004909566596993498,
|
|
"loss": 6.072,
|
|
"mean_token_accuracy": 0.131089448928833,
|
|
"num_tokens": 3633995.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.067581701278686,
|
|
"epoch": 1.622257053291536,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004908720469338988,
|
|
"loss": 5.9294,
|
|
"mean_token_accuracy": 0.14296501129865646,
|
|
"num_tokens": 3643121.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.959115791320801,
|
|
"epoch": 1.6261755485893417,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004907870483700979,
|
|
"loss": 5.8621,
|
|
"mean_token_accuracy": 0.14175319969654082,
|
|
"num_tokens": 3652285.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.929019594192505,
|
|
"epoch": 1.6300940438871474,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004907016641598534,
|
|
"loss": 5.926,
|
|
"mean_token_accuracy": 0.13432129770517348,
|
|
"num_tokens": 3661441.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.0357630252838135,
|
|
"epoch": 1.634012539184953,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004906158944557607,
|
|
"loss": 5.8105,
|
|
"mean_token_accuracy": 0.14453670606017113,
|
|
"num_tokens": 3669364.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.9998321533203125,
|
|
"epoch": 1.6379310344827587,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000490529739411104,
|
|
"loss": 5.9024,
|
|
"mean_token_accuracy": 0.13794894218444825,
|
|
"num_tokens": 3677848.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.93680510520935,
|
|
"epoch": 1.6418495297805644,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004904431991798565,
|
|
"loss": 5.8138,
|
|
"mean_token_accuracy": 0.14087132290005683,
|
|
"num_tokens": 3686617.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.955365371704102,
|
|
"epoch": 1.64576802507837,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004903562739166797,
|
|
"loss": 5.8044,
|
|
"mean_token_accuracy": 0.1422215446829796,
|
|
"num_tokens": 3695106.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.7903650283813475,
|
|
"epoch": 1.6496865203761755,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004902689637769229,
|
|
"loss": 5.7518,
|
|
"mean_token_accuracy": 0.15124305188655854,
|
|
"num_tokens": 3703167.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.9747395515441895,
|
|
"epoch": 1.6536050156739812,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004901812689166237,
|
|
"loss": 5.8666,
|
|
"mean_token_accuracy": 0.13829359784722328,
|
|
"num_tokens": 3711597.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.945003080368042,
|
|
"epoch": 1.657523510971787,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004900931894925069,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.14057869464159012,
|
|
"num_tokens": 3720277.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.949292469024658,
|
|
"epoch": 1.6614420062695925,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004900047256619849,
|
|
"loss": 5.921,
|
|
"mean_token_accuracy": 0.13772802650928498,
|
|
"num_tokens": 3729832.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.9761933326721195,
|
|
"epoch": 1.665360501567398,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004899158775831566,
|
|
"loss": 5.8632,
|
|
"mean_token_accuracy": 0.14447186067700385,
|
|
"num_tokens": 3738390.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.933046817779541,
|
|
"epoch": 1.6692789968652038,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004898266454148081,
|
|
"loss": 5.9185,
|
|
"mean_token_accuracy": 0.14356547445058823,
|
|
"num_tokens": 3747046.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.945548963546753,
|
|
"epoch": 1.6731974921630095,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004897370293164119,
|
|
"loss": 5.8896,
|
|
"mean_token_accuracy": 0.13832552805542947,
|
|
"num_tokens": 3755818.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.093547344207764,
|
|
"epoch": 1.677115987460815,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004896470294481262,
|
|
"loss": 5.9861,
|
|
"mean_token_accuracy": 0.13742954656481743,
|
|
"num_tokens": 3764123.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.950912809371948,
|
|
"epoch": 1.6810344827586206,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895566459707954,
|
|
"loss": 6.035,
|
|
"mean_token_accuracy": 0.12903214767575263,
|
|
"num_tokens": 3773105.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.076629734039306,
|
|
"epoch": 1.6849529780564263,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004894658790459498,
|
|
"loss": 5.8805,
|
|
"mean_token_accuracy": 0.14274725392460824,
|
|
"num_tokens": 3781417.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.058791875839233,
|
|
"epoch": 1.688871473354232,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004893747288358041,
|
|
"loss": 5.9792,
|
|
"mean_token_accuracy": 0.13869686052203178,
|
|
"num_tokens": 3789657.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.9904515743255615,
|
|
"epoch": 1.6927899686520376,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000489283195503259,
|
|
"loss": 5.9875,
|
|
"mean_token_accuracy": 0.13768337592482566,
|
|
"num_tokens": 3798700.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.9594886302948,
|
|
"epoch": 1.6967084639498433,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000489191279211899,
|
|
"loss": 5.8479,
|
|
"mean_token_accuracy": 0.14359756112098693,
|
|
"num_tokens": 3807299.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.006121349334717,
|
|
"epoch": 1.700626959247649,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004890989801259935,
|
|
"loss": 5.9385,
|
|
"mean_token_accuracy": 0.14270951524376868,
|
|
"num_tokens": 3816292.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.049522542953492,
|
|
"epoch": 1.7045454545454546,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000489006298410496,
|
|
"loss": 5.8532,
|
|
"mean_token_accuracy": 0.1399511620402336,
|
|
"num_tokens": 3825070.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.959496307373047,
|
|
"epoch": 1.70846394984326,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004889132342310438,
|
|
"loss": 5.8247,
|
|
"mean_token_accuracy": 0.14473102912306784,
|
|
"num_tokens": 3834016.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.956418132781982,
|
|
"epoch": 1.7123824451410659,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004888197877539577,
|
|
"loss": 5.8849,
|
|
"mean_token_accuracy": 0.14025031253695489,
|
|
"num_tokens": 3842172.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.971714496612549,
|
|
"epoch": 1.7163009404388716,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004887259591462417,
|
|
"loss": 5.798,
|
|
"mean_token_accuracy": 0.14462782070040703,
|
|
"num_tokens": 3851109.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.923008251190185,
|
|
"epoch": 1.7202194357366771,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004886317485755825,
|
|
"loss": 5.924,
|
|
"mean_token_accuracy": 0.14367973506450654,
|
|
"num_tokens": 3859656.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.933929538726806,
|
|
"epoch": 1.7241379310344827,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004885371562103498,
|
|
"loss": 5.7677,
|
|
"mean_token_accuracy": 0.14872414171695708,
|
|
"num_tokens": 3868698.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.8945310592651365,
|
|
"epoch": 1.7280564263322884,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004884421822195957,
|
|
"loss": 5.8545,
|
|
"mean_token_accuracy": 0.1411299616098404,
|
|
"num_tokens": 3877474.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.9916675090789795,
|
|
"epoch": 1.7319749216300941,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004883468267730538,
|
|
"loss": 5.8228,
|
|
"mean_token_accuracy": 0.1443895533680916,
|
|
"num_tokens": 3886328.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.906041145324707,
|
|
"epoch": 1.7358934169278997,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048825109004114006,
|
|
"loss": 5.8058,
|
|
"mean_token_accuracy": 0.1487313315272331,
|
|
"num_tokens": 3894424.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.978779983520508,
|
|
"epoch": 1.7398119122257052,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004881549721949513,
|
|
"loss": 5.8897,
|
|
"mean_token_accuracy": 0.13909043669700621,
|
|
"num_tokens": 3903746.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.851698732376098,
|
|
"epoch": 1.743730407523511,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004880584734062658,
|
|
"loss": 5.8724,
|
|
"mean_token_accuracy": 0.14629912972450257,
|
|
"num_tokens": 3912033.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.883364534378051,
|
|
"epoch": 1.7476489028213167,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004879615938475425,
|
|
"loss": 5.7739,
|
|
"mean_token_accuracy": 0.14114395081996917,
|
|
"num_tokens": 3920388.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.934035730361939,
|
|
"epoch": 1.7515673981191222,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004878643336919209,
|
|
"loss": 5.9772,
|
|
"mean_token_accuracy": 0.139875166118145,
|
|
"num_tokens": 3929926.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.131916475296021,
|
|
"epoch": 1.7554858934169277,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004877666931132206,
|
|
"loss": 5.9601,
|
|
"mean_token_accuracy": 0.13168897181749345,
|
|
"num_tokens": 3938696.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.873245096206665,
|
|
"epoch": 1.7594043887147337,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004876686722859413,
|
|
"loss": 5.7899,
|
|
"mean_token_accuracy": 0.1513090804219246,
|
|
"num_tokens": 3947537.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.992959451675415,
|
|
"epoch": 1.7633228840125392,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000487570271385262,
|
|
"loss": 5.7993,
|
|
"mean_token_accuracy": 0.14130587950348855,
|
|
"num_tokens": 3956539.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.880547904968262,
|
|
"epoch": 1.7672413793103448,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004874714905870411,
|
|
"loss": 5.7667,
|
|
"mean_token_accuracy": 0.14465454295277597,
|
|
"num_tokens": 3964927.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.923993968963623,
|
|
"epoch": 1.7711598746081505,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004873723300678159,
|
|
"loss": 5.9062,
|
|
"mean_token_accuracy": 0.1402622014284134,
|
|
"num_tokens": 3973565.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.974027013778686,
|
|
"epoch": 1.7750783699059562,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048727279000480226,
|
|
"loss": 5.9283,
|
|
"mean_token_accuracy": 0.14056170955300332,
|
|
"num_tokens": 3982683.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.065245008468628,
|
|
"epoch": 1.7789968652037618,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 5.8374,
|
|
"mean_token_accuracy": 0.14206577837467194,
|
|
"num_tokens": 3991462.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.725625848770141,
|
|
"epoch": 1.7829153605015673,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004870725719596648,
|
|
"loss": 5.6081,
|
|
"mean_token_accuracy": 0.15651399940252303,
|
|
"num_tokens": 3999821.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.943557548522949,
|
|
"epoch": 1.786833855799373,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004869718943353631,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.14041255488991738,
|
|
"num_tokens": 4008655.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.972759962081909,
|
|
"epoch": 1.7907523510971788,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048687083788291656,
|
|
"loss": 5.8939,
|
|
"mean_token_accuracy": 0.14818500876426696,
|
|
"num_tokens": 4017317.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.872368335723877,
|
|
"epoch": 1.7946708463949843,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048676940278292953,
|
|
"loss": 5.7282,
|
|
"mean_token_accuracy": 0.14405218958854676,
|
|
"num_tokens": 4025407.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.908810663223266,
|
|
"epoch": 1.7985893416927898,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048666758921668286,
|
|
"loss": 5.7369,
|
|
"mean_token_accuracy": 0.14384883195161818,
|
|
"num_tokens": 4034310.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.879950332641601,
|
|
"epoch": 1.8025078369905956,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048656539736613403,
|
|
"loss": 5.8969,
|
|
"mean_token_accuracy": 0.1422630712389946,
|
|
"num_tokens": 4042326.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.854609394073487,
|
|
"epoch": 1.8064263322884013,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004864628274139164,
|
|
"loss": 5.7595,
|
|
"mean_token_accuracy": 0.1504887729883194,
|
|
"num_tokens": 4051248.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.929189538955688,
|
|
"epoch": 1.8103448275862069,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004863598795433391,
|
|
"loss": 5.8763,
|
|
"mean_token_accuracy": 0.1436061829328537,
|
|
"num_tokens": 4060227.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.035856866836548,
|
|
"epoch": 1.8142633228840124,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048625655393838666,
|
|
"loss": 5.9747,
|
|
"mean_token_accuracy": 0.1343390792608261,
|
|
"num_tokens": 4069700.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.9489781856536865,
|
|
"epoch": 1.8181818181818183,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004861528507837186,
|
|
"loss": 5.8364,
|
|
"mean_token_accuracy": 0.14606306403875352,
|
|
"num_tokens": 4079363.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.038274621963501,
|
|
"epoch": 1.8221003134796239,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004860487702646695,
|
|
"loss": 5.9136,
|
|
"mean_token_accuracy": 0.14080933034420012,
|
|
"num_tokens": 4087944.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.765271520614624,
|
|
"epoch": 1.8260188087774294,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004859443125672479,
|
|
"loss": 5.6074,
|
|
"mean_token_accuracy": 0.16549111306667327,
|
|
"num_tokens": 4096431.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.862937021255493,
|
|
"epoch": 1.8299373040752351,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004858394778781368,
|
|
"loss": 5.7839,
|
|
"mean_token_accuracy": 0.1452955462038517,
|
|
"num_tokens": 4104894.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.929635095596313,
|
|
"epoch": 1.8338557993730409,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004857342663846927,
|
|
"loss": 5.8308,
|
|
"mean_token_accuracy": 0.1402455188333988,
|
|
"num_tokens": 4112878.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.906289291381836,
|
|
"epoch": 1.8377742946708464,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004856286782749456,
|
|
"loss": 5.839,
|
|
"mean_token_accuracy": 0.14992391616106032,
|
|
"num_tokens": 4122501.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.893653440475464,
|
|
"epoch": 1.841692789968652,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004855227137375986,
|
|
"loss": 5.8018,
|
|
"mean_token_accuracy": 0.13767838329076768,
|
|
"num_tokens": 4131301.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.921551322937011,
|
|
"epoch": 1.8456112852664577,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004854163729620275,
|
|
"loss": 5.8349,
|
|
"mean_token_accuracy": 0.14852394610643388,
|
|
"num_tokens": 4140076.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.864505481719971,
|
|
"epoch": 1.8495297805642634,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004853096561382805,
|
|
"loss": 5.7663,
|
|
"mean_token_accuracy": 0.15288633555173875,
|
|
"num_tokens": 4148728.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.914716482162476,
|
|
"epoch": 1.853448275862069,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004852025634570779,
|
|
"loss": 5.9037,
|
|
"mean_token_accuracy": 0.14291643872857093,
|
|
"num_tokens": 4157520.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.905571317672729,
|
|
"epoch": 1.8573667711598745,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004850950951098116,
|
|
"loss": 5.7141,
|
|
"mean_token_accuracy": 0.15238296538591384,
|
|
"num_tokens": 4165547.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.928301191329956,
|
|
"epoch": 1.8612852664576802,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004849872512885451,
|
|
"loss": 5.8628,
|
|
"mean_token_accuracy": 0.1383912220597267,
|
|
"num_tokens": 4174656.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.904009199142456,
|
|
"epoch": 1.865203761755486,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004848790321860127,
|
|
"loss": 5.9161,
|
|
"mean_token_accuracy": 0.1413537159562111,
|
|
"num_tokens": 4183452.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.909180545806885,
|
|
"epoch": 1.8691222570532915,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048477043799561946,
|
|
"loss": 5.7966,
|
|
"mean_token_accuracy": 0.14063763692975045,
|
|
"num_tokens": 4192306.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.8980663299560545,
|
|
"epoch": 1.873040752351097,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004846614689114409,
|
|
"loss": 5.8643,
|
|
"mean_token_accuracy": 0.14282563477754592,
|
|
"num_tokens": 4201201.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.890587997436524,
|
|
"epoch": 1.876959247648903,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004845521251282223,
|
|
"loss": 5.7437,
|
|
"mean_token_accuracy": 0.15185199975967406,
|
|
"num_tokens": 4209374.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.809947061538696,
|
|
"epoch": 1.8808777429467085,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004844424068413789,
|
|
"loss": 5.7215,
|
|
"mean_token_accuracy": 0.1469734065234661,
|
|
"num_tokens": 4217258.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.848538589477539,
|
|
"epoch": 1.884796238244514,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048433231424699504,
|
|
"loss": 5.7551,
|
|
"mean_token_accuracy": 0.1480185203254223,
|
|
"num_tokens": 4226484.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.924857568740845,
|
|
"epoch": 1.8887147335423198,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048422184754182384,
|
|
"loss": 5.7889,
|
|
"mean_token_accuracy": 0.14311063811182975,
|
|
"num_tokens": 4235546.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.950683832168579,
|
|
"epoch": 1.8926332288401255,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004841110069232875,
|
|
"loss": 5.9386,
|
|
"mean_token_accuracy": 0.1451379805803299,
|
|
"num_tokens": 4244549.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.900457239151001,
|
|
"epoch": 1.896551724137931,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048399979258947597,
|
|
"loss": 5.8331,
|
|
"mean_token_accuracy": 0.1464390404522419,
|
|
"num_tokens": 4252918.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.9453812599182125,
|
|
"epoch": 1.9004702194357366,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004838882047391474,
|
|
"loss": 5.8327,
|
|
"mean_token_accuracy": 0.14703057184815407,
|
|
"num_tokens": 4261674.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.875091028213501,
|
|
"epoch": 1.9043887147335423,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048377624357172724,
|
|
"loss": 5.7425,
|
|
"mean_token_accuracy": 0.14618095010519028,
|
|
"num_tokens": 4269986.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.890668630599976,
|
|
"epoch": 1.908307210031348,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048366390928730843,
|
|
"loss": 5.8377,
|
|
"mean_token_accuracy": 0.14733434692025185,
|
|
"num_tokens": 4278131.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.966364622116089,
|
|
"epoch": 1.9122257053291536,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004835512020866504,
|
|
"loss": 5.8243,
|
|
"mean_token_accuracy": 0.14588867947459222,
|
|
"num_tokens": 4287214.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.893796443939209,
|
|
"epoch": 1.9161442006269591,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048343812217117925,
|
|
"loss": 5.7504,
|
|
"mean_token_accuracy": 0.1461929127573967,
|
|
"num_tokens": 4295949.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.786278104782104,
|
|
"epoch": 1.9200626959247649,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048332466974298723,
|
|
"loss": 5.7543,
|
|
"mean_token_accuracy": 0.1465558797121048,
|
|
"num_tokens": 4305014.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.851295328140258,
|
|
"epoch": 1.9239811912225706,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048321084500483203,
|
|
"loss": 5.7265,
|
|
"mean_token_accuracy": 0.14898128807544708,
|
|
"num_tokens": 4313560.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.87207818031311,
|
|
"epoch": 1.9278996865203761,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000483096648160137,
|
|
"loss": 5.8056,
|
|
"mean_token_accuracy": 0.1454322248697281,
|
|
"num_tokens": 4323280.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.872972869873047,
|
|
"epoch": 1.9318181818181817,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048298207941299047,
|
|
"loss": 5.861,
|
|
"mean_token_accuracy": 0.14596157446503638,
|
|
"num_tokens": 4331757.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.8036915302276615,
|
|
"epoch": 1.9357366771159876,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048286713896814536,
|
|
"loss": 5.7876,
|
|
"mean_token_accuracy": 0.14330902695655823,
|
|
"num_tokens": 4341453.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.028292417526245,
|
|
"epoch": 1.9396551724137931,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048275182703101877,
|
|
"loss": 5.7431,
|
|
"mean_token_accuracy": 0.14791915863752364,
|
|
"num_tokens": 4349095.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.750636386871338,
|
|
"epoch": 1.9435736677115987,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048263614380769193,
|
|
"loss": 5.8404,
|
|
"mean_token_accuracy": 0.14387739300727845,
|
|
"num_tokens": 4357913.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.927936792373657,
|
|
"epoch": 1.9474921630094044,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048252008950490957,
|
|
"loss": 5.8083,
|
|
"mean_token_accuracy": 0.14284604787826538,
|
|
"num_tokens": 4366586.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.8729980945587155,
|
|
"epoch": 1.9514106583072102,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048240366433007935,
|
|
"loss": 5.7307,
|
|
"mean_token_accuracy": 0.15130855292081832,
|
|
"num_tokens": 4375393.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.848481798171997,
|
|
"epoch": 1.9553291536050157,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048228686849127213,
|
|
"loss": 5.8159,
|
|
"mean_token_accuracy": 0.15051234513521194,
|
|
"num_tokens": 4383747.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.987670612335205,
|
|
"epoch": 1.9592476489028212,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004821697021972209,
|
|
"loss": 5.7804,
|
|
"mean_token_accuracy": 0.14112372547388077,
|
|
"num_tokens": 4392085.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 1.9592476489028212,
|
|
"eval_entropy": 5.819716327874235,
|
|
"eval_loss": 5.9731364250183105,
|
|
"eval_mean_token_accuracy": 0.14396101977791667,
|
|
"eval_num_tokens": 4392085.0,
|
|
"eval_runtime": 2.8283,
|
|
"eval_samples_per_second": 1457.403,
|
|
"eval_steps_per_second": 182.441,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.882195997238159,
|
|
"epoch": 1.963166144200627,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 5.8043,
|
|
"mean_token_accuracy": 0.14567812159657478,
|
|
"num_tokens": 4400843.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.773218107223511,
|
|
"epoch": 1.9670846394984327,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004819342590816288,
|
|
"loss": 5.7471,
|
|
"mean_token_accuracy": 0.14761917144060135,
|
|
"num_tokens": 4409126.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.89848108291626,
|
|
"epoch": 1.9710031347962382,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004818159826808631,
|
|
"loss": 5.9097,
|
|
"mean_token_accuracy": 0.1417311027646065,
|
|
"num_tokens": 4418401.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.897497129440308,
|
|
"epoch": 1.9749216300940438,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004816973366664026,
|
|
"loss": 5.811,
|
|
"mean_token_accuracy": 0.1456060990691185,
|
|
"num_tokens": 4428141.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.986608266830444,
|
|
"epoch": 1.9788401253918495,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004815783212502871,
|
|
"loss": 5.828,
|
|
"mean_token_accuracy": 0.1428292214870453,
|
|
"num_tokens": 4437158.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.895078134536743,
|
|
"epoch": 1.9827586206896552,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048145893664521645,
|
|
"loss": 5.8973,
|
|
"mean_token_accuracy": 0.13774282485246658,
|
|
"num_tokens": 4445901.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.878189134597778,
|
|
"epoch": 1.9866771159874608,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048133918306455023,
|
|
"loss": 5.7054,
|
|
"mean_token_accuracy": 0.15397633910179137,
|
|
"num_tokens": 4454200.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.8978513240814205,
|
|
"epoch": 1.9905956112852663,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004812190607223075,
|
|
"loss": 5.8244,
|
|
"mean_token_accuracy": 0.1423790991306305,
|
|
"num_tokens": 4463484.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.743094491958618,
|
|
"epoch": 1.9945141065830723,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048109856983316655,
|
|
"loss": 5.6436,
|
|
"mean_token_accuracy": 0.1538312703371048,
|
|
"num_tokens": 4472659.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.894764852523804,
|
|
"epoch": 1.9984326018808778,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000480977710612464,
|
|
"loss": 5.7289,
|
|
"mean_token_accuracy": 0.1462229423224926,
|
|
"num_tokens": 4481355.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.841469097137451,
|
|
"epoch": 2.0023510971786833,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004808564832761948,
|
|
"loss": 5.5879,
|
|
"mean_token_accuracy": 0.1482195809483528,
|
|
"num_tokens": 4490005.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.781640338897705,
|
|
"epoch": 2.006269592476489,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004807348880410119,
|
|
"loss": 5.3827,
|
|
"mean_token_accuracy": 0.1616608127951622,
|
|
"num_tokens": 4498645.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.733411645889282,
|
|
"epoch": 2.010188087774295,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004806129251242258,
|
|
"loss": 5.5173,
|
|
"mean_token_accuracy": 0.15624455660581588,
|
|
"num_tokens": 4509675.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.799925708770752,
|
|
"epoch": 2.0141065830721003,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048049059474380393,
|
|
"loss": 5.3307,
|
|
"mean_token_accuracy": 0.1625298425555229,
|
|
"num_tokens": 4518463.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.857757616043091,
|
|
"epoch": 2.018025078369906,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048036789711837047,
|
|
"loss": 5.4816,
|
|
"mean_token_accuracy": 0.15710628032684326,
|
|
"num_tokens": 4527120.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.79098744392395,
|
|
"epoch": 2.0219435736677114,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048024483246720607,
|
|
"loss": 5.5141,
|
|
"mean_token_accuracy": 0.1521046057343483,
|
|
"num_tokens": 4535962.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.795133399963379,
|
|
"epoch": 2.0258620689655173,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004801214010102472,
|
|
"loss": 5.4748,
|
|
"mean_token_accuracy": 0.15003894567489623,
|
|
"num_tokens": 4545233.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.720596790313721,
|
|
"epoch": 2.029780564263323,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004799976029680858,
|
|
"loss": 5.351,
|
|
"mean_token_accuracy": 0.1570914052426815,
|
|
"num_tokens": 4554658.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.793403244018554,
|
|
"epoch": 2.0336990595611284,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004798734385619691,
|
|
"loss": 5.5496,
|
|
"mean_token_accuracy": 0.1535790517926216,
|
|
"num_tokens": 4563964.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.807430982589722,
|
|
"epoch": 2.0376175548589344,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000479748908013799,
|
|
"loss": 5.4134,
|
|
"mean_token_accuracy": 0.16276097446680068,
|
|
"num_tokens": 4572865.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.672877264022827,
|
|
"epoch": 2.04153605015674,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004796240115461319,
|
|
"loss": 5.3736,
|
|
"mean_token_accuracy": 0.17069609314203263,
|
|
"num_tokens": 4581021.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.710117959976197,
|
|
"epoch": 2.0454545454545454,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004794987493821779,
|
|
"loss": 5.4687,
|
|
"mean_token_accuracy": 0.15729401111602784,
|
|
"num_tokens": 4590779.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.794814777374268,
|
|
"epoch": 2.049373040752351,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047937312174580084,
|
|
"loss": 5.4478,
|
|
"mean_token_accuracy": 0.1574981316924095,
|
|
"num_tokens": 4599831.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.857023143768311,
|
|
"epoch": 2.053291536050157,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004792471288615177,
|
|
"loss": 5.5945,
|
|
"mean_token_accuracy": 0.1500529244542122,
|
|
"num_tokens": 4609264.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.8015649795532225,
|
|
"epoch": 2.0572100313479624,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004791207709544981,
|
|
"loss": 5.5107,
|
|
"mean_token_accuracy": 0.15906523615121843,
|
|
"num_tokens": 4617885.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.6719653606414795,
|
|
"epoch": 2.061128526645768,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047899404825056424,
|
|
"loss": 5.4279,
|
|
"mean_token_accuracy": 0.15619982779026031,
|
|
"num_tokens": 4626536.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.770504665374756,
|
|
"epoch": 2.0650470219435735,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004788669609761901,
|
|
"loss": 5.5118,
|
|
"mean_token_accuracy": 0.1544651284813881,
|
|
"num_tokens": 4635563.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.720648860931396,
|
|
"epoch": 2.0689655172413794,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047873950935850107,
|
|
"loss": 5.4414,
|
|
"mean_token_accuracy": 0.1552010580897331,
|
|
"num_tokens": 4644822.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.75403094291687,
|
|
"epoch": 2.072884012539185,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004786116936252742,
|
|
"loss": 5.4081,
|
|
"mean_token_accuracy": 0.1630728781223297,
|
|
"num_tokens": 4652709.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.651195621490478,
|
|
"epoch": 2.0768025078369905,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004784835140049367,
|
|
"loss": 5.447,
|
|
"mean_token_accuracy": 0.15986524671316146,
|
|
"num_tokens": 4662014.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.7658521175384525,
|
|
"epoch": 2.080721003134796,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004783549707265663,
|
|
"loss": 5.4033,
|
|
"mean_token_accuracy": 0.15730245560407638,
|
|
"num_tokens": 4670618.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.7565501689910885,
|
|
"epoch": 2.084639498432602,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047822606401989084,
|
|
"loss": 5.5384,
|
|
"mean_token_accuracy": 0.14618516638875007,
|
|
"num_tokens": 4679937.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.747328805923462,
|
|
"epoch": 2.0885579937304075,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004780967941152873,
|
|
"loss": 5.3716,
|
|
"mean_token_accuracy": 0.16848112791776657,
|
|
"num_tokens": 4688006.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.598170948028565,
|
|
"epoch": 2.092476489028213,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004779671612437822,
|
|
"loss": 5.3212,
|
|
"mean_token_accuracy": 0.16888306885957718,
|
|
"num_tokens": 4697106.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.709634304046631,
|
|
"epoch": 2.096394984326019,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047783716563705035,
|
|
"loss": 5.4699,
|
|
"mean_token_accuracy": 0.15956653356552125,
|
|
"num_tokens": 4705582.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.789704513549805,
|
|
"epoch": 2.1003134796238245,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000477706807527415,
|
|
"loss": 5.4032,
|
|
"mean_token_accuracy": 0.1633769229054451,
|
|
"num_tokens": 4713640.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.7358156681060795,
|
|
"epoch": 2.10423197492163,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004775760871478472,
|
|
"loss": 5.5193,
|
|
"mean_token_accuracy": 0.1518564686179161,
|
|
"num_tokens": 4722729.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.710189342498779,
|
|
"epoch": 2.1081504702194356,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047744500473196564,
|
|
"loss": 5.4226,
|
|
"mean_token_accuracy": 0.16306858509778976,
|
|
"num_tokens": 4732099.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.778943872451782,
|
|
"epoch": 2.1120689655172415,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047731356051403556,
|
|
"loss": 5.4259,
|
|
"mean_token_accuracy": 0.15491524189710618,
|
|
"num_tokens": 4741766.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.550606107711792,
|
|
"epoch": 2.115987460815047,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004771817547289693,
|
|
"loss": 5.3501,
|
|
"mean_token_accuracy": 0.16072812229394912,
|
|
"num_tokens": 4750890.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.691572904586792,
|
|
"epoch": 2.1199059561128526,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004770495876123251,
|
|
"loss": 5.4713,
|
|
"mean_token_accuracy": 0.15760626047849655,
|
|
"num_tokens": 4760063.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.7919927597045895,
|
|
"epoch": 2.123824451410658,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004769170594003071,
|
|
"loss": 5.4737,
|
|
"mean_token_accuracy": 0.16406295895576478,
|
|
"num_tokens": 4768928.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.65070915222168,
|
|
"epoch": 2.127742946708464,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047678417032976457,
|
|
"loss": 5.4532,
|
|
"mean_token_accuracy": 0.15579911768436433,
|
|
"num_tokens": 4778129.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.69934196472168,
|
|
"epoch": 2.1316614420062696,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004766509206381919,
|
|
"loss": 5.4479,
|
|
"mean_token_accuracy": 0.1599138170480728,
|
|
"num_tokens": 4786759.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.6193219184875485,
|
|
"epoch": 2.135579937304075,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004765173105637279,
|
|
"loss": 5.4547,
|
|
"mean_token_accuracy": 0.16395413279533386,
|
|
"num_tokens": 4795355.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.696455669403076,
|
|
"epoch": 2.139498432601881,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047638334034515547,
|
|
"loss": 5.3964,
|
|
"mean_token_accuracy": 0.15683644711971284,
|
|
"num_tokens": 4804030.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.580397653579712,
|
|
"epoch": 2.1434169278996866,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047624901022190106,
|
|
"loss": 5.3523,
|
|
"mean_token_accuracy": 0.17931961715221406,
|
|
"num_tokens": 4812779.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.7006062984466555,
|
|
"epoch": 2.147335423197492,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.4629,
|
|
"mean_token_accuracy": 0.16019529104232788,
|
|
"num_tokens": 4822292.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.834177780151367,
|
|
"epoch": 2.1512539184952977,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004759792712222679,
|
|
"loss": 5.535,
|
|
"mean_token_accuracy": 0.15739217698574065,
|
|
"num_tokens": 4830524.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.8059389114379885,
|
|
"epoch": 2.1551724137931036,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004758438628279565,
|
|
"loss": 5.5164,
|
|
"mean_token_accuracy": 0.15928612202405928,
|
|
"num_tokens": 4839328.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.638897466659546,
|
|
"epoch": 2.159090909090909,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047570809549309697,
|
|
"loss": 5.3591,
|
|
"mean_token_accuracy": 0.1710207626223564,
|
|
"num_tokens": 4847711.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.774710035324096,
|
|
"epoch": 2.1630094043887147,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004755719694603275,
|
|
"loss": 5.549,
|
|
"mean_token_accuracy": 0.15182463973760604,
|
|
"num_tokens": 4856706.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.728883123397827,
|
|
"epoch": 2.16692789968652,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004754354849729274,
|
|
"loss": 5.5423,
|
|
"mean_token_accuracy": 0.15759230852127076,
|
|
"num_tokens": 4865250.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.702935409545899,
|
|
"epoch": 2.170846394984326,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047529864227481653,
|
|
"loss": 5.4631,
|
|
"mean_token_accuracy": 0.15977989733219147,
|
|
"num_tokens": 4873826.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.781952810287476,
|
|
"epoch": 2.1747648902821317,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000475161441610555,
|
|
"loss": 5.44,
|
|
"mean_token_accuracy": 0.16142217814922333,
|
|
"num_tokens": 4882631.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.665818929672241,
|
|
"epoch": 2.1786833855799372,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004750238832253427,
|
|
"loss": 5.4775,
|
|
"mean_token_accuracy": 0.15787807181477548,
|
|
"num_tokens": 4891007.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.697315788269043,
|
|
"epoch": 2.1826018808777428,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004748859673650187,
|
|
"loss": 5.386,
|
|
"mean_token_accuracy": 0.16155248284339904,
|
|
"num_tokens": 4899692.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.658601570129394,
|
|
"epoch": 2.1865203761755487,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047474769427606115,
|
|
"loss": 5.5095,
|
|
"mean_token_accuracy": 0.15856993645429612,
|
|
"num_tokens": 4909354.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.6818701267242435,
|
|
"epoch": 2.1904388714733543,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004746090642055863,
|
|
"loss": 5.4671,
|
|
"mean_token_accuracy": 0.1562755212187767,
|
|
"num_tokens": 4918400.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.739003992080688,
|
|
"epoch": 2.19435736677116,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047447007740134857,
|
|
"loss": 5.5467,
|
|
"mean_token_accuracy": 0.14892319440841675,
|
|
"num_tokens": 4926852.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.7248610496521,
|
|
"epoch": 2.1982758620689653,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047433073411174,
|
|
"loss": 5.4509,
|
|
"mean_token_accuracy": 0.15806010216474534,
|
|
"num_tokens": 4935717.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.591716861724853,
|
|
"epoch": 2.2021943573667713,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004741910345857896,
|
|
"loss": 5.3813,
|
|
"mean_token_accuracy": 0.16058638840913772,
|
|
"num_tokens": 4945389.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.752324533462525,
|
|
"epoch": 2.206112852664577,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047405097907316315,
|
|
"loss": 5.5201,
|
|
"mean_token_accuracy": 0.1630863979458809,
|
|
"num_tokens": 4953670.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.657805633544922,
|
|
"epoch": 2.2100313479623823,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004739105678241625,
|
|
"loss": 5.5289,
|
|
"mean_token_accuracy": 0.15918288826942445,
|
|
"num_tokens": 4962383.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.662711334228516,
|
|
"epoch": 2.2139498432601883,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004737698010897253,
|
|
"loss": 5.3295,
|
|
"mean_token_accuracy": 0.16085091829299927,
|
|
"num_tokens": 4970672.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.757535028457641,
|
|
"epoch": 2.217868338557994,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004736286791214245,
|
|
"loss": 5.5162,
|
|
"mean_token_accuracy": 0.15798387676477432,
|
|
"num_tokens": 4979303.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.72713942527771,
|
|
"epoch": 2.2217868338557993,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047348720217146807,
|
|
"loss": 5.3865,
|
|
"mean_token_accuracy": 0.1653503268957138,
|
|
"num_tokens": 4988322.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.582199764251709,
|
|
"epoch": 2.225705329153605,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047334537049269806,
|
|
"loss": 5.4754,
|
|
"mean_token_accuracy": 0.15811690539121628,
|
|
"num_tokens": 4997250.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.754626846313476,
|
|
"epoch": 2.229623824451411,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004732031843385909,
|
|
"loss": 5.5191,
|
|
"mean_token_accuracy": 0.151338791847229,
|
|
"num_tokens": 5005459.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.570264291763306,
|
|
"epoch": 2.2335423197492164,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004730606439632562,
|
|
"loss": 5.2819,
|
|
"mean_token_accuracy": 0.17884395718574525,
|
|
"num_tokens": 5013704.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.602528190612793,
|
|
"epoch": 2.237460815047022,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004729177496214367,
|
|
"loss": 5.3976,
|
|
"mean_token_accuracy": 0.16089559048414231,
|
|
"num_tokens": 5022545.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.619579839706421,
|
|
"epoch": 2.2413793103448274,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047277450156850767,
|
|
"loss": 5.4026,
|
|
"mean_token_accuracy": 0.16599346250295638,
|
|
"num_tokens": 5031133.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.6934120655059814,
|
|
"epoch": 2.2452978056426334,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004726309000604768,
|
|
"loss": 5.4422,
|
|
"mean_token_accuracy": 0.15853513032197952,
|
|
"num_tokens": 5039424.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.7841479778289795,
|
|
"epoch": 2.249216300940439,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004724869453539832,
|
|
"loss": 5.6437,
|
|
"mean_token_accuracy": 0.15349650308489798,
|
|
"num_tokens": 5048868.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.743923711776733,
|
|
"epoch": 2.2531347962382444,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004723426377062972,
|
|
"loss": 5.4742,
|
|
"mean_token_accuracy": 0.15738968551158905,
|
|
"num_tokens": 5057532.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.752024412155151,
|
|
"epoch": 2.2570532915360504,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047219797737532,
|
|
"loss": 5.4875,
|
|
"mean_token_accuracy": 0.15550234615802766,
|
|
"num_tokens": 5066088.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.5935587882995605,
|
|
"epoch": 2.260971786833856,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047205296461958314,
|
|
"loss": 5.3623,
|
|
"mean_token_accuracy": 0.1605392187833786,
|
|
"num_tokens": 5074968.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.607887029647827,
|
|
"epoch": 2.2648902821316614,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047190759969824785,
|
|
"loss": 5.4266,
|
|
"mean_token_accuracy": 0.1619516670703888,
|
|
"num_tokens": 5083605.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.665104103088379,
|
|
"epoch": 2.268808777429467,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047176188287110485,
|
|
"loss": 5.4267,
|
|
"mean_token_accuracy": 0.16131291091442107,
|
|
"num_tokens": 5092182.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.784565830230713,
|
|
"epoch": 2.2727272727272725,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004716158143985737,
|
|
"loss": 5.5997,
|
|
"mean_token_accuracy": 0.15428409725427628,
|
|
"num_tokens": 5100910.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.705980825424194,
|
|
"epoch": 2.2766457680250785,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047146939454170245,
|
|
"loss": 5.4627,
|
|
"mean_token_accuracy": 0.16400493532419205,
|
|
"num_tokens": 5109760.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.720902061462402,
|
|
"epoch": 2.280564263322884,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004713226235621672,
|
|
"loss": 5.4957,
|
|
"mean_token_accuracy": 0.15792311280965804,
|
|
"num_tokens": 5119193.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.61166467666626,
|
|
"epoch": 2.2844827586206895,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004711755017222714,
|
|
"loss": 5.4122,
|
|
"mean_token_accuracy": 0.1576576665043831,
|
|
"num_tokens": 5127480.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.743918991088867,
|
|
"epoch": 2.2884012539184955,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047102802928494563,
|
|
"loss": 5.5857,
|
|
"mean_token_accuracy": 0.15232446938753127,
|
|
"num_tokens": 5136648.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.688281393051147,
|
|
"epoch": 2.292319749216301,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004708802065137471,
|
|
"loss": 5.3771,
|
|
"mean_token_accuracy": 0.16786309629678725,
|
|
"num_tokens": 5144708.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.643393802642822,
|
|
"epoch": 2.2962382445141065,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004707320336728591,
|
|
"loss": 5.5237,
|
|
"mean_token_accuracy": 0.15052489414811135,
|
|
"num_tokens": 5153388.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.797506046295166,
|
|
"epoch": 2.300156739811912,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004705835110270904,
|
|
"loss": 5.4824,
|
|
"mean_token_accuracy": 0.15724072754383087,
|
|
"num_tokens": 5161874.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.700078773498535,
|
|
"epoch": 2.304075235109718,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047043463884187517,
|
|
"loss": 5.4791,
|
|
"mean_token_accuracy": 0.16156153082847596,
|
|
"num_tokens": 5170284.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.583881044387818,
|
|
"epoch": 2.3079937304075235,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047028541738327207,
|
|
"loss": 5.44,
|
|
"mean_token_accuracy": 0.15695979446172714,
|
|
"num_tokens": 5179547.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.691956996917725,
|
|
"epoch": 2.311912225705329,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004701358469179641,
|
|
"loss": 5.4472,
|
|
"mean_token_accuracy": 0.16403508335351943,
|
|
"num_tokens": 5188256.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.756633996963501,
|
|
"epoch": 2.3158307210031346,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004699859277132578,
|
|
"loss": 5.5902,
|
|
"mean_token_accuracy": 0.16007334217429162,
|
|
"num_tokens": 5197464.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.734570837020874,
|
|
"epoch": 2.3197492163009406,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046983566003708336,
|
|
"loss": 5.3792,
|
|
"mean_token_accuracy": 0.16429649144411088,
|
|
"num_tokens": 5205989.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.587563323974609,
|
|
"epoch": 2.323667711598746,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00046968504415799325,
|
|
"loss": 5.4346,
|
|
"mean_token_accuracy": 0.1603007957339287,
|
|
"num_tokens": 5214490.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.64924201965332,
|
|
"epoch": 2.3275862068965516,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004695340803451625,
|
|
"loss": 5.464,
|
|
"mean_token_accuracy": 0.16044287830591203,
|
|
"num_tokens": 5223335.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.629110479354859,
|
|
"epoch": 2.3315047021943576,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 5.4147,
|
|
"mean_token_accuracy": 0.15993678867816924,
|
|
"num_tokens": 5231665.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.594243144989013,
|
|
"epoch": 2.335423197492163,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004692311099980878,
|
|
"loss": 5.4005,
|
|
"mean_token_accuracy": 0.16494683176279068,
|
|
"num_tokens": 5239810.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.5794525146484375,
|
|
"epoch": 2.3393416927899686,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046907910400530097,
|
|
"loss": 5.369,
|
|
"mean_token_accuracy": 0.16893674433231354,
|
|
"num_tokens": 5247525.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.594454145431518,
|
|
"epoch": 2.343260188087774,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004689267511616868,
|
|
"loss": 5.4189,
|
|
"mean_token_accuracy": 0.15772538781166076,
|
|
"num_tokens": 5255528.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.617967176437378,
|
|
"epoch": 2.34717868338558,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046877405173952465,
|
|
"loss": 5.3633,
|
|
"mean_token_accuracy": 0.17105703949928283,
|
|
"num_tokens": 5263944.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.657676124572754,
|
|
"epoch": 2.3510971786833856,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000468621006011713,
|
|
"loss": 5.4912,
|
|
"mean_token_accuracy": 0.15778652876615523,
|
|
"num_tokens": 5271724.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.3510971786833856,
|
|
"eval_entropy": 5.565460778021997,
|
|
"eval_loss": 5.932183742523193,
|
|
"eval_mean_token_accuracy": 0.14874126017815614,
|
|
"eval_num_tokens": 5271724.0,
|
|
"eval_runtime": 2.835,
|
|
"eval_samples_per_second": 1453.963,
|
|
"eval_steps_per_second": 182.01,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 12750,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 7109320874112000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|