c867a9a1f1
Model: fpadovani/ind-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.39184952978056425,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 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
|
|
}
|
|
],
|
|
"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": 1156331556864000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|