b5f745adf6
Model: fpadovani/dan-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.2552301255230125,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.768266344070435,
|
|
"epoch": 0.0041841004184100415,
|
|
"grad_norm": 12.875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.3834,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9848.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.7789147853851315,
|
|
"epoch": 0.008368200836820083,
|
|
"grad_norm": 14.3125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.2092,
|
|
"mean_token_accuracy": 0.00013605442363768817,
|
|
"num_tokens": 18196.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.821842193603516,
|
|
"epoch": 0.012552301255230125,
|
|
"grad_norm": 17.25,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.0541,
|
|
"mean_token_accuracy": 0.0002630572533234954,
|
|
"num_tokens": 26927.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.096258974075317,
|
|
"epoch": 0.016736401673640166,
|
|
"grad_norm": 27.125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.3612,
|
|
"mean_token_accuracy": 0.00012658227933570743,
|
|
"num_tokens": 34987.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.207125568389893,
|
|
"epoch": 0.02092050209205021,
|
|
"grad_norm": 19.5,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.9388,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 44936.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.419103336334228,
|
|
"epoch": 0.02510460251046025,
|
|
"grad_norm": 4.25,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.8146,
|
|
"mean_token_accuracy": 0.0020074134168680756,
|
|
"num_tokens": 54562.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.69917049407959,
|
|
"epoch": 0.029288702928870293,
|
|
"grad_norm": 3.296875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.5537,
|
|
"mean_token_accuracy": 0.0153768889605999,
|
|
"num_tokens": 63844.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.692293643951416,
|
|
"epoch": 0.03347280334728033,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2761,
|
|
"mean_token_accuracy": 0.017915738467127084,
|
|
"num_tokens": 74331.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.646587371826172,
|
|
"epoch": 0.03765690376569038,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.904,
|
|
"mean_token_accuracy": 0.028478659689426422,
|
|
"num_tokens": 83028.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.585913181304932,
|
|
"epoch": 0.04184100418410042,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6977,
|
|
"mean_token_accuracy": 0.04541164524853229,
|
|
"num_tokens": 91395.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.563683032989502,
|
|
"epoch": 0.04602510460251046,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5786,
|
|
"mean_token_accuracy": 0.048107235319912435,
|
|
"num_tokens": 101868.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.548775959014893,
|
|
"epoch": 0.0502092050209205,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4624,
|
|
"mean_token_accuracy": 0.04703106284141541,
|
|
"num_tokens": 110403.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.547953414916993,
|
|
"epoch": 0.05439330543933055,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.5071,
|
|
"mean_token_accuracy": 0.041691455990076065,
|
|
"num_tokens": 119557.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.555264663696288,
|
|
"epoch": 0.058577405857740586,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.3084,
|
|
"mean_token_accuracy": 0.04668764472007751,
|
|
"num_tokens": 128903.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.545580863952637,
|
|
"epoch": 0.06276150627615062,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.3022,
|
|
"mean_token_accuracy": 0.04386243894696236,
|
|
"num_tokens": 137630.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.519606494903565,
|
|
"epoch": 0.06694560669456066,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.12,
|
|
"mean_token_accuracy": 0.05209527462720871,
|
|
"num_tokens": 146335.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.482771396636963,
|
|
"epoch": 0.07112970711297072,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0283,
|
|
"mean_token_accuracy": 0.05407022833824158,
|
|
"num_tokens": 155766.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.450575256347657,
|
|
"epoch": 0.07531380753138076,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9477,
|
|
"mean_token_accuracy": 0.05165062099695206,
|
|
"num_tokens": 165844.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.425101280212402,
|
|
"epoch": 0.0794979079497908,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8842,
|
|
"mean_token_accuracy": 0.05370258502662182,
|
|
"num_tokens": 174858.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.32920789718628,
|
|
"epoch": 0.08368200836820083,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.7167,
|
|
"mean_token_accuracy": 0.05981660895049572,
|
|
"num_tokens": 184020.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.181023502349854,
|
|
"epoch": 0.08786610878661087,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.5856,
|
|
"mean_token_accuracy": 0.05290363691747189,
|
|
"num_tokens": 193714.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.198339366912842,
|
|
"epoch": 0.09205020920502092,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.5126,
|
|
"mean_token_accuracy": 0.0547657772898674,
|
|
"num_tokens": 203139.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.105471420288087,
|
|
"epoch": 0.09623430962343096,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.3484,
|
|
"mean_token_accuracy": 0.05634176060557365,
|
|
"num_tokens": 212446.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.886044597625732,
|
|
"epoch": 0.100418410041841,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.1725,
|
|
"mean_token_accuracy": 0.06103258766233921,
|
|
"num_tokens": 220910.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.764387893676759,
|
|
"epoch": 0.10460251046025104,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.096,
|
|
"mean_token_accuracy": 0.053500480577349664,
|
|
"num_tokens": 230522.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.598654460906982,
|
|
"epoch": 0.1087866108786611,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.918,
|
|
"mean_token_accuracy": 0.05621385425329208,
|
|
"num_tokens": 240092.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.302083587646484,
|
|
"epoch": 0.11297071129707113,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.8502,
|
|
"mean_token_accuracy": 0.05275089591741562,
|
|
"num_tokens": 249478.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.09952507019043,
|
|
"epoch": 0.11715481171548117,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.6323,
|
|
"mean_token_accuracy": 0.06236257366836071,
|
|
"num_tokens": 257714.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.845372676849365,
|
|
"epoch": 0.12133891213389121,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.672,
|
|
"mean_token_accuracy": 0.05868846960365772,
|
|
"num_tokens": 266569.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.660739707946778,
|
|
"epoch": 0.12552301255230125,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.5683,
|
|
"mean_token_accuracy": 0.056071584671735765,
|
|
"num_tokens": 276688.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.484380149841309,
|
|
"epoch": 0.1297071129707113,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.4186,
|
|
"mean_token_accuracy": 0.06363016776740552,
|
|
"num_tokens": 285430.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.278593254089355,
|
|
"epoch": 0.13389121338912133,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.4137,
|
|
"mean_token_accuracy": 0.06225575767457485,
|
|
"num_tokens": 295727.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.146096420288085,
|
|
"epoch": 0.13807531380753138,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.3823,
|
|
"mean_token_accuracy": 0.0689712069928646,
|
|
"num_tokens": 304854.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.988676738739014,
|
|
"epoch": 0.14225941422594143,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.3536,
|
|
"mean_token_accuracy": 0.07053346484899521,
|
|
"num_tokens": 313579.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.9785551071167,
|
|
"epoch": 0.14644351464435146,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.4132,
|
|
"mean_token_accuracy": 0.07007367871701717,
|
|
"num_tokens": 323310.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.922500085830689,
|
|
"epoch": 0.1506276150627615,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.2859,
|
|
"mean_token_accuracy": 0.0698877926915884,
|
|
"num_tokens": 331720.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.924635601043701,
|
|
"epoch": 0.15481171548117154,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.24,
|
|
"mean_token_accuracy": 0.07415937893092632,
|
|
"num_tokens": 340331.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.836748027801514,
|
|
"epoch": 0.1589958158995816,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.2493,
|
|
"mean_token_accuracy": 0.07680297121405602,
|
|
"num_tokens": 349677.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.846647310256958,
|
|
"epoch": 0.16317991631799164,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.2344,
|
|
"mean_token_accuracy": 0.06987173184752464,
|
|
"num_tokens": 358674.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.753779888153076,
|
|
"epoch": 0.16736401673640167,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.2296,
|
|
"mean_token_accuracy": 0.072166907787323,
|
|
"num_tokens": 368258.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.816751956939697,
|
|
"epoch": 0.17154811715481172,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.1436,
|
|
"mean_token_accuracy": 0.07943947166204453,
|
|
"num_tokens": 377083.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.698982191085816,
|
|
"epoch": 0.17573221757322174,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.1951,
|
|
"mean_token_accuracy": 0.07452086620032787,
|
|
"num_tokens": 385456.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.709808778762818,
|
|
"epoch": 0.1799163179916318,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.0971,
|
|
"mean_token_accuracy": 0.0767411794513464,
|
|
"num_tokens": 395295.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.692369031906128,
|
|
"epoch": 0.18410041841004185,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.1406,
|
|
"mean_token_accuracy": 0.07074209861457348,
|
|
"num_tokens": 405023.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.62271146774292,
|
|
"epoch": 0.18828451882845187,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.1623,
|
|
"mean_token_accuracy": 0.07522214204072952,
|
|
"num_tokens": 414224.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.621326494216919,
|
|
"epoch": 0.19246861924686193,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.0582,
|
|
"mean_token_accuracy": 0.0759123720228672,
|
|
"num_tokens": 423951.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.696398591995239,
|
|
"epoch": 0.19665271966527198,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.0805,
|
|
"mean_token_accuracy": 0.07548626773059368,
|
|
"num_tokens": 432427.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.593403148651123,
|
|
"epoch": 0.200836820083682,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.0032,
|
|
"mean_token_accuracy": 0.0808411881327629,
|
|
"num_tokens": 441560.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.589205026626587,
|
|
"epoch": 0.20502092050209206,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.092,
|
|
"mean_token_accuracy": 0.0739415742456913,
|
|
"num_tokens": 450364.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.541353750228882,
|
|
"epoch": 0.20920502092050208,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.0399,
|
|
"mean_token_accuracy": 0.0743038758635521,
|
|
"num_tokens": 460076.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.461675310134888,
|
|
"epoch": 0.21338912133891214,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.997,
|
|
"mean_token_accuracy": 0.07904429398477078,
|
|
"num_tokens": 468718.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.617227411270141,
|
|
"epoch": 0.2175732217573222,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.0585,
|
|
"mean_token_accuracy": 0.07864544317126274,
|
|
"num_tokens": 478436.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.475518226623535,
|
|
"epoch": 0.2217573221757322,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0628,
|
|
"mean_token_accuracy": 0.07705017924308777,
|
|
"num_tokens": 486833.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.514133024215698,
|
|
"epoch": 0.22594142259414227,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.9359,
|
|
"mean_token_accuracy": 0.08766618892550468,
|
|
"num_tokens": 495119.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.481584882736206,
|
|
"epoch": 0.2301255230125523,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1039,
|
|
"mean_token_accuracy": 0.08408359885215759,
|
|
"num_tokens": 505396.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.400933456420899,
|
|
"epoch": 0.23430962343096234,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.9713,
|
|
"mean_token_accuracy": 0.08110813125967979,
|
|
"num_tokens": 515259.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.482255935668945,
|
|
"epoch": 0.2384937238493724,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.9695,
|
|
"mean_token_accuracy": 0.08075925037264824,
|
|
"num_tokens": 523608.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.380628252029419,
|
|
"epoch": 0.24267782426778242,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.9614,
|
|
"mean_token_accuracy": 0.08079174906015396,
|
|
"num_tokens": 534613.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.431663274765015,
|
|
"epoch": 0.24686192468619247,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.9779,
|
|
"mean_token_accuracy": 0.08681405410170555,
|
|
"num_tokens": 543347.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.367568683624268,
|
|
"epoch": 0.2510460251046025,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9373,
|
|
"mean_token_accuracy": 0.08151624575257302,
|
|
"num_tokens": 552371.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.431672620773315,
|
|
"epoch": 0.25523012552301255,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.8709,
|
|
"mean_token_accuracy": 0.08783738985657692,
|
|
"num_tokens": 560742.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.372213315963745,
|
|
"epoch": 0.2594142259414226,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.9672,
|
|
"mean_token_accuracy": 0.08577736541628837,
|
|
"num_tokens": 569960.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.366195917129517,
|
|
"epoch": 0.26359832635983266,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.935,
|
|
"mean_token_accuracy": 0.08148402348160744,
|
|
"num_tokens": 578859.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.361504554748535,
|
|
"epoch": 0.26778242677824265,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9323,
|
|
"mean_token_accuracy": 0.08335322514176369,
|
|
"num_tokens": 588146.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.262340354919433,
|
|
"epoch": 0.2719665271966527,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.8837,
|
|
"mean_token_accuracy": 0.08676013201475144,
|
|
"num_tokens": 597268.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.286304664611817,
|
|
"epoch": 0.27615062761506276,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.8065,
|
|
"mean_token_accuracy": 0.09255077987909317,
|
|
"num_tokens": 606199.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.354407405853271,
|
|
"epoch": 0.2803347280334728,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9221,
|
|
"mean_token_accuracy": 0.07822546139359474,
|
|
"num_tokens": 615966.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.328773784637451,
|
|
"epoch": 0.28451882845188287,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.8936,
|
|
"mean_token_accuracy": 0.08315553963184356,
|
|
"num_tokens": 625689.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.292076635360718,
|
|
"epoch": 0.28870292887029286,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9258,
|
|
"mean_token_accuracy": 0.08275936022400857,
|
|
"num_tokens": 634877.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.311211967468262,
|
|
"epoch": 0.2928870292887029,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.8606,
|
|
"mean_token_accuracy": 0.08924969360232353,
|
|
"num_tokens": 644042.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.213941431045532,
|
|
"epoch": 0.29707112970711297,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.8394,
|
|
"mean_token_accuracy": 0.08540538996458054,
|
|
"num_tokens": 653072.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.225866985321045,
|
|
"epoch": 0.301255230125523,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8411,
|
|
"mean_token_accuracy": 0.08791597038507462,
|
|
"num_tokens": 663686.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.237736654281616,
|
|
"epoch": 0.3054393305439331,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.774,
|
|
"mean_token_accuracy": 0.0887495793402195,
|
|
"num_tokens": 672293.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.061402034759522,
|
|
"epoch": 0.30962343096234307,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7483,
|
|
"mean_token_accuracy": 0.09100355952978134,
|
|
"num_tokens": 681758.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.178809404373169,
|
|
"epoch": 0.3138075313807531,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7736,
|
|
"mean_token_accuracy": 0.09110940396785736,
|
|
"num_tokens": 690727.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.200173044204712,
|
|
"epoch": 0.3179916317991632,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.7292,
|
|
"mean_token_accuracy": 0.08735076114535331,
|
|
"num_tokens": 701210.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.065630149841309,
|
|
"epoch": 0.32217573221757323,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.7168,
|
|
"mean_token_accuracy": 0.09401161968708038,
|
|
"num_tokens": 710869.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.200049877166748,
|
|
"epoch": 0.3263598326359833,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8296,
|
|
"mean_token_accuracy": 0.08618490770459175,
|
|
"num_tokens": 721455.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.104738330841064,
|
|
"epoch": 0.3305439330543933,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7259,
|
|
"mean_token_accuracy": 0.09676406383514405,
|
|
"num_tokens": 730917.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.051578664779663,
|
|
"epoch": 0.33472803347280333,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.7446,
|
|
"mean_token_accuracy": 0.09235388711094857,
|
|
"num_tokens": 739302.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.140300941467285,
|
|
"epoch": 0.3389121338912134,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7141,
|
|
"mean_token_accuracy": 0.08636429682374,
|
|
"num_tokens": 748785.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.019905233383179,
|
|
"epoch": 0.34309623430962344,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.6677,
|
|
"mean_token_accuracy": 0.09201755002140999,
|
|
"num_tokens": 757215.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.04399356842041,
|
|
"epoch": 0.3472803347280335,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.5945,
|
|
"mean_token_accuracy": 0.09794526174664497,
|
|
"num_tokens": 765682.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.094897365570068,
|
|
"epoch": 0.3514644351464435,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7956,
|
|
"mean_token_accuracy": 0.08788713663816453,
|
|
"num_tokens": 774814.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.0791693210601805,
|
|
"epoch": 0.35564853556485354,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7096,
|
|
"mean_token_accuracy": 0.09370236843824387,
|
|
"num_tokens": 784019.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.943140363693237,
|
|
"epoch": 0.3598326359832636,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.5796,
|
|
"mean_token_accuracy": 0.09640125781297684,
|
|
"num_tokens": 793237.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.025385618209839,
|
|
"epoch": 0.36401673640167365,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6309,
|
|
"mean_token_accuracy": 0.09518588036298752,
|
|
"num_tokens": 801548.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.028780126571656,
|
|
"epoch": 0.3682008368200837,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.6643,
|
|
"mean_token_accuracy": 0.09595847800374031,
|
|
"num_tokens": 810974.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.958902835845947,
|
|
"epoch": 0.3723849372384937,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.5681,
|
|
"mean_token_accuracy": 0.09765126407146454,
|
|
"num_tokens": 820023.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.949744319915771,
|
|
"epoch": 0.37656903765690375,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6524,
|
|
"mean_token_accuracy": 0.09723440259695053,
|
|
"num_tokens": 828297.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.0269598960876465,
|
|
"epoch": 0.3807531380753138,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7028,
|
|
"mean_token_accuracy": 0.08575191274285317,
|
|
"num_tokens": 838268.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.016048622131348,
|
|
"epoch": 0.38493723849372385,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.623,
|
|
"mean_token_accuracy": 0.09345634952187538,
|
|
"num_tokens": 846491.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.9051624774932865,
|
|
"epoch": 0.3891213389121339,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.5585,
|
|
"mean_token_accuracy": 0.10123546048998833,
|
|
"num_tokens": 856501.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.930023145675659,
|
|
"epoch": 0.39330543933054396,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6253,
|
|
"mean_token_accuracy": 0.09574304819107056,
|
|
"num_tokens": 867054.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.9083311557769775,
|
|
"epoch": 0.39748953974895396,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6316,
|
|
"mean_token_accuracy": 0.09689460024237632,
|
|
"num_tokens": 875956.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.80507345199585,
|
|
"epoch": 0.401673640167364,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.5549,
|
|
"mean_token_accuracy": 0.09844572842121124,
|
|
"num_tokens": 884427.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.988656139373779,
|
|
"epoch": 0.40585774058577406,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6402,
|
|
"mean_token_accuracy": 0.10031970292329788,
|
|
"num_tokens": 894181.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.854498291015625,
|
|
"epoch": 0.4100418410041841,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.4914,
|
|
"mean_token_accuracy": 0.10208117067813874,
|
|
"num_tokens": 904076.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.82157187461853,
|
|
"epoch": 0.41422594142259417,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6406,
|
|
"mean_token_accuracy": 0.10029161274433136,
|
|
"num_tokens": 913915.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.97306637763977,
|
|
"epoch": 0.41841004184100417,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6935,
|
|
"mean_token_accuracy": 0.09564930200576782,
|
|
"num_tokens": 922876.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.41841004184100417,
|
|
"eval_entropy": 6.967948366527908,
|
|
"eval_loss": 6.639475345611572,
|
|
"eval_mean_token_accuracy": 0.09896371554743293,
|
|
"eval_num_tokens": 922876.0,
|
|
"eval_runtime": 2.8273,
|
|
"eval_samples_per_second": 1381.864,
|
|
"eval_steps_per_second": 172.954,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.84868803024292,
|
|
"epoch": 0.4225941422594142,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.4933,
|
|
"mean_token_accuracy": 0.10384752005338668,
|
|
"num_tokens": 931998.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.951689100265503,
|
|
"epoch": 0.42677824267782427,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.5497,
|
|
"mean_token_accuracy": 0.09913533478975296,
|
|
"num_tokens": 941387.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.808173084259034,
|
|
"epoch": 0.4309623430962343,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5857,
|
|
"mean_token_accuracy": 0.09512314647436142,
|
|
"num_tokens": 951866.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.7702148914337155,
|
|
"epoch": 0.4351464435146444,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5134,
|
|
"mean_token_accuracy": 0.1027239017188549,
|
|
"num_tokens": 961252.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.861008834838867,
|
|
"epoch": 0.4393305439330544,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5456,
|
|
"mean_token_accuracy": 0.10307287946343421,
|
|
"num_tokens": 969857.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.827121829986572,
|
|
"epoch": 0.4435146443514644,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.5208,
|
|
"mean_token_accuracy": 0.10130328312516212,
|
|
"num_tokens": 978754.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.812819004058838,
|
|
"epoch": 0.4476987447698745,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5024,
|
|
"mean_token_accuracy": 0.10306508988142013,
|
|
"num_tokens": 987935.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.823488569259643,
|
|
"epoch": 0.45188284518828453,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.4771,
|
|
"mean_token_accuracy": 0.1015734501183033,
|
|
"num_tokens": 996873.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.88533844947815,
|
|
"epoch": 0.4560669456066946,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.572,
|
|
"mean_token_accuracy": 0.097062898427248,
|
|
"num_tokens": 1006487.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.632162094116211,
|
|
"epoch": 0.4602510460251046,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.4142,
|
|
"mean_token_accuracy": 0.10585015416145324,
|
|
"num_tokens": 1015715.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.738191556930542,
|
|
"epoch": 0.46443514644351463,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.53,
|
|
"mean_token_accuracy": 0.10212068408727645,
|
|
"num_tokens": 1025091.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.789957094192505,
|
|
"epoch": 0.4686192468619247,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.4542,
|
|
"mean_token_accuracy": 0.10204173475503922,
|
|
"num_tokens": 1034174.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.847775793075561,
|
|
"epoch": 0.47280334728033474,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6568,
|
|
"mean_token_accuracy": 0.08913957327604294,
|
|
"num_tokens": 1043777.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.742847347259522,
|
|
"epoch": 0.4769874476987448,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.495,
|
|
"mean_token_accuracy": 0.09928980246186256,
|
|
"num_tokens": 1052749.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.8090441703796385,
|
|
"epoch": 0.4811715481171548,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5077,
|
|
"mean_token_accuracy": 0.10130611211061477,
|
|
"num_tokens": 1061564.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.75211615562439,
|
|
"epoch": 0.48535564853556484,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.447,
|
|
"mean_token_accuracy": 0.10412005931138993,
|
|
"num_tokens": 1070006.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.732285690307617,
|
|
"epoch": 0.4895397489539749,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4958,
|
|
"mean_token_accuracy": 0.10480489581823349,
|
|
"num_tokens": 1079799.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.671957015991211,
|
|
"epoch": 0.49372384937238495,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.3577,
|
|
"mean_token_accuracy": 0.10448934435844422,
|
|
"num_tokens": 1088842.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.6517304420471195,
|
|
"epoch": 0.497907949790795,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.4098,
|
|
"mean_token_accuracy": 0.11377439573407173,
|
|
"num_tokens": 1097960.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.644801044464112,
|
|
"epoch": 0.502092050209205,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.3789,
|
|
"mean_token_accuracy": 0.10490157902240753,
|
|
"num_tokens": 1106797.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.7212605476379395,
|
|
"epoch": 0.5062761506276151,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.5517,
|
|
"mean_token_accuracy": 0.10722804665565491,
|
|
"num_tokens": 1116384.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.467281103134155,
|
|
"epoch": 0.5104602510460251,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.313,
|
|
"mean_token_accuracy": 0.11083850488066674,
|
|
"num_tokens": 1125206.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.815744638442993,
|
|
"epoch": 0.5146443514644351,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.538,
|
|
"mean_token_accuracy": 0.10858194977045059,
|
|
"num_tokens": 1134110.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.790111780166626,
|
|
"epoch": 0.5188284518828452,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.529,
|
|
"mean_token_accuracy": 0.10404339656233788,
|
|
"num_tokens": 1142833.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.61668643951416,
|
|
"epoch": 0.5230125523012552,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4423,
|
|
"mean_token_accuracy": 0.10611013397574424,
|
|
"num_tokens": 1152006.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.617235422134399,
|
|
"epoch": 0.5271966527196653,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3947,
|
|
"mean_token_accuracy": 0.10517996922135353,
|
|
"num_tokens": 1161061.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.674160814285278,
|
|
"epoch": 0.5313807531380753,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.412,
|
|
"mean_token_accuracy": 0.10514549165964127,
|
|
"num_tokens": 1170754.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.587571859359741,
|
|
"epoch": 0.5355648535564853,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.3534,
|
|
"mean_token_accuracy": 0.11051966845989228,
|
|
"num_tokens": 1180854.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.644711542129516,
|
|
"epoch": 0.5397489539748954,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4216,
|
|
"mean_token_accuracy": 0.10841243639588356,
|
|
"num_tokens": 1190567.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.525138902664184,
|
|
"epoch": 0.5439330543933054,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.2638,
|
|
"mean_token_accuracy": 0.11364024132490158,
|
|
"num_tokens": 1200247.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.562865495681763,
|
|
"epoch": 0.5481171548117155,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.3777,
|
|
"mean_token_accuracy": 0.11130299195647239,
|
|
"num_tokens": 1209358.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.570280265808106,
|
|
"epoch": 0.5523012552301255,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.3962,
|
|
"mean_token_accuracy": 0.11054785400629044,
|
|
"num_tokens": 1217746.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.588881635665894,
|
|
"epoch": 0.5564853556485355,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.3589,
|
|
"mean_token_accuracy": 0.11272291392087937,
|
|
"num_tokens": 1226829.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.451903820037842,
|
|
"epoch": 0.5606694560669456,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.2812,
|
|
"mean_token_accuracy": 0.11265203654766083,
|
|
"num_tokens": 1236642.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.679233074188232,
|
|
"epoch": 0.5648535564853556,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4083,
|
|
"mean_token_accuracy": 0.11789099648594856,
|
|
"num_tokens": 1244934.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.4409881114959715,
|
|
"epoch": 0.5690376569037657,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.2919,
|
|
"mean_token_accuracy": 0.10770596191287041,
|
|
"num_tokens": 1254786.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.535488128662109,
|
|
"epoch": 0.5732217573221757,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.2508,
|
|
"mean_token_accuracy": 0.11365505158901215,
|
|
"num_tokens": 1263745.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.47140793800354,
|
|
"epoch": 0.5774058577405857,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.3904,
|
|
"mean_token_accuracy": 0.11604329645633697,
|
|
"num_tokens": 1273101.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.5898118019104,
|
|
"epoch": 0.5815899581589958,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.3736,
|
|
"mean_token_accuracy": 0.11040713787078857,
|
|
"num_tokens": 1282547.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.532029438018799,
|
|
"epoch": 0.5857740585774058,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4008,
|
|
"mean_token_accuracy": 0.10778395533561706,
|
|
"num_tokens": 1291594.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.550088500976562,
|
|
"epoch": 0.5899581589958159,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.316,
|
|
"mean_token_accuracy": 0.12029297426342964,
|
|
"num_tokens": 1300282.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.516840314865112,
|
|
"epoch": 0.5941422594142259,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3885,
|
|
"mean_token_accuracy": 0.11583921313285828,
|
|
"num_tokens": 1309304.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.46544771194458,
|
|
"epoch": 0.5983263598326359,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.2321,
|
|
"mean_token_accuracy": 0.11730972602963448,
|
|
"num_tokens": 1317638.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.597035646438599,
|
|
"epoch": 0.602510460251046,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4257,
|
|
"mean_token_accuracy": 0.11083399206399917,
|
|
"num_tokens": 1327242.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.4532746315002445,
|
|
"epoch": 0.606694560669456,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.2428,
|
|
"mean_token_accuracy": 0.1156072735786438,
|
|
"num_tokens": 1335880.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.484941577911377,
|
|
"epoch": 0.6108786610878661,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3266,
|
|
"mean_token_accuracy": 0.11275639981031418,
|
|
"num_tokens": 1344909.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.437925052642822,
|
|
"epoch": 0.6150627615062761,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.1876,
|
|
"mean_token_accuracy": 0.11960602849721909,
|
|
"num_tokens": 1353997.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.4058027267456055,
|
|
"epoch": 0.6192468619246861,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3435,
|
|
"mean_token_accuracy": 0.11241159439086915,
|
|
"num_tokens": 1363293.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.570994853973389,
|
|
"epoch": 0.6234309623430963,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3573,
|
|
"mean_token_accuracy": 0.11127854585647583,
|
|
"num_tokens": 1372174.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.500179719924927,
|
|
"epoch": 0.6276150627615062,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.2551,
|
|
"mean_token_accuracy": 0.11865946277976036,
|
|
"num_tokens": 1381067.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.397150754928589,
|
|
"epoch": 0.6317991631799164,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.2352,
|
|
"mean_token_accuracy": 0.12134608998894691,
|
|
"num_tokens": 1390072.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.405498886108399,
|
|
"epoch": 0.6359832635983264,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.2241,
|
|
"mean_token_accuracy": 0.12313714846968651,
|
|
"num_tokens": 1399391.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.417546701431275,
|
|
"epoch": 0.6401673640167364,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.232,
|
|
"mean_token_accuracy": 0.11985773444175721,
|
|
"num_tokens": 1408054.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.511100482940674,
|
|
"epoch": 0.6443514644351465,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.2528,
|
|
"mean_token_accuracy": 0.11063949167728424,
|
|
"num_tokens": 1417406.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.335081768035889,
|
|
"epoch": 0.6485355648535565,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3041,
|
|
"mean_token_accuracy": 0.11019203439354897,
|
|
"num_tokens": 1426907.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.429039239883423,
|
|
"epoch": 0.6527196652719666,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3593,
|
|
"mean_token_accuracy": 0.11048192381858826,
|
|
"num_tokens": 1436085.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.4894092082977295,
|
|
"epoch": 0.6569037656903766,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2572,
|
|
"mean_token_accuracy": 0.11703520268201828,
|
|
"num_tokens": 1444823.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.4137073993682865,
|
|
"epoch": 0.6610878661087866,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3072,
|
|
"mean_token_accuracy": 0.1172407552599907,
|
|
"num_tokens": 1454621.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.3564558029174805,
|
|
"epoch": 0.6652719665271967,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.1842,
|
|
"mean_token_accuracy": 0.11996964290738106,
|
|
"num_tokens": 1463784.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.409063911437988,
|
|
"epoch": 0.6694560669456067,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3005,
|
|
"mean_token_accuracy": 0.11776863262057305,
|
|
"num_tokens": 1473257.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.424042558670044,
|
|
"epoch": 0.6736401673640168,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2477,
|
|
"mean_token_accuracy": 0.11706730797886848,
|
|
"num_tokens": 1482195.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.349983596801758,
|
|
"epoch": 0.6778242677824268,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.286,
|
|
"mean_token_accuracy": 0.11897709146142006,
|
|
"num_tokens": 1491590.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.305568790435791,
|
|
"epoch": 0.6820083682008368,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.161,
|
|
"mean_token_accuracy": 0.12807343378663064,
|
|
"num_tokens": 1500208.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.405850791931153,
|
|
"epoch": 0.6861924686192469,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2892,
|
|
"mean_token_accuracy": 0.11845047548413276,
|
|
"num_tokens": 1509412.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.440489625930786,
|
|
"epoch": 0.6903765690376569,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2768,
|
|
"mean_token_accuracy": 0.11962562501430511,
|
|
"num_tokens": 1518426.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.419729852676392,
|
|
"epoch": 0.694560669456067,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.2377,
|
|
"mean_token_accuracy": 0.11658703610301017,
|
|
"num_tokens": 1528605.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.269686937332153,
|
|
"epoch": 0.698744769874477,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.1758,
|
|
"mean_token_accuracy": 0.12392851561307908,
|
|
"num_tokens": 1537449.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.348258590698242,
|
|
"epoch": 0.702928870292887,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.215,
|
|
"mean_token_accuracy": 0.12626431211829187,
|
|
"num_tokens": 1546743.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.275034379959107,
|
|
"epoch": 0.7071129707112971,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.213,
|
|
"mean_token_accuracy": 0.11712396070361138,
|
|
"num_tokens": 1555691.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.30648365020752,
|
|
"epoch": 0.7112970711297071,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.1749,
|
|
"mean_token_accuracy": 0.12647855207324027,
|
|
"num_tokens": 1564957.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.4026590347290036,
|
|
"epoch": 0.7154811715481172,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.2381,
|
|
"mean_token_accuracy": 0.1252881944179535,
|
|
"num_tokens": 1573901.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.268061876296997,
|
|
"epoch": 0.7196652719665272,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.2031,
|
|
"mean_token_accuracy": 0.12226112931966782,
|
|
"num_tokens": 1582644.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.437113666534424,
|
|
"epoch": 0.7238493723849372,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3427,
|
|
"mean_token_accuracy": 0.11745756193995475,
|
|
"num_tokens": 1592363.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.349365139007569,
|
|
"epoch": 0.7280334728033473,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2791,
|
|
"mean_token_accuracy": 0.12225733995437622,
|
|
"num_tokens": 1602012.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.268002462387085,
|
|
"epoch": 0.7322175732217573,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.1856,
|
|
"mean_token_accuracy": 0.12651218548417092,
|
|
"num_tokens": 1611625.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.227978944778442,
|
|
"epoch": 0.7364016736401674,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.1431,
|
|
"mean_token_accuracy": 0.11568873450160026,
|
|
"num_tokens": 1620941.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.319770097732544,
|
|
"epoch": 0.7405857740585774,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2285,
|
|
"mean_token_accuracy": 0.11750153079628944,
|
|
"num_tokens": 1629225.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.305209541320801,
|
|
"epoch": 0.7447698744769874,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2069,
|
|
"mean_token_accuracy": 0.11938620209693909,
|
|
"num_tokens": 1639551.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.2366736888885494,
|
|
"epoch": 0.7489539748953975,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.1918,
|
|
"mean_token_accuracy": 0.11552652940154076,
|
|
"num_tokens": 1648963.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.2946977615356445,
|
|
"epoch": 0.7531380753138075,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.1081,
|
|
"mean_token_accuracy": 0.12913521379232407,
|
|
"num_tokens": 1657735.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.3238130569458,
|
|
"epoch": 0.7573221757322176,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.155,
|
|
"mean_token_accuracy": 0.12631697207689285,
|
|
"num_tokens": 1666349.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.150327157974243,
|
|
"epoch": 0.7615062761506276,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.0795,
|
|
"mean_token_accuracy": 0.126066704839468,
|
|
"num_tokens": 1675920.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.217988586425781,
|
|
"epoch": 0.7656903765690377,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.113,
|
|
"mean_token_accuracy": 0.11986384242773056,
|
|
"num_tokens": 1685626.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.319421195983887,
|
|
"epoch": 0.7698744769874477,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2357,
|
|
"mean_token_accuracy": 0.13009796813130378,
|
|
"num_tokens": 1694972.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.216734409332275,
|
|
"epoch": 0.7740585774058577,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.1308,
|
|
"mean_token_accuracy": 0.13374294117093086,
|
|
"num_tokens": 1703806.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.2389215469360355,
|
|
"epoch": 0.7782426778242678,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.2127,
|
|
"mean_token_accuracy": 0.12826964035630226,
|
|
"num_tokens": 1712949.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.404123306274414,
|
|
"epoch": 0.7824267782426778,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3299,
|
|
"mean_token_accuracy": 0.125592914968729,
|
|
"num_tokens": 1722328.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.350027418136596,
|
|
"epoch": 0.7866108786610879,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3068,
|
|
"mean_token_accuracy": 0.12012889161705971,
|
|
"num_tokens": 1731880.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.140142631530762,
|
|
"epoch": 0.7907949790794979,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.1248,
|
|
"mean_token_accuracy": 0.12643283903598784,
|
|
"num_tokens": 1740719.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.273271703720093,
|
|
"epoch": 0.7949790794979079,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1513,
|
|
"mean_token_accuracy": 0.12553820610046387,
|
|
"num_tokens": 1749753.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.292899465560913,
|
|
"epoch": 0.799163179916318,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1807,
|
|
"mean_token_accuracy": 0.12436754852533341,
|
|
"num_tokens": 1758913.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.273108959197998,
|
|
"epoch": 0.803347280334728,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1882,
|
|
"mean_token_accuracy": 0.12275903448462486,
|
|
"num_tokens": 1768536.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.233821058273316,
|
|
"epoch": 0.8075313807531381,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1241,
|
|
"mean_token_accuracy": 0.12568357735872268,
|
|
"num_tokens": 1777291.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.211701774597168,
|
|
"epoch": 0.8117154811715481,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2163,
|
|
"mean_token_accuracy": 0.12532426416873932,
|
|
"num_tokens": 1787186.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.19328761100769,
|
|
"epoch": 0.8158995815899581,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.0235,
|
|
"mean_token_accuracy": 0.1305873580276966,
|
|
"num_tokens": 1795581.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.08319125175476,
|
|
"epoch": 0.8200836820083682,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.0153,
|
|
"mean_token_accuracy": 0.13362710550427437,
|
|
"num_tokens": 1804627.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.261023092269897,
|
|
"epoch": 0.8242677824267782,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.1549,
|
|
"mean_token_accuracy": 0.1278303623199463,
|
|
"num_tokens": 1813808.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.165553712844849,
|
|
"epoch": 0.8284518828451883,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1557,
|
|
"mean_token_accuracy": 0.13012527599930762,
|
|
"num_tokens": 1823422.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.19823989868164,
|
|
"epoch": 0.8326359832635983,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1752,
|
|
"mean_token_accuracy": 0.13083416000008583,
|
|
"num_tokens": 1832437.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.2002424716949465,
|
|
"epoch": 0.8368200836820083,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.1044,
|
|
"mean_token_accuracy": 0.12464204803109169,
|
|
"num_tokens": 1842063.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8368200836820083,
|
|
"eval_entropy": 6.097238668391066,
|
|
"eval_loss": 6.15334939956665,
|
|
"eval_mean_token_accuracy": 0.1348609975815307,
|
|
"eval_num_tokens": 1842063.0,
|
|
"eval_runtime": 2.7384,
|
|
"eval_samples_per_second": 1426.743,
|
|
"eval_steps_per_second": 178.571,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.241647672653198,
|
|
"epoch": 0.8410041841004184,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999998518355666,
|
|
"loss": 6.2485,
|
|
"mean_token_accuracy": 0.12185878977179528,
|
|
"num_tokens": 1851043.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.126685523986817,
|
|
"epoch": 0.8451882845188284,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999992499178905,
|
|
"loss": 6.0931,
|
|
"mean_token_accuracy": 0.13160960227251053,
|
|
"num_tokens": 1860268.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.303120422363281,
|
|
"epoch": 0.8493723849372385,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999981849879322,
|
|
"loss": 6.2042,
|
|
"mean_token_accuracy": 0.1309683345258236,
|
|
"num_tokens": 1870096.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.130255317687988,
|
|
"epoch": 0.8535564853556485,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999966570478833,
|
|
"loss": 6.0849,
|
|
"mean_token_accuracy": 0.1364796683192253,
|
|
"num_tokens": 1879164.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.1024699211120605,
|
|
"epoch": 0.8577405857740585,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999946661008878,
|
|
"loss": 6.0328,
|
|
"mean_token_accuracy": 0.13583800122141837,
|
|
"num_tokens": 1888135.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.158143806457519,
|
|
"epoch": 0.8619246861924686,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499992212151043,
|
|
"loss": 6.1092,
|
|
"mean_token_accuracy": 0.1269300736486912,
|
|
"num_tokens": 1898302.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.110213136672973,
|
|
"epoch": 0.8661087866108786,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999892952033987,
|
|
"loss": 6.0712,
|
|
"mean_token_accuracy": 0.129842671751976,
|
|
"num_tokens": 1907753.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.154698228836059,
|
|
"epoch": 0.8702928870292888,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999859152639573,
|
|
"loss": 6.1328,
|
|
"mean_token_accuracy": 0.12858494743704796,
|
|
"num_tokens": 1917543.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.2287651062011715,
|
|
"epoch": 0.8744769874476988,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999820723396745,
|
|
"loss": 6.1718,
|
|
"mean_token_accuracy": 0.12609437778592109,
|
|
"num_tokens": 1927335.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.19956407546997,
|
|
"epoch": 0.8786610878661087,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999777664384581,
|
|
"loss": 6.0957,
|
|
"mean_token_accuracy": 0.13153462409973143,
|
|
"num_tokens": 1936381.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.174496364593506,
|
|
"epoch": 0.8828451882845189,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999729975691692,
|
|
"loss": 6.1341,
|
|
"mean_token_accuracy": 0.12876357659697532,
|
|
"num_tokens": 1945559.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.041715145111084,
|
|
"epoch": 0.8870292887029289,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999677657416212,
|
|
"loss": 6.0012,
|
|
"mean_token_accuracy": 0.12970590144395827,
|
|
"num_tokens": 1954709.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.168726396560669,
|
|
"epoch": 0.891213389121339,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999620709665804,
|
|
"loss": 6.1249,
|
|
"mean_token_accuracy": 0.1281352959573269,
|
|
"num_tokens": 1963664.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.181927251815796,
|
|
"epoch": 0.895397489539749,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999559132557659,
|
|
"loss": 6.171,
|
|
"mean_token_accuracy": 0.12771156281232834,
|
|
"num_tokens": 1974103.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.23676552772522,
|
|
"epoch": 0.899581589958159,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999492926218489,
|
|
"loss": 6.0457,
|
|
"mean_token_accuracy": 0.1328116737306118,
|
|
"num_tokens": 1982924.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.003208255767822,
|
|
"epoch": 0.9037656903765691,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499942209078454,
|
|
"loss": 5.973,
|
|
"mean_token_accuracy": 0.14368802681565285,
|
|
"num_tokens": 1992033.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.048687314987182,
|
|
"epoch": 0.9079497907949791,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999346626401578,
|
|
"loss": 5.9894,
|
|
"mean_token_accuracy": 0.13765854164958,
|
|
"num_tokens": 2001218.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.094419670104981,
|
|
"epoch": 0.9121338912133892,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999266533224897,
|
|
"loss": 6.1441,
|
|
"mean_token_accuracy": 0.13423838838934898,
|
|
"num_tokens": 2010463.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.1051229476928714,
|
|
"epoch": 0.9163179916317992,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004999181811419314,
|
|
"loss": 6.0181,
|
|
"mean_token_accuracy": 0.13275936618447304,
|
|
"num_tokens": 2019246.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.170205354690552,
|
|
"epoch": 0.9205020920502092,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999092461159177,
|
|
"loss": 6.0876,
|
|
"mean_token_accuracy": 0.12886958941817284,
|
|
"num_tokens": 2028083.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.076777601242066,
|
|
"epoch": 0.9246861924686193,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998998482628351,
|
|
"loss": 5.9235,
|
|
"mean_token_accuracy": 0.1374969571828842,
|
|
"num_tokens": 2036631.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.035492610931397,
|
|
"epoch": 0.9288702928870293,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004998899876020229,
|
|
"loss": 6.0327,
|
|
"mean_token_accuracy": 0.133451097458601,
|
|
"num_tokens": 2045910.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.021814250946045,
|
|
"epoch": 0.9330543933054394,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499879664153773,
|
|
"loss": 6.0868,
|
|
"mean_token_accuracy": 0.14059085845947267,
|
|
"num_tokens": 2055959.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.29487099647522,
|
|
"epoch": 0.9372384937238494,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998688779393291,
|
|
"loss": 6.1577,
|
|
"mean_token_accuracy": 0.12911083772778512,
|
|
"num_tokens": 2065475.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 5.898234081268311,
|
|
"epoch": 0.9414225941422594,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998576289808877,
|
|
"loss": 5.9211,
|
|
"mean_token_accuracy": 0.1388360470533371,
|
|
"num_tokens": 2074469.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.0572669982910154,
|
|
"epoch": 0.9456066945606695,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998459173015972,
|
|
"loss": 6.0961,
|
|
"mean_token_accuracy": 0.13587369546294212,
|
|
"num_tokens": 2084604.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.207867860794067,
|
|
"epoch": 0.9497907949790795,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998337429255584,
|
|
"loss": 6.0656,
|
|
"mean_token_accuracy": 0.1341245450079441,
|
|
"num_tokens": 2093377.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.09445858001709,
|
|
"epoch": 0.9539748953974896,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998211058778243,
|
|
"loss": 6.0838,
|
|
"mean_token_accuracy": 0.13606315553188325,
|
|
"num_tokens": 2102798.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.12096939086914,
|
|
"epoch": 0.9581589958158996,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998080061843999,
|
|
"loss": 6.0464,
|
|
"mean_token_accuracy": 0.1335694245994091,
|
|
"num_tokens": 2112516.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.146814584732056,
|
|
"epoch": 0.9623430962343096,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499794443872242,
|
|
"loss": 6.1417,
|
|
"mean_token_accuracy": 0.13098908141255378,
|
|
"num_tokens": 2121677.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.00798225402832,
|
|
"epoch": 0.9665271966527197,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997804189692601,
|
|
"loss": 5.971,
|
|
"mean_token_accuracy": 0.1396520145237446,
|
|
"num_tokens": 2131220.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.118230247497559,
|
|
"epoch": 0.9707112970711297,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997659315043149,
|
|
"loss": 6.1345,
|
|
"mean_token_accuracy": 0.13085383027791977,
|
|
"num_tokens": 2142104.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.051174783706665,
|
|
"epoch": 0.9748953974895398,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997509815072191,
|
|
"loss": 6.0492,
|
|
"mean_token_accuracy": 0.13568617030978203,
|
|
"num_tokens": 2151970.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.027523326873779,
|
|
"epoch": 0.9790794979079498,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997355690087379,
|
|
"loss": 6.0072,
|
|
"mean_token_accuracy": 0.1357894979417324,
|
|
"num_tokens": 2161371.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.045231628417969,
|
|
"epoch": 0.9832635983263598,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997196940405874,
|
|
"loss": 5.9485,
|
|
"mean_token_accuracy": 0.13663906380534172,
|
|
"num_tokens": 2170650.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 5.984046506881714,
|
|
"epoch": 0.9874476987447699,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997033566354358,
|
|
"loss": 6.028,
|
|
"mean_token_accuracy": 0.1392426609992981,
|
|
"num_tokens": 2180934.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.154821681976318,
|
|
"epoch": 0.9916317991631799,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996865568269028,
|
|
"loss": 5.9669,
|
|
"mean_token_accuracy": 0.14529061168432236,
|
|
"num_tokens": 2189894.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 5.945505809783936,
|
|
"epoch": 0.99581589958159,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996692946495597,
|
|
"loss": 5.9377,
|
|
"mean_token_accuracy": 0.14892658293247224,
|
|
"num_tokens": 2200208.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.064693832397461,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004996515701389295,
|
|
"loss": 5.9726,
|
|
"mean_token_accuracy": 0.13724684566259385,
|
|
"num_tokens": 2209322.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 5.993635749816894,
|
|
"epoch": 1.00418410041841,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004996333833314862,
|
|
"loss": 5.7905,
|
|
"mean_token_accuracy": 0.15274174511432648,
|
|
"num_tokens": 2217663.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 5.90445466041565,
|
|
"epoch": 1.00836820083682,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996147342646554,
|
|
"loss": 5.8433,
|
|
"mean_token_accuracy": 0.14284583255648614,
|
|
"num_tokens": 2226559.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.067541551589966,
|
|
"epoch": 1.0125523012552302,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004995956229768137,
|
|
"loss": 5.8828,
|
|
"mean_token_accuracy": 0.14996479004621505,
|
|
"num_tokens": 2235468.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 5.908719444274903,
|
|
"epoch": 1.0167364016736402,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995760495072893,
|
|
"loss": 5.6674,
|
|
"mean_token_accuracy": 0.15197395384311677,
|
|
"num_tokens": 2245176.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 5.926908683776856,
|
|
"epoch": 1.0209205020920502,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995560138963611,
|
|
"loss": 5.8325,
|
|
"mean_token_accuracy": 0.1469671368598938,
|
|
"num_tokens": 2255123.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 5.958147192001343,
|
|
"epoch": 1.0251046025104602,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995355161852593,
|
|
"loss": 5.7525,
|
|
"mean_token_accuracy": 0.1474505454301834,
|
|
"num_tokens": 2263596.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 5.980195045471191,
|
|
"epoch": 1.0292887029288702,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004995145564161648,
|
|
"loss": 5.8006,
|
|
"mean_token_accuracy": 0.13994376361370087,
|
|
"num_tokens": 2272228.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.907618856430053,
|
|
"epoch": 1.0334728033472804,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004994931346322094,
|
|
"loss": 5.8431,
|
|
"mean_token_accuracy": 0.1445181243121624,
|
|
"num_tokens": 2281248.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.041008806228637,
|
|
"epoch": 1.0376569037656904,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004994712508774757,
|
|
"loss": 5.8114,
|
|
"mean_token_accuracy": 0.1452713206410408,
|
|
"num_tokens": 2290958.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.83619909286499,
|
|
"epoch": 1.0418410041841004,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499448905196997,
|
|
"loss": 5.7584,
|
|
"mean_token_accuracy": 0.15415944904088974,
|
|
"num_tokens": 2299763.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.935490417480469,
|
|
"epoch": 1.0460251046025104,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004994260976367573,
|
|
"loss": 5.7788,
|
|
"mean_token_accuracy": 0.15245575234293937,
|
|
"num_tokens": 2309847.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 5.983281230926513,
|
|
"epoch": 1.0502092050209204,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004994028282436905,
|
|
"loss": 5.8069,
|
|
"mean_token_accuracy": 0.14806725829839706,
|
|
"num_tokens": 2317556.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 5.781259489059448,
|
|
"epoch": 1.0543933054393306,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993790970656814,
|
|
"loss": 5.7374,
|
|
"mean_token_accuracy": 0.1501047745347023,
|
|
"num_tokens": 2327151.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.952362155914306,
|
|
"epoch": 1.0585774058577406,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004993549041515652,
|
|
"loss": 5.7418,
|
|
"mean_token_accuracy": 0.1499510034918785,
|
|
"num_tokens": 2336119.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 5.803351783752442,
|
|
"epoch": 1.0627615062761506,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004993302495511268,
|
|
"loss": 5.7755,
|
|
"mean_token_accuracy": 0.14605644345283508,
|
|
"num_tokens": 2345262.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.077183580398559,
|
|
"epoch": 1.0669456066945606,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004993051333151014,
|
|
"loss": 5.8956,
|
|
"mean_token_accuracy": 0.14342452213168144,
|
|
"num_tokens": 2355210.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.791180896759033,
|
|
"epoch": 1.0711297071129706,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004992795554951742,
|
|
"loss": 5.6401,
|
|
"mean_token_accuracy": 0.15315188989043235,
|
|
"num_tokens": 2364599.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.763968133926392,
|
|
"epoch": 1.0753138075313808,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004992535161439803,
|
|
"loss": 5.8054,
|
|
"mean_token_accuracy": 0.1445068120956421,
|
|
"num_tokens": 2374143.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.005795478820801,
|
|
"epoch": 1.0794979079497908,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004992270153151045,
|
|
"loss": 5.8188,
|
|
"mean_token_accuracy": 0.14890599846839905,
|
|
"num_tokens": 2383155.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.011566638946533,
|
|
"epoch": 1.0836820083682008,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004992000530630812,
|
|
"loss": 5.8663,
|
|
"mean_token_accuracy": 0.14473781883716583,
|
|
"num_tokens": 2392542.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.806177711486816,
|
|
"epoch": 1.0878661087866108,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004991726294433943,
|
|
"loss": 5.8161,
|
|
"mean_token_accuracy": 0.1449222221970558,
|
|
"num_tokens": 2401851.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.932667779922485,
|
|
"epoch": 1.0920502092050208,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004991447445124774,
|
|
"loss": 5.7593,
|
|
"mean_token_accuracy": 0.149945068359375,
|
|
"num_tokens": 2410726.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.812711238861084,
|
|
"epoch": 1.096234309623431,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499116398327713,
|
|
"loss": 5.8305,
|
|
"mean_token_accuracy": 0.14005152210593225,
|
|
"num_tokens": 2420702.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.843758773803711,
|
|
"epoch": 1.100418410041841,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004990875909474331,
|
|
"loss": 5.7499,
|
|
"mean_token_accuracy": 0.14834965020418167,
|
|
"num_tokens": 2431335.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 5.966901779174805,
|
|
"epoch": 1.104602510460251,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004990583224309187,
|
|
"loss": 5.7261,
|
|
"mean_token_accuracy": 0.15352486222982406,
|
|
"num_tokens": 2440722.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.845820617675781,
|
|
"epoch": 1.108786610878661,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004990285928383997,
|
|
"loss": 5.8035,
|
|
"mean_token_accuracy": 0.14780879467725755,
|
|
"num_tokens": 2451230.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.904784488677978,
|
|
"epoch": 1.112970711297071,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004989984022310545,
|
|
"loss": 5.8394,
|
|
"mean_token_accuracy": 0.14220331013202667,
|
|
"num_tokens": 2461442.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.87468147277832,
|
|
"epoch": 1.1171548117154813,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004989677506710109,
|
|
"loss": 5.6527,
|
|
"mean_token_accuracy": 0.1563212588429451,
|
|
"num_tokens": 2471381.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.808247089385986,
|
|
"epoch": 1.1213389121338913,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004989366382213449,
|
|
"loss": 5.7156,
|
|
"mean_token_accuracy": 0.15480270683765412,
|
|
"num_tokens": 2479937.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.837823104858399,
|
|
"epoch": 1.1255230125523012,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989050649460806,
|
|
"loss": 5.6731,
|
|
"mean_token_accuracy": 0.1569208934903145,
|
|
"num_tokens": 2488974.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.856475782394409,
|
|
"epoch": 1.1297071129707112,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000498873030910191,
|
|
"loss": 5.8444,
|
|
"mean_token_accuracy": 0.1466200530529022,
|
|
"num_tokens": 2498800.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.822557353973389,
|
|
"epoch": 1.1338912133891212,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004988405361795969,
|
|
"loss": 5.7363,
|
|
"mean_token_accuracy": 0.14498689994215966,
|
|
"num_tokens": 2506580.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.84963550567627,
|
|
"epoch": 1.1380753138075315,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004988075808211672,
|
|
"loss": 5.7934,
|
|
"mean_token_accuracy": 0.15228895395994185,
|
|
"num_tokens": 2516342.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.809200716018677,
|
|
"epoch": 1.1422594142259415,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004987741649027189,
|
|
"loss": 5.8174,
|
|
"mean_token_accuracy": 0.14777714088559152,
|
|
"num_tokens": 2526054.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.951186895370483,
|
|
"epoch": 1.1464435146443515,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004987402884930163,
|
|
"loss": 5.7855,
|
|
"mean_token_accuracy": 0.1517642021179199,
|
|
"num_tokens": 2535556.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.828822660446167,
|
|
"epoch": 1.1506276150627615,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004987059516617717,
|
|
"loss": 5.7459,
|
|
"mean_token_accuracy": 0.1545324891805649,
|
|
"num_tokens": 2544930.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.7893500328063965,
|
|
"epoch": 1.1548117154811715,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004986711544796449,
|
|
"loss": 5.6609,
|
|
"mean_token_accuracy": 0.1535905510187149,
|
|
"num_tokens": 2554362.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.884280490875244,
|
|
"epoch": 1.1589958158995817,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000498635897018243,
|
|
"loss": 5.777,
|
|
"mean_token_accuracy": 0.14912986755371094,
|
|
"num_tokens": 2563499.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.8048481941223145,
|
|
"epoch": 1.1631799163179917,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004986001793501198,
|
|
"loss": 5.7883,
|
|
"mean_token_accuracy": 0.15240272358059884,
|
|
"num_tokens": 2573346.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.857279062271118,
|
|
"epoch": 1.1673640167364017,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004985640015487768,
|
|
"loss": 5.665,
|
|
"mean_token_accuracy": 0.15898251682519912,
|
|
"num_tokens": 2582541.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.832502317428589,
|
|
"epoch": 1.1715481171548117,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004985273636886621,
|
|
"loss": 5.7536,
|
|
"mean_token_accuracy": 0.14916445165872574,
|
|
"num_tokens": 2592015.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.822238969802856,
|
|
"epoch": 1.1757322175732217,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004984902658451704,
|
|
"loss": 5.7545,
|
|
"mean_token_accuracy": 0.15448953062295914,
|
|
"num_tokens": 2601998.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.775497055053711,
|
|
"epoch": 1.1799163179916319,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004984527080946432,
|
|
"loss": 5.6909,
|
|
"mean_token_accuracy": 0.15859137922525407,
|
|
"num_tokens": 2611435.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.902001571655274,
|
|
"epoch": 1.1841004184100419,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004984146905143683,
|
|
"loss": 5.7934,
|
|
"mean_token_accuracy": 0.14567320495843888,
|
|
"num_tokens": 2621639.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.837963247299195,
|
|
"epoch": 1.1882845188284519,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004983762131825797,
|
|
"loss": 5.7495,
|
|
"mean_token_accuracy": 0.1509477600455284,
|
|
"num_tokens": 2631390.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.777963781356812,
|
|
"epoch": 1.1924686192468619,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983372761784576,
|
|
"loss": 5.7968,
|
|
"mean_token_accuracy": 0.14694754034280777,
|
|
"num_tokens": 2640769.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.9535966396331785,
|
|
"epoch": 1.196652719665272,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004982978795821281,
|
|
"loss": 5.7937,
|
|
"mean_token_accuracy": 0.1512495271861553,
|
|
"num_tokens": 2651137.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.794545030593872,
|
|
"epoch": 1.200836820083682,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004982580234746632,
|
|
"loss": 5.7258,
|
|
"mean_token_accuracy": 0.1513078287243843,
|
|
"num_tokens": 2660530.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.867940950393677,
|
|
"epoch": 1.205020920502092,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00049821770793808,
|
|
"loss": 5.784,
|
|
"mean_token_accuracy": 0.1505849204957485,
|
|
"num_tokens": 2669541.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.860204744338989,
|
|
"epoch": 1.209205020920502,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004981769330553418,
|
|
"loss": 5.7822,
|
|
"mean_token_accuracy": 0.1534332662820816,
|
|
"num_tokens": 2678542.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.87125940322876,
|
|
"epoch": 1.213389121338912,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004981356989103566,
|
|
"loss": 5.8384,
|
|
"mean_token_accuracy": 0.15294507145881653,
|
|
"num_tokens": 2687720.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.862550973892212,
|
|
"epoch": 1.217573221757322,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004980940055879774,
|
|
"loss": 5.698,
|
|
"mean_token_accuracy": 0.15199038982391358,
|
|
"num_tokens": 2697641.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.753113889694214,
|
|
"epoch": 1.2217573221757323,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004980518531740027,
|
|
"loss": 5.7504,
|
|
"mean_token_accuracy": 0.14872822910547256,
|
|
"num_tokens": 2706600.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.950168657302856,
|
|
"epoch": 1.2259414225941423,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004980092417551752,
|
|
"loss": 5.8755,
|
|
"mean_token_accuracy": 0.14688518047332763,
|
|
"num_tokens": 2716633.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.817790365219116,
|
|
"epoch": 1.2301255230125523,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004979661714191825,
|
|
"loss": 5.6416,
|
|
"mean_token_accuracy": 0.1666289746761322,
|
|
"num_tokens": 2724829.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.805331134796143,
|
|
"epoch": 1.2343096234309623,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004979226422546563,
|
|
"loss": 5.7465,
|
|
"mean_token_accuracy": 0.15683723539113997,
|
|
"num_tokens": 2734276.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.913634300231934,
|
|
"epoch": 1.2384937238493725,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004978786543511726,
|
|
"loss": 5.8622,
|
|
"mean_token_accuracy": 0.15280352383852006,
|
|
"num_tokens": 2744413.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.847559309005737,
|
|
"epoch": 1.2426778242677825,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004978342077992515,
|
|
"loss": 5.7394,
|
|
"mean_token_accuracy": 0.1537095919251442,
|
|
"num_tokens": 2753242.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.650951862335205,
|
|
"epoch": 1.2468619246861925,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004977893026903568,
|
|
"loss": 5.6605,
|
|
"mean_token_accuracy": 0.16106596887111663,
|
|
"num_tokens": 2761811.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.895610046386719,
|
|
"epoch": 1.2510460251046025,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004977439391168963,
|
|
"loss": 5.7547,
|
|
"mean_token_accuracy": 0.1545099973678589,
|
|
"num_tokens": 2770560.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.8653045177459715,
|
|
"epoch": 1.2552301255230125,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004976981171722205,
|
|
"loss": 5.7531,
|
|
"mean_token_accuracy": 0.15552559345960618,
|
|
"num_tokens": 2780131.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2552301255230125,
|
|
"eval_entropy": 5.61207090565032,
|
|
"eval_loss": 5.8904619216918945,
|
|
"eval_mean_token_accuracy": 0.15481723342205117,
|
|
"eval_num_tokens": 2780131.0,
|
|
"eval_runtime": 2.9514,
|
|
"eval_samples_per_second": 1323.781,
|
|
"eval_steps_per_second": 165.684,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11950,
|
|
"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": 4056256677888000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|