b5f745adf6
Model: fpadovani/dan-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
11156 lines
304 KiB
JSON
11156 lines
304 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.602510460251046,
|
|
"eval_steps": 500,
|
|
"global_step": 5500,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 5.848963212966919,
|
|
"epoch": 1.2594142259414225,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004976518369506237,
|
|
"loss": 5.8545,
|
|
"mean_token_accuracy": 0.1463423416018486,
|
|
"num_tokens": 2789319.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.7695282936096195,
|
|
"epoch": 1.2635983263598327,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004976050985473433,
|
|
"loss": 5.7321,
|
|
"mean_token_accuracy": 0.14833602607250213,
|
|
"num_tokens": 2799253.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.928028345108032,
|
|
"epoch": 1.2677824267782427,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004975579020585593,
|
|
"loss": 5.8401,
|
|
"mean_token_accuracy": 0.14651586264371871,
|
|
"num_tokens": 2808968.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.814293909072876,
|
|
"epoch": 1.2719665271966527,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004975102475813947,
|
|
"loss": 5.7177,
|
|
"mean_token_accuracy": 0.15377511978149414,
|
|
"num_tokens": 2817282.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.87192554473877,
|
|
"epoch": 1.2761506276150627,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004974621352139144,
|
|
"loss": 5.7843,
|
|
"mean_token_accuracy": 0.14812919795513152,
|
|
"num_tokens": 2826395.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.8476917266845705,
|
|
"epoch": 1.280334728033473,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004974135650551262,
|
|
"loss": 5.865,
|
|
"mean_token_accuracy": 0.14753438383340836,
|
|
"num_tokens": 2836195.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.853739261627197,
|
|
"epoch": 1.284518828451883,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004973645372049796,
|
|
"loss": 5.693,
|
|
"mean_token_accuracy": 0.15756383240222932,
|
|
"num_tokens": 2846020.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.688941192626953,
|
|
"epoch": 1.288702928870293,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004973150517643659,
|
|
"loss": 5.6256,
|
|
"mean_token_accuracy": 0.16124313473701476,
|
|
"num_tokens": 2854895.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.844737005233765,
|
|
"epoch": 1.292887029288703,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004972651088351185,
|
|
"loss": 5.6901,
|
|
"mean_token_accuracy": 0.15637971609830856,
|
|
"num_tokens": 2864491.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.811354351043701,
|
|
"epoch": 1.297071129707113,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004972147085200116,
|
|
"loss": 5.6969,
|
|
"mean_token_accuracy": 0.1557008296251297,
|
|
"num_tokens": 2873962.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.800867891311645,
|
|
"epoch": 1.301255230125523,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004971638509227611,
|
|
"loss": 5.7301,
|
|
"mean_token_accuracy": 0.15785083174705505,
|
|
"num_tokens": 2883530.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.771515417098999,
|
|
"epoch": 1.3054393305439331,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004971125361480236,
|
|
"loss": 5.7266,
|
|
"mean_token_accuracy": 0.151897095143795,
|
|
"num_tokens": 2892867.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.756432056427002,
|
|
"epoch": 1.3096234309623431,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004970607643013969,
|
|
"loss": 5.6874,
|
|
"mean_token_accuracy": 0.15923002511262893,
|
|
"num_tokens": 2902320.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.777465343475342,
|
|
"epoch": 1.3138075313807531,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000497008535489419,
|
|
"loss": 5.6413,
|
|
"mean_token_accuracy": 0.15751131474971772,
|
|
"num_tokens": 2911390.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.701142072677612,
|
|
"epoch": 1.3179916317991631,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004969558498195684,
|
|
"loss": 5.673,
|
|
"mean_token_accuracy": 0.16534920781850815,
|
|
"num_tokens": 2921295.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.815394020080566,
|
|
"epoch": 1.3221757322175733,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004969027074002638,
|
|
"loss": 5.6874,
|
|
"mean_token_accuracy": 0.15951785892248155,
|
|
"num_tokens": 2931328.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.713910436630249,
|
|
"epoch": 1.3263598326359833,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004968491083408636,
|
|
"loss": 5.6434,
|
|
"mean_token_accuracy": 0.1613614335656166,
|
|
"num_tokens": 2940193.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.683044862747193,
|
|
"epoch": 1.3305439330543933,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004967950527516663,
|
|
"loss": 5.6748,
|
|
"mean_token_accuracy": 0.15876813232898712,
|
|
"num_tokens": 2949759.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.708407974243164,
|
|
"epoch": 1.3347280334728033,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004967405407439092,
|
|
"loss": 5.5959,
|
|
"mean_token_accuracy": 0.1553261622786522,
|
|
"num_tokens": 2959185.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.833631801605224,
|
|
"epoch": 1.3389121338912133,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004966855724297698,
|
|
"loss": 5.7455,
|
|
"mean_token_accuracy": 0.15250789225101472,
|
|
"num_tokens": 2968945.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.736705303192139,
|
|
"epoch": 1.3430962343096233,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004966301479223637,
|
|
"loss": 5.6967,
|
|
"mean_token_accuracy": 0.1556132599711418,
|
|
"num_tokens": 2978395.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.739750719070434,
|
|
"epoch": 1.3472803347280335,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004965742673357456,
|
|
"loss": 5.6131,
|
|
"mean_token_accuracy": 0.16823979020118712,
|
|
"num_tokens": 2987030.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.7740395069122314,
|
|
"epoch": 1.3514644351464435,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004965179307849088,
|
|
"loss": 5.6609,
|
|
"mean_token_accuracy": 0.15987352132797242,
|
|
"num_tokens": 2995736.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.8161561489105225,
|
|
"epoch": 1.3556485355648535,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000496461138385785,
|
|
"loss": 5.7754,
|
|
"mean_token_accuracy": 0.156988662481308,
|
|
"num_tokens": 3005105.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.745142889022827,
|
|
"epoch": 1.3598326359832635,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004964038902552438,
|
|
"loss": 5.6924,
|
|
"mean_token_accuracy": 0.15291614979505538,
|
|
"num_tokens": 3013857.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.775795555114746,
|
|
"epoch": 1.3640167364016738,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004963461865110926,
|
|
"loss": 5.7171,
|
|
"mean_token_accuracy": 0.15699103474617004,
|
|
"num_tokens": 3023972.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.664225387573242,
|
|
"epoch": 1.3682008368200838,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004962880272720764,
|
|
"loss": 5.6885,
|
|
"mean_token_accuracy": 0.16199921816587448,
|
|
"num_tokens": 3032705.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.7547321796417235,
|
|
"epoch": 1.3723849372384938,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004962294126578775,
|
|
"loss": 5.62,
|
|
"mean_token_accuracy": 0.16171107590198516,
|
|
"num_tokens": 3040900.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.74275279045105,
|
|
"epoch": 1.3765690376569037,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004961703427891154,
|
|
"loss": 5.6599,
|
|
"mean_token_accuracy": 0.16257089227437974,
|
|
"num_tokens": 3050477.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.7742781162261965,
|
|
"epoch": 1.3807531380753137,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004961108177873466,
|
|
"loss": 5.7237,
|
|
"mean_token_accuracy": 0.154131980240345,
|
|
"num_tokens": 3059743.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.735147619247437,
|
|
"epoch": 1.3849372384937237,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004960508377750638,
|
|
"loss": 5.5306,
|
|
"mean_token_accuracy": 0.16080360263586044,
|
|
"num_tokens": 3067801.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.600426530838012,
|
|
"epoch": 1.389121338912134,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004959904028756961,
|
|
"loss": 5.7173,
|
|
"mean_token_accuracy": 0.15346099585294723,
|
|
"num_tokens": 3077404.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.864295959472656,
|
|
"epoch": 1.393305439330544,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004959295132136091,
|
|
"loss": 5.7488,
|
|
"mean_token_accuracy": 0.1578315958380699,
|
|
"num_tokens": 3087130.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.866291618347168,
|
|
"epoch": 1.397489539748954,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004958681689141036,
|
|
"loss": 5.7664,
|
|
"mean_token_accuracy": 0.15363284721970558,
|
|
"num_tokens": 3096038.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.603582286834717,
|
|
"epoch": 1.401673640167364,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004958063701034167,
|
|
"loss": 5.5974,
|
|
"mean_token_accuracy": 0.1607895866036415,
|
|
"num_tokens": 3105668.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.711364984512329,
|
|
"epoch": 1.4058577405857742,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00049574411690872,
|
|
"loss": 5.5904,
|
|
"mean_token_accuracy": 0.15727734118700026,
|
|
"num_tokens": 3114020.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.642215585708618,
|
|
"epoch": 1.4100418410041842,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004956814094581208,
|
|
"loss": 5.6182,
|
|
"mean_token_accuracy": 0.16320676058530809,
|
|
"num_tokens": 3124096.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.794946718215942,
|
|
"epoch": 1.4142259414225942,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000495618247880661,
|
|
"loss": 5.6859,
|
|
"mean_token_accuracy": 0.16073822230100632,
|
|
"num_tokens": 3132806.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.671970987319947,
|
|
"epoch": 1.4184100418410042,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004955546323063168,
|
|
"loss": 5.6278,
|
|
"mean_token_accuracy": 0.16975785344839095,
|
|
"num_tokens": 3142001.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.647975635528565,
|
|
"epoch": 1.4225941422594142,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004954905628659989,
|
|
"loss": 5.5844,
|
|
"mean_token_accuracy": 0.1559284970164299,
|
|
"num_tokens": 3151658.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.718328285217285,
|
|
"epoch": 1.4267782426778242,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004954260396915521,
|
|
"loss": 5.6258,
|
|
"mean_token_accuracy": 0.16052106022834778,
|
|
"num_tokens": 3159573.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.544895315170288,
|
|
"epoch": 1.4309623430962344,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004953610629157544,
|
|
"loss": 5.4718,
|
|
"mean_token_accuracy": 0.16683861315250398,
|
|
"num_tokens": 3168312.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.742662715911865,
|
|
"epoch": 1.4351464435146444,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004952956326723177,
|
|
"loss": 5.5968,
|
|
"mean_token_accuracy": 0.16999766528606414,
|
|
"num_tokens": 3176811.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.708902883529663,
|
|
"epoch": 1.4393305439330544,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000495229749095887,
|
|
"loss": 5.685,
|
|
"mean_token_accuracy": 0.15340494439005853,
|
|
"num_tokens": 3186599.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.844608926773072,
|
|
"epoch": 1.4435146443514644,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00049516341232204,
|
|
"loss": 5.7892,
|
|
"mean_token_accuracy": 0.15978376120328902,
|
|
"num_tokens": 3195905.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.724213314056397,
|
|
"epoch": 1.4476987447698746,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000495096622487287,
|
|
"loss": 5.75,
|
|
"mean_token_accuracy": 0.1553723022341728,
|
|
"num_tokens": 3205132.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.88250732421875,
|
|
"epoch": 1.4518828451882846,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000495029379729071,
|
|
"loss": 5.7631,
|
|
"mean_token_accuracy": 0.14813076555728913,
|
|
"num_tokens": 3214362.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.781597042083741,
|
|
"epoch": 1.4560669456066946,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004949616841857666,
|
|
"loss": 5.7411,
|
|
"mean_token_accuracy": 0.15504500269889832,
|
|
"num_tokens": 3224248.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.69829797744751,
|
|
"epoch": 1.4602510460251046,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004948935359966806,
|
|
"loss": 5.61,
|
|
"mean_token_accuracy": 0.1609645113348961,
|
|
"num_tokens": 3233416.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.607308721542358,
|
|
"epoch": 1.4644351464435146,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004948249353020506,
|
|
"loss": 5.5365,
|
|
"mean_token_accuracy": 0.16924193501472473,
|
|
"num_tokens": 3242354.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.732311677932739,
|
|
"epoch": 1.4686192468619246,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004947558822430461,
|
|
"loss": 5.6397,
|
|
"mean_token_accuracy": 0.16035355031490325,
|
|
"num_tokens": 3251026.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.692071390151978,
|
|
"epoch": 1.4728033472803348,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004946863769617672,
|
|
"loss": 5.5944,
|
|
"mean_token_accuracy": 0.16337772607803344,
|
|
"num_tokens": 3259429.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.646625089645386,
|
|
"epoch": 1.4769874476987448,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004946164196012445,
|
|
"loss": 5.6404,
|
|
"mean_token_accuracy": 0.15840144455432892,
|
|
"num_tokens": 3269305.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.740614318847657,
|
|
"epoch": 1.4811715481171548,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004945460103054389,
|
|
"loss": 5.7426,
|
|
"mean_token_accuracy": 0.14881804287433625,
|
|
"num_tokens": 3278846.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.732423925399781,
|
|
"epoch": 1.4853556485355648,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004944751492192416,
|
|
"loss": 5.65,
|
|
"mean_token_accuracy": 0.16647016108036042,
|
|
"num_tokens": 3287200.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.745698308944702,
|
|
"epoch": 1.489539748953975,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004944038364884734,
|
|
"loss": 5.6791,
|
|
"mean_token_accuracy": 0.15847390592098237,
|
|
"num_tokens": 3296651.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.708291864395141,
|
|
"epoch": 1.493723849372385,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004943320722598842,
|
|
"loss": 5.541,
|
|
"mean_token_accuracy": 0.16429150104522705,
|
|
"num_tokens": 3304887.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.684462928771973,
|
|
"epoch": 1.497907949790795,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004942598566811532,
|
|
"loss": 5.5646,
|
|
"mean_token_accuracy": 0.16510989218950273,
|
|
"num_tokens": 3313925.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.6262242794036865,
|
|
"epoch": 1.502092050209205,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004941871899008887,
|
|
"loss": 5.5645,
|
|
"mean_token_accuracy": 0.16308156549930572,
|
|
"num_tokens": 3322943.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.638538360595703,
|
|
"epoch": 1.506276150627615,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000494114072068627,
|
|
"loss": 5.5779,
|
|
"mean_token_accuracy": 0.16292459666728973,
|
|
"num_tokens": 3332742.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.693442869186401,
|
|
"epoch": 1.510460251046025,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004940405033348332,
|
|
"loss": 5.6596,
|
|
"mean_token_accuracy": 0.15975773185491562,
|
|
"num_tokens": 3342228.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.666954040527344,
|
|
"epoch": 1.514644351464435,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004939664838508996,
|
|
"loss": 5.6349,
|
|
"mean_token_accuracy": 0.15982257425785065,
|
|
"num_tokens": 3351854.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.664078140258789,
|
|
"epoch": 1.5188284518828452,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004938920137691465,
|
|
"loss": 5.6191,
|
|
"mean_token_accuracy": 0.16421182453632355,
|
|
"num_tokens": 3360809.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.664334726333618,
|
|
"epoch": 1.5230125523012552,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004938170932428214,
|
|
"loss": 5.6452,
|
|
"mean_token_accuracy": 0.15385862588882446,
|
|
"num_tokens": 3371033.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.738883924484253,
|
|
"epoch": 1.5271966527196654,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004937417224260984,
|
|
"loss": 5.5209,
|
|
"mean_token_accuracy": 0.16514809876680375,
|
|
"num_tokens": 3379943.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.57040958404541,
|
|
"epoch": 1.5313807531380754,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004936659014740791,
|
|
"loss": 5.6499,
|
|
"mean_token_accuracy": 0.1651262253522873,
|
|
"num_tokens": 3389316.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.77739405632019,
|
|
"epoch": 1.5355648535564854,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004935896305427904,
|
|
"loss": 5.6779,
|
|
"mean_token_accuracy": 0.153752800822258,
|
|
"num_tokens": 3398914.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.85550889968872,
|
|
"epoch": 1.5397489539748954,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004935129097891857,
|
|
"loss": 5.6735,
|
|
"mean_token_accuracy": 0.15657827109098435,
|
|
"num_tokens": 3407967.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.7320945262908936,
|
|
"epoch": 1.5439330543933054,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000493435739371144,
|
|
"loss": 5.6904,
|
|
"mean_token_accuracy": 0.15912245512008666,
|
|
"num_tokens": 3416530.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.685615587234497,
|
|
"epoch": 1.5481171548117154,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004933581194474695,
|
|
"loss": 5.5558,
|
|
"mean_token_accuracy": 0.16052745580673217,
|
|
"num_tokens": 3425762.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.632355833053589,
|
|
"epoch": 1.5523012552301254,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004932800501778918,
|
|
"loss": 5.6357,
|
|
"mean_token_accuracy": 0.15861250311136246,
|
|
"num_tokens": 3434799.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.721937561035157,
|
|
"epoch": 1.5564853556485354,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004932015317230646,
|
|
"loss": 5.6407,
|
|
"mean_token_accuracy": 0.1599887117743492,
|
|
"num_tokens": 3442986.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.6735435962677006,
|
|
"epoch": 1.5606694560669456,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004931225642445665,
|
|
"loss": 5.6696,
|
|
"mean_token_accuracy": 0.16079186052083969,
|
|
"num_tokens": 3452914.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.717952060699463,
|
|
"epoch": 1.5648535564853556,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004930431479048998,
|
|
"loss": 5.5173,
|
|
"mean_token_accuracy": 0.17034173756837845,
|
|
"num_tokens": 3462093.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.666427803039551,
|
|
"epoch": 1.5690376569037658,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004929632828674904,
|
|
"loss": 5.7166,
|
|
"mean_token_accuracy": 0.15459873676300048,
|
|
"num_tokens": 3471259.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.682976150512696,
|
|
"epoch": 1.5732217573221758,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004928829692966879,
|
|
"loss": 5.6492,
|
|
"mean_token_accuracy": 0.16574902534484864,
|
|
"num_tokens": 3480071.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.791259860992431,
|
|
"epoch": 1.5774058577405858,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004928022073577646,
|
|
"loss": 5.6762,
|
|
"mean_token_accuracy": 0.16466853469610215,
|
|
"num_tokens": 3489075.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.68497748374939,
|
|
"epoch": 1.5815899581589958,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004927209972169157,
|
|
"loss": 5.4978,
|
|
"mean_token_accuracy": 0.16591176837682725,
|
|
"num_tokens": 3498495.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.634211444854737,
|
|
"epoch": 1.5857740585774058,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004926393390412586,
|
|
"loss": 5.6434,
|
|
"mean_token_accuracy": 0.15969491004943848,
|
|
"num_tokens": 3508590.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.74862232208252,
|
|
"epoch": 1.5899581589958158,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004925572329988325,
|
|
"loss": 5.6574,
|
|
"mean_token_accuracy": 0.15974607914686204,
|
|
"num_tokens": 3518397.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.652823257446289,
|
|
"epoch": 1.5941422594142258,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004924746792585987,
|
|
"loss": 5.6367,
|
|
"mean_token_accuracy": 0.16384051144123077,
|
|
"num_tokens": 3526620.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.743967056274414,
|
|
"epoch": 1.5983263598326358,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004923916779904394,
|
|
"loss": 5.6546,
|
|
"mean_token_accuracy": 0.163092640042305,
|
|
"num_tokens": 3535534.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.571524715423584,
|
|
"epoch": 1.602510460251046,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004923082293651578,
|
|
"loss": 5.6151,
|
|
"mean_token_accuracy": 0.15783242136240005,
|
|
"num_tokens": 3544771.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.698504400253296,
|
|
"epoch": 1.606694560669456,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000492224333554478,
|
|
"loss": 5.5663,
|
|
"mean_token_accuracy": 0.1609413132071495,
|
|
"num_tokens": 3553970.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.621961355209351,
|
|
"epoch": 1.6108786610878663,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004921399907310437,
|
|
"loss": 5.4899,
|
|
"mean_token_accuracy": 0.16530968993902206,
|
|
"num_tokens": 3563116.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.569674491882324,
|
|
"epoch": 1.6150627615062763,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004920552010684191,
|
|
"loss": 5.6301,
|
|
"mean_token_accuracy": 0.16663024872541427,
|
|
"num_tokens": 3571605.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.673672008514404,
|
|
"epoch": 1.6192468619246863,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004919699647410878,
|
|
"loss": 5.7027,
|
|
"mean_token_accuracy": 0.1581038311123848,
|
|
"num_tokens": 3581837.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.702061128616333,
|
|
"epoch": 1.6234309623430963,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004918842819244523,
|
|
"loss": 5.6532,
|
|
"mean_token_accuracy": 0.1579617217183113,
|
|
"num_tokens": 3590853.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.6623131275177006,
|
|
"epoch": 1.6276150627615062,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004917981527948342,
|
|
"loss": 5.5433,
|
|
"mean_token_accuracy": 0.16482285112142564,
|
|
"num_tokens": 3599708.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.614315223693848,
|
|
"epoch": 1.6317991631799162,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004917115775294731,
|
|
"loss": 5.5778,
|
|
"mean_token_accuracy": 0.1669729769229889,
|
|
"num_tokens": 3608634.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.6976964473724365,
|
|
"epoch": 1.6359832635983262,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004916245563065273,
|
|
"loss": 5.6369,
|
|
"mean_token_accuracy": 0.16016416847705842,
|
|
"num_tokens": 3618776.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.666608476638794,
|
|
"epoch": 1.6401673640167362,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004915370893050722,
|
|
"loss": 5.6426,
|
|
"mean_token_accuracy": 0.15682819336652756,
|
|
"num_tokens": 3628595.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.806681823730469,
|
|
"epoch": 1.6443514644351465,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004914491767051012,
|
|
"loss": 5.7436,
|
|
"mean_token_accuracy": 0.14956202432513238,
|
|
"num_tokens": 3638133.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.697069072723389,
|
|
"epoch": 1.6485355648535565,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004913608186875241,
|
|
"loss": 5.5586,
|
|
"mean_token_accuracy": 0.1624540075659752,
|
|
"num_tokens": 3646679.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.659416246414184,
|
|
"epoch": 1.6527196652719667,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004912720154341674,
|
|
"loss": 5.6739,
|
|
"mean_token_accuracy": 0.15715343728661538,
|
|
"num_tokens": 3656131.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.671878480911255,
|
|
"epoch": 1.6569037656903767,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004911827671277739,
|
|
"loss": 5.568,
|
|
"mean_token_accuracy": 0.16296357065439224,
|
|
"num_tokens": 3665187.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.695562028884888,
|
|
"epoch": 1.6610878661087867,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004910930739520025,
|
|
"loss": 5.6291,
|
|
"mean_token_accuracy": 0.16535532623529434,
|
|
"num_tokens": 3674649.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.7589911937713625,
|
|
"epoch": 1.6652719665271967,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004910029360914271,
|
|
"loss": 5.6536,
|
|
"mean_token_accuracy": 0.15396214723587037,
|
|
"num_tokens": 3684680.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.60883321762085,
|
|
"epoch": 1.6694560669456067,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000490912353731537,
|
|
"loss": 5.5797,
|
|
"mean_token_accuracy": 0.16498577892780303,
|
|
"num_tokens": 3693560.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.607213926315308,
|
|
"epoch": 1.6736401673640167,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004908213270587363,
|
|
"loss": 5.5851,
|
|
"mean_token_accuracy": 0.1629691630601883,
|
|
"num_tokens": 3702632.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6736401673640167,
|
|
"eval_entropy": 5.544569926271653,
|
|
"eval_loss": 5.715911388397217,
|
|
"eval_mean_token_accuracy": 0.16552528923335494,
|
|
"eval_num_tokens": 3702632.0,
|
|
"eval_runtime": 2.7368,
|
|
"eval_samples_per_second": 1427.569,
|
|
"eval_steps_per_second": 178.674,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.604301977157593,
|
|
"epoch": 1.6778242677824267,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004907298562603431,
|
|
"loss": 5.4632,
|
|
"mean_token_accuracy": 0.1718360811471939,
|
|
"num_tokens": 3711022.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.5926025390625,
|
|
"epoch": 1.6820083682008367,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004906379415245896,
|
|
"loss": 5.5684,
|
|
"mean_token_accuracy": 0.16862087249755858,
|
|
"num_tokens": 3719778.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.590682411193848,
|
|
"epoch": 1.6861924686192469,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004905455830406215,
|
|
"loss": 5.5672,
|
|
"mean_token_accuracy": 0.16277657300233841,
|
|
"num_tokens": 3728773.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.677611303329468,
|
|
"epoch": 1.6903765690376569,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004904527809984978,
|
|
"loss": 5.6141,
|
|
"mean_token_accuracy": 0.16766730397939683,
|
|
"num_tokens": 3737781.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.7193811416625975,
|
|
"epoch": 1.694560669456067,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004903595355891902,
|
|
"loss": 5.6613,
|
|
"mean_token_accuracy": 0.15416572988033295,
|
|
"num_tokens": 3748010.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.650969839096069,
|
|
"epoch": 1.698744769874477,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004902658470045826,
|
|
"loss": 5.642,
|
|
"mean_token_accuracy": 0.16055549085140228,
|
|
"num_tokens": 3757894.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.753380298614502,
|
|
"epoch": 1.702928870292887,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004901717154374713,
|
|
"loss": 5.7388,
|
|
"mean_token_accuracy": 0.15718897730112075,
|
|
"num_tokens": 3768760.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.690996360778809,
|
|
"epoch": 1.707112970711297,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004900771410815638,
|
|
"loss": 5.5709,
|
|
"mean_token_accuracy": 0.16574515402317047,
|
|
"num_tokens": 3778273.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.5101611614227295,
|
|
"epoch": 1.711297071129707,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004899821241314788,
|
|
"loss": 5.5613,
|
|
"mean_token_accuracy": 0.1653672218322754,
|
|
"num_tokens": 3787012.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.649552869796753,
|
|
"epoch": 1.715481171548117,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000489886664782746,
|
|
"loss": 5.5247,
|
|
"mean_token_accuracy": 0.1680687978863716,
|
|
"num_tokens": 3795144.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.708580875396729,
|
|
"epoch": 1.719665271966527,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004897907632318055,
|
|
"loss": 5.7534,
|
|
"mean_token_accuracy": 0.15822094976902007,
|
|
"num_tokens": 3804534.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.641533422470093,
|
|
"epoch": 1.723849372384937,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004896944196760071,
|
|
"loss": 5.5263,
|
|
"mean_token_accuracy": 0.1683221310377121,
|
|
"num_tokens": 3813757.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.594599962234497,
|
|
"epoch": 1.7280334728033473,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895976343136106,
|
|
"loss": 5.5548,
|
|
"mean_token_accuracy": 0.1677917420864105,
|
|
"num_tokens": 3823157.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.538334083557129,
|
|
"epoch": 1.7322175732217573,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895004073437844,
|
|
"loss": 5.4542,
|
|
"mean_token_accuracy": 0.16612266898155212,
|
|
"num_tokens": 3831661.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.541912317276001,
|
|
"epoch": 1.7364016736401675,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004894027389666061,
|
|
"loss": 5.4498,
|
|
"mean_token_accuracy": 0.16812364608049393,
|
|
"num_tokens": 3840591.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.639562273025513,
|
|
"epoch": 1.7405857740585775,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004893046293830616,
|
|
"loss": 5.5785,
|
|
"mean_token_accuracy": 0.16422712057828903,
|
|
"num_tokens": 3850094.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.621643018722534,
|
|
"epoch": 1.7447698744769875,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004892060787950447,
|
|
"loss": 5.626,
|
|
"mean_token_accuracy": 0.16461298018693923,
|
|
"num_tokens": 3858973.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.559000730514526,
|
|
"epoch": 1.7489539748953975,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004891070874053564,
|
|
"loss": 5.4022,
|
|
"mean_token_accuracy": 0.1763455167412758,
|
|
"num_tokens": 3868315.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.538459062576294,
|
|
"epoch": 1.7531380753138075,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004890076554177051,
|
|
"loss": 5.5615,
|
|
"mean_token_accuracy": 0.1631178990006447,
|
|
"num_tokens": 3877605.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.700150108337402,
|
|
"epoch": 1.7573221757322175,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004889077830367061,
|
|
"loss": 5.5782,
|
|
"mean_token_accuracy": 0.16976662129163742,
|
|
"num_tokens": 3886998.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.5663464069366455,
|
|
"epoch": 1.7615062761506275,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004888074704678806,
|
|
"loss": 5.5283,
|
|
"mean_token_accuracy": 0.17066608518362045,
|
|
"num_tokens": 3896018.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.621518468856811,
|
|
"epoch": 1.7656903765690377,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004887067179176558,
|
|
"loss": 5.5343,
|
|
"mean_token_accuracy": 0.1670089140534401,
|
|
"num_tokens": 3904957.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.587416505813598,
|
|
"epoch": 1.7698744769874477,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004886055255933639,
|
|
"loss": 5.6542,
|
|
"mean_token_accuracy": 0.16026381626725197,
|
|
"num_tokens": 3913427.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.667502689361572,
|
|
"epoch": 1.7740585774058577,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004885038937032429,
|
|
"loss": 5.5657,
|
|
"mean_token_accuracy": 0.16527705043554305,
|
|
"num_tokens": 3923134.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.622828578948974,
|
|
"epoch": 1.778242677824268,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004884018224564345,
|
|
"loss": 5.4785,
|
|
"mean_token_accuracy": 0.16797400861978531,
|
|
"num_tokens": 3931867.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.58387508392334,
|
|
"epoch": 1.782426778242678,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004882993120629852,
|
|
"loss": 5.5638,
|
|
"mean_token_accuracy": 0.16899107545614242,
|
|
"num_tokens": 3941057.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.744916820526123,
|
|
"epoch": 1.786610878661088,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004881963627338449,
|
|
"loss": 5.6846,
|
|
"mean_token_accuracy": 0.15740710943937303,
|
|
"num_tokens": 3950511.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.63847131729126,
|
|
"epoch": 1.790794979079498,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048809297468086654,
|
|
"loss": 5.581,
|
|
"mean_token_accuracy": 0.1745120495557785,
|
|
"num_tokens": 3959769.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.657891464233399,
|
|
"epoch": 1.794979079497908,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004879891481168063,
|
|
"loss": 5.5191,
|
|
"mean_token_accuracy": 0.1700441509485245,
|
|
"num_tokens": 3968462.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.530794906616211,
|
|
"epoch": 1.799163179916318,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048788488325532235,
|
|
"loss": 5.5703,
|
|
"mean_token_accuracy": 0.16803695857524872,
|
|
"num_tokens": 3977464.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.5986074924469,
|
|
"epoch": 1.803347280334728,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004877801803109753,
|
|
"loss": 5.4767,
|
|
"mean_token_accuracy": 0.1752568155527115,
|
|
"num_tokens": 3986303.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.683440685272217,
|
|
"epoch": 1.8075313807531381,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048767503949922683,
|
|
"loss": 5.5264,
|
|
"mean_token_accuracy": 0.16948044002056123,
|
|
"num_tokens": 3994970.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.605940961837769,
|
|
"epoch": 1.8117154811715481,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004875694610364398,
|
|
"loss": 5.6154,
|
|
"mean_token_accuracy": 0.1611637741327286,
|
|
"num_tokens": 4004679.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.728769540786743,
|
|
"epoch": 1.8158995815899581,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048746344513987775,
|
|
"loss": 5.6381,
|
|
"mean_token_accuracy": 0.1625831186771393,
|
|
"num_tokens": 4013516.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.613612127304077,
|
|
"epoch": 1.8200836820083683,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004873569920277043,
|
|
"loss": 5.5096,
|
|
"mean_token_accuracy": 0.17253974229097366,
|
|
"num_tokens": 4022335.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.533032703399658,
|
|
"epoch": 1.8242677824267783,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048725010191898296,
|
|
"loss": 5.4899,
|
|
"mean_token_accuracy": 0.17203625291585922,
|
|
"num_tokens": 4031295.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.687892150878906,
|
|
"epoch": 1.8284518828451883,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048714277503367627,
|
|
"loss": 5.7066,
|
|
"mean_token_accuracy": 0.1616518810391426,
|
|
"num_tokens": 4040378.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.640391111373901,
|
|
"epoch": 1.8326359832635983,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048703501159264576,
|
|
"loss": 5.553,
|
|
"mean_token_accuracy": 0.15976608246564866,
|
|
"num_tokens": 4050217.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.5045853614807125,
|
|
"epoch": 1.8368200836820083,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004869268118176512,
|
|
"loss": 5.4167,
|
|
"mean_token_accuracy": 0.16942883282899857,
|
|
"num_tokens": 4059088.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.6252336502075195,
|
|
"epoch": 1.8410041841004183,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048681817593135046,
|
|
"loss": 5.5366,
|
|
"mean_token_accuracy": 0.17682300955057145,
|
|
"num_tokens": 4067609.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.567916822433472,
|
|
"epoch": 1.8451882845188283,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048670910415729855,
|
|
"loss": 5.5195,
|
|
"mean_token_accuracy": 0.17112305611371995,
|
|
"num_tokens": 4076763.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.655111455917359,
|
|
"epoch": 1.8493723849372385,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048659959671994786,
|
|
"loss": 5.6785,
|
|
"mean_token_accuracy": 0.16254908740520477,
|
|
"num_tokens": 4086552.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.494708251953125,
|
|
"epoch": 1.8535564853556485,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048648965384464693,
|
|
"loss": 5.4333,
|
|
"mean_token_accuracy": 0.17247406542301177,
|
|
"num_tokens": 4096200.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.529922199249268,
|
|
"epoch": 1.8577405857740585,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048637927575764057,
|
|
"loss": 5.4908,
|
|
"mean_token_accuracy": 0.16739632934331894,
|
|
"num_tokens": 4105080.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.598203134536743,
|
|
"epoch": 1.8619246861924688,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048626846268606913,
|
|
"loss": 5.4519,
|
|
"mean_token_accuracy": 0.17144161313772202,
|
|
"num_tokens": 4113972.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.681439542770386,
|
|
"epoch": 1.8661087866108788,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048615721485796803,
|
|
"loss": 5.6933,
|
|
"mean_token_accuracy": 0.15622689872980117,
|
|
"num_tokens": 4123615.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.63415675163269,
|
|
"epoch": 1.8702928870292888,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004860455325022676,
|
|
"loss": 5.451,
|
|
"mean_token_accuracy": 0.1640198454260826,
|
|
"num_tokens": 4132128.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.658318138122558,
|
|
"epoch": 1.8744769874476988,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000485933415848792,
|
|
"loss": 5.6749,
|
|
"mean_token_accuracy": 0.162701378762722,
|
|
"num_tokens": 4141575.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.6866742134094235,
|
|
"epoch": 1.8786610878661087,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004858208651282593,
|
|
"loss": 5.505,
|
|
"mean_token_accuracy": 0.16838331073522567,
|
|
"num_tokens": 4150372.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.5462789058685305,
|
|
"epoch": 1.8828451882845187,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000485707880572281,
|
|
"loss": 5.6084,
|
|
"mean_token_accuracy": 0.16846281588077544,
|
|
"num_tokens": 4159007.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.625886917114258,
|
|
"epoch": 1.8870292887029287,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004855944624133608,
|
|
"loss": 5.6619,
|
|
"mean_token_accuracy": 0.16254033297300338,
|
|
"num_tokens": 4168646.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.565406274795532,
|
|
"epoch": 1.891213389121339,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048548061088489523,
|
|
"loss": 5.5563,
|
|
"mean_token_accuracy": 0.1624855801463127,
|
|
"num_tokens": 4177979.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.653033447265625,
|
|
"epoch": 1.895397489539749,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048536632622117246,
|
|
"loss": 5.4694,
|
|
"mean_token_accuracy": 0.17232209146022798,
|
|
"num_tokens": 4186781.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.540466260910034,
|
|
"epoch": 1.899581589958159,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048525160865737186,
|
|
"loss": 5.5396,
|
|
"mean_token_accuracy": 0.1641465112566948,
|
|
"num_tokens": 4195661.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.586657428741455,
|
|
"epoch": 1.9037656903765692,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048513645842956384,
|
|
"loss": 5.4563,
|
|
"mean_token_accuracy": 0.1706296131014824,
|
|
"num_tokens": 4203902.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.555630874633789,
|
|
"epoch": 1.9079497907949792,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048502087577470895,
|
|
"loss": 5.5015,
|
|
"mean_token_accuracy": 0.16890660524368287,
|
|
"num_tokens": 4213663.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.649169540405273,
|
|
"epoch": 1.9121338912133892,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048490486093065774,
|
|
"loss": 5.5524,
|
|
"mean_token_accuracy": 0.16829514801502227,
|
|
"num_tokens": 4222152.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.625352954864502,
|
|
"epoch": 1.9163179916317992,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048478841413615017,
|
|
"loss": 5.572,
|
|
"mean_token_accuracy": 0.17131819427013398,
|
|
"num_tokens": 4232347.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.624868249893188,
|
|
"epoch": 1.9205020920502092,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048467153563081487,
|
|
"loss": 5.6332,
|
|
"mean_token_accuracy": 0.16910638511180878,
|
|
"num_tokens": 4240607.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.633149242401123,
|
|
"epoch": 1.9246861924686192,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004845542256551692,
|
|
"loss": 5.6394,
|
|
"mean_token_accuracy": 0.16387657672166825,
|
|
"num_tokens": 4251131.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.637137794494629,
|
|
"epoch": 1.9288702928870292,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004844364844506181,
|
|
"loss": 5.5719,
|
|
"mean_token_accuracy": 0.16235824227333068,
|
|
"num_tokens": 4259955.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.679235124588013,
|
|
"epoch": 1.9330543933054394,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004843183122594541,
|
|
"loss": 5.5448,
|
|
"mean_token_accuracy": 0.1709941640496254,
|
|
"num_tokens": 4268686.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.612339782714844,
|
|
"epoch": 1.9372384937238494,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004841997093248566,
|
|
"loss": 5.5195,
|
|
"mean_token_accuracy": 0.17368945777416228,
|
|
"num_tokens": 4277377.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.483291482925415,
|
|
"epoch": 1.9414225941422594,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004840806758908913,
|
|
"loss": 5.5151,
|
|
"mean_token_accuracy": 0.16756559759378434,
|
|
"num_tokens": 4287037.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.706357431411743,
|
|
"epoch": 1.9456066945606696,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048396121220250983,
|
|
"loss": 5.6606,
|
|
"mean_token_accuracy": 0.1578308656811714,
|
|
"num_tokens": 4296923.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.4903162002563475,
|
|
"epoch": 1.9497907949790796,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004838413185055494,
|
|
"loss": 5.4163,
|
|
"mean_token_accuracy": 0.17858784198760985,
|
|
"num_tokens": 4305873.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.615921258926392,
|
|
"epoch": 1.9539748953974896,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004837209950467318,
|
|
"loss": 5.5754,
|
|
"mean_token_accuracy": 0.16469462513923644,
|
|
"num_tokens": 4314912.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.605984258651733,
|
|
"epoch": 1.9581589958158996,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004836002420736635,
|
|
"loss": 5.5774,
|
|
"mean_token_accuracy": 0.16062707901000978,
|
|
"num_tokens": 4324606.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.546719932556153,
|
|
"epoch": 1.9623430962343096,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004834790598348347,
|
|
"loss": 5.4396,
|
|
"mean_token_accuracy": 0.16764551401138306,
|
|
"num_tokens": 4333076.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.71311993598938,
|
|
"epoch": 1.9665271966527196,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004833574485796188,
|
|
"loss": 5.5834,
|
|
"mean_token_accuracy": 0.1685716599225998,
|
|
"num_tokens": 4341893.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.57413420677185,
|
|
"epoch": 1.9707112970711296,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048323540855827244,
|
|
"loss": 5.5458,
|
|
"mean_token_accuracy": 0.16568866670131682,
|
|
"num_tokens": 4352303.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.654368257522583,
|
|
"epoch": 1.9748953974895398,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004831129400219341,
|
|
"loss": 5.5109,
|
|
"mean_token_accuracy": 0.16720473915338516,
|
|
"num_tokens": 4361166.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.690392589569091,
|
|
"epoch": 1.9790794979079498,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048299004322262457,
|
|
"loss": 5.5567,
|
|
"mean_token_accuracy": 0.16406455636024475,
|
|
"num_tokens": 4369999.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.539315795898437,
|
|
"epoch": 1.9832635983263598,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048286671841324565,
|
|
"loss": 5.5161,
|
|
"mean_token_accuracy": 0.17316748797893525,
|
|
"num_tokens": 4379373.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.598080635070801,
|
|
"epoch": 1.98744769874477,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004827429658475799,
|
|
"loss": 5.5158,
|
|
"mean_token_accuracy": 0.17036739885807037,
|
|
"num_tokens": 4389035.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.6628663539886475,
|
|
"epoch": 1.99163179916318,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048261878578029026,
|
|
"loss": 5.6745,
|
|
"mean_token_accuracy": 0.16741672456264495,
|
|
"num_tokens": 4399366.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.643234777450561,
|
|
"epoch": 1.99581589958159,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004824941784669193,
|
|
"loss": 5.5101,
|
|
"mean_token_accuracy": 0.16699415892362596,
|
|
"num_tokens": 4408183.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.595900821685791,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004823691441638889,
|
|
"loss": 5.6154,
|
|
"mean_token_accuracy": 0.15458238124847412,
|
|
"num_tokens": 4418644.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.52336573600769,
|
|
"epoch": 2.00418410041841,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004822436831284995,
|
|
"loss": 5.2173,
|
|
"mean_token_accuracy": 0.18154313266277314,
|
|
"num_tokens": 4427828.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.49915246963501,
|
|
"epoch": 2.00836820083682,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004821177956189297,
|
|
"loss": 5.1385,
|
|
"mean_token_accuracy": 0.18106852173805238,
|
|
"num_tokens": 4436194.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.449331378936767,
|
|
"epoch": 2.01255230125523,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004819914818942359,
|
|
"loss": 5.2112,
|
|
"mean_token_accuracy": 0.1826639547944069,
|
|
"num_tokens": 4444468.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.534541130065918,
|
|
"epoch": 2.01673640167364,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048186474221435127,
|
|
"loss": 5.1351,
|
|
"mean_token_accuracy": 0.18929099142551423,
|
|
"num_tokens": 4452760.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.447423124313355,
|
|
"epoch": 2.02092050209205,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048173757684008583,
|
|
"loss": 5.2009,
|
|
"mean_token_accuracy": 0.18149912357330322,
|
|
"num_tokens": 4462154.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.4820966720581055,
|
|
"epoch": 2.0251046025104604,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004816099860331255,
|
|
"loss": 5.2825,
|
|
"mean_token_accuracy": 0.17497736364603042,
|
|
"num_tokens": 4471987.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.497165203094482,
|
|
"epoch": 2.0292887029288704,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048148197005603145,
|
|
"loss": 5.2191,
|
|
"mean_token_accuracy": 0.17726924568414687,
|
|
"num_tokens": 4481645.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.368935203552246,
|
|
"epoch": 2.0334728033472804,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048135352917224005,
|
|
"loss": 5.1751,
|
|
"mean_token_accuracy": 0.1868342339992523,
|
|
"num_tokens": 4491166.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.578146934509277,
|
|
"epoch": 2.0376569037656904,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048122466364606214,
|
|
"loss": 5.1973,
|
|
"mean_token_accuracy": 0.18251455277204515,
|
|
"num_tokens": 4500159.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.389235353469848,
|
|
"epoch": 2.0418410041841004,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048109537374268216,
|
|
"loss": 5.1464,
|
|
"mean_token_accuracy": 0.18923695534467697,
|
|
"num_tokens": 4509737.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.387840270996094,
|
|
"epoch": 2.0460251046025104,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00048096565972815793,
|
|
"loss": 5.0715,
|
|
"mean_token_accuracy": 0.18712350428104402,
|
|
"num_tokens": 4518379.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.4111696720123295,
|
|
"epoch": 2.0502092050209204,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004808355218694199,
|
|
"loss": 5.2007,
|
|
"mean_token_accuracy": 0.1853238508105278,
|
|
"num_tokens": 4527885.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.4629090309143065,
|
|
"epoch": 2.0543933054393304,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048070496043427116,
|
|
"loss": 5.2467,
|
|
"mean_token_accuracy": 0.17788249850273133,
|
|
"num_tokens": 4537894.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.447832202911377,
|
|
"epoch": 2.0585774058577404,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048057397569138596,
|
|
"loss": 5.1357,
|
|
"mean_token_accuracy": 0.19053970128297806,
|
|
"num_tokens": 4546462.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.438874244689941,
|
|
"epoch": 2.062761506276151,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004804425679103099,
|
|
"loss": 5.1718,
|
|
"mean_token_accuracy": 0.18075911104679107,
|
|
"num_tokens": 4555970.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.429273414611816,
|
|
"epoch": 2.066945606694561,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004803107373614589,
|
|
"loss": 5.2205,
|
|
"mean_token_accuracy": 0.1768643453717232,
|
|
"num_tokens": 4565799.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.524178123474121,
|
|
"epoch": 2.071129707112971,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004801784843161194,
|
|
"loss": 5.251,
|
|
"mean_token_accuracy": 0.17221496254205704,
|
|
"num_tokens": 4575571.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.416622257232666,
|
|
"epoch": 2.075313807531381,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004800458090464465,
|
|
"loss": 5.1677,
|
|
"mean_token_accuracy": 0.18350304514169694,
|
|
"num_tokens": 4584263.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.466772651672363,
|
|
"epoch": 2.079497907949791,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047991271182546476,
|
|
"loss": 5.1469,
|
|
"mean_token_accuracy": 0.18672786206007003,
|
|
"num_tokens": 4592979.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.362415981292725,
|
|
"epoch": 2.083682008368201,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047977919292706696,
|
|
"loss": 5.1671,
|
|
"mean_token_accuracy": 0.18652344197034837,
|
|
"num_tokens": 4602042.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.459194040298462,
|
|
"epoch": 2.087866108786611,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004796452526260134,
|
|
"loss": 5.2335,
|
|
"mean_token_accuracy": 0.18386663645505905,
|
|
"num_tokens": 4611197.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.290938758850098,
|
|
"epoch": 2.092050209205021,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047951089119793164,
|
|
"loss": 5.1656,
|
|
"mean_token_accuracy": 0.18329917788505554,
|
|
"num_tokens": 4621612.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.092050209205021,
|
|
"eval_entropy": 5.208794407561757,
|
|
"eval_loss": 5.624749183654785,
|
|
"eval_mean_token_accuracy": 0.17307274384854518,
|
|
"eval_num_tokens": 4621612.0,
|
|
"eval_runtime": 2.7319,
|
|
"eval_samples_per_second": 1430.146,
|
|
"eval_steps_per_second": 178.997,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.342732238769531,
|
|
"epoch": 2.096234309623431,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000479376108919316,
|
|
"loss": 5.0527,
|
|
"mean_token_accuracy": 0.19201176315546037,
|
|
"num_tokens": 4630198.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.436562967300415,
|
|
"epoch": 2.100418410041841,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004792409060675267,
|
|
"loss": 5.1714,
|
|
"mean_token_accuracy": 0.1784696400165558,
|
|
"num_tokens": 4638729.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.353607606887818,
|
|
"epoch": 2.104602510460251,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004791052829207895,
|
|
"loss": 5.2745,
|
|
"mean_token_accuracy": 0.18427398651838303,
|
|
"num_tokens": 4648759.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.442644739151001,
|
|
"epoch": 2.1087866108786613,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000478969239758195,
|
|
"loss": 5.2189,
|
|
"mean_token_accuracy": 0.18320283889770508,
|
|
"num_tokens": 4657744.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.3526036739349365,
|
|
"epoch": 2.1129707112970713,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047883277685969815,
|
|
"loss": 5.1865,
|
|
"mean_token_accuracy": 0.18448097854852677,
|
|
"num_tokens": 4666154.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.471697568893433,
|
|
"epoch": 2.1171548117154813,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047869589450611773,
|
|
"loss": 5.2199,
|
|
"mean_token_accuracy": 0.17987029403448104,
|
|
"num_tokens": 4675677.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.375281524658203,
|
|
"epoch": 2.1213389121338913,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004785585929791355,
|
|
"loss": 5.2017,
|
|
"mean_token_accuracy": 0.18070452511310578,
|
|
"num_tokens": 4684733.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.405807685852051,
|
|
"epoch": 2.1255230125523012,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000478420872561296,
|
|
"loss": 5.1302,
|
|
"mean_token_accuracy": 0.18370863795280457,
|
|
"num_tokens": 4694155.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.34947624206543,
|
|
"epoch": 2.1297071129707112,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047828273353600577,
|
|
"loss": 5.1689,
|
|
"mean_token_accuracy": 0.19383873641490937,
|
|
"num_tokens": 4703286.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.456436491012573,
|
|
"epoch": 2.1338912133891212,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047814417618753256,
|
|
"loss": 5.1856,
|
|
"mean_token_accuracy": 0.17915833443403245,
|
|
"num_tokens": 4711666.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.34681134223938,
|
|
"epoch": 2.1380753138075312,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047800520080100526,
|
|
"loss": 5.1984,
|
|
"mean_token_accuracy": 0.17571722120046615,
|
|
"num_tokens": 4721644.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.424099111557007,
|
|
"epoch": 2.1422594142259412,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047786580766241265,
|
|
"loss": 5.1626,
|
|
"mean_token_accuracy": 0.19055590331554412,
|
|
"num_tokens": 4730885.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.332646608352661,
|
|
"epoch": 2.1464435146443517,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004777259970586036,
|
|
"loss": 5.1425,
|
|
"mean_token_accuracy": 0.18602774739265443,
|
|
"num_tokens": 4740871.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.349590539932251,
|
|
"epoch": 2.1506276150627617,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004775857692772859,
|
|
"loss": 5.1157,
|
|
"mean_token_accuracy": 0.18249449878931046,
|
|
"num_tokens": 4750603.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.368098211288452,
|
|
"epoch": 2.1548117154811717,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004774451246070257,
|
|
"loss": 5.2329,
|
|
"mean_token_accuracy": 0.18403066843748092,
|
|
"num_tokens": 4760681.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.442989015579224,
|
|
"epoch": 2.1589958158995817,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004773040633372471,
|
|
"loss": 5.2284,
|
|
"mean_token_accuracy": 0.18556591123342514,
|
|
"num_tokens": 4770237.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.362786769866943,
|
|
"epoch": 2.1631799163179917,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047716258575823147,
|
|
"loss": 5.2329,
|
|
"mean_token_accuracy": 0.19020240604877472,
|
|
"num_tokens": 4779022.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.4511971950531,
|
|
"epoch": 2.1673640167364017,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000477020692161117,
|
|
"loss": 5.2638,
|
|
"mean_token_accuracy": 0.18160027861595154,
|
|
"num_tokens": 4788644.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.375961828231811,
|
|
"epoch": 2.1715481171548117,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004768783828378978,
|
|
"loss": 5.2574,
|
|
"mean_token_accuracy": 0.17766260355710983,
|
|
"num_tokens": 4798628.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.573528003692627,
|
|
"epoch": 2.1757322175732217,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004767356580814237,
|
|
"loss": 5.3659,
|
|
"mean_token_accuracy": 0.17849716544151306,
|
|
"num_tokens": 4807727.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.4480596542358395,
|
|
"epoch": 2.1799163179916317,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004765925181853993,
|
|
"loss": 5.2467,
|
|
"mean_token_accuracy": 0.17864772230386733,
|
|
"num_tokens": 4817896.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.428419685363769,
|
|
"epoch": 2.1841004184100417,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004764489634443834,
|
|
"loss": 5.2058,
|
|
"mean_token_accuracy": 0.18008076697587966,
|
|
"num_tokens": 4827135.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.491842937469483,
|
|
"epoch": 2.1882845188284517,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047630499415378863,
|
|
"loss": 5.2869,
|
|
"mean_token_accuracy": 0.17407437562942504,
|
|
"num_tokens": 4837554.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.330071115493775,
|
|
"epoch": 2.192468619246862,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004761606106098806,
|
|
"loss": 5.1608,
|
|
"mean_token_accuracy": 0.1822814255952835,
|
|
"num_tokens": 4846739.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.404991960525512,
|
|
"epoch": 2.196652719665272,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047601581310977747,
|
|
"loss": 5.1916,
|
|
"mean_token_accuracy": 0.1816651403903961,
|
|
"num_tokens": 4855707.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.412823104858399,
|
|
"epoch": 2.200836820083682,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004758706019514493,
|
|
"loss": 5.207,
|
|
"mean_token_accuracy": 0.18123696148395538,
|
|
"num_tokens": 4865098.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.448468255996704,
|
|
"epoch": 2.205020920502092,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004757249774337172,
|
|
"loss": 5.2198,
|
|
"mean_token_accuracy": 0.1823917880654335,
|
|
"num_tokens": 4874426.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.36552324295044,
|
|
"epoch": 2.209205020920502,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047557893985625293,
|
|
"loss": 5.1819,
|
|
"mean_token_accuracy": 0.18760903626680375,
|
|
"num_tokens": 4883240.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.359978675842285,
|
|
"epoch": 2.213389121338912,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004754324895195786,
|
|
"loss": 5.182,
|
|
"mean_token_accuracy": 0.1838080033659935,
|
|
"num_tokens": 4891989.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.422989845275879,
|
|
"epoch": 2.217573221757322,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004752856267250652,
|
|
"loss": 5.2809,
|
|
"mean_token_accuracy": 0.1831986978650093,
|
|
"num_tokens": 4901546.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.334002828598022,
|
|
"epoch": 2.221757322175732,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047513835177493286,
|
|
"loss": 5.1024,
|
|
"mean_token_accuracy": 0.19226252734661103,
|
|
"num_tokens": 4910222.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.4006264209747314,
|
|
"epoch": 2.225941422594142,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004749906649722498,
|
|
"loss": 5.3129,
|
|
"mean_token_accuracy": 0.17310767769813537,
|
|
"num_tokens": 4920184.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.370958471298218,
|
|
"epoch": 2.2301255230125525,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004748425666209316,
|
|
"loss": 5.2172,
|
|
"mean_token_accuracy": 0.18402200639247895,
|
|
"num_tokens": 4930014.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.380784320831299,
|
|
"epoch": 2.2343096234309625,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004746940570257409,
|
|
"loss": 5.146,
|
|
"mean_token_accuracy": 0.18692640513181685,
|
|
"num_tokens": 4939556.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.322055530548096,
|
|
"epoch": 2.2384937238493725,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004745451364922867,
|
|
"loss": 5.2095,
|
|
"mean_token_accuracy": 0.181628580391407,
|
|
"num_tokens": 4949066.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.443222427368164,
|
|
"epoch": 2.2426778242677825,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004743958053270233,
|
|
"loss": 5.2076,
|
|
"mean_token_accuracy": 0.18355339765548706,
|
|
"num_tokens": 4957881.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.419846630096435,
|
|
"epoch": 2.2468619246861925,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047424606383725037,
|
|
"loss": 5.245,
|
|
"mean_token_accuracy": 0.18528629839420319,
|
|
"num_tokens": 4967105.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.280267333984375,
|
|
"epoch": 2.2510460251046025,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004740959123311118,
|
|
"loss": 5.1591,
|
|
"mean_token_accuracy": 0.1876959204673767,
|
|
"num_tokens": 4977676.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.382284307479859,
|
|
"epoch": 2.2552301255230125,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047394535111759526,
|
|
"loss": 5.173,
|
|
"mean_token_accuracy": 0.1855098471045494,
|
|
"num_tokens": 4986163.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.391265153884888,
|
|
"epoch": 2.2594142259414225,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004737943805065314,
|
|
"loss": 5.1826,
|
|
"mean_token_accuracy": 0.17979367673397065,
|
|
"num_tokens": 4995179.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.460883665084839,
|
|
"epoch": 2.2635983263598325,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047364300080859363,
|
|
"loss": 5.2715,
|
|
"mean_token_accuracy": 0.17535117119550706,
|
|
"num_tokens": 5003856.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.416038179397583,
|
|
"epoch": 2.2677824267782425,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047349121233529695,
|
|
"loss": 5.2948,
|
|
"mean_token_accuracy": 0.1777150511741638,
|
|
"num_tokens": 5013506.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.383950185775757,
|
|
"epoch": 2.2719665271966525,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004733390153989976,
|
|
"loss": 5.2403,
|
|
"mean_token_accuracy": 0.1848796382546425,
|
|
"num_tokens": 5022159.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.3696952819824215,
|
|
"epoch": 2.276150627615063,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047318641031289245,
|
|
"loss": 5.1792,
|
|
"mean_token_accuracy": 0.18405805081129073,
|
|
"num_tokens": 5031869.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.280503368377685,
|
|
"epoch": 2.280334728033473,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004730333973910183,
|
|
"loss": 5.0345,
|
|
"mean_token_accuracy": 0.19599688649177552,
|
|
"num_tokens": 5039904.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.422358751296997,
|
|
"epoch": 2.284518828451883,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004728799769482511,
|
|
"loss": 5.3412,
|
|
"mean_token_accuracy": 0.1808583602309227,
|
|
"num_tokens": 5049144.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.349178791046143,
|
|
"epoch": 2.288702928870293,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047272614930030533,
|
|
"loss": 5.1023,
|
|
"mean_token_accuracy": 0.18753841668367385,
|
|
"num_tokens": 5058302.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.3154381275177,
|
|
"epoch": 2.292887029288703,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047257191476373375,
|
|
"loss": 5.1363,
|
|
"mean_token_accuracy": 0.18648675978183746,
|
|
"num_tokens": 5067443.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.345571088790893,
|
|
"epoch": 2.297071129707113,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047241727365592603,
|
|
"loss": 5.2723,
|
|
"mean_token_accuracy": 0.17538231164216994,
|
|
"num_tokens": 5076415.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.421261310577393,
|
|
"epoch": 2.301255230125523,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004722622262951089,
|
|
"loss": 5.2224,
|
|
"mean_token_accuracy": 0.1793004974722862,
|
|
"num_tokens": 5085882.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.438392972946167,
|
|
"epoch": 2.305439330543933,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004721067730003448,
|
|
"loss": 5.1986,
|
|
"mean_token_accuracy": 0.1918323591351509,
|
|
"num_tokens": 5094685.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.354588603973388,
|
|
"epoch": 2.309623430962343,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004719509140915317,
|
|
"loss": 5.2114,
|
|
"mean_token_accuracy": 0.18272558599710464,
|
|
"num_tokens": 5103885.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.385832977294922,
|
|
"epoch": 2.3138075313807533,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047179464988940205,
|
|
"loss": 5.1342,
|
|
"mean_token_accuracy": 0.19514969885349273,
|
|
"num_tokens": 5112164.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.255523633956909,
|
|
"epoch": 2.3179916317991633,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047163798071552253,
|
|
"loss": 5.1038,
|
|
"mean_token_accuracy": 0.19346790313720702,
|
|
"num_tokens": 5121913.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.389960050582886,
|
|
"epoch": 2.3221757322175733,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004714809068922931,
|
|
"loss": 5.1109,
|
|
"mean_token_accuracy": 0.19163762032985687,
|
|
"num_tokens": 5130447.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.290537691116333,
|
|
"epoch": 2.3263598326359833,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004713234287429465,
|
|
"loss": 5.237,
|
|
"mean_token_accuracy": 0.18625620007514954,
|
|
"num_tokens": 5139863.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.301087760925293,
|
|
"epoch": 2.3305439330543933,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004711655465915473,
|
|
"loss": 5.2041,
|
|
"mean_token_accuracy": 0.1895967036485672,
|
|
"num_tokens": 5149945.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.448004913330078,
|
|
"epoch": 2.3347280334728033,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004710072607629917,
|
|
"loss": 5.1469,
|
|
"mean_token_accuracy": 0.18175724148750305,
|
|
"num_tokens": 5160293.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.332290697097778,
|
|
"epoch": 2.3389121338912133,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047084857158300646,
|
|
"loss": 5.1618,
|
|
"mean_token_accuracy": 0.18061374127864838,
|
|
"num_tokens": 5169896.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.3752429485321045,
|
|
"epoch": 2.3430962343096233,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047068947937814844,
|
|
"loss": 5.2038,
|
|
"mean_token_accuracy": 0.18880946189165115,
|
|
"num_tokens": 5178898.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.346495723724365,
|
|
"epoch": 2.3472803347280333,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047052998447580374,
|
|
"loss": 5.278,
|
|
"mean_token_accuracy": 0.1804632231593132,
|
|
"num_tokens": 5188733.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.341215991973877,
|
|
"epoch": 2.3514644351464433,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047037008720418724,
|
|
"loss": 5.1752,
|
|
"mean_token_accuracy": 0.18327055722475052,
|
|
"num_tokens": 5198117.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.441304492950439,
|
|
"epoch": 2.3556485355648533,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004702097878923419,
|
|
"loss": 5.2315,
|
|
"mean_token_accuracy": 0.18148796558380126,
|
|
"num_tokens": 5206791.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.28036470413208,
|
|
"epoch": 2.3598326359832638,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004700490868701377,
|
|
"loss": 5.1199,
|
|
"mean_token_accuracy": 0.19287294000387192,
|
|
"num_tokens": 5216427.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.296500301361084,
|
|
"epoch": 2.3640167364016738,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004698879844682718,
|
|
"loss": 5.1009,
|
|
"mean_token_accuracy": 0.19283896833658218,
|
|
"num_tokens": 5225983.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.286601161956787,
|
|
"epoch": 2.3682008368200838,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046972648101826695,
|
|
"loss": 5.2163,
|
|
"mean_token_accuracy": 0.18582138568162918,
|
|
"num_tokens": 5234547.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.49201192855835,
|
|
"epoch": 2.3723849372384938,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004695645768524712,
|
|
"loss": 5.2535,
|
|
"mean_token_accuracy": 0.1892945185303688,
|
|
"num_tokens": 5243281.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.31707010269165,
|
|
"epoch": 2.3765690376569037,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004694022723040573,
|
|
"loss": 5.1356,
|
|
"mean_token_accuracy": 0.18640855848789215,
|
|
"num_tokens": 5252318.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.3913795948028564,
|
|
"epoch": 2.3807531380753137,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046923956770702206,
|
|
"loss": 5.2385,
|
|
"mean_token_accuracy": 0.19023721665143967,
|
|
"num_tokens": 5261781.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.43264856338501,
|
|
"epoch": 2.3849372384937237,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004690764633961852,
|
|
"loss": 5.1422,
|
|
"mean_token_accuracy": 0.1864036113023758,
|
|
"num_tokens": 5270409.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.2199077129364015,
|
|
"epoch": 2.3891213389121337,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004689129597071893,
|
|
"loss": 5.1356,
|
|
"mean_token_accuracy": 0.1981827899813652,
|
|
"num_tokens": 5278766.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.398251724243164,
|
|
"epoch": 2.393305439330544,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046874905697649857,
|
|
"loss": 5.3132,
|
|
"mean_token_accuracy": 0.175306336581707,
|
|
"num_tokens": 5287572.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.394223594665528,
|
|
"epoch": 2.397489539748954,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046858475554139857,
|
|
"loss": 5.2127,
|
|
"mean_token_accuracy": 0.18782407343387603,
|
|
"num_tokens": 5297348.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.372149515151977,
|
|
"epoch": 2.401673640167364,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004684200557399952,
|
|
"loss": 5.2302,
|
|
"mean_token_accuracy": 0.18365682363510133,
|
|
"num_tokens": 5305748.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.339284038543701,
|
|
"epoch": 2.405857740585774,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046825495791121414,
|
|
"loss": 5.1709,
|
|
"mean_token_accuracy": 0.19215793162584305,
|
|
"num_tokens": 5314482.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.379046964645386,
|
|
"epoch": 2.410041841004184,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046808946239480017,
|
|
"loss": 5.2464,
|
|
"mean_token_accuracy": 0.17876814305782318,
|
|
"num_tokens": 5325161.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.321932125091553,
|
|
"epoch": 2.414225941422594,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004679235695313165,
|
|
"loss": 5.2372,
|
|
"mean_token_accuracy": 0.17987630516290665,
|
|
"num_tokens": 5335262.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.33101749420166,
|
|
"epoch": 2.418410041841004,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046775727966214393,
|
|
"loss": 5.1546,
|
|
"mean_token_accuracy": 0.18333236128091812,
|
|
"num_tokens": 5345145.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.274477243423462,
|
|
"epoch": 2.422594142259414,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004675905931294803,
|
|
"loss": 5.1414,
|
|
"mean_token_accuracy": 0.18669739812612535,
|
|
"num_tokens": 5354715.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.275536823272705,
|
|
"epoch": 2.426778242677824,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004674235102763398,
|
|
"loss": 5.2024,
|
|
"mean_token_accuracy": 0.18598475754261018,
|
|
"num_tokens": 5364180.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.340838670730591,
|
|
"epoch": 2.430962343096234,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046725603144655195,
|
|
"loss": 5.1738,
|
|
"mean_token_accuracy": 0.18643196821212768,
|
|
"num_tokens": 5372919.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.340966749191284,
|
|
"epoch": 2.435146443514644,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004670881569847612,
|
|
"loss": 5.1475,
|
|
"mean_token_accuracy": 0.19049450010061264,
|
|
"num_tokens": 5383002.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.390149021148682,
|
|
"epoch": 2.439330543933054,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046691988723642636,
|
|
"loss": 5.3378,
|
|
"mean_token_accuracy": 0.17228108942508696,
|
|
"num_tokens": 5393622.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.419896459579467,
|
|
"epoch": 2.4435146443514646,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004667512225478195,
|
|
"loss": 5.2151,
|
|
"mean_token_accuracy": 0.18529651761054994,
|
|
"num_tokens": 5403106.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.36248083114624,
|
|
"epoch": 2.4476987447698746,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046658216326602526,
|
|
"loss": 5.2015,
|
|
"mean_token_accuracy": 0.18339800983667373,
|
|
"num_tokens": 5412596.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.41774263381958,
|
|
"epoch": 2.4518828451882846,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004664127097389406,
|
|
"loss": 5.2444,
|
|
"mean_token_accuracy": 0.1855345368385315,
|
|
"num_tokens": 5422301.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.411597537994385,
|
|
"epoch": 2.4560669456066946,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046624286231527367,
|
|
"loss": 5.1668,
|
|
"mean_token_accuracy": 0.18422241806983947,
|
|
"num_tokens": 5432178.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.282891702651978,
|
|
"epoch": 2.4602510460251046,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004660726213445431,
|
|
"loss": 5.209,
|
|
"mean_token_accuracy": 0.18362156450748443,
|
|
"num_tokens": 5441162.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.330004358291626,
|
|
"epoch": 2.4644351464435146,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004659019871770774,
|
|
"loss": 5.1474,
|
|
"mean_token_accuracy": 0.19479316771030425,
|
|
"num_tokens": 5450566.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.3660375595092775,
|
|
"epoch": 2.4686192468619246,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046573096016401435,
|
|
"loss": 5.1836,
|
|
"mean_token_accuracy": 0.19586996287107467,
|
|
"num_tokens": 5460289.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.243880128860473,
|
|
"epoch": 2.4728033472803346,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004655595406572999,
|
|
"loss": 5.1583,
|
|
"mean_token_accuracy": 0.19155662804841994,
|
|
"num_tokens": 5469581.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.40550274848938,
|
|
"epoch": 2.476987447698745,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000465387729009688,
|
|
"loss": 5.2454,
|
|
"mean_token_accuracy": 0.189969938993454,
|
|
"num_tokens": 5478005.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.331111335754395,
|
|
"epoch": 2.481171548117155,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046521552557473927,
|
|
"loss": 5.2461,
|
|
"mean_token_accuracy": 0.180669966340065,
|
|
"num_tokens": 5486963.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.326622200012207,
|
|
"epoch": 2.485355648535565,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046504293070682073,
|
|
"loss": 5.1202,
|
|
"mean_token_accuracy": 0.18449053466320037,
|
|
"num_tokens": 5496368.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.282700920104981,
|
|
"epoch": 2.489539748953975,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046486994476110493,
|
|
"loss": 5.1265,
|
|
"mean_token_accuracy": 0.19323625713586806,
|
|
"num_tokens": 5505706.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.259168577194214,
|
|
"epoch": 2.493723849372385,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046469656809356916,
|
|
"loss": 5.1096,
|
|
"mean_token_accuracy": 0.19017785638570786,
|
|
"num_tokens": 5513945.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.2196814060211185,
|
|
"epoch": 2.497907949790795,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046452280106099454,
|
|
"loss": 5.1553,
|
|
"mean_token_accuracy": 0.18404532372951507,
|
|
"num_tokens": 5523831.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.420633029937744,
|
|
"epoch": 2.502092050209205,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004643486440209659,
|
|
"loss": 5.2771,
|
|
"mean_token_accuracy": 0.1873683363199234,
|
|
"num_tokens": 5532705.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.390104961395264,
|
|
"epoch": 2.506276150627615,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004641740973318702,
|
|
"loss": 5.1953,
|
|
"mean_token_accuracy": 0.19106512367725373,
|
|
"num_tokens": 5542793.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.359776258468628,
|
|
"epoch": 2.510460251046025,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004639991613528967,
|
|
"loss": 5.1935,
|
|
"mean_token_accuracy": 0.1863290324807167,
|
|
"num_tokens": 5551938.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.510460251046025,
|
|
"eval_entropy": 5.169321855884388,
|
|
"eval_loss": 5.5464887619018555,
|
|
"eval_mean_token_accuracy": 0.1789830338918358,
|
|
"eval_num_tokens": 5551938.0,
|
|
"eval_runtime": 3.0168,
|
|
"eval_samples_per_second": 1295.075,
|
|
"eval_steps_per_second": 162.091,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.357166910171509,
|
|
"epoch": 2.514644351464435,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046382383644403543,
|
|
"loss": 5.2321,
|
|
"mean_token_accuracy": 0.18058869987726212,
|
|
"num_tokens": 5561374.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.256979990005493,
|
|
"epoch": 2.518828451882845,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046364812296607676,
|
|
"loss": 5.0951,
|
|
"mean_token_accuracy": 0.1917743131518364,
|
|
"num_tokens": 5570459.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.255487632751465,
|
|
"epoch": 2.523012552301255,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046347202128061094,
|
|
"loss": 5.2105,
|
|
"mean_token_accuracy": 0.17877186685800553,
|
|
"num_tokens": 5579692.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.4036478996276855,
|
|
"epoch": 2.5271966527196654,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004632955317500268,
|
|
"loss": 5.1706,
|
|
"mean_token_accuracy": 0.1878630980849266,
|
|
"num_tokens": 5588338.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.335281419754028,
|
|
"epoch": 2.5313807531380754,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046311865473751154,
|
|
"loss": 5.2403,
|
|
"mean_token_accuracy": 0.1856330528855324,
|
|
"num_tokens": 5598589.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.380405378341675,
|
|
"epoch": 2.5355648535564854,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004629413906070495,
|
|
"loss": 5.2293,
|
|
"mean_token_accuracy": 0.18747957050800323,
|
|
"num_tokens": 5608127.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.384419059753418,
|
|
"epoch": 2.5397489539748954,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046276373972342177,
|
|
"loss": 5.2687,
|
|
"mean_token_accuracy": 0.1905320852994919,
|
|
"num_tokens": 5617728.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.359371852874756,
|
|
"epoch": 2.5439330543933054,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004625857024522054,
|
|
"loss": 5.217,
|
|
"mean_token_accuracy": 0.18671149164438247,
|
|
"num_tokens": 5626823.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.266793966293335,
|
|
"epoch": 2.5481171548117154,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046240727915977236,
|
|
"loss": 5.1969,
|
|
"mean_token_accuracy": 0.19211239665746688,
|
|
"num_tokens": 5636284.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.451792049407959,
|
|
"epoch": 2.5523012552301254,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004622284702132893,
|
|
"loss": 5.3029,
|
|
"mean_token_accuracy": 0.18001856356859208,
|
|
"num_tokens": 5645721.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.336426019668579,
|
|
"epoch": 2.5564853556485354,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046204927598071595,
|
|
"loss": 5.2377,
|
|
"mean_token_accuracy": 0.182869254052639,
|
|
"num_tokens": 5654734.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.366493511199951,
|
|
"epoch": 2.560669456066946,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046186969683080547,
|
|
"loss": 5.2121,
|
|
"mean_token_accuracy": 0.19295602440834045,
|
|
"num_tokens": 5664010.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.277511835098267,
|
|
"epoch": 2.564853556485356,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046168973313310294,
|
|
"loss": 5.2143,
|
|
"mean_token_accuracy": 0.18723931014537812,
|
|
"num_tokens": 5673028.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.325067138671875,
|
|
"epoch": 2.569037656903766,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004615093852579445,
|
|
"loss": 5.1813,
|
|
"mean_token_accuracy": 0.19013588428497313,
|
|
"num_tokens": 5682517.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.233159875869751,
|
|
"epoch": 2.573221757322176,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046132865357645717,
|
|
"loss": 5.1082,
|
|
"mean_token_accuracy": 0.19707662165164946,
|
|
"num_tokens": 5690920.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.275312185287476,
|
|
"epoch": 2.577405857740586,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046114753846055774,
|
|
"loss": 5.1656,
|
|
"mean_token_accuracy": 0.18295206874608994,
|
|
"num_tokens": 5700336.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.367909049987793,
|
|
"epoch": 2.581589958158996,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000460966040282952,
|
|
"loss": 5.3017,
|
|
"mean_token_accuracy": 0.18178110867738723,
|
|
"num_tokens": 5711085.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.30069785118103,
|
|
"epoch": 2.585774058577406,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046078415941713384,
|
|
"loss": 5.1157,
|
|
"mean_token_accuracy": 0.19286441653966904,
|
|
"num_tokens": 5719560.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.313541460037231,
|
|
"epoch": 2.589958158995816,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004606018962373852,
|
|
"loss": 5.0783,
|
|
"mean_token_accuracy": 0.1986253798007965,
|
|
"num_tokens": 5727865.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.304821634292603,
|
|
"epoch": 2.594142259414226,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004604192511187741,
|
|
"loss": 5.2132,
|
|
"mean_token_accuracy": 0.18388793617486954,
|
|
"num_tokens": 5737705.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.3230067729949955,
|
|
"epoch": 2.598326359832636,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046023622443715483,
|
|
"loss": 5.2124,
|
|
"mean_token_accuracy": 0.1778946965932846,
|
|
"num_tokens": 5747057.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.3239351272583,
|
|
"epoch": 2.602510460251046,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046005281656916707,
|
|
"loss": 5.1831,
|
|
"mean_token_accuracy": 0.18459225594997405,
|
|
"num_tokens": 5755793.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.348266983032227,
|
|
"epoch": 2.606694560669456,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004598690278922346,
|
|
"loss": 5.204,
|
|
"mean_token_accuracy": 0.18266982585191727,
|
|
"num_tokens": 5765385.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.192241477966308,
|
|
"epoch": 2.6108786610878663,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045968485878456497,
|
|
"loss": 5.1055,
|
|
"mean_token_accuracy": 0.19464822113513947,
|
|
"num_tokens": 5773947.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.307315492630005,
|
|
"epoch": 2.6150627615062763,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004595003096251486,
|
|
"loss": 5.1045,
|
|
"mean_token_accuracy": 0.1931895688176155,
|
|
"num_tokens": 5782700.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.348392486572266,
|
|
"epoch": 2.6192468619246863,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045931538079375803,
|
|
"loss": 5.2651,
|
|
"mean_token_accuracy": 0.1823725715279579,
|
|
"num_tokens": 5792452.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.342971086502075,
|
|
"epoch": 2.6234309623430963,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045913007267094704,
|
|
"loss": 5.1905,
|
|
"mean_token_accuracy": 0.18951852321624757,
|
|
"num_tokens": 5801270.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.358579015731811,
|
|
"epoch": 2.6276150627615062,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004589443856380499,
|
|
"loss": 5.237,
|
|
"mean_token_accuracy": 0.18819057792425156,
|
|
"num_tokens": 5810704.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.340524053573608,
|
|
"epoch": 2.6317991631799162,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004587583200771807,
|
|
"loss": 5.1769,
|
|
"mean_token_accuracy": 0.18378162533044815,
|
|
"num_tokens": 5819690.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.2421385765075685,
|
|
"epoch": 2.6359832635983262,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004585718763712325,
|
|
"loss": 5.1039,
|
|
"mean_token_accuracy": 0.18966793715953828,
|
|
"num_tokens": 5828912.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.150143051147461,
|
|
"epoch": 2.6401673640167362,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004583850549038764,
|
|
"loss": 5.0406,
|
|
"mean_token_accuracy": 0.19083824753761292,
|
|
"num_tokens": 5836914.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.313408517837525,
|
|
"epoch": 2.6443514644351467,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045819785605956105,
|
|
"loss": 5.2069,
|
|
"mean_token_accuracy": 0.1899588793516159,
|
|
"num_tokens": 5846268.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.314242172241211,
|
|
"epoch": 2.6485355648535567,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004580102802235115,
|
|
"loss": 5.2073,
|
|
"mean_token_accuracy": 0.18737068474292756,
|
|
"num_tokens": 5855029.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.374032211303711,
|
|
"epoch": 2.6527196652719667,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004578223277817286,
|
|
"loss": 5.1984,
|
|
"mean_token_accuracy": 0.1793490469455719,
|
|
"num_tokens": 5865049.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.3609302043914795,
|
|
"epoch": 2.6569037656903767,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045763399912098835,
|
|
"loss": 5.1902,
|
|
"mean_token_accuracy": 0.1795242965221405,
|
|
"num_tokens": 5873593.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.216292810440064,
|
|
"epoch": 2.6610878661087867,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045744529462884086,
|
|
"loss": 5.1208,
|
|
"mean_token_accuracy": 0.18870629817247392,
|
|
"num_tokens": 5882562.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.355183887481689,
|
|
"epoch": 2.6652719665271967,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045725621469360964,
|
|
"loss": 5.2887,
|
|
"mean_token_accuracy": 0.17769926488399507,
|
|
"num_tokens": 5892516.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.322477912902832,
|
|
"epoch": 2.6694560669456067,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004570667597043907,
|
|
"loss": 5.1852,
|
|
"mean_token_accuracy": 0.18842493742704391,
|
|
"num_tokens": 5901494.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.311296653747559,
|
|
"epoch": 2.6736401673640167,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000456876930051052,
|
|
"loss": 5.176,
|
|
"mean_token_accuracy": 0.18503561168909072,
|
|
"num_tokens": 5910750.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.383889198303223,
|
|
"epoch": 2.6778242677824267,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004566867261242325,
|
|
"loss": 5.2186,
|
|
"mean_token_accuracy": 0.188629674911499,
|
|
"num_tokens": 5920277.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.310765027999878,
|
|
"epoch": 2.6820083682008367,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045649614831534127,
|
|
"loss": 5.1952,
|
|
"mean_token_accuracy": 0.18537470698356628,
|
|
"num_tokens": 5929880.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.428402519226074,
|
|
"epoch": 2.6861924686192467,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045630519701655677,
|
|
"loss": 5.3412,
|
|
"mean_token_accuracy": 0.1875407412648201,
|
|
"num_tokens": 5939538.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.308660268783569,
|
|
"epoch": 2.6903765690376567,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045611387262082604,
|
|
"loss": 5.1758,
|
|
"mean_token_accuracy": 0.18673402816057205,
|
|
"num_tokens": 5948747.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.280479192733765,
|
|
"epoch": 2.694560669456067,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045592217552186407,
|
|
"loss": 5.1683,
|
|
"mean_token_accuracy": 0.18951367437839509,
|
|
"num_tokens": 5958209.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.37222466468811,
|
|
"epoch": 2.698744769874477,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004557301061141525,
|
|
"loss": 5.219,
|
|
"mean_token_accuracy": 0.18609079718589783,
|
|
"num_tokens": 5967184.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.268049621582032,
|
|
"epoch": 2.702928870292887,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045553766479293945,
|
|
"loss": 5.2253,
|
|
"mean_token_accuracy": 0.18060262799263,
|
|
"num_tokens": 5976006.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.2162786483764645,
|
|
"epoch": 2.707112970711297,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045534485195423806,
|
|
"loss": 5.0743,
|
|
"mean_token_accuracy": 0.19651037454605103,
|
|
"num_tokens": 5984671.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.418746852874756,
|
|
"epoch": 2.711297071129707,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045515166799482633,
|
|
"loss": 5.2046,
|
|
"mean_token_accuracy": 0.18489584624767302,
|
|
"num_tokens": 5993190.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.286226463317871,
|
|
"epoch": 2.715481171548117,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045495811331224564,
|
|
"loss": 5.1916,
|
|
"mean_token_accuracy": 0.18630554378032685,
|
|
"num_tokens": 6004080.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.320625925064087,
|
|
"epoch": 2.719665271966527,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004547641883048005,
|
|
"loss": 5.2035,
|
|
"mean_token_accuracy": 0.18668723702430726,
|
|
"num_tokens": 6013212.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.352359437942505,
|
|
"epoch": 2.723849372384937,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004545698933715573,
|
|
"loss": 5.1355,
|
|
"mean_token_accuracy": 0.19180063307285308,
|
|
"num_tokens": 6023444.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.286853885650634,
|
|
"epoch": 2.7280334728033475,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045437522891234386,
|
|
"loss": 5.1439,
|
|
"mean_token_accuracy": 0.18751030117273332,
|
|
"num_tokens": 6032217.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.258432769775391,
|
|
"epoch": 2.7322175732217575,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045418019532774827,
|
|
"loss": 5.1081,
|
|
"mean_token_accuracy": 0.1933275803923607,
|
|
"num_tokens": 6040246.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.236212825775146,
|
|
"epoch": 2.7364016736401675,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045398479301911836,
|
|
"loss": 5.1991,
|
|
"mean_token_accuracy": 0.18418097048997878,
|
|
"num_tokens": 6049447.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.191279935836792,
|
|
"epoch": 2.7405857740585775,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045378902238856064,
|
|
"loss": 5.0785,
|
|
"mean_token_accuracy": 0.1938243106007576,
|
|
"num_tokens": 6058294.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.364496278762817,
|
|
"epoch": 2.7447698744769875,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004535928838389396,
|
|
"loss": 5.2146,
|
|
"mean_token_accuracy": 0.1871143564581871,
|
|
"num_tokens": 6067237.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.389237117767334,
|
|
"epoch": 2.7489539748953975,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004533963777738768,
|
|
"loss": 5.2133,
|
|
"mean_token_accuracy": 0.18586805909872056,
|
|
"num_tokens": 6075599.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.320382022857666,
|
|
"epoch": 2.7531380753138075,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045319950459775024,
|
|
"loss": 5.1923,
|
|
"mean_token_accuracy": 0.18377586752176284,
|
|
"num_tokens": 6085131.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.351099586486816,
|
|
"epoch": 2.7573221757322175,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004530022647156931,
|
|
"loss": 5.163,
|
|
"mean_token_accuracy": 0.18796681463718415,
|
|
"num_tokens": 6094745.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.258651208877564,
|
|
"epoch": 2.7615062761506275,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004528046585335935,
|
|
"loss": 5.2053,
|
|
"mean_token_accuracy": 0.18384901583194732,
|
|
"num_tokens": 6104182.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.321671009063721,
|
|
"epoch": 2.7656903765690375,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045260668645809304,
|
|
"loss": 5.0765,
|
|
"mean_token_accuracy": 0.1958131730556488,
|
|
"num_tokens": 6113370.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.219642162322998,
|
|
"epoch": 2.7698744769874475,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004524083488965866,
|
|
"loss": 5.1501,
|
|
"mean_token_accuracy": 0.19209472686052323,
|
|
"num_tokens": 6122202.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.329074144363403,
|
|
"epoch": 2.7740585774058575,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004522096462572208,
|
|
"loss": 5.2094,
|
|
"mean_token_accuracy": 0.17987604737281798,
|
|
"num_tokens": 6131081.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.447030544281006,
|
|
"epoch": 2.778242677824268,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004520105789488939,
|
|
"loss": 5.2811,
|
|
"mean_token_accuracy": 0.1828023836016655,
|
|
"num_tokens": 6141707.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.246972894668579,
|
|
"epoch": 2.782426778242678,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045181114738125427,
|
|
"loss": 5.1455,
|
|
"mean_token_accuracy": 0.18587521314620972,
|
|
"num_tokens": 6151651.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.329950952529908,
|
|
"epoch": 2.786610878661088,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045161135196470015,
|
|
"loss": 5.0965,
|
|
"mean_token_accuracy": 0.196394182741642,
|
|
"num_tokens": 6159928.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.303078079223633,
|
|
"epoch": 2.790794979079498,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004514111931103783,
|
|
"loss": 5.1992,
|
|
"mean_token_accuracy": 0.18262383192777634,
|
|
"num_tokens": 6169115.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.3082962989807125,
|
|
"epoch": 2.794979079497908,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045121067123018353,
|
|
"loss": 5.2198,
|
|
"mean_token_accuracy": 0.18815610855817794,
|
|
"num_tokens": 6177995.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.365166044235229,
|
|
"epoch": 2.799163179916318,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045100978673675755,
|
|
"loss": 5.2073,
|
|
"mean_token_accuracy": 0.18383896201848984,
|
|
"num_tokens": 6188081.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.383021354675293,
|
|
"epoch": 2.803347280334728,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045080854004348836,
|
|
"loss": 5.161,
|
|
"mean_token_accuracy": 0.18370933681726456,
|
|
"num_tokens": 6198010.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.1978137493133545,
|
|
"epoch": 2.8075313807531384,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045060693156450933,
|
|
"loss": 5.1343,
|
|
"mean_token_accuracy": 0.19338563531637193,
|
|
"num_tokens": 6207164.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.36414155960083,
|
|
"epoch": 2.8117154811715483,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004504049617146983,
|
|
"loss": 5.2756,
|
|
"mean_token_accuracy": 0.1865020900964737,
|
|
"num_tokens": 6216512.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.351248121261596,
|
|
"epoch": 2.8158995815899583,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045020263090967674,
|
|
"loss": 5.2344,
|
|
"mean_token_accuracy": 0.18984930515289306,
|
|
"num_tokens": 6226196.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.2253562927246096,
|
|
"epoch": 2.8200836820083683,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004499999395658088,
|
|
"loss": 5.0961,
|
|
"mean_token_accuracy": 0.18750745207071304,
|
|
"num_tokens": 6234820.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.317824411392212,
|
|
"epoch": 2.8242677824267783,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000449796888100201,
|
|
"loss": 5.155,
|
|
"mean_token_accuracy": 0.1907557189464569,
|
|
"num_tokens": 6243813.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.387785243988037,
|
|
"epoch": 2.8284518828451883,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044959347693070026,
|
|
"loss": 5.2613,
|
|
"mean_token_accuracy": 0.1873804971575737,
|
|
"num_tokens": 6253376.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.392985439300537,
|
|
"epoch": 2.8326359832635983,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004493897064758943,
|
|
"loss": 5.2791,
|
|
"mean_token_accuracy": 0.18504671454429628,
|
|
"num_tokens": 6262640.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.400721645355224,
|
|
"epoch": 2.8368200836820083,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004491855771551098,
|
|
"loss": 5.233,
|
|
"mean_token_accuracy": 0.19088169932365417,
|
|
"num_tokens": 6270908.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.2806014060974125,
|
|
"epoch": 2.8410041841004183,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004489810893884121,
|
|
"loss": 5.1434,
|
|
"mean_token_accuracy": 0.1863793686032295,
|
|
"num_tokens": 6280361.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.345275402069092,
|
|
"epoch": 2.8451882845188283,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004487762435966042,
|
|
"loss": 5.155,
|
|
"mean_token_accuracy": 0.18769960552453996,
|
|
"num_tokens": 6288832.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.248450183868409,
|
|
"epoch": 2.8493723849372383,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044857104020122585,
|
|
"loss": 5.1292,
|
|
"mean_token_accuracy": 0.19043443650007247,
|
|
"num_tokens": 6298553.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.235322427749634,
|
|
"epoch": 2.8535564853556483,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004483654796245526,
|
|
"loss": 5.096,
|
|
"mean_token_accuracy": 0.1903188243508339,
|
|
"num_tokens": 6307692.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.379913330078125,
|
|
"epoch": 2.8577405857740583,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000448159562289595,
|
|
"loss": 5.1818,
|
|
"mean_token_accuracy": 0.18876910507678984,
|
|
"num_tokens": 6316729.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.377156972885132,
|
|
"epoch": 2.8619246861924688,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044795328862009794,
|
|
"loss": 5.2586,
|
|
"mean_token_accuracy": 0.17772623151540756,
|
|
"num_tokens": 6326210.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.245429372787475,
|
|
"epoch": 2.8661087866108788,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044774665904053943,
|
|
"loss": 5.1695,
|
|
"mean_token_accuracy": 0.18792026191949845,
|
|
"num_tokens": 6335156.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.2129504680633545,
|
|
"epoch": 2.8702928870292888,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004475396739761299,
|
|
"loss": 5.0415,
|
|
"mean_token_accuracy": 0.1941840723156929,
|
|
"num_tokens": 6344264.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.278776836395264,
|
|
"epoch": 2.8744769874476988,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044733233385281117,
|
|
"loss": 5.1183,
|
|
"mean_token_accuracy": 0.1926687702536583,
|
|
"num_tokens": 6352997.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.243551445007324,
|
|
"epoch": 2.8786610878661087,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044712463909725626,
|
|
"loss": 5.1339,
|
|
"mean_token_accuracy": 0.19510126113891602,
|
|
"num_tokens": 6362312.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.373373699188233,
|
|
"epoch": 2.8828451882845187,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044691659013686727,
|
|
"loss": 5.2584,
|
|
"mean_token_accuracy": 0.18169642835855485,
|
|
"num_tokens": 6372107.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.264970064163208,
|
|
"epoch": 2.8870292887029287,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004467081873997756,
|
|
"loss": 5.0956,
|
|
"mean_token_accuracy": 0.19638821929693223,
|
|
"num_tokens": 6381306.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.207119464874268,
|
|
"epoch": 2.891213389121339,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044649943131484056,
|
|
"loss": 5.029,
|
|
"mean_token_accuracy": 0.19724439084529877,
|
|
"num_tokens": 6390404.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.392963218688965,
|
|
"epoch": 2.895397489539749,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044629032231164874,
|
|
"loss": 5.2513,
|
|
"mean_token_accuracy": 0.1828698620200157,
|
|
"num_tokens": 6400366.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.281077146530151,
|
|
"epoch": 2.899581589958159,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004460808608205127,
|
|
"loss": 5.0682,
|
|
"mean_token_accuracy": 0.19214740991592408,
|
|
"num_tokens": 6408955.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.19082498550415,
|
|
"epoch": 2.903765690376569,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004458710472724706,
|
|
"loss": 5.101,
|
|
"mean_token_accuracy": 0.1934350922703743,
|
|
"num_tokens": 6417172.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.304494142532349,
|
|
"epoch": 2.907949790794979,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044566088209928485,
|
|
"loss": 5.1402,
|
|
"mean_token_accuracy": 0.19179511964321136,
|
|
"num_tokens": 6425898.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.245939922332764,
|
|
"epoch": 2.912133891213389,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044545036573344175,
|
|
"loss": 5.1741,
|
|
"mean_token_accuracy": 0.18676467686891557,
|
|
"num_tokens": 6435940.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.254410743713379,
|
|
"epoch": 2.916317991631799,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044523949860815015,
|
|
"loss": 5.1028,
|
|
"mean_token_accuracy": 0.1937511071562767,
|
|
"num_tokens": 6445119.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.240779972076416,
|
|
"epoch": 2.920502092050209,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004450282811573405,
|
|
"loss": 5.1481,
|
|
"mean_token_accuracy": 0.19389404505491256,
|
|
"num_tokens": 6453986.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.293921089172363,
|
|
"epoch": 2.924686192468619,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044481671381566446,
|
|
"loss": 5.1189,
|
|
"mean_token_accuracy": 0.18862144500017167,
|
|
"num_tokens": 6463658.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.253119277954101,
|
|
"epoch": 2.928870292887029,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004446047970184937,
|
|
"loss": 5.1662,
|
|
"mean_token_accuracy": 0.1902005970478058,
|
|
"num_tokens": 6472370.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.928870292887029,
|
|
"eval_entropy": 5.112072067026712,
|
|
"eval_loss": 5.469865798950195,
|
|
"eval_mean_token_accuracy": 0.18382642931611504,
|
|
"eval_num_tokens": 6472370.0,
|
|
"eval_runtime": 2.7402,
|
|
"eval_samples_per_second": 1425.804,
|
|
"eval_steps_per_second": 178.454,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.3380297183990475,
|
|
"epoch": 2.933054393305439,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044439253120191877,
|
|
"loss": 5.286,
|
|
"mean_token_accuracy": 0.1812704920768738,
|
|
"num_tokens": 6482642.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.325414514541626,
|
|
"epoch": 2.937238493723849,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044417991680274874,
|
|
"loss": 5.1941,
|
|
"mean_token_accuracy": 0.19350529164075853,
|
|
"num_tokens": 6491818.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.322343444824218,
|
|
"epoch": 2.941422594142259,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004439669542585097,
|
|
"loss": 5.1629,
|
|
"mean_token_accuracy": 0.18652729839086532,
|
|
"num_tokens": 6500908.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.245506954193115,
|
|
"epoch": 2.9456066945606696,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004437536440074444,
|
|
"loss": 5.1442,
|
|
"mean_token_accuracy": 0.1886870101094246,
|
|
"num_tokens": 6510182.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.287081003189087,
|
|
"epoch": 2.9497907949790796,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004435399864885111,
|
|
"loss": 5.1246,
|
|
"mean_token_accuracy": 0.1900799185037613,
|
|
"num_tokens": 6519051.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.249131345748902,
|
|
"epoch": 2.9539748953974896,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004433259821413825,
|
|
"loss": 5.063,
|
|
"mean_token_accuracy": 0.20106301158666612,
|
|
"num_tokens": 6527949.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.367303943634033,
|
|
"epoch": 2.9581589958158996,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044311163140644524,
|
|
"loss": 5.3363,
|
|
"mean_token_accuracy": 0.1785028725862503,
|
|
"num_tokens": 6537227.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.265641212463379,
|
|
"epoch": 2.9623430962343096,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044289693472479867,
|
|
"loss": 5.1364,
|
|
"mean_token_accuracy": 0.1846728578209877,
|
|
"num_tokens": 6546938.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.255853223800659,
|
|
"epoch": 2.9665271966527196,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044268189253825397,
|
|
"loss": 5.1348,
|
|
"mean_token_accuracy": 0.19034533649683,
|
|
"num_tokens": 6555750.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.248004579544068,
|
|
"epoch": 2.9707112970711296,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004424665052893334,
|
|
"loss": 5.0754,
|
|
"mean_token_accuracy": 0.19686341136693955,
|
|
"num_tokens": 6565680.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.215185642242432,
|
|
"epoch": 2.97489539748954,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044225077342126935,
|
|
"loss": 5.0488,
|
|
"mean_token_accuracy": 0.19836441725492476,
|
|
"num_tokens": 6574595.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.270293712615967,
|
|
"epoch": 2.97907949790795,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004420346973780033,
|
|
"loss": 5.1859,
|
|
"mean_token_accuracy": 0.18911138027906418,
|
|
"num_tokens": 6583333.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.324273633956909,
|
|
"epoch": 2.98326359832636,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000441818277604185,
|
|
"loss": 5.1999,
|
|
"mean_token_accuracy": 0.19488253444433212,
|
|
"num_tokens": 6592252.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.242170810699463,
|
|
"epoch": 2.98744769874477,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004416015145451715,
|
|
"loss": 5.0397,
|
|
"mean_token_accuracy": 0.19601638615131378,
|
|
"num_tokens": 6600675.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.246265029907226,
|
|
"epoch": 2.99163179916318,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004413844086470264,
|
|
"loss": 5.1178,
|
|
"mean_token_accuracy": 0.19394250065088273,
|
|
"num_tokens": 6609722.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.239301013946533,
|
|
"epoch": 2.99581589958159,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004411669603565187,
|
|
"loss": 5.116,
|
|
"mean_token_accuracy": 0.19368834048509598,
|
|
"num_tokens": 6618415.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.255900573730469,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000440949170121122,
|
|
"loss": 5.1262,
|
|
"mean_token_accuracy": 0.19740014672279357,
|
|
"num_tokens": 6627966.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.307106065750122,
|
|
"epoch": 3.00418410041841,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044073103838901373,
|
|
"loss": 4.7674,
|
|
"mean_token_accuracy": 0.21737998127937316,
|
|
"num_tokens": 6636820.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.154292154312134,
|
|
"epoch": 3.00836820083682,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044051256560907374,
|
|
"loss": 4.7969,
|
|
"mean_token_accuracy": 0.20163721293210984,
|
|
"num_tokens": 6645792.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.330294179916382,
|
|
"epoch": 3.01255230125523,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00044029375223088394,
|
|
"loss": 4.9037,
|
|
"mean_token_accuracy": 0.2003510445356369,
|
|
"num_tokens": 6654850.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.138308477401734,
|
|
"epoch": 3.01673640167364,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044007459870472706,
|
|
"loss": 4.8384,
|
|
"mean_token_accuracy": 0.206855471432209,
|
|
"num_tokens": 6664524.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.0820845603942875,
|
|
"epoch": 3.02092050209205,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004398551054815859,
|
|
"loss": 4.7731,
|
|
"mean_token_accuracy": 0.21012527942657472,
|
|
"num_tokens": 6674330.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.248129415512085,
|
|
"epoch": 3.0251046025104604,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004396352730131422,
|
|
"loss": 4.8629,
|
|
"mean_token_accuracy": 0.20349862724542617,
|
|
"num_tokens": 6683121.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.151824855804444,
|
|
"epoch": 3.0292887029288704,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004394151017517758,
|
|
"loss": 4.8044,
|
|
"mean_token_accuracy": 0.20431603044271468,
|
|
"num_tokens": 6692566.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.2952758312225345,
|
|
"epoch": 3.0334728033472804,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004391945921505638,
|
|
"loss": 4.9109,
|
|
"mean_token_accuracy": 0.19832884669303893,
|
|
"num_tokens": 6701577.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.118036508560181,
|
|
"epoch": 3.0376569037656904,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043897374466327944,
|
|
"loss": 4.7795,
|
|
"mean_token_accuracy": 0.21047478467226027,
|
|
"num_tokens": 6710672.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.161023855209351,
|
|
"epoch": 3.0418410041841004,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043875255974439145,
|
|
"loss": 4.7791,
|
|
"mean_token_accuracy": 0.2087192416191101,
|
|
"num_tokens": 6719412.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.07971568107605,
|
|
"epoch": 3.0460251046025104,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004385310378490627,
|
|
"loss": 4.766,
|
|
"mean_token_accuracy": 0.21152466386556626,
|
|
"num_tokens": 6729169.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.153553295135498,
|
|
"epoch": 3.0502092050209204,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043830917943314977,
|
|
"loss": 4.7201,
|
|
"mean_token_accuracy": 0.2165022686123848,
|
|
"num_tokens": 6738549.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.1479654788970945,
|
|
"epoch": 3.0543933054393304,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043808698495320154,
|
|
"loss": 4.781,
|
|
"mean_token_accuracy": 0.20508224815130233,
|
|
"num_tokens": 6747909.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.120273447036743,
|
|
"epoch": 3.0585774058577404,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043786445486645846,
|
|
"loss": 4.8406,
|
|
"mean_token_accuracy": 0.21035999804735184,
|
|
"num_tokens": 6757125.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.2254575252532955,
|
|
"epoch": 3.062761506276151,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004376415896308518,
|
|
"loss": 4.8399,
|
|
"mean_token_accuracy": 0.20878269672393798,
|
|
"num_tokens": 6765774.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.198799705505371,
|
|
"epoch": 3.066945606694561,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043741838970500234,
|
|
"loss": 4.8817,
|
|
"mean_token_accuracy": 0.20897659063339233,
|
|
"num_tokens": 6774980.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.18216233253479,
|
|
"epoch": 3.071129707112971,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004371948555482196,
|
|
"loss": 4.8892,
|
|
"mean_token_accuracy": 0.20027927160263062,
|
|
"num_tokens": 6784187.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.192535448074341,
|
|
"epoch": 3.075313807531381,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000436970987620501,
|
|
"loss": 4.8918,
|
|
"mean_token_accuracy": 0.2044747143983841,
|
|
"num_tokens": 6792953.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.162596225738525,
|
|
"epoch": 3.079497907949791,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043674678638253066,
|
|
"loss": 4.7624,
|
|
"mean_token_accuracy": 0.21232210844755173,
|
|
"num_tokens": 6803004.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.125282239913941,
|
|
"epoch": 3.083682008368201,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004365222522956788,
|
|
"loss": 4.8164,
|
|
"mean_token_accuracy": 0.2045421451330185,
|
|
"num_tokens": 6811883.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.143026304244995,
|
|
"epoch": 3.087866108786611,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004362973858220004,
|
|
"loss": 4.8514,
|
|
"mean_token_accuracy": 0.21020737737417222,
|
|
"num_tokens": 6820327.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.145738220214843,
|
|
"epoch": 3.092050209205021,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043607218742423447,
|
|
"loss": 4.8285,
|
|
"mean_token_accuracy": 0.2086082473397255,
|
|
"num_tokens": 6829412.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.110401916503906,
|
|
"epoch": 3.096234309623431,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043584665756580323,
|
|
"loss": 4.8135,
|
|
"mean_token_accuracy": 0.20601451992988587,
|
|
"num_tokens": 6838997.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.219425630569458,
|
|
"epoch": 3.100418410041841,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004356207967108106,
|
|
"loss": 4.8808,
|
|
"mean_token_accuracy": 0.1984048917889595,
|
|
"num_tokens": 6848957.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.124497985839843,
|
|
"epoch": 3.104602510460251,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004353946053240422,
|
|
"loss": 4.9201,
|
|
"mean_token_accuracy": 0.198494029045105,
|
|
"num_tokens": 6859688.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.228004789352417,
|
|
"epoch": 3.1087866108786613,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004351680838709635,
|
|
"loss": 4.9495,
|
|
"mean_token_accuracy": 0.19453879296779633,
|
|
"num_tokens": 6869149.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.217283868789673,
|
|
"epoch": 3.1129707112970713,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043494123281771915,
|
|
"loss": 4.9036,
|
|
"mean_token_accuracy": 0.19922375530004502,
|
|
"num_tokens": 6879432.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.145452165603638,
|
|
"epoch": 3.1171548117154813,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043471405263113214,
|
|
"loss": 4.889,
|
|
"mean_token_accuracy": 0.20554056465625764,
|
|
"num_tokens": 6888398.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.200772094726562,
|
|
"epoch": 3.1213389121338913,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004344865437787027,
|
|
"loss": 4.8063,
|
|
"mean_token_accuracy": 0.20465275943279265,
|
|
"num_tokens": 6898082.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.097585439682007,
|
|
"epoch": 3.1255230125523012,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043425870672860774,
|
|
"loss": 4.7834,
|
|
"mean_token_accuracy": 0.21189208179712296,
|
|
"num_tokens": 6907693.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.268562841415405,
|
|
"epoch": 3.1297071129707112,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004340305419496989,
|
|
"loss": 4.9823,
|
|
"mean_token_accuracy": 0.19338904172182084,
|
|
"num_tokens": 6917095.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.17300009727478,
|
|
"epoch": 3.1338912133891212,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004338020499115029,
|
|
"loss": 4.9227,
|
|
"mean_token_accuracy": 0.19877261221408843,
|
|
"num_tokens": 6926952.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.185789775848389,
|
|
"epoch": 3.1380753138075312,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004335732310842196,
|
|
"loss": 4.8669,
|
|
"mean_token_accuracy": 0.21069399565458297,
|
|
"num_tokens": 6936704.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.126188707351685,
|
|
"epoch": 3.1422594142259412,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004333440859387212,
|
|
"loss": 4.851,
|
|
"mean_token_accuracy": 0.20845145583152772,
|
|
"num_tokens": 6946614.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.220112037658692,
|
|
"epoch": 3.1464435146443517,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043311461494655165,
|
|
"loss": 4.86,
|
|
"mean_token_accuracy": 0.21031337976455688,
|
|
"num_tokens": 6955980.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.23923077583313,
|
|
"epoch": 3.1506276150627617,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004328848185799254,
|
|
"loss": 4.86,
|
|
"mean_token_accuracy": 0.2033395618200302,
|
|
"num_tokens": 6964108.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.147957849502563,
|
|
"epoch": 3.1548117154811717,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004326546973117264,
|
|
"loss": 4.8568,
|
|
"mean_token_accuracy": 0.21318745166063308,
|
|
"num_tokens": 6972895.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.144702339172364,
|
|
"epoch": 3.1589958158995817,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043242425161550723,
|
|
"loss": 4.8673,
|
|
"mean_token_accuracy": 0.20916283875703812,
|
|
"num_tokens": 6982222.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.250490617752075,
|
|
"epoch": 3.1631799163179917,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043219348196548807,
|
|
"loss": 4.8643,
|
|
"mean_token_accuracy": 0.2065814509987831,
|
|
"num_tokens": 6990820.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.1154365062713625,
|
|
"epoch": 3.1673640167364017,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004319623888365557,
|
|
"loss": 4.8932,
|
|
"mean_token_accuracy": 0.20653683990240096,
|
|
"num_tokens": 7000846.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.156445837020874,
|
|
"epoch": 3.1715481171548117,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043173097270426275,
|
|
"loss": 4.818,
|
|
"mean_token_accuracy": 0.19972968697547913,
|
|
"num_tokens": 7009086.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.341412830352783,
|
|
"epoch": 3.1757322175732217,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043149923404482634,
|
|
"loss": 4.8918,
|
|
"mean_token_accuracy": 0.20719874650239944,
|
|
"num_tokens": 7017752.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.087004375457764,
|
|
"epoch": 3.1799163179916317,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043126717333512733,
|
|
"loss": 4.8702,
|
|
"mean_token_accuracy": 0.21366522163152696,
|
|
"num_tokens": 7026535.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.183447265625,
|
|
"epoch": 3.1841004184100417,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004310347910527094,
|
|
"loss": 4.9067,
|
|
"mean_token_accuracy": 0.20852705091238022,
|
|
"num_tokens": 7035462.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.155086708068848,
|
|
"epoch": 3.1882845188284517,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004308020876757778,
|
|
"loss": 4.8141,
|
|
"mean_token_accuracy": 0.20730070620775223,
|
|
"num_tokens": 7043978.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.098757219314575,
|
|
"epoch": 3.192468619246862,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000430569063683199,
|
|
"loss": 4.8421,
|
|
"mean_token_accuracy": 0.20978484451770782,
|
|
"num_tokens": 7053757.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.060821485519409,
|
|
"epoch": 3.196652719665272,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004303357195544986,
|
|
"loss": 4.7797,
|
|
"mean_token_accuracy": 0.2038672685623169,
|
|
"num_tokens": 7062491.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.155342483520508,
|
|
"epoch": 3.200836820083682,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004301020557698615,
|
|
"loss": 4.8104,
|
|
"mean_token_accuracy": 0.21007771641016007,
|
|
"num_tokens": 7072157.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.083573055267334,
|
|
"epoch": 3.205020920502092,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042986807281013013,
|
|
"loss": 4.787,
|
|
"mean_token_accuracy": 0.2048204466700554,
|
|
"num_tokens": 7080631.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.189060211181641,
|
|
"epoch": 3.209205020920502,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004296337711568037,
|
|
"loss": 4.8535,
|
|
"mean_token_accuracy": 0.2146065503358841,
|
|
"num_tokens": 7089944.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.165537309646607,
|
|
"epoch": 3.213389121338912,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042939915129203745,
|
|
"loss": 4.9119,
|
|
"mean_token_accuracy": 0.20280353128910064,
|
|
"num_tokens": 7099190.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.042298698425293,
|
|
"epoch": 3.217573221757322,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004291642136986413,
|
|
"loss": 4.8578,
|
|
"mean_token_accuracy": 0.20252264589071273,
|
|
"num_tokens": 7107944.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.1508770942687985,
|
|
"epoch": 3.221757322175732,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000428928958860079,
|
|
"loss": 4.7282,
|
|
"mean_token_accuracy": 0.2113970160484314,
|
|
"num_tokens": 7116953.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.040340375900269,
|
|
"epoch": 3.225941422594142,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004286933872604673,
|
|
"loss": 4.7871,
|
|
"mean_token_accuracy": 0.20862122923135756,
|
|
"num_tokens": 7126236.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.137906646728515,
|
|
"epoch": 3.2301255230125525,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004284574993845746,
|
|
"loss": 4.8376,
|
|
"mean_token_accuracy": 0.20907163321971894,
|
|
"num_tokens": 7136286.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.138406753540039,
|
|
"epoch": 3.2343096234309625,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00042822129571782025,
|
|
"loss": 4.9046,
|
|
"mean_token_accuracy": 0.2057375729084015,
|
|
"num_tokens": 7145551.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.142768621444702,
|
|
"epoch": 3.2384937238493725,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042798477674627343,
|
|
"loss": 4.8536,
|
|
"mean_token_accuracy": 0.2094752386212349,
|
|
"num_tokens": 7155486.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.1550507068634035,
|
|
"epoch": 3.2426778242677825,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042774794295665213,
|
|
"loss": 4.8394,
|
|
"mean_token_accuracy": 0.20709673911333085,
|
|
"num_tokens": 7163984.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.073590469360352,
|
|
"epoch": 3.2468619246861925,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004275107948363223,
|
|
"loss": 4.8255,
|
|
"mean_token_accuracy": 0.20394547134637833,
|
|
"num_tokens": 7172854.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.051211786270142,
|
|
"epoch": 3.2510460251046025,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004272733328732965,
|
|
"loss": 4.8097,
|
|
"mean_token_accuracy": 0.20123495161533356,
|
|
"num_tokens": 7182202.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.189564371109009,
|
|
"epoch": 3.2552301255230125,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042703555755623337,
|
|
"loss": 4.9453,
|
|
"mean_token_accuracy": 0.19725567251443862,
|
|
"num_tokens": 7192062.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.240469741821289,
|
|
"epoch": 3.2594142259414225,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004267974693744362,
|
|
"loss": 4.9448,
|
|
"mean_token_accuracy": 0.19944138526916505,
|
|
"num_tokens": 7201766.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.149724340438842,
|
|
"epoch": 3.2635983263598325,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004265590688178523,
|
|
"loss": 4.8514,
|
|
"mean_token_accuracy": 0.20332693308591843,
|
|
"num_tokens": 7211361.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.131576490402222,
|
|
"epoch": 3.2677824267782425,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004263203563770716,
|
|
"loss": 4.8244,
|
|
"mean_token_accuracy": 0.211550971865654,
|
|
"num_tokens": 7220221.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.139615726470947,
|
|
"epoch": 3.2719665271966525,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000426081332543326,
|
|
"loss": 4.9433,
|
|
"mean_token_accuracy": 0.1962783083319664,
|
|
"num_tokens": 7230134.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.044312191009522,
|
|
"epoch": 3.276150627615063,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004258419978084881,
|
|
"loss": 4.7615,
|
|
"mean_token_accuracy": 0.2103862628340721,
|
|
"num_tokens": 7238713.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.153055715560913,
|
|
"epoch": 3.280334728033473,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004256023526650702,
|
|
"loss": 4.8643,
|
|
"mean_token_accuracy": 0.20162613242864608,
|
|
"num_tokens": 7247603.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.111010646820068,
|
|
"epoch": 3.284518828451883,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004253623976062236,
|
|
"loss": 4.9139,
|
|
"mean_token_accuracy": 0.20576744973659516,
|
|
"num_tokens": 7257452.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.260893726348877,
|
|
"epoch": 3.288702928870293,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004251221331257372,
|
|
"loss": 4.9114,
|
|
"mean_token_accuracy": 0.20624379366636275,
|
|
"num_tokens": 7266401.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.113575983047485,
|
|
"epoch": 3.292887029288703,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004248815597180366,
|
|
"loss": 4.8568,
|
|
"mean_token_accuracy": 0.20980830490589142,
|
|
"num_tokens": 7275836.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.1434567928314205,
|
|
"epoch": 3.297071129707113,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042464067787818333,
|
|
"loss": 4.9104,
|
|
"mean_token_accuracy": 0.20328453928232193,
|
|
"num_tokens": 7284854.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.127180099487305,
|
|
"epoch": 3.301255230125523,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004243994881018735,
|
|
"loss": 4.8693,
|
|
"mean_token_accuracy": 0.20316764414310456,
|
|
"num_tokens": 7293421.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.1008624076843265,
|
|
"epoch": 3.305439330543933,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004241579908854367,
|
|
"loss": 4.8025,
|
|
"mean_token_accuracy": 0.20773349106311798,
|
|
"num_tokens": 7302184.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.18611330986023,
|
|
"epoch": 3.309623430962343,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042391618672583546,
|
|
"loss": 4.9577,
|
|
"mean_token_accuracy": 0.20182376950979233,
|
|
"num_tokens": 7311596.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.101362419128418,
|
|
"epoch": 3.3138075313807533,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00042367407612066383,
|
|
"loss": 4.816,
|
|
"mean_token_accuracy": 0.21087881773710251,
|
|
"num_tokens": 7319866.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.126628875732422,
|
|
"epoch": 3.3179916317991633,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004234316595681468,
|
|
"loss": 4.8493,
|
|
"mean_token_accuracy": 0.2091270938515663,
|
|
"num_tokens": 7329327.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.227209281921387,
|
|
"epoch": 3.3221757322175733,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004231889375671382,
|
|
"loss": 5.0337,
|
|
"mean_token_accuracy": 0.19003052413463592,
|
|
"num_tokens": 7339717.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.191570091247558,
|
|
"epoch": 3.3263598326359833,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042294591061712124,
|
|
"loss": 4.8485,
|
|
"mean_token_accuracy": 0.20856435000896453,
|
|
"num_tokens": 7348850.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.077865982055664,
|
|
"epoch": 3.3305439330543933,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042270257921820623,
|
|
"loss": 4.8025,
|
|
"mean_token_accuracy": 0.21396382451057433,
|
|
"num_tokens": 7357081.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 4.985082817077637,
|
|
"epoch": 3.3347280334728033,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00042245894387112987,
|
|
"loss": 4.8151,
|
|
"mean_token_accuracy": 0.21312166899442672,
|
|
"num_tokens": 7365762.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.124319648742675,
|
|
"epoch": 3.3389121338912133,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004222150050772548,
|
|
"loss": 4.8817,
|
|
"mean_token_accuracy": 0.20862279534339906,
|
|
"num_tokens": 7374870.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.0922142505645756,
|
|
"epoch": 3.3430962343096233,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004219707633385677,
|
|
"loss": 4.7662,
|
|
"mean_token_accuracy": 0.20609999001026152,
|
|
"num_tokens": 7383836.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.08710560798645,
|
|
"epoch": 3.3472803347280333,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042172621915767886,
|
|
"loss": 4.8635,
|
|
"mean_token_accuracy": 0.20860992074012757,
|
|
"num_tokens": 7394059.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.3472803347280333,
|
|
"eval_entropy": 4.985736576326055,
|
|
"eval_loss": 5.46319055557251,
|
|
"eval_mean_token_accuracy": 0.18651489741292712,
|
|
"eval_num_tokens": 7394059.0,
|
|
"eval_runtime": 2.7561,
|
|
"eval_samples_per_second": 1417.58,
|
|
"eval_steps_per_second": 177.424,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.0861451625823975,
|
|
"epoch": 3.3514644351464433,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042148137303782077,
|
|
"loss": 4.8307,
|
|
"mean_token_accuracy": 0.21213750094175338,
|
|
"num_tokens": 7402947.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.149941492080688,
|
|
"epoch": 3.3556485355648533,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004212362254828475,
|
|
"loss": 4.8958,
|
|
"mean_token_accuracy": 0.2033420354127884,
|
|
"num_tokens": 7413430.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.202378416061402,
|
|
"epoch": 3.3598326359832638,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042099077699723326,
|
|
"loss": 4.8931,
|
|
"mean_token_accuracy": 0.20540930330753326,
|
|
"num_tokens": 7423239.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.094257879257202,
|
|
"epoch": 3.3640167364016738,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004207450280860716,
|
|
"loss": 4.8709,
|
|
"mean_token_accuracy": 0.20930316001176835,
|
|
"num_tokens": 7432519.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.083430433273316,
|
|
"epoch": 3.3682008368200838,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004204989792550742,
|
|
"loss": 4.8364,
|
|
"mean_token_accuracy": 0.20555925667285918,
|
|
"num_tokens": 7442089.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.190230035781861,
|
|
"epoch": 3.3723849372384938,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00042025263101057016,
|
|
"loss": 4.8583,
|
|
"mean_token_accuracy": 0.20592181831598283,
|
|
"num_tokens": 7451059.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.11042799949646,
|
|
"epoch": 3.3765690376569037,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00042000598385950427,
|
|
"loss": 4.8144,
|
|
"mean_token_accuracy": 0.21088250875473022,
|
|
"num_tokens": 7458960.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.126355361938477,
|
|
"epoch": 3.3807531380753137,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000419759038309437,
|
|
"loss": 5.0011,
|
|
"mean_token_accuracy": 0.19898598790168762,
|
|
"num_tokens": 7468863.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.155329084396362,
|
|
"epoch": 3.3849372384937237,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041951179486854253,
|
|
"loss": 4.9566,
|
|
"mean_token_accuracy": 0.20642842054367067,
|
|
"num_tokens": 7477204.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.11786527633667,
|
|
"epoch": 3.3891213389121337,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000419264254045608,
|
|
"loss": 4.7906,
|
|
"mean_token_accuracy": 0.20480405390262604,
|
|
"num_tokens": 7487367.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 4.98615870475769,
|
|
"epoch": 3.393305439330544,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004190164163500326,
|
|
"loss": 4.7855,
|
|
"mean_token_accuracy": 0.2164631724357605,
|
|
"num_tokens": 7495808.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.120810890197754,
|
|
"epoch": 3.397489539748954,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004187682822918266,
|
|
"loss": 4.9104,
|
|
"mean_token_accuracy": 0.1951922059059143,
|
|
"num_tokens": 7506119.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.126148271560669,
|
|
"epoch": 3.401673640167364,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00041851985238161,
|
|
"loss": 4.8846,
|
|
"mean_token_accuracy": 0.20199777334928512,
|
|
"num_tokens": 7514931.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.0606081008911135,
|
|
"epoch": 3.405857740585774,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00041827112713061155,
|
|
"loss": 4.6942,
|
|
"mean_token_accuracy": 0.22149741649627686,
|
|
"num_tokens": 7523753.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.064794969558716,
|
|
"epoch": 3.410041841004184,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00041802210705066775,
|
|
"loss": 4.7682,
|
|
"mean_token_accuracy": 0.21452270895242692,
|
|
"num_tokens": 7532447.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.082556819915771,
|
|
"epoch": 3.414225941422594,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000417772792654222,
|
|
"loss": 4.9248,
|
|
"mean_token_accuracy": 0.20070126205682753,
|
|
"num_tokens": 7541388.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.150190734863282,
|
|
"epoch": 3.418410041841004,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041752318445432323,
|
|
"loss": 4.9239,
|
|
"mean_token_accuracy": 0.19895025342702866,
|
|
"num_tokens": 7550422.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.1157999515533445,
|
|
"epoch": 3.422594142259414,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000417273282964625,
|
|
"loss": 4.7817,
|
|
"mean_token_accuracy": 0.21615640223026275,
|
|
"num_tokens": 7559522.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.112621021270752,
|
|
"epoch": 3.426778242677824,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004170230886993844,
|
|
"loss": 4.902,
|
|
"mean_token_accuracy": 0.19627136141061782,
|
|
"num_tokens": 7568785.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.065780305862427,
|
|
"epoch": 3.430962343096234,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041677260217346093,
|
|
"loss": 4.8267,
|
|
"mean_token_accuracy": 0.21189531534910203,
|
|
"num_tokens": 7579013.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.111838912963867,
|
|
"epoch": 3.435146443514644,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041652182390231556,
|
|
"loss": 4.8605,
|
|
"mean_token_accuracy": 0.20376810282468796,
|
|
"num_tokens": 7588577.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.090034055709839,
|
|
"epoch": 3.439330543933054,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041627075440200977,
|
|
"loss": 4.7785,
|
|
"mean_token_accuracy": 0.2134248748421669,
|
|
"num_tokens": 7597702.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.063868522644043,
|
|
"epoch": 3.4435146443514646,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004160193941892043,
|
|
"loss": 4.8422,
|
|
"mean_token_accuracy": 0.20544036775827407,
|
|
"num_tokens": 7607156.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.107327699661255,
|
|
"epoch": 3.4476987447698746,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004157677437811578,
|
|
"loss": 4.9251,
|
|
"mean_token_accuracy": 0.2133098989725113,
|
|
"num_tokens": 7615936.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.157761907577514,
|
|
"epoch": 3.4518828451882846,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004155158036957265,
|
|
"loss": 4.9077,
|
|
"mean_token_accuracy": 0.2041935294866562,
|
|
"num_tokens": 7626568.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.1917156219482425,
|
|
"epoch": 3.4560669456066946,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004152635744513626,
|
|
"loss": 4.8458,
|
|
"mean_token_accuracy": 0.21043034940958022,
|
|
"num_tokens": 7635102.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.013117122650146,
|
|
"epoch": 3.4602510460251046,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004150110565671133,
|
|
"loss": 4.7563,
|
|
"mean_token_accuracy": 0.22028307914733886,
|
|
"num_tokens": 7644204.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.116233921051025,
|
|
"epoch": 3.4644351464435146,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041475825056261985,
|
|
"loss": 4.8843,
|
|
"mean_token_accuracy": 0.20191183984279631,
|
|
"num_tokens": 7653585.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.189920139312744,
|
|
"epoch": 3.4686192468619246,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004145051569581162,
|
|
"loss": 4.8768,
|
|
"mean_token_accuracy": 0.19981294125318527,
|
|
"num_tokens": 7662734.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.151976680755615,
|
|
"epoch": 3.4728033472803346,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004142517762744286,
|
|
"loss": 4.8322,
|
|
"mean_token_accuracy": 0.2101568877696991,
|
|
"num_tokens": 7671682.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.149965047836304,
|
|
"epoch": 3.476987447698745,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004139981090329735,
|
|
"loss": 4.9305,
|
|
"mean_token_accuracy": 0.20690995901823045,
|
|
"num_tokens": 7681551.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.124179649353027,
|
|
"epoch": 3.481171548117155,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004137441557557575,
|
|
"loss": 4.9015,
|
|
"mean_token_accuracy": 0.20703185349702835,
|
|
"num_tokens": 7690943.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.073544120788574,
|
|
"epoch": 3.485355648535565,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041348991696537533,
|
|
"loss": 4.793,
|
|
"mean_token_accuracy": 0.21441176235675813,
|
|
"num_tokens": 7700376.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.148408889770508,
|
|
"epoch": 3.489539748953975,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004132353931850098,
|
|
"loss": 4.9043,
|
|
"mean_token_accuracy": 0.19997784048318862,
|
|
"num_tokens": 7709845.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.187413167953491,
|
|
"epoch": 3.493723849372385,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004129805849384299,
|
|
"loss": 4.9811,
|
|
"mean_token_accuracy": 0.20034596025943757,
|
|
"num_tokens": 7719908.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.0858958721160885,
|
|
"epoch": 3.497907949790795,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004127254927499898,
|
|
"loss": 4.8971,
|
|
"mean_token_accuracy": 0.20275071114301682,
|
|
"num_tokens": 7729543.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.146850728988648,
|
|
"epoch": 3.502092050209205,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004124701171446286,
|
|
"loss": 4.9043,
|
|
"mean_token_accuracy": 0.2049371689558029,
|
|
"num_tokens": 7739371.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.10733208656311,
|
|
"epoch": 3.506276150627615,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00041221445864786794,
|
|
"loss": 4.8187,
|
|
"mean_token_accuracy": 0.20833780020475387,
|
|
"num_tokens": 7748027.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.0194145202636715,
|
|
"epoch": 3.510460251046025,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000411958517785812,
|
|
"loss": 4.7581,
|
|
"mean_token_accuracy": 0.21444778442382811,
|
|
"num_tokens": 7756588.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.1101737976074215,
|
|
"epoch": 3.514644351464435,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041170229508514584,
|
|
"loss": 4.7468,
|
|
"mean_token_accuracy": 0.20663689076900482,
|
|
"num_tokens": 7765222.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.092128467559815,
|
|
"epoch": 3.518828451882845,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004114457910731347,
|
|
"loss": 4.8657,
|
|
"mean_token_accuracy": 0.2106161281466484,
|
|
"num_tokens": 7774139.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.026708507537842,
|
|
"epoch": 3.523012552301255,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004111890062776226,
|
|
"loss": 4.7452,
|
|
"mean_token_accuracy": 0.20863335132598876,
|
|
"num_tokens": 7783597.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.045945262908935,
|
|
"epoch": 3.5271966527196654,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004109319412270314,
|
|
"loss": 4.8409,
|
|
"mean_token_accuracy": 0.21286385655403137,
|
|
"num_tokens": 7792711.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.2014275074005125,
|
|
"epoch": 3.5313807531380754,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004106745964503594,
|
|
"loss": 4.8983,
|
|
"mean_token_accuracy": 0.2045663923025131,
|
|
"num_tokens": 7801878.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.030341577529907,
|
|
"epoch": 3.5355648535564854,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000410416972477181,
|
|
"loss": 4.798,
|
|
"mean_token_accuracy": 0.21313095539808274,
|
|
"num_tokens": 7810484.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.141058349609375,
|
|
"epoch": 3.5397489539748954,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00041015906983764495,
|
|
"loss": 4.9184,
|
|
"mean_token_accuracy": 0.204750457406044,
|
|
"num_tokens": 7820521.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.168326950073242,
|
|
"epoch": 3.5439330543933054,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00040990088906247326,
|
|
"loss": 4.8305,
|
|
"mean_token_accuracy": 0.20982770025730133,
|
|
"num_tokens": 7829218.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.05530309677124,
|
|
"epoch": 3.5481171548117154,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004096424306829605,
|
|
"loss": 4.8607,
|
|
"mean_token_accuracy": 0.20657587945461273,
|
|
"num_tokens": 7837795.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.165108299255371,
|
|
"epoch": 3.5523012552301254,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004093836952309726,
|
|
"loss": 4.8744,
|
|
"mean_token_accuracy": 0.2080150991678238,
|
|
"num_tokens": 7846973.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.1759199619293215,
|
|
"epoch": 3.5564853556485354,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004091246832389453,
|
|
"loss": 4.9133,
|
|
"mean_token_accuracy": 0.20285059064626693,
|
|
"num_tokens": 7856627.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.04175763130188,
|
|
"epoch": 3.560669456066946,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004088653952398837,
|
|
"loss": 4.8365,
|
|
"mean_token_accuracy": 0.2079515963792801,
|
|
"num_tokens": 7866308.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.077424430847168,
|
|
"epoch": 3.564853556485356,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040860583176736097,
|
|
"loss": 4.8262,
|
|
"mean_token_accuracy": 0.20950917750597,
|
|
"num_tokens": 7875350.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.077731943130493,
|
|
"epoch": 3.569037656903766,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004083459933555169,
|
|
"loss": 4.8298,
|
|
"mean_token_accuracy": 0.21197184771299363,
|
|
"num_tokens": 7884276.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.107650852203369,
|
|
"epoch": 3.573221757322176,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004080858805390571,
|
|
"loss": 4.8995,
|
|
"mean_token_accuracy": 0.20368795841932297,
|
|
"num_tokens": 7893836.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.1801831245422365,
|
|
"epoch": 3.577405857740586,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000407825493853252,
|
|
"loss": 4.8734,
|
|
"mean_token_accuracy": 0.2066332772374153,
|
|
"num_tokens": 7903249.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.120818758010865,
|
|
"epoch": 3.581589958158996,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040756483383393556,
|
|
"loss": 4.9225,
|
|
"mean_token_accuracy": 0.203904427587986,
|
|
"num_tokens": 7912217.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.086220693588257,
|
|
"epoch": 3.585774058577406,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004073039010175041,
|
|
"loss": 4.9629,
|
|
"mean_token_accuracy": 0.2006915271282196,
|
|
"num_tokens": 7922241.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.186107063293457,
|
|
"epoch": 3.589958158995816,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040704269594091556,
|
|
"loss": 4.871,
|
|
"mean_token_accuracy": 0.2071731850504875,
|
|
"num_tokens": 7931866.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.1948097229003904,
|
|
"epoch": 3.594142259414226,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004067812191416878,
|
|
"loss": 4.8986,
|
|
"mean_token_accuracy": 0.20188879817724228,
|
|
"num_tokens": 7941787.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.117114782333374,
|
|
"epoch": 3.598326359832636,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004065194711578982,
|
|
"loss": 4.9136,
|
|
"mean_token_accuracy": 0.20612358748912812,
|
|
"num_tokens": 7950832.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.1436402797698975,
|
|
"epoch": 3.602510460251046,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004062574525281819,
|
|
"loss": 4.8968,
|
|
"mean_token_accuracy": 0.20418329536914825,
|
|
"num_tokens": 7960059.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.134668636322021,
|
|
"epoch": 3.606694560669456,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00040599516379173113,
|
|
"loss": 4.8861,
|
|
"mean_token_accuracy": 0.20005105286836625,
|
|
"num_tokens": 7969411.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.116720819473267,
|
|
"epoch": 3.6108786610878663,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040573260548829417,
|
|
"loss": 4.8842,
|
|
"mean_token_accuracy": 0.20714796483516693,
|
|
"num_tokens": 7978446.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.015983247756958,
|
|
"epoch": 3.6150627615062763,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004054697781581734,
|
|
"loss": 4.7564,
|
|
"mean_token_accuracy": 0.20707345455884935,
|
|
"num_tokens": 7987117.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.107610368728638,
|
|
"epoch": 3.6192468619246863,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004052066823422255,
|
|
"loss": 4.8457,
|
|
"mean_token_accuracy": 0.20731030106544496,
|
|
"num_tokens": 7995989.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.082603549957275,
|
|
"epoch": 3.6234309623430963,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004049433185818593,
|
|
"loss": 4.9229,
|
|
"mean_token_accuracy": 0.20139078050851822,
|
|
"num_tokens": 8005486.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.078189849853516,
|
|
"epoch": 3.6276150627615062,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040467968741903507,
|
|
"loss": 4.7734,
|
|
"mean_token_accuracy": 0.21146121770143508,
|
|
"num_tokens": 8014853.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.125499629974366,
|
|
"epoch": 3.6317991631799162,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004044157893962633,
|
|
"loss": 4.8429,
|
|
"mean_token_accuracy": 0.21341473758220672,
|
|
"num_tokens": 8023931.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.103664350509644,
|
|
"epoch": 3.6359832635983262,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004041516250566037,
|
|
"loss": 4.8827,
|
|
"mean_token_accuracy": 0.21109170764684676,
|
|
"num_tokens": 8032659.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.145339918136597,
|
|
"epoch": 3.6401673640167362,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040388719494366404,
|
|
"loss": 4.9573,
|
|
"mean_token_accuracy": 0.20322352200746535,
|
|
"num_tokens": 8041857.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.213011121749878,
|
|
"epoch": 3.6443514644351467,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00040362249960159904,
|
|
"loss": 4.9744,
|
|
"mean_token_accuracy": 0.20129672586917877,
|
|
"num_tokens": 8051186.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.097787809371948,
|
|
"epoch": 3.6485355648535567,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00040335753957510906,
|
|
"loss": 4.8951,
|
|
"mean_token_accuracy": 0.19750130027532578,
|
|
"num_tokens": 8060396.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.073494338989258,
|
|
"epoch": 3.6527196652719667,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040309231540943937,
|
|
"loss": 4.8079,
|
|
"mean_token_accuracy": 0.21439304649829866,
|
|
"num_tokens": 8068106.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.120754909515381,
|
|
"epoch": 3.6569037656903767,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004028268276503784,
|
|
"loss": 4.8483,
|
|
"mean_token_accuracy": 0.2051659792661667,
|
|
"num_tokens": 8077878.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.1757666110992435,
|
|
"epoch": 3.6610878661087867,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004025610768442576,
|
|
"loss": 4.9738,
|
|
"mean_token_accuracy": 0.1989295795559883,
|
|
"num_tokens": 8086968.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.131380891799926,
|
|
"epoch": 3.6652719665271967,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004022950635379492,
|
|
"loss": 4.7689,
|
|
"mean_token_accuracy": 0.22497869580984114,
|
|
"num_tokens": 8095999.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.033745956420899,
|
|
"epoch": 3.6694560669456067,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040202878827886587,
|
|
"loss": 4.7985,
|
|
"mean_token_accuracy": 0.2179463043808937,
|
|
"num_tokens": 8105376.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.106619548797608,
|
|
"epoch": 3.6736401673640167,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004017622516149592,
|
|
"loss": 4.8912,
|
|
"mean_token_accuracy": 0.20658889710903167,
|
|
"num_tokens": 8115024.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.121318626403808,
|
|
"epoch": 3.6778242677824267,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004014954540947189,
|
|
"loss": 4.8889,
|
|
"mean_token_accuracy": 0.20341428220272065,
|
|
"num_tokens": 8124306.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.090257740020752,
|
|
"epoch": 3.6820083682008367,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00040122839626717137,
|
|
"loss": 4.8792,
|
|
"mean_token_accuracy": 0.21472734659910203,
|
|
"num_tokens": 8133151.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.089275932312011,
|
|
"epoch": 3.6861924686192467,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004009610786818786,
|
|
"loss": 4.8903,
|
|
"mean_token_accuracy": 0.2134702667593956,
|
|
"num_tokens": 8141745.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.059766674041748,
|
|
"epoch": 3.6903765690376567,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040069350188893735,
|
|
"loss": 4.7383,
|
|
"mean_token_accuracy": 0.215824893116951,
|
|
"num_tokens": 8150884.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.093609046936035,
|
|
"epoch": 3.694560669456067,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040042566643897745,
|
|
"loss": 4.8414,
|
|
"mean_token_accuracy": 0.20642719566822051,
|
|
"num_tokens": 8160138.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.150986671447754,
|
|
"epoch": 3.698744769874477,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004001575728831613,
|
|
"loss": 4.9199,
|
|
"mean_token_accuracy": 0.20834485590457916,
|
|
"num_tokens": 8169655.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.213423728942871,
|
|
"epoch": 3.702928870292887,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003998892217731824,
|
|
"loss": 4.9225,
|
|
"mean_token_accuracy": 0.21138543933629989,
|
|
"num_tokens": 8178846.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.066442012786865,
|
|
"epoch": 3.707112970711297,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000399620613661264,
|
|
"loss": 4.833,
|
|
"mean_token_accuracy": 0.2046512633562088,
|
|
"num_tokens": 8187525.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.121527099609375,
|
|
"epoch": 3.711297071129707,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003993517491001586,
|
|
"loss": 4.9248,
|
|
"mean_token_accuracy": 0.20239897221326827,
|
|
"num_tokens": 8196855.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.039672756195069,
|
|
"epoch": 3.715481171548117,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003990826286431462,
|
|
"loss": 4.8132,
|
|
"mean_token_accuracy": 0.204982253909111,
|
|
"num_tokens": 8205530.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.113843774795532,
|
|
"epoch": 3.719665271966527,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003988132528440333,
|
|
"loss": 4.8777,
|
|
"mean_token_accuracy": 0.2079408660531044,
|
|
"num_tokens": 8215066.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.095618438720703,
|
|
"epoch": 3.723849372384937,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003985436222571521,
|
|
"loss": 4.8609,
|
|
"mean_token_accuracy": 0.20079515129327774,
|
|
"num_tokens": 8224585.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.1642131328582765,
|
|
"epoch": 3.7280334728033475,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000398273737437359,
|
|
"loss": 4.9373,
|
|
"mean_token_accuracy": 0.20152941048145295,
|
|
"num_tokens": 8233334.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.148773813247681,
|
|
"epoch": 3.7322175732217575,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039800359894003355,
|
|
"loss": 4.8705,
|
|
"mean_token_accuracy": 0.2125294029712677,
|
|
"num_tokens": 8241636.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.136564540863037,
|
|
"epoch": 3.7364016736401675,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003977332073210774,
|
|
"loss": 4.9004,
|
|
"mean_token_accuracy": 0.2041999727487564,
|
|
"num_tokens": 8250273.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.139496755599976,
|
|
"epoch": 3.7405857740585775,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003974625631369129,
|
|
"loss": 4.9344,
|
|
"mean_token_accuracy": 0.1999513953924179,
|
|
"num_tokens": 8260349.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.108516979217529,
|
|
"epoch": 3.7447698744769875,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003971916669444824,
|
|
"loss": 4.929,
|
|
"mean_token_accuracy": 0.20351004600524902,
|
|
"num_tokens": 8270505.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.197675275802612,
|
|
"epoch": 3.7489539748953975,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00039692051930124663,
|
|
"loss": 4.8964,
|
|
"mean_token_accuracy": 0.20123675167560579,
|
|
"num_tokens": 8278651.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.111096525192261,
|
|
"epoch": 3.7531380753138075,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003966491207651838,
|
|
"loss": 4.8475,
|
|
"mean_token_accuracy": 0.21011456847190857,
|
|
"num_tokens": 8287173.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.171531820297242,
|
|
"epoch": 3.7573221757322175,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039637747189478866,
|
|
"loss": 4.8576,
|
|
"mean_token_accuracy": 0.2036084309220314,
|
|
"num_tokens": 8295785.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.143098258972168,
|
|
"epoch": 3.7615062761506275,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003961055732490708,
|
|
"loss": 4.9552,
|
|
"mean_token_accuracy": 0.2031768247485161,
|
|
"num_tokens": 8305534.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.098934125900269,
|
|
"epoch": 3.7656903765690375,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039583342538755374,
|
|
"loss": 4.844,
|
|
"mean_token_accuracy": 0.2070768177509308,
|
|
"num_tokens": 8314232.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.7656903765690375,
|
|
"eval_entropy": 4.91763174119415,
|
|
"eval_loss": 5.411562919616699,
|
|
"eval_mean_token_accuracy": 0.19110058825501147,
|
|
"eval_num_tokens": 8314232.0,
|
|
"eval_runtime": 2.956,
|
|
"eval_samples_per_second": 1321.726,
|
|
"eval_steps_per_second": 165.427,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.1339836597442625,
|
|
"epoch": 3.7698744769874475,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039556102887027423,
|
|
"loss": 4.9124,
|
|
"mean_token_accuracy": 0.1987938776612282,
|
|
"num_tokens": 8323230.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.180046796798706,
|
|
"epoch": 3.7740585774058575,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039528838425778053,
|
|
"loss": 4.9411,
|
|
"mean_token_accuracy": 0.20595014244318008,
|
|
"num_tokens": 8333072.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.1110107421875,
|
|
"epoch": 3.778242677824268,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003950154921111313,
|
|
"loss": 4.9007,
|
|
"mean_token_accuracy": 0.20191781520843505,
|
|
"num_tokens": 8342953.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.114612770080567,
|
|
"epoch": 3.782426778242678,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039474235299189484,
|
|
"loss": 4.8899,
|
|
"mean_token_accuracy": 0.20523059368133545,
|
|
"num_tokens": 8351729.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.192487192153931,
|
|
"epoch": 3.786610878661088,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003944689674621474,
|
|
"loss": 4.9468,
|
|
"mean_token_accuracy": 0.20631079077720643,
|
|
"num_tokens": 8360929.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.154665756225586,
|
|
"epoch": 3.790794979079498,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039419533608447244,
|
|
"loss": 4.9408,
|
|
"mean_token_accuracy": 0.20518142879009246,
|
|
"num_tokens": 8369506.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.153623580932617,
|
|
"epoch": 3.794979079497908,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003939214594219595,
|
|
"loss": 4.8527,
|
|
"mean_token_accuracy": 0.20913135707378389,
|
|
"num_tokens": 8378583.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.1313371658325195,
|
|
"epoch": 3.799163179916318,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039364733803820263,
|
|
"loss": 4.8364,
|
|
"mean_token_accuracy": 0.21072372943162918,
|
|
"num_tokens": 8388078.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.020518255233765,
|
|
"epoch": 3.803347280334728,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003933729724972996,
|
|
"loss": 4.8356,
|
|
"mean_token_accuracy": 0.20935306698083878,
|
|
"num_tokens": 8397086.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.069676208496094,
|
|
"epoch": 3.8075313807531384,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039309836336385044,
|
|
"loss": 4.829,
|
|
"mean_token_accuracy": 0.22389160692691804,
|
|
"num_tokens": 8405621.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.102335405349732,
|
|
"epoch": 3.8117154811715483,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003928235112029568,
|
|
"loss": 4.9006,
|
|
"mean_token_accuracy": 0.2056551069021225,
|
|
"num_tokens": 8415220.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.144475269317627,
|
|
"epoch": 3.8158995815899583,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003925484165802201,
|
|
"loss": 4.9142,
|
|
"mean_token_accuracy": 0.20348594039678575,
|
|
"num_tokens": 8424719.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.063912963867187,
|
|
"epoch": 3.8200836820083683,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003922730800617408,
|
|
"loss": 4.85,
|
|
"mean_token_accuracy": 0.20849055647850037,
|
|
"num_tokens": 8433844.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.007762098312378,
|
|
"epoch": 3.8242677824267783,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003919975022141174,
|
|
"loss": 4.7826,
|
|
"mean_token_accuracy": 0.21511289924383165,
|
|
"num_tokens": 8442700.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.22186312675476,
|
|
"epoch": 3.8284518828451883,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00039172168360444463,
|
|
"loss": 5.0249,
|
|
"mean_token_accuracy": 0.19310083240270615,
|
|
"num_tokens": 8452246.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.129202032089234,
|
|
"epoch": 3.8326359832635983,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039144562480031297,
|
|
"loss": 4.8596,
|
|
"mean_token_accuracy": 0.21053662449121474,
|
|
"num_tokens": 8461977.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.026481199264526,
|
|
"epoch": 3.8368200836820083,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039116932636980685,
|
|
"loss": 4.7689,
|
|
"mean_token_accuracy": 0.20971563160419465,
|
|
"num_tokens": 8470124.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.091988611221313,
|
|
"epoch": 3.8410041841004183,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039089278888150417,
|
|
"loss": 4.9136,
|
|
"mean_token_accuracy": 0.19794375747442244,
|
|
"num_tokens": 8479408.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.071839761734009,
|
|
"epoch": 3.8451882845188283,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003906160129044746,
|
|
"loss": 4.8261,
|
|
"mean_token_accuracy": 0.2049538791179657,
|
|
"num_tokens": 8488239.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.197607517242432,
|
|
"epoch": 3.8493723849372383,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003903389990082787,
|
|
"loss": 4.9037,
|
|
"mean_token_accuracy": 0.19678782671689987,
|
|
"num_tokens": 8497690.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.194812679290772,
|
|
"epoch": 3.8535564853556483,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003900617477629663,
|
|
"loss": 4.9648,
|
|
"mean_token_accuracy": 0.2004719316959381,
|
|
"num_tokens": 8506974.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.1031715869903564,
|
|
"epoch": 3.8577405857740583,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003897842597390761,
|
|
"loss": 4.907,
|
|
"mean_token_accuracy": 0.20890475660562516,
|
|
"num_tokens": 8516066.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.0752955913543705,
|
|
"epoch": 3.8619246861924688,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003895065355076338,
|
|
"loss": 4.8654,
|
|
"mean_token_accuracy": 0.20371763408184052,
|
|
"num_tokens": 8524747.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.065091371536255,
|
|
"epoch": 3.8661087866108788,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003892285756401511,
|
|
"loss": 4.8698,
|
|
"mean_token_accuracy": 0.21580760478973388,
|
|
"num_tokens": 8534274.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.0911235332489015,
|
|
"epoch": 3.8702928870292888,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003889503807086248,
|
|
"loss": 4.8715,
|
|
"mean_token_accuracy": 0.2059857204556465,
|
|
"num_tokens": 8543453.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.114228630065918,
|
|
"epoch": 3.8744769874476988,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038867195128553536,
|
|
"loss": 4.9553,
|
|
"mean_token_accuracy": 0.2071972444653511,
|
|
"num_tokens": 8553190.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.14755859375,
|
|
"epoch": 3.8786610878661087,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003883932879438457,
|
|
"loss": 4.8298,
|
|
"mean_token_accuracy": 0.20431060791015626,
|
|
"num_tokens": 8562814.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.028541326522827,
|
|
"epoch": 3.8828451882845187,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003881143912570002,
|
|
"loss": 4.842,
|
|
"mean_token_accuracy": 0.20079325437545775,
|
|
"num_tokens": 8572131.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.078951597213745,
|
|
"epoch": 3.8870292887029287,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003878352617989235,
|
|
"loss": 4.9203,
|
|
"mean_token_accuracy": 0.1968255579471588,
|
|
"num_tokens": 8582325.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.1008062839508055,
|
|
"epoch": 3.891213389121339,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003875559001440191,
|
|
"loss": 4.8256,
|
|
"mean_token_accuracy": 0.21458019614219664,
|
|
"num_tokens": 8591704.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.137116861343384,
|
|
"epoch": 3.895397489539749,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003872763068671682,
|
|
"loss": 4.8893,
|
|
"mean_token_accuracy": 0.20066926181316375,
|
|
"num_tokens": 8601136.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.052433252334595,
|
|
"epoch": 3.899581589958159,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003869964825437291,
|
|
"loss": 4.8619,
|
|
"mean_token_accuracy": 0.20738904625177385,
|
|
"num_tokens": 8610491.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.124885940551758,
|
|
"epoch": 3.903765690376569,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038671642774953507,
|
|
"loss": 4.8947,
|
|
"mean_token_accuracy": 0.20506638288497925,
|
|
"num_tokens": 8619199.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.103211355209351,
|
|
"epoch": 3.907949790794979,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003864361430608939,
|
|
"loss": 4.9461,
|
|
"mean_token_accuracy": 0.20372416228055953,
|
|
"num_tokens": 8629291.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.114092493057251,
|
|
"epoch": 3.912133891213389,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003861556290545864,
|
|
"loss": 4.9198,
|
|
"mean_token_accuracy": 0.20413174778223037,
|
|
"num_tokens": 8639212.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.108756685256958,
|
|
"epoch": 3.916317991631799,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038587488630786534,
|
|
"loss": 4.8623,
|
|
"mean_token_accuracy": 0.20985756069421768,
|
|
"num_tokens": 8648351.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.134943580627441,
|
|
"epoch": 3.920502092050209,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038559391539845413,
|
|
"loss": 4.9085,
|
|
"mean_token_accuracy": 0.202670918405056,
|
|
"num_tokens": 8657267.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.131587409973145,
|
|
"epoch": 3.924686192468619,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038531271690454567,
|
|
"loss": 5.0421,
|
|
"mean_token_accuracy": 0.19898991137742997,
|
|
"num_tokens": 8666982.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.200097894668579,
|
|
"epoch": 3.928870292887029,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003850312914048015,
|
|
"loss": 4.8429,
|
|
"mean_token_accuracy": 0.2136537253856659,
|
|
"num_tokens": 8676067.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.112923431396484,
|
|
"epoch": 3.933054393305439,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003847496394783496,
|
|
"loss": 4.8878,
|
|
"mean_token_accuracy": 0.21538768112659454,
|
|
"num_tokens": 8685106.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.087848043441772,
|
|
"epoch": 3.937238493723849,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003844677617047848,
|
|
"loss": 4.9409,
|
|
"mean_token_accuracy": 0.1996578350663185,
|
|
"num_tokens": 8695203.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.042820072174072,
|
|
"epoch": 3.941422594142259,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000384185658664166,
|
|
"loss": 4.7758,
|
|
"mean_token_accuracy": 0.21243813037872314,
|
|
"num_tokens": 8703700.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.169864368438721,
|
|
"epoch": 3.9456066945606696,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038390333093701596,
|
|
"loss": 4.9168,
|
|
"mean_token_accuracy": 0.2031609058380127,
|
|
"num_tokens": 8713393.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.109617900848389,
|
|
"epoch": 3.9497907949790796,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038362077910431973,
|
|
"loss": 4.8715,
|
|
"mean_token_accuracy": 0.2068963512778282,
|
|
"num_tokens": 8722263.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.067483520507812,
|
|
"epoch": 3.9539748953974896,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003833380037475236,
|
|
"loss": 4.8639,
|
|
"mean_token_accuracy": 0.2049105256795883,
|
|
"num_tokens": 8732190.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.185487985610962,
|
|
"epoch": 3.9581589958158996,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038305500544853375,
|
|
"loss": 4.8458,
|
|
"mean_token_accuracy": 0.21297864764928817,
|
|
"num_tokens": 8740995.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.216033840179444,
|
|
"epoch": 3.9623430962343096,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038277178478971527,
|
|
"loss": 4.9665,
|
|
"mean_token_accuracy": 0.1954173520207405,
|
|
"num_tokens": 8750768.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.052295589447022,
|
|
"epoch": 3.9665271966527196,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003824883423538905,
|
|
"loss": 4.8918,
|
|
"mean_token_accuracy": 0.2058481052517891,
|
|
"num_tokens": 8760070.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.0752264022827145,
|
|
"epoch": 3.9707112970711296,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038220467872433875,
|
|
"loss": 4.8717,
|
|
"mean_token_accuracy": 0.2065647304058075,
|
|
"num_tokens": 8769313.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.098513650894165,
|
|
"epoch": 3.97489539748954,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003819207944847939,
|
|
"loss": 4.8416,
|
|
"mean_token_accuracy": 0.2117307260632515,
|
|
"num_tokens": 8778328.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.116994905471802,
|
|
"epoch": 3.97907949790795,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038163669021944413,
|
|
"loss": 4.868,
|
|
"mean_token_accuracy": 0.21231727302074432,
|
|
"num_tokens": 8787408.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.121054029464721,
|
|
"epoch": 3.98326359832636,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003813523665129304,
|
|
"loss": 4.8126,
|
|
"mean_token_accuracy": 0.2095250591635704,
|
|
"num_tokens": 8796781.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.096701717376709,
|
|
"epoch": 3.98744769874477,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003810678239503452,
|
|
"loss": 4.8814,
|
|
"mean_token_accuracy": 0.20449732095003129,
|
|
"num_tokens": 8806965.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.084413623809814,
|
|
"epoch": 3.99163179916318,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003807830631172312,
|
|
"loss": 4.8921,
|
|
"mean_token_accuracy": 0.2022484600543976,
|
|
"num_tokens": 8817348.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.101122140884399,
|
|
"epoch": 3.99581589958159,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038049808459958065,
|
|
"loss": 4.9295,
|
|
"mean_token_accuracy": 0.20652856975793837,
|
|
"num_tokens": 8827304.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.091573047637939,
|
|
"epoch": 4.0,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038021288898383343,
|
|
"loss": 4.8331,
|
|
"mean_token_accuracy": 0.2077145606279373,
|
|
"num_tokens": 8837288.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.127749586105347,
|
|
"epoch": 4.00418410041841,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003799274768568761,
|
|
"loss": 4.6296,
|
|
"mean_token_accuracy": 0.22858476340770723,
|
|
"num_tokens": 8846447.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.056386661529541,
|
|
"epoch": 4.00836820083682,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037964184880604125,
|
|
"loss": 4.5441,
|
|
"mean_token_accuracy": 0.23339543342590333,
|
|
"num_tokens": 8855990.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.014510631561279,
|
|
"epoch": 4.01255230125523,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037935600541910523,
|
|
"loss": 4.5512,
|
|
"mean_token_accuracy": 0.22543267756700516,
|
|
"num_tokens": 8864949.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.000003862380981,
|
|
"epoch": 4.01673640167364,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003790699472842879,
|
|
"loss": 4.4939,
|
|
"mean_token_accuracy": 0.23468135446310043,
|
|
"num_tokens": 8873501.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.025322151184082,
|
|
"epoch": 4.02092050209205,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037878367499025086,
|
|
"loss": 4.6044,
|
|
"mean_token_accuracy": 0.22417306303977966,
|
|
"num_tokens": 8882920.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.038575887680054,
|
|
"epoch": 4.02510460251046,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003784971891260965,
|
|
"loss": 4.5254,
|
|
"mean_token_accuracy": 0.22962529212236404,
|
|
"num_tokens": 8893617.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.070232295989991,
|
|
"epoch": 4.02928870292887,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037821049028136664,
|
|
"loss": 4.5512,
|
|
"mean_token_accuracy": 0.2259914293885231,
|
|
"num_tokens": 8902558.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.077103900909424,
|
|
"epoch": 4.03347280334728,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003779235790460414,
|
|
"loss": 4.573,
|
|
"mean_token_accuracy": 0.2270744636654854,
|
|
"num_tokens": 8912477.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.044887256622315,
|
|
"epoch": 4.03765690376569,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003776364560105379,
|
|
"loss": 4.5799,
|
|
"mean_token_accuracy": 0.2272309586405754,
|
|
"num_tokens": 8922313.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 4.988963603973389,
|
|
"epoch": 4.0418410041841,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00037734912176570935,
|
|
"loss": 4.5391,
|
|
"mean_token_accuracy": 0.2340757817029953,
|
|
"num_tokens": 8932011.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.025544023513794,
|
|
"epoch": 4.046025104602511,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037706157690284335,
|
|
"loss": 4.5751,
|
|
"mean_token_accuracy": 0.22412413209676743,
|
|
"num_tokens": 8941022.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.0532307624816895,
|
|
"epoch": 4.050209205020921,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037677382201366103,
|
|
"loss": 4.5941,
|
|
"mean_token_accuracy": 0.22880506813526152,
|
|
"num_tokens": 8950825.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.049771642684936,
|
|
"epoch": 4.054393305439331,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037648585769031555,
|
|
"loss": 4.6131,
|
|
"mean_token_accuracy": 0.21904713809490203,
|
|
"num_tokens": 8961035.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.098345184326172,
|
|
"epoch": 4.058577405857741,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003761976845253915,
|
|
"loss": 4.6507,
|
|
"mean_token_accuracy": 0.2194043755531311,
|
|
"num_tokens": 8970535.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 4.985476875305176,
|
|
"epoch": 4.062761506276151,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003759093031119025,
|
|
"loss": 4.5822,
|
|
"mean_token_accuracy": 0.22069828063249589,
|
|
"num_tokens": 8979404.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.081757593154907,
|
|
"epoch": 4.066945606694561,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037562071404329147,
|
|
"loss": 4.5815,
|
|
"mean_token_accuracy": 0.23617115765810012,
|
|
"num_tokens": 8989164.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.132647085189819,
|
|
"epoch": 4.071129707112971,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037533191791342817,
|
|
"loss": 4.6384,
|
|
"mean_token_accuracy": 0.21816497445106506,
|
|
"num_tokens": 8998630.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.041156482696533,
|
|
"epoch": 4.075313807531381,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037504291531660866,
|
|
"loss": 4.5583,
|
|
"mean_token_accuracy": 0.2322463497519493,
|
|
"num_tokens": 9008235.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.041400671005249,
|
|
"epoch": 4.079497907949791,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003747537068475538,
|
|
"loss": 4.676,
|
|
"mean_token_accuracy": 0.2232330098748207,
|
|
"num_tokens": 9018175.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.117913818359375,
|
|
"epoch": 4.083682008368201,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003744642931014082,
|
|
"loss": 4.6922,
|
|
"mean_token_accuracy": 0.21891210973262787,
|
|
"num_tokens": 9028159.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 4.971291971206665,
|
|
"epoch": 4.087866108786611,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003741746746737386,
|
|
"loss": 4.6022,
|
|
"mean_token_accuracy": 0.2239283248782158,
|
|
"num_tokens": 9038263.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.066917133331299,
|
|
"epoch": 4.092050209205021,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003738848521605333,
|
|
"loss": 4.5974,
|
|
"mean_token_accuracy": 0.22447667866945267,
|
|
"num_tokens": 9048023.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.051163101196289,
|
|
"epoch": 4.096234309623431,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003735948261582005,
|
|
"loss": 4.6578,
|
|
"mean_token_accuracy": 0.21464162617921828,
|
|
"num_tokens": 9056988.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.029681873321533,
|
|
"epoch": 4.100418410041841,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000373304597263567,
|
|
"loss": 4.628,
|
|
"mean_token_accuracy": 0.22242659330368042,
|
|
"num_tokens": 9066595.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.066319227218628,
|
|
"epoch": 4.104602510460251,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037301416607387716,
|
|
"loss": 4.6563,
|
|
"mean_token_accuracy": 0.22113464921712875,
|
|
"num_tokens": 9075034.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 4.993513917922973,
|
|
"epoch": 4.108786610878661,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037272353318679185,
|
|
"loss": 4.5948,
|
|
"mean_token_accuracy": 0.22850108593702317,
|
|
"num_tokens": 9084574.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 4.98863115310669,
|
|
"epoch": 4.112970711297071,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003724326992003867,
|
|
"loss": 4.5698,
|
|
"mean_token_accuracy": 0.22540607750415803,
|
|
"num_tokens": 9093591.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.04891004562378,
|
|
"epoch": 4.117154811715481,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003721416647131512,
|
|
"loss": 4.59,
|
|
"mean_token_accuracy": 0.2303481638431549,
|
|
"num_tokens": 9102439.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 4.949215269088745,
|
|
"epoch": 4.121338912133891,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003718504303239876,
|
|
"loss": 4.4807,
|
|
"mean_token_accuracy": 0.23733771443367005,
|
|
"num_tokens": 9111363.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.053757667541504,
|
|
"epoch": 4.125523012552302,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003715589966322097,
|
|
"loss": 4.6627,
|
|
"mean_token_accuracy": 0.2233867108821869,
|
|
"num_tokens": 9120417.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.102218723297119,
|
|
"epoch": 4.129707112970712,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003712673642375409,
|
|
"loss": 4.6351,
|
|
"mean_token_accuracy": 0.21361656337976456,
|
|
"num_tokens": 9130622.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 4.996326017379761,
|
|
"epoch": 4.133891213389122,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003709755337401138,
|
|
"loss": 4.6061,
|
|
"mean_token_accuracy": 0.2224092349410057,
|
|
"num_tokens": 9139992.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 4.954313564300537,
|
|
"epoch": 4.138075313807532,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00037068350574046867,
|
|
"loss": 4.5954,
|
|
"mean_token_accuracy": 0.2263863727450371,
|
|
"num_tokens": 9149883.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.00340518951416,
|
|
"epoch": 4.142259414225942,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037039128083955227,
|
|
"loss": 4.5567,
|
|
"mean_token_accuracy": 0.23284687846899033,
|
|
"num_tokens": 9159633.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.013108205795288,
|
|
"epoch": 4.146443514644352,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003700988596387163,
|
|
"loss": 4.5665,
|
|
"mean_token_accuracy": 0.22669409215450287,
|
|
"num_tokens": 9167780.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.017022705078125,
|
|
"epoch": 4.150627615062762,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003698062427397168,
|
|
"loss": 4.596,
|
|
"mean_token_accuracy": 0.22676117569208146,
|
|
"num_tokens": 9176808.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.0242228507995605,
|
|
"epoch": 4.154811715481172,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036951343074471217,
|
|
"loss": 4.6018,
|
|
"mean_token_accuracy": 0.22619294226169587,
|
|
"num_tokens": 9186097.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.026171493530273,
|
|
"epoch": 4.158995815899582,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003692204242562624,
|
|
"loss": 4.6249,
|
|
"mean_token_accuracy": 0.21765639185905455,
|
|
"num_tokens": 9194428.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.0025266170501705,
|
|
"epoch": 4.163179916317992,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003689272238773278,
|
|
"loss": 4.6335,
|
|
"mean_token_accuracy": 0.2212685227394104,
|
|
"num_tokens": 9203156.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.00650429725647,
|
|
"epoch": 4.167364016736402,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036863383021126756,
|
|
"loss": 4.6364,
|
|
"mean_token_accuracy": 0.2184078186750412,
|
|
"num_tokens": 9212102.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.009310245513916,
|
|
"epoch": 4.171548117154812,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036834024386183875,
|
|
"loss": 4.6296,
|
|
"mean_token_accuracy": 0.22262282073497772,
|
|
"num_tokens": 9221507.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.03646674156189,
|
|
"epoch": 4.175732217573222,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003680464654331947,
|
|
"loss": 4.5685,
|
|
"mean_token_accuracy": 0.230113685131073,
|
|
"num_tokens": 9230543.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.064976406097412,
|
|
"epoch": 4.179916317991632,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003677524955298844,
|
|
"loss": 4.6704,
|
|
"mean_token_accuracy": 0.2220829501748085,
|
|
"num_tokens": 9240092.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.0096848487854,
|
|
"epoch": 4.184100418410042,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003674583347568505,
|
|
"loss": 4.6337,
|
|
"mean_token_accuracy": 0.2181374341249466,
|
|
"num_tokens": 9250145.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.184100418410042,
|
|
"eval_entropy": 4.8839097281418935,
|
|
"eval_loss": 5.404016494750977,
|
|
"eval_mean_token_accuracy": 0.19186635220952805,
|
|
"eval_num_tokens": 9250145.0,
|
|
"eval_runtime": 2.7597,
|
|
"eval_samples_per_second": 1415.754,
|
|
"eval_steps_per_second": 177.196,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.024817848205567,
|
|
"epoch": 4.188284518828452,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036716398371942866,
|
|
"loss": 4.5989,
|
|
"mean_token_accuracy": 0.22544457763433456,
|
|
"num_tokens": 9259643.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.045625019073486,
|
|
"epoch": 4.192468619246862,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003668694430233459,
|
|
"loss": 4.6235,
|
|
"mean_token_accuracy": 0.22249772399663925,
|
|
"num_tokens": 9268471.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.036555576324463,
|
|
"epoch": 4.196652719665272,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003665747132747198,
|
|
"loss": 4.5753,
|
|
"mean_token_accuracy": 0.22026388943195344,
|
|
"num_tokens": 9276916.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 4.919180202484131,
|
|
"epoch": 4.200836820083682,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036627979508005654,
|
|
"loss": 4.5937,
|
|
"mean_token_accuracy": 0.22548947632312774,
|
|
"num_tokens": 9285867.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.073176574707031,
|
|
"epoch": 4.205020920502092,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003659846890462506,
|
|
"loss": 4.7031,
|
|
"mean_token_accuracy": 0.2204178974032402,
|
|
"num_tokens": 9295044.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.1380829334259035,
|
|
"epoch": 4.209205020920502,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003656893957805826,
|
|
"loss": 4.7088,
|
|
"mean_token_accuracy": 0.22122974395751954,
|
|
"num_tokens": 9303744.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.020355606079102,
|
|
"epoch": 4.2133891213389125,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003653939158907186,
|
|
"loss": 4.5924,
|
|
"mean_token_accuracy": 0.22269472479820251,
|
|
"num_tokens": 9312996.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 4.980009078979492,
|
|
"epoch": 4.2175732217573225,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003650982499847088,
|
|
"loss": 4.6009,
|
|
"mean_token_accuracy": 0.2312599629163742,
|
|
"num_tokens": 9321912.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.018081760406494,
|
|
"epoch": 4.2217573221757325,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000364802398670986,
|
|
"loss": 4.6189,
|
|
"mean_token_accuracy": 0.22419243007898332,
|
|
"num_tokens": 9331798.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 4.992781448364258,
|
|
"epoch": 4.2259414225941425,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036450636255836474,
|
|
"loss": 4.6671,
|
|
"mean_token_accuracy": 0.21717500537633896,
|
|
"num_tokens": 9342064.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.078842115402222,
|
|
"epoch": 4.2301255230125525,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00036421014225603964,
|
|
"loss": 4.5761,
|
|
"mean_token_accuracy": 0.22376060783863067,
|
|
"num_tokens": 9351494.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 4.899067115783692,
|
|
"epoch": 4.2343096234309625,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036391373837358455,
|
|
"loss": 4.5336,
|
|
"mean_token_accuracy": 0.23086315095424653,
|
|
"num_tokens": 9360683.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 4.962994050979614,
|
|
"epoch": 4.2384937238493725,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00036361715152095095,
|
|
"loss": 4.621,
|
|
"mean_token_accuracy": 0.22050356566905976,
|
|
"num_tokens": 9370784.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.07359504699707,
|
|
"epoch": 4.2426778242677825,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003633203823084669,
|
|
"loss": 4.6723,
|
|
"mean_token_accuracy": 0.22397183179855346,
|
|
"num_tokens": 9380673.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.03209867477417,
|
|
"epoch": 4.2468619246861925,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00036302343134683566,
|
|
"loss": 4.6556,
|
|
"mean_token_accuracy": 0.22542175203561782,
|
|
"num_tokens": 9390260.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.101150369644165,
|
|
"epoch": 4.2510460251046025,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003627262992471348,
|
|
"loss": 4.7013,
|
|
"mean_token_accuracy": 0.2191369652748108,
|
|
"num_tokens": 9400643.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 4.979807043075562,
|
|
"epoch": 4.2552301255230125,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003624289866208142,
|
|
"loss": 4.5681,
|
|
"mean_token_accuracy": 0.22356537431478501,
|
|
"num_tokens": 9408898.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 4.990454626083374,
|
|
"epoch": 4.2594142259414225,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036213149407969547,
|
|
"loss": 4.6421,
|
|
"mean_token_accuracy": 0.2180890902876854,
|
|
"num_tokens": 9418024.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.1459509372711185,
|
|
"epoch": 4.2635983263598325,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003618338222359705,
|
|
"loss": 4.7253,
|
|
"mean_token_accuracy": 0.211072239279747,
|
|
"num_tokens": 9427548.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.031407833099365,
|
|
"epoch": 4.2677824267782425,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036153597170220005,
|
|
"loss": 4.5436,
|
|
"mean_token_accuracy": 0.22402570843696595,
|
|
"num_tokens": 9436370.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.033519315719604,
|
|
"epoch": 4.2719665271966525,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003612379430913126,
|
|
"loss": 4.6129,
|
|
"mean_token_accuracy": 0.21965315490961074,
|
|
"num_tokens": 9445506.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.0202953815460205,
|
|
"epoch": 4.2761506276150625,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003609397370166031,
|
|
"loss": 4.547,
|
|
"mean_token_accuracy": 0.23238158524036406,
|
|
"num_tokens": 9455288.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 4.938917922973633,
|
|
"epoch": 4.2803347280334725,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003606413540917319,
|
|
"loss": 4.6767,
|
|
"mean_token_accuracy": 0.22185649573802949,
|
|
"num_tokens": 9464366.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 4.981463098526001,
|
|
"epoch": 4.2845188284518825,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003603427949307228,
|
|
"loss": 4.5957,
|
|
"mean_token_accuracy": 0.2296289473772049,
|
|
"num_tokens": 9474205.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 4.9770636558532715,
|
|
"epoch": 4.2887029288702925,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036004406014796274,
|
|
"loss": 4.5906,
|
|
"mean_token_accuracy": 0.22608524858951567,
|
|
"num_tokens": 9483181.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 4.976647901535034,
|
|
"epoch": 4.292887029288703,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003597451503581998,
|
|
"loss": 4.5965,
|
|
"mean_token_accuracy": 0.22682426422834395,
|
|
"num_tokens": 9492394.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.072396516799927,
|
|
"epoch": 4.297071129707113,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035944606617654236,
|
|
"loss": 4.6818,
|
|
"mean_token_accuracy": 0.21135516166687013,
|
|
"num_tokens": 9502184.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 4.970218467712402,
|
|
"epoch": 4.301255230125523,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00035914680821845746,
|
|
"loss": 4.6521,
|
|
"mean_token_accuracy": 0.22258177250623704,
|
|
"num_tokens": 9511399.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.029709339141846,
|
|
"epoch": 4.305439330543933,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035884737709976995,
|
|
"loss": 4.6562,
|
|
"mean_token_accuracy": 0.227021224796772,
|
|
"num_tokens": 9520710.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 4.999623441696167,
|
|
"epoch": 4.309623430962343,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00035854777343666093,
|
|
"loss": 4.528,
|
|
"mean_token_accuracy": 0.2403809756040573,
|
|
"num_tokens": 9529992.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 4.950124216079712,
|
|
"epoch": 4.313807531380753,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003582479978456665,
|
|
"loss": 4.5511,
|
|
"mean_token_accuracy": 0.23041269630193711,
|
|
"num_tokens": 9538279.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 4.979808139801025,
|
|
"epoch": 4.317991631799163,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003579480509436766,
|
|
"loss": 4.5346,
|
|
"mean_token_accuracy": 0.22957701981067657,
|
|
"num_tokens": 9547425.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 4.970987844467163,
|
|
"epoch": 4.322175732217573,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00035764793334793393,
|
|
"loss": 4.6028,
|
|
"mean_token_accuracy": 0.2294863075017929,
|
|
"num_tokens": 9556657.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.0658666610717775,
|
|
"epoch": 4.326359832635983,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00035734764567603216,
|
|
"loss": 4.6611,
|
|
"mean_token_accuracy": 0.22321824729442596,
|
|
"num_tokens": 9565989.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 4.995517873764038,
|
|
"epoch": 4.330543933054393,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003570471885459149,
|
|
"loss": 4.6182,
|
|
"mean_token_accuracy": 0.22329700589179993,
|
|
"num_tokens": 9575246.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 4.950409412384033,
|
|
"epoch": 4.334728033472803,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003567465625758748,
|
|
"loss": 4.5388,
|
|
"mean_token_accuracy": 0.23305031210184096,
|
|
"num_tokens": 9586185.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 4.993305301666259,
|
|
"epoch": 4.338912133891213,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003564457683845516,
|
|
"loss": 4.5495,
|
|
"mean_token_accuracy": 0.23107281774282457,
|
|
"num_tokens": 9595222.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.056625509262085,
|
|
"epoch": 4.343096234309623,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003561448065909315,
|
|
"loss": 4.6905,
|
|
"mean_token_accuracy": 0.22466468065977097,
|
|
"num_tokens": 9604866.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 4.967604446411133,
|
|
"epoch": 4.347280334728033,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003558436778143455,
|
|
"loss": 4.5297,
|
|
"mean_token_accuracy": 0.2330329164862633,
|
|
"num_tokens": 9614522.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.036429595947266,
|
|
"epoch": 4.351464435146443,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003555423826744682,
|
|
"loss": 4.7085,
|
|
"mean_token_accuracy": 0.21232665479183196,
|
|
"num_tokens": 9623308.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.027107048034668,
|
|
"epoch": 4.355648535564853,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035524092179131645,
|
|
"loss": 4.693,
|
|
"mean_token_accuracy": 0.21849237978458405,
|
|
"num_tokens": 9632887.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 4.985956048965454,
|
|
"epoch": 4.359832635983263,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003549392957852485,
|
|
"loss": 4.6535,
|
|
"mean_token_accuracy": 0.22133904844522476,
|
|
"num_tokens": 9643218.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.01773476600647,
|
|
"epoch": 4.364016736401673,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003546375052769619,
|
|
"loss": 4.6584,
|
|
"mean_token_accuracy": 0.2160348728299141,
|
|
"num_tokens": 9653175.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 4.973669195175171,
|
|
"epoch": 4.368200836820083,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003543355508874932,
|
|
"loss": 4.561,
|
|
"mean_token_accuracy": 0.22843843400478364,
|
|
"num_tokens": 9662431.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.0396506786346436,
|
|
"epoch": 4.372384937238493,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003540334332382159,
|
|
"loss": 4.6049,
|
|
"mean_token_accuracy": 0.22674299478530885,
|
|
"num_tokens": 9671191.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 4.90244722366333,
|
|
"epoch": 4.376569037656903,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003537311529508396,
|
|
"loss": 4.5638,
|
|
"mean_token_accuracy": 0.22337159663438796,
|
|
"num_tokens": 9680637.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 4.976162052154541,
|
|
"epoch": 4.380753138075314,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035342871064740846,
|
|
"loss": 4.5225,
|
|
"mean_token_accuracy": 0.2310200423002243,
|
|
"num_tokens": 9689325.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 4.999366378784179,
|
|
"epoch": 4.384937238493724,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035312610695030017,
|
|
"loss": 4.6383,
|
|
"mean_token_accuracy": 0.2239434614777565,
|
|
"num_tokens": 9698335.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 4.9663115501403805,
|
|
"epoch": 4.389121338912134,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003528233424822245,
|
|
"loss": 4.593,
|
|
"mean_token_accuracy": 0.22254677563905717,
|
|
"num_tokens": 9706647.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.103724622726441,
|
|
"epoch": 4.393305439330544,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000352520417866222,
|
|
"loss": 4.6231,
|
|
"mean_token_accuracy": 0.22940755188465117,
|
|
"num_tokens": 9715941.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 4.981633567810059,
|
|
"epoch": 4.397489539748954,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003522173337256628,
|
|
"loss": 4.6518,
|
|
"mean_token_accuracy": 0.21918769776821137,
|
|
"num_tokens": 9724693.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 4.985882234573364,
|
|
"epoch": 4.401673640167364,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003519140906842455,
|
|
"loss": 4.6694,
|
|
"mean_token_accuracy": 0.21712317317724228,
|
|
"num_tokens": 9733314.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.123903465270996,
|
|
"epoch": 4.405857740585774,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003516106893659953,
|
|
"loss": 4.693,
|
|
"mean_token_accuracy": 0.21850016415119172,
|
|
"num_tokens": 9742844.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 4.985010385513306,
|
|
"epoch": 4.410041841004184,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003513071303952635,
|
|
"loss": 4.5888,
|
|
"mean_token_accuracy": 0.22708119601011276,
|
|
"num_tokens": 9752164.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 4.900617074966431,
|
|
"epoch": 4.414225941422594,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003510034143967256,
|
|
"loss": 4.5845,
|
|
"mean_token_accuracy": 0.22353900372982025,
|
|
"num_tokens": 9761483.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.002999877929687,
|
|
"epoch": 4.418410041841004,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00035069954199538027,
|
|
"loss": 4.5528,
|
|
"mean_token_accuracy": 0.22828604876995087,
|
|
"num_tokens": 9769998.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.072404527664185,
|
|
"epoch": 4.422594142259414,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00035039551381654816,
|
|
"loss": 4.628,
|
|
"mean_token_accuracy": 0.22308734506368638,
|
|
"num_tokens": 9778965.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.040173721313477,
|
|
"epoch": 4.426778242677824,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035009133048587024,
|
|
"loss": 4.6381,
|
|
"mean_token_accuracy": 0.22403745204210282,
|
|
"num_tokens": 9788543.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 4.962851285934448,
|
|
"epoch": 4.430962343096234,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003497869926293069,
|
|
"loss": 4.5158,
|
|
"mean_token_accuracy": 0.22720393687486648,
|
|
"num_tokens": 9797744.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 4.895706510543823,
|
|
"epoch": 4.435146443514644,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003494825008731367,
|
|
"loss": 4.5796,
|
|
"mean_token_accuracy": 0.22817811220884324,
|
|
"num_tokens": 9807528.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 4.910212659835816,
|
|
"epoch": 4.439330543933054,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034917785584395457,
|
|
"loss": 4.6079,
|
|
"mean_token_accuracy": 0.21853889375925065,
|
|
"num_tokens": 9816078.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.032774496078491,
|
|
"epoch": 4.443514644351464,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000348873058168671,
|
|
"loss": 4.5618,
|
|
"mean_token_accuracy": 0.2248278096318245,
|
|
"num_tokens": 9825199.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.07826509475708,
|
|
"epoch": 4.447698744769874,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003485681084745106,
|
|
"loss": 4.7152,
|
|
"mean_token_accuracy": 0.2193989485502243,
|
|
"num_tokens": 9835294.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 4.9364392280578615,
|
|
"epoch": 4.451882845188284,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000348263007389011,
|
|
"loss": 4.5814,
|
|
"mean_token_accuracy": 0.2208610162138939,
|
|
"num_tokens": 9844449.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 4.966502475738525,
|
|
"epoch": 4.456066945606695,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00034795775554002093,
|
|
"loss": 4.6054,
|
|
"mean_token_accuracy": 0.23577258586883545,
|
|
"num_tokens": 9852930.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 4.98392243385315,
|
|
"epoch": 4.460251046025105,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00034765235355569974,
|
|
"loss": 4.5416,
|
|
"mean_token_accuracy": 0.23519645780324935,
|
|
"num_tokens": 9862168.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 4.985634708404541,
|
|
"epoch": 4.464435146443515,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003473468020645155,
|
|
"loss": 4.6888,
|
|
"mean_token_accuracy": 0.22639351189136506,
|
|
"num_tokens": 9871702.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.078781032562256,
|
|
"epoch": 4.468619246861925,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003470411016952442,
|
|
"loss": 4.6932,
|
|
"mean_token_accuracy": 0.2196798101067543,
|
|
"num_tokens": 9880502.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.055650043487549,
|
|
"epoch": 4.472803347280335,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000346735253076968,
|
|
"loss": 4.6389,
|
|
"mean_token_accuracy": 0.22310524582862853,
|
|
"num_tokens": 9889093.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 4.9169941425323485,
|
|
"epoch": 4.476987447698745,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003464292568390743,
|
|
"loss": 4.5337,
|
|
"mean_token_accuracy": 0.23203957229852676,
|
|
"num_tokens": 9897113.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 4.872857284545899,
|
|
"epoch": 4.481171548117155,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003461231136112539,
|
|
"loss": 4.5793,
|
|
"mean_token_accuracy": 0.22614129185676574,
|
|
"num_tokens": 9906478.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.02797155380249,
|
|
"epoch": 4.485355648535565,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034581682402350057,
|
|
"loss": 4.632,
|
|
"mean_token_accuracy": 0.21706513166427613,
|
|
"num_tokens": 9916143.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.045882654190064,
|
|
"epoch": 4.489539748953975,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034551038870610885,
|
|
"loss": 4.6058,
|
|
"mean_token_accuracy": 0.22375112771987915,
|
|
"num_tokens": 9925675.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 4.957107448577881,
|
|
"epoch": 4.493723849372385,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034520380828967346,
|
|
"loss": 4.5428,
|
|
"mean_token_accuracy": 0.23210841417312622,
|
|
"num_tokens": 9934398.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 4.948310041427613,
|
|
"epoch": 4.497907949790795,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003448970834050877,
|
|
"loss": 4.5963,
|
|
"mean_token_accuracy": 0.22628725022077562,
|
|
"num_tokens": 9943066.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 4.866590547561645,
|
|
"epoch": 4.502092050209205,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003445902146835419,
|
|
"loss": 4.4997,
|
|
"mean_token_accuracy": 0.23174125850200653,
|
|
"num_tokens": 9951565.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 4.948875570297242,
|
|
"epoch": 4.506276150627615,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003442832027565227,
|
|
"loss": 4.6335,
|
|
"mean_token_accuracy": 0.22018599212169648,
|
|
"num_tokens": 9960997.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.06408462524414,
|
|
"epoch": 4.510460251046025,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034397604825581115,
|
|
"loss": 4.6398,
|
|
"mean_token_accuracy": 0.22715661227703093,
|
|
"num_tokens": 9970188.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 4.997549295425415,
|
|
"epoch": 4.514644351464435,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003436687518134819,
|
|
"loss": 4.63,
|
|
"mean_token_accuracy": 0.22984076738357545,
|
|
"num_tokens": 9978749.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.036268424987793,
|
|
"epoch": 4.518828451882845,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003433613140619016,
|
|
"loss": 4.6966,
|
|
"mean_token_accuracy": 0.22537703067064285,
|
|
"num_tokens": 9988107.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 4.89153733253479,
|
|
"epoch": 4.523012552301255,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003430537356337277,
|
|
"loss": 4.5219,
|
|
"mean_token_accuracy": 0.22330804765224457,
|
|
"num_tokens": 9996127.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 4.989614391326905,
|
|
"epoch": 4.527196652719665,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003427460171619073,
|
|
"loss": 4.6394,
|
|
"mean_token_accuracy": 0.219285224378109,
|
|
"num_tokens": 10005482.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.041315746307373,
|
|
"epoch": 4.531380753138075,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00034243815927967536,
|
|
"loss": 4.6911,
|
|
"mean_token_accuracy": 0.21671502888202668,
|
|
"num_tokens": 10014765.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.0097565174102785,
|
|
"epoch": 4.535564853556485,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034213016262055396,
|
|
"loss": 4.6227,
|
|
"mean_token_accuracy": 0.21503682881593705,
|
|
"num_tokens": 10022897.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.044774103164673,
|
|
"epoch": 4.539748953974895,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003418220278183507,
|
|
"loss": 4.6959,
|
|
"mean_token_accuracy": 0.2108847975730896,
|
|
"num_tokens": 10032281.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 4.891918468475342,
|
|
"epoch": 4.543933054393305,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034151375550715753,
|
|
"loss": 4.5326,
|
|
"mean_token_accuracy": 0.23642961382865907,
|
|
"num_tokens": 10041194.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 4.897344636917114,
|
|
"epoch": 4.548117154811716,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00034120534632134917,
|
|
"loss": 4.6057,
|
|
"mean_token_accuracy": 0.21950752884149552,
|
|
"num_tokens": 10050062.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.0067027568817135,
|
|
"epoch": 4.552301255230126,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00034089680089558226,
|
|
"loss": 4.6858,
|
|
"mean_token_accuracy": 0.22794269323348998,
|
|
"num_tokens": 10059747.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.0438157081604,
|
|
"epoch": 4.556485355648536,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00034058811986479367,
|
|
"loss": 4.711,
|
|
"mean_token_accuracy": 0.21663242131471633,
|
|
"num_tokens": 10069351.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.003661346435547,
|
|
"epoch": 4.560669456066946,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034027930386419935,
|
|
"loss": 4.6373,
|
|
"mean_token_accuracy": 0.21767364144325257,
|
|
"num_tokens": 10078727.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 4.928648948669434,
|
|
"epoch": 4.564853556485356,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033997035352929285,
|
|
"loss": 4.5883,
|
|
"mean_token_accuracy": 0.22172482460737228,
|
|
"num_tokens": 10088151.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 4.917711782455444,
|
|
"epoch": 4.569037656903766,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003396612694958444,
|
|
"loss": 4.5452,
|
|
"mean_token_accuracy": 0.2294139802455902,
|
|
"num_tokens": 10097311.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 4.9842822551727295,
|
|
"epoch": 4.573221757322176,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033935205239989927,
|
|
"loss": 4.6796,
|
|
"mean_token_accuracy": 0.2146899789571762,
|
|
"num_tokens": 10107474.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.0064507007598875,
|
|
"epoch": 4.577405857740586,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033904270287777645,
|
|
"loss": 4.6116,
|
|
"mean_token_accuracy": 0.23307846784591674,
|
|
"num_tokens": 10116305.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.010985851287842,
|
|
"epoch": 4.581589958158996,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003387332215660675,
|
|
"loss": 4.6489,
|
|
"mean_token_accuracy": 0.22137546241283418,
|
|
"num_tokens": 10125947.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 4.944061613082885,
|
|
"epoch": 4.585774058577406,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00033842360910163536,
|
|
"loss": 4.6351,
|
|
"mean_token_accuracy": 0.22578810006380082,
|
|
"num_tokens": 10134510.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.044047784805298,
|
|
"epoch": 4.589958158995816,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003381138661216125,
|
|
"loss": 4.6945,
|
|
"mean_token_accuracy": 0.217996309697628,
|
|
"num_tokens": 10143429.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.028554058074951,
|
|
"epoch": 4.594142259414226,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033780399326340026,
|
|
"loss": 4.6737,
|
|
"mean_token_accuracy": 0.220823672413826,
|
|
"num_tokens": 10153405.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.046740102767944,
|
|
"epoch": 4.598326359832636,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033749399116466723,
|
|
"loss": 4.6279,
|
|
"mean_token_accuracy": 0.23109050989151,
|
|
"num_tokens": 10162334.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 4.98068675994873,
|
|
"epoch": 4.602510460251046,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003371838604633479,
|
|
"loss": 4.6394,
|
|
"mean_token_accuracy": 0.22259229719638823,
|
|
"num_tokens": 10171604.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.602510460251046,
|
|
"eval_entropy": 4.798924718166422,
|
|
"eval_loss": 5.38128662109375,
|
|
"eval_mean_token_accuracy": 0.19562271357740604,
|
|
"eval_num_tokens": 10171604.0,
|
|
"eval_runtime": 2.7933,
|
|
"eval_samples_per_second": 1398.7,
|
|
"eval_steps_per_second": 175.061,
|
|
"step": 5500
|
|
}
|
|
],
|
|
"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": 1.482828198912e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|