6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.3462204270051933,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692008876800537,
|
|
"epoch": 0.0005770340450086555,
|
|
"grad_norm": 11.4375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8655,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 14330.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691947269439698,
|
|
"epoch": 0.001154068090017311,
|
|
"grad_norm": 11.8125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7846,
|
|
"mean_token_accuracy": 0.0006837096530944109,
|
|
"num_tokens": 26284.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690341567993164,
|
|
"epoch": 0.0017311021350259665,
|
|
"grad_norm": 9.6875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5317,
|
|
"mean_token_accuracy": 0.03542185090482235,
|
|
"num_tokens": 38394.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.66337537765503,
|
|
"epoch": 0.002308136180034622,
|
|
"grad_norm": 5.96875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2064,
|
|
"mean_token_accuracy": 0.0529673807322979,
|
|
"num_tokens": 51341.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.496922397613526,
|
|
"epoch": 0.0028851702250432777,
|
|
"grad_norm": 3.84375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9191,
|
|
"mean_token_accuracy": 0.05405417382717133,
|
|
"num_tokens": 64559.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.455469608306885,
|
|
"epoch": 0.003462204270051933,
|
|
"grad_norm": 3.46875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.8404,
|
|
"mean_token_accuracy": 0.053551160171628,
|
|
"num_tokens": 76688.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.447058010101319,
|
|
"epoch": 0.004039238315060588,
|
|
"grad_norm": 2.9375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.7034,
|
|
"mean_token_accuracy": 0.055309614166617393,
|
|
"num_tokens": 88392.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.431883907318115,
|
|
"epoch": 0.004616272360069244,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.7162,
|
|
"mean_token_accuracy": 0.053266770020127294,
|
|
"num_tokens": 100542.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.33115463256836,
|
|
"epoch": 0.0051933064050779,
|
|
"grad_norm": 3.09375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.5797,
|
|
"mean_token_accuracy": 0.07032700143754482,
|
|
"num_tokens": 113190.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.309079837799072,
|
|
"epoch": 0.005770340450086555,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5553,
|
|
"mean_token_accuracy": 0.0662531953305006,
|
|
"num_tokens": 125554.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.260576152801514,
|
|
"epoch": 0.006347374495095211,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.514,
|
|
"mean_token_accuracy": 0.07373123690485954,
|
|
"num_tokens": 138527.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.316655826568603,
|
|
"epoch": 0.006924408540103866,
|
|
"grad_norm": 2.828125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5893,
|
|
"mean_token_accuracy": 0.06340348795056343,
|
|
"num_tokens": 150404.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.265781021118164,
|
|
"epoch": 0.0075014425851125215,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4827,
|
|
"mean_token_accuracy": 0.0759716846048832,
|
|
"num_tokens": 163743.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.188546752929687,
|
|
"epoch": 0.008078476630121177,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.4526,
|
|
"mean_token_accuracy": 0.07428089305758476,
|
|
"num_tokens": 175657.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.151709461212159,
|
|
"epoch": 0.008655510675129832,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.3966,
|
|
"mean_token_accuracy": 0.07655606269836426,
|
|
"num_tokens": 188756.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.180650806427002,
|
|
"epoch": 0.009232544720138488,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.3212,
|
|
"mean_token_accuracy": 0.07740879319608211,
|
|
"num_tokens": 200548.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.025209522247314,
|
|
"epoch": 0.009809578765147143,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.2239,
|
|
"mean_token_accuracy": 0.08616762235760689,
|
|
"num_tokens": 213307.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.14529628753662,
|
|
"epoch": 0.0103866128101558,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.2001,
|
|
"mean_token_accuracy": 0.07975278049707413,
|
|
"num_tokens": 226863.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.057593822479248,
|
|
"epoch": 0.010963646855164455,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.1455,
|
|
"mean_token_accuracy": 0.08079247176647186,
|
|
"num_tokens": 238655.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 9.97190818786621,
|
|
"epoch": 0.01154068090017311,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.0904,
|
|
"mean_token_accuracy": 0.08307576477527619,
|
|
"num_tokens": 251118.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.988478946685792,
|
|
"epoch": 0.012117714945181766,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.9828,
|
|
"mean_token_accuracy": 0.08939878344535827,
|
|
"num_tokens": 263665.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.928913402557374,
|
|
"epoch": 0.012694748990190421,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.9232,
|
|
"mean_token_accuracy": 0.08705427050590515,
|
|
"num_tokens": 276531.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.803722476959228,
|
|
"epoch": 0.013271783035199077,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.8978,
|
|
"mean_token_accuracy": 0.08507851585745811,
|
|
"num_tokens": 289878.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.810256195068359,
|
|
"epoch": 0.013848817080207732,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8167,
|
|
"mean_token_accuracy": 0.08298168927431107,
|
|
"num_tokens": 302705.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.63891897201538,
|
|
"epoch": 0.014425851125216388,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.6927,
|
|
"mean_token_accuracy": 0.09015745520591736,
|
|
"num_tokens": 314775.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.604006576538087,
|
|
"epoch": 0.015002885170225043,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.7246,
|
|
"mean_token_accuracy": 0.08897400721907615,
|
|
"num_tokens": 328563.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.50387487411499,
|
|
"epoch": 0.015579919215233698,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.5573,
|
|
"mean_token_accuracy": 0.09593394622206688,
|
|
"num_tokens": 341820.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.313443374633788,
|
|
"epoch": 0.016156953260242354,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.5692,
|
|
"mean_token_accuracy": 0.1013730101287365,
|
|
"num_tokens": 354994.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.28393840789795,
|
|
"epoch": 0.01673398730525101,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.4769,
|
|
"mean_token_accuracy": 0.09589103534817696,
|
|
"num_tokens": 365852.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.20916519165039,
|
|
"epoch": 0.017311021350259664,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.4884,
|
|
"mean_token_accuracy": 0.0926565632224083,
|
|
"num_tokens": 379147.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.129295921325683,
|
|
"epoch": 0.01788805539526832,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.472,
|
|
"mean_token_accuracy": 0.08658494129776954,
|
|
"num_tokens": 392257.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.01758165359497,
|
|
"epoch": 0.018465089440276975,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.4153,
|
|
"mean_token_accuracy": 0.08768738433718681,
|
|
"num_tokens": 405249.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.947623825073242,
|
|
"epoch": 0.019042123485285632,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.389,
|
|
"mean_token_accuracy": 0.09221537113189697,
|
|
"num_tokens": 417819.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.845267486572265,
|
|
"epoch": 0.019619157530294286,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.2969,
|
|
"mean_token_accuracy": 0.09195446819067002,
|
|
"num_tokens": 429364.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.706775093078614,
|
|
"epoch": 0.020196191575302943,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.3186,
|
|
"mean_token_accuracy": 0.1003533847630024,
|
|
"num_tokens": 441804.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.740819454193115,
|
|
"epoch": 0.0207732256203116,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.2603,
|
|
"mean_token_accuracy": 0.10172701776027679,
|
|
"num_tokens": 453886.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.652677536010742,
|
|
"epoch": 0.021350259665320254,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.2733,
|
|
"mean_token_accuracy": 0.09830424413084984,
|
|
"num_tokens": 465828.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.614245128631591,
|
|
"epoch": 0.02192729371032891,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.225,
|
|
"mean_token_accuracy": 0.10284605026245117,
|
|
"num_tokens": 478783.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.633440303802491,
|
|
"epoch": 0.022504327755337564,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.2896,
|
|
"mean_token_accuracy": 0.0949929729104042,
|
|
"num_tokens": 491167.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.589673137664795,
|
|
"epoch": 0.02308136180034622,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.2636,
|
|
"mean_token_accuracy": 0.09787702932953835,
|
|
"num_tokens": 503511.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.475285720825195,
|
|
"epoch": 0.023658395845354875,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.1585,
|
|
"mean_token_accuracy": 0.10205093994736672,
|
|
"num_tokens": 515816.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.644243526458741,
|
|
"epoch": 0.024235429890363532,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.2575,
|
|
"mean_token_accuracy": 0.09533519297838211,
|
|
"num_tokens": 528919.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.444662475585938,
|
|
"epoch": 0.024812463935372186,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.1863,
|
|
"mean_token_accuracy": 0.10425740778446198,
|
|
"num_tokens": 540973.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.465065479278564,
|
|
"epoch": 0.025389497980380843,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.1542,
|
|
"mean_token_accuracy": 0.11338075250387192,
|
|
"num_tokens": 553265.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.467651462554931,
|
|
"epoch": 0.025966532025389497,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.2439,
|
|
"mean_token_accuracy": 0.10033949688076974,
|
|
"num_tokens": 567534.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.495089054107666,
|
|
"epoch": 0.026543566070398154,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.1763,
|
|
"mean_token_accuracy": 0.10402834340929985,
|
|
"num_tokens": 579297.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.415564250946044,
|
|
"epoch": 0.027120600115406807,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.1508,
|
|
"mean_token_accuracy": 0.10633992031216621,
|
|
"num_tokens": 592554.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.438064002990723,
|
|
"epoch": 0.027697634160415464,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.1491,
|
|
"mean_token_accuracy": 0.10545074120163918,
|
|
"num_tokens": 605038.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.381492137908936,
|
|
"epoch": 0.02827466820542412,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.163,
|
|
"mean_token_accuracy": 0.1005297139286995,
|
|
"num_tokens": 617172.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.391489028930664,
|
|
"epoch": 0.028851702250432775,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.1113,
|
|
"mean_token_accuracy": 0.1038644127547741,
|
|
"num_tokens": 629243.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.385975456237793,
|
|
"epoch": 0.029428736295441432,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.1635,
|
|
"mean_token_accuracy": 0.10401127263903617,
|
|
"num_tokens": 641487.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.447768592834473,
|
|
"epoch": 0.030005770340450086,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.1661,
|
|
"mean_token_accuracy": 0.09748540446162224,
|
|
"num_tokens": 653214.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.27870225906372,
|
|
"epoch": 0.030582804385458743,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.1277,
|
|
"mean_token_accuracy": 0.11644875779747962,
|
|
"num_tokens": 666549.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.414948844909668,
|
|
"epoch": 0.031159838430467397,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.1369,
|
|
"mean_token_accuracy": 0.1066818192601204,
|
|
"num_tokens": 679443.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.385531330108643,
|
|
"epoch": 0.031736872475476054,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.1374,
|
|
"mean_token_accuracy": 0.1021499790251255,
|
|
"num_tokens": 692929.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.398066520690918,
|
|
"epoch": 0.03231390652048471,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.0569,
|
|
"mean_token_accuracy": 0.10605769827961922,
|
|
"num_tokens": 705605.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.333725929260254,
|
|
"epoch": 0.03289094056549336,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.1521,
|
|
"mean_token_accuracy": 0.09875996261835099,
|
|
"num_tokens": 719064.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.305869674682617,
|
|
"epoch": 0.03346797461050202,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.0664,
|
|
"mean_token_accuracy": 0.10697638019919395,
|
|
"num_tokens": 732572.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.336920833587646,
|
|
"epoch": 0.034045008655510675,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.0627,
|
|
"mean_token_accuracy": 0.10701068118214607,
|
|
"num_tokens": 745161.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.26529426574707,
|
|
"epoch": 0.03462204270051933,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.0361,
|
|
"mean_token_accuracy": 0.11285013034939766,
|
|
"num_tokens": 756180.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.337069225311279,
|
|
"epoch": 0.03519907674552799,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.0545,
|
|
"mean_token_accuracy": 0.11181231290102005,
|
|
"num_tokens": 768568.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.28462495803833,
|
|
"epoch": 0.03577611079053664,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.0859,
|
|
"mean_token_accuracy": 0.11087586879730224,
|
|
"num_tokens": 781580.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.21790370941162,
|
|
"epoch": 0.0363531448355453,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.9589,
|
|
"mean_token_accuracy": 0.12124751061201096,
|
|
"num_tokens": 793486.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.240262222290038,
|
|
"epoch": 0.03693017888055395,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.0306,
|
|
"mean_token_accuracy": 0.11093810424208642,
|
|
"num_tokens": 805997.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.202302837371827,
|
|
"epoch": 0.03750721292556261,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.9966,
|
|
"mean_token_accuracy": 0.11470017209649086,
|
|
"num_tokens": 817681.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.279125785827636,
|
|
"epoch": 0.038084246970571264,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.0151,
|
|
"mean_token_accuracy": 0.1159965842962265,
|
|
"num_tokens": 830027.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.221384429931641,
|
|
"epoch": 0.03866128101557992,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.0353,
|
|
"mean_token_accuracy": 0.11259357258677483,
|
|
"num_tokens": 842862.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.216114330291749,
|
|
"epoch": 0.03923831506058857,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.9479,
|
|
"mean_token_accuracy": 0.11576894521713257,
|
|
"num_tokens": 856548.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.159035778045654,
|
|
"epoch": 0.03981534910559723,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.9681,
|
|
"mean_token_accuracy": 0.11400512978434563,
|
|
"num_tokens": 869651.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.185454750061036,
|
|
"epoch": 0.040392383150605886,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.9938,
|
|
"mean_token_accuracy": 0.1144215151667595,
|
|
"num_tokens": 880717.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.218715572357178,
|
|
"epoch": 0.04096941719561454,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.9851,
|
|
"mean_token_accuracy": 0.11426586061716079,
|
|
"num_tokens": 894066.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.077167415618897,
|
|
"epoch": 0.0415464512406232,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.8762,
|
|
"mean_token_accuracy": 0.11363315507769585,
|
|
"num_tokens": 905287.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.201995038986206,
|
|
"epoch": 0.042123485285631854,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.9672,
|
|
"mean_token_accuracy": 0.1200343832373619,
|
|
"num_tokens": 917512.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.16402292251587,
|
|
"epoch": 0.04270051933064051,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.9206,
|
|
"mean_token_accuracy": 0.11295781657099724,
|
|
"num_tokens": 930263.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.1646258354187,
|
|
"epoch": 0.04327755337564916,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.9761,
|
|
"mean_token_accuracy": 0.11365053132176399,
|
|
"num_tokens": 942330.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.16142988204956,
|
|
"epoch": 0.04385458742065782,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.9558,
|
|
"mean_token_accuracy": 0.11354472935199737,
|
|
"num_tokens": 954005.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.192191314697265,
|
|
"epoch": 0.044431621465666475,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.9435,
|
|
"mean_token_accuracy": 0.11344988718628883,
|
|
"num_tokens": 966725.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.170158672332764,
|
|
"epoch": 0.04500865551067513,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.9346,
|
|
"mean_token_accuracy": 0.11046166941523552,
|
|
"num_tokens": 978021.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.039873075485229,
|
|
"epoch": 0.04558568955568378,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.9217,
|
|
"mean_token_accuracy": 0.11965367197990417,
|
|
"num_tokens": 991321.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.160380363464355,
|
|
"epoch": 0.04616272360069244,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.9176,
|
|
"mean_token_accuracy": 0.11871105283498765,
|
|
"num_tokens": 1004172.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.075436639785767,
|
|
"epoch": 0.0467397576457011,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.8652,
|
|
"mean_token_accuracy": 0.12333920821547509,
|
|
"num_tokens": 1016513.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.040183401107788,
|
|
"epoch": 0.04731679169070975,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.867,
|
|
"mean_token_accuracy": 0.11452069729566575,
|
|
"num_tokens": 1028536.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.143816423416137,
|
|
"epoch": 0.04789382573571841,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.7972,
|
|
"mean_token_accuracy": 0.12931199073791505,
|
|
"num_tokens": 1040534.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.921964454650879,
|
|
"epoch": 0.048470859780727064,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.9327,
|
|
"mean_token_accuracy": 0.1181611604988575,
|
|
"num_tokens": 1054995.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.17293095588684,
|
|
"epoch": 0.04904789382573572,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.9133,
|
|
"mean_token_accuracy": 0.11562721282243729,
|
|
"num_tokens": 1067722.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.100650596618653,
|
|
"epoch": 0.04962492787074437,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.8862,
|
|
"mean_token_accuracy": 0.111819938570261,
|
|
"num_tokens": 1079788.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.100652170181274,
|
|
"epoch": 0.05020196191575303,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.885,
|
|
"mean_token_accuracy": 0.11935903802514077,
|
|
"num_tokens": 1092016.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.106025123596192,
|
|
"epoch": 0.050778995960761686,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.901,
|
|
"mean_token_accuracy": 0.11573071926832199,
|
|
"num_tokens": 1104522.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.052202367782593,
|
|
"epoch": 0.05135603000577034,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.9235,
|
|
"mean_token_accuracy": 0.12538544684648514,
|
|
"num_tokens": 1117239.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.048323249816894,
|
|
"epoch": 0.05193306405077899,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.8395,
|
|
"mean_token_accuracy": 0.12211849763989449,
|
|
"num_tokens": 1129061.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.052234315872193,
|
|
"epoch": 0.052510098095787654,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.828,
|
|
"mean_token_accuracy": 0.1126674197614193,
|
|
"num_tokens": 1141409.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.968437433242798,
|
|
"epoch": 0.05308713214079631,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.8315,
|
|
"mean_token_accuracy": 0.11563115417957306,
|
|
"num_tokens": 1153623.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.061407089233398,
|
|
"epoch": 0.05366416618580496,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.8652,
|
|
"mean_token_accuracy": 0.11920796036720276,
|
|
"num_tokens": 1166877.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.019161272048951,
|
|
"epoch": 0.054241200230813615,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.8554,
|
|
"mean_token_accuracy": 0.11903531998395919,
|
|
"num_tokens": 1179625.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.995475769042969,
|
|
"epoch": 0.054818234275822275,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.8848,
|
|
"mean_token_accuracy": 0.11980156600475311,
|
|
"num_tokens": 1193586.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.04285216331482,
|
|
"epoch": 0.05539526832083093,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.8237,
|
|
"mean_token_accuracy": 0.12064610123634338,
|
|
"num_tokens": 1207086.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.9670654296875,
|
|
"epoch": 0.05597230236583958,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.8129,
|
|
"mean_token_accuracy": 0.12205322310328484,
|
|
"num_tokens": 1219389.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.9849029064178465,
|
|
"epoch": 0.05654933641084824,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.7593,
|
|
"mean_token_accuracy": 0.12576186433434486,
|
|
"num_tokens": 1232080.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.957043218612671,
|
|
"epoch": 0.0571263704558569,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.8116,
|
|
"mean_token_accuracy": 0.1276965133845806,
|
|
"num_tokens": 1246775.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.927796268463135,
|
|
"epoch": 0.05770340450086555,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.8609,
|
|
"mean_token_accuracy": 0.11414324194192886,
|
|
"num_tokens": 1258839.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.979376649856567,
|
|
"epoch": 0.058280438545874204,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.7701,
|
|
"mean_token_accuracy": 0.12441443428397178,
|
|
"num_tokens": 1270710.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.965421342849732,
|
|
"epoch": 0.058857472590882864,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.7192,
|
|
"mean_token_accuracy": 0.12516708746552468,
|
|
"num_tokens": 1281930.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.895893859863281,
|
|
"epoch": 0.05943450663589152,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.8013,
|
|
"mean_token_accuracy": 0.11694373786449433,
|
|
"num_tokens": 1294548.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.908454704284668,
|
|
"epoch": 0.06001154068090017,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.7466,
|
|
"mean_token_accuracy": 0.11997460946440697,
|
|
"num_tokens": 1307588.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.922945499420166,
|
|
"epoch": 0.060588574725908825,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.7252,
|
|
"mean_token_accuracy": 0.12717078104615212,
|
|
"num_tokens": 1319478.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.9428062915802,
|
|
"epoch": 0.061165608770917486,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.8123,
|
|
"mean_token_accuracy": 0.11468452289700508,
|
|
"num_tokens": 1332950.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.8876683712005615,
|
|
"epoch": 0.06174264281592614,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.6891,
|
|
"mean_token_accuracy": 0.12758101224899293,
|
|
"num_tokens": 1345510.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.72719554901123,
|
|
"epoch": 0.06231967686093479,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.7493,
|
|
"mean_token_accuracy": 0.1305077627301216,
|
|
"num_tokens": 1359388.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.881310319900512,
|
|
"epoch": 0.06289671090594345,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.7725,
|
|
"mean_token_accuracy": 0.1230618417263031,
|
|
"num_tokens": 1373067.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.837239837646484,
|
|
"epoch": 0.06347374495095211,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.7374,
|
|
"mean_token_accuracy": 0.1261225961148739,
|
|
"num_tokens": 1385718.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.000623750686646,
|
|
"epoch": 0.06405077899596076,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.7379,
|
|
"mean_token_accuracy": 0.12663036063313485,
|
|
"num_tokens": 1398788.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.7889786720275875,
|
|
"epoch": 0.06462781304096941,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.7291,
|
|
"mean_token_accuracy": 0.13189933598041534,
|
|
"num_tokens": 1411752.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.891875267028809,
|
|
"epoch": 0.06520484708597807,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.7763,
|
|
"mean_token_accuracy": 0.12392229437828065,
|
|
"num_tokens": 1423979.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.946758174896241,
|
|
"epoch": 0.06578188113098672,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.828,
|
|
"mean_token_accuracy": 0.1189330443739891,
|
|
"num_tokens": 1437009.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.795953941345215,
|
|
"epoch": 0.06635891517599539,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.6487,
|
|
"mean_token_accuracy": 0.12712273001670837,
|
|
"num_tokens": 1448220.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.787666893005371,
|
|
"epoch": 0.06693594922100404,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.6992,
|
|
"mean_token_accuracy": 0.12536668553948402,
|
|
"num_tokens": 1460024.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.011966657638549,
|
|
"epoch": 0.0675129832660127,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.7021,
|
|
"mean_token_accuracy": 0.11621926128864288,
|
|
"num_tokens": 1472283.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.664339017868042,
|
|
"epoch": 0.06809001731102135,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.7415,
|
|
"mean_token_accuracy": 0.12770119607448577,
|
|
"num_tokens": 1486565.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.810535287857055,
|
|
"epoch": 0.06866705135603,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.7145,
|
|
"mean_token_accuracy": 0.1263989932835102,
|
|
"num_tokens": 1498708.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.804998970031738,
|
|
"epoch": 0.06924408540103866,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.6492,
|
|
"mean_token_accuracy": 0.1294543094933033,
|
|
"num_tokens": 1511318.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.817861795425415,
|
|
"epoch": 0.06982111944604731,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.6533,
|
|
"mean_token_accuracy": 0.13287195861339568,
|
|
"num_tokens": 1525415.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.86554970741272,
|
|
"epoch": 0.07039815349105598,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.6892,
|
|
"mean_token_accuracy": 0.12285038828849792,
|
|
"num_tokens": 1538757.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.643300628662109,
|
|
"epoch": 0.07097518753606463,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.671,
|
|
"mean_token_accuracy": 0.12466116324067116,
|
|
"num_tokens": 1550619.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.822739982604981,
|
|
"epoch": 0.07155222158107329,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.7067,
|
|
"mean_token_accuracy": 0.12112424820661545,
|
|
"num_tokens": 1562588.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.776826238632202,
|
|
"epoch": 0.07212925562608194,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.6167,
|
|
"mean_token_accuracy": 0.1296200931072235,
|
|
"num_tokens": 1574892.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.708193683624268,
|
|
"epoch": 0.0727062896710906,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.6351,
|
|
"mean_token_accuracy": 0.1296207919716835,
|
|
"num_tokens": 1587525.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.8489649295806885,
|
|
"epoch": 0.07328332371609925,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.6498,
|
|
"mean_token_accuracy": 0.11944840252399444,
|
|
"num_tokens": 1600350.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.635661983489991,
|
|
"epoch": 0.0738603577611079,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.6298,
|
|
"mean_token_accuracy": 0.13184600844979286,
|
|
"num_tokens": 1613318.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.869474792480469,
|
|
"epoch": 0.07443739180611655,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.7555,
|
|
"mean_token_accuracy": 0.11981712207198143,
|
|
"num_tokens": 1626757.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.726322650909424,
|
|
"epoch": 0.07501442585112522,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.5653,
|
|
"mean_token_accuracy": 0.1351322740316391,
|
|
"num_tokens": 1639185.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.705139207839966,
|
|
"epoch": 0.07559145989613388,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.6865,
|
|
"mean_token_accuracy": 0.12470589056611062,
|
|
"num_tokens": 1652009.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.701083326339722,
|
|
"epoch": 0.07616849394114253,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.6647,
|
|
"mean_token_accuracy": 0.12210506573319435,
|
|
"num_tokens": 1665604.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.796258974075317,
|
|
"epoch": 0.07674552798615118,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.6319,
|
|
"mean_token_accuracy": 0.12362316846847535,
|
|
"num_tokens": 1679036.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.649466943740845,
|
|
"epoch": 0.07732256203115984,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.6317,
|
|
"mean_token_accuracy": 0.12975930497050286,
|
|
"num_tokens": 1691512.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.84274697303772,
|
|
"epoch": 0.07789959607616849,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.6818,
|
|
"mean_token_accuracy": 0.12658526599407197,
|
|
"num_tokens": 1704589.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.613417482376098,
|
|
"epoch": 0.07847663012117714,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.6798,
|
|
"mean_token_accuracy": 0.11807645931839943,
|
|
"num_tokens": 1718612.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.623522710800171,
|
|
"epoch": 0.07905366416618581,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.5748,
|
|
"mean_token_accuracy": 0.12185449972748756,
|
|
"num_tokens": 1730472.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.759632921218872,
|
|
"epoch": 0.07963069821119446,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.7018,
|
|
"mean_token_accuracy": 0.12128116115927697,
|
|
"num_tokens": 1742535.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.77196855545044,
|
|
"epoch": 0.08020773225620312,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.6766,
|
|
"mean_token_accuracy": 0.11788606792688369,
|
|
"num_tokens": 1753877.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.763478708267212,
|
|
"epoch": 0.08078476630121177,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.5707,
|
|
"mean_token_accuracy": 0.12885044515132904,
|
|
"num_tokens": 1766413.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.513013744354248,
|
|
"epoch": 0.08136180034622043,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.4623,
|
|
"mean_token_accuracy": 0.13055645897984505,
|
|
"num_tokens": 1779719.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.673223543167114,
|
|
"epoch": 0.08193883439122908,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.5946,
|
|
"mean_token_accuracy": 0.12953749522566796,
|
|
"num_tokens": 1791448.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.66050238609314,
|
|
"epoch": 0.08251586843623773,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.603,
|
|
"mean_token_accuracy": 0.12697791904211045,
|
|
"num_tokens": 1803802.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.657990217208862,
|
|
"epoch": 0.0830929024812464,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.6079,
|
|
"mean_token_accuracy": 0.12700373977422713,
|
|
"num_tokens": 1817545.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.667493295669556,
|
|
"epoch": 0.08366993652625505,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.5869,
|
|
"mean_token_accuracy": 0.12516693994402886,
|
|
"num_tokens": 1830115.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.611949396133423,
|
|
"epoch": 0.08424697057126371,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.5303,
|
|
"mean_token_accuracy": 0.1249528095126152,
|
|
"num_tokens": 1844707.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.6815104484558105,
|
|
"epoch": 0.08482400461627236,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.6093,
|
|
"mean_token_accuracy": 0.12226399332284928,
|
|
"num_tokens": 1858719.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.659861135482788,
|
|
"epoch": 0.08540103866128101,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.5551,
|
|
"mean_token_accuracy": 0.1302394263446331,
|
|
"num_tokens": 1870421.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.612351417541504,
|
|
"epoch": 0.08597807270628967,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.6212,
|
|
"mean_token_accuracy": 0.11757046580314637,
|
|
"num_tokens": 1883792.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.606675052642823,
|
|
"epoch": 0.08655510675129832,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.598,
|
|
"mean_token_accuracy": 0.12712156772613525,
|
|
"num_tokens": 1896209.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.588115072250366,
|
|
"epoch": 0.08713214079630698,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.5331,
|
|
"mean_token_accuracy": 0.12872409150004388,
|
|
"num_tokens": 1908520.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.640614938735962,
|
|
"epoch": 0.08770917484131564,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.5346,
|
|
"mean_token_accuracy": 0.12833384424448013,
|
|
"num_tokens": 1922489.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.590028858184814,
|
|
"epoch": 0.0882862088863243,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.5305,
|
|
"mean_token_accuracy": 0.12443057373166085,
|
|
"num_tokens": 1935206.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.588243770599365,
|
|
"epoch": 0.08886324293133295,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.5068,
|
|
"mean_token_accuracy": 0.1315755218267441,
|
|
"num_tokens": 1948798.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.5549592018127445,
|
|
"epoch": 0.0894402769763416,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.4928,
|
|
"mean_token_accuracy": 0.13577663153409958,
|
|
"num_tokens": 1961095.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.596438980102539,
|
|
"epoch": 0.09001731102135026,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.5125,
|
|
"mean_token_accuracy": 0.1260148160159588,
|
|
"num_tokens": 1974229.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.607128286361695,
|
|
"epoch": 0.09059434506635891,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.602,
|
|
"mean_token_accuracy": 0.12650037407875062,
|
|
"num_tokens": 1986334.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.584266185760498,
|
|
"epoch": 0.09117137911136756,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.5522,
|
|
"mean_token_accuracy": 0.1282362699508667,
|
|
"num_tokens": 2000337.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.498491811752319,
|
|
"epoch": 0.09174841315637623,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.4676,
|
|
"mean_token_accuracy": 0.13595186024904252,
|
|
"num_tokens": 2012048.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.588355207443238,
|
|
"epoch": 0.09232544720138489,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.497,
|
|
"mean_token_accuracy": 0.13125768825411796,
|
|
"num_tokens": 2025244.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.5243137836456295,
|
|
"epoch": 0.09290248124639354,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.5419,
|
|
"mean_token_accuracy": 0.13079462423920632,
|
|
"num_tokens": 2036542.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.59242262840271,
|
|
"epoch": 0.0934795152914022,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.484,
|
|
"mean_token_accuracy": 0.13099602833390236,
|
|
"num_tokens": 2049358.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.438653898239136,
|
|
"epoch": 0.09405654933641085,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.4969,
|
|
"mean_token_accuracy": 0.13118922114372253,
|
|
"num_tokens": 2062114.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.643336772918701,
|
|
"epoch": 0.0946335833814195,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.4629,
|
|
"mean_token_accuracy": 0.1343265689909458,
|
|
"num_tokens": 2074236.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.525158262252807,
|
|
"epoch": 0.09521061742642815,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.3893,
|
|
"mean_token_accuracy": 0.135568980127573,
|
|
"num_tokens": 2086320.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.457302284240723,
|
|
"epoch": 0.09578765147143682,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.4473,
|
|
"mean_token_accuracy": 0.13218377977609636,
|
|
"num_tokens": 2099119.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.532028770446777,
|
|
"epoch": 0.09636468551644548,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.4827,
|
|
"mean_token_accuracy": 0.13144656643271446,
|
|
"num_tokens": 2111795.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.413727283477783,
|
|
"epoch": 0.09694171956145413,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.4238,
|
|
"mean_token_accuracy": 0.12527592554688455,
|
|
"num_tokens": 2126621.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.471304702758789,
|
|
"epoch": 0.09751875360646278,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.3631,
|
|
"mean_token_accuracy": 0.14259372502565384,
|
|
"num_tokens": 2139770.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.440826320648194,
|
|
"epoch": 0.09809578765147144,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.4611,
|
|
"mean_token_accuracy": 0.13174420446157456,
|
|
"num_tokens": 2151683.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.430302906036377,
|
|
"epoch": 0.09867282169648009,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.421,
|
|
"mean_token_accuracy": 0.13163666427135468,
|
|
"num_tokens": 2163178.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.475621461868286,
|
|
"epoch": 0.09924985574148874,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.5021,
|
|
"mean_token_accuracy": 0.1217731073498726,
|
|
"num_tokens": 2176312.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.50553822517395,
|
|
"epoch": 0.0998268897864974,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.4503,
|
|
"mean_token_accuracy": 0.12792299836874008,
|
|
"num_tokens": 2189045.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.380361652374267,
|
|
"epoch": 0.10040392383150606,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.4504,
|
|
"mean_token_accuracy": 0.1204420991241932,
|
|
"num_tokens": 2202083.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.474799299240113,
|
|
"epoch": 0.10098095787651472,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.4886,
|
|
"mean_token_accuracy": 0.12477894797921181,
|
|
"num_tokens": 2215775.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.536693000793457,
|
|
"epoch": 0.10155799192152337,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.4308,
|
|
"mean_token_accuracy": 0.12408801987767219,
|
|
"num_tokens": 2228387.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.429737758636475,
|
|
"epoch": 0.10213502596653203,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.442,
|
|
"mean_token_accuracy": 0.13106903955340385,
|
|
"num_tokens": 2241491.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.284702110290527,
|
|
"epoch": 0.10271206001154068,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.3722,
|
|
"mean_token_accuracy": 0.13908286914229392,
|
|
"num_tokens": 2255282.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.476779127120972,
|
|
"epoch": 0.10328909405654933,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.3868,
|
|
"mean_token_accuracy": 0.1294294334948063,
|
|
"num_tokens": 2268081.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.410451793670655,
|
|
"epoch": 0.10386612810155799,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.3463,
|
|
"mean_token_accuracy": 0.1367862269282341,
|
|
"num_tokens": 2281069.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.424945640563965,
|
|
"epoch": 0.10444316214656665,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.4009,
|
|
"mean_token_accuracy": 0.12864845171570777,
|
|
"num_tokens": 2293606.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.475167083740234,
|
|
"epoch": 0.10502019619157531,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.3905,
|
|
"mean_token_accuracy": 0.13015836849808693,
|
|
"num_tokens": 2304973.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.403925657272339,
|
|
"epoch": 0.10559723023658396,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.426,
|
|
"mean_token_accuracy": 0.12412591949105263,
|
|
"num_tokens": 2319016.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.38353796005249,
|
|
"epoch": 0.10617426428159261,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.402,
|
|
"mean_token_accuracy": 0.1354122206568718,
|
|
"num_tokens": 2330599.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.44637622833252,
|
|
"epoch": 0.10675129832660127,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.4054,
|
|
"mean_token_accuracy": 0.12546970024704934,
|
|
"num_tokens": 2343153.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.42939600944519,
|
|
"epoch": 0.10732833237160992,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.4317,
|
|
"mean_token_accuracy": 0.12410960122942924,
|
|
"num_tokens": 2356762.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.285542058944702,
|
|
"epoch": 0.10790536641661858,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.4542,
|
|
"mean_token_accuracy": 0.1292865566909313,
|
|
"num_tokens": 2369581.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.475341939926148,
|
|
"epoch": 0.10848240046162723,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.3583,
|
|
"mean_token_accuracy": 0.12862989977002143,
|
|
"num_tokens": 2381419.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.332505321502685,
|
|
"epoch": 0.1090594345066359,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.3724,
|
|
"mean_token_accuracy": 0.13676412105560304,
|
|
"num_tokens": 2393586.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.300458240509033,
|
|
"epoch": 0.10963646855164455,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.36,
|
|
"mean_token_accuracy": 0.1377156250178814,
|
|
"num_tokens": 2407720.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.401245880126953,
|
|
"epoch": 0.1102135025966532,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.4057,
|
|
"mean_token_accuracy": 0.12623452842235566,
|
|
"num_tokens": 2421089.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.3679827690124515,
|
|
"epoch": 0.11079053664166186,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.3399,
|
|
"mean_token_accuracy": 0.1357784502208233,
|
|
"num_tokens": 2432859.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.317124748229981,
|
|
"epoch": 0.11136757068667051,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.3411,
|
|
"mean_token_accuracy": 0.12867245748639106,
|
|
"num_tokens": 2445072.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.359953451156616,
|
|
"epoch": 0.11194460473167916,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.3532,
|
|
"mean_token_accuracy": 0.13644537180662156,
|
|
"num_tokens": 2458724.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.275381755828858,
|
|
"epoch": 0.11252163877668782,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.4164,
|
|
"mean_token_accuracy": 0.12978286817669868,
|
|
"num_tokens": 2472976.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.473965215682983,
|
|
"epoch": 0.11309867282169649,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.371,
|
|
"mean_token_accuracy": 0.12792500779032706,
|
|
"num_tokens": 2486164.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.275361776351929,
|
|
"epoch": 0.11367570686670514,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.3328,
|
|
"mean_token_accuracy": 0.13169763311743737,
|
|
"num_tokens": 2499395.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.323497009277344,
|
|
"epoch": 0.1142527409117138,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.3271,
|
|
"mean_token_accuracy": 0.13341858312487603,
|
|
"num_tokens": 2512163.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.386171340942383,
|
|
"epoch": 0.11482977495672245,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.3263,
|
|
"mean_token_accuracy": 0.12971674054861068,
|
|
"num_tokens": 2523634.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.3099339485168455,
|
|
"epoch": 0.1154068090017311,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.362,
|
|
"mean_token_accuracy": 0.135257351398468,
|
|
"num_tokens": 2536237.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.379648447036743,
|
|
"epoch": 0.11598384304673975,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999999975783157,
|
|
"loss": 7.3765,
|
|
"mean_token_accuracy": 0.12056859806180001,
|
|
"num_tokens": 2548035.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.317667388916016,
|
|
"epoch": 0.11656087709174841,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999999877402236,
|
|
"loss": 7.3159,
|
|
"mean_token_accuracy": 0.13096051216125487,
|
|
"num_tokens": 2561670.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.273052406311035,
|
|
"epoch": 0.11713791113675708,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999999703343686,
|
|
"loss": 7.3317,
|
|
"mean_token_accuracy": 0.13321302831172943,
|
|
"num_tokens": 2573471.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.3107616901397705,
|
|
"epoch": 0.11771494518176573,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999999453607515,
|
|
"loss": 7.282,
|
|
"mean_token_accuracy": 0.13478338941931725,
|
|
"num_tokens": 2586534.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.346148490905762,
|
|
"epoch": 0.11829197922677438,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999999128193729,
|
|
"loss": 7.3461,
|
|
"mean_token_accuracy": 0.12729120627045631,
|
|
"num_tokens": 2597970.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.172978067398072,
|
|
"epoch": 0.11886901327178304,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000499999872710234,
|
|
"loss": 7.2802,
|
|
"mean_token_accuracy": 0.13685771971940994,
|
|
"num_tokens": 2610924.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.363209962844849,
|
|
"epoch": 0.11944604731679169,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999998250333361,
|
|
"loss": 7.3474,
|
|
"mean_token_accuracy": 0.13248208165168762,
|
|
"num_tokens": 2623571.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.261402225494384,
|
|
"epoch": 0.12002308136180034,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499999769788681,
|
|
"loss": 7.3617,
|
|
"mean_token_accuracy": 0.1302515096962452,
|
|
"num_tokens": 2636063.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.279189252853394,
|
|
"epoch": 0.120600115406809,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999997069762703,
|
|
"loss": 7.3021,
|
|
"mean_token_accuracy": 0.13231852501630784,
|
|
"num_tokens": 2648621.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.2357665538787845,
|
|
"epoch": 0.12117714945181765,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999996365961062,
|
|
"loss": 7.2375,
|
|
"mean_token_accuracy": 0.13813495635986328,
|
|
"num_tokens": 2661794.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.320775318145752,
|
|
"epoch": 0.12175418349682632,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999995586481912,
|
|
"loss": 7.3105,
|
|
"mean_token_accuracy": 0.12649577111005783,
|
|
"num_tokens": 2674048.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.145959424972534,
|
|
"epoch": 0.12233121754183497,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999994731325276,
|
|
"loss": 7.2144,
|
|
"mean_token_accuracy": 0.13311017528176308,
|
|
"num_tokens": 2685510.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.320645141601562,
|
|
"epoch": 0.12290825158684363,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999993800491187,
|
|
"loss": 7.3029,
|
|
"mean_token_accuracy": 0.13155850619077683,
|
|
"num_tokens": 2698414.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.2215817928314205,
|
|
"epoch": 0.12348528563185228,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999992793979672,
|
|
"loss": 7.301,
|
|
"mean_token_accuracy": 0.13250066041946412,
|
|
"num_tokens": 2711343.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.241006183624267,
|
|
"epoch": 0.12406231967686093,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999991711790769,
|
|
"loss": 7.1878,
|
|
"mean_token_accuracy": 0.14295068979263306,
|
|
"num_tokens": 2723123.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.133680200576782,
|
|
"epoch": 0.12463935372186959,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999990553924511,
|
|
"loss": 7.2722,
|
|
"mean_token_accuracy": 0.1344312846660614,
|
|
"num_tokens": 2736015.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.196067523956299,
|
|
"epoch": 0.12521638776687824,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999989320380939,
|
|
"loss": 7.221,
|
|
"mean_token_accuracy": 0.1397103875875473,
|
|
"num_tokens": 2747561.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.292922401428223,
|
|
"epoch": 0.1257934218118869,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999988011160094,
|
|
"loss": 7.2605,
|
|
"mean_token_accuracy": 0.1328883707523346,
|
|
"num_tokens": 2758992.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.279149007797241,
|
|
"epoch": 0.12637045585689555,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499998662626202,
|
|
"loss": 7.2514,
|
|
"mean_token_accuracy": 0.13322949036955833,
|
|
"num_tokens": 2771164.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.044620227813721,
|
|
"epoch": 0.12694748990190421,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999985165686764,
|
|
"loss": 7.2315,
|
|
"mean_token_accuracy": 0.1337655045092106,
|
|
"num_tokens": 2784400.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.231164073944091,
|
|
"epoch": 0.12752452394691285,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999983629434373,
|
|
"loss": 7.1957,
|
|
"mean_token_accuracy": 0.13645816892385482,
|
|
"num_tokens": 2795729.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.156470489501953,
|
|
"epoch": 0.12810155799192152,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999982017504901,
|
|
"loss": 7.2668,
|
|
"mean_token_accuracy": 0.13351675048470496,
|
|
"num_tokens": 2808118.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.241878986358643,
|
|
"epoch": 0.1286785920369302,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999980329898401,
|
|
"loss": 7.227,
|
|
"mean_token_accuracy": 0.1321822077035904,
|
|
"num_tokens": 2821118.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.150159502029419,
|
|
"epoch": 0.12925562608193883,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000499997856661493,
|
|
"loss": 7.2141,
|
|
"mean_token_accuracy": 0.1326787978410721,
|
|
"num_tokens": 2833636.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.185760831832885,
|
|
"epoch": 0.1298326601269475,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499997672765455,
|
|
"loss": 7.2273,
|
|
"mean_token_accuracy": 0.13806429505348206,
|
|
"num_tokens": 2846617.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.1420997142791744,
|
|
"epoch": 0.13040969417195614,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999974813017318,
|
|
"loss": 7.1761,
|
|
"mean_token_accuracy": 0.14047402441501616,
|
|
"num_tokens": 2859191.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.267583703994751,
|
|
"epoch": 0.1309867282169648,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999972822703301,
|
|
"loss": 7.2047,
|
|
"mean_token_accuracy": 0.13429140374064447,
|
|
"num_tokens": 2872592.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.123564100265503,
|
|
"epoch": 0.13156376226197344,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999970756712567,
|
|
"loss": 7.1782,
|
|
"mean_token_accuracy": 0.13929741457104683,
|
|
"num_tokens": 2884094.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.225417804718018,
|
|
"epoch": 0.1321407963069821,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999968615045183,
|
|
"loss": 7.2401,
|
|
"mean_token_accuracy": 0.13268111422657966,
|
|
"num_tokens": 2896947.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.190128946304322,
|
|
"epoch": 0.13271783035199078,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999966397701222,
|
|
"loss": 7.2378,
|
|
"mean_token_accuracy": 0.13291095346212387,
|
|
"num_tokens": 2909029.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.222978162765503,
|
|
"epoch": 0.13329486439699942,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999964104680759,
|
|
"loss": 7.2202,
|
|
"mean_token_accuracy": 0.1370919555425644,
|
|
"num_tokens": 2920778.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.057752561569214,
|
|
"epoch": 0.13387189844200809,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999961735983871,
|
|
"loss": 7.1546,
|
|
"mean_token_accuracy": 0.1401558116078377,
|
|
"num_tokens": 2933754.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.285988759994507,
|
|
"epoch": 0.13444893248701673,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999959291610639,
|
|
"loss": 7.2337,
|
|
"mean_token_accuracy": 0.1410619392991066,
|
|
"num_tokens": 2946019.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.11617488861084,
|
|
"epoch": 0.1350259665320254,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999956771561143,
|
|
"loss": 7.1804,
|
|
"mean_token_accuracy": 0.13369629979133607,
|
|
"num_tokens": 2959688.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.10788106918335,
|
|
"epoch": 0.13560300057703403,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999954175835469,
|
|
"loss": 7.0996,
|
|
"mean_token_accuracy": 0.1448257327079773,
|
|
"num_tokens": 2971211.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.1773261547088625,
|
|
"epoch": 0.1361800346220427,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999951504433703,
|
|
"loss": 7.2158,
|
|
"mean_token_accuracy": 0.13084226623177528,
|
|
"num_tokens": 2983559.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.126594400405883,
|
|
"epoch": 0.13675706866705137,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999948757355935,
|
|
"loss": 7.168,
|
|
"mean_token_accuracy": 0.13138265088200568,
|
|
"num_tokens": 2995053.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.131065607070923,
|
|
"epoch": 0.13733410271206,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499994593460226,
|
|
"loss": 7.1255,
|
|
"mean_token_accuracy": 0.13711200654506683,
|
|
"num_tokens": 3006143.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.1007500171661375,
|
|
"epoch": 0.13791113675706868,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999943036172771,
|
|
"loss": 7.1097,
|
|
"mean_token_accuracy": 0.1373543344438076,
|
|
"num_tokens": 3018793.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.153094959259033,
|
|
"epoch": 0.13848817080207732,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999940062067565,
|
|
"loss": 7.1354,
|
|
"mean_token_accuracy": 0.13795246854424476,
|
|
"num_tokens": 3031547.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.04004921913147,
|
|
"epoch": 0.13906520484708598,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999937012286742,
|
|
"loss": 7.1537,
|
|
"mean_token_accuracy": 0.13225894793868065,
|
|
"num_tokens": 3044035.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.132466888427734,
|
|
"epoch": 0.13964223889209462,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999933886830405,
|
|
"loss": 7.1779,
|
|
"mean_token_accuracy": 0.14098356738686563,
|
|
"num_tokens": 3056238.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.119182538986206,
|
|
"epoch": 0.1402192729371033,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000499993068569866,
|
|
"loss": 7.1719,
|
|
"mean_token_accuracy": 0.1401755228638649,
|
|
"num_tokens": 3069250.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.180674600601196,
|
|
"epoch": 0.14079630698211196,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999927408891614,
|
|
"loss": 7.1522,
|
|
"mean_token_accuracy": 0.13382151871919631,
|
|
"num_tokens": 3082048.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.034345865249634,
|
|
"epoch": 0.1413733410271206,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999924056409378,
|
|
"loss": 7.2487,
|
|
"mean_token_accuracy": 0.13513218089938164,
|
|
"num_tokens": 3094588.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.2652326107025145,
|
|
"epoch": 0.14195037507212926,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999920628252063,
|
|
"loss": 7.2225,
|
|
"mean_token_accuracy": 0.12983713001012803,
|
|
"num_tokens": 3108067.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.027235317230224,
|
|
"epoch": 0.1425274091171379,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999917124419785,
|
|
"loss": 7.1671,
|
|
"mean_token_accuracy": 0.14140997901558877,
|
|
"num_tokens": 3120524.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.2310967445373535,
|
|
"epoch": 0.14310444316214657,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999913544912664,
|
|
"loss": 7.2371,
|
|
"mean_token_accuracy": 0.12993111908435823,
|
|
"num_tokens": 3133533.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.224728441238403,
|
|
"epoch": 0.1436814772071552,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999909889730817,
|
|
"loss": 7.1937,
|
|
"mean_token_accuracy": 0.13602788001298904,
|
|
"num_tokens": 3146183.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.092638444900513,
|
|
"epoch": 0.14425851125216388,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999906158874368,
|
|
"loss": 7.1575,
|
|
"mean_token_accuracy": 0.1345686912536621,
|
|
"num_tokens": 3158436.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.0961510181427006,
|
|
"epoch": 0.14483554529717255,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999902352343444,
|
|
"loss": 7.1168,
|
|
"mean_token_accuracy": 0.14122596234083176,
|
|
"num_tokens": 3170426.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.156266593933106,
|
|
"epoch": 0.1454125793421812,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999898470138171,
|
|
"loss": 7.1086,
|
|
"mean_token_accuracy": 0.1393971025943756,
|
|
"num_tokens": 3182865.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.043174409866333,
|
|
"epoch": 0.14598961338718985,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999894512258681,
|
|
"loss": 7.1419,
|
|
"mean_token_accuracy": 0.13401855304837226,
|
|
"num_tokens": 3195674.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.135420846939087,
|
|
"epoch": 0.1465666474321985,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999890478705107,
|
|
"loss": 7.0969,
|
|
"mean_token_accuracy": 0.14015892669558525,
|
|
"num_tokens": 3208848.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.114065217971802,
|
|
"epoch": 0.14714368147720716,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999886369477585,
|
|
"loss": 7.1859,
|
|
"mean_token_accuracy": 0.13531602919101715,
|
|
"num_tokens": 3221179.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.15838098526001,
|
|
"epoch": 0.1477207155222158,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999882184576251,
|
|
"loss": 7.2691,
|
|
"mean_token_accuracy": 0.12990250065922737,
|
|
"num_tokens": 3233671.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.2466796875,
|
|
"epoch": 0.14829774956722447,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999877924001248,
|
|
"loss": 7.1856,
|
|
"mean_token_accuracy": 0.1305832877755165,
|
|
"num_tokens": 3247219.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.049865818023681,
|
|
"epoch": 0.1488747836122331,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999873587752718,
|
|
"loss": 7.1495,
|
|
"mean_token_accuracy": 0.13346763998270034,
|
|
"num_tokens": 3259444.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.969862937927246,
|
|
"epoch": 0.14945181765724178,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999869175830808,
|
|
"loss": 7.0034,
|
|
"mean_token_accuracy": 0.14465287700295448,
|
|
"num_tokens": 3272092.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.140999794006348,
|
|
"epoch": 0.15002885170225044,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999864688235666,
|
|
"loss": 7.1902,
|
|
"mean_token_accuracy": 0.12994412407279016,
|
|
"num_tokens": 3284260.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.1015129566192625,
|
|
"epoch": 0.15060588574725908,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999860124967442,
|
|
"loss": 7.1399,
|
|
"mean_token_accuracy": 0.1303492307662964,
|
|
"num_tokens": 3296662.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.115274858474732,
|
|
"epoch": 0.15118291979226775,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999855486026291,
|
|
"loss": 7.098,
|
|
"mean_token_accuracy": 0.1382736712694168,
|
|
"num_tokens": 3307782.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.117005920410156,
|
|
"epoch": 0.1517599538372764,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999850771412369,
|
|
"loss": 7.1608,
|
|
"mean_token_accuracy": 0.13215686306357383,
|
|
"num_tokens": 3320519.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.089527368545532,
|
|
"epoch": 0.15233698788228506,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999845981125832,
|
|
"loss": 7.0264,
|
|
"mean_token_accuracy": 0.14284324571490287,
|
|
"num_tokens": 3331722.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.069760704040528,
|
|
"epoch": 0.1529140219272937,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999841115166844,
|
|
"loss": 7.1303,
|
|
"mean_token_accuracy": 0.13235283344984056,
|
|
"num_tokens": 3345143.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.030734205245972,
|
|
"epoch": 0.15349105597230236,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999836173535568,
|
|
"loss": 7.0524,
|
|
"mean_token_accuracy": 0.14473133385181428,
|
|
"num_tokens": 3358946.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.128999328613281,
|
|
"epoch": 0.15406809001731103,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999831156232169,
|
|
"loss": 7.1604,
|
|
"mean_token_accuracy": 0.1378837525844574,
|
|
"num_tokens": 3372465.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.039134931564331,
|
|
"epoch": 0.15464512406231967,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999826063256818,
|
|
"loss": 7.0408,
|
|
"mean_token_accuracy": 0.14155885428190232,
|
|
"num_tokens": 3384392.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.1258001804351805,
|
|
"epoch": 0.15522215810732834,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999820894609683,
|
|
"loss": 7.1384,
|
|
"mean_token_accuracy": 0.1228943333029747,
|
|
"num_tokens": 3398137.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.124754333496094,
|
|
"epoch": 0.15579919215233698,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999815650290941,
|
|
"loss": 7.0811,
|
|
"mean_token_accuracy": 0.14019765704870224,
|
|
"num_tokens": 3410353.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.00805253982544,
|
|
"epoch": 0.15637622619734565,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999810330300766,
|
|
"loss": 7.0326,
|
|
"mean_token_accuracy": 0.13864756077528,
|
|
"num_tokens": 3423206.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.053879833221435,
|
|
"epoch": 0.1569532602423543,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000499980493463934,
|
|
"loss": 7.1112,
|
|
"mean_token_accuracy": 0.1342466413974762,
|
|
"num_tokens": 3436780.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.1501076221466064,
|
|
"epoch": 0.15753029428736295,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999799463306841,
|
|
"loss": 7.1779,
|
|
"mean_token_accuracy": 0.13237905949354173,
|
|
"num_tokens": 3450285.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.058084535598755,
|
|
"epoch": 0.15810732833237162,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999793916303455,
|
|
"loss": 7.0564,
|
|
"mean_token_accuracy": 0.14176423102617264,
|
|
"num_tokens": 3462157.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.970040607452392,
|
|
"epoch": 0.15868436237738026,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999788293629368,
|
|
"loss": 7.0133,
|
|
"mean_token_accuracy": 0.1461627095937729,
|
|
"num_tokens": 3475027.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.019241905212402,
|
|
"epoch": 0.15926139642238893,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499978259528477,
|
|
"loss": 7.1074,
|
|
"mean_token_accuracy": 0.13902512937784195,
|
|
"num_tokens": 3489111.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.062412309646606,
|
|
"epoch": 0.15983843046739757,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999776821269852,
|
|
"loss": 7.0485,
|
|
"mean_token_accuracy": 0.1354866899549961,
|
|
"num_tokens": 3501673.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.9125723361969,
|
|
"epoch": 0.16041546451240624,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999770971584807,
|
|
"loss": 6.972,
|
|
"mean_token_accuracy": 0.14499091804027558,
|
|
"num_tokens": 3513511.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.126408863067627,
|
|
"epoch": 0.16099249855741488,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999765046229834,
|
|
"loss": 7.1008,
|
|
"mean_token_accuracy": 0.13360448479652404,
|
|
"num_tokens": 3525218.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.0275006771087645,
|
|
"epoch": 0.16156953260242354,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499975904520513,
|
|
"loss": 6.9918,
|
|
"mean_token_accuracy": 0.1486022636294365,
|
|
"num_tokens": 3537006.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.0024169921875,
|
|
"epoch": 0.1621465666474322,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00049997529685109,
|
|
"loss": 7.0915,
|
|
"mean_token_accuracy": 0.13563401028513908,
|
|
"num_tokens": 3549873.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.1586450099945065,
|
|
"epoch": 0.16272360069244085,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999746816147344,
|
|
"loss": 7.1743,
|
|
"mean_token_accuracy": 0.1260405257344246,
|
|
"num_tokens": 3563542.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.972772789001465,
|
|
"epoch": 0.16330063473744952,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999740588114673,
|
|
"loss": 7.0563,
|
|
"mean_token_accuracy": 0.13719577193260193,
|
|
"num_tokens": 3575679.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.14780478477478,
|
|
"epoch": 0.16387766878245816,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999734284413095,
|
|
"loss": 7.0748,
|
|
"mean_token_accuracy": 0.1393434651196003,
|
|
"num_tokens": 3588605.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.959795999526977,
|
|
"epoch": 0.16445470282746683,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000499972790504282,
|
|
"loss": 7.075,
|
|
"mean_token_accuracy": 0.14719273895025253,
|
|
"num_tokens": 3601894.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.970956611633301,
|
|
"epoch": 0.16503173687247547,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004999721450004067,
|
|
"loss": 7.0138,
|
|
"mean_token_accuracy": 0.14499847069382668,
|
|
"num_tokens": 3614436.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.0098734378814695,
|
|
"epoch": 0.16560877091748413,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999714919297049,
|
|
"loss": 6.9708,
|
|
"mean_token_accuracy": 0.14259159415960312,
|
|
"num_tokens": 3627590.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.064096879959107,
|
|
"epoch": 0.1661858049624928,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999708312921987,
|
|
"loss": 7.1157,
|
|
"mean_token_accuracy": 0.13143253847956657,
|
|
"num_tokens": 3640834.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.9391326904296875,
|
|
"epoch": 0.16676283900750144,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999701630879103,
|
|
"loss": 6.9743,
|
|
"mean_token_accuracy": 0.1408776670694351,
|
|
"num_tokens": 3653024.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.055517578125,
|
|
"epoch": 0.1673398730525101,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999694873168623,
|
|
"loss": 7.0344,
|
|
"mean_token_accuracy": 0.1418595716357231,
|
|
"num_tokens": 3665311.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.973097419738769,
|
|
"epoch": 0.16791690709751875,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999688039790773,
|
|
"loss": 7.0708,
|
|
"mean_token_accuracy": 0.13803807273507118,
|
|
"num_tokens": 3679102.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.027426385879517,
|
|
"epoch": 0.16849394114252741,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999681130745784,
|
|
"loss": 6.9891,
|
|
"mean_token_accuracy": 0.1482289746403694,
|
|
"num_tokens": 3690046.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.917978143692016,
|
|
"epoch": 0.16907097518753605,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999674146033888,
|
|
"loss": 7.015,
|
|
"mean_token_accuracy": 0.1324767976999283,
|
|
"num_tokens": 3703668.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.034569263458252,
|
|
"epoch": 0.16964800923254472,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999667085655318,
|
|
"loss": 7.0303,
|
|
"mean_token_accuracy": 0.14069233685731888,
|
|
"num_tokens": 3715852.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.9502979755401615,
|
|
"epoch": 0.1702250432775534,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999659949610313,
|
|
"loss": 6.9468,
|
|
"mean_token_accuracy": 0.14053553566336632,
|
|
"num_tokens": 3728310.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.0809468746185305,
|
|
"epoch": 0.17080207732256203,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999652737899114,
|
|
"loss": 7.0286,
|
|
"mean_token_accuracy": 0.1336084134876728,
|
|
"num_tokens": 3741323.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.936049604415894,
|
|
"epoch": 0.1713791113675707,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999645450521963,
|
|
"loss": 7.0266,
|
|
"mean_token_accuracy": 0.13839906752109526,
|
|
"num_tokens": 3752667.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.046130800247193,
|
|
"epoch": 0.17195614541257934,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999638087479104,
|
|
"loss": 7.0423,
|
|
"mean_token_accuracy": 0.1376166321337223,
|
|
"num_tokens": 3764282.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.975060272216797,
|
|
"epoch": 0.172533179457588,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999630648770786,
|
|
"loss": 7.0681,
|
|
"mean_token_accuracy": 0.13423443585634232,
|
|
"num_tokens": 3778280.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.946654510498047,
|
|
"epoch": 0.17311021350259664,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999623134397257,
|
|
"loss": 7.0722,
|
|
"mean_token_accuracy": 0.1414807364344597,
|
|
"num_tokens": 3791010.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.078482675552368,
|
|
"epoch": 0.1736872475476053,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999615544358774,
|
|
"loss": 7.0158,
|
|
"mean_token_accuracy": 0.13697310537099838,
|
|
"num_tokens": 3802738.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.945161008834839,
|
|
"epoch": 0.17426428159261395,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999607878655587,
|
|
"loss": 7.0416,
|
|
"mean_token_accuracy": 0.13953944891691208,
|
|
"num_tokens": 3815448.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.016398239135742,
|
|
"epoch": 0.17484131563762262,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999600137287958,
|
|
"loss": 7.0208,
|
|
"mean_token_accuracy": 0.1362545184791088,
|
|
"num_tokens": 3829833.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.95634651184082,
|
|
"epoch": 0.17541834968263129,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999592320256146,
|
|
"loss": 6.9977,
|
|
"mean_token_accuracy": 0.1375351645052433,
|
|
"num_tokens": 3841544.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.058304262161255,
|
|
"epoch": 0.17599538372763993,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999584427560412,
|
|
"loss": 6.9914,
|
|
"mean_token_accuracy": 0.1338309071958065,
|
|
"num_tokens": 3853886.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.900669813156128,
|
|
"epoch": 0.1765724177726486,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999576459201024,
|
|
"loss": 7.0318,
|
|
"mean_token_accuracy": 0.14436644166707993,
|
|
"num_tokens": 3867942.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.937264537811279,
|
|
"epoch": 0.17714945181765723,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499956841517825,
|
|
"loss": 7.0233,
|
|
"mean_token_accuracy": 0.1413443475961685,
|
|
"num_tokens": 3881764.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.968937110900879,
|
|
"epoch": 0.1777264858626659,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000499956029549236,
|
|
"loss": 6.9303,
|
|
"mean_token_accuracy": 0.14205588772892952,
|
|
"num_tokens": 3893979.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.96471152305603,
|
|
"epoch": 0.17830351990767454,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999552100143625,
|
|
"loss": 7.0387,
|
|
"mean_token_accuracy": 0.13929225131869316,
|
|
"num_tokens": 3905934.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.040468406677246,
|
|
"epoch": 0.1788805539526832,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999543829132324,
|
|
"loss": 6.9983,
|
|
"mean_token_accuracy": 0.14421933889389038,
|
|
"num_tokens": 3918348.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.9538825988769535,
|
|
"epoch": 0.17945758799769188,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999535482458734,
|
|
"loss": 6.973,
|
|
"mean_token_accuracy": 0.1384164996445179,
|
|
"num_tokens": 3931863.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.961157894134521,
|
|
"epoch": 0.18003462204270052,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999527060123135,
|
|
"loss": 6.9652,
|
|
"mean_token_accuracy": 0.1431449018418789,
|
|
"num_tokens": 3945108.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.975248908996582,
|
|
"epoch": 0.18061165608770918,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999518562125811,
|
|
"loss": 6.9944,
|
|
"mean_token_accuracy": 0.13739961683750151,
|
|
"num_tokens": 3956695.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.972767305374146,
|
|
"epoch": 0.18118869013271782,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004999509988467047,
|
|
"loss": 6.9805,
|
|
"mean_token_accuracy": 0.1392670303583145,
|
|
"num_tokens": 3970458.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.973606538772583,
|
|
"epoch": 0.1817657241777265,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999501339147132,
|
|
"loss": 6.9695,
|
|
"mean_token_accuracy": 0.13941276222467422,
|
|
"num_tokens": 3981962.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.010713863372803,
|
|
"epoch": 0.18234275822273513,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999492614166357,
|
|
"loss": 6.9822,
|
|
"mean_token_accuracy": 0.13373472541570663,
|
|
"num_tokens": 3993512.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.8247644901275635,
|
|
"epoch": 0.1829197922677438,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999483813525014,
|
|
"loss": 6.875,
|
|
"mean_token_accuracy": 0.15325912684202195,
|
|
"num_tokens": 4005826.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.960187387466431,
|
|
"epoch": 0.18349682631275246,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999474937223403,
|
|
"loss": 6.9475,
|
|
"mean_token_accuracy": 0.13995122984051706,
|
|
"num_tokens": 4018927.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.905779933929443,
|
|
"epoch": 0.1840738603577611,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999465985261818,
|
|
"loss": 6.9818,
|
|
"mean_token_accuracy": 0.13842669874429703,
|
|
"num_tokens": 4030970.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.920939159393311,
|
|
"epoch": 0.18465089440276977,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999456957640562,
|
|
"loss": 6.9621,
|
|
"mean_token_accuracy": 0.14341214746236802,
|
|
"num_tokens": 4043396.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.979000473022461,
|
|
"epoch": 0.1852279284477784,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999447854359939,
|
|
"loss": 7.0026,
|
|
"mean_token_accuracy": 0.13570645451545715,
|
|
"num_tokens": 4055588.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.951703596115112,
|
|
"epoch": 0.18580496249278708,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999438675420255,
|
|
"loss": 6.9723,
|
|
"mean_token_accuracy": 0.13750323951244353,
|
|
"num_tokens": 4068036.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.912143850326538,
|
|
"epoch": 0.18638199653779572,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999429420821818,
|
|
"loss": 7.0134,
|
|
"mean_token_accuracy": 0.13561285212635993,
|
|
"num_tokens": 4080531.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.996648979187012,
|
|
"epoch": 0.1869590305828044,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499942009056494,
|
|
"loss": 6.9186,
|
|
"mean_token_accuracy": 0.14655348733067514,
|
|
"num_tokens": 4093614.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.9176637649536135,
|
|
"epoch": 0.18753606462781305,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999410684649935,
|
|
"loss": 7.0373,
|
|
"mean_token_accuracy": 0.13838913813233375,
|
|
"num_tokens": 4106836.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.0432507514953615,
|
|
"epoch": 0.1881130986728217,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999401203077119,
|
|
"loss": 6.9509,
|
|
"mean_token_accuracy": 0.14006586968898774,
|
|
"num_tokens": 4118739.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.852195692062378,
|
|
"epoch": 0.18869013271783036,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000499939164584681,
|
|
"loss": 6.9308,
|
|
"mean_token_accuracy": 0.14347120001912117,
|
|
"num_tokens": 4131174.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.005696773529053,
|
|
"epoch": 0.189267166762839,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999382012959331,
|
|
"loss": 6.9709,
|
|
"mean_token_accuracy": 0.133350370824337,
|
|
"num_tokens": 4143459.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.931919956207276,
|
|
"epoch": 0.18984420080784767,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999372304415005,
|
|
"loss": 6.9939,
|
|
"mean_token_accuracy": 0.13698216155171394,
|
|
"num_tokens": 4155864.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.837768316268921,
|
|
"epoch": 0.1904212348528563,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999362520214159,
|
|
"loss": 6.7786,
|
|
"mean_token_accuracy": 0.14995110183954238,
|
|
"num_tokens": 4168433.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.960832262039185,
|
|
"epoch": 0.19099826889786498,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999352660357122,
|
|
"loss": 6.9764,
|
|
"mean_token_accuracy": 0.13728797063231468,
|
|
"num_tokens": 4180852.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.950531387329102,
|
|
"epoch": 0.19157530294287364,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999342724844226,
|
|
"loss": 6.9155,
|
|
"mean_token_accuracy": 0.14255120754241943,
|
|
"num_tokens": 4194712.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.91253252029419,
|
|
"epoch": 0.19215233698788228,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999332713675804,
|
|
"loss": 6.9909,
|
|
"mean_token_accuracy": 0.14385711327195166,
|
|
"num_tokens": 4207970.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.85596489906311,
|
|
"epoch": 0.19272937103289095,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999322626852193,
|
|
"loss": 6.9123,
|
|
"mean_token_accuracy": 0.14430793896317481,
|
|
"num_tokens": 4220893.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.888790082931519,
|
|
"epoch": 0.1933064050778996,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999312464373734,
|
|
"loss": 6.9127,
|
|
"mean_token_accuracy": 0.14029934853315354,
|
|
"num_tokens": 4232463.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.904188966751098,
|
|
"epoch": 0.19388343912290826,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999302226240767,
|
|
"loss": 6.8553,
|
|
"mean_token_accuracy": 0.13843169063329697,
|
|
"num_tokens": 4244604.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.918503475189209,
|
|
"epoch": 0.1944604731679169,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999291912453637,
|
|
"loss": 6.9512,
|
|
"mean_token_accuracy": 0.14121533036231995,
|
|
"num_tokens": 4257260.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.981924057006836,
|
|
"epoch": 0.19503750721292556,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999281523012691,
|
|
"loss": 6.8856,
|
|
"mean_token_accuracy": 0.14021847546100616,
|
|
"num_tokens": 4269885.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.918470287322998,
|
|
"epoch": 0.1956145412579342,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999271057918278,
|
|
"loss": 7.0177,
|
|
"mean_token_accuracy": 0.137399822473526,
|
|
"num_tokens": 4281560.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.923596143722534,
|
|
"epoch": 0.19619157530294287,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999260517170751,
|
|
"loss": 6.8986,
|
|
"mean_token_accuracy": 0.1467658117413521,
|
|
"num_tokens": 4293746.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.947674751281738,
|
|
"epoch": 0.19676860934795154,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999249900770463,
|
|
"loss": 6.9248,
|
|
"mean_token_accuracy": 0.1379011631011963,
|
|
"num_tokens": 4306595.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.960206651687622,
|
|
"epoch": 0.19734564339296018,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999239208717772,
|
|
"loss": 6.8912,
|
|
"mean_token_accuracy": 0.1443597435951233,
|
|
"num_tokens": 4318705.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.834646701812744,
|
|
"epoch": 0.19792267743796885,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999228441013037,
|
|
"loss": 6.9075,
|
|
"mean_token_accuracy": 0.14063191488385202,
|
|
"num_tokens": 4331321.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.875298166275025,
|
|
"epoch": 0.1984997114829775,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499921759765662,
|
|
"loss": 6.875,
|
|
"mean_token_accuracy": 0.14705403074622153,
|
|
"num_tokens": 4343517.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.939173650741577,
|
|
"epoch": 0.19907674552798615,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999206678648888,
|
|
"loss": 6.9152,
|
|
"mean_token_accuracy": 0.14792341142892837,
|
|
"num_tokens": 4356342.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.834136724472046,
|
|
"epoch": 0.1996537795729948,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999195683990206,
|
|
"loss": 6.9055,
|
|
"mean_token_accuracy": 0.14260428175330161,
|
|
"num_tokens": 4369078.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.930724620819092,
|
|
"epoch": 0.20023081361800346,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999184613680945,
|
|
"loss": 6.8893,
|
|
"mean_token_accuracy": 0.13664604276418685,
|
|
"num_tokens": 4382091.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.834902429580689,
|
|
"epoch": 0.20080784766301213,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999173467721476,
|
|
"loss": 6.916,
|
|
"mean_token_accuracy": 0.13967763930559157,
|
|
"num_tokens": 4393919.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.9305174350738525,
|
|
"epoch": 0.20138488170802077,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004999162246112175,
|
|
"loss": 6.8841,
|
|
"mean_token_accuracy": 0.14394750744104384,
|
|
"num_tokens": 4407054.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.847874784469605,
|
|
"epoch": 0.20196191575302944,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004999150948853419,
|
|
"loss": 6.9343,
|
|
"mean_token_accuracy": 0.13633092492818832,
|
|
"num_tokens": 4420180.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.021355628967285,
|
|
"epoch": 0.20253894979803808,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999139575945587,
|
|
"loss": 6.9167,
|
|
"mean_token_accuracy": 0.15015274733304979,
|
|
"num_tokens": 4433248.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.800896453857422,
|
|
"epoch": 0.20311598384304674,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999128127389065,
|
|
"loss": 6.9057,
|
|
"mean_token_accuracy": 0.14164429455995559,
|
|
"num_tokens": 4446340.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.900873231887817,
|
|
"epoch": 0.20369301788805538,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999116603184233,
|
|
"loss": 6.9154,
|
|
"mean_token_accuracy": 0.14202600717544556,
|
|
"num_tokens": 4459918.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.862395143508911,
|
|
"epoch": 0.20427005193306405,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999105003331482,
|
|
"loss": 6.8724,
|
|
"mean_token_accuracy": 0.14486078545451164,
|
|
"num_tokens": 4472020.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.871878528594971,
|
|
"epoch": 0.20484708597807272,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999093327831202,
|
|
"loss": 6.844,
|
|
"mean_token_accuracy": 0.14406751170754434,
|
|
"num_tokens": 4484269.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.906937932968139,
|
|
"epoch": 0.20542412002308136,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999081576683784,
|
|
"loss": 6.944,
|
|
"mean_token_accuracy": 0.13962593898177147,
|
|
"num_tokens": 4498040.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.899182319641113,
|
|
"epoch": 0.20600115406809003,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999069749889626,
|
|
"loss": 6.8898,
|
|
"mean_token_accuracy": 0.14287936463952064,
|
|
"num_tokens": 4509701.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.898169612884521,
|
|
"epoch": 0.20657818811309867,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999057847449123,
|
|
"loss": 6.9164,
|
|
"mean_token_accuracy": 0.14265610203146933,
|
|
"num_tokens": 4521840.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.880409812927246,
|
|
"epoch": 0.20715522215810733,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999045869362678,
|
|
"loss": 6.9135,
|
|
"mean_token_accuracy": 0.141028805822134,
|
|
"num_tokens": 4534241.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.891494417190552,
|
|
"epoch": 0.20773225620311597,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499903381563069,
|
|
"loss": 6.8726,
|
|
"mean_token_accuracy": 0.14260033294558525,
|
|
"num_tokens": 4547255.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.858333587646484,
|
|
"epoch": 0.20830929024812464,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999021686253568,
|
|
"loss": 6.8877,
|
|
"mean_token_accuracy": 0.14118088632822037,
|
|
"num_tokens": 4559538.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.860760068893432,
|
|
"epoch": 0.2088863242931333,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999009481231719,
|
|
"loss": 6.8935,
|
|
"mean_token_accuracy": 0.14185215309262275,
|
|
"num_tokens": 4571444.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.703266668319702,
|
|
"epoch": 0.20946335833814195,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004998997200565553,
|
|
"loss": 6.7323,
|
|
"mean_token_accuracy": 0.15689657777547836,
|
|
"num_tokens": 4586294.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.907558965682983,
|
|
"epoch": 0.21004039238315061,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998984844255483,
|
|
"loss": 6.8867,
|
|
"mean_token_accuracy": 0.14380680844187738,
|
|
"num_tokens": 4599661.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.882632923126221,
|
|
"epoch": 0.21061742642815925,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004998972412301925,
|
|
"loss": 6.892,
|
|
"mean_token_accuracy": 0.14093836694955825,
|
|
"num_tokens": 4611305.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.85354471206665,
|
|
"epoch": 0.21119446047316792,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004998959904705297,
|
|
"loss": 6.9001,
|
|
"mean_token_accuracy": 0.13364512771368026,
|
|
"num_tokens": 4623992.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.925228023529053,
|
|
"epoch": 0.21177149451817656,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998947321466021,
|
|
"loss": 6.8803,
|
|
"mean_token_accuracy": 0.13986791148781777,
|
|
"num_tokens": 4636495.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.903949499130249,
|
|
"epoch": 0.21234852856318523,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004998934662584518,
|
|
"loss": 6.9039,
|
|
"mean_token_accuracy": 0.13886573910713196,
|
|
"num_tokens": 4650266.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.853269004821778,
|
|
"epoch": 0.2129255626081939,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998921928061214,
|
|
"loss": 6.859,
|
|
"mean_token_accuracy": 0.14313986301422119,
|
|
"num_tokens": 4663238.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.901078176498413,
|
|
"epoch": 0.21350259665320254,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998909117896539,
|
|
"loss": 6.8956,
|
|
"mean_token_accuracy": 0.1408998563885689,
|
|
"num_tokens": 4675501.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.818858623504639,
|
|
"epoch": 0.2140796306982112,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998896232090922,
|
|
"loss": 6.8711,
|
|
"mean_token_accuracy": 0.13704144060611725,
|
|
"num_tokens": 4688996.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.915656042098999,
|
|
"epoch": 0.21465666474321984,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998883270644798,
|
|
"loss": 6.9764,
|
|
"mean_token_accuracy": 0.13644033223390578,
|
|
"num_tokens": 4703226.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.90679931640625,
|
|
"epoch": 0.2152336987882285,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998870233558602,
|
|
"loss": 6.8537,
|
|
"mean_token_accuracy": 0.13797650784254073,
|
|
"num_tokens": 4715825.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.776429557800293,
|
|
"epoch": 0.21581073283323715,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998857120832774,
|
|
"loss": 6.8472,
|
|
"mean_token_accuracy": 0.14446549713611603,
|
|
"num_tokens": 4727819.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.890902948379517,
|
|
"epoch": 0.21638776687824582,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004998843932467751,
|
|
"loss": 6.8847,
|
|
"mean_token_accuracy": 0.13867277577519416,
|
|
"num_tokens": 4742890.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.844753170013428,
|
|
"epoch": 0.21696480092325446,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998830668463982,
|
|
"loss": 6.868,
|
|
"mean_token_accuracy": 0.1419196143746376,
|
|
"num_tokens": 4755609.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.797444152832031,
|
|
"epoch": 0.21754183496826313,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004998817328821909,
|
|
"loss": 6.8404,
|
|
"mean_token_accuracy": 0.1449926421046257,
|
|
"num_tokens": 4766674.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.869100189208984,
|
|
"epoch": 0.2181188690132718,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998803913541983,
|
|
"loss": 6.8862,
|
|
"mean_token_accuracy": 0.13758110478520394,
|
|
"num_tokens": 4779639.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 7.015814256668091,
|
|
"epoch": 0.21869590305828043,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998790422624654,
|
|
"loss": 6.9639,
|
|
"mean_token_accuracy": 0.13711576908826828,
|
|
"num_tokens": 4791702.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.843172836303711,
|
|
"epoch": 0.2192729371032891,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004998776856070376,
|
|
"loss": 6.7892,
|
|
"mean_token_accuracy": 0.14893252551555633,
|
|
"num_tokens": 4805756.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.778265953063965,
|
|
"epoch": 0.21984997114829774,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998763213879606,
|
|
"loss": 6.8403,
|
|
"mean_token_accuracy": 0.1457715541124344,
|
|
"num_tokens": 4817941.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.812149238586426,
|
|
"epoch": 0.2204270051933064,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998749496052803,
|
|
"loss": 6.7324,
|
|
"mean_token_accuracy": 0.14796641916036607,
|
|
"num_tokens": 4832298.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.744766473770142,
|
|
"epoch": 0.22100403923831505,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998735702590426,
|
|
"loss": 6.8345,
|
|
"mean_token_accuracy": 0.14761534780263902,
|
|
"num_tokens": 4844814.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.905437231063843,
|
|
"epoch": 0.22158107328332372,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998721833492942,
|
|
"loss": 6.8911,
|
|
"mean_token_accuracy": 0.14497228264808654,
|
|
"num_tokens": 4857888.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.804248714447022,
|
|
"epoch": 0.22215810732833238,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998707888760816,
|
|
"loss": 6.7916,
|
|
"mean_token_accuracy": 0.14039909094572067,
|
|
"num_tokens": 4870479.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.816269254684448,
|
|
"epoch": 0.22273514137334102,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998693868394516,
|
|
"loss": 6.784,
|
|
"mean_token_accuracy": 0.14926855266094208,
|
|
"num_tokens": 4883294.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.94655408859253,
|
|
"epoch": 0.2233121754183497,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004998679772394516,
|
|
"loss": 6.9359,
|
|
"mean_token_accuracy": 0.13703567907214165,
|
|
"num_tokens": 4896331.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.7918534755706785,
|
|
"epoch": 0.22388920946335833,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004998665600761289,
|
|
"loss": 6.8752,
|
|
"mean_token_accuracy": 0.14121760800480843,
|
|
"num_tokens": 4908857.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.851579332351685,
|
|
"epoch": 0.224466243508367,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499865135349531,
|
|
"loss": 6.8359,
|
|
"mean_token_accuracy": 0.14525341987609863,
|
|
"num_tokens": 4921564.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.8388258934021,
|
|
"epoch": 0.22504327755337564,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998637030597061,
|
|
"loss": 6.9006,
|
|
"mean_token_accuracy": 0.1439589023590088,
|
|
"num_tokens": 4934960.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.740708255767823,
|
|
"epoch": 0.2256203115983843,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998622632067022,
|
|
"loss": 6.7348,
|
|
"mean_token_accuracy": 0.14645355492830275,
|
|
"num_tokens": 4946942.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.884095239639282,
|
|
"epoch": 0.22619734564339297,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998608157905678,
|
|
"loss": 6.8547,
|
|
"mean_token_accuracy": 0.137442384660244,
|
|
"num_tokens": 4960751.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.83891282081604,
|
|
"epoch": 0.2267743796884016,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998593608113515,
|
|
"loss": 6.8662,
|
|
"mean_token_accuracy": 0.13838402479887008,
|
|
"num_tokens": 4973551.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.7111108779907225,
|
|
"epoch": 0.22735141373341028,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004998578982691024,
|
|
"loss": 6.884,
|
|
"mean_token_accuracy": 0.14016093909740449,
|
|
"num_tokens": 4986224.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.898041009902954,
|
|
"epoch": 0.22792844777841892,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998564281638694,
|
|
"loss": 6.8416,
|
|
"mean_token_accuracy": 0.14575736969709396,
|
|
"num_tokens": 4999874.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.8169921875,
|
|
"epoch": 0.2285054818234276,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998549504957023,
|
|
"loss": 6.8364,
|
|
"mean_token_accuracy": 0.14460800364613532,
|
|
"num_tokens": 5013106.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.807040786743164,
|
|
"epoch": 0.22908251586843623,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998534652646506,
|
|
"loss": 6.8062,
|
|
"mean_token_accuracy": 0.13764368891716003,
|
|
"num_tokens": 5026145.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.798084878921509,
|
|
"epoch": 0.2296595499134449,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998519724707642,
|
|
"loss": 6.8203,
|
|
"mean_token_accuracy": 0.14574942290782927,
|
|
"num_tokens": 5039216.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.829916763305664,
|
|
"epoch": 0.23023658395845356,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004998504721140934,
|
|
"loss": 6.7989,
|
|
"mean_token_accuracy": 0.14435165673494338,
|
|
"num_tokens": 5051956.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.8158831119537355,
|
|
"epoch": 0.2308136180034622,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004998489641946887,
|
|
"loss": 6.8865,
|
|
"mean_token_accuracy": 0.14350505471229552,
|
|
"num_tokens": 5065325.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.828835344314575,
|
|
"epoch": 0.23139065204847087,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998474487126009,
|
|
"loss": 6.8822,
|
|
"mean_token_accuracy": 0.14764170944690705,
|
|
"num_tokens": 5077776.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.852951622009277,
|
|
"epoch": 0.2319676860934795,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998459256678806,
|
|
"loss": 6.8817,
|
|
"mean_token_accuracy": 0.1433338388800621,
|
|
"num_tokens": 5091167.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.881577014923096,
|
|
"epoch": 0.23254472013848818,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998443950605796,
|
|
"loss": 6.8069,
|
|
"mean_token_accuracy": 0.14394941478967666,
|
|
"num_tokens": 5103321.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.797431755065918,
|
|
"epoch": 0.23312175418349682,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004998428568907488,
|
|
"loss": 6.8253,
|
|
"mean_token_accuracy": 0.14137339890003203,
|
|
"num_tokens": 5116999.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.807014989852905,
|
|
"epoch": 0.23369878822850548,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998413111584403,
|
|
"loss": 6.8226,
|
|
"mean_token_accuracy": 0.14501554518938065,
|
|
"num_tokens": 5130429.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.7783522605896,
|
|
"epoch": 0.23427582227351415,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998397578637059,
|
|
"loss": 6.762,
|
|
"mean_token_accuracy": 0.14727519750595092,
|
|
"num_tokens": 5142728.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.840787601470947,
|
|
"epoch": 0.2348528563185228,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499838197006598,
|
|
"loss": 6.873,
|
|
"mean_token_accuracy": 0.14228396862745285,
|
|
"num_tokens": 5155105.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.886496639251709,
|
|
"epoch": 0.23542989036353146,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000499836628587169,
|
|
"loss": 6.8909,
|
|
"mean_token_accuracy": 0.13143049255013467,
|
|
"num_tokens": 5169206.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.709402132034302,
|
|
"epoch": 0.2360069244085401,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998350526054716,
|
|
"loss": 6.7657,
|
|
"mean_token_accuracy": 0.14668394550681113,
|
|
"num_tokens": 5182007.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.844460344314575,
|
|
"epoch": 0.23658395845354876,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998334690615591,
|
|
"loss": 6.797,
|
|
"mean_token_accuracy": 0.1384980671107769,
|
|
"num_tokens": 5195137.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.796845769882202,
|
|
"epoch": 0.2371609924985574,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004998318779554844,
|
|
"loss": 6.8063,
|
|
"mean_token_accuracy": 0.1452823057770729,
|
|
"num_tokens": 5207713.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.778965282440185,
|
|
"epoch": 0.23773802654356607,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998302792873012,
|
|
"loss": 6.7537,
|
|
"mean_token_accuracy": 0.1435099482536316,
|
|
"num_tokens": 5220125.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.790195322036743,
|
|
"epoch": 0.23831506058857474,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998286730570631,
|
|
"loss": 6.8133,
|
|
"mean_token_accuracy": 0.14231673255562782,
|
|
"num_tokens": 5232142.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.807083463668823,
|
|
"epoch": 0.23889209463358338,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998270592648245,
|
|
"loss": 6.8058,
|
|
"mean_token_accuracy": 0.14608865603804588,
|
|
"num_tokens": 5245517.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.799246835708618,
|
|
"epoch": 0.23946912867859205,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998254379106394,
|
|
"loss": 6.7838,
|
|
"mean_token_accuracy": 0.14370430558919906,
|
|
"num_tokens": 5257862.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.738052463531494,
|
|
"epoch": 0.2400461627236007,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998238089945622,
|
|
"loss": 6.7372,
|
|
"mean_token_accuracy": 0.1498562604188919,
|
|
"num_tokens": 5270141.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.822539758682251,
|
|
"epoch": 0.24062319676860935,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004998221725166479,
|
|
"loss": 6.7559,
|
|
"mean_token_accuracy": 0.15351167917251587,
|
|
"num_tokens": 5283143.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.763333368301391,
|
|
"epoch": 0.241200230813618,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004998205284769516,
|
|
"loss": 6.8284,
|
|
"mean_token_accuracy": 0.14074681848287582,
|
|
"num_tokens": 5295340.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.8454938411712645,
|
|
"epoch": 0.24177726485862666,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998188768755285,
|
|
"loss": 6.7681,
|
|
"mean_token_accuracy": 0.14496009051799774,
|
|
"num_tokens": 5306821.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.715377140045166,
|
|
"epoch": 0.2423542989036353,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998172177124341,
|
|
"loss": 6.7732,
|
|
"mean_token_accuracy": 0.1452890232205391,
|
|
"num_tokens": 5319428.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.903585720062256,
|
|
"epoch": 0.24293133294864397,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998155509877242,
|
|
"loss": 6.8206,
|
|
"mean_token_accuracy": 0.1386203348636627,
|
|
"num_tokens": 5331777.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.615080881118774,
|
|
"epoch": 0.24350836699365264,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000499813876701455,
|
|
"loss": 6.7726,
|
|
"mean_token_accuracy": 0.14410800784826278,
|
|
"num_tokens": 5344533.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.7770771980285645,
|
|
"epoch": 0.24408540103866128,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998121948536828,
|
|
"loss": 6.718,
|
|
"mean_token_accuracy": 0.1547897882759571,
|
|
"num_tokens": 5357123.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.803248691558838,
|
|
"epoch": 0.24466243508366994,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998105054444639,
|
|
"loss": 6.8454,
|
|
"mean_token_accuracy": 0.14135158509016038,
|
|
"num_tokens": 5370123.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.810746097564698,
|
|
"epoch": 0.24523946912867858,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998088084738555,
|
|
"loss": 6.7767,
|
|
"mean_token_accuracy": 0.15171893686056137,
|
|
"num_tokens": 5382870.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.694212627410889,
|
|
"epoch": 0.24581650317368725,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998071039419144,
|
|
"loss": 6.761,
|
|
"mean_token_accuracy": 0.1418367989361286,
|
|
"num_tokens": 5395052.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.8836406707763675,
|
|
"epoch": 0.2463935372186959,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000499805391848698,
|
|
"loss": 6.7966,
|
|
"mean_token_accuracy": 0.1489194765686989,
|
|
"num_tokens": 5406720.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.736827850341797,
|
|
"epoch": 0.24697057126370456,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000499803672194264,
|
|
"loss": 6.7875,
|
|
"mean_token_accuracy": 0.14699700772762297,
|
|
"num_tokens": 5419465.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.800150632858276,
|
|
"epoch": 0.24754760530871323,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004998019449786701,
|
|
"loss": 6.7748,
|
|
"mean_token_accuracy": 0.13936677277088166,
|
|
"num_tokens": 5432434.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.758311748504639,
|
|
"epoch": 0.24812463935372187,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998002102019744,
|
|
"loss": 6.7772,
|
|
"mean_token_accuracy": 0.1484916940331459,
|
|
"num_tokens": 5445249.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.888931179046631,
|
|
"epoch": 0.24870167339873053,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004997984678642354,
|
|
"loss": 6.7955,
|
|
"mean_token_accuracy": 0.14284422919154166,
|
|
"num_tokens": 5458996.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.7189394474029545,
|
|
"epoch": 0.24927870744373917,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004997967179655115,
|
|
"loss": 6.7892,
|
|
"mean_token_accuracy": 0.13899215385317804,
|
|
"num_tokens": 5472174.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.748557662963867,
|
|
"epoch": 0.24985574148874784,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004997949605058617,
|
|
"loss": 6.7129,
|
|
"mean_token_accuracy": 0.1528759181499481,
|
|
"num_tokens": 5484464.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.79465274810791,
|
|
"epoch": 0.2504327755337565,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997931954853451,
|
|
"loss": 6.8143,
|
|
"mean_token_accuracy": 0.14372445791959762,
|
|
"num_tokens": 5497278.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.7927593231201175,
|
|
"epoch": 0.2510098095787651,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499791422904021,
|
|
"loss": 6.8328,
|
|
"mean_token_accuracy": 0.1454320654273033,
|
|
"num_tokens": 5510847.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.7520914554595945,
|
|
"epoch": 0.2515868436237738,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997896427619491,
|
|
"loss": 6.7613,
|
|
"mean_token_accuracy": 0.14879355430603028,
|
|
"num_tokens": 5522543.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.717813110351562,
|
|
"epoch": 0.25216387766878245,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004997878550591892,
|
|
"loss": 6.7648,
|
|
"mean_token_accuracy": 0.1428240269422531,
|
|
"num_tokens": 5534856.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.851426029205323,
|
|
"epoch": 0.2527409117137911,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004997860597958013,
|
|
"loss": 6.6801,
|
|
"mean_token_accuracy": 0.15747882872819902,
|
|
"num_tokens": 5546881.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.717418527603149,
|
|
"epoch": 0.2533179457587998,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004997842569718461,
|
|
"loss": 6.8,
|
|
"mean_token_accuracy": 0.14309746026992798,
|
|
"num_tokens": 5559711.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.8028624057769775,
|
|
"epoch": 0.25389497980380843,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499782446587384,
|
|
"loss": 6.783,
|
|
"mean_token_accuracy": 0.1497884564101696,
|
|
"num_tokens": 5573553.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.793797731399536,
|
|
"epoch": 0.25447201384881707,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499780628642476,
|
|
"loss": 6.7494,
|
|
"mean_token_accuracy": 0.1393908143043518,
|
|
"num_tokens": 5586046.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.702590990066528,
|
|
"epoch": 0.2550490478938257,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997788031371834,
|
|
"loss": 6.7686,
|
|
"mean_token_accuracy": 0.14913811907172203,
|
|
"num_tokens": 5597876.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.8008434772491455,
|
|
"epoch": 0.2556260819388344,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997769700715672,
|
|
"loss": 6.6819,
|
|
"mean_token_accuracy": 0.1523703783750534,
|
|
"num_tokens": 5611264.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.685852336883545,
|
|
"epoch": 0.25620311598384304,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004997751294456892,
|
|
"loss": 6.7773,
|
|
"mean_token_accuracy": 0.14500191956758499,
|
|
"num_tokens": 5625093.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.798491287231445,
|
|
"epoch": 0.2567801500288517,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997732812596114,
|
|
"loss": 6.7895,
|
|
"mean_token_accuracy": 0.14235708564519883,
|
|
"num_tokens": 5637206.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.776926708221436,
|
|
"epoch": 0.2573571840738604,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997714255133961,
|
|
"loss": 6.6586,
|
|
"mean_token_accuracy": 0.15341545939445494,
|
|
"num_tokens": 5649136.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.628551292419433,
|
|
"epoch": 0.257934218118869,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997695622071054,
|
|
"loss": 6.7041,
|
|
"mean_token_accuracy": 0.14951958358287812,
|
|
"num_tokens": 5660714.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.831679010391236,
|
|
"epoch": 0.25851125216387766,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997676913408021,
|
|
"loss": 6.7764,
|
|
"mean_token_accuracy": 0.14415197670459748,
|
|
"num_tokens": 5673295.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.752670621871948,
|
|
"epoch": 0.2590882862088863,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004997658129145493,
|
|
"loss": 6.7818,
|
|
"mean_token_accuracy": 0.1467595562338829,
|
|
"num_tokens": 5686822.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.774938583374023,
|
|
"epoch": 0.259665320253895,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997639269284099,
|
|
"loss": 6.8382,
|
|
"mean_token_accuracy": 0.14073604345321655,
|
|
"num_tokens": 5700621.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.839014863967895,
|
|
"epoch": 0.26024235429890363,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997620333824476,
|
|
"loss": 6.7463,
|
|
"mean_token_accuracy": 0.1523364171385765,
|
|
"num_tokens": 5713565.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.72583498954773,
|
|
"epoch": 0.2608193883439123,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997601322767258,
|
|
"loss": 6.7916,
|
|
"mean_token_accuracy": 0.14135704040527344,
|
|
"num_tokens": 5726487.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.8859046459197994,
|
|
"epoch": 0.26139642238892097,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997582236113087,
|
|
"loss": 6.8826,
|
|
"mean_token_accuracy": 0.13812644258141518,
|
|
"num_tokens": 5739934.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.76940541267395,
|
|
"epoch": 0.2619734564339296,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004997563073862603,
|
|
"loss": 6.8123,
|
|
"mean_token_accuracy": 0.13768922835588454,
|
|
"num_tokens": 5752187.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.70945725440979,
|
|
"epoch": 0.26255049047893825,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004997543836016453,
|
|
"loss": 6.7426,
|
|
"mean_token_accuracy": 0.14485511556267738,
|
|
"num_tokens": 5764850.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.867229843139649,
|
|
"epoch": 0.2631275245239469,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004997524522575281,
|
|
"loss": 6.8221,
|
|
"mean_token_accuracy": 0.14024255499243737,
|
|
"num_tokens": 5778577.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.7879596710205075,
|
|
"epoch": 0.2637045585689556,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004997505133539738,
|
|
"loss": 6.7548,
|
|
"mean_token_accuracy": 0.14954013526439666,
|
|
"num_tokens": 5791699.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.717269659042358,
|
|
"epoch": 0.2642815926139642,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004997485668910476,
|
|
"loss": 6.7576,
|
|
"mean_token_accuracy": 0.14114084988832473,
|
|
"num_tokens": 5804553.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.840640878677368,
|
|
"epoch": 0.26485862665897286,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997466128688151,
|
|
"loss": 6.7691,
|
|
"mean_token_accuracy": 0.1443391412496567,
|
|
"num_tokens": 5816758.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.754163122177124,
|
|
"epoch": 0.26543566070398156,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000499744651287342,
|
|
"loss": 6.7999,
|
|
"mean_token_accuracy": 0.13716064393520355,
|
|
"num_tokens": 5831048.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.835683441162109,
|
|
"epoch": 0.2660126947489902,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499742682146694,
|
|
"loss": 6.7946,
|
|
"mean_token_accuracy": 0.13617580756545067,
|
|
"num_tokens": 5843159.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.710077857971191,
|
|
"epoch": 0.26658972879399884,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004997407054469377,
|
|
"loss": 6.6863,
|
|
"mean_token_accuracy": 0.15132941156625748,
|
|
"num_tokens": 5856078.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.773683309555054,
|
|
"epoch": 0.2671667628390075,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004997387211881392,
|
|
"loss": 6.7288,
|
|
"mean_token_accuracy": 0.145392032712698,
|
|
"num_tokens": 5867835.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.747898149490356,
|
|
"epoch": 0.26774379688401617,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004997367293703656,
|
|
"loss": 6.7683,
|
|
"mean_token_accuracy": 0.14005454927682875,
|
|
"num_tokens": 5881179.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.79307336807251,
|
|
"epoch": 0.2683208309290248,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997347299936837,
|
|
"loss": 6.7228,
|
|
"mean_token_accuracy": 0.14137931019067765,
|
|
"num_tokens": 5892774.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.781385231018066,
|
|
"epoch": 0.26889786497403345,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004997327230581608,
|
|
"loss": 6.7838,
|
|
"mean_token_accuracy": 0.14386948049068451,
|
|
"num_tokens": 5905857.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.713219213485718,
|
|
"epoch": 0.26947489901904215,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997307085638644,
|
|
"loss": 6.7246,
|
|
"mean_token_accuracy": 0.14458930790424346,
|
|
"num_tokens": 5918128.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.815898609161377,
|
|
"epoch": 0.2700519330640508,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004997286865108621,
|
|
"loss": 6.8516,
|
|
"mean_token_accuracy": 0.14051285684108733,
|
|
"num_tokens": 5930606.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.749667167663574,
|
|
"epoch": 0.2706289671090594,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997266568992222,
|
|
"loss": 6.7854,
|
|
"mean_token_accuracy": 0.13582591190934182,
|
|
"num_tokens": 5944139.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.768195295333863,
|
|
"epoch": 0.27120600115406807,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004997246197290128,
|
|
"loss": 6.7608,
|
|
"mean_token_accuracy": 0.14503443688154222,
|
|
"num_tokens": 5956779.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.704603576660157,
|
|
"epoch": 0.27178303519907676,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997225750003024,
|
|
"loss": 6.7297,
|
|
"mean_token_accuracy": 0.15043272450566292,
|
|
"num_tokens": 5971510.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.63649697303772,
|
|
"epoch": 0.2723600692440854,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997205227131599,
|
|
"loss": 6.7573,
|
|
"mean_token_accuracy": 0.14194964990019798,
|
|
"num_tokens": 5983877.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.898113250732422,
|
|
"epoch": 0.27293710328909404,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004997184628676542,
|
|
"loss": 6.7795,
|
|
"mean_token_accuracy": 0.14526699259877204,
|
|
"num_tokens": 5996739.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.708590412139893,
|
|
"epoch": 0.27351413733410274,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004997163954638546,
|
|
"loss": 6.7437,
|
|
"mean_token_accuracy": 0.1482561096549034,
|
|
"num_tokens": 6009289.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.8309389591217045,
|
|
"epoch": 0.2740911713791114,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997143205018306,
|
|
"loss": 6.811,
|
|
"mean_token_accuracy": 0.13893268257379532,
|
|
"num_tokens": 6021375.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.804889583587647,
|
|
"epoch": 0.27466820542412,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997122379816523,
|
|
"loss": 6.7297,
|
|
"mean_token_accuracy": 0.14610403627157212,
|
|
"num_tokens": 6034182.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.675373125076294,
|
|
"epoch": 0.27524523946912866,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004997101479033893,
|
|
"loss": 6.7187,
|
|
"mean_token_accuracy": 0.14495936110615731,
|
|
"num_tokens": 6046486.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.791097116470337,
|
|
"epoch": 0.27582227351413735,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997080502671122,
|
|
"loss": 6.7782,
|
|
"mean_token_accuracy": 0.14465533643960954,
|
|
"num_tokens": 6060867.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.686882209777832,
|
|
"epoch": 0.276399307559146,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997059450728913,
|
|
"loss": 6.6263,
|
|
"mean_token_accuracy": 0.15024522840976715,
|
|
"num_tokens": 6073000.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.704486465454101,
|
|
"epoch": 0.27697634160415463,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004997038323207977,
|
|
"loss": 6.6973,
|
|
"mean_token_accuracy": 0.14645934849977493,
|
|
"num_tokens": 6085884.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.675950574874878,
|
|
"epoch": 0.2775533756491633,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997017120109022,
|
|
"loss": 6.6314,
|
|
"mean_token_accuracy": 0.14963351488113402,
|
|
"num_tokens": 6098230.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.721731472015381,
|
|
"epoch": 0.27813040969417197,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004996995841432762,
|
|
"loss": 6.6896,
|
|
"mean_token_accuracy": 0.14837536364793777,
|
|
"num_tokens": 6111568.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.646577262878418,
|
|
"epoch": 0.2787074437391806,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004996974487179915,
|
|
"loss": 6.6649,
|
|
"mean_token_accuracy": 0.14184629917144775,
|
|
"num_tokens": 6123992.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.754623317718506,
|
|
"epoch": 0.27928447778418924,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004996953057351195,
|
|
"loss": 6.6896,
|
|
"mean_token_accuracy": 0.1484291136264801,
|
|
"num_tokens": 6135452.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.6486248016357425,
|
|
"epoch": 0.27986151182919794,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996931551947327,
|
|
"loss": 6.6125,
|
|
"mean_token_accuracy": 0.156364855915308,
|
|
"num_tokens": 6150144.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.7660589694976805,
|
|
"epoch": 0.2804385458742066,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004996909970969031,
|
|
"loss": 6.7401,
|
|
"mean_token_accuracy": 0.14714512154459952,
|
|
"num_tokens": 6163397.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.687129402160645,
|
|
"epoch": 0.2810155799192152,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996888314417034,
|
|
"loss": 6.731,
|
|
"mean_token_accuracy": 0.1419209510087967,
|
|
"num_tokens": 6175579.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.683587169647216,
|
|
"epoch": 0.2815926139642239,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004996866582292067,
|
|
"loss": 6.7079,
|
|
"mean_token_accuracy": 0.14961353838443756,
|
|
"num_tokens": 6189229.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.73968095779419,
|
|
"epoch": 0.28216964800923255,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996844774594856,
|
|
"loss": 6.7092,
|
|
"mean_token_accuracy": 0.14912996739149093,
|
|
"num_tokens": 6203817.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.717224407196045,
|
|
"epoch": 0.2827466820542412,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004996822891326138,
|
|
"loss": 6.7278,
|
|
"mean_token_accuracy": 0.14414327293634416,
|
|
"num_tokens": 6216098.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.753392219543457,
|
|
"epoch": 0.28332371609924983,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004996800932486648,
|
|
"loss": 6.7286,
|
|
"mean_token_accuracy": 0.1483657404780388,
|
|
"num_tokens": 6228987.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.664376258850098,
|
|
"epoch": 0.28390075014425853,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004996778898077126,
|
|
"loss": 6.675,
|
|
"mean_token_accuracy": 0.1498671442270279,
|
|
"num_tokens": 6241676.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.794411563873291,
|
|
"epoch": 0.28447778418926717,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004996756788098309,
|
|
"loss": 6.7131,
|
|
"mean_token_accuracy": 0.14685458093881607,
|
|
"num_tokens": 6254744.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.630081748962402,
|
|
"epoch": 0.2850548182342758,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004996734602550945,
|
|
"loss": 6.6722,
|
|
"mean_token_accuracy": 0.14627297297120095,
|
|
"num_tokens": 6268242.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.659547185897827,
|
|
"epoch": 0.2856318522792845,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004996712341435779,
|
|
"loss": 6.6072,
|
|
"mean_token_accuracy": 0.15240462720394135,
|
|
"num_tokens": 6279904.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.643069171905518,
|
|
"epoch": 0.28620888632429314,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004996690004753559,
|
|
"loss": 6.7204,
|
|
"mean_token_accuracy": 0.1448551282286644,
|
|
"num_tokens": 6292885.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.64668550491333,
|
|
"epoch": 0.2867859203693018,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004996667592505037,
|
|
"loss": 6.7049,
|
|
"mean_token_accuracy": 0.1487411305308342,
|
|
"num_tokens": 6306278.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.800214529037476,
|
|
"epoch": 0.2873629544143104,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004996645104690967,
|
|
"loss": 6.7361,
|
|
"mean_token_accuracy": 0.13756746500730516,
|
|
"num_tokens": 6320360.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.657628870010376,
|
|
"epoch": 0.2879399884593191,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004996622541312103,
|
|
"loss": 6.7216,
|
|
"mean_token_accuracy": 0.145097079873085,
|
|
"num_tokens": 6332778.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.7686714172363285,
|
|
"epoch": 0.28851702250432776,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004996599902369207,
|
|
"loss": 6.7012,
|
|
"mean_token_accuracy": 0.1465804323554039,
|
|
"num_tokens": 6345028.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.653385210037231,
|
|
"epoch": 0.2890940565493364,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996577187863039,
|
|
"loss": 6.6454,
|
|
"mean_token_accuracy": 0.15272147804498673,
|
|
"num_tokens": 6356931.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.664360809326172,
|
|
"epoch": 0.2896710905943451,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004996554397794364,
|
|
"loss": 6.5927,
|
|
"mean_token_accuracy": 0.157369065284729,
|
|
"num_tokens": 6370247.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.653738355636596,
|
|
"epoch": 0.29024812463935373,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004996531532163947,
|
|
"loss": 6.6727,
|
|
"mean_token_accuracy": 0.14644310176372527,
|
|
"num_tokens": 6382352.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.664462614059448,
|
|
"epoch": 0.2908251586843624,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004996508590972558,
|
|
"loss": 6.6628,
|
|
"mean_token_accuracy": 0.1447147697210312,
|
|
"num_tokens": 6395949.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.685240316390991,
|
|
"epoch": 0.291402192729371,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996485574220969,
|
|
"loss": 6.6785,
|
|
"mean_token_accuracy": 0.1504662573337555,
|
|
"num_tokens": 6408469.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.677483606338501,
|
|
"epoch": 0.2919792267743797,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004996462481909953,
|
|
"loss": 6.675,
|
|
"mean_token_accuracy": 0.1545150399208069,
|
|
"num_tokens": 6419953.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.695938968658448,
|
|
"epoch": 0.29255626081938835,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004996439314040287,
|
|
"loss": 6.7126,
|
|
"mean_token_accuracy": 0.1428425744175911,
|
|
"num_tokens": 6432567.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.797707271575928,
|
|
"epoch": 0.293133294864397,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000499641607061275,
|
|
"loss": 6.6713,
|
|
"mean_token_accuracy": 0.14336236119270324,
|
|
"num_tokens": 6445951.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.6457275390625,
|
|
"epoch": 0.2937103289094057,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996392751628126,
|
|
"loss": 6.7311,
|
|
"mean_token_accuracy": 0.15187496691942215,
|
|
"num_tokens": 6458799.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.727105379104614,
|
|
"epoch": 0.2942873629544143,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996369357087196,
|
|
"loss": 6.7279,
|
|
"mean_token_accuracy": 0.14346445798873902,
|
|
"num_tokens": 6471423.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.745425605773926,
|
|
"epoch": 0.29486439699942296,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499634588699075,
|
|
"loss": 6.7509,
|
|
"mean_token_accuracy": 0.14119785130023957,
|
|
"num_tokens": 6484248.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.744411516189575,
|
|
"epoch": 0.2954414310444316,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004996322341339575,
|
|
"loss": 6.6769,
|
|
"mean_token_accuracy": 0.1400275394320488,
|
|
"num_tokens": 6497048.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.662609720230103,
|
|
"epoch": 0.2960184650894403,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996298720134465,
|
|
"loss": 6.6918,
|
|
"mean_token_accuracy": 0.1442575104534626,
|
|
"num_tokens": 6508840.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.695674324035645,
|
|
"epoch": 0.29659549913444894,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004996275023376213,
|
|
"loss": 6.6791,
|
|
"mean_token_accuracy": 0.14943985342979432,
|
|
"num_tokens": 6520976.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.68662805557251,
|
|
"epoch": 0.2971725331794576,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004996251251065615,
|
|
"loss": 6.6945,
|
|
"mean_token_accuracy": 0.14707806557416916,
|
|
"num_tokens": 6533009.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.712952613830566,
|
|
"epoch": 0.2977495672244662,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996227403203473,
|
|
"loss": 6.6792,
|
|
"mean_token_accuracy": 0.14307745099067687,
|
|
"num_tokens": 6545196.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.744017934799194,
|
|
"epoch": 0.2983266012694749,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004996203479790589,
|
|
"loss": 6.7119,
|
|
"mean_token_accuracy": 0.1456460326910019,
|
|
"num_tokens": 6557479.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.709696102142334,
|
|
"epoch": 0.29890363531448355,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996179480827764,
|
|
"loss": 6.733,
|
|
"mean_token_accuracy": 0.14646259546279908,
|
|
"num_tokens": 6569832.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.684864521026611,
|
|
"epoch": 0.2994806693594922,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000499615540631581,
|
|
"loss": 6.6935,
|
|
"mean_token_accuracy": 0.14482910484075545,
|
|
"num_tokens": 6581651.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.737748479843139,
|
|
"epoch": 0.3000577034045009,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004996131256255534,
|
|
"loss": 6.7498,
|
|
"mean_token_accuracy": 0.14754320159554482,
|
|
"num_tokens": 6596331.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.737265586853027,
|
|
"epoch": 0.3006347374495095,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004996107030647749,
|
|
"loss": 6.6931,
|
|
"mean_token_accuracy": 0.14396552741527557,
|
|
"num_tokens": 6609712.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.733648252487183,
|
|
"epoch": 0.30121177149451817,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004996082729493269,
|
|
"loss": 6.71,
|
|
"mean_token_accuracy": 0.14992823898792268,
|
|
"num_tokens": 6622780.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.6975846767425535,
|
|
"epoch": 0.3017888055395268,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996058352792913,
|
|
"loss": 6.7664,
|
|
"mean_token_accuracy": 0.14214256927371025,
|
|
"num_tokens": 6636288.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.7836583137512205,
|
|
"epoch": 0.3023658395845355,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00049960339005475,
|
|
"loss": 6.7468,
|
|
"mean_token_accuracy": 0.1360473372042179,
|
|
"num_tokens": 6649883.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.70995774269104,
|
|
"epoch": 0.30294287362954414,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004996009372757852,
|
|
"loss": 6.6904,
|
|
"mean_token_accuracy": 0.14994267225265503,
|
|
"num_tokens": 6662241.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.6174835681915285,
|
|
"epoch": 0.3035199076745528,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004995984769424795,
|
|
"loss": 6.6436,
|
|
"mean_token_accuracy": 0.148514224588871,
|
|
"num_tokens": 6674790.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.731441926956177,
|
|
"epoch": 0.3040969417195615,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004995960090549155,
|
|
"loss": 6.67,
|
|
"mean_token_accuracy": 0.14512893110513686,
|
|
"num_tokens": 6687713.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.580911350250244,
|
|
"epoch": 0.3046739757645701,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995935336131764,
|
|
"loss": 6.6733,
|
|
"mean_token_accuracy": 0.1467847667634487,
|
|
"num_tokens": 6701123.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.748194313049316,
|
|
"epoch": 0.30525100980957875,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004995910506173452,
|
|
"loss": 6.6264,
|
|
"mean_token_accuracy": 0.152266626060009,
|
|
"num_tokens": 6712674.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.585151863098145,
|
|
"epoch": 0.3058280438545874,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004995885600675059,
|
|
"loss": 6.6317,
|
|
"mean_token_accuracy": 0.15141176879405976,
|
|
"num_tokens": 6724840.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.713297176361084,
|
|
"epoch": 0.3064050778995961,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004995860619637415,
|
|
"loss": 6.7572,
|
|
"mean_token_accuracy": 0.1347401686012745,
|
|
"num_tokens": 6738746.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.730117273330689,
|
|
"epoch": 0.30698211194460473,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004995835563061367,
|
|
"loss": 6.6619,
|
|
"mean_token_accuracy": 0.14460855424404145,
|
|
"num_tokens": 6751620.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.723760747909546,
|
|
"epoch": 0.30755914598961337,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004995810430947756,
|
|
"loss": 6.6421,
|
|
"mean_token_accuracy": 0.14792002290487288,
|
|
"num_tokens": 6763957.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.4763280868530275,
|
|
"epoch": 0.30813618003462206,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004995785223297426,
|
|
"loss": 6.551,
|
|
"mean_token_accuracy": 0.1542297527194023,
|
|
"num_tokens": 6776211.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.775838565826416,
|
|
"epoch": 0.3087132140796307,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004995759940111225,
|
|
"loss": 6.8122,
|
|
"mean_token_accuracy": 0.13833045810461045,
|
|
"num_tokens": 6789745.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.757865047454834,
|
|
"epoch": 0.30929024812463934,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995734581390005,
|
|
"loss": 6.7322,
|
|
"mean_token_accuracy": 0.14661592841148377,
|
|
"num_tokens": 6801771.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.666882467269898,
|
|
"epoch": 0.309867282169648,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004995709147134617,
|
|
"loss": 6.67,
|
|
"mean_token_accuracy": 0.14816711097955704,
|
|
"num_tokens": 6813993.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.75975227355957,
|
|
"epoch": 0.3104443162146567,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004995683637345919,
|
|
"loss": 6.728,
|
|
"mean_token_accuracy": 0.14804089665412903,
|
|
"num_tokens": 6826833.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.66631441116333,
|
|
"epoch": 0.3110213502596653,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004995658052024765,
|
|
"loss": 6.6693,
|
|
"mean_token_accuracy": 0.1456070676445961,
|
|
"num_tokens": 6838947.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.758131980895996,
|
|
"epoch": 0.31159838430467396,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004995632391172019,
|
|
"loss": 6.7287,
|
|
"mean_token_accuracy": 0.14437391310930253,
|
|
"num_tokens": 6851781.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.707075405120849,
|
|
"epoch": 0.31217541834968265,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004995606654788543,
|
|
"loss": 6.6256,
|
|
"mean_token_accuracy": 0.15065254122018815,
|
|
"num_tokens": 6863479.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.62445650100708,
|
|
"epoch": 0.3127524523946913,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004995580842875202,
|
|
"loss": 6.6145,
|
|
"mean_token_accuracy": 0.15116894394159316,
|
|
"num_tokens": 6877000.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.720065450668335,
|
|
"epoch": 0.31332948643969993,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995554955432866,
|
|
"loss": 6.7174,
|
|
"mean_token_accuracy": 0.14094419330358504,
|
|
"num_tokens": 6888878.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.668194341659546,
|
|
"epoch": 0.3139065204847086,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004995528992462404,
|
|
"loss": 6.6328,
|
|
"mean_token_accuracy": 0.14614609777927398,
|
|
"num_tokens": 6902147.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.681904458999634,
|
|
"epoch": 0.31448355452971727,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000499550295396469,
|
|
"loss": 6.6788,
|
|
"mean_token_accuracy": 0.1509232111275196,
|
|
"num_tokens": 6914927.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.710317325592041,
|
|
"epoch": 0.3150605885747259,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004995476839940598,
|
|
"loss": 6.6955,
|
|
"mean_token_accuracy": 0.1472512513399124,
|
|
"num_tokens": 6927757.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.674422359466552,
|
|
"epoch": 0.31563762261973455,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000499545065039101,
|
|
"loss": 6.6768,
|
|
"mean_token_accuracy": 0.14419350624084473,
|
|
"num_tokens": 6941039.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.694507312774658,
|
|
"epoch": 0.31621465666474324,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995424385316803,
|
|
"loss": 6.7503,
|
|
"mean_token_accuracy": 0.14313241690397263,
|
|
"num_tokens": 6954876.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.665238857269287,
|
|
"epoch": 0.3167916907097519,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004995398044718863,
|
|
"loss": 6.596,
|
|
"mean_token_accuracy": 0.15183332860469817,
|
|
"num_tokens": 6966889.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.698918628692627,
|
|
"epoch": 0.3173687247547605,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004995371628598074,
|
|
"loss": 6.6259,
|
|
"mean_token_accuracy": 0.1496454581618309,
|
|
"num_tokens": 6979894.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.705928373336792,
|
|
"epoch": 0.31794575879976916,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995345136955328,
|
|
"loss": 6.6708,
|
|
"mean_token_accuracy": 0.143413520604372,
|
|
"num_tokens": 6992734.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.664940214157104,
|
|
"epoch": 0.31852279284477786,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004995318569791512,
|
|
"loss": 6.6779,
|
|
"mean_token_accuracy": 0.14752028435468673,
|
|
"num_tokens": 7006504.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.694177198410034,
|
|
"epoch": 0.3190998268897865,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004995291927107522,
|
|
"loss": 6.6083,
|
|
"mean_token_accuracy": 0.15231579244136811,
|
|
"num_tokens": 7018978.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.5929046154022215,
|
|
"epoch": 0.31967686093479514,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004995265208904252,
|
|
"loss": 6.6973,
|
|
"mean_token_accuracy": 0.14533187747001647,
|
|
"num_tokens": 7031294.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.704652452468872,
|
|
"epoch": 0.32025389497980383,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004995238415182603,
|
|
"loss": 6.6554,
|
|
"mean_token_accuracy": 0.1433829165995121,
|
|
"num_tokens": 7044139.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.583083963394165,
|
|
"epoch": 0.3208309290248125,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004995211545943476,
|
|
"loss": 6.5525,
|
|
"mean_token_accuracy": 0.15347641259431838,
|
|
"num_tokens": 7056232.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.695966291427612,
|
|
"epoch": 0.3214079630698211,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004995184601187772,
|
|
"loss": 6.7123,
|
|
"mean_token_accuracy": 0.14398207515478134,
|
|
"num_tokens": 7070125.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.615304565429687,
|
|
"epoch": 0.32198499711482975,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00049951575809164,
|
|
"loss": 6.5848,
|
|
"mean_token_accuracy": 0.15029205679893493,
|
|
"num_tokens": 7083535.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.624817371368408,
|
|
"epoch": 0.32256203115983845,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004995130485130269,
|
|
"loss": 6.5539,
|
|
"mean_token_accuracy": 0.1532074198126793,
|
|
"num_tokens": 7095081.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.699851179122925,
|
|
"epoch": 0.3231390652048471,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004995103313830289,
|
|
"loss": 6.683,
|
|
"mean_token_accuracy": 0.14111172929406166,
|
|
"num_tokens": 7107764.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.6601025581359865,
|
|
"epoch": 0.3237160992498557,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004995076067017373,
|
|
"loss": 6.6111,
|
|
"mean_token_accuracy": 0.15362927615642546,
|
|
"num_tokens": 7119754.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.564910221099853,
|
|
"epoch": 0.3242931332948644,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004995048744692439,
|
|
"loss": 6.6472,
|
|
"mean_token_accuracy": 0.1543910101056099,
|
|
"num_tokens": 7132087.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.642766809463501,
|
|
"epoch": 0.32487016733987306,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004995021346856405,
|
|
"loss": 6.6196,
|
|
"mean_token_accuracy": 0.15014703422784806,
|
|
"num_tokens": 7143325.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.652205085754394,
|
|
"epoch": 0.3254472013848817,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004994993873510196,
|
|
"loss": 6.5793,
|
|
"mean_token_accuracy": 0.15422357767820358,
|
|
"num_tokens": 7154771.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.642023944854737,
|
|
"epoch": 0.32602423542989034,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004994966324654731,
|
|
"loss": 6.693,
|
|
"mean_token_accuracy": 0.1453319787979126,
|
|
"num_tokens": 7168352.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.497804021835327,
|
|
"epoch": 0.32660126947489904,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000499493870029094,
|
|
"loss": 6.4934,
|
|
"mean_token_accuracy": 0.16205645650625228,
|
|
"num_tokens": 7180618.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.666966104507447,
|
|
"epoch": 0.3271783035199077,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004994911000419749,
|
|
"loss": 6.6267,
|
|
"mean_token_accuracy": 0.14406555965542794,
|
|
"num_tokens": 7192634.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.592815494537353,
|
|
"epoch": 0.3277553375649163,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004994883225042093,
|
|
"loss": 6.6478,
|
|
"mean_token_accuracy": 0.14309148415923117,
|
|
"num_tokens": 7205839.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.6875081062316895,
|
|
"epoch": 0.328332371609925,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004994855374158905,
|
|
"loss": 6.6635,
|
|
"mean_token_accuracy": 0.14771109595894813,
|
|
"num_tokens": 7218371.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.6291303634643555,
|
|
"epoch": 0.32890940565493365,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.000499482744777112,
|
|
"loss": 6.4748,
|
|
"mean_token_accuracy": 0.16835684180259705,
|
|
"num_tokens": 7231506.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.602149820327758,
|
|
"epoch": 0.3294864396999423,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000499479944587968,
|
|
"loss": 6.6033,
|
|
"mean_token_accuracy": 0.15234598368406296,
|
|
"num_tokens": 7243088.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.582425212860107,
|
|
"epoch": 0.33006347374495093,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004994771368485526,
|
|
"loss": 6.5306,
|
|
"mean_token_accuracy": 0.1502147063612938,
|
|
"num_tokens": 7255383.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.583856105804443,
|
|
"epoch": 0.3306405077899596,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004994743215589602,
|
|
"loss": 6.6303,
|
|
"mean_token_accuracy": 0.15170817896723748,
|
|
"num_tokens": 7268270.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.590276622772217,
|
|
"epoch": 0.33121754183496827,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004994714987192854,
|
|
"loss": 6.543,
|
|
"mean_token_accuracy": 0.15047916322946547,
|
|
"num_tokens": 7281108.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.596184539794922,
|
|
"epoch": 0.3317945758799769,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004994686683296234,
|
|
"loss": 6.6151,
|
|
"mean_token_accuracy": 0.14714079722762108,
|
|
"num_tokens": 7293961.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.654780054092408,
|
|
"epoch": 0.3323716099249856,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004994658303900691,
|
|
"loss": 6.6303,
|
|
"mean_token_accuracy": 0.14734134078025818,
|
|
"num_tokens": 7306574.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.68011884689331,
|
|
"epoch": 0.33294864396999424,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004994629849007182,
|
|
"loss": 6.667,
|
|
"mean_token_accuracy": 0.14702263921499253,
|
|
"num_tokens": 7319046.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.692756223678589,
|
|
"epoch": 0.3335256780150029,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994601318616663,
|
|
"loss": 6.659,
|
|
"mean_token_accuracy": 0.14255366772413253,
|
|
"num_tokens": 7332033.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.608484172821045,
|
|
"epoch": 0.3341027120600115,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004994572712730092,
|
|
"loss": 6.6756,
|
|
"mean_token_accuracy": 0.13977887481451035,
|
|
"num_tokens": 7344658.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.74255256652832,
|
|
"epoch": 0.3346797461050202,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004994544031348434,
|
|
"loss": 6.5531,
|
|
"mean_token_accuracy": 0.1509678602218628,
|
|
"num_tokens": 7357252.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.453712368011475,
|
|
"epoch": 0.33525678015002885,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004994515274472653,
|
|
"loss": 6.5003,
|
|
"mean_token_accuracy": 0.16169244945049285,
|
|
"num_tokens": 7369280.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.535965251922607,
|
|
"epoch": 0.3358338141950375,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004994486442103715,
|
|
"loss": 6.5023,
|
|
"mean_token_accuracy": 0.15722685903310776,
|
|
"num_tokens": 7383017.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.617752981185913,
|
|
"epoch": 0.3364108482400462,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499445753424259,
|
|
"loss": 6.6474,
|
|
"mean_token_accuracy": 0.13711773306131364,
|
|
"num_tokens": 7395634.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.697910356521606,
|
|
"epoch": 0.33698788228505483,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004994428550890251,
|
|
"loss": 6.62,
|
|
"mean_token_accuracy": 0.1489083245396614,
|
|
"num_tokens": 7408590.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.6144743919372555,
|
|
"epoch": 0.33756491633006347,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004994399492047672,
|
|
"loss": 6.6102,
|
|
"mean_token_accuracy": 0.14934324771165847,
|
|
"num_tokens": 7420202.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.6217461109161375,
|
|
"epoch": 0.3381419503750721,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994370357715832,
|
|
"loss": 6.5721,
|
|
"mean_token_accuracy": 0.15716550201177598,
|
|
"num_tokens": 7432664.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.596539640426636,
|
|
"epoch": 0.3387189844200808,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000499434114789571,
|
|
"loss": 6.7155,
|
|
"mean_token_accuracy": 0.1429716408252716,
|
|
"num_tokens": 7446084.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.627397680282593,
|
|
"epoch": 0.33929601846508944,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004994311862588286,
|
|
"loss": 6.5659,
|
|
"mean_token_accuracy": 0.15205595940351485,
|
|
"num_tokens": 7458314.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.650746536254883,
|
|
"epoch": 0.3398730525100981,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004994282501794551,
|
|
"loss": 6.6073,
|
|
"mean_token_accuracy": 0.15533693358302117,
|
|
"num_tokens": 7471995.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.6287164211273195,
|
|
"epoch": 0.3404500865551068,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004994253065515486,
|
|
"loss": 6.6467,
|
|
"mean_token_accuracy": 0.14468009397387505,
|
|
"num_tokens": 7485676.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.698169898986817,
|
|
"epoch": 0.3410271206001154,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004994223553752084,
|
|
"loss": 6.5347,
|
|
"mean_token_accuracy": 0.15355831310153006,
|
|
"num_tokens": 7497057.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.529919147491455,
|
|
"epoch": 0.34160415464512406,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004994193966505339,
|
|
"loss": 6.5581,
|
|
"mean_token_accuracy": 0.152175223082304,
|
|
"num_tokens": 7510625.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.542059373855591,
|
|
"epoch": 0.3421811886901327,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004994164303776244,
|
|
"loss": 6.5599,
|
|
"mean_token_accuracy": 0.15199756473302842,
|
|
"num_tokens": 7522677.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.536736965179443,
|
|
"epoch": 0.3427582227351414,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004994134565565797,
|
|
"loss": 6.4882,
|
|
"mean_token_accuracy": 0.15326575115323066,
|
|
"num_tokens": 7535406.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.621460723876953,
|
|
"epoch": 0.34333525678015003,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004994104751875,
|
|
"loss": 6.6714,
|
|
"mean_token_accuracy": 0.14512530714273453,
|
|
"num_tokens": 7547577.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.65846061706543,
|
|
"epoch": 0.3439122908251587,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004994074862704854,
|
|
"loss": 6.6405,
|
|
"mean_token_accuracy": 0.14858071729540825,
|
|
"num_tokens": 7561734.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.637954330444336,
|
|
"epoch": 0.3444893248701673,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004994044898056363,
|
|
"loss": 6.7036,
|
|
"mean_token_accuracy": 0.1463548108935356,
|
|
"num_tokens": 7574256.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.763088846206665,
|
|
"epoch": 0.345066358915176,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004994014857930538,
|
|
"loss": 6.6147,
|
|
"mean_token_accuracy": 0.15213310569524766,
|
|
"num_tokens": 7587149.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.554739332199096,
|
|
"epoch": 0.34564339296018465,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004993984742328387,
|
|
"loss": 6.5758,
|
|
"mean_token_accuracy": 0.14600939601659774,
|
|
"num_tokens": 7599124.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.577654266357422,
|
|
"epoch": 0.3462204270051933,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004993954551250925,
|
|
"loss": 6.5727,
|
|
"mean_token_accuracy": 0.15178048834204674,
|
|
"num_tokens": 7612070.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.3462204270051933,
|
|
"eval_entropy": 6.630148336634502,
|
|
"eval_loss": 6.675414085388184,
|
|
"eval_mean_token_accuracy": 0.15012675653883586,
|
|
"eval_num_tokens": 7612070.0,
|
|
"eval_runtime": 23.0236,
|
|
"eval_samples_per_second": 1222.048,
|
|
"eval_steps_per_second": 152.756,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 86650,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 3000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.1231850295296e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|