4035 lines
109 KiB
JSON
4035 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.10270103728047653,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742582702636719,
|
|
"epoch": 0.00025675259320119136,
|
|
"grad_norm": 5.15625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7385,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 8466.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742582893371582,
|
|
"epoch": 0.0005135051864023827,
|
|
"grad_norm": 5.375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7383,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 17819.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742578315734864,
|
|
"epoch": 0.000770257779603574,
|
|
"grad_norm": 5.15625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7074,
|
|
"mean_token_accuracy": 8.474576170556248e-05,
|
|
"num_tokens": 26701.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.7426007270813,
|
|
"epoch": 0.0010270103728047654,
|
|
"grad_norm": 4.96875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6576,
|
|
"mean_token_accuracy": 0.0011235648184083403,
|
|
"num_tokens": 35386.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742522716522217,
|
|
"epoch": 0.0012837629660059566,
|
|
"grad_norm": 4.8125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5556,
|
|
"mean_token_accuracy": 0.009902489895466715,
|
|
"num_tokens": 44013.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742104721069335,
|
|
"epoch": 0.001540515559207148,
|
|
"grad_norm": 4.03125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4463,
|
|
"mean_token_accuracy": 0.03455784637480974,
|
|
"num_tokens": 52692.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740787982940674,
|
|
"epoch": 0.0017972681524083393,
|
|
"grad_norm": 3.6875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.2846,
|
|
"mean_token_accuracy": 0.04273700974881649,
|
|
"num_tokens": 60741.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.736649703979491,
|
|
"epoch": 0.002054020745609531,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1801,
|
|
"mean_token_accuracy": 0.04606366418302059,
|
|
"num_tokens": 69580.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.729912948608398,
|
|
"epoch": 0.002310773338810722,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.036,
|
|
"mean_token_accuracy": 0.04552499800920486,
|
|
"num_tokens": 77450.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.72356538772583,
|
|
"epoch": 0.002567525932011913,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9714,
|
|
"mean_token_accuracy": 0.04347024708986282,
|
|
"num_tokens": 85519.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.719997215270997,
|
|
"epoch": 0.0028242785252131047,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.8927,
|
|
"mean_token_accuracy": 0.04672105684876442,
|
|
"num_tokens": 93628.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.717273998260499,
|
|
"epoch": 0.003081031118414296,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8626,
|
|
"mean_token_accuracy": 0.04250718653202057,
|
|
"num_tokens": 102441.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.714400959014892,
|
|
"epoch": 0.0033377837116154874,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.7868,
|
|
"mean_token_accuracy": 0.04955284185707569,
|
|
"num_tokens": 111166.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.709486484527588,
|
|
"epoch": 0.0035945363048166786,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7258,
|
|
"mean_token_accuracy": 0.04404304698109627,
|
|
"num_tokens": 119300.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.703517818450928,
|
|
"epoch": 0.00385128889801787,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.6956,
|
|
"mean_token_accuracy": 0.04594614543020725,
|
|
"num_tokens": 127959.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.697421169281006,
|
|
"epoch": 0.004108041491219062,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6217,
|
|
"mean_token_accuracy": 0.047571105882525444,
|
|
"num_tokens": 136631.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.68674201965332,
|
|
"epoch": 0.004364794084420253,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5111,
|
|
"mean_token_accuracy": 0.057168309763073924,
|
|
"num_tokens": 145023.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.670315933227538,
|
|
"epoch": 0.004621546677621444,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4523,
|
|
"mean_token_accuracy": 0.05657172314822674,
|
|
"num_tokens": 154075.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.65313549041748,
|
|
"epoch": 0.004878299270822635,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3482,
|
|
"mean_token_accuracy": 0.06485468745231629,
|
|
"num_tokens": 162006.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.625462627410888,
|
|
"epoch": 0.005135051864023826,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2717,
|
|
"mean_token_accuracy": 0.06588986031711101,
|
|
"num_tokens": 170230.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.582051181793213,
|
|
"epoch": 0.005391804457225018,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2423,
|
|
"mean_token_accuracy": 0.05897444784641266,
|
|
"num_tokens": 179215.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.52384090423584,
|
|
"epoch": 0.0056485570504262094,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1095,
|
|
"mean_token_accuracy": 0.06802426725625992,
|
|
"num_tokens": 187858.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.461370754241944,
|
|
"epoch": 0.005905309643627401,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0136,
|
|
"mean_token_accuracy": 0.06227913834154606,
|
|
"num_tokens": 196198.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.423934841156006,
|
|
"epoch": 0.006162062236828592,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8603,
|
|
"mean_token_accuracy": 0.06186963692307472,
|
|
"num_tokens": 204940.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.381671047210693,
|
|
"epoch": 0.006418814830029783,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7743,
|
|
"mean_token_accuracy": 0.0610227856785059,
|
|
"num_tokens": 213092.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.294575500488282,
|
|
"epoch": 0.006675567423230975,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.5688,
|
|
"mean_token_accuracy": 0.06969367675483226,
|
|
"num_tokens": 220978.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.18680181503296,
|
|
"epoch": 0.006932320016432166,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.5845,
|
|
"mean_token_accuracy": 0.06069661751389503,
|
|
"num_tokens": 229852.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.096609592437744,
|
|
"epoch": 0.007189072609633357,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.4392,
|
|
"mean_token_accuracy": 0.06324404701590539,
|
|
"num_tokens": 238908.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.960357856750488,
|
|
"epoch": 0.007445825202834548,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.3607,
|
|
"mean_token_accuracy": 0.06009394600987435,
|
|
"num_tokens": 248473.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.86625270843506,
|
|
"epoch": 0.00770257779603574,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.2554,
|
|
"mean_token_accuracy": 0.06259206980466843,
|
|
"num_tokens": 257210.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.693745231628418,
|
|
"epoch": 0.007959330389236931,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.1019,
|
|
"mean_token_accuracy": 0.06411121524870396,
|
|
"num_tokens": 266677.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.491777038574218,
|
|
"epoch": 0.008216082982438123,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.9959,
|
|
"mean_token_accuracy": 0.06184820123016834,
|
|
"num_tokens": 275216.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.286455821990966,
|
|
"epoch": 0.008472835575639314,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.9137,
|
|
"mean_token_accuracy": 0.06522414833307266,
|
|
"num_tokens": 283770.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.112856197357178,
|
|
"epoch": 0.008729588168840506,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.8781,
|
|
"mean_token_accuracy": 0.06161203049123287,
|
|
"num_tokens": 292414.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.901847839355469,
|
|
"epoch": 0.008986340762041696,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.7992,
|
|
"mean_token_accuracy": 0.06540008261799812,
|
|
"num_tokens": 300569.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.703197193145751,
|
|
"epoch": 0.009243093355242888,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.6644,
|
|
"mean_token_accuracy": 0.06989255100488663,
|
|
"num_tokens": 308285.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.601799583435058,
|
|
"epoch": 0.00949984594844408,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7802,
|
|
"mean_token_accuracy": 0.06652609743177891,
|
|
"num_tokens": 317558.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.49561529159546,
|
|
"epoch": 0.00975659854164527,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.6158,
|
|
"mean_token_accuracy": 0.06920971758663655,
|
|
"num_tokens": 326377.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.364750003814697,
|
|
"epoch": 0.010013351134846462,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.7265,
|
|
"mean_token_accuracy": 0.06648653373122215,
|
|
"num_tokens": 334669.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.294552993774413,
|
|
"epoch": 0.010270103728047653,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.5956,
|
|
"mean_token_accuracy": 0.07071267962455749,
|
|
"num_tokens": 343100.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.269872093200684,
|
|
"epoch": 0.010526856321248845,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.6108,
|
|
"mean_token_accuracy": 0.07263344749808312,
|
|
"num_tokens": 351447.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.213550281524657,
|
|
"epoch": 0.010783608914450037,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.5708,
|
|
"mean_token_accuracy": 0.06988951861858368,
|
|
"num_tokens": 360908.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.182147026062012,
|
|
"epoch": 0.011040361507651227,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.5837,
|
|
"mean_token_accuracy": 0.06750233620405197,
|
|
"num_tokens": 370075.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.132856750488282,
|
|
"epoch": 0.011297114100852419,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.4565,
|
|
"mean_token_accuracy": 0.07424333952367305,
|
|
"num_tokens": 378598.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.098768615722657,
|
|
"epoch": 0.01155386669405361,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.5533,
|
|
"mean_token_accuracy": 0.07423756159842014,
|
|
"num_tokens": 387177.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.190262031555175,
|
|
"epoch": 0.011810619287254801,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.6521,
|
|
"mean_token_accuracy": 0.06974112130701542,
|
|
"num_tokens": 396017.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.12903299331665,
|
|
"epoch": 0.012067371880455993,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.5765,
|
|
"mean_token_accuracy": 0.07110274769365788,
|
|
"num_tokens": 405324.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.074048614501953,
|
|
"epoch": 0.012324124473657183,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.5065,
|
|
"mean_token_accuracy": 0.07165832705795765,
|
|
"num_tokens": 414633.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.055518102645873,
|
|
"epoch": 0.012580877066858375,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.558,
|
|
"mean_token_accuracy": 0.06795871779322624,
|
|
"num_tokens": 424088.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.072611761093139,
|
|
"epoch": 0.012837629660059566,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.5031,
|
|
"mean_token_accuracy": 0.0740334302186966,
|
|
"num_tokens": 432868.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.065638113021851,
|
|
"epoch": 0.013094382253260758,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.4989,
|
|
"mean_token_accuracy": 0.07697901576757431,
|
|
"num_tokens": 442282.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.062051105499268,
|
|
"epoch": 0.01335113484646195,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.4639,
|
|
"mean_token_accuracy": 0.07005516327917576,
|
|
"num_tokens": 451600.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.881219291687012,
|
|
"epoch": 0.01360788743966314,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.4421,
|
|
"mean_token_accuracy": 0.07474162802100182,
|
|
"num_tokens": 461072.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.112028455734253,
|
|
"epoch": 0.013864640032864332,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.5392,
|
|
"mean_token_accuracy": 0.07459446638822556,
|
|
"num_tokens": 469793.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.97756552696228,
|
|
"epoch": 0.014121392626065524,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.4663,
|
|
"mean_token_accuracy": 0.07501543574035167,
|
|
"num_tokens": 479147.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.046127462387085,
|
|
"epoch": 0.014378145219266714,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.491,
|
|
"mean_token_accuracy": 0.07903510183095933,
|
|
"num_tokens": 487684.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.984290552139282,
|
|
"epoch": 0.014634897812467906,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.4167,
|
|
"mean_token_accuracy": 0.07433984354138375,
|
|
"num_tokens": 496806.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.933219766616821,
|
|
"epoch": 0.014891650405669097,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.4893,
|
|
"mean_token_accuracy": 0.07593215778470039,
|
|
"num_tokens": 505937.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.9124115943908695,
|
|
"epoch": 0.015148402998870289,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.3239,
|
|
"mean_token_accuracy": 0.07792473435401917,
|
|
"num_tokens": 515503.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.034483766555786,
|
|
"epoch": 0.01540515559207148,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.3811,
|
|
"mean_token_accuracy": 0.07625493034720421,
|
|
"num_tokens": 524506.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.836033296585083,
|
|
"epoch": 0.01566190818527267,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.3815,
|
|
"mean_token_accuracy": 0.0733937170356512,
|
|
"num_tokens": 534009.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.863447093963623,
|
|
"epoch": 0.015918660778473863,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.266,
|
|
"mean_token_accuracy": 0.07975648492574691,
|
|
"num_tokens": 543104.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.837308692932129,
|
|
"epoch": 0.016175413371675055,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.2217,
|
|
"mean_token_accuracy": 0.0892428256571293,
|
|
"num_tokens": 551303.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.922349071502685,
|
|
"epoch": 0.016432165964876247,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.4125,
|
|
"mean_token_accuracy": 0.08186463341116905,
|
|
"num_tokens": 559361.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.849794530868531,
|
|
"epoch": 0.016688918558077435,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.333,
|
|
"mean_token_accuracy": 0.08158622235059738,
|
|
"num_tokens": 568629.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.817071199417114,
|
|
"epoch": 0.016945671151278627,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.2763,
|
|
"mean_token_accuracy": 0.08513593450188636,
|
|
"num_tokens": 576776.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.845745038986206,
|
|
"epoch": 0.01720242374447982,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.3776,
|
|
"mean_token_accuracy": 0.07514606490731239,
|
|
"num_tokens": 585970.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.911917018890381,
|
|
"epoch": 0.01745917633768101,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.4101,
|
|
"mean_token_accuracy": 0.07996492646634579,
|
|
"num_tokens": 594837.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.7797877311706545,
|
|
"epoch": 0.017715928930882203,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.3007,
|
|
"mean_token_accuracy": 0.08453992381691933,
|
|
"num_tokens": 602584.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.790009212493897,
|
|
"epoch": 0.017972681524083392,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.3112,
|
|
"mean_token_accuracy": 0.08209161311388016,
|
|
"num_tokens": 611461.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.816317987442017,
|
|
"epoch": 0.018229434117284584,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.2513,
|
|
"mean_token_accuracy": 0.0919821560382843,
|
|
"num_tokens": 619873.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.7200432300567625,
|
|
"epoch": 0.018486186710485776,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.2821,
|
|
"mean_token_accuracy": 0.08631467223167419,
|
|
"num_tokens": 628487.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.797741556167603,
|
|
"epoch": 0.018742939303686968,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.2589,
|
|
"mean_token_accuracy": 0.08670427426695823,
|
|
"num_tokens": 638162.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.808457946777343,
|
|
"epoch": 0.01899969189688816,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.153,
|
|
"mean_token_accuracy": 0.08735950589179993,
|
|
"num_tokens": 646802.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.710161352157593,
|
|
"epoch": 0.01925644449008935,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.2292,
|
|
"mean_token_accuracy": 0.0810220967978239,
|
|
"num_tokens": 655527.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.7011494636535645,
|
|
"epoch": 0.01951319708329054,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.2127,
|
|
"mean_token_accuracy": 0.08935043439269066,
|
|
"num_tokens": 664001.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.72194185256958,
|
|
"epoch": 0.019769949676491733,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.1847,
|
|
"mean_token_accuracy": 0.09060925990343094,
|
|
"num_tokens": 672972.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.752814483642578,
|
|
"epoch": 0.020026702269692925,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.1859,
|
|
"mean_token_accuracy": 0.08640384674072266,
|
|
"num_tokens": 681841.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.733091974258423,
|
|
"epoch": 0.020283454862894117,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.2036,
|
|
"mean_token_accuracy": 0.08569250404834747,
|
|
"num_tokens": 690903.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.624167728424072,
|
|
"epoch": 0.020540207456095305,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.208,
|
|
"mean_token_accuracy": 0.08355596587061882,
|
|
"num_tokens": 700411.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.67492446899414,
|
|
"epoch": 0.020796960049296497,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.107,
|
|
"mean_token_accuracy": 0.08860503733158112,
|
|
"num_tokens": 709062.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.748419570922851,
|
|
"epoch": 0.02105371264249769,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.2563,
|
|
"mean_token_accuracy": 0.0841764748096466,
|
|
"num_tokens": 718726.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.661073493957519,
|
|
"epoch": 0.02131046523569888,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.1345,
|
|
"mean_token_accuracy": 0.0915317751467228,
|
|
"num_tokens": 726710.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.460268259048462,
|
|
"epoch": 0.021567217828900073,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.1529,
|
|
"mean_token_accuracy": 0.09219859838485718,
|
|
"num_tokens": 735145.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.778389644622803,
|
|
"epoch": 0.021823970422101262,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.2522,
|
|
"mean_token_accuracy": 0.0854285180568695,
|
|
"num_tokens": 744267.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.646036958694458,
|
|
"epoch": 0.022080723015302454,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.1583,
|
|
"mean_token_accuracy": 0.08593368083238602,
|
|
"num_tokens": 753253.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.666396236419677,
|
|
"epoch": 0.022337475608503646,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.1065,
|
|
"mean_token_accuracy": 0.09268246218562126,
|
|
"num_tokens": 761697.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.581840705871582,
|
|
"epoch": 0.022594228201704838,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.0921,
|
|
"mean_token_accuracy": 0.09590916931629181,
|
|
"num_tokens": 769745.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.527474021911621,
|
|
"epoch": 0.02285098079490603,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0425,
|
|
"mean_token_accuracy": 0.0901702880859375,
|
|
"num_tokens": 778474.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.621454095840454,
|
|
"epoch": 0.02310773338810722,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.0576,
|
|
"mean_token_accuracy": 0.09153895452618599,
|
|
"num_tokens": 786581.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.626477289199829,
|
|
"epoch": 0.02336448598130841,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.1844,
|
|
"mean_token_accuracy": 0.08645047172904015,
|
|
"num_tokens": 795359.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.494252109527588,
|
|
"epoch": 0.023621238574509602,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.031,
|
|
"mean_token_accuracy": 0.09843903109431267,
|
|
"num_tokens": 802836.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.607794904708863,
|
|
"epoch": 0.023877991167710794,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0743,
|
|
"mean_token_accuracy": 0.09356487020850182,
|
|
"num_tokens": 811749.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.594829416275024,
|
|
"epoch": 0.024134743760911986,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.0839,
|
|
"mean_token_accuracy": 0.09629017040133477,
|
|
"num_tokens": 820315.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.577248907089233,
|
|
"epoch": 0.024391496354113175,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.0196,
|
|
"mean_token_accuracy": 0.09056874141097068,
|
|
"num_tokens": 828983.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.597412490844727,
|
|
"epoch": 0.024648248947314367,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.1117,
|
|
"mean_token_accuracy": 0.09410159885883332,
|
|
"num_tokens": 838622.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.545994901657105,
|
|
"epoch": 0.02490500154051556,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.0061,
|
|
"mean_token_accuracy": 0.09935126826167107,
|
|
"num_tokens": 848037.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.420415019989013,
|
|
"epoch": 0.02516175413371675,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.9918,
|
|
"mean_token_accuracy": 0.09526313245296478,
|
|
"num_tokens": 857048.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.4686620235443115,
|
|
"epoch": 0.025418506726917943,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.907,
|
|
"mean_token_accuracy": 0.09526302069425582,
|
|
"num_tokens": 865120.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.445110416412353,
|
|
"epoch": 0.02567525932011913,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.0197,
|
|
"mean_token_accuracy": 0.0922734022140503,
|
|
"num_tokens": 874061.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.49036660194397,
|
|
"epoch": 0.025932011913320323,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.969,
|
|
"mean_token_accuracy": 0.10023247674107552,
|
|
"num_tokens": 883216.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.455933570861816,
|
|
"epoch": 0.026188764506521515,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.0138,
|
|
"mean_token_accuracy": 0.09378844946622848,
|
|
"num_tokens": 892350.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.489746046066284,
|
|
"epoch": 0.026445517099722708,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.9823,
|
|
"mean_token_accuracy": 0.09784873127937317,
|
|
"num_tokens": 900558.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.49373345375061,
|
|
"epoch": 0.0267022696929239,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.9777,
|
|
"mean_token_accuracy": 0.0942293494939804,
|
|
"num_tokens": 909947.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.483462619781494,
|
|
"epoch": 0.02695902228612509,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.051,
|
|
"mean_token_accuracy": 0.09402686804533004,
|
|
"num_tokens": 918739.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.447958755493164,
|
|
"epoch": 0.02721577487932628,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.9791,
|
|
"mean_token_accuracy": 0.09842836782336235,
|
|
"num_tokens": 927362.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.415892648696899,
|
|
"epoch": 0.027472527472527472,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.0044,
|
|
"mean_token_accuracy": 0.10114640071988105,
|
|
"num_tokens": 936383.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.461966943740845,
|
|
"epoch": 0.027729280065728664,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8998,
|
|
"mean_token_accuracy": 0.10296619608998299,
|
|
"num_tokens": 944997.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.497038841247559,
|
|
"epoch": 0.027986032658929856,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.059,
|
|
"mean_token_accuracy": 0.09460930153727531,
|
|
"num_tokens": 954077.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.522246599197388,
|
|
"epoch": 0.028242785252131048,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.0347,
|
|
"mean_token_accuracy": 0.09197534620761871,
|
|
"num_tokens": 964255.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.422016334533692,
|
|
"epoch": 0.028499537845332237,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.9696,
|
|
"mean_token_accuracy": 0.09874921962618828,
|
|
"num_tokens": 973018.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.3553742408752445,
|
|
"epoch": 0.02875629043853343,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.9592,
|
|
"mean_token_accuracy": 0.09615055844187737,
|
|
"num_tokens": 981991.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.295046949386597,
|
|
"epoch": 0.02901304303173462,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.8329,
|
|
"mean_token_accuracy": 0.10283219069242477,
|
|
"num_tokens": 990627.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.459612798690796,
|
|
"epoch": 0.029269795624935813,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.8763,
|
|
"mean_token_accuracy": 0.09862258359789848,
|
|
"num_tokens": 998596.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.385720443725586,
|
|
"epoch": 0.029526548218137005,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.973,
|
|
"mean_token_accuracy": 0.09788134768605232,
|
|
"num_tokens": 1007175.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.259206295013428,
|
|
"epoch": 0.029783300811338193,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.9401,
|
|
"mean_token_accuracy": 0.09557250440120697,
|
|
"num_tokens": 1015721.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.413903093338012,
|
|
"epoch": 0.030040053404539385,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.993,
|
|
"mean_token_accuracy": 0.09768485501408578,
|
|
"num_tokens": 1024636.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.447688007354737,
|
|
"epoch": 0.030296805997740577,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.9748,
|
|
"mean_token_accuracy": 0.09863441213965415,
|
|
"num_tokens": 1033345.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.267191457748413,
|
|
"epoch": 0.03055355859094177,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.8654,
|
|
"mean_token_accuracy": 0.10642952471971512,
|
|
"num_tokens": 1042145.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.3619465827941895,
|
|
"epoch": 0.03081031118414296,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.881,
|
|
"mean_token_accuracy": 0.10219442248344421,
|
|
"num_tokens": 1050989.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.355409574508667,
|
|
"epoch": 0.03106706377734415,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.8932,
|
|
"mean_token_accuracy": 0.10146859213709832,
|
|
"num_tokens": 1059523.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.273333930969239,
|
|
"epoch": 0.03132381637054534,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.9519,
|
|
"mean_token_accuracy": 0.09742964655160904,
|
|
"num_tokens": 1068597.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.319173955917359,
|
|
"epoch": 0.03158056896374654,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.8821,
|
|
"mean_token_accuracy": 0.1009488008916378,
|
|
"num_tokens": 1076687.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.361953592300415,
|
|
"epoch": 0.031837321556947726,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.8987,
|
|
"mean_token_accuracy": 0.10232065096497536,
|
|
"num_tokens": 1085473.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.328115177154541,
|
|
"epoch": 0.032094074150148914,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.9003,
|
|
"mean_token_accuracy": 0.10789056047797203,
|
|
"num_tokens": 1093755.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.14094033241272,
|
|
"epoch": 0.03235082674335011,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7555,
|
|
"mean_token_accuracy": 0.10103646516799927,
|
|
"num_tokens": 1103555.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.497932863235474,
|
|
"epoch": 0.0326075793365513,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.8988,
|
|
"mean_token_accuracy": 0.10374749824404716,
|
|
"num_tokens": 1112158.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.159998369216919,
|
|
"epoch": 0.032864331929752494,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7945,
|
|
"mean_token_accuracy": 0.10642132759094239,
|
|
"num_tokens": 1120512.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.276958751678467,
|
|
"epoch": 0.03312108452295368,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.9121,
|
|
"mean_token_accuracy": 0.1034335196018219,
|
|
"num_tokens": 1128879.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.273133230209351,
|
|
"epoch": 0.03337783711615487,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.9686,
|
|
"mean_token_accuracy": 0.09752831980586052,
|
|
"num_tokens": 1137366.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.269780445098877,
|
|
"epoch": 0.033634589709356066,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.7687,
|
|
"mean_token_accuracy": 0.10863290056586265,
|
|
"num_tokens": 1146535.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.186447620391846,
|
|
"epoch": 0.033891342302557255,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.8351,
|
|
"mean_token_accuracy": 0.10125730335712432,
|
|
"num_tokens": 1155043.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.206981992721557,
|
|
"epoch": 0.03414809489575845,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.8185,
|
|
"mean_token_accuracy": 0.10186112374067306,
|
|
"num_tokens": 1163703.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.267451667785645,
|
|
"epoch": 0.03440484748895964,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.8533,
|
|
"mean_token_accuracy": 0.10344401970505715,
|
|
"num_tokens": 1171770.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.236594724655151,
|
|
"epoch": 0.03466160008216083,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.7989,
|
|
"mean_token_accuracy": 0.10623008012771606,
|
|
"num_tokens": 1180717.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.21160855293274,
|
|
"epoch": 0.03491835267536202,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.7809,
|
|
"mean_token_accuracy": 0.10603002682328225,
|
|
"num_tokens": 1188913.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.069119596481324,
|
|
"epoch": 0.03517510526856321,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.7651,
|
|
"mean_token_accuracy": 0.10698620229959488,
|
|
"num_tokens": 1197572.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.312754535675049,
|
|
"epoch": 0.03543185786176441,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7779,
|
|
"mean_token_accuracy": 0.10408765450119972,
|
|
"num_tokens": 1206248.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.221107006072998,
|
|
"epoch": 0.035688610454965596,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.8614,
|
|
"mean_token_accuracy": 0.10458310097455978,
|
|
"num_tokens": 1214834.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.180823087692261,
|
|
"epoch": 0.035945363048166784,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.8282,
|
|
"mean_token_accuracy": 0.10219828709959984,
|
|
"num_tokens": 1224493.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.227695560455322,
|
|
"epoch": 0.03620211564136798,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.7849,
|
|
"mean_token_accuracy": 0.10087160542607307,
|
|
"num_tokens": 1233706.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.126480054855347,
|
|
"epoch": 0.03645886823456917,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.8267,
|
|
"mean_token_accuracy": 0.10416910126805305,
|
|
"num_tokens": 1242897.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.2293792247772215,
|
|
"epoch": 0.036715620827770364,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.7923,
|
|
"mean_token_accuracy": 0.1014643669128418,
|
|
"num_tokens": 1250920.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.1777020454406735,
|
|
"epoch": 0.03697237342097155,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.7983,
|
|
"mean_token_accuracy": 0.10312379300594329,
|
|
"num_tokens": 1258978.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.225130081176758,
|
|
"epoch": 0.03722912601417274,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.7935,
|
|
"mean_token_accuracy": 0.10582225993275643,
|
|
"num_tokens": 1267302.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.090721082687378,
|
|
"epoch": 0.037485878607373936,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6958,
|
|
"mean_token_accuracy": 0.10774607136845589,
|
|
"num_tokens": 1275469.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.144723510742187,
|
|
"epoch": 0.037742631200575125,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.7851,
|
|
"mean_token_accuracy": 0.10810397937893867,
|
|
"num_tokens": 1283247.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.263766145706176,
|
|
"epoch": 0.03799938379377632,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.772,
|
|
"mean_token_accuracy": 0.1056891031563282,
|
|
"num_tokens": 1291501.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.1570065975189205,
|
|
"epoch": 0.03825613638697751,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.8218,
|
|
"mean_token_accuracy": 0.10072930157184601,
|
|
"num_tokens": 1300677.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.055219650268555,
|
|
"epoch": 0.0385128889801787,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.7675,
|
|
"mean_token_accuracy": 0.10408244505524636,
|
|
"num_tokens": 1309499.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.173019313812256,
|
|
"epoch": 0.03876964157337989,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6451,
|
|
"mean_token_accuracy": 0.10914059057831764,
|
|
"num_tokens": 1317929.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.017005825042725,
|
|
"epoch": 0.03902639416658108,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.6827,
|
|
"mean_token_accuracy": 0.11490903943777084,
|
|
"num_tokens": 1326707.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.176887035369873,
|
|
"epoch": 0.03928314675978228,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.7504,
|
|
"mean_token_accuracy": 0.1038425773382187,
|
|
"num_tokens": 1335481.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.086644124984741,
|
|
"epoch": 0.039539899352983465,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.6711,
|
|
"mean_token_accuracy": 0.10754814520478248,
|
|
"num_tokens": 1344880.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.088477706909179,
|
|
"epoch": 0.039796651946184654,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.7537,
|
|
"mean_token_accuracy": 0.1024178683757782,
|
|
"num_tokens": 1353711.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.03897066116333,
|
|
"epoch": 0.04005340453938585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.6974,
|
|
"mean_token_accuracy": 0.10224836692214012,
|
|
"num_tokens": 1363159.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.968532657623291,
|
|
"epoch": 0.04031015713258704,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.594,
|
|
"mean_token_accuracy": 0.11433639004826546,
|
|
"num_tokens": 1371396.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.022575330734253,
|
|
"epoch": 0.04056690972578823,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5903,
|
|
"mean_token_accuracy": 0.11093990504741669,
|
|
"num_tokens": 1379862.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.060764265060425,
|
|
"epoch": 0.04082366231898942,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.6821,
|
|
"mean_token_accuracy": 0.11073108166456222,
|
|
"num_tokens": 1388220.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.100030374526978,
|
|
"epoch": 0.04108041491219061,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.834,
|
|
"mean_token_accuracy": 0.10184895843267441,
|
|
"num_tokens": 1397409.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.060917663574219,
|
|
"epoch": 0.041337167505391806,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.659,
|
|
"mean_token_accuracy": 0.11565897613763809,
|
|
"num_tokens": 1405967.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.036740016937256,
|
|
"epoch": 0.041593920098592994,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.7056,
|
|
"mean_token_accuracy": 0.10764685571193695,
|
|
"num_tokens": 1414107.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.039219760894776,
|
|
"epoch": 0.04185067269179419,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.6238,
|
|
"mean_token_accuracy": 0.10947153940796853,
|
|
"num_tokens": 1422834.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.04460973739624,
|
|
"epoch": 0.04210742528499538,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.7356,
|
|
"mean_token_accuracy": 0.10661220997571945,
|
|
"num_tokens": 1432053.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.139359998703003,
|
|
"epoch": 0.04236417787819657,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6804,
|
|
"mean_token_accuracy": 0.10453324764966965,
|
|
"num_tokens": 1440562.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.938592052459716,
|
|
"epoch": 0.04262093047139776,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.6066,
|
|
"mean_token_accuracy": 0.11469024196267127,
|
|
"num_tokens": 1449202.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.033561897277832,
|
|
"epoch": 0.04287768306459895,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.6811,
|
|
"mean_token_accuracy": 0.10556919798254967,
|
|
"num_tokens": 1457994.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.035749006271362,
|
|
"epoch": 0.043134435657800146,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5594,
|
|
"mean_token_accuracy": 0.11265857443213463,
|
|
"num_tokens": 1467610.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.930410718917846,
|
|
"epoch": 0.043391188251001335,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.6115,
|
|
"mean_token_accuracy": 0.11150972470641136,
|
|
"num_tokens": 1475070.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.036338949203492,
|
|
"epoch": 0.043647940844202524,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.6788,
|
|
"mean_token_accuracy": 0.10261262133717537,
|
|
"num_tokens": 1483813.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.053100967407227,
|
|
"epoch": 0.04390469343740372,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.6764,
|
|
"mean_token_accuracy": 0.10689589381217957,
|
|
"num_tokens": 1492261.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.972715282440186,
|
|
"epoch": 0.04416144603060491,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.6628,
|
|
"mean_token_accuracy": 0.1079073965549469,
|
|
"num_tokens": 1501036.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.978304815292359,
|
|
"epoch": 0.0444181986238061,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.6607,
|
|
"mean_token_accuracy": 0.1067041240632534,
|
|
"num_tokens": 1509755.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.03783950805664,
|
|
"epoch": 0.04467495121700729,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.662,
|
|
"mean_token_accuracy": 0.10334146693348885,
|
|
"num_tokens": 1518433.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.073136329650879,
|
|
"epoch": 0.04493170381020848,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.7077,
|
|
"mean_token_accuracy": 0.1048253409564495,
|
|
"num_tokens": 1527501.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.931898355484009,
|
|
"epoch": 0.045188456403409676,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5618,
|
|
"mean_token_accuracy": 0.11353855803608895,
|
|
"num_tokens": 1536299.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.8974997997283936,
|
|
"epoch": 0.045445208996610864,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.6404,
|
|
"mean_token_accuracy": 0.10956602469086647,
|
|
"num_tokens": 1544408.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.950430822372437,
|
|
"epoch": 0.04570196158981206,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.5004,
|
|
"mean_token_accuracy": 0.11693055927753448,
|
|
"num_tokens": 1552564.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.900494909286499,
|
|
"epoch": 0.04595871418301325,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.5678,
|
|
"mean_token_accuracy": 0.11564396619796753,
|
|
"num_tokens": 1561378.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.850385761260986,
|
|
"epoch": 0.04621546677621444,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5308,
|
|
"mean_token_accuracy": 0.11212190836668015,
|
|
"num_tokens": 1570663.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.92402753829956,
|
|
"epoch": 0.04647221936941563,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.629,
|
|
"mean_token_accuracy": 0.11024776846170425,
|
|
"num_tokens": 1579681.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.058414793014526,
|
|
"epoch": 0.04672897196261682,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.6784,
|
|
"mean_token_accuracy": 0.10603147745132446,
|
|
"num_tokens": 1588680.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.741462564468383,
|
|
"epoch": 0.046985724555818016,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4856,
|
|
"mean_token_accuracy": 0.11496321633458137,
|
|
"num_tokens": 1597141.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.009930515289307,
|
|
"epoch": 0.047242477149019205,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5692,
|
|
"mean_token_accuracy": 0.11231728866696358,
|
|
"num_tokens": 1606314.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.935445070266724,
|
|
"epoch": 0.04749922974222039,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.705,
|
|
"mean_token_accuracy": 0.10736741870641708,
|
|
"num_tokens": 1615822.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.000515937805176,
|
|
"epoch": 0.04775598233542159,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.6088,
|
|
"mean_token_accuracy": 0.10640651285648346,
|
|
"num_tokens": 1624630.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.91909384727478,
|
|
"epoch": 0.04801273492862278,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.5477,
|
|
"mean_token_accuracy": 0.11673416793346406,
|
|
"num_tokens": 1632416.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.916115617752075,
|
|
"epoch": 0.04826948752182397,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.5873,
|
|
"mean_token_accuracy": 0.10929609909653663,
|
|
"num_tokens": 1641619.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.8772941589355465,
|
|
"epoch": 0.04852624011502516,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.5304,
|
|
"mean_token_accuracy": 0.11052975505590439,
|
|
"num_tokens": 1649752.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.90340223312378,
|
|
"epoch": 0.04878299270822635,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.5347,
|
|
"mean_token_accuracy": 0.11766225099563599,
|
|
"num_tokens": 1658074.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.747793102264405,
|
|
"epoch": 0.049039745301427545,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4964,
|
|
"mean_token_accuracy": 0.11006889939308166,
|
|
"num_tokens": 1666767.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.807773399353027,
|
|
"epoch": 0.049296497894628734,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4424,
|
|
"mean_token_accuracy": 0.1198193997144699,
|
|
"num_tokens": 1674952.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.811061191558838,
|
|
"epoch": 0.04955325048782993,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.5249,
|
|
"mean_token_accuracy": 0.11521950364112854,
|
|
"num_tokens": 1683642.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.836861085891724,
|
|
"epoch": 0.04981000308103112,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.5997,
|
|
"mean_token_accuracy": 0.1094896785914898,
|
|
"num_tokens": 1692501.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.962518167495728,
|
|
"epoch": 0.050066755674232306,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.5998,
|
|
"mean_token_accuracy": 0.109241334348917,
|
|
"num_tokens": 1702210.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.7919018268585205,
|
|
"epoch": 0.0503235082674335,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5486,
|
|
"mean_token_accuracy": 0.10999819412827491,
|
|
"num_tokens": 1710942.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.740083837509156,
|
|
"epoch": 0.05058026086063469,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.4648,
|
|
"mean_token_accuracy": 0.11652028635144233,
|
|
"num_tokens": 1719876.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.961598348617554,
|
|
"epoch": 0.050837013453835886,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.5059,
|
|
"mean_token_accuracy": 0.11878032237291336,
|
|
"num_tokens": 1728259.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.795040130615234,
|
|
"epoch": 0.051093766047037074,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.5472,
|
|
"mean_token_accuracy": 0.10858382433652877,
|
|
"num_tokens": 1737165.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.943842172622681,
|
|
"epoch": 0.05135051864023826,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.6405,
|
|
"mean_token_accuracy": 0.10827328190207482,
|
|
"num_tokens": 1746090.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.866076374053955,
|
|
"epoch": 0.05160727123343946,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.58,
|
|
"mean_token_accuracy": 0.10802414938807488,
|
|
"num_tokens": 1754140.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.794606924057007,
|
|
"epoch": 0.05186402382664065,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.497,
|
|
"mean_token_accuracy": 0.11030312180519104,
|
|
"num_tokens": 1763724.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.799782228469849,
|
|
"epoch": 0.05212077641984184,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.4175,
|
|
"mean_token_accuracy": 0.1173188902437687,
|
|
"num_tokens": 1771634.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.750743007659912,
|
|
"epoch": 0.05237752901304303,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.5607,
|
|
"mean_token_accuracy": 0.1105339378118515,
|
|
"num_tokens": 1780380.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.859860372543335,
|
|
"epoch": 0.052634281606244226,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.5258,
|
|
"mean_token_accuracy": 0.11754016950726509,
|
|
"num_tokens": 1788716.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.786990356445313,
|
|
"epoch": 0.052891034199445415,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.4883,
|
|
"mean_token_accuracy": 0.11224759444594383,
|
|
"num_tokens": 1797666.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.84674482345581,
|
|
"epoch": 0.053147786792646604,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.6177,
|
|
"mean_token_accuracy": 0.11337128654122353,
|
|
"num_tokens": 1806953.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.859242725372314,
|
|
"epoch": 0.0534045393858478,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.6235,
|
|
"mean_token_accuracy": 0.10895855575799943,
|
|
"num_tokens": 1815832.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.89027624130249,
|
|
"epoch": 0.05366129197904899,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.5896,
|
|
"mean_token_accuracy": 0.11460649147629738,
|
|
"num_tokens": 1825147.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.851807546615601,
|
|
"epoch": 0.05391804457225018,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.5497,
|
|
"mean_token_accuracy": 0.11144849956035614,
|
|
"num_tokens": 1833943.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.800374555587768,
|
|
"epoch": 0.05417479716545137,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.5816,
|
|
"mean_token_accuracy": 0.10899447724223137,
|
|
"num_tokens": 1841946.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.796955204010009,
|
|
"epoch": 0.05443154975865256,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.4747,
|
|
"mean_token_accuracy": 0.1164444699883461,
|
|
"num_tokens": 1850603.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.83609676361084,
|
|
"epoch": 0.054688302351853756,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.5192,
|
|
"mean_token_accuracy": 0.10918731167912483,
|
|
"num_tokens": 1859738.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.747317695617676,
|
|
"epoch": 0.054945054945054944,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.4835,
|
|
"mean_token_accuracy": 0.11204507797956467,
|
|
"num_tokens": 1868173.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.896972227096557,
|
|
"epoch": 0.05520180753825614,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.5569,
|
|
"mean_token_accuracy": 0.11272180378437043,
|
|
"num_tokens": 1876573.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.717895078659057,
|
|
"epoch": 0.05545856013145733,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.4727,
|
|
"mean_token_accuracy": 0.1115611620247364,
|
|
"num_tokens": 1886083.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.798813199996948,
|
|
"epoch": 0.05571531272465852,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.4569,
|
|
"mean_token_accuracy": 0.11852912381291389,
|
|
"num_tokens": 1894914.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.8052410125732425,
|
|
"epoch": 0.05597206531785971,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.6038,
|
|
"mean_token_accuracy": 0.11169130429625511,
|
|
"num_tokens": 1904106.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.71564884185791,
|
|
"epoch": 0.0562288179110609,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.3691,
|
|
"mean_token_accuracy": 0.11713010892271995,
|
|
"num_tokens": 1912805.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.893805932998657,
|
|
"epoch": 0.056485570504262096,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.6222,
|
|
"mean_token_accuracy": 0.11934545859694481,
|
|
"num_tokens": 1920866.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.685084056854248,
|
|
"epoch": 0.056742323097463285,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.4528,
|
|
"mean_token_accuracy": 0.11949897482991219,
|
|
"num_tokens": 1929134.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.812266635894775,
|
|
"epoch": 0.05699907569066447,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.3809,
|
|
"mean_token_accuracy": 0.11728307008743286,
|
|
"num_tokens": 1937812.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.649289751052857,
|
|
"epoch": 0.05725582828386567,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.4805,
|
|
"mean_token_accuracy": 0.1128297932446003,
|
|
"num_tokens": 1947243.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.804856681823731,
|
|
"epoch": 0.05751258087706686,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.501,
|
|
"mean_token_accuracy": 0.11157863885164261,
|
|
"num_tokens": 1956202.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.721863031387329,
|
|
"epoch": 0.05776933347026805,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.4611,
|
|
"mean_token_accuracy": 0.11436681747436524,
|
|
"num_tokens": 1964633.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.773561477661133,
|
|
"epoch": 0.05802608606346924,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.4021,
|
|
"mean_token_accuracy": 0.11736030727624894,
|
|
"num_tokens": 1973184.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.543147563934326,
|
|
"epoch": 0.05828283865667043,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.3447,
|
|
"mean_token_accuracy": 0.11813951656222343,
|
|
"num_tokens": 1982315.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.890185117721558,
|
|
"epoch": 0.058539591249871625,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.4687,
|
|
"mean_token_accuracy": 0.11018439456820488,
|
|
"num_tokens": 1990956.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.554495763778687,
|
|
"epoch": 0.058796343843072814,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.4443,
|
|
"mean_token_accuracy": 0.11464114412665367,
|
|
"num_tokens": 2000475.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.679226207733154,
|
|
"epoch": 0.05905309643627401,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.4058,
|
|
"mean_token_accuracy": 0.11374457627534866,
|
|
"num_tokens": 2009803.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.617973184585571,
|
|
"epoch": 0.0593098490294752,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.369,
|
|
"mean_token_accuracy": 0.12533641532063483,
|
|
"num_tokens": 2019107.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.765725994110108,
|
|
"epoch": 0.059566601622676386,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.4291,
|
|
"mean_token_accuracy": 0.12166900932788849,
|
|
"num_tokens": 2027702.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.625709915161133,
|
|
"epoch": 0.05982335421587758,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.4263,
|
|
"mean_token_accuracy": 0.11392560228705406,
|
|
"num_tokens": 2037027.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.8339520454406735,
|
|
"epoch": 0.06008010680907877,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.446,
|
|
"mean_token_accuracy": 0.11708763837814332,
|
|
"num_tokens": 2045583.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.497016763687133,
|
|
"epoch": 0.060336859402279966,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.3324,
|
|
"mean_token_accuracy": 0.12295132800936699,
|
|
"num_tokens": 2053645.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.738676404953003,
|
|
"epoch": 0.060593611995481154,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.4177,
|
|
"mean_token_accuracy": 0.12177978456020355,
|
|
"num_tokens": 2062365.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.693480539321899,
|
|
"epoch": 0.06085036458868234,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.5582,
|
|
"mean_token_accuracy": 0.11437630504369736,
|
|
"num_tokens": 2072356.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.697644424438477,
|
|
"epoch": 0.06110711718188354,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.3728,
|
|
"mean_token_accuracy": 0.12430283352732659,
|
|
"num_tokens": 2080885.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.703748273849487,
|
|
"epoch": 0.06136386977508473,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.4242,
|
|
"mean_token_accuracy": 0.11977330669760704,
|
|
"num_tokens": 2089747.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.676883506774902,
|
|
"epoch": 0.06162062236828592,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.5105,
|
|
"mean_token_accuracy": 0.11225128918886185,
|
|
"num_tokens": 2099092.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.754091215133667,
|
|
"epoch": 0.06187737496148711,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.3861,
|
|
"mean_token_accuracy": 0.1250210888683796,
|
|
"num_tokens": 2107452.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.581334209442138,
|
|
"epoch": 0.0621341275546883,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.3821,
|
|
"mean_token_accuracy": 0.11902668997645378,
|
|
"num_tokens": 2115711.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.75599627494812,
|
|
"epoch": 0.062390880147889495,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.3802,
|
|
"mean_token_accuracy": 0.12134013175964356,
|
|
"num_tokens": 2123662.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.690904331207276,
|
|
"epoch": 0.06264763274109068,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.3967,
|
|
"mean_token_accuracy": 0.12150893285870552,
|
|
"num_tokens": 2132328.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.722704267501831,
|
|
"epoch": 0.06290438533429188,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.4939,
|
|
"mean_token_accuracy": 0.12254033461213112,
|
|
"num_tokens": 2141110.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.65714168548584,
|
|
"epoch": 0.06316113792749307,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.2619,
|
|
"mean_token_accuracy": 0.12005072087049484,
|
|
"num_tokens": 2150337.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.558941221237182,
|
|
"epoch": 0.06341789052069426,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.3837,
|
|
"mean_token_accuracy": 0.12301670834422111,
|
|
"num_tokens": 2159125.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.614150428771973,
|
|
"epoch": 0.06367464311389545,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.341,
|
|
"mean_token_accuracy": 0.1229495607316494,
|
|
"num_tokens": 2168176.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.664106941223144,
|
|
"epoch": 0.06393139570709665,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.3347,
|
|
"mean_token_accuracy": 0.11931615918874741,
|
|
"num_tokens": 2176789.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.543559122085571,
|
|
"epoch": 0.06418814830029783,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.3282,
|
|
"mean_token_accuracy": 0.11372191831469536,
|
|
"num_tokens": 2185424.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.579632949829102,
|
|
"epoch": 0.06444490089349902,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.2639,
|
|
"mean_token_accuracy": 0.1248880796134472,
|
|
"num_tokens": 2194379.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.624179458618164,
|
|
"epoch": 0.06470165348670022,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.3308,
|
|
"mean_token_accuracy": 0.12357598170638084,
|
|
"num_tokens": 2203098.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.557571840286255,
|
|
"epoch": 0.0649584060799014,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.3617,
|
|
"mean_token_accuracy": 0.11703156158328057,
|
|
"num_tokens": 2212019.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.752705526351929,
|
|
"epoch": 0.0652151586731026,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.3258,
|
|
"mean_token_accuracy": 0.1183007724583149,
|
|
"num_tokens": 2221139.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.4972742080688475,
|
|
"epoch": 0.06547191126630379,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.226,
|
|
"mean_token_accuracy": 0.12040054500102997,
|
|
"num_tokens": 2229506.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.584129858016968,
|
|
"epoch": 0.06572866385950499,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.3198,
|
|
"mean_token_accuracy": 0.12376804128289223,
|
|
"num_tokens": 2238134.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.553785991668701,
|
|
"epoch": 0.06598541645270617,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.2643,
|
|
"mean_token_accuracy": 0.1191803939640522,
|
|
"num_tokens": 2245338.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.744561767578125,
|
|
"epoch": 0.06624216904590736,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.4527,
|
|
"mean_token_accuracy": 0.1175347812473774,
|
|
"num_tokens": 2254440.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.616067266464233,
|
|
"epoch": 0.06649892163910856,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.3764,
|
|
"mean_token_accuracy": 0.11979993656277657,
|
|
"num_tokens": 2263530.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.696237468719483,
|
|
"epoch": 0.06675567423230974,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.453,
|
|
"mean_token_accuracy": 0.11864392161369323,
|
|
"num_tokens": 2272520.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.517717504501343,
|
|
"epoch": 0.06701242682551094,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.2568,
|
|
"mean_token_accuracy": 0.1262233830988407,
|
|
"num_tokens": 2280720.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.69150938987732,
|
|
"epoch": 0.06726917941871213,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.3982,
|
|
"mean_token_accuracy": 0.12398588731884956,
|
|
"num_tokens": 2289381.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.549073600769043,
|
|
"epoch": 0.06752593201191331,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.3307,
|
|
"mean_token_accuracy": 0.12101859003305435,
|
|
"num_tokens": 2298281.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.666070938110352,
|
|
"epoch": 0.06778268460511451,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.3453,
|
|
"mean_token_accuracy": 0.12255151793360711,
|
|
"num_tokens": 2307158.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.485422849655151,
|
|
"epoch": 0.0680394371983157,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.2792,
|
|
"mean_token_accuracy": 0.12860881313681602,
|
|
"num_tokens": 2316080.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.554500436782837,
|
|
"epoch": 0.0682961897915169,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.278,
|
|
"mean_token_accuracy": 0.12202934175729752,
|
|
"num_tokens": 2325553.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.5212348937988285,
|
|
"epoch": 0.06855294238471808,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.2912,
|
|
"mean_token_accuracy": 0.12347806841135026,
|
|
"num_tokens": 2333889.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.611920976638794,
|
|
"epoch": 0.06880969497791928,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.254,
|
|
"mean_token_accuracy": 0.12139070555567741,
|
|
"num_tokens": 2342963.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.4503942966461185,
|
|
"epoch": 0.06906644757112047,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.2802,
|
|
"mean_token_accuracy": 0.12291336581110954,
|
|
"num_tokens": 2352165.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.6296186447143555,
|
|
"epoch": 0.06932320016432165,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.3895,
|
|
"mean_token_accuracy": 0.1163854293525219,
|
|
"num_tokens": 2361407.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.474624252319336,
|
|
"epoch": 0.06957995275752285,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.2308,
|
|
"mean_token_accuracy": 0.126140808314085,
|
|
"num_tokens": 2370274.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.568676233291626,
|
|
"epoch": 0.06983670535072405,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.2486,
|
|
"mean_token_accuracy": 0.12546533793210984,
|
|
"num_tokens": 2379393.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.570184230804443,
|
|
"epoch": 0.07009345794392523,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.3764,
|
|
"mean_token_accuracy": 0.12241309732198716,
|
|
"num_tokens": 2389523.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.524905490875244,
|
|
"epoch": 0.07035021053712642,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.2468,
|
|
"mean_token_accuracy": 0.13126732558012008,
|
|
"num_tokens": 2398330.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.623579740524292,
|
|
"epoch": 0.07060696313032762,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.3661,
|
|
"mean_token_accuracy": 0.12497142404317856,
|
|
"num_tokens": 2406205.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.502502012252807,
|
|
"epoch": 0.07086371572352881,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.281,
|
|
"mean_token_accuracy": 0.11763083785772324,
|
|
"num_tokens": 2415283.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.6591963291168215,
|
|
"epoch": 0.07112046831673,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.3112,
|
|
"mean_token_accuracy": 0.11914715766906739,
|
|
"num_tokens": 2424470.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.566618394851685,
|
|
"epoch": 0.07137722090993119,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.3308,
|
|
"mean_token_accuracy": 0.11918682977557182,
|
|
"num_tokens": 2432984.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.516817951202393,
|
|
"epoch": 0.07163397350313239,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.2568,
|
|
"mean_token_accuracy": 0.11953618749976158,
|
|
"num_tokens": 2441918.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.526464319229126,
|
|
"epoch": 0.07189072609633357,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.2549,
|
|
"mean_token_accuracy": 0.12539261057972909,
|
|
"num_tokens": 2450159.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.465778255462647,
|
|
"epoch": 0.07214747868953476,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.1942,
|
|
"mean_token_accuracy": 0.1224373534321785,
|
|
"num_tokens": 2458625.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.572404193878174,
|
|
"epoch": 0.07240423128273596,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.3143,
|
|
"mean_token_accuracy": 0.12915984988212587,
|
|
"num_tokens": 2466979.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.493269205093384,
|
|
"epoch": 0.07266098387593714,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.2283,
|
|
"mean_token_accuracy": 0.12615607306361198,
|
|
"num_tokens": 2475509.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.4519850730896,
|
|
"epoch": 0.07291773646913834,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.2588,
|
|
"mean_token_accuracy": 0.12535108551383017,
|
|
"num_tokens": 2484232.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.5603673458099365,
|
|
"epoch": 0.07317448906233953,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.258,
|
|
"mean_token_accuracy": 0.12554760426282882,
|
|
"num_tokens": 2492466.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.505724287033081,
|
|
"epoch": 0.07343124165554073,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.2029,
|
|
"mean_token_accuracy": 0.126475278288126,
|
|
"num_tokens": 2500656.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.416388511657715,
|
|
"epoch": 0.07368799424874191,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.1126,
|
|
"mean_token_accuracy": 0.13116336688399316,
|
|
"num_tokens": 2509192.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.421076345443725,
|
|
"epoch": 0.0739447468419431,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.2209,
|
|
"mean_token_accuracy": 0.1266927421092987,
|
|
"num_tokens": 2517600.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.463506746292114,
|
|
"epoch": 0.0742014994351443,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.0954,
|
|
"mean_token_accuracy": 0.13116538003087044,
|
|
"num_tokens": 2526063.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.522179746627808,
|
|
"epoch": 0.07445825202834548,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.2054,
|
|
"mean_token_accuracy": 0.13104380965232848,
|
|
"num_tokens": 2535038.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.385414266586304,
|
|
"epoch": 0.07471500462154668,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.0917,
|
|
"mean_token_accuracy": 0.1255633920431137,
|
|
"num_tokens": 2543779.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.491334104537964,
|
|
"epoch": 0.07497175721474787,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.2104,
|
|
"mean_token_accuracy": 0.12376456633210182,
|
|
"num_tokens": 2552584.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.575016212463379,
|
|
"epoch": 0.07522850980794905,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.3035,
|
|
"mean_token_accuracy": 0.1276801899075508,
|
|
"num_tokens": 2560543.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.361924266815185,
|
|
"epoch": 0.07548526240115025,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.1464,
|
|
"mean_token_accuracy": 0.1312977321445942,
|
|
"num_tokens": 2569103.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.535239553451538,
|
|
"epoch": 0.07574201499435144,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.2299,
|
|
"mean_token_accuracy": 0.13009746447205545,
|
|
"num_tokens": 2576672.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.404985857009888,
|
|
"epoch": 0.07599876758755264,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.2071,
|
|
"mean_token_accuracy": 0.12912483885884285,
|
|
"num_tokens": 2585098.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.518115711212158,
|
|
"epoch": 0.07625552018075382,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.2212,
|
|
"mean_token_accuracy": 0.1286090739071369,
|
|
"num_tokens": 2593885.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.507031106948853,
|
|
"epoch": 0.07651227277395502,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.1966,
|
|
"mean_token_accuracy": 0.13003123849630355,
|
|
"num_tokens": 2602818.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.414033508300781,
|
|
"epoch": 0.07676902536715621,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.2693,
|
|
"mean_token_accuracy": 0.12209416180849075,
|
|
"num_tokens": 2611792.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.601748752593994,
|
|
"epoch": 0.0770257779603574,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.2734,
|
|
"mean_token_accuracy": 0.12304662093520165,
|
|
"num_tokens": 2620208.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.451296186447143,
|
|
"epoch": 0.07728253055355859,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 6.2473,
|
|
"mean_token_accuracy": 0.12112687975168228,
|
|
"num_tokens": 2629358.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.426006269454956,
|
|
"epoch": 0.07753928314675979,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.1278,
|
|
"mean_token_accuracy": 0.13388494476675988,
|
|
"num_tokens": 2637253.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.387893152236939,
|
|
"epoch": 0.07779603573996097,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.106,
|
|
"mean_token_accuracy": 0.1332385241985321,
|
|
"num_tokens": 2645840.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.491432237625122,
|
|
"epoch": 0.07805278833316216,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.1346,
|
|
"mean_token_accuracy": 0.12694116234779357,
|
|
"num_tokens": 2655280.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.4490635871887205,
|
|
"epoch": 0.07830954092636336,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.2092,
|
|
"mean_token_accuracy": 0.12281604930758476,
|
|
"num_tokens": 2664611.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.398184156417846,
|
|
"epoch": 0.07856629351956455,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.2421,
|
|
"mean_token_accuracy": 0.12447008788585663,
|
|
"num_tokens": 2673175.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.463336324691772,
|
|
"epoch": 0.07882304611276574,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.2439,
|
|
"mean_token_accuracy": 0.1250321723520756,
|
|
"num_tokens": 2681814.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.457613706588745,
|
|
"epoch": 0.07907979870596693,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.1399,
|
|
"mean_token_accuracy": 0.13413845673203467,
|
|
"num_tokens": 2690836.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.483572340011596,
|
|
"epoch": 0.07933655129916813,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.2204,
|
|
"mean_token_accuracy": 0.12513797655701636,
|
|
"num_tokens": 2699452.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.477361106872559,
|
|
"epoch": 0.07959330389236931,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.2412,
|
|
"mean_token_accuracy": 0.12649901136755942,
|
|
"num_tokens": 2708537.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.574354267120361,
|
|
"epoch": 0.0798500564855705,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.1869,
|
|
"mean_token_accuracy": 0.12899849116802214,
|
|
"num_tokens": 2716694.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.290395975112915,
|
|
"epoch": 0.0801068090787717,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.0791,
|
|
"mean_token_accuracy": 0.12540727928280831,
|
|
"num_tokens": 2725818.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.355536270141601,
|
|
"epoch": 0.08036356167197288,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 6.1347,
|
|
"mean_token_accuracy": 0.138398939371109,
|
|
"num_tokens": 2734297.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.482157802581787,
|
|
"epoch": 0.08062031426517408,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 6.1795,
|
|
"mean_token_accuracy": 0.12599481418728828,
|
|
"num_tokens": 2742983.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.387547016143799,
|
|
"epoch": 0.08087706685837527,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.1209,
|
|
"mean_token_accuracy": 0.1289542905986309,
|
|
"num_tokens": 2751826.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.356708908081055,
|
|
"epoch": 0.08113381945157647,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 5.9788,
|
|
"mean_token_accuracy": 0.13213314935564996,
|
|
"num_tokens": 2760037.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.388432455062866,
|
|
"epoch": 0.08139057204477765,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 6.1962,
|
|
"mean_token_accuracy": 0.13228445798158645,
|
|
"num_tokens": 2768861.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.343868684768677,
|
|
"epoch": 0.08164732463797884,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.1449,
|
|
"mean_token_accuracy": 0.12374370470643044,
|
|
"num_tokens": 2777452.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.464413261413574,
|
|
"epoch": 0.08190407723118004,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.2313,
|
|
"mean_token_accuracy": 0.1225876472890377,
|
|
"num_tokens": 2785913.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.525475645065308,
|
|
"epoch": 0.08216082982438122,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 6.0815,
|
|
"mean_token_accuracy": 0.13376436457037927,
|
|
"num_tokens": 2794800.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.417443084716797,
|
|
"epoch": 0.08241758241758242,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 6.2696,
|
|
"mean_token_accuracy": 0.12571724206209184,
|
|
"num_tokens": 2804228.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.354932069778442,
|
|
"epoch": 0.08267433501078361,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 6.0795,
|
|
"mean_token_accuracy": 0.12980363443493842,
|
|
"num_tokens": 2813184.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.376421117782593,
|
|
"epoch": 0.08293108760398481,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.2246,
|
|
"mean_token_accuracy": 0.12621781155467032,
|
|
"num_tokens": 2821952.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.44928126335144,
|
|
"epoch": 0.08318784019718599,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.083,
|
|
"mean_token_accuracy": 0.13047630861401557,
|
|
"num_tokens": 2830007.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.352209997177124,
|
|
"epoch": 0.08344459279038718,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 6.1489,
|
|
"mean_token_accuracy": 0.12876534834504128,
|
|
"num_tokens": 2838768.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.418527412414551,
|
|
"epoch": 0.08370134538358838,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.12,
|
|
"mean_token_accuracy": 0.1371123783290386,
|
|
"num_tokens": 2846980.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.330585241317749,
|
|
"epoch": 0.08395809797678956,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 6.1187,
|
|
"mean_token_accuracy": 0.13255744129419328,
|
|
"num_tokens": 2855217.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.4154908657073975,
|
|
"epoch": 0.08421485056999076,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 6.1406,
|
|
"mean_token_accuracy": 0.1382390558719635,
|
|
"num_tokens": 2863528.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.360077333450318,
|
|
"epoch": 0.08447160316319195,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 6.1123,
|
|
"mean_token_accuracy": 0.12890905141830444,
|
|
"num_tokens": 2872747.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.347766017913818,
|
|
"epoch": 0.08472835575639313,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.1221,
|
|
"mean_token_accuracy": 0.13139552399516105,
|
|
"num_tokens": 2880923.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.41855821609497,
|
|
"epoch": 0.08498510834959433,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 6.1571,
|
|
"mean_token_accuracy": 0.13074238151311873,
|
|
"num_tokens": 2889250.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.365989875793457,
|
|
"epoch": 0.08524186094279552,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 5.9928,
|
|
"mean_token_accuracy": 0.13420796319842337,
|
|
"num_tokens": 2897826.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.372472143173217,
|
|
"epoch": 0.08549861353599672,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 6.2305,
|
|
"mean_token_accuracy": 0.12156680822372437,
|
|
"num_tokens": 2907594.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.427253770828247,
|
|
"epoch": 0.0857553661291979,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.1515,
|
|
"mean_token_accuracy": 0.12236132472753525,
|
|
"num_tokens": 2916539.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.29747371673584,
|
|
"epoch": 0.0860121187223991,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.0278,
|
|
"mean_token_accuracy": 0.1366209253668785,
|
|
"num_tokens": 2925175.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.463992881774902,
|
|
"epoch": 0.08626887131560029,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 6.2366,
|
|
"mean_token_accuracy": 0.12456757575273514,
|
|
"num_tokens": 2934052.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.342855167388916,
|
|
"epoch": 0.08652562390880147,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 6.1099,
|
|
"mean_token_accuracy": 0.12803666070103645,
|
|
"num_tokens": 2944380.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.324626159667969,
|
|
"epoch": 0.08678237650200267,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.0308,
|
|
"mean_token_accuracy": 0.13413935974240304,
|
|
"num_tokens": 2953345.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.359399318695068,
|
|
"epoch": 0.08703912909520387,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.1745,
|
|
"mean_token_accuracy": 0.13524795323610306,
|
|
"num_tokens": 2961421.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.340810394287109,
|
|
"epoch": 0.08729588168840505,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 6.1289,
|
|
"mean_token_accuracy": 0.13010590076446532,
|
|
"num_tokens": 2970059.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.530434894561767,
|
|
"epoch": 0.08755263428160624,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 6.2301,
|
|
"mean_token_accuracy": 0.11859835758805275,
|
|
"num_tokens": 2978485.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.417358636856079,
|
|
"epoch": 0.08780938687480744,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 6.1942,
|
|
"mean_token_accuracy": 0.12599658593535423,
|
|
"num_tokens": 2987858.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.359304523468017,
|
|
"epoch": 0.08806613946800863,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.0983,
|
|
"mean_token_accuracy": 0.13075019642710686,
|
|
"num_tokens": 2997065.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.3911253929138185,
|
|
"epoch": 0.08832289206120982,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 6.0722,
|
|
"mean_token_accuracy": 0.12853528559207916,
|
|
"num_tokens": 3005507.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.378786230087281,
|
|
"epoch": 0.08857964465441101,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 6.2589,
|
|
"mean_token_accuracy": 0.12528994753956796,
|
|
"num_tokens": 3014149.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.483067560195923,
|
|
"epoch": 0.0888363972476122,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 6.1815,
|
|
"mean_token_accuracy": 0.12748926654458045,
|
|
"num_tokens": 3022402.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.346529102325439,
|
|
"epoch": 0.08909314984081339,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.1301,
|
|
"mean_token_accuracy": 0.1264743335545063,
|
|
"num_tokens": 3030869.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.4348961353302006,
|
|
"epoch": 0.08934990243401458,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.1996,
|
|
"mean_token_accuracy": 0.1253055900335312,
|
|
"num_tokens": 3040834.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.423256969451904,
|
|
"epoch": 0.08960665502721578,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 6.1437,
|
|
"mean_token_accuracy": 0.13117995485663414,
|
|
"num_tokens": 3048466.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.4277565479278564,
|
|
"epoch": 0.08986340762041696,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.2085,
|
|
"mean_token_accuracy": 0.12590872794389724,
|
|
"num_tokens": 3057933.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.446882009506226,
|
|
"epoch": 0.09012016021361816,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.1383,
|
|
"mean_token_accuracy": 0.12370797023177146,
|
|
"num_tokens": 3066629.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.415470123291016,
|
|
"epoch": 0.09037691280681935,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.1672,
|
|
"mean_token_accuracy": 0.12792096361517907,
|
|
"num_tokens": 3075747.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.402537679672241,
|
|
"epoch": 0.09063366540002055,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.1817,
|
|
"mean_token_accuracy": 0.1291731856763363,
|
|
"num_tokens": 3084777.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.296290302276612,
|
|
"epoch": 0.09089041799322173,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 6.1562,
|
|
"mean_token_accuracy": 0.13283068910241128,
|
|
"num_tokens": 3094410.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.3334144115447994,
|
|
"epoch": 0.09114717058642292,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.0026,
|
|
"mean_token_accuracy": 0.1360984228551388,
|
|
"num_tokens": 3103386.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.285834741592407,
|
|
"epoch": 0.09140392317962412,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 6.0554,
|
|
"mean_token_accuracy": 0.13300737217068673,
|
|
"num_tokens": 3112282.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.401378536224366,
|
|
"epoch": 0.0916606757728253,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 6.1387,
|
|
"mean_token_accuracy": 0.12468569204211236,
|
|
"num_tokens": 3120755.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.490727806091309,
|
|
"epoch": 0.0919174283660265,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.2275,
|
|
"mean_token_accuracy": 0.12739756479859352,
|
|
"num_tokens": 3130337.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.295879077911377,
|
|
"epoch": 0.09217418095922769,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 6.0651,
|
|
"mean_token_accuracy": 0.1281225398182869,
|
|
"num_tokens": 3139056.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.4862042427062985,
|
|
"epoch": 0.09243093355242887,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.2646,
|
|
"mean_token_accuracy": 0.12719134613871574,
|
|
"num_tokens": 3148038.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.377978038787842,
|
|
"epoch": 0.09268768614563007,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.0078,
|
|
"mean_token_accuracy": 0.1361188441514969,
|
|
"num_tokens": 3156260.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.344951486587524,
|
|
"epoch": 0.09294443873883126,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 6.1439,
|
|
"mean_token_accuracy": 0.12833771407604216,
|
|
"num_tokens": 3164920.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.380027103424072,
|
|
"epoch": 0.09320119133203246,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 6.1055,
|
|
"mean_token_accuracy": 0.12662351056933402,
|
|
"num_tokens": 3174546.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.338020277023316,
|
|
"epoch": 0.09345794392523364,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 6.0621,
|
|
"mean_token_accuracy": 0.12475857213139534,
|
|
"num_tokens": 3184250.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.213906717300415,
|
|
"epoch": 0.09371469651843484,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 5.9632,
|
|
"mean_token_accuracy": 0.13960219770669938,
|
|
"num_tokens": 3193086.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.320476341247558,
|
|
"epoch": 0.09397144911163603,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 6.1122,
|
|
"mean_token_accuracy": 0.12685565352439881,
|
|
"num_tokens": 3201864.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.426863718032837,
|
|
"epoch": 0.09422820170483721,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.0851,
|
|
"mean_token_accuracy": 0.13300562500953675,
|
|
"num_tokens": 3210606.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.315115261077881,
|
|
"epoch": 0.09448495429803841,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 6.1144,
|
|
"mean_token_accuracy": 0.12918654158711435,
|
|
"num_tokens": 3219494.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.3636940002441404,
|
|
"epoch": 0.0947417068912396,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.1018,
|
|
"mean_token_accuracy": 0.13404014483094215,
|
|
"num_tokens": 3227653.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.304843378067017,
|
|
"epoch": 0.09499845948444079,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 6.059,
|
|
"mean_token_accuracy": 0.13522360175848008,
|
|
"num_tokens": 3235872.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.377466201782227,
|
|
"epoch": 0.09525521207764198,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 6.1122,
|
|
"mean_token_accuracy": 0.13392022475600243,
|
|
"num_tokens": 3244776.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.367177104949951,
|
|
"epoch": 0.09551196467084318,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 6.0808,
|
|
"mean_token_accuracy": 0.1407839186489582,
|
|
"num_tokens": 3253306.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.283543062210083,
|
|
"epoch": 0.09576871726404437,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.0556,
|
|
"mean_token_accuracy": 0.128963503241539,
|
|
"num_tokens": 3261585.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.352444124221802,
|
|
"epoch": 0.09602546985724555,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 6.0892,
|
|
"mean_token_accuracy": 0.1352520227432251,
|
|
"num_tokens": 3269770.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.3088125705719,
|
|
"epoch": 0.09628222245044675,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 6.0762,
|
|
"mean_token_accuracy": 0.13414381369948386,
|
|
"num_tokens": 3279429.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.259514236450196,
|
|
"epoch": 0.09653897504364795,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.0067,
|
|
"mean_token_accuracy": 0.14171726107597352,
|
|
"num_tokens": 3288317.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.239880990982056,
|
|
"epoch": 0.09679572763684913,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 5.9918,
|
|
"mean_token_accuracy": 0.1375198930501938,
|
|
"num_tokens": 3296606.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.2098822593688965,
|
|
"epoch": 0.09705248023005032,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.13969532400369644,
|
|
"num_tokens": 3304597.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.332381629943848,
|
|
"epoch": 0.09730923282325152,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.0584,
|
|
"mean_token_accuracy": 0.13430723026394845,
|
|
"num_tokens": 3313032.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.298563385009766,
|
|
"epoch": 0.0975659854164527,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 6.0022,
|
|
"mean_token_accuracy": 0.13629518523812295,
|
|
"num_tokens": 3321435.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.2789918899536135,
|
|
"epoch": 0.0978227380096539,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 6.133,
|
|
"mean_token_accuracy": 0.12246991395950317,
|
|
"num_tokens": 3330216.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.437214040756226,
|
|
"epoch": 0.09807949060285509,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.1324,
|
|
"mean_token_accuracy": 0.12947221025824546,
|
|
"num_tokens": 3340020.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.362472677230835,
|
|
"epoch": 0.09833624319605629,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 6.1259,
|
|
"mean_token_accuracy": 0.1282772108912468,
|
|
"num_tokens": 3349165.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.338744640350342,
|
|
"epoch": 0.09859299578925747,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.0265,
|
|
"mean_token_accuracy": 0.13221975266933442,
|
|
"num_tokens": 3357531.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.300238609313965,
|
|
"epoch": 0.09884974838245866,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 5.9466,
|
|
"mean_token_accuracy": 0.14144921079277992,
|
|
"num_tokens": 3366154.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.274377489089966,
|
|
"epoch": 0.09910650097565986,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.0445,
|
|
"mean_token_accuracy": 0.13802527487277985,
|
|
"num_tokens": 3374698.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.176395320892334,
|
|
"epoch": 0.09936325356886104,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 5.9103,
|
|
"mean_token_accuracy": 0.1357658825814724,
|
|
"num_tokens": 3383820.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.295207834243774,
|
|
"epoch": 0.09962000616206224,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.0326,
|
|
"mean_token_accuracy": 0.1353134199976921,
|
|
"num_tokens": 3392738.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.222642707824707,
|
|
"epoch": 0.09987675875526343,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 6.0702,
|
|
"mean_token_accuracy": 0.13356536477804185,
|
|
"num_tokens": 3401799.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.338154554367065,
|
|
"epoch": 0.10013351134846461,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 6.0875,
|
|
"mean_token_accuracy": 0.1363699369132519,
|
|
"num_tokens": 3410612.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.297544336318969,
|
|
"epoch": 0.10039026394166581,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 6.115,
|
|
"mean_token_accuracy": 0.1259643740952015,
|
|
"num_tokens": 3419640.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.362210083007812,
|
|
"epoch": 0.100647016534867,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.0073,
|
|
"mean_token_accuracy": 0.1342238776385784,
|
|
"num_tokens": 3428391.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.292529296875,
|
|
"epoch": 0.1009037691280682,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 6.1107,
|
|
"mean_token_accuracy": 0.1288607709109783,
|
|
"num_tokens": 3437770.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.205248165130615,
|
|
"epoch": 0.10116052172126938,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 5.9252,
|
|
"mean_token_accuracy": 0.13843688890337943,
|
|
"num_tokens": 3446267.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.229200935363769,
|
|
"epoch": 0.10141727431447058,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 5.9945,
|
|
"mean_token_accuracy": 0.13976073116064072,
|
|
"num_tokens": 3454677.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.376470565795898,
|
|
"epoch": 0.10167402690767177,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 6.0858,
|
|
"mean_token_accuracy": 0.1412315510213375,
|
|
"num_tokens": 3463057.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.365133237838745,
|
|
"epoch": 0.10193077950087295,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.104,
|
|
"mean_token_accuracy": 0.13191207125782967,
|
|
"num_tokens": 3471704.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.223714017868042,
|
|
"epoch": 0.10218753209407415,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 6.0632,
|
|
"mean_token_accuracy": 0.1316584102809429,
|
|
"num_tokens": 3480709.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.347341060638428,
|
|
"epoch": 0.10244428468727534,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 6.0386,
|
|
"mean_token_accuracy": 0.13127250745892524,
|
|
"num_tokens": 3489255.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.267368030548096,
|
|
"epoch": 0.10270103728047653,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 6.0371,
|
|
"mean_token_accuracy": 0.13498617261648177,
|
|
"num_tokens": 3498398.0,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 708717507379200.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|