1cad190a13
Model: fpadovani/nld-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
7112 lines
193 KiB
JSON
7112 lines
193 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.179009541117674,
|
|
"eval_steps": 500,
|
|
"global_step": 3500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.784345436096191,
|
|
"epoch": 0.004543389368468878,
|
|
"grad_norm": 17.75,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.1752,
|
|
"mean_token_accuracy": 0.0001060445443727076,
|
|
"num_tokens": 10131.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.809862565994263,
|
|
"epoch": 0.009086778736937756,
|
|
"grad_norm": 19.25,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.099,
|
|
"mean_token_accuracy": 0.00031843508477322755,
|
|
"num_tokens": 20042.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.916619968414307,
|
|
"epoch": 0.013630168105406633,
|
|
"grad_norm": 23.5,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.7398,
|
|
"mean_token_accuracy": 0.00030120814335532486,
|
|
"num_tokens": 30596.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.580635833740234,
|
|
"epoch": 0.01817355747387551,
|
|
"grad_norm": 31.125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 12.8465,
|
|
"mean_token_accuracy": 0.00010810811072587967,
|
|
"num_tokens": 40667.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 9.254024171829224,
|
|
"epoch": 0.02271694684234439,
|
|
"grad_norm": 4.125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.1582,
|
|
"mean_token_accuracy": 0.00011574074160307646,
|
|
"num_tokens": 51027.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.672628784179688,
|
|
"epoch": 0.027260336210813266,
|
|
"grad_norm": 3.578125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.7082,
|
|
"mean_token_accuracy": 0.004387588211102411,
|
|
"num_tokens": 61860.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.695150661468507,
|
|
"epoch": 0.03180372557928214,
|
|
"grad_norm": 3.078125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.4675,
|
|
"mean_token_accuracy": 0.01812011804431677,
|
|
"num_tokens": 72136.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.66664924621582,
|
|
"epoch": 0.03634711494775102,
|
|
"grad_norm": 2.8125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.0872,
|
|
"mean_token_accuracy": 0.020329243503510952,
|
|
"num_tokens": 82611.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.566537189483643,
|
|
"epoch": 0.0408905043162199,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8543,
|
|
"mean_token_accuracy": 0.030497603956609963,
|
|
"num_tokens": 92722.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.554127979278565,
|
|
"epoch": 0.04543389368468878,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6461,
|
|
"mean_token_accuracy": 0.04449060596525669,
|
|
"num_tokens": 103258.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.527712154388428,
|
|
"epoch": 0.049977283053157656,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5766,
|
|
"mean_token_accuracy": 0.0401018675416708,
|
|
"num_tokens": 114277.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.55856409072876,
|
|
"epoch": 0.05452067242162653,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4598,
|
|
"mean_token_accuracy": 0.03970269616693258,
|
|
"num_tokens": 124098.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.568616485595703,
|
|
"epoch": 0.059064061790095414,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3543,
|
|
"mean_token_accuracy": 0.03490773383527994,
|
|
"num_tokens": 133775.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.547126579284669,
|
|
"epoch": 0.06360745115856428,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.3169,
|
|
"mean_token_accuracy": 0.039617203176021576,
|
|
"num_tokens": 144765.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.492327785491943,
|
|
"epoch": 0.06815084052703317,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.191,
|
|
"mean_token_accuracy": 0.04435103014111519,
|
|
"num_tokens": 154641.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.46142349243164,
|
|
"epoch": 0.07269422989550205,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.1009,
|
|
"mean_token_accuracy": 0.04529869481921196,
|
|
"num_tokens": 164728.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.45467119216919,
|
|
"epoch": 0.07723761926397092,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0406,
|
|
"mean_token_accuracy": 0.04549358636140823,
|
|
"num_tokens": 174739.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.434930515289306,
|
|
"epoch": 0.0817810086324398,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9082,
|
|
"mean_token_accuracy": 0.05227328687906265,
|
|
"num_tokens": 184580.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.37447452545166,
|
|
"epoch": 0.08632439800090867,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8648,
|
|
"mean_token_accuracy": 0.05065051428973675,
|
|
"num_tokens": 195243.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.303302764892578,
|
|
"epoch": 0.09086778736937756,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.736,
|
|
"mean_token_accuracy": 0.05876764878630638,
|
|
"num_tokens": 205751.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.217681503295898,
|
|
"epoch": 0.09541117673784644,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.5864,
|
|
"mean_token_accuracy": 0.05198401734232903,
|
|
"num_tokens": 217283.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.115850257873536,
|
|
"epoch": 0.09995456610631531,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4609,
|
|
"mean_token_accuracy": 0.04754493683576584,
|
|
"num_tokens": 226909.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.03682804107666,
|
|
"epoch": 0.10449795547478419,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.3981,
|
|
"mean_token_accuracy": 0.05233983248472214,
|
|
"num_tokens": 237343.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.905374145507812,
|
|
"epoch": 0.10904134484325306,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.2219,
|
|
"mean_token_accuracy": 0.05766485221683979,
|
|
"num_tokens": 246601.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.75998296737671,
|
|
"epoch": 0.11358473421172194,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.1089,
|
|
"mean_token_accuracy": 0.061960494890809056,
|
|
"num_tokens": 256818.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.538305950164794,
|
|
"epoch": 0.11812812358019083,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.9286,
|
|
"mean_token_accuracy": 0.05742153190076351,
|
|
"num_tokens": 267157.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.336606979370117,
|
|
"epoch": 0.1226715129486597,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.8459,
|
|
"mean_token_accuracy": 0.06012442037463188,
|
|
"num_tokens": 277370.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.107532691955566,
|
|
"epoch": 0.12721490231712856,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.759,
|
|
"mean_token_accuracy": 0.054580170288681985,
|
|
"num_tokens": 287851.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.893938255310058,
|
|
"epoch": 0.13175829168559747,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.7274,
|
|
"mean_token_accuracy": 0.06078527122735977,
|
|
"num_tokens": 299043.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.632384300231934,
|
|
"epoch": 0.13630168105406634,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.6541,
|
|
"mean_token_accuracy": 0.06055265218019486,
|
|
"num_tokens": 308525.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.41025161743164,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.5306,
|
|
"mean_token_accuracy": 0.06570497304201126,
|
|
"num_tokens": 318301.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.30331039428711,
|
|
"epoch": 0.1453884597910041,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.527,
|
|
"mean_token_accuracy": 0.07010898999869823,
|
|
"num_tokens": 329599.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.167072677612305,
|
|
"epoch": 0.14993184915947297,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.4514,
|
|
"mean_token_accuracy": 0.07434111759066582,
|
|
"num_tokens": 339446.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.028205299377442,
|
|
"epoch": 0.15447523852794184,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.5229,
|
|
"mean_token_accuracy": 0.07283368557691575,
|
|
"num_tokens": 349579.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.01085844039917,
|
|
"epoch": 0.15901862789641072,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.4528,
|
|
"mean_token_accuracy": 0.07020434252917766,
|
|
"num_tokens": 359452.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.793022012710571,
|
|
"epoch": 0.1635620172648796,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.3541,
|
|
"mean_token_accuracy": 0.07883949726819992,
|
|
"num_tokens": 369558.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.859193992614746,
|
|
"epoch": 0.16810540663334847,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.3863,
|
|
"mean_token_accuracy": 0.07790203467011451,
|
|
"num_tokens": 380458.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.857694339752197,
|
|
"epoch": 0.17264879600181735,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3652,
|
|
"mean_token_accuracy": 0.0808223120868206,
|
|
"num_tokens": 390051.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.768624973297119,
|
|
"epoch": 0.17719218537028622,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3239,
|
|
"mean_token_accuracy": 0.08125012665987015,
|
|
"num_tokens": 400343.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.734456300735474,
|
|
"epoch": 0.18173557473875512,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.3367,
|
|
"mean_token_accuracy": 0.07972371056675912,
|
|
"num_tokens": 410004.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.739763355255127,
|
|
"epoch": 0.186278964107224,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.3202,
|
|
"mean_token_accuracy": 0.0789877288043499,
|
|
"num_tokens": 421286.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.8016503810882565,
|
|
"epoch": 0.19082235347569287,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3443,
|
|
"mean_token_accuracy": 0.07945962622761726,
|
|
"num_tokens": 432351.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.740051794052124,
|
|
"epoch": 0.19536574284416175,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3457,
|
|
"mean_token_accuracy": 0.08271278738975525,
|
|
"num_tokens": 442843.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.63407711982727,
|
|
"epoch": 0.19990913221263062,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2133,
|
|
"mean_token_accuracy": 0.07842602878808975,
|
|
"num_tokens": 453370.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.7399622917175295,
|
|
"epoch": 0.2044525215810995,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3604,
|
|
"mean_token_accuracy": 0.07667806819081306,
|
|
"num_tokens": 464715.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.678428506851196,
|
|
"epoch": 0.20899591094956838,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2099,
|
|
"mean_token_accuracy": 0.08242459893226624,
|
|
"num_tokens": 475086.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.593945550918579,
|
|
"epoch": 0.21353930031803725,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2476,
|
|
"mean_token_accuracy": 0.09175414070487023,
|
|
"num_tokens": 486088.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.680302572250366,
|
|
"epoch": 0.21808268968650613,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.2214,
|
|
"mean_token_accuracy": 0.08108382746577263,
|
|
"num_tokens": 496978.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.542342758178711,
|
|
"epoch": 0.222626079054975,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.0185,
|
|
"mean_token_accuracy": 0.09117448925971985,
|
|
"num_tokens": 507794.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.60085506439209,
|
|
"epoch": 0.22716946842344388,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1885,
|
|
"mean_token_accuracy": 0.08196934387087822,
|
|
"num_tokens": 517763.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.550417375564575,
|
|
"epoch": 0.23171285779191278,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.0802,
|
|
"mean_token_accuracy": 0.08660967722535133,
|
|
"num_tokens": 527374.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.5772816181182865,
|
|
"epoch": 0.23625624716038165,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1524,
|
|
"mean_token_accuracy": 0.08742348551750183,
|
|
"num_tokens": 537435.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.5303443431854244,
|
|
"epoch": 0.24079963652885053,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1058,
|
|
"mean_token_accuracy": 0.08927300423383713,
|
|
"num_tokens": 547236.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.422381019592285,
|
|
"epoch": 0.2453430258973194,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0979,
|
|
"mean_token_accuracy": 0.08617129400372506,
|
|
"num_tokens": 557130.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.546438264846802,
|
|
"epoch": 0.24988641526578828,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0892,
|
|
"mean_token_accuracy": 0.09092889800667762,
|
|
"num_tokens": 567482.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.45471682548523,
|
|
"epoch": 0.25442980463425713,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.0032,
|
|
"mean_token_accuracy": 0.08620407655835152,
|
|
"num_tokens": 577084.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.413665533065796,
|
|
"epoch": 0.25897319400272606,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1008,
|
|
"mean_token_accuracy": 0.0921616092324257,
|
|
"num_tokens": 587980.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.384335041046143,
|
|
"epoch": 0.26351658337119493,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.9798,
|
|
"mean_token_accuracy": 0.09184363558888435,
|
|
"num_tokens": 598989.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.457988500595093,
|
|
"epoch": 0.2680599727396638,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1774,
|
|
"mean_token_accuracy": 0.08681531026959419,
|
|
"num_tokens": 608599.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.4478963851928714,
|
|
"epoch": 0.2726033621081327,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9794,
|
|
"mean_token_accuracy": 0.09082643985748291,
|
|
"num_tokens": 619267.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.406017208099366,
|
|
"epoch": 0.27714675147660156,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.1453,
|
|
"mean_token_accuracy": 0.08878572881221772,
|
|
"num_tokens": 629991.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.374471712112427,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.1133,
|
|
"mean_token_accuracy": 0.08506234511733055,
|
|
"num_tokens": 641311.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.490564584732056,
|
|
"epoch": 0.2862335302135393,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0629,
|
|
"mean_token_accuracy": 0.08935049623250961,
|
|
"num_tokens": 651812.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.428491497039795,
|
|
"epoch": 0.2907769195820082,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0138,
|
|
"mean_token_accuracy": 0.09055089727044105,
|
|
"num_tokens": 662387.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.328471040725708,
|
|
"epoch": 0.29532030895047706,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9974,
|
|
"mean_token_accuracy": 0.09098516553640365,
|
|
"num_tokens": 673018.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.389373064041138,
|
|
"epoch": 0.29986369831894594,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9366,
|
|
"mean_token_accuracy": 0.09445346742868424,
|
|
"num_tokens": 683192.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.305378770828247,
|
|
"epoch": 0.3044070876874148,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.0207,
|
|
"mean_token_accuracy": 0.09813978150486946,
|
|
"num_tokens": 692747.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.231426000595093,
|
|
"epoch": 0.3089504770558837,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0149,
|
|
"mean_token_accuracy": 0.0843175120651722,
|
|
"num_tokens": 705109.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.3670117378234865,
|
|
"epoch": 0.31349386642435256,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9399,
|
|
"mean_token_accuracy": 0.08961606845259666,
|
|
"num_tokens": 715658.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.264230108261108,
|
|
"epoch": 0.31803725579282144,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9492,
|
|
"mean_token_accuracy": 0.1010621391236782,
|
|
"num_tokens": 724961.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.207327604293823,
|
|
"epoch": 0.3225806451612903,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9033,
|
|
"mean_token_accuracy": 0.10158559232950211,
|
|
"num_tokens": 735416.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.289533853530884,
|
|
"epoch": 0.3271240345297592,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9908,
|
|
"mean_token_accuracy": 0.09567400589585304,
|
|
"num_tokens": 745005.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.433129930496216,
|
|
"epoch": 0.33166742389822806,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0279,
|
|
"mean_token_accuracy": 0.09788246005773545,
|
|
"num_tokens": 754743.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.176350069046021,
|
|
"epoch": 0.33621081326669694,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.8811,
|
|
"mean_token_accuracy": 0.09762823358178138,
|
|
"num_tokens": 764897.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.201433515548706,
|
|
"epoch": 0.3407542026351658,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.942,
|
|
"mean_token_accuracy": 0.09215436205267906,
|
|
"num_tokens": 775302.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.281723833084106,
|
|
"epoch": 0.3452975920036347,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8712,
|
|
"mean_token_accuracy": 0.09897375032305718,
|
|
"num_tokens": 785694.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.198097848892212,
|
|
"epoch": 0.34984098137210357,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8274,
|
|
"mean_token_accuracy": 0.1060432367026806,
|
|
"num_tokens": 795045.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.149559593200683,
|
|
"epoch": 0.35438437074057244,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8887,
|
|
"mean_token_accuracy": 0.09728073477745056,
|
|
"num_tokens": 805579.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.184843063354492,
|
|
"epoch": 0.35892776010904137,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8013,
|
|
"mean_token_accuracy": 0.10386217460036277,
|
|
"num_tokens": 816743.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.237153148651123,
|
|
"epoch": 0.36347114947751025,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.9312,
|
|
"mean_token_accuracy": 0.09671236723661422,
|
|
"num_tokens": 827217.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.121163940429687,
|
|
"epoch": 0.3680145388459791,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7826,
|
|
"mean_token_accuracy": 0.10176689177751541,
|
|
"num_tokens": 837555.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.082342529296875,
|
|
"epoch": 0.372557928214448,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8467,
|
|
"mean_token_accuracy": 0.09997194483876229,
|
|
"num_tokens": 846726.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.264329671859741,
|
|
"epoch": 0.3771013175829169,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8527,
|
|
"mean_token_accuracy": 0.0990760438144207,
|
|
"num_tokens": 856535.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.163194799423218,
|
|
"epoch": 0.38164470695138575,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9053,
|
|
"mean_token_accuracy": 0.09648826271295548,
|
|
"num_tokens": 866973.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.096738195419311,
|
|
"epoch": 0.3861880963198546,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8393,
|
|
"mean_token_accuracy": 0.09683325067162514,
|
|
"num_tokens": 877873.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.196438980102539,
|
|
"epoch": 0.3907314856883235,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8366,
|
|
"mean_token_accuracy": 0.09828796163201332,
|
|
"num_tokens": 887135.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.189782333374024,
|
|
"epoch": 0.3952748750567924,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8437,
|
|
"mean_token_accuracy": 0.09872046262025833,
|
|
"num_tokens": 897206.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.112062072753906,
|
|
"epoch": 0.39981826442526125,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.8243,
|
|
"mean_token_accuracy": 0.09822067096829415,
|
|
"num_tokens": 906808.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.018064546585083,
|
|
"epoch": 0.4043616537937301,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.742,
|
|
"mean_token_accuracy": 0.10562362670898437,
|
|
"num_tokens": 917177.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.033844232559204,
|
|
"epoch": 0.408905043162199,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7756,
|
|
"mean_token_accuracy": 0.10635508671402931,
|
|
"num_tokens": 927229.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.1084190845489506,
|
|
"epoch": 0.4134484325306679,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7814,
|
|
"mean_token_accuracy": 0.10521619990468026,
|
|
"num_tokens": 937832.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.975793027877808,
|
|
"epoch": 0.41799182189913675,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6997,
|
|
"mean_token_accuracy": 0.10499884337186813,
|
|
"num_tokens": 948038.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.101250648498535,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8235,
|
|
"mean_token_accuracy": 0.09736672714352608,
|
|
"num_tokens": 957924.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.0696635246276855,
|
|
"epoch": 0.4270786006360745,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7782,
|
|
"mean_token_accuracy": 0.1001742161810398,
|
|
"num_tokens": 968497.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.0578991889953615,
|
|
"epoch": 0.4316219900045434,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7031,
|
|
"mean_token_accuracy": 0.10739717781543731,
|
|
"num_tokens": 978878.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.926171064376831,
|
|
"epoch": 0.43616537937301225,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7936,
|
|
"mean_token_accuracy": 0.10459667965769767,
|
|
"num_tokens": 989984.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.042437887191772,
|
|
"epoch": 0.4407087687414811,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7195,
|
|
"mean_token_accuracy": 0.108768729865551,
|
|
"num_tokens": 1000254.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.930255270004272,
|
|
"epoch": 0.44525215810995,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7759,
|
|
"mean_token_accuracy": 0.10471918955445289,
|
|
"num_tokens": 1010697.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.141830158233643,
|
|
"epoch": 0.4497955474784189,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8673,
|
|
"mean_token_accuracy": 0.10067511573433877,
|
|
"num_tokens": 1020892.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.046828079223633,
|
|
"epoch": 0.45433893684688775,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.7463,
|
|
"mean_token_accuracy": 0.11069609895348549,
|
|
"num_tokens": 1030380.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.45433893684688775,
|
|
"eval_entropy": 6.98116321341936,
|
|
"eval_loss": 6.841126441955566,
|
|
"eval_mean_token_accuracy": 0.10453861841801987,
|
|
"eval_num_tokens": 1030380.0,
|
|
"eval_runtime": 2.6082,
|
|
"eval_samples_per_second": 1316.98,
|
|
"eval_steps_per_second": 164.862,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.005007743835449,
|
|
"epoch": 0.45888232621535663,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7946,
|
|
"mean_token_accuracy": 0.10074228718876839,
|
|
"num_tokens": 1040301.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.971395969390869,
|
|
"epoch": 0.46342571558382556,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6608,
|
|
"mean_token_accuracy": 0.1083550937473774,
|
|
"num_tokens": 1051999.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.075518608093262,
|
|
"epoch": 0.46796910495229443,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.8761,
|
|
"mean_token_accuracy": 0.09833496063947678,
|
|
"num_tokens": 1063133.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.913373708724976,
|
|
"epoch": 0.4725124943207633,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6895,
|
|
"mean_token_accuracy": 0.10114651918411255,
|
|
"num_tokens": 1073972.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.906210899353027,
|
|
"epoch": 0.4770558836892322,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6763,
|
|
"mean_token_accuracy": 0.10489214286208152,
|
|
"num_tokens": 1084050.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.034184503555298,
|
|
"epoch": 0.48159927305770106,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7713,
|
|
"mean_token_accuracy": 0.11127977892756462,
|
|
"num_tokens": 1094415.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.822019624710083,
|
|
"epoch": 0.48614266242616994,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6952,
|
|
"mean_token_accuracy": 0.10635157525539399,
|
|
"num_tokens": 1105133.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.941200923919678,
|
|
"epoch": 0.4906860517946388,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6581,
|
|
"mean_token_accuracy": 0.1116144098341465,
|
|
"num_tokens": 1115266.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.84721941947937,
|
|
"epoch": 0.4952294411631077,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6292,
|
|
"mean_token_accuracy": 0.10852839723229409,
|
|
"num_tokens": 1124775.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.842454719543457,
|
|
"epoch": 0.49977283053157656,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6842,
|
|
"mean_token_accuracy": 0.10360125750303269,
|
|
"num_tokens": 1134498.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.993335723876953,
|
|
"epoch": 0.5043162199000454,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.726,
|
|
"mean_token_accuracy": 0.10071266666054726,
|
|
"num_tokens": 1146067.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.975860595703125,
|
|
"epoch": 0.5088596092685143,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.7304,
|
|
"mean_token_accuracy": 0.10556481257081032,
|
|
"num_tokens": 1156349.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.9192403793334964,
|
|
"epoch": 0.5134029986369832,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6519,
|
|
"mean_token_accuracy": 0.10416188091039658,
|
|
"num_tokens": 1166851.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.879871797561646,
|
|
"epoch": 0.5179463880054521,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.6014,
|
|
"mean_token_accuracy": 0.10705872029066085,
|
|
"num_tokens": 1176899.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.788269662857056,
|
|
"epoch": 0.522489777373921,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6955,
|
|
"mean_token_accuracy": 0.10475884526968002,
|
|
"num_tokens": 1187383.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.945191860198975,
|
|
"epoch": 0.5270331667423899,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6781,
|
|
"mean_token_accuracy": 0.11020390838384628,
|
|
"num_tokens": 1197836.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.8903790473937985,
|
|
"epoch": 0.5315765561108587,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6212,
|
|
"mean_token_accuracy": 0.1158889003098011,
|
|
"num_tokens": 1208533.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.8323503017425535,
|
|
"epoch": 0.5361199454793276,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6672,
|
|
"mean_token_accuracy": 0.10879800394177437,
|
|
"num_tokens": 1219397.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.859197521209717,
|
|
"epoch": 0.5406633348477965,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6478,
|
|
"mean_token_accuracy": 0.10693326815962792,
|
|
"num_tokens": 1229991.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.72550892829895,
|
|
"epoch": 0.5452067242162654,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.5657,
|
|
"mean_token_accuracy": 0.1022007167339325,
|
|
"num_tokens": 1240343.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.842637062072754,
|
|
"epoch": 0.5497501135847342,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6323,
|
|
"mean_token_accuracy": 0.1036772184073925,
|
|
"num_tokens": 1250432.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.746218585968018,
|
|
"epoch": 0.5542935029532031,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.5642,
|
|
"mean_token_accuracy": 0.11089006587862968,
|
|
"num_tokens": 1260744.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.836953592300415,
|
|
"epoch": 0.558836892321672,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5926,
|
|
"mean_token_accuracy": 0.11175156757235527,
|
|
"num_tokens": 1272155.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.6719419956207275,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.4653,
|
|
"mean_token_accuracy": 0.11243500337004661,
|
|
"num_tokens": 1282240.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.832616186141967,
|
|
"epoch": 0.5679236710586097,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6481,
|
|
"mean_token_accuracy": 0.10761656239628792,
|
|
"num_tokens": 1292169.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.71432147026062,
|
|
"epoch": 0.5724670604270786,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.5575,
|
|
"mean_token_accuracy": 0.11017942726612091,
|
|
"num_tokens": 1302010.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.755236959457397,
|
|
"epoch": 0.5770104497955475,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5962,
|
|
"mean_token_accuracy": 0.11409324407577515,
|
|
"num_tokens": 1312596.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.760840702056885,
|
|
"epoch": 0.5815538391640164,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5452,
|
|
"mean_token_accuracy": 0.10920280888676644,
|
|
"num_tokens": 1321753.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.732819080352783,
|
|
"epoch": 0.5860972285324852,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5939,
|
|
"mean_token_accuracy": 0.11155394092202187,
|
|
"num_tokens": 1332059.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.753859329223633,
|
|
"epoch": 0.5906406179009541,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5566,
|
|
"mean_token_accuracy": 0.11300203204154968,
|
|
"num_tokens": 1341517.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.82979154586792,
|
|
"epoch": 0.595184007269423,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.642,
|
|
"mean_token_accuracy": 0.11486971005797386,
|
|
"num_tokens": 1351439.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.628808450698853,
|
|
"epoch": 0.5997273966378919,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5154,
|
|
"mean_token_accuracy": 0.11203483566641807,
|
|
"num_tokens": 1360745.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.692475748062134,
|
|
"epoch": 0.6042707860063607,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5149,
|
|
"mean_token_accuracy": 0.10992752835154533,
|
|
"num_tokens": 1371931.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.6035096645355225,
|
|
"epoch": 0.6088141753748296,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4581,
|
|
"mean_token_accuracy": 0.11140664517879487,
|
|
"num_tokens": 1382040.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.6743144512176515,
|
|
"epoch": 0.6133575647432985,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5347,
|
|
"mean_token_accuracy": 0.10514698475599289,
|
|
"num_tokens": 1392690.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.752614450454712,
|
|
"epoch": 0.6179009541117674,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5393,
|
|
"mean_token_accuracy": 0.11459807455539703,
|
|
"num_tokens": 1403060.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.716874790191651,
|
|
"epoch": 0.6224443434802363,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.5432,
|
|
"mean_token_accuracy": 0.11251721158623695,
|
|
"num_tokens": 1412895.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.682245874404908,
|
|
"epoch": 0.6269877328487051,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4939,
|
|
"mean_token_accuracy": 0.12101733833551406,
|
|
"num_tokens": 1423024.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.651356983184814,
|
|
"epoch": 0.631531122217174,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5011,
|
|
"mean_token_accuracy": 0.11385466754436493,
|
|
"num_tokens": 1433578.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.630496883392334,
|
|
"epoch": 0.6360745115856429,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5554,
|
|
"mean_token_accuracy": 0.11394480168819428,
|
|
"num_tokens": 1444473.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.694334983825684,
|
|
"epoch": 0.6406179009541118,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5722,
|
|
"mean_token_accuracy": 0.1110615760087967,
|
|
"num_tokens": 1454553.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.6838539123535154,
|
|
"epoch": 0.6451612903225806,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4358,
|
|
"mean_token_accuracy": 0.11738718673586845,
|
|
"num_tokens": 1464162.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.616205263137817,
|
|
"epoch": 0.6497046796910495,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5065,
|
|
"mean_token_accuracy": 0.11549382582306862,
|
|
"num_tokens": 1474691.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.63447003364563,
|
|
"epoch": 0.6542480690595184,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5666,
|
|
"mean_token_accuracy": 0.11404236555099487,
|
|
"num_tokens": 1484650.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.727664947509766,
|
|
"epoch": 0.6587914584279873,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5574,
|
|
"mean_token_accuracy": 0.11143433079123496,
|
|
"num_tokens": 1494788.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.63223786354065,
|
|
"epoch": 0.6633348477964561,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4911,
|
|
"mean_token_accuracy": 0.11522913351655006,
|
|
"num_tokens": 1506574.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.614343595504761,
|
|
"epoch": 0.667878237164925,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5088,
|
|
"mean_token_accuracy": 0.11446577459573745,
|
|
"num_tokens": 1517505.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.602818012237549,
|
|
"epoch": 0.6724216265333939,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3841,
|
|
"mean_token_accuracy": 0.12834027037024498,
|
|
"num_tokens": 1526662.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.56980414390564,
|
|
"epoch": 0.6769650159018628,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4927,
|
|
"mean_token_accuracy": 0.11517568081617355,
|
|
"num_tokens": 1536798.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.6200847148895265,
|
|
"epoch": 0.6815084052703316,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4135,
|
|
"mean_token_accuracy": 0.12157793939113617,
|
|
"num_tokens": 1546673.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.557360887527466,
|
|
"epoch": 0.6860517946388005,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4385,
|
|
"mean_token_accuracy": 0.12250431329011917,
|
|
"num_tokens": 1556793.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.632262897491455,
|
|
"epoch": 0.6905951840072694,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4835,
|
|
"mean_token_accuracy": 0.12135926857590676,
|
|
"num_tokens": 1566450.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.563928413391113,
|
|
"epoch": 0.6951385733757383,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5083,
|
|
"mean_token_accuracy": 0.11970612406730652,
|
|
"num_tokens": 1576353.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.583096694946289,
|
|
"epoch": 0.6996819627442071,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3948,
|
|
"mean_token_accuracy": 0.11896727532148361,
|
|
"num_tokens": 1586318.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.583036708831787,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4942,
|
|
"mean_token_accuracy": 0.11539135798811913,
|
|
"num_tokens": 1596839.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.5770739078521725,
|
|
"epoch": 0.7087687414811449,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.502,
|
|
"mean_token_accuracy": 0.11619984358549118,
|
|
"num_tokens": 1606843.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.488235092163086,
|
|
"epoch": 0.7133121308496138,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4432,
|
|
"mean_token_accuracy": 0.11896191388368607,
|
|
"num_tokens": 1617048.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.551537752151489,
|
|
"epoch": 0.7178555202180827,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.37,
|
|
"mean_token_accuracy": 0.12010857686400414,
|
|
"num_tokens": 1627472.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.584505224227906,
|
|
"epoch": 0.7223989095865516,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4742,
|
|
"mean_token_accuracy": 0.11549579650163651,
|
|
"num_tokens": 1637716.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.581101369857788,
|
|
"epoch": 0.7269422989550205,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.4833,
|
|
"mean_token_accuracy": 0.11456933096051217,
|
|
"num_tokens": 1648119.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.526739311218262,
|
|
"epoch": 0.7314856883234894,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4152,
|
|
"mean_token_accuracy": 0.1224575825035572,
|
|
"num_tokens": 1658214.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.412497568130493,
|
|
"epoch": 0.7360290776919582,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.389,
|
|
"mean_token_accuracy": 0.12186362668871879,
|
|
"num_tokens": 1668725.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.541046619415283,
|
|
"epoch": 0.7405724670604271,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4727,
|
|
"mean_token_accuracy": 0.12043284252285957,
|
|
"num_tokens": 1678891.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.5487672805786135,
|
|
"epoch": 0.745115856428896,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3587,
|
|
"mean_token_accuracy": 0.12360751256346703,
|
|
"num_tokens": 1688749.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.534564065933227,
|
|
"epoch": 0.7496592457973649,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4691,
|
|
"mean_token_accuracy": 0.11970914229750633,
|
|
"num_tokens": 1699166.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.5155445575714115,
|
|
"epoch": 0.7542026351658337,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.399,
|
|
"mean_token_accuracy": 0.11641391217708588,
|
|
"num_tokens": 1710690.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.578842306137085,
|
|
"epoch": 0.7587460245343026,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4944,
|
|
"mean_token_accuracy": 0.11359859183430672,
|
|
"num_tokens": 1721169.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.4253908634185795,
|
|
"epoch": 0.7632894139027715,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3678,
|
|
"mean_token_accuracy": 0.12227956652641296,
|
|
"num_tokens": 1731523.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.496607065200806,
|
|
"epoch": 0.7678328032712404,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.3485,
|
|
"mean_token_accuracy": 0.12179554626345634,
|
|
"num_tokens": 1742316.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.514575099945068,
|
|
"epoch": 0.7723761926397092,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.496,
|
|
"mean_token_accuracy": 0.11687857434153556,
|
|
"num_tokens": 1753424.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.440532588958741,
|
|
"epoch": 0.7769195820081781,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.3356,
|
|
"mean_token_accuracy": 0.12124308049678803,
|
|
"num_tokens": 1762814.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.488030433654785,
|
|
"epoch": 0.781462971376647,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4869,
|
|
"mean_token_accuracy": 0.12215745151042938,
|
|
"num_tokens": 1773096.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.467377185821533,
|
|
"epoch": 0.7860063607451159,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3854,
|
|
"mean_token_accuracy": 0.12040901333093643,
|
|
"num_tokens": 1783499.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.660170555114746,
|
|
"epoch": 0.7905497501135847,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.57,
|
|
"mean_token_accuracy": 0.11177526637911797,
|
|
"num_tokens": 1794107.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.487541723251343,
|
|
"epoch": 0.7950931394820536,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4467,
|
|
"mean_token_accuracy": 0.11407195180654525,
|
|
"num_tokens": 1805151.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.467250680923462,
|
|
"epoch": 0.7996365288505225,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4294,
|
|
"mean_token_accuracy": 0.11573711559176444,
|
|
"num_tokens": 1815205.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.4450499534606935,
|
|
"epoch": 0.8041799182189914,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4028,
|
|
"mean_token_accuracy": 0.12317447438836097,
|
|
"num_tokens": 1825301.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.493942308425903,
|
|
"epoch": 0.8087233075874602,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4263,
|
|
"mean_token_accuracy": 0.11993300542235374,
|
|
"num_tokens": 1835634.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.460993242263794,
|
|
"epoch": 0.8132666969559291,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4401,
|
|
"mean_token_accuracy": 0.11308973282575607,
|
|
"num_tokens": 1846855.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.423117923736572,
|
|
"epoch": 0.817810086324398,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4106,
|
|
"mean_token_accuracy": 0.11578994616866112,
|
|
"num_tokens": 1857747.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.491503715515137,
|
|
"epoch": 0.8223534756928669,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.443,
|
|
"mean_token_accuracy": 0.11862753704190254,
|
|
"num_tokens": 1868098.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.407588958740234,
|
|
"epoch": 0.8268968650613358,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3501,
|
|
"mean_token_accuracy": 0.1208985224366188,
|
|
"num_tokens": 1879682.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.500104999542236,
|
|
"epoch": 0.8314402544298046,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3779,
|
|
"mean_token_accuracy": 0.12159892171621323,
|
|
"num_tokens": 1890605.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.395988321304321,
|
|
"epoch": 0.8359836437982735,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4181,
|
|
"mean_token_accuracy": 0.12678836286067963,
|
|
"num_tokens": 1900861.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.448094415664673,
|
|
"epoch": 0.8405270331667424,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3823,
|
|
"mean_token_accuracy": 0.12325264140963554,
|
|
"num_tokens": 1910867.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.489271545410157,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3943,
|
|
"mean_token_accuracy": 0.11608804985880852,
|
|
"num_tokens": 1921424.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.461827802658081,
|
|
"epoch": 0.8496138119036801,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3656,
|
|
"mean_token_accuracy": 0.12558601945638656,
|
|
"num_tokens": 1930784.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.335020542144775,
|
|
"epoch": 0.854157201272149,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3751,
|
|
"mean_token_accuracy": 0.12744139283895492,
|
|
"num_tokens": 1940325.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.403874206542969,
|
|
"epoch": 0.8587005906406179,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.3801,
|
|
"mean_token_accuracy": 0.1256302960216999,
|
|
"num_tokens": 1951799.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.47987380027771,
|
|
"epoch": 0.8632439800090868,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3566,
|
|
"mean_token_accuracy": 0.12004562765359879,
|
|
"num_tokens": 1962311.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.4173583984375,
|
|
"epoch": 0.8677873693775556,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3695,
|
|
"mean_token_accuracy": 0.11883095428347587,
|
|
"num_tokens": 1974128.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.301603317260742,
|
|
"epoch": 0.8723307587460245,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.2904,
|
|
"mean_token_accuracy": 0.127122513204813,
|
|
"num_tokens": 1983765.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.427798891067505,
|
|
"epoch": 0.8768741481144934,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3609,
|
|
"mean_token_accuracy": 0.12233026251196862,
|
|
"num_tokens": 1994045.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.435657501220703,
|
|
"epoch": 0.8814175374829623,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4492,
|
|
"mean_token_accuracy": 0.11813291311264038,
|
|
"num_tokens": 2005353.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.304826259613037,
|
|
"epoch": 0.8859609268514311,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.2567,
|
|
"mean_token_accuracy": 0.12729906663298607,
|
|
"num_tokens": 2015808.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.371440172195435,
|
|
"epoch": 0.8905043162199,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.3167,
|
|
"mean_token_accuracy": 0.12886400520801544,
|
|
"num_tokens": 2025824.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.341680526733398,
|
|
"epoch": 0.8950477055883689,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3246,
|
|
"mean_token_accuracy": 0.12846749201416968,
|
|
"num_tokens": 2036924.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.34727668762207,
|
|
"epoch": 0.8995910949568378,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3602,
|
|
"mean_token_accuracy": 0.1250719718635082,
|
|
"num_tokens": 2047302.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.36696982383728,
|
|
"epoch": 0.9041344843253066,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3297,
|
|
"mean_token_accuracy": 0.12433951273560524,
|
|
"num_tokens": 2057972.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.26378173828125,
|
|
"epoch": 0.9086778736937755,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.1684,
|
|
"mean_token_accuracy": 0.1347182147204876,
|
|
"num_tokens": 2067916.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9086778736937755,
|
|
"eval_entropy": 6.301827516112216,
|
|
"eval_loss": 6.447918891906738,
|
|
"eval_mean_token_accuracy": 0.12384221866553606,
|
|
"eval_num_tokens": 2067916.0,
|
|
"eval_runtime": 2.6053,
|
|
"eval_samples_per_second": 1318.451,
|
|
"eval_steps_per_second": 165.046,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.22369966506958,
|
|
"epoch": 0.9132212630622444,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999998223471442,
|
|
"loss": 6.3624,
|
|
"mean_token_accuracy": 0.12653156146407127,
|
|
"num_tokens": 2079119.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.363337469100952,
|
|
"epoch": 0.9177646524307133,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999991006328981,
|
|
"loss": 6.259,
|
|
"mean_token_accuracy": 0.12525053545832635,
|
|
"num_tokens": 2090080.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.3007762908935545,
|
|
"epoch": 0.9223080417991822,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999978237557377,
|
|
"loss": 6.3041,
|
|
"mean_token_accuracy": 0.1320334978401661,
|
|
"num_tokens": 2100541.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.342982530593872,
|
|
"epoch": 0.9268514311676511,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999959917188137,
|
|
"loss": 6.2584,
|
|
"mean_token_accuracy": 0.12421070411801338,
|
|
"num_tokens": 2111213.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.306941318511963,
|
|
"epoch": 0.93139482053612,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999936045266462,
|
|
"loss": 6.3575,
|
|
"mean_token_accuracy": 0.12838359400629998,
|
|
"num_tokens": 2121714.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.346536684036255,
|
|
"epoch": 0.9359382099045889,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999906621851255,
|
|
"loss": 6.233,
|
|
"mean_token_accuracy": 0.11972849667072297,
|
|
"num_tokens": 2131226.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.327339315414429,
|
|
"epoch": 0.9404815992730577,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999871647015115,
|
|
"loss": 6.3215,
|
|
"mean_token_accuracy": 0.11869452074170113,
|
|
"num_tokens": 2143179.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.292591142654419,
|
|
"epoch": 0.9450249886415266,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999831120844339,
|
|
"loss": 6.2961,
|
|
"mean_token_accuracy": 0.12834577411413192,
|
|
"num_tokens": 2153646.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.243809938430786,
|
|
"epoch": 0.9495683780099955,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999785043438922,
|
|
"loss": 6.2479,
|
|
"mean_token_accuracy": 0.12622877582907677,
|
|
"num_tokens": 2163267.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.311278581619263,
|
|
"epoch": 0.9541117673784644,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999733414912555,
|
|
"loss": 6.1592,
|
|
"mean_token_accuracy": 0.13357205241918563,
|
|
"num_tokens": 2173282.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.2701067447662355,
|
|
"epoch": 0.9586551567469332,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999676235392626,
|
|
"loss": 6.2547,
|
|
"mean_token_accuracy": 0.12008880004286766,
|
|
"num_tokens": 2184202.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.331092643737793,
|
|
"epoch": 0.9631985461154021,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999613505020219,
|
|
"loss": 6.2234,
|
|
"mean_token_accuracy": 0.1308535598218441,
|
|
"num_tokens": 2194342.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.211825513839722,
|
|
"epoch": 0.967741935483871,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999545223950116,
|
|
"loss": 6.238,
|
|
"mean_token_accuracy": 0.13356681689620017,
|
|
"num_tokens": 2204817.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.29638900756836,
|
|
"epoch": 0.9722853248523399,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999471392350795,
|
|
"loss": 6.316,
|
|
"mean_token_accuracy": 0.13001345917582513,
|
|
"num_tokens": 2215281.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.394066953659058,
|
|
"epoch": 0.9768287142208087,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999392010404426,
|
|
"loss": 6.2985,
|
|
"mean_token_accuracy": 0.12989716455340386,
|
|
"num_tokens": 2225784.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.192068195343017,
|
|
"epoch": 0.9813721035892776,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999307078306876,
|
|
"loss": 6.2269,
|
|
"mean_token_accuracy": 0.1377440184354782,
|
|
"num_tokens": 2235357.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.349479484558105,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999216596267708,
|
|
"loss": 6.2666,
|
|
"mean_token_accuracy": 0.1292445532977581,
|
|
"num_tokens": 2246438.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.231703758239746,
|
|
"epoch": 0.9904588823262154,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999120564510177,
|
|
"loss": 6.2487,
|
|
"mean_token_accuracy": 0.12972478568553925,
|
|
"num_tokens": 2257528.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.269535779953003,
|
|
"epoch": 0.9950022716946842,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000499901898327123,
|
|
"loss": 6.2736,
|
|
"mean_token_accuracy": 0.13083961233496666,
|
|
"num_tokens": 2267572.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.2416705131530765,
|
|
"epoch": 0.9995456610631531,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998911852801512,
|
|
"loss": 6.2093,
|
|
"mean_token_accuracy": 0.1358557313680649,
|
|
"num_tokens": 2278520.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.27468204498291,
|
|
"epoch": 1.003634711494775,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998799173365355,
|
|
"loss": 6.1262,
|
|
"mean_token_accuracy": 0.1319936273826493,
|
|
"num_tokens": 2286991.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.2702305793762205,
|
|
"epoch": 1.008178100863244,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998680945240783,
|
|
"loss": 6.151,
|
|
"mean_token_accuracy": 0.12948329746723175,
|
|
"num_tokens": 2297857.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.24632716178894,
|
|
"epoch": 1.0127214902317128,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998557168719513,
|
|
"loss": 6.2073,
|
|
"mean_token_accuracy": 0.13167741447687148,
|
|
"num_tokens": 2308621.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.22882342338562,
|
|
"epoch": 1.0172648796001817,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998427844106954,
|
|
"loss": 6.0279,
|
|
"mean_token_accuracy": 0.140340007096529,
|
|
"num_tokens": 2318654.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.193508720397949,
|
|
"epoch": 1.0218082689686505,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998292971722197,
|
|
"loss": 6.0539,
|
|
"mean_token_accuracy": 0.13239168524742126,
|
|
"num_tokens": 2329064.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.191493892669678,
|
|
"epoch": 1.0263516583371195,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499815255189803,
|
|
"loss": 6.1278,
|
|
"mean_token_accuracy": 0.1320304274559021,
|
|
"num_tokens": 2339843.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.224408054351807,
|
|
"epoch": 1.0308950477055883,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998006584980922,
|
|
"loss": 6.0619,
|
|
"mean_token_accuracy": 0.13299913033843042,
|
|
"num_tokens": 2350715.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.196117687225342,
|
|
"epoch": 1.0354384370740572,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997855071331034,
|
|
"loss": 6.0276,
|
|
"mean_token_accuracy": 0.13162992000579835,
|
|
"num_tokens": 2360635.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.174038362503052,
|
|
"epoch": 1.039981826442526,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997698011322211,
|
|
"loss": 6.052,
|
|
"mean_token_accuracy": 0.1327362336218357,
|
|
"num_tokens": 2370627.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.229216814041138,
|
|
"epoch": 1.044525215810995,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499753540534198,
|
|
"loss": 6.0767,
|
|
"mean_token_accuracy": 0.13144420832395554,
|
|
"num_tokens": 2381790.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.041177749633789,
|
|
"epoch": 1.049068605179464,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499736725379156,
|
|
"loss": 6.0296,
|
|
"mean_token_accuracy": 0.13484439700841905,
|
|
"num_tokens": 2392685.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.155071020126343,
|
|
"epoch": 1.0536119945479328,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997193557085844,
|
|
"loss": 6.07,
|
|
"mean_token_accuracy": 0.13602437153458596,
|
|
"num_tokens": 2403722.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.1968738555908205,
|
|
"epoch": 1.0581553839164017,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004997014315653413,
|
|
"loss": 6.1232,
|
|
"mean_token_accuracy": 0.12442034780979157,
|
|
"num_tokens": 2415311.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.174206924438477,
|
|
"epoch": 1.0626987732848705,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004996829529936527,
|
|
"loss": 6.0091,
|
|
"mean_token_accuracy": 0.14058221727609635,
|
|
"num_tokens": 2425116.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.210444021224975,
|
|
"epoch": 1.0672421626533395,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996639200391126,
|
|
"loss": 6.1816,
|
|
"mean_token_accuracy": 0.13474437668919564,
|
|
"num_tokens": 2435916.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.097880458831787,
|
|
"epoch": 1.0717855520218083,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996443327486831,
|
|
"loss": 6.0697,
|
|
"mean_token_accuracy": 0.13110727071762085,
|
|
"num_tokens": 2446920.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.164422178268433,
|
|
"epoch": 1.0763289413902772,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996241911706937,
|
|
"loss": 6.0117,
|
|
"mean_token_accuracy": 0.1407651871442795,
|
|
"num_tokens": 2456367.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.152524423599243,
|
|
"epoch": 1.080872330758746,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004996034953548418,
|
|
"loss": 5.9799,
|
|
"mean_token_accuracy": 0.1370386466383934,
|
|
"num_tokens": 2466790.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.116287517547607,
|
|
"epoch": 1.085415720127215,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995822453521924,
|
|
"loss": 6.0155,
|
|
"mean_token_accuracy": 0.1394006423652172,
|
|
"num_tokens": 2476419.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.150832033157348,
|
|
"epoch": 1.0899591094956838,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004995604412151776,
|
|
"loss": 6.1381,
|
|
"mean_token_accuracy": 0.13063833564519883,
|
|
"num_tokens": 2487295.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.16593828201294,
|
|
"epoch": 1.0945024988641527,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499538082997597,
|
|
"loss": 6.0124,
|
|
"mean_token_accuracy": 0.13131626322865486,
|
|
"num_tokens": 2497545.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.1826817989349365,
|
|
"epoch": 1.0990458882326215,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995151707546172,
|
|
"loss": 6.1714,
|
|
"mean_token_accuracy": 0.13120559602975845,
|
|
"num_tokens": 2507805.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.146686792373657,
|
|
"epoch": 1.1035892776010905,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004994917045427721,
|
|
"loss": 6.0463,
|
|
"mean_token_accuracy": 0.1330270990729332,
|
|
"num_tokens": 2518474.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.248140716552735,
|
|
"epoch": 1.1081326669695593,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004994676844199621,
|
|
"loss": 6.0825,
|
|
"mean_token_accuracy": 0.1410455100238323,
|
|
"num_tokens": 2528974.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.093984794616699,
|
|
"epoch": 1.1126760563380282,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004994431104454543,
|
|
"loss": 6.1104,
|
|
"mean_token_accuracy": 0.13412080109119415,
|
|
"num_tokens": 2540675.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.076309013366699,
|
|
"epoch": 1.117219445706497,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004994179826798828,
|
|
"loss": 5.9887,
|
|
"mean_token_accuracy": 0.13977164998650551,
|
|
"num_tokens": 2551367.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.169183015823364,
|
|
"epoch": 1.121762835074966,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004993923011852477,
|
|
"loss": 6.0996,
|
|
"mean_token_accuracy": 0.1306145392358303,
|
|
"num_tokens": 2562096.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.169359540939331,
|
|
"epoch": 1.1263062244434348,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004993660660249158,
|
|
"loss": 6.1549,
|
|
"mean_token_accuracy": 0.1343975305557251,
|
|
"num_tokens": 2572917.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.109742593765259,
|
|
"epoch": 1.1308496138119037,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004993392772636194,
|
|
"loss": 6.0009,
|
|
"mean_token_accuracy": 0.13797525987029075,
|
|
"num_tokens": 2582971.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.253916501998901,
|
|
"epoch": 1.1353930031803725,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004993119349674573,
|
|
"loss": 6.1411,
|
|
"mean_token_accuracy": 0.13633586689829827,
|
|
"num_tokens": 2593186.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.126196384429932,
|
|
"epoch": 1.1399363925488415,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004992840392038938,
|
|
"loss": 6.1267,
|
|
"mean_token_accuracy": 0.13513572216033937,
|
|
"num_tokens": 2604359.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.173921966552735,
|
|
"epoch": 1.1444797819173103,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499255590041759,
|
|
"loss": 6.0446,
|
|
"mean_token_accuracy": 0.13811376094818115,
|
|
"num_tokens": 2614436.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.065403413772583,
|
|
"epoch": 1.1490231712857792,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004992265875512485,
|
|
"loss": 6.0111,
|
|
"mean_token_accuracy": 0.14057485163211822,
|
|
"num_tokens": 2624175.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.13704743385315,
|
|
"epoch": 1.153566560654248,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004991970318039228,
|
|
"loss": 6.0504,
|
|
"mean_token_accuracy": 0.14092691838741303,
|
|
"num_tokens": 2634898.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.221057033538818,
|
|
"epoch": 1.158109950022717,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991669228727078,
|
|
"loss": 6.0714,
|
|
"mean_token_accuracy": 0.14015807211399078,
|
|
"num_tokens": 2645590.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.16869707107544,
|
|
"epoch": 1.1626533393911858,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004991362608318947,
|
|
"loss": 6.065,
|
|
"mean_token_accuracy": 0.13930616527795792,
|
|
"num_tokens": 2656115.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.07602276802063,
|
|
"epoch": 1.1671967287596547,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991050457571384,
|
|
"loss": 6.0508,
|
|
"mean_token_accuracy": 0.1356099359691143,
|
|
"num_tokens": 2666760.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.166289806365967,
|
|
"epoch": 1.1717401181281235,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004990732777254594,
|
|
"loss": 6.0015,
|
|
"mean_token_accuracy": 0.1397775262594223,
|
|
"num_tokens": 2676763.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.027611351013183,
|
|
"epoch": 1.1762835074965925,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004990409568152422,
|
|
"loss": 6.0354,
|
|
"mean_token_accuracy": 0.13115530014038085,
|
|
"num_tokens": 2686598.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.110258722305298,
|
|
"epoch": 1.1808268968650613,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004990080831062352,
|
|
"loss": 5.9848,
|
|
"mean_token_accuracy": 0.14191189110279084,
|
|
"num_tokens": 2697767.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.085461187362671,
|
|
"epoch": 1.1853702862335302,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004989746566795512,
|
|
"loss": 5.9869,
|
|
"mean_token_accuracy": 0.13826760947704314,
|
|
"num_tokens": 2707081.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.039988088607788,
|
|
"epoch": 1.189913675601999,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004989406776176665,
|
|
"loss": 6.0147,
|
|
"mean_token_accuracy": 0.1434862032532692,
|
|
"num_tokens": 2717312.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.1803795337677006,
|
|
"epoch": 1.194457064970468,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000498906146004421,
|
|
"loss": 6.0388,
|
|
"mean_token_accuracy": 0.13803832307457925,
|
|
"num_tokens": 2728108.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.206027221679688,
|
|
"epoch": 1.1990004543389368,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004988710619250181,
|
|
"loss": 6.2267,
|
|
"mean_token_accuracy": 0.13564569801092147,
|
|
"num_tokens": 2738860.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.112774801254273,
|
|
"epoch": 1.2035438437074057,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004988354254660244,
|
|
"loss": 6.0746,
|
|
"mean_token_accuracy": 0.1385789170861244,
|
|
"num_tokens": 2749786.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.077335548400879,
|
|
"epoch": 1.2080872330758745,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004987992367153691,
|
|
"loss": 5.9793,
|
|
"mean_token_accuracy": 0.13855826556682588,
|
|
"num_tokens": 2760018.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.050770044326782,
|
|
"epoch": 1.2126306224443435,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004987624957623443,
|
|
"loss": 6.0223,
|
|
"mean_token_accuracy": 0.13678463250398637,
|
|
"num_tokens": 2770572.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.047999429702759,
|
|
"epoch": 1.2171740118128123,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004987252026976049,
|
|
"loss": 5.9931,
|
|
"mean_token_accuracy": 0.14768794775009156,
|
|
"num_tokens": 2780297.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.132225608825683,
|
|
"epoch": 1.2217174011812812,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004986873576131676,
|
|
"loss": 6.0959,
|
|
"mean_token_accuracy": 0.14233560636639594,
|
|
"num_tokens": 2790454.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.069753742218017,
|
|
"epoch": 1.22626079054975,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004986489606024116,
|
|
"loss": 5.9649,
|
|
"mean_token_accuracy": 0.14346184730529785,
|
|
"num_tokens": 2800034.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.9972669124603275,
|
|
"epoch": 1.230804179918219,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004986100117600776,
|
|
"loss": 5.9658,
|
|
"mean_token_accuracy": 0.14471054077148438,
|
|
"num_tokens": 2810157.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.0917027473449705,
|
|
"epoch": 1.235347569286688,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000498570511182268,
|
|
"loss": 5.9956,
|
|
"mean_token_accuracy": 0.13442177325487137,
|
|
"num_tokens": 2819404.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.028844976425171,
|
|
"epoch": 1.2398909586551567,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004985304589664465,
|
|
"loss": 5.9316,
|
|
"mean_token_accuracy": 0.1416188210248947,
|
|
"num_tokens": 2829551.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.041907739639282,
|
|
"epoch": 1.2444343480236255,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004984898552114381,
|
|
"loss": 6.0028,
|
|
"mean_token_accuracy": 0.13631764501333238,
|
|
"num_tokens": 2840381.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.0108194828033445,
|
|
"epoch": 1.2489777373920945,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004984487000174283,
|
|
"loss": 5.9934,
|
|
"mean_token_accuracy": 0.1371256522834301,
|
|
"num_tokens": 2850279.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.18448691368103,
|
|
"epoch": 1.2535211267605635,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004984069934859637,
|
|
"loss": 6.0117,
|
|
"mean_token_accuracy": 0.13990146294236183,
|
|
"num_tokens": 2860209.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.987959003448486,
|
|
"epoch": 1.2580645161290323,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004983647357199509,
|
|
"loss": 6.0216,
|
|
"mean_token_accuracy": 0.14364159107208252,
|
|
"num_tokens": 2870932.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.155229139328003,
|
|
"epoch": 1.262607905497501,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983219268236567,
|
|
"loss": 6.0152,
|
|
"mean_token_accuracy": 0.13599357455968858,
|
|
"num_tokens": 2881139.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.115420246124268,
|
|
"epoch": 1.26715129486597,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004982785669027079,
|
|
"loss": 6.0034,
|
|
"mean_token_accuracy": 0.14194265007972717,
|
|
"num_tokens": 2890774.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.017755174636841,
|
|
"epoch": 1.271694684234439,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004982346560640907,
|
|
"loss": 5.954,
|
|
"mean_token_accuracy": 0.15009693652391434,
|
|
"num_tokens": 2900259.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.056285047531128,
|
|
"epoch": 1.2762380736029078,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004981901944161507,
|
|
"loss": 5.9704,
|
|
"mean_token_accuracy": 0.14565521255135536,
|
|
"num_tokens": 2911365.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.081052160263061,
|
|
"epoch": 1.2807814629713765,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004981451820685928,
|
|
"loss": 5.9768,
|
|
"mean_token_accuracy": 0.13834436386823654,
|
|
"num_tokens": 2921189.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.95786681175232,
|
|
"epoch": 1.2853248523398455,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004980996191324802,
|
|
"loss": 5.9971,
|
|
"mean_token_accuracy": 0.1435582846403122,
|
|
"num_tokens": 2931271.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.106979179382324,
|
|
"epoch": 1.2898682417083145,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004980535057202351,
|
|
"loss": 6.0026,
|
|
"mean_token_accuracy": 0.1380285695195198,
|
|
"num_tokens": 2941557.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.976951837539673,
|
|
"epoch": 1.2944116310767833,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004980068419456378,
|
|
"loss": 5.9231,
|
|
"mean_token_accuracy": 0.15052516162395477,
|
|
"num_tokens": 2951305.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.0214759349823,
|
|
"epoch": 1.2989550204452522,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004979596279238265,
|
|
"loss": 6.0034,
|
|
"mean_token_accuracy": 0.14994281977415086,
|
|
"num_tokens": 2961948.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.065707969665527,
|
|
"epoch": 1.303498409813721,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004979118637712969,
|
|
"loss": 6.0482,
|
|
"mean_token_accuracy": 0.13576218634843826,
|
|
"num_tokens": 2973300.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.1157904148101805,
|
|
"epoch": 1.30804179918219,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004978635496059025,
|
|
"loss": 6.0069,
|
|
"mean_token_accuracy": 0.1413679003715515,
|
|
"num_tokens": 2984114.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.9245984077453615,
|
|
"epoch": 1.3125851885506588,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004978146855468537,
|
|
"loss": 5.8607,
|
|
"mean_token_accuracy": 0.14962516129016876,
|
|
"num_tokens": 2993940.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.042234468460083,
|
|
"epoch": 1.3171285779191277,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004977652717147177,
|
|
"loss": 6.0009,
|
|
"mean_token_accuracy": 0.14177796244621277,
|
|
"num_tokens": 3004093.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.918850517272949,
|
|
"epoch": 1.3216719672875965,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004977153082314182,
|
|
"loss": 5.9063,
|
|
"mean_token_accuracy": 0.14089421778917313,
|
|
"num_tokens": 3013537.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.048681545257568,
|
|
"epoch": 1.3262153566560655,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000497664795220235,
|
|
"loss": 5.9193,
|
|
"mean_token_accuracy": 0.14717012643814087,
|
|
"num_tokens": 3023502.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.0000035762786865,
|
|
"epoch": 1.3307587460245343,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004976137328058042,
|
|
"loss": 5.9669,
|
|
"mean_token_accuracy": 0.13930393680930137,
|
|
"num_tokens": 3033910.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.938542127609253,
|
|
"epoch": 1.3353021353930032,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497562121114117,
|
|
"loss": 6.0137,
|
|
"mean_token_accuracy": 0.143656387925148,
|
|
"num_tokens": 3043500.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.067913389205932,
|
|
"epoch": 1.339845524761472,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004975099602725202,
|
|
"loss": 5.9781,
|
|
"mean_token_accuracy": 0.14194607511162757,
|
|
"num_tokens": 3054776.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.02539324760437,
|
|
"epoch": 1.344388914129941,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004974572504097154,
|
|
"loss": 5.9354,
|
|
"mean_token_accuracy": 0.14674000740051268,
|
|
"num_tokens": 3065403.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.979982328414917,
|
|
"epoch": 1.3489323034984098,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004974039916557591,
|
|
"loss": 5.9775,
|
|
"mean_token_accuracy": 0.1461941346526146,
|
|
"num_tokens": 3074661.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.957530736923218,
|
|
"epoch": 1.3534756928668787,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000497350184142062,
|
|
"loss": 5.9289,
|
|
"mean_token_accuracy": 0.14567428082227707,
|
|
"num_tokens": 3085319.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.022159147262573,
|
|
"epoch": 1.3580190822353475,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004972958280013886,
|
|
"loss": 5.9129,
|
|
"mean_token_accuracy": 0.1448183536529541,
|
|
"num_tokens": 3095115.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.046401786804199,
|
|
"epoch": 1.3625624716038165,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004972409233678575,
|
|
"loss": 6.0999,
|
|
"mean_token_accuracy": 0.1356892429292202,
|
|
"num_tokens": 3104841.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.3625624716038165,
|
|
"eval_entropy": 6.051776233939237,
|
|
"eval_loss": 6.2143402099609375,
|
|
"eval_mean_token_accuracy": 0.13918755460790425,
|
|
"eval_num_tokens": 3104841.0,
|
|
"eval_runtime": 2.5942,
|
|
"eval_samples_per_second": 1324.128,
|
|
"eval_steps_per_second": 165.757,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.050335216522217,
|
|
"epoch": 1.3671058609722853,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004971854703769401,
|
|
"loss": 5.9548,
|
|
"mean_token_accuracy": 0.14158371612429618,
|
|
"num_tokens": 3115098.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.984045791625976,
|
|
"epoch": 1.3716492503407542,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004971294691654616,
|
|
"loss": 5.862,
|
|
"mean_token_accuracy": 0.1499510645866394,
|
|
"num_tokens": 3124831.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.947478723526001,
|
|
"epoch": 1.376192639709223,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497072919871599,
|
|
"loss": 6.027,
|
|
"mean_token_accuracy": 0.1404582031071186,
|
|
"num_tokens": 3136113.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.205549621582032,
|
|
"epoch": 1.380736029077692,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004970158226348824,
|
|
"loss": 5.9554,
|
|
"mean_token_accuracy": 0.15334268063306808,
|
|
"num_tokens": 3145617.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.889804887771606,
|
|
"epoch": 1.3852794184461608,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004969581775961932,
|
|
"loss": 6.0339,
|
|
"mean_token_accuracy": 0.13728870823979378,
|
|
"num_tokens": 3156913.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.111681795120239,
|
|
"epoch": 1.3898228078146297,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004968999848977651,
|
|
"loss": 5.972,
|
|
"mean_token_accuracy": 0.14302797764539718,
|
|
"num_tokens": 3167506.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.898993492126465,
|
|
"epoch": 1.3943661971830985,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004968412446831827,
|
|
"loss": 5.864,
|
|
"mean_token_accuracy": 0.14338979199528695,
|
|
"num_tokens": 3177956.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.047224426269532,
|
|
"epoch": 1.3989095865515675,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004967819570973816,
|
|
"loss": 6.1159,
|
|
"mean_token_accuracy": 0.1445922702550888,
|
|
"num_tokens": 3189172.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.124930667877197,
|
|
"epoch": 1.4034529759200365,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000496722122286648,
|
|
"loss": 6.0225,
|
|
"mean_token_accuracy": 0.14140935614705086,
|
|
"num_tokens": 3200188.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.934967756271362,
|
|
"epoch": 1.4079963652885052,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004966617403986185,
|
|
"loss": 5.8657,
|
|
"mean_token_accuracy": 0.13875720053911209,
|
|
"num_tokens": 3210171.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.086915922164917,
|
|
"epoch": 1.412539754656974,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004966008115822793,
|
|
"loss": 6.1055,
|
|
"mean_token_accuracy": 0.13564804196357727,
|
|
"num_tokens": 3221201.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.940374565124512,
|
|
"epoch": 1.417083144025443,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004965393359879662,
|
|
"loss": 5.9362,
|
|
"mean_token_accuracy": 0.14272732883691788,
|
|
"num_tokens": 3231825.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.1064142227172855,
|
|
"epoch": 1.421626533393912,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004964773137673641,
|
|
"loss": 6.1405,
|
|
"mean_token_accuracy": 0.13389862030744554,
|
|
"num_tokens": 3241860.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.945473718643188,
|
|
"epoch": 1.4261699227623807,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004964147450735067,
|
|
"loss": 5.7977,
|
|
"mean_token_accuracy": 0.15062655210494996,
|
|
"num_tokens": 3251893.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.019118165969848,
|
|
"epoch": 1.4307133121308495,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000496351630060776,
|
|
"loss": 5.9307,
|
|
"mean_token_accuracy": 0.1430902898311615,
|
|
"num_tokens": 3262748.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.037960863113403,
|
|
"epoch": 1.4352567014993185,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004962879688849021,
|
|
"loss": 6.0417,
|
|
"mean_token_accuracy": 0.13982143104076386,
|
|
"num_tokens": 3273035.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.080470561981201,
|
|
"epoch": 1.4398000908677875,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004962237617029625,
|
|
"loss": 5.9723,
|
|
"mean_token_accuracy": 0.14449140578508377,
|
|
"num_tokens": 3283413.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.956830739974976,
|
|
"epoch": 1.4443434802362562,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004961590086733821,
|
|
"loss": 5.9563,
|
|
"mean_token_accuracy": 0.1427573099732399,
|
|
"num_tokens": 3294628.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.9700994968414305,
|
|
"epoch": 1.448886869604725,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004960937099559326,
|
|
"loss": 5.9326,
|
|
"mean_token_accuracy": 0.15429284125566484,
|
|
"num_tokens": 3304631.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.979518175125122,
|
|
"epoch": 1.453430258973194,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004960278657117321,
|
|
"loss": 5.956,
|
|
"mean_token_accuracy": 0.14319687187671662,
|
|
"num_tokens": 3315481.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.004811382293701,
|
|
"epoch": 1.457973648341663,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004959614761032446,
|
|
"loss": 5.8905,
|
|
"mean_token_accuracy": 0.1432957947254181,
|
|
"num_tokens": 3326493.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.090801811218261,
|
|
"epoch": 1.4625170377101318,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00049589454129428,
|
|
"loss": 6.0285,
|
|
"mean_token_accuracy": 0.14221277907490731,
|
|
"num_tokens": 3338358.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.009984302520752,
|
|
"epoch": 1.4670604270786005,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004958270614499932,
|
|
"loss": 6.0025,
|
|
"mean_token_accuracy": 0.13881809562444686,
|
|
"num_tokens": 3347961.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.023412275314331,
|
|
"epoch": 1.4716038164470695,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004957590367368841,
|
|
"loss": 5.9063,
|
|
"mean_token_accuracy": 0.1509341448545456,
|
|
"num_tokens": 3358544.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.927429676055908,
|
|
"epoch": 1.4761472058155385,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004956904673227968,
|
|
"loss": 5.9387,
|
|
"mean_token_accuracy": 0.14586805701255798,
|
|
"num_tokens": 3368233.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.989821195602417,
|
|
"epoch": 1.4806905951840073,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004956213533769195,
|
|
"loss": 5.8955,
|
|
"mean_token_accuracy": 0.15005389750003814,
|
|
"num_tokens": 3379065.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.039559459686279,
|
|
"epoch": 1.485233984552476,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004955516950697843,
|
|
"loss": 5.9877,
|
|
"mean_token_accuracy": 0.1411728210747242,
|
|
"num_tokens": 3389251.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.919572973251343,
|
|
"epoch": 1.489777373920945,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004954814925732657,
|
|
"loss": 5.9118,
|
|
"mean_token_accuracy": 0.1507558375597,
|
|
"num_tokens": 3399660.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.050065708160401,
|
|
"epoch": 1.494320763289414,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004954107460605816,
|
|
"loss": 5.9473,
|
|
"mean_token_accuracy": 0.14760428965091704,
|
|
"num_tokens": 3409395.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.003555774688721,
|
|
"epoch": 1.4988641526578828,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004953394557062922,
|
|
"loss": 5.9364,
|
|
"mean_token_accuracy": 0.1427161142230034,
|
|
"num_tokens": 3419881.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.981671142578125,
|
|
"epoch": 1.5034075420263515,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004952676216862989,
|
|
"loss": 5.9546,
|
|
"mean_token_accuracy": 0.14186897575855256,
|
|
"num_tokens": 3430316.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.940034961700439,
|
|
"epoch": 1.5079509313948205,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004951952441778453,
|
|
"loss": 5.8452,
|
|
"mean_token_accuracy": 0.15199173986911774,
|
|
"num_tokens": 3440174.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.9946540832519535,
|
|
"epoch": 1.5124943207632895,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004951223233595158,
|
|
"loss": 5.989,
|
|
"mean_token_accuracy": 0.14114882349967955,
|
|
"num_tokens": 3451141.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.006590938568115,
|
|
"epoch": 1.5170377101317583,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000495048859411235,
|
|
"loss": 5.892,
|
|
"mean_token_accuracy": 0.14501796513795853,
|
|
"num_tokens": 3461767.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.894695568084717,
|
|
"epoch": 1.521581099500227,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004949748525142681,
|
|
"loss": 5.8518,
|
|
"mean_token_accuracy": 0.1538882002234459,
|
|
"num_tokens": 3470937.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.002769804000854,
|
|
"epoch": 1.526124488868696,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004949003028512195,
|
|
"loss": 5.9313,
|
|
"mean_token_accuracy": 0.144069167971611,
|
|
"num_tokens": 3482130.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.932562208175659,
|
|
"epoch": 1.530667878237165,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004948252106060334,
|
|
"loss": 5.9906,
|
|
"mean_token_accuracy": 0.13914068043231964,
|
|
"num_tokens": 3491992.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.911404800415039,
|
|
"epoch": 1.5352112676056338,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004947495759639923,
|
|
"loss": 5.8496,
|
|
"mean_token_accuracy": 0.15234332978725434,
|
|
"num_tokens": 3501437.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.877619218826294,
|
|
"epoch": 1.5397546569741025,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004946733991117171,
|
|
"loss": 5.9091,
|
|
"mean_token_accuracy": 0.15192992836236954,
|
|
"num_tokens": 3511625.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.954566287994385,
|
|
"epoch": 1.5442980463425715,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004945966802371668,
|
|
"loss": 5.7032,
|
|
"mean_token_accuracy": 0.15604728013277053,
|
|
"num_tokens": 3520428.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.913967275619507,
|
|
"epoch": 1.5488414357110405,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004945194195296374,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.14569951593875885,
|
|
"num_tokens": 3532218.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.993043708801269,
|
|
"epoch": 1.5533848250795095,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000494441617179762,
|
|
"loss": 5.923,
|
|
"mean_token_accuracy": 0.14351641461253167,
|
|
"num_tokens": 3543177.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.0035828113555905,
|
|
"epoch": 1.5579282144479782,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004943632733795104,
|
|
"loss": 5.8716,
|
|
"mean_token_accuracy": 0.15494320765137673,
|
|
"num_tokens": 3552833.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.769624853134156,
|
|
"epoch": 1.562471603816447,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004942843883221879,
|
|
"loss": 5.9311,
|
|
"mean_token_accuracy": 0.1427910655736923,
|
|
"num_tokens": 3563558.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.974128246307373,
|
|
"epoch": 1.567014993184916,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004942049622024357,
|
|
"loss": 5.8835,
|
|
"mean_token_accuracy": 0.15125887244939804,
|
|
"num_tokens": 3574993.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.954115867614746,
|
|
"epoch": 1.571558382553385,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004941249952162298,
|
|
"loss": 5.9015,
|
|
"mean_token_accuracy": 0.14047565683722496,
|
|
"num_tokens": 3584586.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.850758361816406,
|
|
"epoch": 1.5761017719218537,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004940444875608807,
|
|
"loss": 5.8515,
|
|
"mean_token_accuracy": 0.1485603392124176,
|
|
"num_tokens": 3595029.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.991342544555664,
|
|
"epoch": 1.5806451612903225,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004939634394350332,
|
|
"loss": 5.924,
|
|
"mean_token_accuracy": 0.14164209067821504,
|
|
"num_tokens": 3606034.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.945333099365234,
|
|
"epoch": 1.5851885506587915,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004938818510386654,
|
|
"loss": 5.8472,
|
|
"mean_token_accuracy": 0.15045266672968866,
|
|
"num_tokens": 3615551.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.950373554229737,
|
|
"epoch": 1.5897319400272605,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004937997225730886,
|
|
"loss": 5.9905,
|
|
"mean_token_accuracy": 0.1494086466729641,
|
|
"num_tokens": 3626811.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.850814914703369,
|
|
"epoch": 1.5942753293957292,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004937170542409464,
|
|
"loss": 5.8531,
|
|
"mean_token_accuracy": 0.14677981585264205,
|
|
"num_tokens": 3637338.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.857586288452149,
|
|
"epoch": 1.598818718764198,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000493633846246215,
|
|
"loss": 5.7922,
|
|
"mean_token_accuracy": 0.15139450058341025,
|
|
"num_tokens": 3648650.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.970294570922851,
|
|
"epoch": 1.603362108132667,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004935500987942018,
|
|
"loss": 5.959,
|
|
"mean_token_accuracy": 0.14909740760922433,
|
|
"num_tokens": 3659331.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.033718061447144,
|
|
"epoch": 1.607905497501136,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004934658120915452,
|
|
"loss": 6.005,
|
|
"mean_token_accuracy": 0.1395428881049156,
|
|
"num_tokens": 3670757.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.917076873779297,
|
|
"epoch": 1.6124488868696047,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004933809863462142,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.15006528198719024,
|
|
"num_tokens": 3681286.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.8674839496612545,
|
|
"epoch": 1.6169922762380735,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004932956217675082,
|
|
"loss": 5.8518,
|
|
"mean_token_accuracy": 0.15662607550621033,
|
|
"num_tokens": 3690733.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.03576340675354,
|
|
"epoch": 1.6215356656065425,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004932097185660555,
|
|
"loss": 6.019,
|
|
"mean_token_accuracy": 0.14538080394268035,
|
|
"num_tokens": 3701263.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.9428391456604,
|
|
"epoch": 1.6260790549750115,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004931232769538138,
|
|
"loss": 5.9292,
|
|
"mean_token_accuracy": 0.14668467864394188,
|
|
"num_tokens": 3711578.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.033915615081787,
|
|
"epoch": 1.6306224443434802,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004930362971440692,
|
|
"loss": 5.9245,
|
|
"mean_token_accuracy": 0.14553339928388595,
|
|
"num_tokens": 3722140.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.790395975112915,
|
|
"epoch": 1.635165833711949,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004929487793514358,
|
|
"loss": 5.6341,
|
|
"mean_token_accuracy": 0.15669778138399124,
|
|
"num_tokens": 3732019.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.913341903686524,
|
|
"epoch": 1.639709223080418,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000492860723791855,
|
|
"loss": 5.8385,
|
|
"mean_token_accuracy": 0.1515497460961342,
|
|
"num_tokens": 3741912.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.881040906906128,
|
|
"epoch": 1.644252612448887,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004927721306825952,
|
|
"loss": 5.8319,
|
|
"mean_token_accuracy": 0.1510633796453476,
|
|
"num_tokens": 3751365.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.899726629257202,
|
|
"epoch": 1.6487960018173557,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004926830002422509,
|
|
"loss": 5.8163,
|
|
"mean_token_accuracy": 0.14396188259124756,
|
|
"num_tokens": 3762160.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.8803504467010494,
|
|
"epoch": 1.6533393911858245,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000492593332690743,
|
|
"loss": 5.845,
|
|
"mean_token_accuracy": 0.15151019394397736,
|
|
"num_tokens": 3772608.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.916077995300293,
|
|
"epoch": 1.6578827805542935,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004925031282493167,
|
|
"loss": 5.8965,
|
|
"mean_token_accuracy": 0.1424654796719551,
|
|
"num_tokens": 3783344.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.949909305572509,
|
|
"epoch": 1.6624261699227625,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004924123871405428,
|
|
"loss": 5.8992,
|
|
"mean_token_accuracy": 0.145297309756279,
|
|
"num_tokens": 3794103.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.804653549194336,
|
|
"epoch": 1.6669695592912313,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004923211095883161,
|
|
"loss": 5.8398,
|
|
"mean_token_accuracy": 0.15114678516983987,
|
|
"num_tokens": 3804537.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.938943338394165,
|
|
"epoch": 1.6715129486597,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004922292958178546,
|
|
"loss": 5.8405,
|
|
"mean_token_accuracy": 0.1490088000893593,
|
|
"num_tokens": 3814147.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.955902051925659,
|
|
"epoch": 1.676056338028169,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004921369460557,
|
|
"loss": 5.9434,
|
|
"mean_token_accuracy": 0.14378588423132896,
|
|
"num_tokens": 3823640.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.904258441925049,
|
|
"epoch": 1.680599727396638,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004920440605297159,
|
|
"loss": 5.7868,
|
|
"mean_token_accuracy": 0.15925955027341843,
|
|
"num_tokens": 3833418.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.84422812461853,
|
|
"epoch": 1.6851431167651068,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000491950639469088,
|
|
"loss": 5.7977,
|
|
"mean_token_accuracy": 0.15323520004749297,
|
|
"num_tokens": 3844998.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.8427070617675785,
|
|
"epoch": 1.6896865061335755,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004918566831043238,
|
|
"loss": 5.7205,
|
|
"mean_token_accuracy": 0.15908368229866027,
|
|
"num_tokens": 3855549.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.833364629745484,
|
|
"epoch": 1.6942298955020445,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000491762191667251,
|
|
"loss": 5.8521,
|
|
"mean_token_accuracy": 0.15248288363218307,
|
|
"num_tokens": 3865287.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.9676250457763675,
|
|
"epoch": 1.6987732848705135,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000491667165391018,
|
|
"loss": 5.9003,
|
|
"mean_token_accuracy": 0.14770784750580787,
|
|
"num_tokens": 3875914.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.967091417312622,
|
|
"epoch": 1.7033166742389823,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004915716045100927,
|
|
"loss": 5.8958,
|
|
"mean_token_accuracy": 0.14882167130708696,
|
|
"num_tokens": 3885196.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.962908554077148,
|
|
"epoch": 1.707860063607451,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004914755092602621,
|
|
"loss": 5.9387,
|
|
"mean_token_accuracy": 0.13579298481345176,
|
|
"num_tokens": 3895227.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.8981880187988285,
|
|
"epoch": 1.71240345297592,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004913788798786314,
|
|
"loss": 5.8048,
|
|
"mean_token_accuracy": 0.1507883608341217,
|
|
"num_tokens": 3905926.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.861813402175903,
|
|
"epoch": 1.716946842344389,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004912817166036243,
|
|
"loss": 5.843,
|
|
"mean_token_accuracy": 0.15541554391384124,
|
|
"num_tokens": 3916982.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.811162424087525,
|
|
"epoch": 1.7214902317128578,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004911840196749813,
|
|
"loss": 5.7675,
|
|
"mean_token_accuracy": 0.15070637315511703,
|
|
"num_tokens": 3926889.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.922468662261963,
|
|
"epoch": 1.7260336210813265,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004910857893337602,
|
|
"loss": 5.8046,
|
|
"mean_token_accuracy": 0.1469975158572197,
|
|
"num_tokens": 3936734.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.8749395370483395,
|
|
"epoch": 1.7305770104497955,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004909870258223342,
|
|
"loss": 5.9333,
|
|
"mean_token_accuracy": 0.14144831746816636,
|
|
"num_tokens": 3948001.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.931931734085083,
|
|
"epoch": 1.7351203998182645,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004908877293843926,
|
|
"loss": 5.814,
|
|
"mean_token_accuracy": 0.14603467136621476,
|
|
"num_tokens": 3957433.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.943529415130615,
|
|
"epoch": 1.7396637891867333,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004907879002649395,
|
|
"loss": 5.8902,
|
|
"mean_token_accuracy": 0.1463154137134552,
|
|
"num_tokens": 3967375.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.874042415618897,
|
|
"epoch": 1.744207178555202,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004906875387102935,
|
|
"loss": 5.8723,
|
|
"mean_token_accuracy": 0.14370815306901932,
|
|
"num_tokens": 3977190.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.898986196517944,
|
|
"epoch": 1.748750567923671,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004905866449680864,
|
|
"loss": 5.8513,
|
|
"mean_token_accuracy": 0.15149876177310945,
|
|
"num_tokens": 3987878.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.870923376083374,
|
|
"epoch": 1.75329395729214,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004904852192872638,
|
|
"loss": 5.8404,
|
|
"mean_token_accuracy": 0.15223072916269303,
|
|
"num_tokens": 3997166.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.91577672958374,
|
|
"epoch": 1.757837346660609,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004903832619180834,
|
|
"loss": 5.837,
|
|
"mean_token_accuracy": 0.14691349267959594,
|
|
"num_tokens": 4007711.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.944367170333862,
|
|
"epoch": 1.7623807360290777,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004902807731121147,
|
|
"loss": 5.9863,
|
|
"mean_token_accuracy": 0.14061963111162185,
|
|
"num_tokens": 4018907.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.853556299209595,
|
|
"epoch": 1.7669241253975465,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004901777531222389,
|
|
"loss": 5.8637,
|
|
"mean_token_accuracy": 0.15473533272743226,
|
|
"num_tokens": 4030006.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.048425912857056,
|
|
"epoch": 1.7714675147660155,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004900742022026474,
|
|
"loss": 5.9613,
|
|
"mean_token_accuracy": 0.15112675130367278,
|
|
"num_tokens": 4040609.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.814242553710938,
|
|
"epoch": 1.7760109041344845,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004899701206088419,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.14815148264169692,
|
|
"num_tokens": 4050392.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.914210462570191,
|
|
"epoch": 1.7805542935029532,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004898655085976333,
|
|
"loss": 5.8899,
|
|
"mean_token_accuracy": 0.14553061574697496,
|
|
"num_tokens": 4060156.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.895521211624145,
|
|
"epoch": 1.785097682871422,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004897603664271413,
|
|
"loss": 5.7612,
|
|
"mean_token_accuracy": 0.15392354279756545,
|
|
"num_tokens": 4069670.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.820160341262818,
|
|
"epoch": 1.789641072239891,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000489654694356794,
|
|
"loss": 5.7833,
|
|
"mean_token_accuracy": 0.14432441145181657,
|
|
"num_tokens": 4080738.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.783408451080322,
|
|
"epoch": 1.79418446160836,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004895484926473265,
|
|
"loss": 5.7708,
|
|
"mean_token_accuracy": 0.15193571746349335,
|
|
"num_tokens": 4091122.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.979312753677368,
|
|
"epoch": 1.7987278509768287,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000489441761560781,
|
|
"loss": 5.8977,
|
|
"mean_token_accuracy": 0.14632806181907654,
|
|
"num_tokens": 4101055.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.851382875442505,
|
|
"epoch": 1.8032712403452975,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000489334501360506,
|
|
"loss": 5.8219,
|
|
"mean_token_accuracy": 0.15745959430933,
|
|
"num_tokens": 4110631.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.856267642974854,
|
|
"epoch": 1.8078146297137665,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004892267123111551,
|
|
"loss": 5.8472,
|
|
"mean_token_accuracy": 0.14736875146627426,
|
|
"num_tokens": 4121120.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.869805097579956,
|
|
"epoch": 1.8123580190822355,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004891183946786873,
|
|
"loss": 5.7925,
|
|
"mean_token_accuracy": 0.15268601775169371,
|
|
"num_tokens": 4132106.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.855297231674195,
|
|
"epoch": 1.8169014084507042,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004890095487303655,
|
|
"loss": 5.8248,
|
|
"mean_token_accuracy": 0.1471826359629631,
|
|
"num_tokens": 4142216.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.8169014084507042,
|
|
"eval_entropy": 5.837889374134153,
|
|
"eval_loss": 6.0509233474731445,
|
|
"eval_mean_token_accuracy": 0.14518940613366837,
|
|
"eval_num_tokens": 4142216.0,
|
|
"eval_runtime": 2.6012,
|
|
"eval_samples_per_second": 1320.568,
|
|
"eval_steps_per_second": 165.311,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.895492601394653,
|
|
"epoch": 1.821444797819173,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004889001747347562,
|
|
"loss": 5.8892,
|
|
"mean_token_accuracy": 0.147642320394516,
|
|
"num_tokens": 4152168.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.9000341415405275,
|
|
"epoch": 1.825988187187642,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000488790272961729,
|
|
"loss": 5.868,
|
|
"mean_token_accuracy": 0.14695144891738893,
|
|
"num_tokens": 4162085.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.931619119644165,
|
|
"epoch": 1.830531576556111,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004886798436824556,
|
|
"loss": 5.8809,
|
|
"mean_token_accuracy": 0.1446850225329399,
|
|
"num_tokens": 4171168.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.913224792480468,
|
|
"epoch": 1.8350749659245797,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004885688871694091,
|
|
"loss": 5.8345,
|
|
"mean_token_accuracy": 0.14913656413555146,
|
|
"num_tokens": 4181804.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.916212749481201,
|
|
"epoch": 1.8396183552930485,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004884574036963639,
|
|
"loss": 5.8917,
|
|
"mean_token_accuracy": 0.15118832141160965,
|
|
"num_tokens": 4192483.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.876117420196533,
|
|
"epoch": 1.8441617446615175,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004883453935383941,
|
|
"loss": 5.781,
|
|
"mean_token_accuracy": 0.15238289684057235,
|
|
"num_tokens": 4202214.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.908911275863647,
|
|
"epoch": 1.8487051340299865,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004882328569718738,
|
|
"loss": 5.8631,
|
|
"mean_token_accuracy": 0.15410126149654388,
|
|
"num_tokens": 4212440.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.819844532012939,
|
|
"epoch": 1.8532485233984552,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004881197942744758,
|
|
"loss": 5.8134,
|
|
"mean_token_accuracy": 0.15207200050354003,
|
|
"num_tokens": 4222780.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.854392099380493,
|
|
"epoch": 1.857791912766924,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004880062057251711,
|
|
"loss": 5.7624,
|
|
"mean_token_accuracy": 0.1560590997338295,
|
|
"num_tokens": 4232895.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.804136753082275,
|
|
"epoch": 1.862335302135393,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000487892091604228,
|
|
"loss": 5.8453,
|
|
"mean_token_accuracy": 0.1527414858341217,
|
|
"num_tokens": 4244287.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.900206899642944,
|
|
"epoch": 1.866878691503862,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048777745219321186,
|
|
"loss": 5.8138,
|
|
"mean_token_accuracy": 0.14360963478684424,
|
|
"num_tokens": 4254327.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.857224082946777,
|
|
"epoch": 1.8714220808723308,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048766228777498405,
|
|
"loss": 5.8308,
|
|
"mean_token_accuracy": 0.15000951290130615,
|
|
"num_tokens": 4264540.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.8516613960266115,
|
|
"epoch": 1.8759654702407995,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048754659863370125,
|
|
"loss": 5.8031,
|
|
"mean_token_accuracy": 0.14837189614772797,
|
|
"num_tokens": 4274946.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.839747142791748,
|
|
"epoch": 1.8805088596092685,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048743038505481503,
|
|
"loss": 5.7397,
|
|
"mean_token_accuracy": 0.15348523259162902,
|
|
"num_tokens": 4284395.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.937459135055542,
|
|
"epoch": 1.8850522489777375,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004873136473250708,
|
|
"loss": 5.852,
|
|
"mean_token_accuracy": 0.1507784515619278,
|
|
"num_tokens": 4294887.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.854568386077881,
|
|
"epoch": 1.8895956383462063,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048719638573250715,
|
|
"loss": 5.8508,
|
|
"mean_token_accuracy": 0.14535905569791793,
|
|
"num_tokens": 4304794.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.849844598770142,
|
|
"epoch": 1.894139027714675,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004870786005664555,
|
|
"loss": 5.7642,
|
|
"mean_token_accuracy": 0.1498329222202301,
|
|
"num_tokens": 4314810.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.864306831359864,
|
|
"epoch": 1.898682417083144,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004869602921175392,
|
|
"loss": 5.8218,
|
|
"mean_token_accuracy": 0.1472187742590904,
|
|
"num_tokens": 4325444.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.866009330749511,
|
|
"epoch": 1.903225806451613,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004868414606776724,
|
|
"loss": 5.8394,
|
|
"mean_token_accuracy": 0.15019661486148833,
|
|
"num_tokens": 4335620.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.87330436706543,
|
|
"epoch": 1.9077691958200818,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048672210654005996,
|
|
"loss": 5.8265,
|
|
"mean_token_accuracy": 0.1510421395301819,
|
|
"num_tokens": 4345006.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.823740673065186,
|
|
"epoch": 1.9123125851885505,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048660222999919616,
|
|
"loss": 5.742,
|
|
"mean_token_accuracy": 0.151262167096138,
|
|
"num_tokens": 4354805.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.876954221725464,
|
|
"epoch": 1.9168559745570195,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004864818313508647,
|
|
"loss": 5.8996,
|
|
"mean_token_accuracy": 0.14643090963363647,
|
|
"num_tokens": 4365881.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.9119120121002195,
|
|
"epoch": 1.9213993639254885,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004863609108921372,
|
|
"loss": 5.8801,
|
|
"mean_token_accuracy": 0.146506667137146,
|
|
"num_tokens": 4376368.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.90466079711914,
|
|
"epoch": 1.9259427532939573,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004862394689213728,
|
|
"loss": 5.8044,
|
|
"mean_token_accuracy": 0.15287439674139022,
|
|
"num_tokens": 4386695.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.881112480163575,
|
|
"epoch": 1.930486142662426,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004861175057382175,
|
|
"loss": 5.8068,
|
|
"mean_token_accuracy": 0.14728754609823227,
|
|
"num_tokens": 4396837.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.846132373809814,
|
|
"epoch": 1.935029532030895,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004859950216436034,
|
|
"loss": 5.7957,
|
|
"mean_token_accuracy": 0.15695781856775284,
|
|
"num_tokens": 4407310.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.851901197433472,
|
|
"epoch": 1.939572921399364,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004858720169397478,
|
|
"loss": 5.8284,
|
|
"mean_token_accuracy": 0.15003812313079834,
|
|
"num_tokens": 4417544.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.854311227798462,
|
|
"epoch": 1.9441163107678328,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004857484919301525,
|
|
"loss": 5.764,
|
|
"mean_token_accuracy": 0.15185186862945557,
|
|
"num_tokens": 4428228.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.822315454483032,
|
|
"epoch": 1.9486597001363015,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004856244469196034,
|
|
"loss": 5.7732,
|
|
"mean_token_accuracy": 0.1535707250237465,
|
|
"num_tokens": 4439088.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.876076984405517,
|
|
"epoch": 1.9532030895047705,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000485499882214169,
|
|
"loss": 5.8997,
|
|
"mean_token_accuracy": 0.1445956900715828,
|
|
"num_tokens": 4449715.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.836854267120361,
|
|
"epoch": 1.9577464788732395,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004853747981212003,
|
|
"loss": 5.7448,
|
|
"mean_token_accuracy": 0.15456032007932663,
|
|
"num_tokens": 4460233.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.791164922714233,
|
|
"epoch": 1.9622898682417085,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004852491949493301,
|
|
"loss": 5.6796,
|
|
"mean_token_accuracy": 0.15854840278625487,
|
|
"num_tokens": 4471401.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.759272289276123,
|
|
"epoch": 1.9668332576101772,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004851230730084716,
|
|
"loss": 5.7466,
|
|
"mean_token_accuracy": 0.16108963042497634,
|
|
"num_tokens": 4481185.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.874443531036377,
|
|
"epoch": 1.971376646978646,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048499643260981816,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.1526338279247284,
|
|
"num_tokens": 4492342.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.89804744720459,
|
|
"epoch": 1.975920036347115,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004848692740658424,
|
|
"loss": 5.8617,
|
|
"mean_token_accuracy": 0.15097771733999252,
|
|
"num_tokens": 4504277.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.749866342544555,
|
|
"epoch": 1.980463425715584,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004847415976902954,
|
|
"loss": 5.8212,
|
|
"mean_token_accuracy": 0.15068218410015105,
|
|
"num_tokens": 4514178.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.8503865718841555,
|
|
"epoch": 1.9850068150840527,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048461340379820587,
|
|
"loss": 5.8005,
|
|
"mean_token_accuracy": 0.15338014960289,
|
|
"num_tokens": 4524559.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.838710737228394,
|
|
"epoch": 1.9895502044525215,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048448469270587945,
|
|
"loss": 5.75,
|
|
"mean_token_accuracy": 0.15777102559804917,
|
|
"num_tokens": 4535168.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.845028257369995,
|
|
"epoch": 1.9940935938209905,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048435546473089814,
|
|
"loss": 5.8299,
|
|
"mean_token_accuracy": 0.1515686810016632,
|
|
"num_tokens": 4546105.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.907456493377685,
|
|
"epoch": 1.9986369831894595,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048422572019211896,
|
|
"loss": 5.7757,
|
|
"mean_token_accuracy": 0.15573390573263168,
|
|
"num_tokens": 4556498.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.662998570336236,
|
|
"epoch": 2.0027260336210815,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048409545940967375,
|
|
"loss": 5.5679,
|
|
"mean_token_accuracy": 0.15117573903666603,
|
|
"num_tokens": 4565029.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.9106169700622555,
|
|
"epoch": 2.00726942298955,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000483964682704968,
|
|
"loss": 5.5324,
|
|
"mean_token_accuracy": 0.15834336578845978,
|
|
"num_tokens": 4575232.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.754504060745239,
|
|
"epoch": 2.011812812358019,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048383339040068024,
|
|
"loss": 5.5059,
|
|
"mean_token_accuracy": 0.15715845078229904,
|
|
"num_tokens": 4585818.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.704641056060791,
|
|
"epoch": 2.016356201726488,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004837015828207612,
|
|
"loss": 5.4472,
|
|
"mean_token_accuracy": 0.1678731009364128,
|
|
"num_tokens": 4595823.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.736291837692261,
|
|
"epoch": 2.020899591094957,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004835692602904329,
|
|
"loss": 5.4278,
|
|
"mean_token_accuracy": 0.16101882308721543,
|
|
"num_tokens": 4606537.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.808406639099121,
|
|
"epoch": 2.0254429804634255,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004834364231361882,
|
|
"loss": 5.4763,
|
|
"mean_token_accuracy": 0.1621137738227844,
|
|
"num_tokens": 4616096.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.736590147018433,
|
|
"epoch": 2.0299863698318945,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004833030716857894,
|
|
"loss": 5.4994,
|
|
"mean_token_accuracy": 0.1609218567609787,
|
|
"num_tokens": 4626287.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.7770387649536135,
|
|
"epoch": 2.0345297592003635,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048316920626826807,
|
|
"loss": 5.5033,
|
|
"mean_token_accuracy": 0.16166141033172607,
|
|
"num_tokens": 4636898.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.712182950973511,
|
|
"epoch": 2.0390731485688325,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048303482721392363,
|
|
"loss": 5.4521,
|
|
"mean_token_accuracy": 0.16646481603384017,
|
|
"num_tokens": 4646263.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.791868829727173,
|
|
"epoch": 2.043616537937301,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048289993485432325,
|
|
"loss": 5.4946,
|
|
"mean_token_accuracy": 0.16336590498685838,
|
|
"num_tokens": 4656463.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.70614013671875,
|
|
"epoch": 2.04815992730577,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048276452952230036,
|
|
"loss": 5.5244,
|
|
"mean_token_accuracy": 0.16146081537008286,
|
|
"num_tokens": 4666824.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.673251724243164,
|
|
"epoch": 2.052703316674239,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048262861155195407,
|
|
"loss": 5.4108,
|
|
"mean_token_accuracy": 0.16472615003585817,
|
|
"num_tokens": 4676793.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.780643081665039,
|
|
"epoch": 2.057246706042708,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004824921812786486,
|
|
"loss": 5.5636,
|
|
"mean_token_accuracy": 0.1657853752374649,
|
|
"num_tokens": 4687124.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.774113988876342,
|
|
"epoch": 2.0617900954111765,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000482355239039012,
|
|
"loss": 5.7144,
|
|
"mean_token_accuracy": 0.15666003823280333,
|
|
"num_tokens": 4698074.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.819989347457886,
|
|
"epoch": 2.0663334847796455,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004822177851709356,
|
|
"loss": 5.5795,
|
|
"mean_token_accuracy": 0.1618112653493881,
|
|
"num_tokens": 4709177.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.782473134994507,
|
|
"epoch": 2.0708768741481145,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048207982001357327,
|
|
"loss": 5.5005,
|
|
"mean_token_accuracy": 0.15604408085346222,
|
|
"num_tokens": 4719649.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.762876033782959,
|
|
"epoch": 2.0754202635165835,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004819413439073402,
|
|
"loss": 5.5714,
|
|
"mean_token_accuracy": 0.15874197632074355,
|
|
"num_tokens": 4731685.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.70639214515686,
|
|
"epoch": 2.079963652885052,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004818023571939125,
|
|
"loss": 5.5115,
|
|
"mean_token_accuracy": 0.163041552901268,
|
|
"num_tokens": 4742854.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.732694911956787,
|
|
"epoch": 2.084507042253521,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000481662860216226,
|
|
"loss": 5.4915,
|
|
"mean_token_accuracy": 0.1624978080391884,
|
|
"num_tokens": 4754088.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.785396528244019,
|
|
"epoch": 2.08905043162199,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048152285331847574,
|
|
"loss": 5.5248,
|
|
"mean_token_accuracy": 0.1613600268959999,
|
|
"num_tokens": 4764090.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.633577919006347,
|
|
"epoch": 2.093593820990459,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004813823368461148,
|
|
"loss": 5.5049,
|
|
"mean_token_accuracy": 0.1666693389415741,
|
|
"num_tokens": 4773704.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.793992424011231,
|
|
"epoch": 2.098137210358928,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004812413111458534,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.16529950201511384,
|
|
"num_tokens": 4783678.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.665829753875732,
|
|
"epoch": 2.1026805997273965,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004810997765656587,
|
|
"loss": 5.4364,
|
|
"mean_token_accuracy": 0.16927295178174973,
|
|
"num_tokens": 4794183.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.650082588195801,
|
|
"epoch": 2.1072239890958655,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048095773345475294,
|
|
"loss": 5.4112,
|
|
"mean_token_accuracy": 0.1630910024046898,
|
|
"num_tokens": 4803670.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.67885012626648,
|
|
"epoch": 2.1117673784643345,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048081518216361367,
|
|
"loss": 5.4301,
|
|
"mean_token_accuracy": 0.16907580941915512,
|
|
"num_tokens": 4813611.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.687435245513916,
|
|
"epoch": 2.1163107678328035,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004806721230439719,
|
|
"loss": 5.4996,
|
|
"mean_token_accuracy": 0.155875563621521,
|
|
"num_tokens": 4824356.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.800898027420044,
|
|
"epoch": 2.120854157201272,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004805285564488117,
|
|
"loss": 5.5454,
|
|
"mean_token_accuracy": 0.16431267261505128,
|
|
"num_tokens": 4834410.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.618631792068482,
|
|
"epoch": 2.125397546569741,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004803844827323696,
|
|
"loss": 5.3835,
|
|
"mean_token_accuracy": 0.16828938573598862,
|
|
"num_tokens": 4844975.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.658405733108521,
|
|
"epoch": 2.12994093593821,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004802399022501329,
|
|
"loss": 5.5071,
|
|
"mean_token_accuracy": 0.16457533091306686,
|
|
"num_tokens": 4856419.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.652211427688599,
|
|
"epoch": 2.134484325306679,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004800948153588397,
|
|
"loss": 5.4191,
|
|
"mean_token_accuracy": 0.1696417048573494,
|
|
"num_tokens": 4867115.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.60864143371582,
|
|
"epoch": 2.1390277146751475,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047994922241647746,
|
|
"loss": 5.5083,
|
|
"mean_token_accuracy": 0.16505912691354752,
|
|
"num_tokens": 4876675.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.731504678726196,
|
|
"epoch": 2.1435711040436165,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047980312378228244,
|
|
"loss": 5.5283,
|
|
"mean_token_accuracy": 0.16072432398796083,
|
|
"num_tokens": 4887868.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.667622852325439,
|
|
"epoch": 2.1481144934120855,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047965651981673823,
|
|
"loss": 5.5019,
|
|
"mean_token_accuracy": 0.1720751479268074,
|
|
"num_tokens": 4896878.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.687494850158691,
|
|
"epoch": 2.1526578827805545,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047950941088157576,
|
|
"loss": 5.4555,
|
|
"mean_token_accuracy": 0.16290661543607712,
|
|
"num_tokens": 4906683.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.657924699783325,
|
|
"epoch": 2.157201272149023,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004793617973397715,
|
|
"loss": 5.4584,
|
|
"mean_token_accuracy": 0.17150721251964568,
|
|
"num_tokens": 4917184.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.65337290763855,
|
|
"epoch": 2.161744661517492,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047921367955554745,
|
|
"loss": 5.479,
|
|
"mean_token_accuracy": 0.16076952517032622,
|
|
"num_tokens": 4927337.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.713319635391235,
|
|
"epoch": 2.166288050885961,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004790650578943693,
|
|
"loss": 5.5032,
|
|
"mean_token_accuracy": 0.16063827648758888,
|
|
"num_tokens": 4937661.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.715535974502563,
|
|
"epoch": 2.17083144025443,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047891593272294636,
|
|
"loss": 5.537,
|
|
"mean_token_accuracy": 0.15505636632442474,
|
|
"num_tokens": 4948068.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.670109033584595,
|
|
"epoch": 2.1753748296228985,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047876630440923,
|
|
"loss": 5.4769,
|
|
"mean_token_accuracy": 0.16679450422525405,
|
|
"num_tokens": 4958703.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.609113645553589,
|
|
"epoch": 2.1799182189913675,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004786161733224134,
|
|
"loss": 5.4898,
|
|
"mean_token_accuracy": 0.16843014657497407,
|
|
"num_tokens": 4969190.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.687240028381348,
|
|
"epoch": 2.1844616083598365,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047846553983293006,
|
|
"loss": 5.3759,
|
|
"mean_token_accuracy": 0.171466401219368,
|
|
"num_tokens": 4978663.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.73572187423706,
|
|
"epoch": 2.1890049977283055,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047831440431245307,
|
|
"loss": 5.5569,
|
|
"mean_token_accuracy": 0.16112247109413147,
|
|
"num_tokens": 4989839.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.587356090545654,
|
|
"epoch": 2.193548387096774,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047816276713389424,
|
|
"loss": 5.3899,
|
|
"mean_token_accuracy": 0.16882797181606293,
|
|
"num_tokens": 5000030.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.696376991271973,
|
|
"epoch": 2.198091776465243,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004780106286714034,
|
|
"loss": 5.5726,
|
|
"mean_token_accuracy": 0.16172168999910355,
|
|
"num_tokens": 5010229.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.749238204956055,
|
|
"epoch": 2.202635165833712,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000477857989300367,
|
|
"loss": 5.4941,
|
|
"mean_token_accuracy": 0.16530462354421616,
|
|
"num_tokens": 5021000.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.611483812332153,
|
|
"epoch": 2.207178555202181,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047770484939740757,
|
|
"loss": 5.3904,
|
|
"mean_token_accuracy": 0.1689944565296173,
|
|
"num_tokens": 5031319.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.611922311782837,
|
|
"epoch": 2.2117219445706495,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004775512093403824,
|
|
"loss": 5.444,
|
|
"mean_token_accuracy": 0.16835010796785355,
|
|
"num_tokens": 5041892.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.623884820938111,
|
|
"epoch": 2.2162653339391185,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047739706950838327,
|
|
"loss": 5.5131,
|
|
"mean_token_accuracy": 0.16812876760959625,
|
|
"num_tokens": 5052363.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.776235961914063,
|
|
"epoch": 2.2208087233075875,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047724243028173484,
|
|
"loss": 5.5431,
|
|
"mean_token_accuracy": 0.15931915789842604,
|
|
"num_tokens": 5063342.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.621704530715943,
|
|
"epoch": 2.2253521126760565,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047708729204199406,
|
|
"loss": 5.4533,
|
|
"mean_token_accuracy": 0.15909151136875152,
|
|
"num_tokens": 5073393.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.606422805786133,
|
|
"epoch": 2.229895502044525,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000476931655171949,
|
|
"loss": 5.4175,
|
|
"mean_token_accuracy": 0.17029767036437987,
|
|
"num_tokens": 5082417.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.781980419158936,
|
|
"epoch": 2.234438891412994,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004767755200556183,
|
|
"loss": 5.5759,
|
|
"mean_token_accuracy": 0.15932134985923768,
|
|
"num_tokens": 5093383.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.61558165550232,
|
|
"epoch": 2.238982280781463,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004766188870782497,
|
|
"loss": 5.3894,
|
|
"mean_token_accuracy": 0.17075251340866088,
|
|
"num_tokens": 5104346.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.590832090377807,
|
|
"epoch": 2.243525670149932,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047646175662631967,
|
|
"loss": 5.5415,
|
|
"mean_token_accuracy": 0.1631665825843811,
|
|
"num_tokens": 5114375.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.737929391860962,
|
|
"epoch": 2.2480690595184005,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047630412908753183,
|
|
"loss": 5.5227,
|
|
"mean_token_accuracy": 0.16462021470069885,
|
|
"num_tokens": 5125035.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.671669578552246,
|
|
"epoch": 2.2526124488868695,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047614600485081657,
|
|
"loss": 5.5414,
|
|
"mean_token_accuracy": 0.15197518169879914,
|
|
"num_tokens": 5135869.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.737759590148926,
|
|
"epoch": 2.2571558382553385,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047598738430632963,
|
|
"loss": 5.5018,
|
|
"mean_token_accuracy": 0.16437771171331406,
|
|
"num_tokens": 5146615.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.535083913803101,
|
|
"epoch": 2.2616992276238075,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047582826784545146,
|
|
"loss": 5.3978,
|
|
"mean_token_accuracy": 0.16925125420093537,
|
|
"num_tokens": 5157490.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.6835729598999025,
|
|
"epoch": 2.266242616992276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047566865586078615,
|
|
"loss": 5.524,
|
|
"mean_token_accuracy": 0.15676333606243134,
|
|
"num_tokens": 5168693.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.7088587284088135,
|
|
"epoch": 2.270786006360745,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047550854874616047,
|
|
"loss": 5.4961,
|
|
"mean_token_accuracy": 0.1648841142654419,
|
|
"num_tokens": 5179232.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.270786006360745,
|
|
"eval_entropy": 5.442809280129366,
|
|
"eval_loss": 5.97740364074707,
|
|
"eval_mean_token_accuracy": 0.15040350478402403,
|
|
"eval_num_tokens": 5179232.0,
|
|
"eval_runtime": 2.5897,
|
|
"eval_samples_per_second": 1326.387,
|
|
"eval_steps_per_second": 166.04,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.637321901321411,
|
|
"epoch": 2.275329395729214,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047534794689662257,
|
|
"loss": 5.5455,
|
|
"mean_token_accuracy": 0.1695609301328659,
|
|
"num_tokens": 5188613.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.586072683334351,
|
|
"epoch": 2.279872785097683,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004751868507084418,
|
|
"loss": 5.427,
|
|
"mean_token_accuracy": 0.1748349979519844,
|
|
"num_tokens": 5198441.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.612478017807007,
|
|
"epoch": 2.284416174466152,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004750252605791068,
|
|
"loss": 5.5062,
|
|
"mean_token_accuracy": 0.16696918308734893,
|
|
"num_tokens": 5208466.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.606588888168335,
|
|
"epoch": 2.2889595638346205,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004748631769073254,
|
|
"loss": 5.4763,
|
|
"mean_token_accuracy": 0.1626569002866745,
|
|
"num_tokens": 5218847.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.688857889175415,
|
|
"epoch": 2.2935029532030895,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004747006000930227,
|
|
"loss": 5.5083,
|
|
"mean_token_accuracy": 0.15848715156316756,
|
|
"num_tokens": 5228870.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.752295398712159,
|
|
"epoch": 2.2980463425715585,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004745375305373409,
|
|
"loss": 5.4704,
|
|
"mean_token_accuracy": 0.16673778593540192,
|
|
"num_tokens": 5238186.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.567574834823608,
|
|
"epoch": 2.302589731940027,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047437396864263804,
|
|
"loss": 5.5189,
|
|
"mean_token_accuracy": 0.16429466456174852,
|
|
"num_tokens": 5247478.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.7289081573486325,
|
|
"epoch": 2.307133121308496,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004742099148124866,
|
|
"loss": 5.4394,
|
|
"mean_token_accuracy": 0.16676439940929413,
|
|
"num_tokens": 5258519.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.672767400741577,
|
|
"epoch": 2.311676510676965,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004740453694516734,
|
|
"loss": 5.4892,
|
|
"mean_token_accuracy": 0.16587547361850738,
|
|
"num_tokens": 5268218.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.544440603256225,
|
|
"epoch": 2.316219900045434,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047388033296619746,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.16287526488304138,
|
|
"num_tokens": 5277926.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.615134143829346,
|
|
"epoch": 2.320763289413903,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004737148057632701,
|
|
"loss": 5.4785,
|
|
"mean_token_accuracy": 0.16812876164913176,
|
|
"num_tokens": 5287735.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.57270884513855,
|
|
"epoch": 2.3253066787823715,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004735487882513131,
|
|
"loss": 5.4068,
|
|
"mean_token_accuracy": 0.17444850504398346,
|
|
"num_tokens": 5297120.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.565961790084839,
|
|
"epoch": 2.3298500681508405,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004733822808399583,
|
|
"loss": 5.4264,
|
|
"mean_token_accuracy": 0.17079222202301025,
|
|
"num_tokens": 5307448.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.6533262729644775,
|
|
"epoch": 2.3343934575193095,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047321528394004615,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.16939034909009934,
|
|
"num_tokens": 5317914.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.6184889793396,
|
|
"epoch": 2.3389368468877785,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047304779796362485,
|
|
"loss": 5.4275,
|
|
"mean_token_accuracy": 0.16940037310123443,
|
|
"num_tokens": 5328471.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.551731777191162,
|
|
"epoch": 2.343480236256247,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047287982332394943,
|
|
"loss": 5.4419,
|
|
"mean_token_accuracy": 0.1600135535001755,
|
|
"num_tokens": 5339118.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.649324369430542,
|
|
"epoch": 2.348023625624716,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047271136043548057,
|
|
"loss": 5.5506,
|
|
"mean_token_accuracy": 0.1595440998673439,
|
|
"num_tokens": 5350183.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.637921380996704,
|
|
"epoch": 2.352567014993185,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004725424097138838,
|
|
"loss": 5.4716,
|
|
"mean_token_accuracy": 0.16334169656038283,
|
|
"num_tokens": 5361317.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.61320652961731,
|
|
"epoch": 2.357110404361654,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047237297157602816,
|
|
"loss": 5.4796,
|
|
"mean_token_accuracy": 0.1689859852194786,
|
|
"num_tokens": 5371338.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.735874176025391,
|
|
"epoch": 2.3616537937301225,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004722030464399854,
|
|
"loss": 5.5966,
|
|
"mean_token_accuracy": 0.15847954750061036,
|
|
"num_tokens": 5382043.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.696183872222901,
|
|
"epoch": 2.3661971830985915,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000472032634725029,
|
|
"loss": 5.4905,
|
|
"mean_token_accuracy": 0.16926760077476502,
|
|
"num_tokens": 5392547.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.486810922622681,
|
|
"epoch": 2.3707405724670605,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047186173685163273,
|
|
"loss": 5.3763,
|
|
"mean_token_accuracy": 0.1735202193260193,
|
|
"num_tokens": 5402455.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.590288305282593,
|
|
"epoch": 2.3752839618355295,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047169035324147023,
|
|
"loss": 5.5437,
|
|
"mean_token_accuracy": 0.16168220490217208,
|
|
"num_tokens": 5413925.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.714783763885498,
|
|
"epoch": 2.379827351203998,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047151848431741363,
|
|
"loss": 5.5642,
|
|
"mean_token_accuracy": 0.15730200707912445,
|
|
"num_tokens": 5425344.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.611330604553222,
|
|
"epoch": 2.384370740572467,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004713461305035322,
|
|
"loss": 5.5096,
|
|
"mean_token_accuracy": 0.16345102787017823,
|
|
"num_tokens": 5435197.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.728695440292358,
|
|
"epoch": 2.388914129940936,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000471173292225092,
|
|
"loss": 5.4892,
|
|
"mean_token_accuracy": 0.16780440211296083,
|
|
"num_tokens": 5445683.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.5622600555419925,
|
|
"epoch": 2.393457519309405,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047099996990855424,
|
|
"loss": 5.3659,
|
|
"mean_token_accuracy": 0.1666329324245453,
|
|
"num_tokens": 5456846.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.600703525543213,
|
|
"epoch": 2.3980009086778735,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047082616398157456,
|
|
"loss": 5.4361,
|
|
"mean_token_accuracy": 0.1714688628911972,
|
|
"num_tokens": 5467701.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.630035829544068,
|
|
"epoch": 2.4025442980463425,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004706518748730017,
|
|
"loss": 5.4687,
|
|
"mean_token_accuracy": 0.16582391560077667,
|
|
"num_tokens": 5478246.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.619263315200806,
|
|
"epoch": 2.4070876874148115,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004704771030128768,
|
|
"loss": 5.4702,
|
|
"mean_token_accuracy": 0.16932938992977142,
|
|
"num_tokens": 5488107.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.654132986068726,
|
|
"epoch": 2.4116310767832805,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00047030184883243203,
|
|
"loss": 5.472,
|
|
"mean_token_accuracy": 0.1618760645389557,
|
|
"num_tokens": 5498521.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.59398832321167,
|
|
"epoch": 2.416174466151749,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047012611276408964,
|
|
"loss": 5.5127,
|
|
"mean_token_accuracy": 0.1568394958972931,
|
|
"num_tokens": 5509450.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.671771097183227,
|
|
"epoch": 2.420717855520218,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004699498952414609,
|
|
"loss": 5.4232,
|
|
"mean_token_accuracy": 0.17102673798799514,
|
|
"num_tokens": 5519662.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.523070812225342,
|
|
"epoch": 2.425261244888687,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046977319669934506,
|
|
"loss": 5.4078,
|
|
"mean_token_accuracy": 0.16589158028364182,
|
|
"num_tokens": 5529499.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.589613771438598,
|
|
"epoch": 2.429804634257156,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046959601757372814,
|
|
"loss": 5.5843,
|
|
"mean_token_accuracy": 0.15429533272981644,
|
|
"num_tokens": 5540938.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.609573173522949,
|
|
"epoch": 2.4343480236256245,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00046941835830178216,
|
|
"loss": 5.4384,
|
|
"mean_token_accuracy": 0.1675647258758545,
|
|
"num_tokens": 5552120.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.555629444122315,
|
|
"epoch": 2.4388914129940935,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046924021932186357,
|
|
"loss": 5.4176,
|
|
"mean_token_accuracy": 0.16815963238477707,
|
|
"num_tokens": 5561590.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.562342023849487,
|
|
"epoch": 2.4434348023625625,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046906160107351263,
|
|
"loss": 5.4005,
|
|
"mean_token_accuracy": 0.1693267121911049,
|
|
"num_tokens": 5571682.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.618103313446045,
|
|
"epoch": 2.4479781917310315,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046888250399745217,
|
|
"loss": 5.4587,
|
|
"mean_token_accuracy": 0.17296975404024123,
|
|
"num_tokens": 5581299.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.609625291824341,
|
|
"epoch": 2.4525215810995,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046870292853558646,
|
|
"loss": 5.3986,
|
|
"mean_token_accuracy": 0.17211423963308334,
|
|
"num_tokens": 5590766.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.584544038772583,
|
|
"epoch": 2.457064970467969,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046852287513099997,
|
|
"loss": 5.5055,
|
|
"mean_token_accuracy": 0.16718498468399048,
|
|
"num_tokens": 5600898.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.661745214462281,
|
|
"epoch": 2.461608359836438,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004683423442279566,
|
|
"loss": 5.4841,
|
|
"mean_token_accuracy": 0.16790943294763566,
|
|
"num_tokens": 5612052.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.639920616149903,
|
|
"epoch": 2.466151749204907,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004681613362718985,
|
|
"loss": 5.5635,
|
|
"mean_token_accuracy": 0.171800135076046,
|
|
"num_tokens": 5622847.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.5262782096862795,
|
|
"epoch": 2.470695138573376,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004679798517094447,
|
|
"loss": 5.3964,
|
|
"mean_token_accuracy": 0.17257711589336394,
|
|
"num_tokens": 5633134.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.655404424667358,
|
|
"epoch": 2.4752385279418445,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046779789098839043,
|
|
"loss": 5.5613,
|
|
"mean_token_accuracy": 0.1665353626012802,
|
|
"num_tokens": 5644223.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.557641506195068,
|
|
"epoch": 2.4797819173103135,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046761545455770555,
|
|
"loss": 5.4872,
|
|
"mean_token_accuracy": 0.1618534117937088,
|
|
"num_tokens": 5653620.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.666029930114746,
|
|
"epoch": 2.4843253066787825,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004674325428675339,
|
|
"loss": 5.4051,
|
|
"mean_token_accuracy": 0.17211906909942626,
|
|
"num_tokens": 5663800.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.5644697666168215,
|
|
"epoch": 2.488868696047251,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004672491563691919,
|
|
"loss": 5.4688,
|
|
"mean_token_accuracy": 0.17142953872680664,
|
|
"num_tokens": 5674606.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.631884145736694,
|
|
"epoch": 2.49341208541572,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046706529551516753,
|
|
"loss": 5.5675,
|
|
"mean_token_accuracy": 0.16467656344175338,
|
|
"num_tokens": 5684975.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.620426988601684,
|
|
"epoch": 2.497955474784189,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00046688096075911905,
|
|
"loss": 5.4161,
|
|
"mean_token_accuracy": 0.16923390179872513,
|
|
"num_tokens": 5695336.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.6551628589630125,
|
|
"epoch": 2.502498864152658,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004666961525558743,
|
|
"loss": 5.4471,
|
|
"mean_token_accuracy": 0.17089568674564362,
|
|
"num_tokens": 5706027.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.592363452911377,
|
|
"epoch": 2.507042253521127,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000466510871361429,
|
|
"loss": 5.4779,
|
|
"mean_token_accuracy": 0.1645099937915802,
|
|
"num_tokens": 5716026.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.592696189880371,
|
|
"epoch": 2.5115856428895955,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046632511763294615,
|
|
"loss": 5.4679,
|
|
"mean_token_accuracy": 0.1719784304499626,
|
|
"num_tokens": 5726055.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.672763252258301,
|
|
"epoch": 2.5161290322580645,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004661388918287546,
|
|
"loss": 5.5295,
|
|
"mean_token_accuracy": 0.16350285410881044,
|
|
"num_tokens": 5736562.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.671252298355102,
|
|
"epoch": 2.5206724216265335,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004659521944083481,
|
|
"loss": 5.5616,
|
|
"mean_token_accuracy": 0.1590777099132538,
|
|
"num_tokens": 5747297.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.650782060623169,
|
|
"epoch": 2.525215810995002,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004657650258323838,
|
|
"loss": 5.5125,
|
|
"mean_token_accuracy": 0.16581853926181794,
|
|
"num_tokens": 5756801.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.615206527709961,
|
|
"epoch": 2.529759200363471,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046557738656268166,
|
|
"loss": 5.5188,
|
|
"mean_token_accuracy": 0.16645000576972963,
|
|
"num_tokens": 5767493.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.599960851669311,
|
|
"epoch": 2.53430258973194,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000465389277062223,
|
|
"loss": 5.4055,
|
|
"mean_token_accuracy": 0.1837794825434685,
|
|
"num_tokens": 5777539.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.599096584320068,
|
|
"epoch": 2.538845979100409,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046520069779514917,
|
|
"loss": 5.4309,
|
|
"mean_token_accuracy": 0.17846813946962356,
|
|
"num_tokens": 5787202.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.586429595947266,
|
|
"epoch": 2.543389368468878,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004650116492267609,
|
|
"loss": 5.4795,
|
|
"mean_token_accuracy": 0.16962323635816573,
|
|
"num_tokens": 5798023.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.52474684715271,
|
|
"epoch": 2.5479327578373465,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004648221318235165,
|
|
"loss": 5.3787,
|
|
"mean_token_accuracy": 0.1730539843440056,
|
|
"num_tokens": 5807711.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.605598783493042,
|
|
"epoch": 2.5524761472058155,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004646321460530316,
|
|
"loss": 5.3983,
|
|
"mean_token_accuracy": 0.17484969943761824,
|
|
"num_tokens": 5817053.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.511811065673828,
|
|
"epoch": 2.5570195365742845,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000464441692384077,
|
|
"loss": 5.4836,
|
|
"mean_token_accuracy": 0.1696155324578285,
|
|
"num_tokens": 5827792.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.5640058517456055,
|
|
"epoch": 2.561562925942753,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004642507712865784,
|
|
"loss": 5.3951,
|
|
"mean_token_accuracy": 0.17243893146514894,
|
|
"num_tokens": 5837702.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.640513277053833,
|
|
"epoch": 2.566106315311222,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046405938323161453,
|
|
"loss": 5.5285,
|
|
"mean_token_accuracy": 0.16529992073774338,
|
|
"num_tokens": 5847973.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.639587306976319,
|
|
"epoch": 2.570649704679691,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046386752869141624,
|
|
"loss": 5.5266,
|
|
"mean_token_accuracy": 0.1601054772734642,
|
|
"num_tokens": 5857591.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.619012546539307,
|
|
"epoch": 2.57519309404816,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004636752081393658,
|
|
"loss": 5.4852,
|
|
"mean_token_accuracy": 0.16269528716802598,
|
|
"num_tokens": 5868616.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.64703950881958,
|
|
"epoch": 2.579736483416629,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046348242204999487,
|
|
"loss": 5.4101,
|
|
"mean_token_accuracy": 0.17154187560081482,
|
|
"num_tokens": 5878526.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.537039709091187,
|
|
"epoch": 2.5842798727850975,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046328917089898396,
|
|
"loss": 5.4367,
|
|
"mean_token_accuracy": 0.17496657520532607,
|
|
"num_tokens": 5887844.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.63788480758667,
|
|
"epoch": 2.5888232621535665,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004630954551631612,
|
|
"loss": 5.5269,
|
|
"mean_token_accuracy": 0.16755133047699927,
|
|
"num_tokens": 5898523.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.636503982543945,
|
|
"epoch": 2.5933666515220355,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046290127532050087,
|
|
"loss": 5.5259,
|
|
"mean_token_accuracy": 0.16437408030033113,
|
|
"num_tokens": 5909671.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.5825988292694095,
|
|
"epoch": 2.5979100408905045,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004627066318501223,
|
|
"loss": 5.4603,
|
|
"mean_token_accuracy": 0.1605490505695343,
|
|
"num_tokens": 5920451.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.564822149276734,
|
|
"epoch": 2.602453430258973,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046251152523228906,
|
|
"loss": 5.449,
|
|
"mean_token_accuracy": 0.1752915248274803,
|
|
"num_tokens": 5931218.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.493123292922974,
|
|
"epoch": 2.606996819627442,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004623159559484074,
|
|
"loss": 5.4567,
|
|
"mean_token_accuracy": 0.16683945208787918,
|
|
"num_tokens": 5941380.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.650035810470581,
|
|
"epoch": 2.611540208995911,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004621199244810248,
|
|
"loss": 5.4091,
|
|
"mean_token_accuracy": 0.1655109629034996,
|
|
"num_tokens": 5951989.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.537979030609131,
|
|
"epoch": 2.61608359836438,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046192343131382964,
|
|
"loss": 5.4806,
|
|
"mean_token_accuracy": 0.1636374205350876,
|
|
"num_tokens": 5961982.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.626474237442016,
|
|
"epoch": 2.620626987732849,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046172647693164926,
|
|
"loss": 5.5172,
|
|
"mean_token_accuracy": 0.16451143324375153,
|
|
"num_tokens": 5972184.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.645173501968384,
|
|
"epoch": 2.6251703771013175,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004615290618204489,
|
|
"loss": 5.4413,
|
|
"mean_token_accuracy": 0.1672982707619667,
|
|
"num_tokens": 5981912.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.587864542007447,
|
|
"epoch": 2.6297137664697865,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004613311864673309,
|
|
"loss": 5.4216,
|
|
"mean_token_accuracy": 0.16713207811117173,
|
|
"num_tokens": 5992460.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.602197599411011,
|
|
"epoch": 2.6342571558382555,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046113285136053286,
|
|
"loss": 5.4841,
|
|
"mean_token_accuracy": 0.16885117888450624,
|
|
"num_tokens": 6002170.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.539362096786499,
|
|
"epoch": 2.638800545206724,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000460934056989427,
|
|
"loss": 5.4002,
|
|
"mean_token_accuracy": 0.16570922881364822,
|
|
"num_tokens": 6012240.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.582073974609375,
|
|
"epoch": 2.643343934575193,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004607348038445187,
|
|
"loss": 5.3953,
|
|
"mean_token_accuracy": 0.1713549479842186,
|
|
"num_tokens": 6022289.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.645311164855957,
|
|
"epoch": 2.647887323943662,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046053509241744526,
|
|
"loss": 5.4783,
|
|
"mean_token_accuracy": 0.16424850523471832,
|
|
"num_tokens": 6032268.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.665022563934326,
|
|
"epoch": 2.652430713312131,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046033492320097483,
|
|
"loss": 5.5584,
|
|
"mean_token_accuracy": 0.1588386908173561,
|
|
"num_tokens": 6042574.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.699222660064697,
|
|
"epoch": 2.6569741026806,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004601342966890049,
|
|
"loss": 5.5385,
|
|
"mean_token_accuracy": 0.1621340826153755,
|
|
"num_tokens": 6052508.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.465180206298828,
|
|
"epoch": 2.6615174920490685,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045993321337656163,
|
|
"loss": 5.3873,
|
|
"mean_token_accuracy": 0.17350232750177383,
|
|
"num_tokens": 6062541.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.58162899017334,
|
|
"epoch": 2.6660608814175375,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000459731673759798,
|
|
"loss": 5.5411,
|
|
"mean_token_accuracy": 0.16393526643514633,
|
|
"num_tokens": 6072532.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.687358331680298,
|
|
"epoch": 2.6706042707860065,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045952967833599303,
|
|
"loss": 5.4213,
|
|
"mean_token_accuracy": 0.17391514033079147,
|
|
"num_tokens": 6082518.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.56571774482727,
|
|
"epoch": 2.675147660154475,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004593272276035502,
|
|
"loss": 5.4315,
|
|
"mean_token_accuracy": 0.1704488754272461,
|
|
"num_tokens": 6092356.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.622656726837159,
|
|
"epoch": 2.679691049522944,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004591243220619969,
|
|
"loss": 5.5268,
|
|
"mean_token_accuracy": 0.17083123177289963,
|
|
"num_tokens": 6103450.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.6321862697601315,
|
|
"epoch": 2.684234438891413,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000458920962211982,
|
|
"loss": 5.4077,
|
|
"mean_token_accuracy": 0.1748711422085762,
|
|
"num_tokens": 6113708.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.523899412155151,
|
|
"epoch": 2.688777828259882,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004587171485552759,
|
|
"loss": 5.4863,
|
|
"mean_token_accuracy": 0.16334034949541093,
|
|
"num_tokens": 6124508.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.657847929000854,
|
|
"epoch": 2.693321217628351,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045851288159476854,
|
|
"loss": 5.4578,
|
|
"mean_token_accuracy": 0.17719073742628097,
|
|
"num_tokens": 6135516.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.564949369430542,
|
|
"epoch": 2.6978646069968195,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004583081618344684,
|
|
"loss": 5.3917,
|
|
"mean_token_accuracy": 0.1639246791601181,
|
|
"num_tokens": 6146206.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.440062475204468,
|
|
"epoch": 2.7024079963652885,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000458102989779501,
|
|
"loss": 5.3413,
|
|
"mean_token_accuracy": 0.17632217854261398,
|
|
"num_tokens": 6156244.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.440253639221192,
|
|
"epoch": 2.7069513857337575,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004578973659361081,
|
|
"loss": 5.3882,
|
|
"mean_token_accuracy": 0.16691307574510575,
|
|
"num_tokens": 6167091.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.617421102523804,
|
|
"epoch": 2.711494775102226,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045769129081164596,
|
|
"loss": 5.4674,
|
|
"mean_token_accuracy": 0.16855571717023848,
|
|
"num_tokens": 6177993.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.552607917785645,
|
|
"epoch": 2.716038164470695,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004574847649145845,
|
|
"loss": 5.5018,
|
|
"mean_token_accuracy": 0.17167451977729797,
|
|
"num_tokens": 6187963.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.556172895431518,
|
|
"epoch": 2.720581553839164,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000457277788754506,
|
|
"loss": 5.454,
|
|
"mean_token_accuracy": 0.16785339564085006,
|
|
"num_tokens": 6198442.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.592342567443848,
|
|
"epoch": 2.725124943207633,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045707036284210344,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.16420092135667802,
|
|
"num_tokens": 6209332.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.725124943207633,
|
|
"eval_entropy": 5.527031234253284,
|
|
"eval_loss": 5.894404411315918,
|
|
"eval_mean_token_accuracy": 0.15553997422373572,
|
|
"eval_num_tokens": 6209332.0,
|
|
"eval_runtime": 2.5988,
|
|
"eval_samples_per_second": 1321.764,
|
|
"eval_steps_per_second": 165.461,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.5462010383605955,
|
|
"epoch": 2.729668332576102,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004568624876891796,
|
|
"loss": 5.4423,
|
|
"mean_token_accuracy": 0.16724530458450318,
|
|
"num_tokens": 6219467.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.579170989990234,
|
|
"epoch": 2.7342117219445705,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045665416380864587,
|
|
"loss": 5.4374,
|
|
"mean_token_accuracy": 0.17122004181146622,
|
|
"num_tokens": 6229277.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.5708277225494385,
|
|
"epoch": 2.7387551113130395,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004564453917145206,
|
|
"loss": 5.394,
|
|
"mean_token_accuracy": 0.1746645987033844,
|
|
"num_tokens": 6239191.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.563258123397827,
|
|
"epoch": 2.7432985006815085,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004562361719219283,
|
|
"loss": 5.5064,
|
|
"mean_token_accuracy": 0.1637481302022934,
|
|
"num_tokens": 6249539.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.659433746337891,
|
|
"epoch": 2.747841890049977,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045602650494709786,
|
|
"loss": 5.4998,
|
|
"mean_token_accuracy": 0.16153956204652786,
|
|
"num_tokens": 6260896.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.644665479660034,
|
|
"epoch": 2.752385279418446,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045581639130736173,
|
|
"loss": 5.476,
|
|
"mean_token_accuracy": 0.16394941359758378,
|
|
"num_tokens": 6271828.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.599518489837647,
|
|
"epoch": 2.756928668786915,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004556058315211546,
|
|
"loss": 5.5235,
|
|
"mean_token_accuracy": 0.16828653514385222,
|
|
"num_tokens": 6282698.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.628439044952392,
|
|
"epoch": 2.761472058155384,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045539482610801156,
|
|
"loss": 5.4145,
|
|
"mean_token_accuracy": 0.17144882977008818,
|
|
"num_tokens": 6291905.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.503455400466919,
|
|
"epoch": 2.766015447523853,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004551833755885677,
|
|
"loss": 5.4649,
|
|
"mean_token_accuracy": 0.16770863384008408,
|
|
"num_tokens": 6301579.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.570836782455444,
|
|
"epoch": 2.7705588368923215,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004549714804845561,
|
|
"loss": 5.4228,
|
|
"mean_token_accuracy": 0.16588374674320222,
|
|
"num_tokens": 6311209.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.54651141166687,
|
|
"epoch": 2.7751022262607905,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004547591413188068,
|
|
"loss": 5.4086,
|
|
"mean_token_accuracy": 0.16484267264604568,
|
|
"num_tokens": 6321601.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.617306756973266,
|
|
"epoch": 2.7796456156292595,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004545463586152456,
|
|
"loss": 5.5221,
|
|
"mean_token_accuracy": 0.1671125575900078,
|
|
"num_tokens": 6331070.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.5911578178405765,
|
|
"epoch": 2.784189004997728,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004543331328988927,
|
|
"loss": 5.476,
|
|
"mean_token_accuracy": 0.16390688121318817,
|
|
"num_tokens": 6341772.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.492143678665161,
|
|
"epoch": 2.788732394366197,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004541194646958614,
|
|
"loss": 5.4319,
|
|
"mean_token_accuracy": 0.16603079289197922,
|
|
"num_tokens": 6352237.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.63241138458252,
|
|
"epoch": 2.793275783734666,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045390535453335666,
|
|
"loss": 5.5512,
|
|
"mean_token_accuracy": 0.1627603605389595,
|
|
"num_tokens": 6363645.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.59042501449585,
|
|
"epoch": 2.797819173103135,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045369080293967414,
|
|
"loss": 5.4235,
|
|
"mean_token_accuracy": 0.16862290501594543,
|
|
"num_tokens": 6373559.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.6492146968841555,
|
|
"epoch": 2.802362562471604,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045347581044419846,
|
|
"loss": 5.4196,
|
|
"mean_token_accuracy": 0.173454749584198,
|
|
"num_tokens": 6384197.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.477000188827515,
|
|
"epoch": 2.806905951840073,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045326037757740216,
|
|
"loss": 5.3318,
|
|
"mean_token_accuracy": 0.17231505811214448,
|
|
"num_tokens": 6394834.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.52566499710083,
|
|
"epoch": 2.8114493412085415,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004530445048708447,
|
|
"loss": 5.4303,
|
|
"mean_token_accuracy": 0.16809161901473998,
|
|
"num_tokens": 6405515.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.65869402885437,
|
|
"epoch": 2.8159927305770105,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045282819285717024,
|
|
"loss": 5.4597,
|
|
"mean_token_accuracy": 0.169706130027771,
|
|
"num_tokens": 6415885.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.540473604202271,
|
|
"epoch": 2.8205361199454795,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045261144207010736,
|
|
"loss": 5.3728,
|
|
"mean_token_accuracy": 0.1802276164293289,
|
|
"num_tokens": 6426053.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.538764762878418,
|
|
"epoch": 2.825079509313948,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045239425304446684,
|
|
"loss": 5.5089,
|
|
"mean_token_accuracy": 0.16386046707630159,
|
|
"num_tokens": 6436024.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.579290723800659,
|
|
"epoch": 2.829622898682417,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004521766263161413,
|
|
"loss": 5.4725,
|
|
"mean_token_accuracy": 0.1655261844396591,
|
|
"num_tokens": 6446847.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.543673324584961,
|
|
"epoch": 2.834166288050886,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000451958562422103,
|
|
"loss": 5.379,
|
|
"mean_token_accuracy": 0.1740669474005699,
|
|
"num_tokens": 6456901.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.544238519668579,
|
|
"epoch": 2.838709677419355,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045174006190040267,
|
|
"loss": 5.4251,
|
|
"mean_token_accuracy": 0.16738705039024354,
|
|
"num_tokens": 6467308.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.54875636100769,
|
|
"epoch": 2.843253066787824,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000451521125290169,
|
|
"loss": 5.4073,
|
|
"mean_token_accuracy": 0.17252165377140044,
|
|
"num_tokens": 6477071.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.542594861984253,
|
|
"epoch": 2.8477964561562925,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000451301753131606,
|
|
"loss": 5.4321,
|
|
"mean_token_accuracy": 0.1710267886519432,
|
|
"num_tokens": 6488300.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.561252880096435,
|
|
"epoch": 2.8523398455247615,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045108194596599286,
|
|
"loss": 5.4988,
|
|
"mean_token_accuracy": 0.15834234803915023,
|
|
"num_tokens": 6499515.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.497963094711304,
|
|
"epoch": 2.8568832348932305,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004508617043356818,
|
|
"loss": 5.4346,
|
|
"mean_token_accuracy": 0.1675139918923378,
|
|
"num_tokens": 6509167.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.589042901992798,
|
|
"epoch": 2.861426624261699,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004506410287840972,
|
|
"loss": 5.4516,
|
|
"mean_token_accuracy": 0.16526902765035628,
|
|
"num_tokens": 6520899.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.555674028396607,
|
|
"epoch": 2.865970013630168,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000450419919855734,
|
|
"loss": 5.4509,
|
|
"mean_token_accuracy": 0.16406622231006623,
|
|
"num_tokens": 6531293.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.555219125747681,
|
|
"epoch": 2.870513402998637,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004501983780961566,
|
|
"loss": 5.4563,
|
|
"mean_token_accuracy": 0.16738714277744293,
|
|
"num_tokens": 6541039.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.515122032165527,
|
|
"epoch": 2.875056792367106,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004499764040519972,
|
|
"loss": 5.4368,
|
|
"mean_token_accuracy": 0.17328928411006927,
|
|
"num_tokens": 6551948.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.565078020095825,
|
|
"epoch": 2.879600181735575,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044975399827095467,
|
|
"loss": 5.3574,
|
|
"mean_token_accuracy": 0.1725606456398964,
|
|
"num_tokens": 6562158.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.572651004791259,
|
|
"epoch": 2.8841435711040435,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044953116130179313,
|
|
"loss": 5.463,
|
|
"mean_token_accuracy": 0.16340922117233275,
|
|
"num_tokens": 6572605.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.467506408691406,
|
|
"epoch": 2.8886869604725125,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004493078936943408,
|
|
"loss": 5.4388,
|
|
"mean_token_accuracy": 0.1738500714302063,
|
|
"num_tokens": 6583247.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.667569875717163,
|
|
"epoch": 2.8932303498409815,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004490841959994882,
|
|
"loss": 5.5563,
|
|
"mean_token_accuracy": 0.16030332893133165,
|
|
"num_tokens": 6593371.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.568409204483032,
|
|
"epoch": 2.89777373920945,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044886006876918716,
|
|
"loss": 5.3323,
|
|
"mean_token_accuracy": 0.17091696262359618,
|
|
"num_tokens": 6603777.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.627681541442871,
|
|
"epoch": 2.902317128577919,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004486355125564493,
|
|
"loss": 5.5913,
|
|
"mean_token_accuracy": 0.16564136743545532,
|
|
"num_tokens": 6615654.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.6277184009552,
|
|
"epoch": 2.906860517946388,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004484105279153449,
|
|
"loss": 5.4397,
|
|
"mean_token_accuracy": 0.16940508633852006,
|
|
"num_tokens": 6626079.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.555370140075683,
|
|
"epoch": 2.911403907314857,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000448185115401001,
|
|
"loss": 5.3549,
|
|
"mean_token_accuracy": 0.17818198949098588,
|
|
"num_tokens": 6636047.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.543598699569702,
|
|
"epoch": 2.915947296683326,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004479592755696008,
|
|
"loss": 5.5192,
|
|
"mean_token_accuracy": 0.1616763636469841,
|
|
"num_tokens": 6646670.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.628071880340576,
|
|
"epoch": 2.9204906860517945,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044773300897838137,
|
|
"loss": 5.5334,
|
|
"mean_token_accuracy": 0.16706586331129075,
|
|
"num_tokens": 6658027.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.603945541381836,
|
|
"epoch": 2.9250340754202635,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004475063161856332,
|
|
"loss": 5.4451,
|
|
"mean_token_accuracy": 0.16946922540664672,
|
|
"num_tokens": 6668446.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.500796699523926,
|
|
"epoch": 2.9295774647887325,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004472791977506983,
|
|
"loss": 5.4775,
|
|
"mean_token_accuracy": 0.16555911302566528,
|
|
"num_tokens": 6679038.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.541210556030274,
|
|
"epoch": 2.934120854157201,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044705165423396855,
|
|
"loss": 5.3854,
|
|
"mean_token_accuracy": 0.17950762212276458,
|
|
"num_tokens": 6689207.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.539836978912353,
|
|
"epoch": 2.93866424352567,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044682368619688523,
|
|
"loss": 5.3844,
|
|
"mean_token_accuracy": 0.17160203605890273,
|
|
"num_tokens": 6699264.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.519908666610718,
|
|
"epoch": 2.943207632894139,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004465952942019367,
|
|
"loss": 5.4348,
|
|
"mean_token_accuracy": 0.16802884489297867,
|
|
"num_tokens": 6709309.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.590364980697632,
|
|
"epoch": 2.947751022262608,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044636647881265745,
|
|
"loss": 5.4041,
|
|
"mean_token_accuracy": 0.17207185178995132,
|
|
"num_tokens": 6719302.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.581957149505615,
|
|
"epoch": 2.952294411631077,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004461372405936269,
|
|
"loss": 5.3721,
|
|
"mean_token_accuracy": 0.16619062423706055,
|
|
"num_tokens": 6730668.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.516962671279908,
|
|
"epoch": 2.9568378009995455,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004459075801104673,
|
|
"loss": 5.377,
|
|
"mean_token_accuracy": 0.1696368098258972,
|
|
"num_tokens": 6740140.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.483997964859009,
|
|
"epoch": 2.9613811903680145,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044567749792984325,
|
|
"loss": 5.4051,
|
|
"mean_token_accuracy": 0.16935156732797624,
|
|
"num_tokens": 6751864.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.591496467590332,
|
|
"epoch": 2.9659245797364835,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044544699461945966,
|
|
"loss": 5.5374,
|
|
"mean_token_accuracy": 0.16746223866939544,
|
|
"num_tokens": 6762375.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.616140413284302,
|
|
"epoch": 2.970467969104952,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044521607074806037,
|
|
"loss": 5.4464,
|
|
"mean_token_accuracy": 0.17296652346849442,
|
|
"num_tokens": 6772181.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.586779308319092,
|
|
"epoch": 2.975011358473421,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004449847268854272,
|
|
"loss": 5.4632,
|
|
"mean_token_accuracy": 0.16927452385425568,
|
|
"num_tokens": 6782402.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.530501651763916,
|
|
"epoch": 2.97955474784189,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004447529636023782,
|
|
"loss": 5.4093,
|
|
"mean_token_accuracy": 0.17009349018335343,
|
|
"num_tokens": 6792671.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.490527057647705,
|
|
"epoch": 2.984098137210359,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00044452078147076614,
|
|
"loss": 5.3129,
|
|
"mean_token_accuracy": 0.1772904470562935,
|
|
"num_tokens": 6802111.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.503754377365112,
|
|
"epoch": 2.988641526578828,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004442881810634774,
|
|
"loss": 5.3987,
|
|
"mean_token_accuracy": 0.17522549331188203,
|
|
"num_tokens": 6812651.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.563858890533448,
|
|
"epoch": 2.9931849159472965,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004440551629544303,
|
|
"loss": 5.431,
|
|
"mean_token_accuracy": 0.16292740553617477,
|
|
"num_tokens": 6823208.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.582638597488403,
|
|
"epoch": 2.9977283053157655,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044382172771857383,
|
|
"loss": 5.4322,
|
|
"mean_token_accuracy": 0.16947678625583648,
|
|
"num_tokens": 6833412.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.572803603278266,
|
|
"epoch": 3.0018173557473875,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004435878759318864,
|
|
"loss": 5.295,
|
|
"mean_token_accuracy": 0.17448057896561092,
|
|
"num_tokens": 6841977.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.513389444351196,
|
|
"epoch": 3.0063607451158565,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004433536081713738,
|
|
"loss": 5.0925,
|
|
"mean_token_accuracy": 0.18363199532032012,
|
|
"num_tokens": 6852571.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.507623910903931,
|
|
"epoch": 3.0109041344843255,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004431189250150686,
|
|
"loss": 5.1497,
|
|
"mean_token_accuracy": 0.17519737929105758,
|
|
"num_tokens": 6861451.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.493114280700683,
|
|
"epoch": 3.015447523852794,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044288382704202816,
|
|
"loss": 5.092,
|
|
"mean_token_accuracy": 0.18401553481817245,
|
|
"num_tokens": 6872134.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.452037525177002,
|
|
"epoch": 3.019990913221263,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044264831483233335,
|
|
"loss": 5.1569,
|
|
"mean_token_accuracy": 0.18173797726631163,
|
|
"num_tokens": 6882477.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.477185869216919,
|
|
"epoch": 3.024534302589732,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044241238896708716,
|
|
"loss": 5.0955,
|
|
"mean_token_accuracy": 0.18331550806760788,
|
|
"num_tokens": 6892233.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.482796096801758,
|
|
"epoch": 3.029077691958201,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004421760500284131,
|
|
"loss": 5.0585,
|
|
"mean_token_accuracy": 0.18502746373414994,
|
|
"num_tokens": 6902090.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.496952819824219,
|
|
"epoch": 3.0336210813266695,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004419392985994541,
|
|
"loss": 5.157,
|
|
"mean_token_accuracy": 0.18697426468133926,
|
|
"num_tokens": 6912295.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.59736328125,
|
|
"epoch": 3.0381644706951385,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044170213526437076,
|
|
"loss": 5.1875,
|
|
"mean_token_accuracy": 0.1673503816127777,
|
|
"num_tokens": 6923064.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.50259313583374,
|
|
"epoch": 3.0427078600636075,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00044146456060833994,
|
|
"loss": 5.1104,
|
|
"mean_token_accuracy": 0.18162370026111602,
|
|
"num_tokens": 6932825.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.384890604019165,
|
|
"epoch": 3.0472512494320765,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044122657521755373,
|
|
"loss": 5.0209,
|
|
"mean_token_accuracy": 0.1872776612639427,
|
|
"num_tokens": 6943188.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.418858766555786,
|
|
"epoch": 3.051794638800545,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044098817967921716,
|
|
"loss": 5.128,
|
|
"mean_token_accuracy": 0.1796145498752594,
|
|
"num_tokens": 6953864.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.49640302658081,
|
|
"epoch": 3.056338028169014,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044074937458154763,
|
|
"loss": 5.1319,
|
|
"mean_token_accuracy": 0.18452672213315963,
|
|
"num_tokens": 6963745.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.48064546585083,
|
|
"epoch": 3.060881417537483,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044051016051377303,
|
|
"loss": 5.1411,
|
|
"mean_token_accuracy": 0.18625538051128387,
|
|
"num_tokens": 6973933.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.396935606002808,
|
|
"epoch": 3.065424806905952,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004402705380661302,
|
|
"loss": 5.0877,
|
|
"mean_token_accuracy": 0.18050813674926758,
|
|
"num_tokens": 6984314.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.498983907699585,
|
|
"epoch": 3.0699681962744205,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00044003050782986377,
|
|
"loss": 5.2505,
|
|
"mean_token_accuracy": 0.17716064900159836,
|
|
"num_tokens": 6995921.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.510509300231933,
|
|
"epoch": 3.0745115856428895,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004397900703972245,
|
|
"loss": 5.2135,
|
|
"mean_token_accuracy": 0.17649488300085067,
|
|
"num_tokens": 7006910.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.521936988830566,
|
|
"epoch": 3.0790549750113585,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004395492263614677,
|
|
"loss": 5.1042,
|
|
"mean_token_accuracy": 0.18011764138936998,
|
|
"num_tokens": 7017229.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.445307111740112,
|
|
"epoch": 3.0835983643798275,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043930797631685234,
|
|
"loss": 5.1786,
|
|
"mean_token_accuracy": 0.17943109273910524,
|
|
"num_tokens": 7027934.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.499941682815551,
|
|
"epoch": 3.088141753748296,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043906632085863876,
|
|
"loss": 5.1163,
|
|
"mean_token_accuracy": 0.18496674299240112,
|
|
"num_tokens": 7037964.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.480854845046997,
|
|
"epoch": 3.092685143116765,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043882426058308775,
|
|
"loss": 5.1236,
|
|
"mean_token_accuracy": 0.18577734380960464,
|
|
"num_tokens": 7048275.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.425022602081299,
|
|
"epoch": 3.097228532485234,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00043858179608745915,
|
|
"loss": 5.1273,
|
|
"mean_token_accuracy": 0.18461983352899553,
|
|
"num_tokens": 7058488.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.51739182472229,
|
|
"epoch": 3.101771921853703,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004383389279700098,
|
|
"loss": 5.1769,
|
|
"mean_token_accuracy": 0.18376078605651855,
|
|
"num_tokens": 7069195.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.409678363800049,
|
|
"epoch": 3.1063153112221715,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043809565682999274,
|
|
"loss": 5.0946,
|
|
"mean_token_accuracy": 0.1833580732345581,
|
|
"num_tokens": 7078858.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.454770278930664,
|
|
"epoch": 3.1108587005906405,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043785198326765535,
|
|
"loss": 5.1247,
|
|
"mean_token_accuracy": 0.18185268938541413,
|
|
"num_tokens": 7088551.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.414833450317383,
|
|
"epoch": 3.1154020899591095,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043760790788423785,
|
|
"loss": 5.0929,
|
|
"mean_token_accuracy": 0.18560312688350677,
|
|
"num_tokens": 7098987.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.430917882919312,
|
|
"epoch": 3.1199454793275785,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043736343128197207,
|
|
"loss": 5.1322,
|
|
"mean_token_accuracy": 0.1848243921995163,
|
|
"num_tokens": 7108337.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.454144763946533,
|
|
"epoch": 3.124488868696047,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004371185540640796,
|
|
"loss": 5.1493,
|
|
"mean_token_accuracy": 0.18822747319936753,
|
|
"num_tokens": 7118267.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.460283041000366,
|
|
"epoch": 3.129032258064516,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043687327683477077,
|
|
"loss": 5.1943,
|
|
"mean_token_accuracy": 0.18505121618509293,
|
|
"num_tokens": 7129780.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.449014472961426,
|
|
"epoch": 3.133575647432985,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004366276001992427,
|
|
"loss": 5.156,
|
|
"mean_token_accuracy": 0.18080466389656066,
|
|
"num_tokens": 7139581.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.426307964324951,
|
|
"epoch": 3.138119036801454,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000436381524763678,
|
|
"loss": 5.1393,
|
|
"mean_token_accuracy": 0.18389866799116134,
|
|
"num_tokens": 7150093.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.425113153457642,
|
|
"epoch": 3.142662426169923,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004361350511352435,
|
|
"loss": 5.0859,
|
|
"mean_token_accuracy": 0.1798602297902107,
|
|
"num_tokens": 7159668.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.452692174911499,
|
|
"epoch": 3.1472058155383915,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004358881799220882,
|
|
"loss": 5.1848,
|
|
"mean_token_accuracy": 0.17840422242879866,
|
|
"num_tokens": 7169713.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.462359285354614,
|
|
"epoch": 3.1517492049068605,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043564091173334236,
|
|
"loss": 5.1764,
|
|
"mean_token_accuracy": 0.17656957358121872,
|
|
"num_tokens": 7180681.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.44673752784729,
|
|
"epoch": 3.1562925942753295,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043539324717911573,
|
|
"loss": 5.0349,
|
|
"mean_token_accuracy": 0.1882912516593933,
|
|
"num_tokens": 7189922.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.412583446502685,
|
|
"epoch": 3.1608359836437985,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043514518687049587,
|
|
"loss": 5.1277,
|
|
"mean_token_accuracy": 0.19167211502790452,
|
|
"num_tokens": 7200388.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.464343547821045,
|
|
"epoch": 3.165379373012267,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000434896731419547,
|
|
"loss": 5.1568,
|
|
"mean_token_accuracy": 0.18717652112245559,
|
|
"num_tokens": 7211219.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.338333940505981,
|
|
"epoch": 3.169922762380736,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004346478814393082,
|
|
"loss": 4.9798,
|
|
"mean_token_accuracy": 0.19143124520778657,
|
|
"num_tokens": 7220585.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.444310474395752,
|
|
"epoch": 3.174466151749205,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004343986375437921,
|
|
"loss": 5.1396,
|
|
"mean_token_accuracy": 0.1862209677696228,
|
|
"num_tokens": 7231123.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.397456312179566,
|
|
"epoch": 3.179009541117674,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043414900034798314,
|
|
"loss": 5.142,
|
|
"mean_token_accuracy": 0.18140021562576295,
|
|
"num_tokens": 7241802.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.179009541117674,
|
|
"eval_entropy": 5.283362882081852,
|
|
"eval_loss": 5.8798322677612305,
|
|
"eval_mean_token_accuracy": 0.15679230539258135,
|
|
"eval_num_tokens": 7241802.0,
|
|
"eval_runtime": 2.5833,
|
|
"eval_samples_per_second": 1329.671,
|
|
"eval_steps_per_second": 166.451,
|
|
"step": 3500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.0472512685184e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|