1cad190a13
Model: fpadovani/nld-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.45433893684688775,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 1490466908160000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|