1cad190a13
Model: fpadovani/nld-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.3625624716038165,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 4496087207808000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|