3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.08779631255487269,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691977596282959,
|
|
"epoch": 0.000292654375182909,
|
|
"grad_norm": 12.75,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8274,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10517.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691955089569092,
|
|
"epoch": 0.000585308750365818,
|
|
"grad_norm": 12.875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7547,
|
|
"mean_token_accuracy": 0.00010537407360970974,
|
|
"num_tokens": 20499.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690744304656983,
|
|
"epoch": 0.000877963125548727,
|
|
"grad_norm": 10.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.507,
|
|
"mean_token_accuracy": 0.03299602484330535,
|
|
"num_tokens": 30396.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.663488483428955,
|
|
"epoch": 0.001170617500731636,
|
|
"grad_norm": 6.0625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.196,
|
|
"mean_token_accuracy": 0.05248659364879131,
|
|
"num_tokens": 40862.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.466671466827393,
|
|
"epoch": 0.0014632718759145448,
|
|
"grad_norm": 4.125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9118,
|
|
"mean_token_accuracy": 0.05587219782173634,
|
|
"num_tokens": 50482.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.479176902770996,
|
|
"epoch": 0.001755926251097454,
|
|
"grad_norm": 3.28125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7666,
|
|
"mean_token_accuracy": 0.05055557526648045,
|
|
"num_tokens": 59557.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.499413967132568,
|
|
"epoch": 0.002048580626280363,
|
|
"grad_norm": 2.84375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.7212,
|
|
"mean_token_accuracy": 0.05302470289170742,
|
|
"num_tokens": 69864.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.425041770935058,
|
|
"epoch": 0.002341235001463272,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.6782,
|
|
"mean_token_accuracy": 0.05624178126454353,
|
|
"num_tokens": 80632.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.474136352539062,
|
|
"epoch": 0.0026338893766461808,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6706,
|
|
"mean_token_accuracy": 0.04874711595475674,
|
|
"num_tokens": 90889.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.464450550079345,
|
|
"epoch": 0.0029265437518290896,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5772,
|
|
"mean_token_accuracy": 0.05688605234026909,
|
|
"num_tokens": 101136.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.375558757781983,
|
|
"epoch": 0.003219198127011999,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5567,
|
|
"mean_token_accuracy": 0.06301373578608035,
|
|
"num_tokens": 111743.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.292755794525146,
|
|
"epoch": 0.003511852502194908,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5051,
|
|
"mean_token_accuracy": 0.06023350767791271,
|
|
"num_tokens": 121968.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.296923160552979,
|
|
"epoch": 0.0038045068773778167,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4037,
|
|
"mean_token_accuracy": 0.06824495643377304,
|
|
"num_tokens": 131917.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.250029182434082,
|
|
"epoch": 0.004097161252560726,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.3308,
|
|
"mean_token_accuracy": 0.06861314773559571,
|
|
"num_tokens": 141562.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.270350456237793,
|
|
"epoch": 0.004389815627743635,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.323,
|
|
"mean_token_accuracy": 0.06920466646552086,
|
|
"num_tokens": 152163.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.280463314056396,
|
|
"epoch": 0.004682470002926544,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.301,
|
|
"mean_token_accuracy": 0.06251571290194988,
|
|
"num_tokens": 162684.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.195634841918945,
|
|
"epoch": 0.004975124378109453,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1337,
|
|
"mean_token_accuracy": 0.07077112905681134,
|
|
"num_tokens": 173234.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.168700313568115,
|
|
"epoch": 0.0052677787532923615,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.1283,
|
|
"mean_token_accuracy": 0.06726612299680709,
|
|
"num_tokens": 183645.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.11437921524048,
|
|
"epoch": 0.00556043312847527,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.0472,
|
|
"mean_token_accuracy": 0.06981992349028587,
|
|
"num_tokens": 193874.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.146056175231934,
|
|
"epoch": 0.005853087503658179,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.9144,
|
|
"mean_token_accuracy": 0.0724099788814783,
|
|
"num_tokens": 203965.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.023962688446044,
|
|
"epoch": 0.006145741878841089,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.8606,
|
|
"mean_token_accuracy": 0.07363765798509121,
|
|
"num_tokens": 214334.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.978809070587157,
|
|
"epoch": 0.006438396254023998,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.7784,
|
|
"mean_token_accuracy": 0.07168939225375652,
|
|
"num_tokens": 225435.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.947597980499268,
|
|
"epoch": 0.006731050629206907,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.6424,
|
|
"mean_token_accuracy": 0.06688602603971958,
|
|
"num_tokens": 235107.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.726509189605713,
|
|
"epoch": 0.007023705004389816,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.5683,
|
|
"mean_token_accuracy": 0.07460153996944427,
|
|
"num_tokens": 245660.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.732983684539795,
|
|
"epoch": 0.0073163593795727245,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.4869,
|
|
"mean_token_accuracy": 0.07104945406317711,
|
|
"num_tokens": 256077.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.572216033935547,
|
|
"epoch": 0.007609013754755633,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.4352,
|
|
"mean_token_accuracy": 0.06920053586363792,
|
|
"num_tokens": 265702.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.483925437927246,
|
|
"epoch": 0.007901668129938543,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.3711,
|
|
"mean_token_accuracy": 0.0720779214054346,
|
|
"num_tokens": 276739.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.266712093353272,
|
|
"epoch": 0.008194322505121452,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.2541,
|
|
"mean_token_accuracy": 0.07680930271744728,
|
|
"num_tokens": 285370.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.12420482635498,
|
|
"epoch": 0.008486976880304361,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.2031,
|
|
"mean_token_accuracy": 0.07261989042162895,
|
|
"num_tokens": 295080.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.998386192321778,
|
|
"epoch": 0.00877963125548727,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.1432,
|
|
"mean_token_accuracy": 0.07861845716834068,
|
|
"num_tokens": 304509.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.892388534545898,
|
|
"epoch": 0.009072285630670179,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.1203,
|
|
"mean_token_accuracy": 0.07556253299117088,
|
|
"num_tokens": 315214.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.732898807525634,
|
|
"epoch": 0.009364940005853088,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.1005,
|
|
"mean_token_accuracy": 0.07594769522547722,
|
|
"num_tokens": 325865.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.679925060272216,
|
|
"epoch": 0.009657594381035996,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.0335,
|
|
"mean_token_accuracy": 0.07580512613058091,
|
|
"num_tokens": 335490.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.627653217315673,
|
|
"epoch": 0.009950248756218905,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.1191,
|
|
"mean_token_accuracy": 0.06616722382605075,
|
|
"num_tokens": 345507.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.511928844451905,
|
|
"epoch": 0.010242903131401814,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.0372,
|
|
"mean_token_accuracy": 0.07108715996146202,
|
|
"num_tokens": 355129.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.449845886230468,
|
|
"epoch": 0.010535557506584723,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.9472,
|
|
"mean_token_accuracy": 0.07913482263684272,
|
|
"num_tokens": 364700.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.450096702575683,
|
|
"epoch": 0.010828211881767632,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.0521,
|
|
"mean_token_accuracy": 0.07427797615528106,
|
|
"num_tokens": 374920.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.375695610046387,
|
|
"epoch": 0.01112086625695054,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.8379,
|
|
"mean_token_accuracy": 0.07953086122870445,
|
|
"num_tokens": 384517.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.32225456237793,
|
|
"epoch": 0.01141352063213345,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.0024,
|
|
"mean_token_accuracy": 0.0835827887058258,
|
|
"num_tokens": 394118.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.35567274093628,
|
|
"epoch": 0.011706175007316359,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.9942,
|
|
"mean_token_accuracy": 0.07656502649188042,
|
|
"num_tokens": 404135.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.368758583068848,
|
|
"epoch": 0.01199882938249927,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.9689,
|
|
"mean_token_accuracy": 0.0749567061662674,
|
|
"num_tokens": 414198.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.292402458190917,
|
|
"epoch": 0.012291483757682178,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.9303,
|
|
"mean_token_accuracy": 0.0756587840616703,
|
|
"num_tokens": 424747.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.321221828460693,
|
|
"epoch": 0.012584138132865087,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.9038,
|
|
"mean_token_accuracy": 0.07831631563603877,
|
|
"num_tokens": 435597.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.24533395767212,
|
|
"epoch": 0.012876792508047996,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.9501,
|
|
"mean_token_accuracy": 0.07723408415913582,
|
|
"num_tokens": 445262.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.260603141784667,
|
|
"epoch": 0.013169446883230905,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.0105,
|
|
"mean_token_accuracy": 0.06991274990141391,
|
|
"num_tokens": 456455.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.28931884765625,
|
|
"epoch": 0.013462101258413814,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.8175,
|
|
"mean_token_accuracy": 0.07421639449894428,
|
|
"num_tokens": 465696.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.135773086547852,
|
|
"epoch": 0.013754755633596722,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.89,
|
|
"mean_token_accuracy": 0.079552548751235,
|
|
"num_tokens": 475171.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.152732849121094,
|
|
"epoch": 0.014047410008779631,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.8955,
|
|
"mean_token_accuracy": 0.07634073123335838,
|
|
"num_tokens": 484852.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.248970890045166,
|
|
"epoch": 0.01434006438396254,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.8986,
|
|
"mean_token_accuracy": 0.07723180837929249,
|
|
"num_tokens": 494539.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.244297409057618,
|
|
"epoch": 0.014632718759145449,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.8722,
|
|
"mean_token_accuracy": 0.07467923238873482,
|
|
"num_tokens": 504406.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.212084579467774,
|
|
"epoch": 0.014925373134328358,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.8875,
|
|
"mean_token_accuracy": 0.07326231747865677,
|
|
"num_tokens": 514905.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.192084980010986,
|
|
"epoch": 0.015218027509511267,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.841,
|
|
"mean_token_accuracy": 0.07844860479235649,
|
|
"num_tokens": 524904.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.152733612060548,
|
|
"epoch": 0.015510681884694176,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.8749,
|
|
"mean_token_accuracy": 0.07390310689806938,
|
|
"num_tokens": 534423.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.228134441375733,
|
|
"epoch": 0.015803336259877086,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.9388,
|
|
"mean_token_accuracy": 0.06983838640153409,
|
|
"num_tokens": 544468.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.17867660522461,
|
|
"epoch": 0.016095990635059995,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.8759,
|
|
"mean_token_accuracy": 0.07778191342949867,
|
|
"num_tokens": 555873.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.1208966255188,
|
|
"epoch": 0.016388645010242904,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.8485,
|
|
"mean_token_accuracy": 0.07848807573318481,
|
|
"num_tokens": 566228.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.147705793380737,
|
|
"epoch": 0.016681299385425813,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.7519,
|
|
"mean_token_accuracy": 0.07823162600398063,
|
|
"num_tokens": 576087.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.103727769851684,
|
|
"epoch": 0.016973953760608722,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.8782,
|
|
"mean_token_accuracy": 0.07339520268142223,
|
|
"num_tokens": 586768.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.19735746383667,
|
|
"epoch": 0.01726660813579163,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.871,
|
|
"mean_token_accuracy": 0.07267906814813614,
|
|
"num_tokens": 597352.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.021362733840942,
|
|
"epoch": 0.01755926251097454,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.7407,
|
|
"mean_token_accuracy": 0.07754108160734177,
|
|
"num_tokens": 607460.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.081702327728271,
|
|
"epoch": 0.01785191688615745,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.7764,
|
|
"mean_token_accuracy": 0.07714812532067299,
|
|
"num_tokens": 618882.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.187482166290284,
|
|
"epoch": 0.018144571261340357,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.8853,
|
|
"mean_token_accuracy": 0.07371802367269993,
|
|
"num_tokens": 628378.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.044809103012085,
|
|
"epoch": 0.018437225636523266,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.8403,
|
|
"mean_token_accuracy": 0.07703564129769802,
|
|
"num_tokens": 637339.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.112459707260133,
|
|
"epoch": 0.018729880011706175,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.8195,
|
|
"mean_token_accuracy": 0.0773365393280983,
|
|
"num_tokens": 648190.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.131685829162597,
|
|
"epoch": 0.019022534386889084,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.7489,
|
|
"mean_token_accuracy": 0.07892031595110893,
|
|
"num_tokens": 658704.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.087395620346069,
|
|
"epoch": 0.019315188762071993,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.7497,
|
|
"mean_token_accuracy": 0.07515155747532845,
|
|
"num_tokens": 668888.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.038054466247559,
|
|
"epoch": 0.0196078431372549,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.7818,
|
|
"mean_token_accuracy": 0.07353989705443383,
|
|
"num_tokens": 679294.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.008176279067992,
|
|
"epoch": 0.01990049751243781,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.7553,
|
|
"mean_token_accuracy": 0.07883304283022881,
|
|
"num_tokens": 689662.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.033820152282715,
|
|
"epoch": 0.02019315188762072,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.7645,
|
|
"mean_token_accuracy": 0.07818084880709648,
|
|
"num_tokens": 700356.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.944976902008056,
|
|
"epoch": 0.02048580626280363,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.6998,
|
|
"mean_token_accuracy": 0.07956961244344711,
|
|
"num_tokens": 710831.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.022411108016968,
|
|
"epoch": 0.020778460637986537,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.6943,
|
|
"mean_token_accuracy": 0.07844458818435669,
|
|
"num_tokens": 720445.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.978969573974609,
|
|
"epoch": 0.021071115013169446,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.6882,
|
|
"mean_token_accuracy": 0.08231284841895103,
|
|
"num_tokens": 730355.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.976030063629151,
|
|
"epoch": 0.021363769388352355,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.714,
|
|
"mean_token_accuracy": 0.08171503022313117,
|
|
"num_tokens": 740397.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.986613845825195,
|
|
"epoch": 0.021656423763535264,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.718,
|
|
"mean_token_accuracy": 0.07903816923499107,
|
|
"num_tokens": 750277.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.9489898681640625,
|
|
"epoch": 0.021949078138718173,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.6953,
|
|
"mean_token_accuracy": 0.07957587093114853,
|
|
"num_tokens": 759764.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.996847534179688,
|
|
"epoch": 0.02224173251390108,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.7437,
|
|
"mean_token_accuracy": 0.07961214706301689,
|
|
"num_tokens": 769709.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.960434198379517,
|
|
"epoch": 0.02253438688908399,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.6888,
|
|
"mean_token_accuracy": 0.08052227124571801,
|
|
"num_tokens": 780267.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.887183332443238,
|
|
"epoch": 0.0228270412642669,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.6611,
|
|
"mean_token_accuracy": 0.0823482058942318,
|
|
"num_tokens": 790184.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.074115419387818,
|
|
"epoch": 0.023119695639449808,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.7323,
|
|
"mean_token_accuracy": 0.07797951698303222,
|
|
"num_tokens": 800645.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.919103336334229,
|
|
"epoch": 0.023412350014632717,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.6645,
|
|
"mean_token_accuracy": 0.08313405811786652,
|
|
"num_tokens": 810873.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.905709600448608,
|
|
"epoch": 0.02370500438981563,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.6646,
|
|
"mean_token_accuracy": 0.08137974441051483,
|
|
"num_tokens": 821661.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.993849468231201,
|
|
"epoch": 0.02399765876499854,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.6417,
|
|
"mean_token_accuracy": 0.08281447514891624,
|
|
"num_tokens": 831818.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.86600136756897,
|
|
"epoch": 0.024290313140181447,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.6784,
|
|
"mean_token_accuracy": 0.08450354561209679,
|
|
"num_tokens": 843030.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.940770626068115,
|
|
"epoch": 0.024582967515364356,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.6337,
|
|
"mean_token_accuracy": 0.09171406775712967,
|
|
"num_tokens": 852477.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.925455904006958,
|
|
"epoch": 0.024875621890547265,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.6457,
|
|
"mean_token_accuracy": 0.08530597761273384,
|
|
"num_tokens": 863018.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.887920331954956,
|
|
"epoch": 0.025168276265730174,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.5837,
|
|
"mean_token_accuracy": 0.07957525178790092,
|
|
"num_tokens": 872732.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.84278564453125,
|
|
"epoch": 0.025460930640913083,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.6344,
|
|
"mean_token_accuracy": 0.0791689209640026,
|
|
"num_tokens": 883189.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.75295729637146,
|
|
"epoch": 0.02575358501609599,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.6724,
|
|
"mean_token_accuracy": 0.08212153613567352,
|
|
"num_tokens": 893874.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.893164730072021,
|
|
"epoch": 0.0260462393912789,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.5997,
|
|
"mean_token_accuracy": 0.08359748423099518,
|
|
"num_tokens": 904424.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.8800774097442625,
|
|
"epoch": 0.02633889376646181,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.6555,
|
|
"mean_token_accuracy": 0.07997112423181534,
|
|
"num_tokens": 914601.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.880753421783448,
|
|
"epoch": 0.026631548141644718,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.6627,
|
|
"mean_token_accuracy": 0.08024522922933101,
|
|
"num_tokens": 925162.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.87500581741333,
|
|
"epoch": 0.026924202516827627,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.5647,
|
|
"mean_token_accuracy": 0.08628017753362656,
|
|
"num_tokens": 935182.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.85167350769043,
|
|
"epoch": 0.027216856892010536,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.6007,
|
|
"mean_token_accuracy": 0.07952202931046486,
|
|
"num_tokens": 944812.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.8027184963226315,
|
|
"epoch": 0.027509511267193445,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.6031,
|
|
"mean_token_accuracy": 0.0839798703789711,
|
|
"num_tokens": 954532.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.882350301742553,
|
|
"epoch": 0.027802165642376354,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.557,
|
|
"mean_token_accuracy": 0.08790333345532417,
|
|
"num_tokens": 963611.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.754200458526611,
|
|
"epoch": 0.028094820017559263,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.6005,
|
|
"mean_token_accuracy": 0.08356534838676452,
|
|
"num_tokens": 973177.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.877309036254883,
|
|
"epoch": 0.02838747439274217,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.5999,
|
|
"mean_token_accuracy": 0.08032175153493881,
|
|
"num_tokens": 983623.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.859593296051026,
|
|
"epoch": 0.02868012876792508,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.6515,
|
|
"mean_token_accuracy": 0.0777646329253912,
|
|
"num_tokens": 993810.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.860140466690064,
|
|
"epoch": 0.02897278314310799,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.5529,
|
|
"mean_token_accuracy": 0.08074892833828926,
|
|
"num_tokens": 1004155.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.809819650650025,
|
|
"epoch": 0.029265437518290898,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.5186,
|
|
"mean_token_accuracy": 0.08852879330515862,
|
|
"num_tokens": 1014896.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.734996938705445,
|
|
"epoch": 0.029558091893473807,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.4593,
|
|
"mean_token_accuracy": 0.0899660475552082,
|
|
"num_tokens": 1024991.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.704321193695068,
|
|
"epoch": 0.029850746268656716,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.5035,
|
|
"mean_token_accuracy": 0.09011007472872734,
|
|
"num_tokens": 1036697.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.777565670013428,
|
|
"epoch": 0.030143400643839625,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.5742,
|
|
"mean_token_accuracy": 0.07923320233821869,
|
|
"num_tokens": 1046245.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.677716588973999,
|
|
"epoch": 0.030436055019022534,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.4666,
|
|
"mean_token_accuracy": 0.09066108837723733,
|
|
"num_tokens": 1056232.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.591713285446167,
|
|
"epoch": 0.030728709394205442,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.3564,
|
|
"mean_token_accuracy": 0.09806667268276215,
|
|
"num_tokens": 1065536.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.824580192565918,
|
|
"epoch": 0.03102136376938835,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.4864,
|
|
"mean_token_accuracy": 0.0877640463411808,
|
|
"num_tokens": 1075135.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.644569540023804,
|
|
"epoch": 0.031314018144571264,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.4849,
|
|
"mean_token_accuracy": 0.08468610793352127,
|
|
"num_tokens": 1085498.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.75826244354248,
|
|
"epoch": 0.03160667251975417,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.5173,
|
|
"mean_token_accuracy": 0.08989619985222816,
|
|
"num_tokens": 1095559.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.702898693084717,
|
|
"epoch": 0.03189932689493708,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.4577,
|
|
"mean_token_accuracy": 0.08639004975557327,
|
|
"num_tokens": 1105701.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.664012336730957,
|
|
"epoch": 0.03219198127011999,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.449,
|
|
"mean_token_accuracy": 0.08857948780059814,
|
|
"num_tokens": 1116455.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.666244554519653,
|
|
"epoch": 0.0324846356453029,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.479,
|
|
"mean_token_accuracy": 0.08700974211096764,
|
|
"num_tokens": 1126596.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.726414442062378,
|
|
"epoch": 0.03277729002048581,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.499,
|
|
"mean_token_accuracy": 0.08912580162286758,
|
|
"num_tokens": 1136969.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.644306802749634,
|
|
"epoch": 0.03306994439566872,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.4698,
|
|
"mean_token_accuracy": 0.0905322976410389,
|
|
"num_tokens": 1147175.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.675201559066773,
|
|
"epoch": 0.033362598770851626,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.4342,
|
|
"mean_token_accuracy": 0.08494588285684586,
|
|
"num_tokens": 1156539.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.602833318710327,
|
|
"epoch": 0.033655253146034535,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.4346,
|
|
"mean_token_accuracy": 0.08530361130833626,
|
|
"num_tokens": 1166630.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.671648740768433,
|
|
"epoch": 0.033947907521217444,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.4563,
|
|
"mean_token_accuracy": 0.0869609259068966,
|
|
"num_tokens": 1176279.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.705862236022949,
|
|
"epoch": 0.03424056189640035,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.4396,
|
|
"mean_token_accuracy": 0.08494879379868507,
|
|
"num_tokens": 1186321.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.618758630752564,
|
|
"epoch": 0.03453321627158326,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.4286,
|
|
"mean_token_accuracy": 0.082433769851923,
|
|
"num_tokens": 1198287.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.56975564956665,
|
|
"epoch": 0.03482587064676617,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.4261,
|
|
"mean_token_accuracy": 0.0872905969619751,
|
|
"num_tokens": 1208618.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.6115447044372555,
|
|
"epoch": 0.03511852502194908,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.4087,
|
|
"mean_token_accuracy": 0.08459671512246132,
|
|
"num_tokens": 1217932.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.66593861579895,
|
|
"epoch": 0.03541117939713199,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.4552,
|
|
"mean_token_accuracy": 0.08583111017942428,
|
|
"num_tokens": 1227965.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.641590595245361,
|
|
"epoch": 0.0357038337723149,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.3849,
|
|
"mean_token_accuracy": 0.08682549521327018,
|
|
"num_tokens": 1237879.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.628591918945313,
|
|
"epoch": 0.035996488147497806,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.3709,
|
|
"mean_token_accuracy": 0.08815298303961754,
|
|
"num_tokens": 1248547.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.592710399627686,
|
|
"epoch": 0.036289142522680715,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.4141,
|
|
"mean_token_accuracy": 0.08911648765206337,
|
|
"num_tokens": 1259462.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.500692319869995,
|
|
"epoch": 0.036581796897863623,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.4601,
|
|
"mean_token_accuracy": 0.0926029309630394,
|
|
"num_tokens": 1270668.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.620847511291504,
|
|
"epoch": 0.03687445127304653,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.4281,
|
|
"mean_token_accuracy": 0.08673678189516068,
|
|
"num_tokens": 1280459.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.60734920501709,
|
|
"epoch": 0.03716710564822944,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.3529,
|
|
"mean_token_accuracy": 0.09062747210264206,
|
|
"num_tokens": 1291114.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.678114891052246,
|
|
"epoch": 0.03745976002341235,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.4891,
|
|
"mean_token_accuracy": 0.0807405211031437,
|
|
"num_tokens": 1301442.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.498232316970825,
|
|
"epoch": 0.03775241439859526,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.3042,
|
|
"mean_token_accuracy": 0.09260110929608345,
|
|
"num_tokens": 1311315.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.427506780624389,
|
|
"epoch": 0.03804506877377817,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.3316,
|
|
"mean_token_accuracy": 0.0925542175769806,
|
|
"num_tokens": 1320962.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.54851884841919,
|
|
"epoch": 0.03833772314896108,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.3727,
|
|
"mean_token_accuracy": 0.09268264397978783,
|
|
"num_tokens": 1331006.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.576120758056641,
|
|
"epoch": 0.038630377524143986,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.3409,
|
|
"mean_token_accuracy": 0.08663004487752915,
|
|
"num_tokens": 1341342.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.465879917144775,
|
|
"epoch": 0.038923031899326895,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.3029,
|
|
"mean_token_accuracy": 0.08978775963187217,
|
|
"num_tokens": 1350484.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.51578164100647,
|
|
"epoch": 0.0392156862745098,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.3375,
|
|
"mean_token_accuracy": 0.08680702522397041,
|
|
"num_tokens": 1360422.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.3830320835113525,
|
|
"epoch": 0.03950834064969271,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.2542,
|
|
"mean_token_accuracy": 0.09665400013327599,
|
|
"num_tokens": 1370653.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.505912113189697,
|
|
"epoch": 0.03980099502487562,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.3277,
|
|
"mean_token_accuracy": 0.09258134663105011,
|
|
"num_tokens": 1379904.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.57382869720459,
|
|
"epoch": 0.04009364940005853,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.3169,
|
|
"mean_token_accuracy": 0.09115454629063606,
|
|
"num_tokens": 1390292.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.463971281051636,
|
|
"epoch": 0.04038630377524144,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.3081,
|
|
"mean_token_accuracy": 0.0934259407222271,
|
|
"num_tokens": 1400089.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.4088707447052,
|
|
"epoch": 0.04067895815042435,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.3249,
|
|
"mean_token_accuracy": 0.08856789469718933,
|
|
"num_tokens": 1409486.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.534200525283813,
|
|
"epoch": 0.04097161252560726,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.3246,
|
|
"mean_token_accuracy": 0.08936321660876274,
|
|
"num_tokens": 1419834.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.549503469467163,
|
|
"epoch": 0.041264266900790166,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.3435,
|
|
"mean_token_accuracy": 0.08608863651752471,
|
|
"num_tokens": 1430045.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.384866046905517,
|
|
"epoch": 0.041556921275973074,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.2372,
|
|
"mean_token_accuracy": 0.09240642040967942,
|
|
"num_tokens": 1439394.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.43756628036499,
|
|
"epoch": 0.04184957565115598,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.3577,
|
|
"mean_token_accuracy": 0.0927269198000431,
|
|
"num_tokens": 1449768.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.4466043472290036,
|
|
"epoch": 0.04214223002633889,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.2984,
|
|
"mean_token_accuracy": 0.08740061819553376,
|
|
"num_tokens": 1460716.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.405421400070191,
|
|
"epoch": 0.0424348844015218,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.2418,
|
|
"mean_token_accuracy": 0.09399772509932518,
|
|
"num_tokens": 1470447.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.345859813690185,
|
|
"epoch": 0.04272753877670471,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.2734,
|
|
"mean_token_accuracy": 0.09072801098227501,
|
|
"num_tokens": 1481195.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.407094430923462,
|
|
"epoch": 0.04302019315188762,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.256,
|
|
"mean_token_accuracy": 0.0962791122496128,
|
|
"num_tokens": 1491679.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.357693719863891,
|
|
"epoch": 0.04331284752707053,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.2081,
|
|
"mean_token_accuracy": 0.09399823993444442,
|
|
"num_tokens": 1501444.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.368088722229004,
|
|
"epoch": 0.043605501902253437,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.2364,
|
|
"mean_token_accuracy": 0.09220209643244744,
|
|
"num_tokens": 1511299.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.387484216690064,
|
|
"epoch": 0.043898156277436345,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.2527,
|
|
"mean_token_accuracy": 0.08768362179398537,
|
|
"num_tokens": 1522389.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.322992515563965,
|
|
"epoch": 0.044190810652619254,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.2208,
|
|
"mean_token_accuracy": 0.08979862108826638,
|
|
"num_tokens": 1533493.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.447886848449707,
|
|
"epoch": 0.04448346502780216,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.2967,
|
|
"mean_token_accuracy": 0.08659840822219848,
|
|
"num_tokens": 1542668.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.391131734848022,
|
|
"epoch": 0.04477611940298507,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.1518,
|
|
"mean_token_accuracy": 0.09064248204231262,
|
|
"num_tokens": 1552221.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.3629687309265135,
|
|
"epoch": 0.04506877377816798,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.3092,
|
|
"mean_token_accuracy": 0.0847119078040123,
|
|
"num_tokens": 1562807.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.3367757320404055,
|
|
"epoch": 0.04536142815335089,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.2427,
|
|
"mean_token_accuracy": 0.09314708188176155,
|
|
"num_tokens": 1572751.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.391009855270386,
|
|
"epoch": 0.0456540825285338,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.289,
|
|
"mean_token_accuracy": 0.08479087874293327,
|
|
"num_tokens": 1582802.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.323002338409424,
|
|
"epoch": 0.04594673690371671,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.2001,
|
|
"mean_token_accuracy": 0.0916834220290184,
|
|
"num_tokens": 1593780.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.306653308868408,
|
|
"epoch": 0.046239391278899616,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.2784,
|
|
"mean_token_accuracy": 0.08860109895467758,
|
|
"num_tokens": 1604386.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.427073383331299,
|
|
"epoch": 0.046532045654082525,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.2719,
|
|
"mean_token_accuracy": 0.08890935704112053,
|
|
"num_tokens": 1614359.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.268821048736572,
|
|
"epoch": 0.046824700029265434,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.1332,
|
|
"mean_token_accuracy": 0.09368834868073464,
|
|
"num_tokens": 1624389.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.353055620193482,
|
|
"epoch": 0.04711735440444834,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.2184,
|
|
"mean_token_accuracy": 0.09352055266499519,
|
|
"num_tokens": 1634608.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.236748743057251,
|
|
"epoch": 0.04741000877963126,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.1634,
|
|
"mean_token_accuracy": 0.095853441208601,
|
|
"num_tokens": 1645723.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.2720245838165285,
|
|
"epoch": 0.04770266315481417,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.1303,
|
|
"mean_token_accuracy": 0.09833073690533638,
|
|
"num_tokens": 1655182.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.243285274505615,
|
|
"epoch": 0.04799531752999708,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.1475,
|
|
"mean_token_accuracy": 0.09099882766604424,
|
|
"num_tokens": 1664597.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.277136993408203,
|
|
"epoch": 0.048287971905179985,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.2512,
|
|
"mean_token_accuracy": 0.09426932334899903,
|
|
"num_tokens": 1674809.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.181485414505005,
|
|
"epoch": 0.048580626280362894,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.1055,
|
|
"mean_token_accuracy": 0.09550172835588455,
|
|
"num_tokens": 1684082.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.367832040786743,
|
|
"epoch": 0.0488732806555458,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.2189,
|
|
"mean_token_accuracy": 0.0909538634121418,
|
|
"num_tokens": 1695547.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.3647455215454105,
|
|
"epoch": 0.04916593503072871,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.2685,
|
|
"mean_token_accuracy": 0.09123990088701248,
|
|
"num_tokens": 1706434.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.249982166290283,
|
|
"epoch": 0.04945858940591162,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.187,
|
|
"mean_token_accuracy": 0.09113817065954208,
|
|
"num_tokens": 1718063.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.216959714889526,
|
|
"epoch": 0.04975124378109453,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.187,
|
|
"mean_token_accuracy": 0.08529561832547188,
|
|
"num_tokens": 1728594.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.320889902114868,
|
|
"epoch": 0.05004389815627744,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.1762,
|
|
"mean_token_accuracy": 0.0950587272644043,
|
|
"num_tokens": 1738879.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.196149730682373,
|
|
"epoch": 0.05033655253146035,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.1489,
|
|
"mean_token_accuracy": 0.09129266664385796,
|
|
"num_tokens": 1748850.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.239317321777344,
|
|
"epoch": 0.050629206906643257,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.1758,
|
|
"mean_token_accuracy": 0.09727160260081291,
|
|
"num_tokens": 1759992.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.350429534912109,
|
|
"epoch": 0.050921861281826165,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.207,
|
|
"mean_token_accuracy": 0.08741379529237747,
|
|
"num_tokens": 1769879.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.124841737747192,
|
|
"epoch": 0.051214515657009074,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.0808,
|
|
"mean_token_accuracy": 0.09819410294294358,
|
|
"num_tokens": 1779429.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.272202873229981,
|
|
"epoch": 0.05150717003219198,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.1918,
|
|
"mean_token_accuracy": 0.08553909882903099,
|
|
"num_tokens": 1788655.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.256893968582153,
|
|
"epoch": 0.05179982440737489,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.1638,
|
|
"mean_token_accuracy": 0.08843306303024293,
|
|
"num_tokens": 1798454.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.1225560188293455,
|
|
"epoch": 0.0520924787825578,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.139,
|
|
"mean_token_accuracy": 0.09109530672430992,
|
|
"num_tokens": 1809016.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.287998104095459,
|
|
"epoch": 0.05238513315774071,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.0465,
|
|
"mean_token_accuracy": 0.10102619305253029,
|
|
"num_tokens": 1819992.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.165624284744263,
|
|
"epoch": 0.05267778753292362,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.1035,
|
|
"mean_token_accuracy": 0.09398501217365265,
|
|
"num_tokens": 1829932.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.105218172073364,
|
|
"epoch": 0.05297044190810653,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.0717,
|
|
"mean_token_accuracy": 0.09618882760405541,
|
|
"num_tokens": 1839877.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.160112476348877,
|
|
"epoch": 0.053263096283289436,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.0921,
|
|
"mean_token_accuracy": 0.0896080605685711,
|
|
"num_tokens": 1849619.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.986614036560058,
|
|
"epoch": 0.053555750658472345,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.0567,
|
|
"mean_token_accuracy": 0.09908407628536224,
|
|
"num_tokens": 1859513.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.1867334842681885,
|
|
"epoch": 0.053848405033655254,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.0246,
|
|
"mean_token_accuracy": 0.09950580522418022,
|
|
"num_tokens": 1869849.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.17880163192749,
|
|
"epoch": 0.05414105940883816,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.1353,
|
|
"mean_token_accuracy": 0.09082436189055443,
|
|
"num_tokens": 1879385.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.163400745391845,
|
|
"epoch": 0.05443371378402107,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.1464,
|
|
"mean_token_accuracy": 0.0945731982588768,
|
|
"num_tokens": 1889415.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.191874408721924,
|
|
"epoch": 0.05472636815920398,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.2114,
|
|
"mean_token_accuracy": 0.0922793872654438,
|
|
"num_tokens": 1901145.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.162847709655762,
|
|
"epoch": 0.05501902253438689,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.0702,
|
|
"mean_token_accuracy": 0.10264224112033844,
|
|
"num_tokens": 1910355.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.219426870346069,
|
|
"epoch": 0.0553116769095698,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.1286,
|
|
"mean_token_accuracy": 0.0911145493388176,
|
|
"num_tokens": 1920526.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.0836316585540775,
|
|
"epoch": 0.05560433128475271,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.0834,
|
|
"mean_token_accuracy": 0.09749370738863945,
|
|
"num_tokens": 1929884.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.146844291687012,
|
|
"epoch": 0.055896985659935616,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.0832,
|
|
"mean_token_accuracy": 0.09699373468756675,
|
|
"num_tokens": 1940707.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.1908472061157225,
|
|
"epoch": 0.056189640035118525,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.1551,
|
|
"mean_token_accuracy": 0.09065189436078072,
|
|
"num_tokens": 1951024.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.026748418807983,
|
|
"epoch": 0.056482294410301434,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.0405,
|
|
"mean_token_accuracy": 0.08979763388633728,
|
|
"num_tokens": 1961553.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.184598064422607,
|
|
"epoch": 0.05677494878548434,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.1215,
|
|
"mean_token_accuracy": 0.08966808468103409,
|
|
"num_tokens": 1971297.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.9619499206542965,
|
|
"epoch": 0.05706760316066725,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.0252,
|
|
"mean_token_accuracy": 0.09121646285057068,
|
|
"num_tokens": 1981753.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.195439386367798,
|
|
"epoch": 0.05736025753585016,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.1494,
|
|
"mean_token_accuracy": 0.09855610951781273,
|
|
"num_tokens": 1992590.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.115369653701782,
|
|
"epoch": 0.05765291191103307,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.1144,
|
|
"mean_token_accuracy": 0.09149665981531144,
|
|
"num_tokens": 2001745.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.156594371795654,
|
|
"epoch": 0.05794556628621598,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.0449,
|
|
"mean_token_accuracy": 0.09658506661653518,
|
|
"num_tokens": 2011171.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.011155891418457,
|
|
"epoch": 0.05823822066139889,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.0054,
|
|
"mean_token_accuracy": 0.09929059594869613,
|
|
"num_tokens": 2020732.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.176776313781739,
|
|
"epoch": 0.058530875036581796,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.0523,
|
|
"mean_token_accuracy": 0.09472576081752777,
|
|
"num_tokens": 2030702.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.000023984909058,
|
|
"epoch": 0.058823529411764705,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 7.0649,
|
|
"mean_token_accuracy": 0.09343343526124955,
|
|
"num_tokens": 2040933.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.050817203521729,
|
|
"epoch": 0.059116183786947614,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 7.1486,
|
|
"mean_token_accuracy": 0.09363606870174408,
|
|
"num_tokens": 2051083.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.166114139556885,
|
|
"epoch": 0.05940883816213052,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 7.0404,
|
|
"mean_token_accuracy": 0.09443358294665813,
|
|
"num_tokens": 2060290.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.024473714828491,
|
|
"epoch": 0.05970149253731343,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 7.0755,
|
|
"mean_token_accuracy": 0.10160979479551316,
|
|
"num_tokens": 2070537.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.96842851638794,
|
|
"epoch": 0.05999414691249634,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 7.02,
|
|
"mean_token_accuracy": 0.09730740785598754,
|
|
"num_tokens": 2081020.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.130616950988769,
|
|
"epoch": 0.06028680128767925,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 7.0202,
|
|
"mean_token_accuracy": 0.09943432137370109,
|
|
"num_tokens": 2090731.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.189221143722534,
|
|
"epoch": 0.06057945566286216,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 7.1187,
|
|
"mean_token_accuracy": 0.0950706273317337,
|
|
"num_tokens": 2100248.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.975552558898926,
|
|
"epoch": 0.06087211003804507,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 7.0041,
|
|
"mean_token_accuracy": 0.09668137729167939,
|
|
"num_tokens": 2110518.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.152935314178467,
|
|
"epoch": 0.061164764413227976,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 7.0801,
|
|
"mean_token_accuracy": 0.09166607484221459,
|
|
"num_tokens": 2121424.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.9373194694519045,
|
|
"epoch": 0.061457418788410885,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 7.0011,
|
|
"mean_token_accuracy": 0.1005364179611206,
|
|
"num_tokens": 2130904.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.107814741134644,
|
|
"epoch": 0.061750073163593794,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 7.0425,
|
|
"mean_token_accuracy": 0.09852371588349343,
|
|
"num_tokens": 2140710.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.131130743026733,
|
|
"epoch": 0.0620427275387767,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 7.0957,
|
|
"mean_token_accuracy": 0.09656500145792961,
|
|
"num_tokens": 2150767.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.963807916641235,
|
|
"epoch": 0.06233538191395961,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 7.0583,
|
|
"mean_token_accuracy": 0.09317740201950073,
|
|
"num_tokens": 2161414.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.0348228931427,
|
|
"epoch": 0.06262803628914253,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.9704,
|
|
"mean_token_accuracy": 0.10106262564659119,
|
|
"num_tokens": 2171960.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.988269376754761,
|
|
"epoch": 0.06292069066432543,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.9545,
|
|
"mean_token_accuracy": 0.0978710524737835,
|
|
"num_tokens": 2182344.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.8844482421875,
|
|
"epoch": 0.06321334503950835,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.9697,
|
|
"mean_token_accuracy": 0.09841145277023315,
|
|
"num_tokens": 2194829.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.943915987014771,
|
|
"epoch": 0.06350599941469125,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.9232,
|
|
"mean_token_accuracy": 0.09941138327121735,
|
|
"num_tokens": 2204662.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.095290327072144,
|
|
"epoch": 0.06379865378987416,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.9566,
|
|
"mean_token_accuracy": 0.10324475392699242,
|
|
"num_tokens": 2213815.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.923392009735108,
|
|
"epoch": 0.06409130816505706,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.9752,
|
|
"mean_token_accuracy": 0.10176795944571496,
|
|
"num_tokens": 2223026.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.005344009399414,
|
|
"epoch": 0.06438396254023998,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 7.0232,
|
|
"mean_token_accuracy": 0.09601744785904884,
|
|
"num_tokens": 2233575.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.986445569992066,
|
|
"epoch": 0.06467661691542288,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.8971,
|
|
"mean_token_accuracy": 0.10455278530716897,
|
|
"num_tokens": 2244171.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.964025020599365,
|
|
"epoch": 0.0649692712906058,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.9553,
|
|
"mean_token_accuracy": 0.09998015984892845,
|
|
"num_tokens": 2253972.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.934032917022705,
|
|
"epoch": 0.0652619256657887,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.9949,
|
|
"mean_token_accuracy": 0.09778623878955842,
|
|
"num_tokens": 2264699.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.02061276435852,
|
|
"epoch": 0.06555458004097162,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.9781,
|
|
"mean_token_accuracy": 0.10324280336499214,
|
|
"num_tokens": 2274894.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.9424807071685795,
|
|
"epoch": 0.06584723441615452,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 7.0036,
|
|
"mean_token_accuracy": 0.09519628286361695,
|
|
"num_tokens": 2285541.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.030115222930908,
|
|
"epoch": 0.06613988879133743,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.954,
|
|
"mean_token_accuracy": 0.10234859511256218,
|
|
"num_tokens": 2295443.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.947086143493652,
|
|
"epoch": 0.06643254316652034,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 7.0239,
|
|
"mean_token_accuracy": 0.09602486193180085,
|
|
"num_tokens": 2306149.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.944507074356079,
|
|
"epoch": 0.06672519754170325,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.9592,
|
|
"mean_token_accuracy": 0.0979753352701664,
|
|
"num_tokens": 2316589.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.021292734146118,
|
|
"epoch": 0.06701785191688615,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.9439,
|
|
"mean_token_accuracy": 0.10417687743902207,
|
|
"num_tokens": 2326340.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.961699962615967,
|
|
"epoch": 0.06731050629206907,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.917,
|
|
"mean_token_accuracy": 0.10509163066744805,
|
|
"num_tokens": 2336078.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.891680383682251,
|
|
"epoch": 0.06760316066725197,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.8219,
|
|
"mean_token_accuracy": 0.10515628084540367,
|
|
"num_tokens": 2346017.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.886036777496338,
|
|
"epoch": 0.06789581504243489,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.9646,
|
|
"mean_token_accuracy": 0.09433479383587837,
|
|
"num_tokens": 2355970.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.976067066192627,
|
|
"epoch": 0.06818846941761779,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.9037,
|
|
"mean_token_accuracy": 0.09823015183210373,
|
|
"num_tokens": 2366010.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.902979373931885,
|
|
"epoch": 0.0684811237928007,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.9054,
|
|
"mean_token_accuracy": 0.1001129500567913,
|
|
"num_tokens": 2376431.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.957536602020264,
|
|
"epoch": 0.0687737781679836,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.8969,
|
|
"mean_token_accuracy": 0.10259418487548828,
|
|
"num_tokens": 2387027.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.923643636703491,
|
|
"epoch": 0.06906643254316652,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.8952,
|
|
"mean_token_accuracy": 0.10383450165390969,
|
|
"num_tokens": 2396746.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.907520866394043,
|
|
"epoch": 0.06935908691834942,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.9158,
|
|
"mean_token_accuracy": 0.10082735046744347,
|
|
"num_tokens": 2406340.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.893206644058227,
|
|
"epoch": 0.06965174129353234,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.9427,
|
|
"mean_token_accuracy": 0.0992270477116108,
|
|
"num_tokens": 2417441.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.953333759307862,
|
|
"epoch": 0.06994439566871524,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.8218,
|
|
"mean_token_accuracy": 0.10669703856110573,
|
|
"num_tokens": 2427365.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.861691522598266,
|
|
"epoch": 0.07023705004389816,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.8665,
|
|
"mean_token_accuracy": 0.10322953760623932,
|
|
"num_tokens": 2437236.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.906563425064087,
|
|
"epoch": 0.07052970441908106,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.973,
|
|
"mean_token_accuracy": 0.09446094781160355,
|
|
"num_tokens": 2448000.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.908754634857178,
|
|
"epoch": 0.07082235879426398,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.8887,
|
|
"mean_token_accuracy": 0.10634658485651016,
|
|
"num_tokens": 2458627.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.876393127441406,
|
|
"epoch": 0.07111501316944688,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.8194,
|
|
"mean_token_accuracy": 0.10663582906126975,
|
|
"num_tokens": 2468110.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.933553409576416,
|
|
"epoch": 0.0714076675446298,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.9829,
|
|
"mean_token_accuracy": 0.09609689638018608,
|
|
"num_tokens": 2479527.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.877217864990234,
|
|
"epoch": 0.0717003219198127,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.8725,
|
|
"mean_token_accuracy": 0.10952854081988335,
|
|
"num_tokens": 2489856.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.883581113815308,
|
|
"epoch": 0.07199297629499561,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.9196,
|
|
"mean_token_accuracy": 0.09842127859592438,
|
|
"num_tokens": 2500164.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.001914691925049,
|
|
"epoch": 0.07228563067017851,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.9188,
|
|
"mean_token_accuracy": 0.10088242888450623,
|
|
"num_tokens": 2510535.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.857234334945678,
|
|
"epoch": 0.07257828504536143,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.8733,
|
|
"mean_token_accuracy": 0.1038922868669033,
|
|
"num_tokens": 2521334.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.899796676635742,
|
|
"epoch": 0.07287093942054433,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.8342,
|
|
"mean_token_accuracy": 0.10259378775954246,
|
|
"num_tokens": 2531990.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.847463941574096,
|
|
"epoch": 0.07316359379572725,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.9435,
|
|
"mean_token_accuracy": 0.1009743221104145,
|
|
"num_tokens": 2542418.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.958188772201538,
|
|
"epoch": 0.07345624817091015,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.9085,
|
|
"mean_token_accuracy": 0.10628803819417953,
|
|
"num_tokens": 2553944.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.8211475849151615,
|
|
"epoch": 0.07374890254609306,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.8592,
|
|
"mean_token_accuracy": 0.10203179717063904,
|
|
"num_tokens": 2564577.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.976000070571899,
|
|
"epoch": 0.07404155692127597,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.9663,
|
|
"mean_token_accuracy": 0.09566880464553833,
|
|
"num_tokens": 2574003.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.839417171478272,
|
|
"epoch": 0.07433421129645888,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.8109,
|
|
"mean_token_accuracy": 0.11298123002052307,
|
|
"num_tokens": 2585367.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.835225009918213,
|
|
"epoch": 0.07462686567164178,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.8087,
|
|
"mean_token_accuracy": 0.10525748059153557,
|
|
"num_tokens": 2595282.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.860079383850097,
|
|
"epoch": 0.0749195200468247,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.7604,
|
|
"mean_token_accuracy": 0.11382328793406486,
|
|
"num_tokens": 2605220.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.8381617069244385,
|
|
"epoch": 0.0752121744220076,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.8551,
|
|
"mean_token_accuracy": 0.10523293167352676,
|
|
"num_tokens": 2616492.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.865453815460205,
|
|
"epoch": 0.07550482879719052,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.8451,
|
|
"mean_token_accuracy": 0.09958744794130325,
|
|
"num_tokens": 2626133.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.7963536262512205,
|
|
"epoch": 0.07579748317237343,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.8325,
|
|
"mean_token_accuracy": 0.1026077076792717,
|
|
"num_tokens": 2636661.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.933389186859131,
|
|
"epoch": 0.07609013754755634,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.816,
|
|
"mean_token_accuracy": 0.10291560664772988,
|
|
"num_tokens": 2647598.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.722589731216431,
|
|
"epoch": 0.07638279192273925,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.7453,
|
|
"mean_token_accuracy": 0.10669632852077485,
|
|
"num_tokens": 2657267.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.8958821296691895,
|
|
"epoch": 0.07667544629792215,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.8478,
|
|
"mean_token_accuracy": 0.10577277392148972,
|
|
"num_tokens": 2667048.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.9029340744018555,
|
|
"epoch": 0.07696810067310507,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.8436,
|
|
"mean_token_accuracy": 0.0981109656393528,
|
|
"num_tokens": 2677386.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.724293518066406,
|
|
"epoch": 0.07726075504828797,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.8244,
|
|
"mean_token_accuracy": 0.10608431845903396,
|
|
"num_tokens": 2687655.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.951364278793335,
|
|
"epoch": 0.07755340942347089,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.7947,
|
|
"mean_token_accuracy": 0.10765203386545182,
|
|
"num_tokens": 2698098.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.722763299942017,
|
|
"epoch": 0.07784606379865379,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.7949,
|
|
"mean_token_accuracy": 0.10944582521915436,
|
|
"num_tokens": 2708119.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.886203479766846,
|
|
"epoch": 0.0781387181738367,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.7975,
|
|
"mean_token_accuracy": 0.10819103494286537,
|
|
"num_tokens": 2717036.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.829187726974487,
|
|
"epoch": 0.0784313725490196,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.9166,
|
|
"mean_token_accuracy": 0.10675505623221397,
|
|
"num_tokens": 2728290.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.83986463546753,
|
|
"epoch": 0.07872402692420252,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.8117,
|
|
"mean_token_accuracy": 0.10610796362161637,
|
|
"num_tokens": 2738373.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.891552734375,
|
|
"epoch": 0.07901668129938542,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.8265,
|
|
"mean_token_accuracy": 0.10829621031880379,
|
|
"num_tokens": 2748232.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.822065448760986,
|
|
"epoch": 0.07930933567456834,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.8269,
|
|
"mean_token_accuracy": 0.10751490145921708,
|
|
"num_tokens": 2758138.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.817276525497436,
|
|
"epoch": 0.07960199004975124,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.7964,
|
|
"mean_token_accuracy": 0.11103422194719315,
|
|
"num_tokens": 2768497.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.761292934417725,
|
|
"epoch": 0.07989464442493416,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.828,
|
|
"mean_token_accuracy": 0.10593293607234955,
|
|
"num_tokens": 2778956.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.919249629974365,
|
|
"epoch": 0.08018729880011706,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.7689,
|
|
"mean_token_accuracy": 0.10828186571598053,
|
|
"num_tokens": 2788295.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.8148205280303955,
|
|
"epoch": 0.08047995317529998,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.8132,
|
|
"mean_token_accuracy": 0.10210561156272888,
|
|
"num_tokens": 2797736.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.762246179580688,
|
|
"epoch": 0.08077260755048288,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.6702,
|
|
"mean_token_accuracy": 0.1148024320602417,
|
|
"num_tokens": 2806933.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.808951377868652,
|
|
"epoch": 0.0810652619256658,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.8339,
|
|
"mean_token_accuracy": 0.10306724533438683,
|
|
"num_tokens": 2817471.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.849059438705444,
|
|
"epoch": 0.0813579163008487,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.9718,
|
|
"mean_token_accuracy": 0.10684928670525551,
|
|
"num_tokens": 2828016.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.848748636245728,
|
|
"epoch": 0.08165057067603161,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.8005,
|
|
"mean_token_accuracy": 0.1023244470357895,
|
|
"num_tokens": 2838011.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.727694749832153,
|
|
"epoch": 0.08194322505121451,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.6735,
|
|
"mean_token_accuracy": 0.11005943566560746,
|
|
"num_tokens": 2848281.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.743996810913086,
|
|
"epoch": 0.08223587942639743,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.8155,
|
|
"mean_token_accuracy": 0.107390346378088,
|
|
"num_tokens": 2859338.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.874584150314331,
|
|
"epoch": 0.08252853380158033,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.8131,
|
|
"mean_token_accuracy": 0.10883597433567047,
|
|
"num_tokens": 2869238.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.764737939834594,
|
|
"epoch": 0.08282118817676325,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.7939,
|
|
"mean_token_accuracy": 0.10805475264787674,
|
|
"num_tokens": 2879312.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.771303415298462,
|
|
"epoch": 0.08311384255194615,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.7424,
|
|
"mean_token_accuracy": 0.10402957499027252,
|
|
"num_tokens": 2888974.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.8599330425262455,
|
|
"epoch": 0.08340649692712906,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.7749,
|
|
"mean_token_accuracy": 0.10574989691376686,
|
|
"num_tokens": 2899501.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.72014741897583,
|
|
"epoch": 0.08369915130231197,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.7974,
|
|
"mean_token_accuracy": 0.10439777374267578,
|
|
"num_tokens": 2909210.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.787353420257569,
|
|
"epoch": 0.08399180567749488,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.7491,
|
|
"mean_token_accuracy": 0.11018471121788025,
|
|
"num_tokens": 2919470.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.730296850204468,
|
|
"epoch": 0.08428446005267778,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.7117,
|
|
"mean_token_accuracy": 0.11215005367994309,
|
|
"num_tokens": 2929526.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.720075511932373,
|
|
"epoch": 0.0845771144278607,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.7261,
|
|
"mean_token_accuracy": 0.11107333600521088,
|
|
"num_tokens": 2938944.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.758396100997925,
|
|
"epoch": 0.0848697688030436,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.748,
|
|
"mean_token_accuracy": 0.10751529335975647,
|
|
"num_tokens": 2948334.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.71314549446106,
|
|
"epoch": 0.08516242317822652,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.7049,
|
|
"mean_token_accuracy": 0.11244085505604744,
|
|
"num_tokens": 2957687.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.85503659248352,
|
|
"epoch": 0.08545507755340942,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.7059,
|
|
"mean_token_accuracy": 0.11085794866085052,
|
|
"num_tokens": 2968291.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.69832329750061,
|
|
"epoch": 0.08574773192859234,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.6821,
|
|
"mean_token_accuracy": 0.11803077980875969,
|
|
"num_tokens": 2978809.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.829856252670288,
|
|
"epoch": 0.08604038630377524,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.8476,
|
|
"mean_token_accuracy": 0.10481944605708123,
|
|
"num_tokens": 2988839.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.744125032424927,
|
|
"epoch": 0.08633304067895815,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.7904,
|
|
"mean_token_accuracy": 0.10630232319235802,
|
|
"num_tokens": 2999380.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.794864654541016,
|
|
"epoch": 0.08662569505414106,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.6954,
|
|
"mean_token_accuracy": 0.10977205708622932,
|
|
"num_tokens": 3009298.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.670021867752075,
|
|
"epoch": 0.08691834942932397,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.7028,
|
|
"mean_token_accuracy": 0.11015302389860153,
|
|
"num_tokens": 3019747.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.7533143043518065,
|
|
"epoch": 0.08721100380450687,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.7011,
|
|
"mean_token_accuracy": 0.11383544877171517,
|
|
"num_tokens": 3029777.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.704556035995483,
|
|
"epoch": 0.08750365817968979,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.8275,
|
|
"mean_token_accuracy": 0.10207732617855073,
|
|
"num_tokens": 3040184.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.819152307510376,
|
|
"epoch": 0.08779631255487269,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.6749,
|
|
"mean_token_accuracy": 0.11216761693358421,
|
|
"num_tokens": 3049183.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 4544627816448000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|