6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.26492405510420347,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692046070098877,
|
|
"epoch": 0.0004415400918403391,
|
|
"grad_norm": 15.5625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7987,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9390.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69192419052124,
|
|
"epoch": 0.0008830801836806782,
|
|
"grad_norm": 14.4375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7211,
|
|
"mean_token_accuracy": 0.0002551274374127388,
|
|
"num_tokens": 18671.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.69034719467163,
|
|
"epoch": 0.0013246202755210173,
|
|
"grad_norm": 10.1875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4504,
|
|
"mean_token_accuracy": 0.04107050695456564,
|
|
"num_tokens": 27614.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.65851650238037,
|
|
"epoch": 0.0017661603673613563,
|
|
"grad_norm": 6.09375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1437,
|
|
"mean_token_accuracy": 0.05260300412774086,
|
|
"num_tokens": 37850.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.480317497253418,
|
|
"epoch": 0.0022077004592016957,
|
|
"grad_norm": 4.21875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.7713,
|
|
"mean_token_accuracy": 0.06779314689338208,
|
|
"num_tokens": 47166.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.35213108062744,
|
|
"epoch": 0.0026492405510420347,
|
|
"grad_norm": 2.921875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.6925,
|
|
"mean_token_accuracy": 0.0573563989251852,
|
|
"num_tokens": 55500.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.413340187072754,
|
|
"epoch": 0.0030907806428823736,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.676,
|
|
"mean_token_accuracy": 0.05297967046499252,
|
|
"num_tokens": 63851.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.392519474029541,
|
|
"epoch": 0.0035323207347227126,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.5804,
|
|
"mean_token_accuracy": 0.06258798092603683,
|
|
"num_tokens": 73697.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.351652717590332,
|
|
"epoch": 0.003973860826563052,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.5487,
|
|
"mean_token_accuracy": 0.06010213047266007,
|
|
"num_tokens": 83000.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.341690158843994,
|
|
"epoch": 0.004415400918403391,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5288,
|
|
"mean_token_accuracy": 0.06254246681928635,
|
|
"num_tokens": 92982.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.24167251586914,
|
|
"epoch": 0.00485694101024373,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4211,
|
|
"mean_token_accuracy": 0.06874397993087769,
|
|
"num_tokens": 101455.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.19759111404419,
|
|
"epoch": 0.005298481102084069,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.3764,
|
|
"mean_token_accuracy": 0.07629412487149238,
|
|
"num_tokens": 110782.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.167214965820312,
|
|
"epoch": 0.005740021193924408,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.2641,
|
|
"mean_token_accuracy": 0.07399890869855881,
|
|
"num_tokens": 119241.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.210993480682372,
|
|
"epoch": 0.006181561285764747,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2782,
|
|
"mean_token_accuracy": 0.07282476425170899,
|
|
"num_tokens": 127903.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.159809684753418,
|
|
"epoch": 0.006623101377605086,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2258,
|
|
"mean_token_accuracy": 0.07543315626680851,
|
|
"num_tokens": 137370.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.111583232879639,
|
|
"epoch": 0.007064641469445425,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.0774,
|
|
"mean_token_accuracy": 0.08292015641927719,
|
|
"num_tokens": 146582.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.07847490310669,
|
|
"epoch": 0.007506181561285765,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0108,
|
|
"mean_token_accuracy": 0.07970988750457764,
|
|
"num_tokens": 154933.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.015530204772949,
|
|
"epoch": 0.007947721653126103,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9747,
|
|
"mean_token_accuracy": 0.07985232844948768,
|
|
"num_tokens": 165157.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.035838222503662,
|
|
"epoch": 0.008389261744966443,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8768,
|
|
"mean_token_accuracy": 0.07754012271761894,
|
|
"num_tokens": 174958.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 9.947721004486084,
|
|
"epoch": 0.008830801836806783,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.7872,
|
|
"mean_token_accuracy": 0.0828177772462368,
|
|
"num_tokens": 184256.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.94132833480835,
|
|
"epoch": 0.009272341928647121,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.6971,
|
|
"mean_token_accuracy": 0.07942965477705002,
|
|
"num_tokens": 192894.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.842441368103028,
|
|
"epoch": 0.00971388202048746,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.6172,
|
|
"mean_token_accuracy": 0.08055685237050056,
|
|
"num_tokens": 202675.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.828944206237793,
|
|
"epoch": 0.010155422112327799,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.5621,
|
|
"mean_token_accuracy": 0.07997862100601197,
|
|
"num_tokens": 212261.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.671414470672607,
|
|
"epoch": 0.010596962204168139,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.5494,
|
|
"mean_token_accuracy": 0.07975570932030678,
|
|
"num_tokens": 222329.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.579044532775878,
|
|
"epoch": 0.011038502296008477,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.392,
|
|
"mean_token_accuracy": 0.07828540578484536,
|
|
"num_tokens": 231247.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.464293384552002,
|
|
"epoch": 0.011480042387848817,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.2789,
|
|
"mean_token_accuracy": 0.08522977083921432,
|
|
"num_tokens": 239978.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.329948711395264,
|
|
"epoch": 0.011921582479689156,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.243,
|
|
"mean_token_accuracy": 0.08045366927981376,
|
|
"num_tokens": 249735.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.195873355865478,
|
|
"epoch": 0.012363122571529495,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.1804,
|
|
"mean_token_accuracy": 0.07878185361623764,
|
|
"num_tokens": 259369.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.076773929595948,
|
|
"epoch": 0.012804662663369834,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.165,
|
|
"mean_token_accuracy": 0.07605967372655868,
|
|
"num_tokens": 268645.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.913009357452392,
|
|
"epoch": 0.013246202755210172,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.992,
|
|
"mean_token_accuracy": 0.08179110959172249,
|
|
"num_tokens": 276667.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.75396728515625,
|
|
"epoch": 0.013687742847050512,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.9969,
|
|
"mean_token_accuracy": 0.08649424239993095,
|
|
"num_tokens": 286017.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.627446365356445,
|
|
"epoch": 0.01412928293889085,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.9781,
|
|
"mean_token_accuracy": 0.08121571242809296,
|
|
"num_tokens": 295631.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.693611431121827,
|
|
"epoch": 0.01457082303073119,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.9589,
|
|
"mean_token_accuracy": 0.08223926201462746,
|
|
"num_tokens": 304704.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.534437942504884,
|
|
"epoch": 0.01501236312257153,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.9731,
|
|
"mean_token_accuracy": 0.08499160557985305,
|
|
"num_tokens": 314195.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.449780178070068,
|
|
"epoch": 0.015453903214411868,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.9605,
|
|
"mean_token_accuracy": 0.08167596161365509,
|
|
"num_tokens": 323379.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.427653789520264,
|
|
"epoch": 0.015895443306252206,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.864,
|
|
"mean_token_accuracy": 0.08790193870663643,
|
|
"num_tokens": 332322.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.308454895019532,
|
|
"epoch": 0.016336983398092548,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.8241,
|
|
"mean_token_accuracy": 0.0907207302749157,
|
|
"num_tokens": 341735.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.393166542053223,
|
|
"epoch": 0.016778523489932886,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.9195,
|
|
"mean_token_accuracy": 0.08468585535883903,
|
|
"num_tokens": 351209.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.27011661529541,
|
|
"epoch": 0.017220063581773224,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.8614,
|
|
"mean_token_accuracy": 0.08361415192484856,
|
|
"num_tokens": 360467.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.242918300628663,
|
|
"epoch": 0.017661603673613566,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.8241,
|
|
"mean_token_accuracy": 0.0869070716202259,
|
|
"num_tokens": 370361.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.279992771148681,
|
|
"epoch": 0.018103143765453904,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.8877,
|
|
"mean_token_accuracy": 0.07924672663211822,
|
|
"num_tokens": 380366.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.198458003997803,
|
|
"epoch": 0.018544683857294242,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.7928,
|
|
"mean_token_accuracy": 0.08724351823329926,
|
|
"num_tokens": 390690.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.174153900146484,
|
|
"epoch": 0.01898622394913458,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.8526,
|
|
"mean_token_accuracy": 0.0823112078011036,
|
|
"num_tokens": 400722.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.113383865356445,
|
|
"epoch": 0.01942776404097492,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.733,
|
|
"mean_token_accuracy": 0.08983569592237473,
|
|
"num_tokens": 410223.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.137540912628173,
|
|
"epoch": 0.01986930413281526,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.7948,
|
|
"mean_token_accuracy": 0.0847058854997158,
|
|
"num_tokens": 420214.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.165675449371339,
|
|
"epoch": 0.020310844224655598,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.7491,
|
|
"mean_token_accuracy": 0.08784600198268891,
|
|
"num_tokens": 429407.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.056314134597779,
|
|
"epoch": 0.02075238431649594,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.7019,
|
|
"mean_token_accuracy": 0.08565770387649536,
|
|
"num_tokens": 438403.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.126753520965575,
|
|
"epoch": 0.021193924408336277,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.7374,
|
|
"mean_token_accuracy": 0.09219109117984772,
|
|
"num_tokens": 447466.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.062481260299682,
|
|
"epoch": 0.021635464500176615,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.8271,
|
|
"mean_token_accuracy": 0.0802032507956028,
|
|
"num_tokens": 457141.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.097463226318359,
|
|
"epoch": 0.022077004592016954,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.698,
|
|
"mean_token_accuracy": 0.0921065405011177,
|
|
"num_tokens": 465708.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.170325756072998,
|
|
"epoch": 0.022518544683857295,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.7169,
|
|
"mean_token_accuracy": 0.0848769947886467,
|
|
"num_tokens": 475369.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.963189935684204,
|
|
"epoch": 0.022960084775697633,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.6855,
|
|
"mean_token_accuracy": 0.08922288343310356,
|
|
"num_tokens": 484249.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.03909511566162,
|
|
"epoch": 0.02340162486753797,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.7881,
|
|
"mean_token_accuracy": 0.08452328145503998,
|
|
"num_tokens": 493303.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.062015914916993,
|
|
"epoch": 0.023843164959378313,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.758,
|
|
"mean_token_accuracy": 0.09125194177031518,
|
|
"num_tokens": 501503.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.085032892227172,
|
|
"epoch": 0.02428470505121865,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.6942,
|
|
"mean_token_accuracy": 0.09186801388859749,
|
|
"num_tokens": 509661.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.047321224212647,
|
|
"epoch": 0.02472624514305899,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.6684,
|
|
"mean_token_accuracy": 0.09005074873566628,
|
|
"num_tokens": 519464.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.028418731689452,
|
|
"epoch": 0.025167785234899327,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.6342,
|
|
"mean_token_accuracy": 0.09150514975190163,
|
|
"num_tokens": 527968.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.062326526641845,
|
|
"epoch": 0.02560932532673967,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.6649,
|
|
"mean_token_accuracy": 0.09316564500331878,
|
|
"num_tokens": 537234.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.969491100311279,
|
|
"epoch": 0.026050865418580007,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.6521,
|
|
"mean_token_accuracy": 0.09308369681239129,
|
|
"num_tokens": 546398.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.048774671554565,
|
|
"epoch": 0.026492405510420345,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.6327,
|
|
"mean_token_accuracy": 0.09298400059342385,
|
|
"num_tokens": 555362.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.990510559082031,
|
|
"epoch": 0.026933945602260687,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.6861,
|
|
"mean_token_accuracy": 0.08575507774949073,
|
|
"num_tokens": 564575.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.113833093643189,
|
|
"epoch": 0.027375485694101025,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.7137,
|
|
"mean_token_accuracy": 0.08935339227318764,
|
|
"num_tokens": 573915.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.953329849243164,
|
|
"epoch": 0.027817025785941363,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.6052,
|
|
"mean_token_accuracy": 0.08935642167925835,
|
|
"num_tokens": 583216.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.974963426589966,
|
|
"epoch": 0.0282585658777817,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.6365,
|
|
"mean_token_accuracy": 0.09493646398186684,
|
|
"num_tokens": 591955.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.945090007781983,
|
|
"epoch": 0.028700105969622042,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.6372,
|
|
"mean_token_accuracy": 0.09564759358763694,
|
|
"num_tokens": 600999.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.945532560348511,
|
|
"epoch": 0.02914164606146238,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.5347,
|
|
"mean_token_accuracy": 0.09901952669024468,
|
|
"num_tokens": 609478.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.826081418991089,
|
|
"epoch": 0.02958318615330272,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.6044,
|
|
"mean_token_accuracy": 0.09094477742910385,
|
|
"num_tokens": 618348.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.00728144645691,
|
|
"epoch": 0.03002472624514306,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.651,
|
|
"mean_token_accuracy": 0.08828822076320648,
|
|
"num_tokens": 628548.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.951943874359131,
|
|
"epoch": 0.0304662663369834,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.5861,
|
|
"mean_token_accuracy": 0.09196597188711167,
|
|
"num_tokens": 637489.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.842890119552612,
|
|
"epoch": 0.030907806428823736,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.5698,
|
|
"mean_token_accuracy": 0.08872186839580536,
|
|
"num_tokens": 646715.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.839594936370849,
|
|
"epoch": 0.031349346520664075,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.5673,
|
|
"mean_token_accuracy": 0.09486312419176102,
|
|
"num_tokens": 656858.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.985370254516601,
|
|
"epoch": 0.03179088661250441,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.5492,
|
|
"mean_token_accuracy": 0.09201375618577004,
|
|
"num_tokens": 665968.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.856255674362183,
|
|
"epoch": 0.03223242670434476,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.5346,
|
|
"mean_token_accuracy": 0.0938662439584732,
|
|
"num_tokens": 674295.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.819755506515503,
|
|
"epoch": 0.032673966796185096,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.4683,
|
|
"mean_token_accuracy": 0.0921003058552742,
|
|
"num_tokens": 683559.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.767789649963379,
|
|
"epoch": 0.033115506888025434,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.4958,
|
|
"mean_token_accuracy": 0.09682972505688667,
|
|
"num_tokens": 692402.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.937490701675415,
|
|
"epoch": 0.03355704697986577,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.61,
|
|
"mean_token_accuracy": 0.08481517881155014,
|
|
"num_tokens": 702052.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.88536114692688,
|
|
"epoch": 0.03399858707170611,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.5255,
|
|
"mean_token_accuracy": 0.09476071372628211,
|
|
"num_tokens": 711926.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.827258634567261,
|
|
"epoch": 0.03444012716354645,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.4746,
|
|
"mean_token_accuracy": 0.09309235811233521,
|
|
"num_tokens": 720948.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.764995431900024,
|
|
"epoch": 0.034881667255386786,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.5254,
|
|
"mean_token_accuracy": 0.09631742537021637,
|
|
"num_tokens": 729611.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.932223796844482,
|
|
"epoch": 0.03532320734722713,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.4985,
|
|
"mean_token_accuracy": 0.08739121779799461,
|
|
"num_tokens": 738433.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.759926271438599,
|
|
"epoch": 0.03576474743906747,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.5344,
|
|
"mean_token_accuracy": 0.09274654388427735,
|
|
"num_tokens": 747310.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.756983041763306,
|
|
"epoch": 0.03620628753090781,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.5047,
|
|
"mean_token_accuracy": 0.09618922770023346,
|
|
"num_tokens": 756362.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.847079801559448,
|
|
"epoch": 0.036647827622748146,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.4946,
|
|
"mean_token_accuracy": 0.09153542071580886,
|
|
"num_tokens": 764978.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.769409275054931,
|
|
"epoch": 0.037089367714588484,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.4616,
|
|
"mean_token_accuracy": 0.09671959578990937,
|
|
"num_tokens": 774058.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.735736560821533,
|
|
"epoch": 0.03753090780642882,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.4465,
|
|
"mean_token_accuracy": 0.09103933200240136,
|
|
"num_tokens": 783332.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.767506504058838,
|
|
"epoch": 0.03797244789826916,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.5161,
|
|
"mean_token_accuracy": 0.09249499961733817,
|
|
"num_tokens": 792481.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.697516012191772,
|
|
"epoch": 0.038413987990109505,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.4117,
|
|
"mean_token_accuracy": 0.09788367450237274,
|
|
"num_tokens": 801396.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.726156330108642,
|
|
"epoch": 0.03885552808194984,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.4781,
|
|
"mean_token_accuracy": 0.09258785918354988,
|
|
"num_tokens": 809903.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.754117822647094,
|
|
"epoch": 0.03929706817379018,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.4228,
|
|
"mean_token_accuracy": 0.10010684058070182,
|
|
"num_tokens": 819144.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.638237190246582,
|
|
"epoch": 0.03973860826563052,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.4501,
|
|
"mean_token_accuracy": 0.09609700441360473,
|
|
"num_tokens": 828881.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.699933958053589,
|
|
"epoch": 0.04018014835747086,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.3763,
|
|
"mean_token_accuracy": 0.09211291894316673,
|
|
"num_tokens": 837588.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.64010763168335,
|
|
"epoch": 0.040621688449311195,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.3807,
|
|
"mean_token_accuracy": 0.09671871289610863,
|
|
"num_tokens": 847002.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.623350667953491,
|
|
"epoch": 0.041063228541151534,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.3849,
|
|
"mean_token_accuracy": 0.0986923448741436,
|
|
"num_tokens": 855791.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.721793460845947,
|
|
"epoch": 0.04150476863299188,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.4131,
|
|
"mean_token_accuracy": 0.09923509359359742,
|
|
"num_tokens": 865392.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.608292579650879,
|
|
"epoch": 0.04194630872483222,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.478,
|
|
"mean_token_accuracy": 0.09382415115833283,
|
|
"num_tokens": 874807.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.754652500152588,
|
|
"epoch": 0.042387848816672555,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.4339,
|
|
"mean_token_accuracy": 0.09926938116550446,
|
|
"num_tokens": 883928.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.652652740478516,
|
|
"epoch": 0.04282938890851289,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.3376,
|
|
"mean_token_accuracy": 0.09957752376794815,
|
|
"num_tokens": 893047.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.5953545570373535,
|
|
"epoch": 0.04327092900035323,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.3515,
|
|
"mean_token_accuracy": 0.10079265013337135,
|
|
"num_tokens": 901645.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.616002416610717,
|
|
"epoch": 0.04371246909219357,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.3691,
|
|
"mean_token_accuracy": 0.09735967218875885,
|
|
"num_tokens": 911169.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.688420724868775,
|
|
"epoch": 0.04415400918403391,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.3811,
|
|
"mean_token_accuracy": 0.0928712822496891,
|
|
"num_tokens": 921382.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.501060581207275,
|
|
"epoch": 0.04459554927587425,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.3012,
|
|
"mean_token_accuracy": 0.09976874813437461,
|
|
"num_tokens": 930409.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.606061887741089,
|
|
"epoch": 0.04503708936771459,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.283,
|
|
"mean_token_accuracy": 0.10316867306828499,
|
|
"num_tokens": 939049.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.589950037002564,
|
|
"epoch": 0.04547862945955493,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.2954,
|
|
"mean_token_accuracy": 0.10237202495336532,
|
|
"num_tokens": 947575.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.631234502792358,
|
|
"epoch": 0.045920169551395267,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.4624,
|
|
"mean_token_accuracy": 0.09285921081900597,
|
|
"num_tokens": 957848.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.598208427429199,
|
|
"epoch": 0.046361709643235605,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.2919,
|
|
"mean_token_accuracy": 0.10156237781047821,
|
|
"num_tokens": 966521.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.606314086914063,
|
|
"epoch": 0.04680324973507594,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.3541,
|
|
"mean_token_accuracy": 0.10115546062588691,
|
|
"num_tokens": 975827.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.526746702194214,
|
|
"epoch": 0.04724478982691628,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.2985,
|
|
"mean_token_accuracy": 0.10082201957702637,
|
|
"num_tokens": 985292.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.502959203720093,
|
|
"epoch": 0.047686329918756626,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.2896,
|
|
"mean_token_accuracy": 0.10343632996082305,
|
|
"num_tokens": 994791.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.516558980941772,
|
|
"epoch": 0.048127870010596964,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.2553,
|
|
"mean_token_accuracy": 0.10770378857851029,
|
|
"num_tokens": 1003700.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.503448867797852,
|
|
"epoch": 0.0485694101024373,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.2947,
|
|
"mean_token_accuracy": 0.10270922258496284,
|
|
"num_tokens": 1012682.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.590410900115967,
|
|
"epoch": 0.04901095019427764,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.3023,
|
|
"mean_token_accuracy": 0.10066464468836785,
|
|
"num_tokens": 1021018.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.514344882965088,
|
|
"epoch": 0.04945249028611798,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.1687,
|
|
"mean_token_accuracy": 0.10436531826853752,
|
|
"num_tokens": 1029744.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.328309202194214,
|
|
"epoch": 0.049894030377958316,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.2636,
|
|
"mean_token_accuracy": 0.11038385182619095,
|
|
"num_tokens": 1038805.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.525923824310302,
|
|
"epoch": 0.050335570469798654,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.3197,
|
|
"mean_token_accuracy": 0.10082013010978699,
|
|
"num_tokens": 1047656.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.603262233734131,
|
|
"epoch": 0.050777110561639,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.3119,
|
|
"mean_token_accuracy": 0.10330245941877365,
|
|
"num_tokens": 1056598.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.506361103057861,
|
|
"epoch": 0.05121865065347934,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.3105,
|
|
"mean_token_accuracy": 0.10376007333397866,
|
|
"num_tokens": 1065226.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.43021969795227,
|
|
"epoch": 0.051660190745319676,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.2312,
|
|
"mean_token_accuracy": 0.10194071978330613,
|
|
"num_tokens": 1074661.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.51248574256897,
|
|
"epoch": 0.052101730837160014,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.3225,
|
|
"mean_token_accuracy": 0.10019931197166443,
|
|
"num_tokens": 1083921.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.486316394805908,
|
|
"epoch": 0.05254327092900035,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.2137,
|
|
"mean_token_accuracy": 0.10983922332525253,
|
|
"num_tokens": 1093399.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.439640522003174,
|
|
"epoch": 0.05298481102084069,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.2835,
|
|
"mean_token_accuracy": 0.10640830993652343,
|
|
"num_tokens": 1104065.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.37196102142334,
|
|
"epoch": 0.05342635111268103,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.1855,
|
|
"mean_token_accuracy": 0.10447794273495674,
|
|
"num_tokens": 1112995.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.374333763122559,
|
|
"epoch": 0.05386789120452137,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.219,
|
|
"mean_token_accuracy": 0.10373939201235771,
|
|
"num_tokens": 1121637.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.4547041893005375,
|
|
"epoch": 0.05430943129636171,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.163,
|
|
"mean_token_accuracy": 0.10560059025883675,
|
|
"num_tokens": 1131166.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.4077800750732425,
|
|
"epoch": 0.05475097138820205,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.1546,
|
|
"mean_token_accuracy": 0.11212797611951827,
|
|
"num_tokens": 1140888.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.458501863479614,
|
|
"epoch": 0.05519251148004239,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.3313,
|
|
"mean_token_accuracy": 0.10084073692560196,
|
|
"num_tokens": 1150278.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.399151372909546,
|
|
"epoch": 0.055634051571882726,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.2699,
|
|
"mean_token_accuracy": 0.10444999486207962,
|
|
"num_tokens": 1159348.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.549579715728759,
|
|
"epoch": 0.056075591663723064,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.2841,
|
|
"mean_token_accuracy": 0.1042952410876751,
|
|
"num_tokens": 1168883.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.408907222747803,
|
|
"epoch": 0.0565171317555634,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.1845,
|
|
"mean_token_accuracy": 0.10902434810996056,
|
|
"num_tokens": 1178572.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.410972738265992,
|
|
"epoch": 0.05695867184740375,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.2662,
|
|
"mean_token_accuracy": 0.10295297279953956,
|
|
"num_tokens": 1188909.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.352515697479248,
|
|
"epoch": 0.057400211939244085,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.1696,
|
|
"mean_token_accuracy": 0.10579917058348656,
|
|
"num_tokens": 1197705.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.353521823883057,
|
|
"epoch": 0.05784175203108442,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.0136,
|
|
"mean_token_accuracy": 0.11067867055535316,
|
|
"num_tokens": 1206351.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.337934875488282,
|
|
"epoch": 0.05828329212292476,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.1286,
|
|
"mean_token_accuracy": 0.1017768256366253,
|
|
"num_tokens": 1214984.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.323618602752686,
|
|
"epoch": 0.0587248322147651,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.22,
|
|
"mean_token_accuracy": 0.0986666016280651,
|
|
"num_tokens": 1224485.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.373797750473022,
|
|
"epoch": 0.05916637230660544,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.2399,
|
|
"mean_token_accuracy": 0.10024571865797043,
|
|
"num_tokens": 1233560.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.387822151184082,
|
|
"epoch": 0.05960791239844578,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.1624,
|
|
"mean_token_accuracy": 0.10045728534460067,
|
|
"num_tokens": 1242628.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.265125274658203,
|
|
"epoch": 0.06004945249028612,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.0658,
|
|
"mean_token_accuracy": 0.1067326933145523,
|
|
"num_tokens": 1251004.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.378535032272339,
|
|
"epoch": 0.06049099258212646,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.2571,
|
|
"mean_token_accuracy": 0.10440935268998146,
|
|
"num_tokens": 1260344.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.322955751419068,
|
|
"epoch": 0.0609325326739668,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.1572,
|
|
"mean_token_accuracy": 0.10920507833361626,
|
|
"num_tokens": 1269988.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.2962963581085205,
|
|
"epoch": 0.061374072765807135,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.1193,
|
|
"mean_token_accuracy": 0.11102982461452485,
|
|
"num_tokens": 1280912.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.252480220794678,
|
|
"epoch": 0.06181561285764747,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.134,
|
|
"mean_token_accuracy": 0.10760174319148064,
|
|
"num_tokens": 1289684.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.354262113571167,
|
|
"epoch": 0.06225715294948781,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.1721,
|
|
"mean_token_accuracy": 0.10398160740733146,
|
|
"num_tokens": 1298853.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.227016019821167,
|
|
"epoch": 0.06269869304132815,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.1271,
|
|
"mean_token_accuracy": 0.1127957746386528,
|
|
"num_tokens": 1309112.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.324345111846924,
|
|
"epoch": 0.0631402331331685,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.1092,
|
|
"mean_token_accuracy": 0.1057639792561531,
|
|
"num_tokens": 1319064.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.188746166229248,
|
|
"epoch": 0.06358177322500883,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.0766,
|
|
"mean_token_accuracy": 0.11342449262738227,
|
|
"num_tokens": 1327889.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.324895286560059,
|
|
"epoch": 0.06402331331684917,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.1054,
|
|
"mean_token_accuracy": 0.10539671033620834,
|
|
"num_tokens": 1337241.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.222482204437256,
|
|
"epoch": 0.06446485340868952,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.1636,
|
|
"mean_token_accuracy": 0.10495321676135064,
|
|
"num_tokens": 1346527.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.305452728271485,
|
|
"epoch": 0.06490639350052985,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.1429,
|
|
"mean_token_accuracy": 0.10441825911402702,
|
|
"num_tokens": 1355677.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.290005493164062,
|
|
"epoch": 0.06534793359237019,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.1298,
|
|
"mean_token_accuracy": 0.10936881825327874,
|
|
"num_tokens": 1364874.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.279782056808472,
|
|
"epoch": 0.06578947368421052,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.0692,
|
|
"mean_token_accuracy": 0.11476822644472122,
|
|
"num_tokens": 1373717.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.331944179534912,
|
|
"epoch": 0.06623101377605087,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.0294,
|
|
"mean_token_accuracy": 0.11214973554015159,
|
|
"num_tokens": 1382767.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.091894626617432,
|
|
"epoch": 0.0666725538678912,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.0713,
|
|
"mean_token_accuracy": 0.11100590229034424,
|
|
"num_tokens": 1391190.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.175401639938355,
|
|
"epoch": 0.06711409395973154,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.0763,
|
|
"mean_token_accuracy": 0.10921204090118408,
|
|
"num_tokens": 1400722.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.318567132949829,
|
|
"epoch": 0.06755563405157189,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.0907,
|
|
"mean_token_accuracy": 0.10661459416151046,
|
|
"num_tokens": 1409328.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.106989192962646,
|
|
"epoch": 0.06799717414341222,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.967,
|
|
"mean_token_accuracy": 0.10996146872639656,
|
|
"num_tokens": 1418504.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.248307847976685,
|
|
"epoch": 0.06843871423525257,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.1104,
|
|
"mean_token_accuracy": 0.10902469977736473,
|
|
"num_tokens": 1427690.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.239201641082763,
|
|
"epoch": 0.0688802543270929,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.091,
|
|
"mean_token_accuracy": 0.11118610724806785,
|
|
"num_tokens": 1436798.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.1204283237457275,
|
|
"epoch": 0.06932179441893324,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.0933,
|
|
"mean_token_accuracy": 0.10568799898028373,
|
|
"num_tokens": 1446357.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.168965148925781,
|
|
"epoch": 0.06976333451077357,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.0073,
|
|
"mean_token_accuracy": 0.11038993671536446,
|
|
"num_tokens": 1455998.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.229979085922241,
|
|
"epoch": 0.07020487460261392,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.0599,
|
|
"mean_token_accuracy": 0.10410151407122611,
|
|
"num_tokens": 1465237.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.097687721252441,
|
|
"epoch": 0.07064641469445426,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.024,
|
|
"mean_token_accuracy": 0.11839348077774048,
|
|
"num_tokens": 1474363.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.138771963119507,
|
|
"epoch": 0.0710879547862946,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.0872,
|
|
"mean_token_accuracy": 0.1139365516602993,
|
|
"num_tokens": 1483379.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.267211008071899,
|
|
"epoch": 0.07152949487813494,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.0619,
|
|
"mean_token_accuracy": 0.10780827179551125,
|
|
"num_tokens": 1492507.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.073677682876587,
|
|
"epoch": 0.07197103496997527,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.9431,
|
|
"mean_token_accuracy": 0.11598889455199242,
|
|
"num_tokens": 1500888.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.079563140869141,
|
|
"epoch": 0.07241257506181561,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.9983,
|
|
"mean_token_accuracy": 0.11227924451231956,
|
|
"num_tokens": 1510310.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.121931362152099,
|
|
"epoch": 0.07285411515365595,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.9855,
|
|
"mean_token_accuracy": 0.11764683723449706,
|
|
"num_tokens": 1519427.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.14368166923523,
|
|
"epoch": 0.07329565524549629,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.0087,
|
|
"mean_token_accuracy": 0.10792600363492966,
|
|
"num_tokens": 1529456.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.080411815643311,
|
|
"epoch": 0.07373719533733664,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.9375,
|
|
"mean_token_accuracy": 0.11380904093384743,
|
|
"num_tokens": 1537695.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.071755599975586,
|
|
"epoch": 0.07417873542917697,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.0784,
|
|
"mean_token_accuracy": 0.11253369823098183,
|
|
"num_tokens": 1547511.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.204211902618408,
|
|
"epoch": 0.07462027552101731,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.0709,
|
|
"mean_token_accuracy": 0.11463971808552742,
|
|
"num_tokens": 1557035.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.083140897750854,
|
|
"epoch": 0.07506181561285764,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.0065,
|
|
"mean_token_accuracy": 0.11546228975057601,
|
|
"num_tokens": 1566773.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.061517763137817,
|
|
"epoch": 0.07550335570469799,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.0081,
|
|
"mean_token_accuracy": 0.10918990075588227,
|
|
"num_tokens": 1576873.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.128572607040406,
|
|
"epoch": 0.07594489579653832,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.0375,
|
|
"mean_token_accuracy": 0.10865660831332206,
|
|
"num_tokens": 1586170.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.218125915527343,
|
|
"epoch": 0.07638643588837866,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.0778,
|
|
"mean_token_accuracy": 0.11018856391310691,
|
|
"num_tokens": 1596054.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.02945466041565,
|
|
"epoch": 0.07682797598021901,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.9583,
|
|
"mean_token_accuracy": 0.11728574708104134,
|
|
"num_tokens": 1604544.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.105824756622314,
|
|
"epoch": 0.07726951607205934,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.9789,
|
|
"mean_token_accuracy": 0.1162014789879322,
|
|
"num_tokens": 1614580.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.058492374420166,
|
|
"epoch": 0.07771105616389969,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.0641,
|
|
"mean_token_accuracy": 0.11256316900253296,
|
|
"num_tokens": 1624701.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.0211296558380125,
|
|
"epoch": 0.07815259625574002,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.9931,
|
|
"mean_token_accuracy": 0.11021944805979729,
|
|
"num_tokens": 1634085.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.09322075843811,
|
|
"epoch": 0.07859413634758036,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.9645,
|
|
"mean_token_accuracy": 0.1174542985856533,
|
|
"num_tokens": 1643190.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.966829776763916,
|
|
"epoch": 0.0790356764394207,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.006,
|
|
"mean_token_accuracy": 0.11372746750712395,
|
|
"num_tokens": 1652705.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.127198505401611,
|
|
"epoch": 0.07947721653126104,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.977,
|
|
"mean_token_accuracy": 0.10978531762957573,
|
|
"num_tokens": 1662210.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.1131603717803955,
|
|
"epoch": 0.07991875662310138,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.0268,
|
|
"mean_token_accuracy": 0.10922672674059868,
|
|
"num_tokens": 1671893.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.049384355545044,
|
|
"epoch": 0.08036029671494171,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.9779,
|
|
"mean_token_accuracy": 0.11501859724521638,
|
|
"num_tokens": 1681217.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.015186929702759,
|
|
"epoch": 0.08080183680678206,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.988,
|
|
"mean_token_accuracy": 0.1133071318268776,
|
|
"num_tokens": 1690447.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.087394189834595,
|
|
"epoch": 0.08124337689862239,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.0741,
|
|
"mean_token_accuracy": 0.10595951378345489,
|
|
"num_tokens": 1700355.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.045471048355102,
|
|
"epoch": 0.08168491699046274,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.9851,
|
|
"mean_token_accuracy": 0.10544866621494293,
|
|
"num_tokens": 1709449.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.048018169403076,
|
|
"epoch": 0.08212645708230307,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.0167,
|
|
"mean_token_accuracy": 0.105999456346035,
|
|
"num_tokens": 1718838.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.981620121002197,
|
|
"epoch": 0.08256799717414341,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.9721,
|
|
"mean_token_accuracy": 0.11307689547538757,
|
|
"num_tokens": 1728594.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.088650274276733,
|
|
"epoch": 0.08300953726598376,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.0148,
|
|
"mean_token_accuracy": 0.11065684705972671,
|
|
"num_tokens": 1738814.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.006063842773438,
|
|
"epoch": 0.08345107735782409,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.9401,
|
|
"mean_token_accuracy": 0.11795193180441857,
|
|
"num_tokens": 1747644.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.048807621002197,
|
|
"epoch": 0.08389261744966443,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.9822,
|
|
"mean_token_accuracy": 0.11285707354545593,
|
|
"num_tokens": 1757489.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.994456195831299,
|
|
"epoch": 0.08433415754150476,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.8881,
|
|
"mean_token_accuracy": 0.12073725908994674,
|
|
"num_tokens": 1767546.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.929489183425903,
|
|
"epoch": 0.08477569763334511,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.9528,
|
|
"mean_token_accuracy": 0.10833125934004784,
|
|
"num_tokens": 1776035.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.031561088562012,
|
|
"epoch": 0.08521723772518544,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.0163,
|
|
"mean_token_accuracy": 0.10963534265756607,
|
|
"num_tokens": 1786161.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.13347601890564,
|
|
"epoch": 0.08565877781702579,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.0758,
|
|
"mean_token_accuracy": 0.10902466773986816,
|
|
"num_tokens": 1796093.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.994558525085449,
|
|
"epoch": 0.08610031790886613,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.9267,
|
|
"mean_token_accuracy": 0.10800953432917595,
|
|
"num_tokens": 1805574.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.005662727355957,
|
|
"epoch": 0.08654185800070646,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.9528,
|
|
"mean_token_accuracy": 0.11506856232881546,
|
|
"num_tokens": 1815175.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.939174652099609,
|
|
"epoch": 0.08698339809254681,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.8892,
|
|
"mean_token_accuracy": 0.11544275358319282,
|
|
"num_tokens": 1824683.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.9920509338378904,
|
|
"epoch": 0.08742493818438714,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.8652,
|
|
"mean_token_accuracy": 0.1142121821641922,
|
|
"num_tokens": 1833174.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.92423644065857,
|
|
"epoch": 0.08786647827622748,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.9598,
|
|
"mean_token_accuracy": 0.11292968466877937,
|
|
"num_tokens": 1842409.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.785798788070679,
|
|
"epoch": 0.08830801836806781,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.8616,
|
|
"mean_token_accuracy": 0.11990160048007965,
|
|
"num_tokens": 1851441.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.146481227874756,
|
|
"epoch": 0.08874955845990816,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499999998589561,
|
|
"loss": 6.9982,
|
|
"mean_token_accuracy": 0.11039657667279243,
|
|
"num_tokens": 1861188.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.901903963088989,
|
|
"epoch": 0.0891910985517485,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999999928596523,
|
|
"loss": 6.9301,
|
|
"mean_token_accuracy": 0.11363181099295616,
|
|
"num_tokens": 1870284.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.84821515083313,
|
|
"epoch": 0.08963263864358884,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999999827221219,
|
|
"loss": 6.8983,
|
|
"mean_token_accuracy": 0.12304212301969528,
|
|
"num_tokens": 1879744.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.960662460327148,
|
|
"epoch": 0.09007417873542918,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999999681769696,
|
|
"loss": 6.8983,
|
|
"mean_token_accuracy": 0.11716285720467567,
|
|
"num_tokens": 1889241.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.971296596527099,
|
|
"epoch": 0.09051571882726951,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499999949224196,
|
|
"loss": 6.9084,
|
|
"mean_token_accuracy": 0.1200313277542591,
|
|
"num_tokens": 1898247.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.955649042129517,
|
|
"epoch": 0.09095725891910986,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999999258638013,
|
|
"loss": 6.8611,
|
|
"mean_token_accuracy": 0.11644072383642197,
|
|
"num_tokens": 1907559.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.793795824050903,
|
|
"epoch": 0.09139879901095019,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999998980957861,
|
|
"loss": 6.8669,
|
|
"mean_token_accuracy": 0.12169807255268097,
|
|
"num_tokens": 1917072.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.9457625389099125,
|
|
"epoch": 0.09184033910279053,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999998659201508,
|
|
"loss": 6.8238,
|
|
"mean_token_accuracy": 0.1175873689353466,
|
|
"num_tokens": 1926597.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.854114770889282,
|
|
"epoch": 0.09228187919463088,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999998293368961,
|
|
"loss": 6.8766,
|
|
"mean_token_accuracy": 0.11491318792104721,
|
|
"num_tokens": 1935978.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.869282388687134,
|
|
"epoch": 0.09272341928647121,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999997883460227,
|
|
"loss": 6.8304,
|
|
"mean_token_accuracy": 0.11738137528300285,
|
|
"num_tokens": 1944424.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.931444597244263,
|
|
"epoch": 0.09316495937831155,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999997429475314,
|
|
"loss": 6.7926,
|
|
"mean_token_accuracy": 0.1264521062374115,
|
|
"num_tokens": 1953844.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.838017129898072,
|
|
"epoch": 0.09360649947015189,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999996931414232,
|
|
"loss": 6.8929,
|
|
"mean_token_accuracy": 0.12035574093461036,
|
|
"num_tokens": 1963974.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.908746671676636,
|
|
"epoch": 0.09404803956199223,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999996389276988,
|
|
"loss": 6.8573,
|
|
"mean_token_accuracy": 0.12281356453895569,
|
|
"num_tokens": 1973466.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.815357208251953,
|
|
"epoch": 0.09448957965383256,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999995803063596,
|
|
"loss": 6.8649,
|
|
"mean_token_accuracy": 0.11969125345349312,
|
|
"num_tokens": 1983478.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.947906064987182,
|
|
"epoch": 0.0949311197456729,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999995172774065,
|
|
"loss": 6.9373,
|
|
"mean_token_accuracy": 0.11548090726137161,
|
|
"num_tokens": 1992775.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.954066181182862,
|
|
"epoch": 0.09537265983751325,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999994498408408,
|
|
"loss": 6.9065,
|
|
"mean_token_accuracy": 0.11163463667035103,
|
|
"num_tokens": 2002526.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.880347061157226,
|
|
"epoch": 0.09581419992935358,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999993779966639,
|
|
"loss": 6.9226,
|
|
"mean_token_accuracy": 0.11343899965286255,
|
|
"num_tokens": 2012476.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.772713136672974,
|
|
"epoch": 0.09625574002119393,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999993017448771,
|
|
"loss": 6.7619,
|
|
"mean_token_accuracy": 0.1315491996705532,
|
|
"num_tokens": 2021252.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.972479009628296,
|
|
"epoch": 0.09669728011303426,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999992210854821,
|
|
"loss": 6.9025,
|
|
"mean_token_accuracy": 0.1197818472981453,
|
|
"num_tokens": 2031438.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.9433167457580565,
|
|
"epoch": 0.0971388202048746,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999991360184801,
|
|
"loss": 6.9497,
|
|
"mean_token_accuracy": 0.11331594586372376,
|
|
"num_tokens": 2041319.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.875679302215576,
|
|
"epoch": 0.09758036029671494,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999990465438731,
|
|
"loss": 6.9417,
|
|
"mean_token_accuracy": 0.11621764153242112,
|
|
"num_tokens": 2052060.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.868831157684326,
|
|
"epoch": 0.09802190038855528,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999989526616628,
|
|
"loss": 6.8207,
|
|
"mean_token_accuracy": 0.12271343320608138,
|
|
"num_tokens": 2061331.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.84274697303772,
|
|
"epoch": 0.09846344048039563,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999988543718509,
|
|
"loss": 6.8327,
|
|
"mean_token_accuracy": 0.1182619109749794,
|
|
"num_tokens": 2070006.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.921954727172851,
|
|
"epoch": 0.09890498057223596,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999987516744394,
|
|
"loss": 6.7824,
|
|
"mean_token_accuracy": 0.12447450086474418,
|
|
"num_tokens": 2079089.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.797588872909546,
|
|
"epoch": 0.0993465206640763,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999986445694303,
|
|
"loss": 6.778,
|
|
"mean_token_accuracy": 0.118412546813488,
|
|
"num_tokens": 2087237.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.792767906188965,
|
|
"epoch": 0.09978806075591663,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999985330568258,
|
|
"loss": 6.7448,
|
|
"mean_token_accuracy": 0.12214136868715286,
|
|
"num_tokens": 2096598.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.833769607543945,
|
|
"epoch": 0.10022960084775698,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999984171366278,
|
|
"loss": 6.8458,
|
|
"mean_token_accuracy": 0.11508344933390617,
|
|
"num_tokens": 2106106.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.860817527770996,
|
|
"epoch": 0.10067114093959731,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499998296808839,
|
|
"loss": 6.8066,
|
|
"mean_token_accuracy": 0.11775731146335602,
|
|
"num_tokens": 2115478.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.8433990478515625,
|
|
"epoch": 0.10111268103143765,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004999981720734615,
|
|
"loss": 6.8429,
|
|
"mean_token_accuracy": 0.11517784297466278,
|
|
"num_tokens": 2124439.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.893320178985595,
|
|
"epoch": 0.101554221123278,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999980429304977,
|
|
"loss": 6.861,
|
|
"mean_token_accuracy": 0.11661542877554894,
|
|
"num_tokens": 2133718.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.720952081680298,
|
|
"epoch": 0.10199576121511833,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999979093799502,
|
|
"loss": 6.7669,
|
|
"mean_token_accuracy": 0.1219759725034237,
|
|
"num_tokens": 2142978.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.826765871047973,
|
|
"epoch": 0.10243730130695868,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999977714218217,
|
|
"loss": 6.8287,
|
|
"mean_token_accuracy": 0.11483320519328118,
|
|
"num_tokens": 2152250.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.841332817077637,
|
|
"epoch": 0.102878841398799,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999976290561147,
|
|
"loss": 6.8186,
|
|
"mean_token_accuracy": 0.11421233564615249,
|
|
"num_tokens": 2161620.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.8951427936553955,
|
|
"epoch": 0.10332038149063935,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999974822828322,
|
|
"loss": 6.8321,
|
|
"mean_token_accuracy": 0.11468955874443054,
|
|
"num_tokens": 2170856.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.783407306671142,
|
|
"epoch": 0.10376192158247968,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499997331101977,
|
|
"loss": 6.8226,
|
|
"mean_token_accuracy": 0.1140426091849804,
|
|
"num_tokens": 2180926.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.841979598999023,
|
|
"epoch": 0.10420346167432003,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499997175513552,
|
|
"loss": 6.7918,
|
|
"mean_token_accuracy": 0.11650402471423149,
|
|
"num_tokens": 2190248.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.812575197219848,
|
|
"epoch": 0.10464500176616037,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999970155175603,
|
|
"loss": 6.8152,
|
|
"mean_token_accuracy": 0.12006641253829002,
|
|
"num_tokens": 2199833.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.79837212562561,
|
|
"epoch": 0.1050865418580007,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499996851114005,
|
|
"loss": 6.7952,
|
|
"mean_token_accuracy": 0.12233175113797187,
|
|
"num_tokens": 2208240.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.796438360214234,
|
|
"epoch": 0.10552808194984105,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999966823028894,
|
|
"loss": 6.8365,
|
|
"mean_token_accuracy": 0.11369045302271844,
|
|
"num_tokens": 2218758.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.935247182846069,
|
|
"epoch": 0.10596962204168138,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999965090842168,
|
|
"loss": 6.8302,
|
|
"mean_token_accuracy": 0.12288962900638581,
|
|
"num_tokens": 2228443.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.712041759490967,
|
|
"epoch": 0.10641116213352173,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999963314579905,
|
|
"loss": 6.6787,
|
|
"mean_token_accuracy": 0.13284276723861693,
|
|
"num_tokens": 2236787.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.800426197052002,
|
|
"epoch": 0.10685270222536206,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999961494242139,
|
|
"loss": 6.8593,
|
|
"mean_token_accuracy": 0.11500039100646972,
|
|
"num_tokens": 2247089.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.786765146255493,
|
|
"epoch": 0.1072942423172024,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999959629828908,
|
|
"loss": 6.7664,
|
|
"mean_token_accuracy": 0.11581304892897606,
|
|
"num_tokens": 2256045.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.886348628997803,
|
|
"epoch": 0.10773578240904275,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999957721340248,
|
|
"loss": 6.8384,
|
|
"mean_token_accuracy": 0.12041957154870034,
|
|
"num_tokens": 2265835.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.722246551513672,
|
|
"epoch": 0.10817732250088308,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999955768776195,
|
|
"loss": 6.7786,
|
|
"mean_token_accuracy": 0.12380016893148423,
|
|
"num_tokens": 2275318.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.808089828491211,
|
|
"epoch": 0.10861886259272342,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999953772136788,
|
|
"loss": 6.7658,
|
|
"mean_token_accuracy": 0.12278728261590004,
|
|
"num_tokens": 2284821.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.8324737548828125,
|
|
"epoch": 0.10906040268456375,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999951731422068,
|
|
"loss": 6.7482,
|
|
"mean_token_accuracy": 0.11858468577265739,
|
|
"num_tokens": 2294013.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.786966562271118,
|
|
"epoch": 0.1095019427764041,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999949646632072,
|
|
"loss": 6.73,
|
|
"mean_token_accuracy": 0.12245445623993874,
|
|
"num_tokens": 2302727.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.705034589767456,
|
|
"epoch": 0.10994348286824443,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999947517766842,
|
|
"loss": 6.7649,
|
|
"mean_token_accuracy": 0.12241661176085472,
|
|
"num_tokens": 2312032.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.982846593856811,
|
|
"epoch": 0.11038502296008477,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499994534482642,
|
|
"loss": 6.8443,
|
|
"mean_token_accuracy": 0.11787378117442131,
|
|
"num_tokens": 2321839.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.719681882858277,
|
|
"epoch": 0.11082656305192512,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999943127810847,
|
|
"loss": 6.8137,
|
|
"mean_token_accuracy": 0.11633599549531937,
|
|
"num_tokens": 2331255.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.744579744338989,
|
|
"epoch": 0.11126810314376545,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999940866720169,
|
|
"loss": 6.6363,
|
|
"mean_token_accuracy": 0.13151753917336464,
|
|
"num_tokens": 2340038.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.680500268936157,
|
|
"epoch": 0.1117096432356058,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999938561554429,
|
|
"loss": 6.7486,
|
|
"mean_token_accuracy": 0.1244722306728363,
|
|
"num_tokens": 2348901.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.772228145599366,
|
|
"epoch": 0.11215118332744613,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999936212313672,
|
|
"loss": 6.827,
|
|
"mean_token_accuracy": 0.11700050979852676,
|
|
"num_tokens": 2358842.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.788944578170776,
|
|
"epoch": 0.11259272341928647,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999933818997943,
|
|
"loss": 6.7218,
|
|
"mean_token_accuracy": 0.1224422350525856,
|
|
"num_tokens": 2368650.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.765577077865601,
|
|
"epoch": 0.1130342635111268,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999931381607292,
|
|
"loss": 6.7811,
|
|
"mean_token_accuracy": 0.12511691078543663,
|
|
"num_tokens": 2377916.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.782421112060547,
|
|
"epoch": 0.11347580360296715,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999928900141764,
|
|
"loss": 6.7279,
|
|
"mean_token_accuracy": 0.12245306149125099,
|
|
"num_tokens": 2387507.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.7623984813690186,
|
|
"epoch": 0.1139173436948075,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499992637460141,
|
|
"loss": 6.7762,
|
|
"mean_token_accuracy": 0.1228838860988617,
|
|
"num_tokens": 2396148.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.7144958019256595,
|
|
"epoch": 0.11435888378664782,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999923804986275,
|
|
"loss": 6.6356,
|
|
"mean_token_accuracy": 0.12269090339541436,
|
|
"num_tokens": 2404891.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.786207437515259,
|
|
"epoch": 0.11480042387848817,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999921191296415,
|
|
"loss": 6.6766,
|
|
"mean_token_accuracy": 0.12154053971171379,
|
|
"num_tokens": 2414406.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.703838205337524,
|
|
"epoch": 0.1152419639703285,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999918533531877,
|
|
"loss": 6.7829,
|
|
"mean_token_accuracy": 0.12308658063411712,
|
|
"num_tokens": 2424363.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.811527252197266,
|
|
"epoch": 0.11568350406216885,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999915831692714,
|
|
"loss": 6.7213,
|
|
"mean_token_accuracy": 0.12458684146404267,
|
|
"num_tokens": 2433753.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.642784214019775,
|
|
"epoch": 0.11612504415400918,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999913085778981,
|
|
"loss": 6.7353,
|
|
"mean_token_accuracy": 0.1179688699543476,
|
|
"num_tokens": 2443275.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.845086908340454,
|
|
"epoch": 0.11656658424584952,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999910295790729,
|
|
"loss": 6.7522,
|
|
"mean_token_accuracy": 0.116317979991436,
|
|
"num_tokens": 2452510.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.789072132110595,
|
|
"epoch": 0.11700812433768987,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999907461728014,
|
|
"loss": 6.8496,
|
|
"mean_token_accuracy": 0.1186776876449585,
|
|
"num_tokens": 2462742.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.73516001701355,
|
|
"epoch": 0.1174496644295302,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999904583590893,
|
|
"loss": 6.7087,
|
|
"mean_token_accuracy": 0.11719698682427407,
|
|
"num_tokens": 2471409.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.80999174118042,
|
|
"epoch": 0.11789120452137054,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999901661379418,
|
|
"loss": 6.66,
|
|
"mean_token_accuracy": 0.12834221944212915,
|
|
"num_tokens": 2481011.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.656212949752808,
|
|
"epoch": 0.11833274461321087,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999898695093652,
|
|
"loss": 6.7655,
|
|
"mean_token_accuracy": 0.1210456795990467,
|
|
"num_tokens": 2490664.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.774575805664062,
|
|
"epoch": 0.11877428470505122,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999895684733648,
|
|
"loss": 6.6871,
|
|
"mean_token_accuracy": 0.12933790683746338,
|
|
"num_tokens": 2499799.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.75890417098999,
|
|
"epoch": 0.11921582479689156,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004999892630299467,
|
|
"loss": 6.6744,
|
|
"mean_token_accuracy": 0.12494029551744461,
|
|
"num_tokens": 2508780.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.714679288864136,
|
|
"epoch": 0.1196573648887319,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999889531791171,
|
|
"loss": 6.6833,
|
|
"mean_token_accuracy": 0.12272755652666092,
|
|
"num_tokens": 2517741.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.662789583206177,
|
|
"epoch": 0.12009890498057224,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999886389208817,
|
|
"loss": 6.7716,
|
|
"mean_token_accuracy": 0.12046518027782441,
|
|
"num_tokens": 2528742.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.69284987449646,
|
|
"epoch": 0.12054044507241257,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999883202552468,
|
|
"loss": 6.7043,
|
|
"mean_token_accuracy": 0.12755436450242996,
|
|
"num_tokens": 2538609.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.828431463241577,
|
|
"epoch": 0.12098198516425292,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999879971822189,
|
|
"loss": 6.684,
|
|
"mean_token_accuracy": 0.11924384832382202,
|
|
"num_tokens": 2547772.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.571238899230957,
|
|
"epoch": 0.12142352525609325,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999876697018038,
|
|
"loss": 6.6582,
|
|
"mean_token_accuracy": 0.12748303934931754,
|
|
"num_tokens": 2556114.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.8254400253295895,
|
|
"epoch": 0.1218650653479336,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999873378140085,
|
|
"loss": 6.7179,
|
|
"mean_token_accuracy": 0.1250918261706829,
|
|
"num_tokens": 2566814.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.614608812332153,
|
|
"epoch": 0.12230660543977394,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999870015188389,
|
|
"loss": 6.6667,
|
|
"mean_token_accuracy": 0.12528225034475327,
|
|
"num_tokens": 2576030.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.711950302124023,
|
|
"epoch": 0.12274814553161427,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999866608163021,
|
|
"loss": 6.689,
|
|
"mean_token_accuracy": 0.12896783277392387,
|
|
"num_tokens": 2585756.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.719356441497803,
|
|
"epoch": 0.12318968562345461,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999863157064045,
|
|
"loss": 6.7433,
|
|
"mean_token_accuracy": 0.12513794749975204,
|
|
"num_tokens": 2595676.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.758820867538452,
|
|
"epoch": 0.12363122571529495,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999859661891529,
|
|
"loss": 6.7445,
|
|
"mean_token_accuracy": 0.12377320975065231,
|
|
"num_tokens": 2606197.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.740264415740967,
|
|
"epoch": 0.12407276580713529,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999856122645543,
|
|
"loss": 6.6646,
|
|
"mean_token_accuracy": 0.12415967881679535,
|
|
"num_tokens": 2615311.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.632203722000122,
|
|
"epoch": 0.12451430589897562,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999852539326154,
|
|
"loss": 6.5799,
|
|
"mean_token_accuracy": 0.1287233628332615,
|
|
"num_tokens": 2624074.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.623841571807861,
|
|
"epoch": 0.12495584599081597,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999848911933434,
|
|
"loss": 6.7141,
|
|
"mean_token_accuracy": 0.13292624652385712,
|
|
"num_tokens": 2633877.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.68168888092041,
|
|
"epoch": 0.1253973860826563,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999845240467453,
|
|
"loss": 6.6526,
|
|
"mean_token_accuracy": 0.12299527004361152,
|
|
"num_tokens": 2643330.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.702590560913086,
|
|
"epoch": 0.12583892617449666,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999841524928282,
|
|
"loss": 6.6348,
|
|
"mean_token_accuracy": 0.12881363406777382,
|
|
"num_tokens": 2652070.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.6873023986816404,
|
|
"epoch": 0.126280466266337,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999837765315997,
|
|
"loss": 6.6718,
|
|
"mean_token_accuracy": 0.12390449419617652,
|
|
"num_tokens": 2660546.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.522921085357666,
|
|
"epoch": 0.12672200635817732,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999833961630669,
|
|
"loss": 6.5631,
|
|
"mean_token_accuracy": 0.13146114349365234,
|
|
"num_tokens": 2669938.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.8073399543762205,
|
|
"epoch": 0.12716354645001765,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999830113872374,
|
|
"loss": 6.6919,
|
|
"mean_token_accuracy": 0.12384787425398827,
|
|
"num_tokens": 2679814.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.547947025299072,
|
|
"epoch": 0.127605086541858,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999826222041186,
|
|
"loss": 6.6066,
|
|
"mean_token_accuracy": 0.1329873576760292,
|
|
"num_tokens": 2688733.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.649660348892212,
|
|
"epoch": 0.12804662663369834,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999822286137182,
|
|
"loss": 6.5858,
|
|
"mean_token_accuracy": 0.1287431985139847,
|
|
"num_tokens": 2697744.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.6083179950714115,
|
|
"epoch": 0.12848816672553867,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999818306160439,
|
|
"loss": 6.5587,
|
|
"mean_token_accuracy": 0.13241407200694083,
|
|
"num_tokens": 2707037.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.707999324798584,
|
|
"epoch": 0.12892970681737903,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999814282111034,
|
|
"loss": 6.7159,
|
|
"mean_token_accuracy": 0.12165460363030434,
|
|
"num_tokens": 2717345.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.706764125823975,
|
|
"epoch": 0.12937124690921936,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999810213989047,
|
|
"loss": 6.6771,
|
|
"mean_token_accuracy": 0.11996371075510978,
|
|
"num_tokens": 2726892.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.606377267837525,
|
|
"epoch": 0.1298127870010597,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999806101794558,
|
|
"loss": 6.6282,
|
|
"mean_token_accuracy": 0.12699908390641212,
|
|
"num_tokens": 2736479.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.772217559814453,
|
|
"epoch": 0.13025432709290002,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999801945527648,
|
|
"loss": 6.6764,
|
|
"mean_token_accuracy": 0.12172863110899926,
|
|
"num_tokens": 2745998.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.60609245300293,
|
|
"epoch": 0.13069586718474038,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999797745188395,
|
|
"loss": 6.6578,
|
|
"mean_token_accuracy": 0.12872695848345755,
|
|
"num_tokens": 2754346.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.701602792739868,
|
|
"epoch": 0.13113740727658071,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999793500776886,
|
|
"loss": 6.6094,
|
|
"mean_token_accuracy": 0.13089651241898537,
|
|
"num_tokens": 2763391.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.633615016937256,
|
|
"epoch": 0.13157894736842105,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999789212293201,
|
|
"loss": 6.669,
|
|
"mean_token_accuracy": 0.12867793068289757,
|
|
"num_tokens": 2772764.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.7330268859863285,
|
|
"epoch": 0.1320204874602614,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999784879737423,
|
|
"loss": 6.7564,
|
|
"mean_token_accuracy": 0.1226215623319149,
|
|
"num_tokens": 2782312.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.67516393661499,
|
|
"epoch": 0.13246202755210174,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999780503109642,
|
|
"loss": 6.6484,
|
|
"mean_token_accuracy": 0.12383122742176056,
|
|
"num_tokens": 2791159.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.610153341293335,
|
|
"epoch": 0.13290356764394207,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999776082409939,
|
|
"loss": 6.4972,
|
|
"mean_token_accuracy": 0.13475448414683341,
|
|
"num_tokens": 2799319.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.595534706115723,
|
|
"epoch": 0.1333451077357824,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999771617638401,
|
|
"loss": 6.6108,
|
|
"mean_token_accuracy": 0.1235594168305397,
|
|
"num_tokens": 2807401.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.707736253738403,
|
|
"epoch": 0.13378664782762276,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999767108795118,
|
|
"loss": 6.661,
|
|
"mean_token_accuracy": 0.1273370273411274,
|
|
"num_tokens": 2817734.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.612422895431519,
|
|
"epoch": 0.1342281879194631,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999762555880176,
|
|
"loss": 6.6679,
|
|
"mean_token_accuracy": 0.12645771428942681,
|
|
"num_tokens": 2828235.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.702644395828247,
|
|
"epoch": 0.13466972801130342,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999757958893666,
|
|
"loss": 6.6872,
|
|
"mean_token_accuracy": 0.12617624551057816,
|
|
"num_tokens": 2837453.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.566971969604492,
|
|
"epoch": 0.13511126810314378,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999753317835677,
|
|
"loss": 6.6608,
|
|
"mean_token_accuracy": 0.12031328305602074,
|
|
"num_tokens": 2847055.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.747590446472168,
|
|
"epoch": 0.1355528081949841,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999748632706299,
|
|
"loss": 6.6333,
|
|
"mean_token_accuracy": 0.12874888330698014,
|
|
"num_tokens": 2857101.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.58017520904541,
|
|
"epoch": 0.13599434828682444,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999743903505626,
|
|
"loss": 6.5493,
|
|
"mean_token_accuracy": 0.12921440675854684,
|
|
"num_tokens": 2866685.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.686506080627441,
|
|
"epoch": 0.13643588837866477,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999739130233749,
|
|
"loss": 6.6483,
|
|
"mean_token_accuracy": 0.12603347525000572,
|
|
"num_tokens": 2876022.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.676938915252686,
|
|
"epoch": 0.13687742847050513,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999734312890761,
|
|
"loss": 6.5654,
|
|
"mean_token_accuracy": 0.1277615949511528,
|
|
"num_tokens": 2885560.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.552069282531738,
|
|
"epoch": 0.13731896856234546,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999729451476757,
|
|
"loss": 6.6064,
|
|
"mean_token_accuracy": 0.1283570185303688,
|
|
"num_tokens": 2894686.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.67106065750122,
|
|
"epoch": 0.1377605086541858,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999724545991835,
|
|
"loss": 6.6135,
|
|
"mean_token_accuracy": 0.13272278234362603,
|
|
"num_tokens": 2904390.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.641124057769775,
|
|
"epoch": 0.13820204874602615,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999719596436086,
|
|
"loss": 6.6751,
|
|
"mean_token_accuracy": 0.12594080567359925,
|
|
"num_tokens": 2913311.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.69909701347351,
|
|
"epoch": 0.13864358883786648,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999714602809611,
|
|
"loss": 6.5549,
|
|
"mean_token_accuracy": 0.13006579801440238,
|
|
"num_tokens": 2923196.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.554364538192749,
|
|
"epoch": 0.1390851289297068,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999709565112506,
|
|
"loss": 6.6195,
|
|
"mean_token_accuracy": 0.12607380226254464,
|
|
"num_tokens": 2932813.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.618018245697021,
|
|
"epoch": 0.13952666902154715,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000499970448334487,
|
|
"loss": 6.5517,
|
|
"mean_token_accuracy": 0.1182281494140625,
|
|
"num_tokens": 2942694.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.567258405685425,
|
|
"epoch": 0.1399682091133875,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999699357506803,
|
|
"loss": 6.4833,
|
|
"mean_token_accuracy": 0.13758974224328996,
|
|
"num_tokens": 2950932.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.665033578872681,
|
|
"epoch": 0.14040974920522784,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999694187598406,
|
|
"loss": 6.5959,
|
|
"mean_token_accuracy": 0.13096074685454367,
|
|
"num_tokens": 2960102.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.5607537746429445,
|
|
"epoch": 0.14085128929706817,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004999688973619777,
|
|
"loss": 6.5666,
|
|
"mean_token_accuracy": 0.12616149932146073,
|
|
"num_tokens": 2969968.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.565665245056152,
|
|
"epoch": 0.14129282938890853,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999683715571022,
|
|
"loss": 6.5376,
|
|
"mean_token_accuracy": 0.13295454159379005,
|
|
"num_tokens": 2978880.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.6195861339569095,
|
|
"epoch": 0.14173436948074886,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999678413452242,
|
|
"loss": 6.5692,
|
|
"mean_token_accuracy": 0.1263133704662323,
|
|
"num_tokens": 2988369.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.635553073883057,
|
|
"epoch": 0.1421759095725892,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999673067263542,
|
|
"loss": 6.6144,
|
|
"mean_token_accuracy": 0.12454515546560288,
|
|
"num_tokens": 2997070.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.616920137405396,
|
|
"epoch": 0.14261744966442952,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999667677005026,
|
|
"loss": 6.5553,
|
|
"mean_token_accuracy": 0.13325243517756463,
|
|
"num_tokens": 3006547.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.574477910995483,
|
|
"epoch": 0.14305898975626988,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999662242676799,
|
|
"loss": 6.5792,
|
|
"mean_token_accuracy": 0.12996308133006096,
|
|
"num_tokens": 3015821.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.618298435211182,
|
|
"epoch": 0.1435005298481102,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999656764278968,
|
|
"loss": 6.5491,
|
|
"mean_token_accuracy": 0.13459995537996292,
|
|
"num_tokens": 3024750.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.57071361541748,
|
|
"epoch": 0.14394206993995054,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999651241811642,
|
|
"loss": 6.5191,
|
|
"mean_token_accuracy": 0.13164762035012245,
|
|
"num_tokens": 3033345.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.575192975997925,
|
|
"epoch": 0.1443836100317909,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999645675274925,
|
|
"loss": 6.4902,
|
|
"mean_token_accuracy": 0.13214875981211663,
|
|
"num_tokens": 3042060.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.564555883407593,
|
|
"epoch": 0.14482515012363123,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999640064668931,
|
|
"loss": 6.6404,
|
|
"mean_token_accuracy": 0.12763455510139465,
|
|
"num_tokens": 3052490.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.715090990066528,
|
|
"epoch": 0.14526669021547156,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999634409993766,
|
|
"loss": 6.6155,
|
|
"mean_token_accuracy": 0.12552304938435555,
|
|
"num_tokens": 3061934.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.513729286193848,
|
|
"epoch": 0.1457082303073119,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999628711249544,
|
|
"loss": 6.5198,
|
|
"mean_token_accuracy": 0.13856697753071784,
|
|
"num_tokens": 3070890.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.682709264755249,
|
|
"epoch": 0.14614977039915225,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999622968436373,
|
|
"loss": 6.5315,
|
|
"mean_token_accuracy": 0.128802065551281,
|
|
"num_tokens": 3079933.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.47434253692627,
|
|
"epoch": 0.14659131049099258,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999617181554369,
|
|
"loss": 6.6016,
|
|
"mean_token_accuracy": 0.13180896043777465,
|
|
"num_tokens": 3089910.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.710703754425049,
|
|
"epoch": 0.1470328505828329,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999611350603643,
|
|
"loss": 6.5781,
|
|
"mean_token_accuracy": 0.13360659033060074,
|
|
"num_tokens": 3098676.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.531248569488525,
|
|
"epoch": 0.14747439067467327,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000499960547558431,
|
|
"loss": 6.69,
|
|
"mean_token_accuracy": 0.12630268037319184,
|
|
"num_tokens": 3109055.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.600419187545777,
|
|
"epoch": 0.1479159307665136,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999599556496486,
|
|
"loss": 6.5276,
|
|
"mean_token_accuracy": 0.13437643125653267,
|
|
"num_tokens": 3117517.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.542807102203369,
|
|
"epoch": 0.14835747085835393,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999593593340286,
|
|
"loss": 6.5288,
|
|
"mean_token_accuracy": 0.13295082822442056,
|
|
"num_tokens": 3126606.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.650139379501343,
|
|
"epoch": 0.14879901095019427,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999587586115826,
|
|
"loss": 6.551,
|
|
"mean_token_accuracy": 0.13533952906727792,
|
|
"num_tokens": 3135444.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.524611854553223,
|
|
"epoch": 0.14924055104203462,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999581534823226,
|
|
"loss": 6.4885,
|
|
"mean_token_accuracy": 0.13537175804376603,
|
|
"num_tokens": 3144967.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.606992101669311,
|
|
"epoch": 0.14968209113387496,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999575439462601,
|
|
"loss": 6.5042,
|
|
"mean_token_accuracy": 0.1286872260272503,
|
|
"num_tokens": 3153898.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.504314661026001,
|
|
"epoch": 0.1501236312257153,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999569300034075,
|
|
"loss": 6.5644,
|
|
"mean_token_accuracy": 0.12845081239938735,
|
|
"num_tokens": 3162341.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.507896661758423,
|
|
"epoch": 0.15056517131755565,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999563116537764,
|
|
"loss": 6.3613,
|
|
"mean_token_accuracy": 0.14074554145336152,
|
|
"num_tokens": 3171420.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.5167927742004395,
|
|
"epoch": 0.15100671140939598,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999556888973792,
|
|
"loss": 6.4165,
|
|
"mean_token_accuracy": 0.12971381321549416,
|
|
"num_tokens": 3180221.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.600985956192017,
|
|
"epoch": 0.1514482515012363,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999550617342279,
|
|
"loss": 6.5615,
|
|
"mean_token_accuracy": 0.12731998935341834,
|
|
"num_tokens": 3189156.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.536636877059936,
|
|
"epoch": 0.15188979159307664,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499954430164335,
|
|
"loss": 6.4905,
|
|
"mean_token_accuracy": 0.1309817299246788,
|
|
"num_tokens": 3199870.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.541285371780395,
|
|
"epoch": 0.152331331684917,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999537941877127,
|
|
"loss": 6.4727,
|
|
"mean_token_accuracy": 0.13528374806046486,
|
|
"num_tokens": 3208815.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.487531661987305,
|
|
"epoch": 0.15277287177675733,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999531538043735,
|
|
"loss": 6.5498,
|
|
"mean_token_accuracy": 0.12755031064152716,
|
|
"num_tokens": 3218692.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.591389942169189,
|
|
"epoch": 0.15321441186859766,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999525090143298,
|
|
"loss": 6.5177,
|
|
"mean_token_accuracy": 0.13854537010192872,
|
|
"num_tokens": 3227604.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.57657470703125,
|
|
"epoch": 0.15365595196043802,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999518598175946,
|
|
"loss": 6.5291,
|
|
"mean_token_accuracy": 0.13887950852513314,
|
|
"num_tokens": 3237912.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.481386184692383,
|
|
"epoch": 0.15409749205227835,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999512062141805,
|
|
"loss": 6.3914,
|
|
"mean_token_accuracy": 0.1368810623884201,
|
|
"num_tokens": 3246003.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.51813645362854,
|
|
"epoch": 0.15453903214411868,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999505482040999,
|
|
"loss": 6.5332,
|
|
"mean_token_accuracy": 0.13183941319584846,
|
|
"num_tokens": 3256363.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.553296089172363,
|
|
"epoch": 0.154980572235959,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999498857873662,
|
|
"loss": 6.5312,
|
|
"mean_token_accuracy": 0.1321260340511799,
|
|
"num_tokens": 3265822.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.417472457885742,
|
|
"epoch": 0.15542211232779937,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999492189639921,
|
|
"loss": 6.424,
|
|
"mean_token_accuracy": 0.13568418473005295,
|
|
"num_tokens": 3274614.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.586168956756592,
|
|
"epoch": 0.1558636524196397,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999485477339907,
|
|
"loss": 6.5175,
|
|
"mean_token_accuracy": 0.12809595912694932,
|
|
"num_tokens": 3283800.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.485697317123413,
|
|
"epoch": 0.15630519251148003,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999478720973753,
|
|
"loss": 6.4297,
|
|
"mean_token_accuracy": 0.13804005309939385,
|
|
"num_tokens": 3293221.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.522076082229614,
|
|
"epoch": 0.1567467326033204,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000499947192054159,
|
|
"loss": 6.5876,
|
|
"mean_token_accuracy": 0.12661009952425956,
|
|
"num_tokens": 3302852.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.621862745285034,
|
|
"epoch": 0.15718827269516072,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499946507604355,
|
|
"loss": 6.4486,
|
|
"mean_token_accuracy": 0.1398267850279808,
|
|
"num_tokens": 3311752.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.4172522068023685,
|
|
"epoch": 0.15762981278700106,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499945818747977,
|
|
"loss": 6.5267,
|
|
"mean_token_accuracy": 0.13450878411531447,
|
|
"num_tokens": 3321339.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.623373413085938,
|
|
"epoch": 0.1580713528788414,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999451254850383,
|
|
"loss": 6.5377,
|
|
"mean_token_accuracy": 0.13095242530107498,
|
|
"num_tokens": 3330269.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.57389874458313,
|
|
"epoch": 0.15851289297068175,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999444278155525,
|
|
"loss": 6.4619,
|
|
"mean_token_accuracy": 0.13360615670681,
|
|
"num_tokens": 3340770.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.484994888305664,
|
|
"epoch": 0.15895443306252208,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999437257395333,
|
|
"loss": 6.5612,
|
|
"mean_token_accuracy": 0.1361308626830578,
|
|
"num_tokens": 3349976.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.629813623428345,
|
|
"epoch": 0.1593959731543624,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999430192569944,
|
|
"loss": 6.5942,
|
|
"mean_token_accuracy": 0.126756651699543,
|
|
"num_tokens": 3359764.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.493142127990723,
|
|
"epoch": 0.15983751324620277,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999423083679498,
|
|
"loss": 6.4589,
|
|
"mean_token_accuracy": 0.1307877480983734,
|
|
"num_tokens": 3369939.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.5387725830078125,
|
|
"epoch": 0.1602790533380431,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999415930724133,
|
|
"loss": 6.6238,
|
|
"mean_token_accuracy": 0.12685370370745658,
|
|
"num_tokens": 3381326.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.630081462860107,
|
|
"epoch": 0.16072059342988343,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999408733703988,
|
|
"loss": 6.5692,
|
|
"mean_token_accuracy": 0.12697295919060708,
|
|
"num_tokens": 3391016.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.515750217437744,
|
|
"epoch": 0.16116213352172376,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999401492619207,
|
|
"loss": 6.3301,
|
|
"mean_token_accuracy": 0.1411547876894474,
|
|
"num_tokens": 3400294.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.519891738891602,
|
|
"epoch": 0.16160367361356412,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999394207469928,
|
|
"loss": 6.5452,
|
|
"mean_token_accuracy": 0.12329790964722634,
|
|
"num_tokens": 3409685.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.627039623260498,
|
|
"epoch": 0.16204521370540445,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999386878256297,
|
|
"loss": 6.4467,
|
|
"mean_token_accuracy": 0.12930482253432274,
|
|
"num_tokens": 3418946.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.345337200164795,
|
|
"epoch": 0.16248675379724478,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999379504978457,
|
|
"loss": 6.4523,
|
|
"mean_token_accuracy": 0.1312091715633869,
|
|
"num_tokens": 3428245.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.535007095336914,
|
|
"epoch": 0.16292829388908514,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499937208763655,
|
|
"loss": 6.4627,
|
|
"mean_token_accuracy": 0.13302566036581992,
|
|
"num_tokens": 3437580.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.5879792213439945,
|
|
"epoch": 0.16336983398092547,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999364626230724,
|
|
"loss": 6.4411,
|
|
"mean_token_accuracy": 0.1374020665884018,
|
|
"num_tokens": 3446393.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.394599485397339,
|
|
"epoch": 0.1638113740727658,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999357120761124,
|
|
"loss": 6.5361,
|
|
"mean_token_accuracy": 0.13167317807674409,
|
|
"num_tokens": 3455956.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.6326652526855465,
|
|
"epoch": 0.16425291416460613,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999349571227898,
|
|
"loss": 6.5281,
|
|
"mean_token_accuracy": 0.13483646959066392,
|
|
"num_tokens": 3465722.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.507793617248535,
|
|
"epoch": 0.1646944542564465,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999341977631193,
|
|
"loss": 6.5072,
|
|
"mean_token_accuracy": 0.12958121970295905,
|
|
"num_tokens": 3475467.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.5095072269439695,
|
|
"epoch": 0.16513599434828682,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999334339971157,
|
|
"loss": 6.3989,
|
|
"mean_token_accuracy": 0.13358662649989128,
|
|
"num_tokens": 3484931.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.522356176376343,
|
|
"epoch": 0.16557753444012716,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999326658247942,
|
|
"loss": 6.5123,
|
|
"mean_token_accuracy": 0.13453342840075494,
|
|
"num_tokens": 3494001.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.49297513961792,
|
|
"epoch": 0.16601907453196751,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999318932461696,
|
|
"loss": 6.4281,
|
|
"mean_token_accuracy": 0.1413221076130867,
|
|
"num_tokens": 3503021.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.447078704833984,
|
|
"epoch": 0.16646061462380785,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004999311162612571,
|
|
"loss": 6.4761,
|
|
"mean_token_accuracy": 0.13532513231039048,
|
|
"num_tokens": 3513005.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.60153341293335,
|
|
"epoch": 0.16690215471564818,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499930334870072,
|
|
"loss": 6.5511,
|
|
"mean_token_accuracy": 0.13551064282655717,
|
|
"num_tokens": 3523219.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.511942768096924,
|
|
"epoch": 0.1673436948074885,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999295490726296,
|
|
"loss": 6.5013,
|
|
"mean_token_accuracy": 0.130006618052721,
|
|
"num_tokens": 3532917.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.566938781738282,
|
|
"epoch": 0.16778523489932887,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999287588689453,
|
|
"loss": 6.496,
|
|
"mean_token_accuracy": 0.1396396428346634,
|
|
"num_tokens": 3542845.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.521204996109009,
|
|
"epoch": 0.1682267749911692,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999279642590344,
|
|
"loss": 6.5214,
|
|
"mean_token_accuracy": 0.13577667698264123,
|
|
"num_tokens": 3552126.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.5692566394805905,
|
|
"epoch": 0.16866831508300953,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999271652429127,
|
|
"loss": 6.5265,
|
|
"mean_token_accuracy": 0.127306517213583,
|
|
"num_tokens": 3561254.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.499909734725952,
|
|
"epoch": 0.1691098551748499,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999263618205958,
|
|
"loss": 6.3733,
|
|
"mean_token_accuracy": 0.14404519647359848,
|
|
"num_tokens": 3569781.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.53024697303772,
|
|
"epoch": 0.16955139526669022,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999255539920993,
|
|
"loss": 6.4874,
|
|
"mean_token_accuracy": 0.13062172681093215,
|
|
"num_tokens": 3579664.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.532351350784301,
|
|
"epoch": 0.16999293535853055,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999247417574391,
|
|
"loss": 6.526,
|
|
"mean_token_accuracy": 0.13393994346261023,
|
|
"num_tokens": 3588671.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.538648414611816,
|
|
"epoch": 0.17043447545037088,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999239251166312,
|
|
"loss": 6.4128,
|
|
"mean_token_accuracy": 0.14050136953592302,
|
|
"num_tokens": 3597656.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.45431957244873,
|
|
"epoch": 0.17087601554221124,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999231040696914,
|
|
"loss": 6.4342,
|
|
"mean_token_accuracy": 0.14062088206410409,
|
|
"num_tokens": 3608017.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.46071605682373,
|
|
"epoch": 0.17131755563405157,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999222786166361,
|
|
"loss": 6.5009,
|
|
"mean_token_accuracy": 0.13390257805585862,
|
|
"num_tokens": 3618189.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.564482021331787,
|
|
"epoch": 0.1717590957258919,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999214487574812,
|
|
"loss": 6.463,
|
|
"mean_token_accuracy": 0.13240221589803697,
|
|
"num_tokens": 3627211.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.480297803878784,
|
|
"epoch": 0.17220063581773226,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999206144922431,
|
|
"loss": 6.4038,
|
|
"mean_token_accuracy": 0.13147775381803511,
|
|
"num_tokens": 3636781.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.488903188705445,
|
|
"epoch": 0.1726421759095726,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499919775820938,
|
|
"loss": 6.4913,
|
|
"mean_token_accuracy": 0.14189594313502313,
|
|
"num_tokens": 3644891.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.561635589599609,
|
|
"epoch": 0.17308371600141292,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999189327435825,
|
|
"loss": 6.5195,
|
|
"mean_token_accuracy": 0.13184107840061188,
|
|
"num_tokens": 3655477.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.517252016067505,
|
|
"epoch": 0.17352525609325326,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999180852601929,
|
|
"loss": 6.5422,
|
|
"mean_token_accuracy": 0.13338224366307258,
|
|
"num_tokens": 3664542.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.498584127426147,
|
|
"epoch": 0.17396679618509361,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000499917233370786,
|
|
"loss": 6.4452,
|
|
"mean_token_accuracy": 0.12930241152644156,
|
|
"num_tokens": 3673806.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.516369295120239,
|
|
"epoch": 0.17440833627693395,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999163770753784,
|
|
"loss": 6.4487,
|
|
"mean_token_accuracy": 0.13249206990003587,
|
|
"num_tokens": 3683238.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.473311996459961,
|
|
"epoch": 0.17484987636877428,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004999155163739869,
|
|
"loss": 6.4228,
|
|
"mean_token_accuracy": 0.13501294553279877,
|
|
"num_tokens": 3692161.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.512854528427124,
|
|
"epoch": 0.17529141646061464,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999146512666284,
|
|
"loss": 6.4675,
|
|
"mean_token_accuracy": 0.13027839213609696,
|
|
"num_tokens": 3701431.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.468549728393555,
|
|
"epoch": 0.17573295655245497,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999137817533197,
|
|
"loss": 6.419,
|
|
"mean_token_accuracy": 0.14170578867197037,
|
|
"num_tokens": 3710963.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.4296314239501955,
|
|
"epoch": 0.1761744966442953,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999129078340779,
|
|
"loss": 6.4214,
|
|
"mean_token_accuracy": 0.13799419105052949,
|
|
"num_tokens": 3720177.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.4377001285552975,
|
|
"epoch": 0.17661603673613563,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999120295089202,
|
|
"loss": 6.3718,
|
|
"mean_token_accuracy": 0.14307373464107515,
|
|
"num_tokens": 3728123.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.573485851287842,
|
|
"epoch": 0.177057576827976,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999111467778639,
|
|
"loss": 6.509,
|
|
"mean_token_accuracy": 0.13284810408949851,
|
|
"num_tokens": 3736869.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.46274700164795,
|
|
"epoch": 0.17749911691981632,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000499910259640926,
|
|
"loss": 6.3788,
|
|
"mean_token_accuracy": 0.13863728046417237,
|
|
"num_tokens": 3745830.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.421447610855102,
|
|
"epoch": 0.17794065701165665,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499909368098124,
|
|
"loss": 6.4166,
|
|
"mean_token_accuracy": 0.13812872543931007,
|
|
"num_tokens": 3755019.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.508076333999634,
|
|
"epoch": 0.178382197103497,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999084721494754,
|
|
"loss": 6.3968,
|
|
"mean_token_accuracy": 0.1333928920328617,
|
|
"num_tokens": 3764814.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.345089483261108,
|
|
"epoch": 0.17882373719533734,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999075717949978,
|
|
"loss": 6.3581,
|
|
"mean_token_accuracy": 0.14092704504728318,
|
|
"num_tokens": 3774258.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.421577262878418,
|
|
"epoch": 0.17926527728717767,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999066670347089,
|
|
"loss": 6.3978,
|
|
"mean_token_accuracy": 0.13964474871754645,
|
|
"num_tokens": 3783961.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.40748291015625,
|
|
"epoch": 0.179706817379018,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999057578686261,
|
|
"loss": 6.363,
|
|
"mean_token_accuracy": 0.13734552562236785,
|
|
"num_tokens": 3792417.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.495989513397217,
|
|
"epoch": 0.18014835747085836,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999048442967675,
|
|
"loss": 6.3557,
|
|
"mean_token_accuracy": 0.13433733358979225,
|
|
"num_tokens": 3801479.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.37765908241272,
|
|
"epoch": 0.1805898975626987,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999039263191508,
|
|
"loss": 6.4254,
|
|
"mean_token_accuracy": 0.13214251399040222,
|
|
"num_tokens": 3810799.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.491030645370484,
|
|
"epoch": 0.18103143765453902,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999030039357943,
|
|
"loss": 6.462,
|
|
"mean_token_accuracy": 0.13567366376519202,
|
|
"num_tokens": 3820966.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.463741731643677,
|
|
"epoch": 0.18147297774637938,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999020771467158,
|
|
"loss": 6.4423,
|
|
"mean_token_accuracy": 0.1354211077094078,
|
|
"num_tokens": 3829247.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.572631978988648,
|
|
"epoch": 0.1819145178382197,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999011459519335,
|
|
"loss": 6.3891,
|
|
"mean_token_accuracy": 0.1375864826142788,
|
|
"num_tokens": 3838114.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.325645637512207,
|
|
"epoch": 0.18235605793006004,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999002103514655,
|
|
"loss": 6.4586,
|
|
"mean_token_accuracy": 0.13582065477967262,
|
|
"num_tokens": 3848075.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.570796966552734,
|
|
"epoch": 0.18279759802190038,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998992703453304,
|
|
"loss": 6.4987,
|
|
"mean_token_accuracy": 0.13409382104873657,
|
|
"num_tokens": 3857934.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.490766811370849,
|
|
"epoch": 0.18323913811374073,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998983259335466,
|
|
"loss": 6.3433,
|
|
"mean_token_accuracy": 0.14234771504998206,
|
|
"num_tokens": 3866707.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.456926107406616,
|
|
"epoch": 0.18368067820558107,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998973771161324,
|
|
"loss": 6.4192,
|
|
"mean_token_accuracy": 0.13825157508254052,
|
|
"num_tokens": 3875233.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.485457134246826,
|
|
"epoch": 0.1841222182974214,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998964238931065,
|
|
"loss": 6.4138,
|
|
"mean_token_accuracy": 0.13973441645503043,
|
|
"num_tokens": 3885173.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.477391719818115,
|
|
"epoch": 0.18456375838926176,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998954662644876,
|
|
"loss": 6.3774,
|
|
"mean_token_accuracy": 0.13736600577831268,
|
|
"num_tokens": 3894198.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.421681976318359,
|
|
"epoch": 0.1850052984811021,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998945042302943,
|
|
"loss": 6.373,
|
|
"mean_token_accuracy": 0.13609152138233185,
|
|
"num_tokens": 3904076.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.480407571792602,
|
|
"epoch": 0.18544683857294242,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998935377905457,
|
|
"loss": 6.4802,
|
|
"mean_token_accuracy": 0.1311386428773403,
|
|
"num_tokens": 3913204.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.535258817672729,
|
|
"epoch": 0.18588837866478275,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998925669452605,
|
|
"loss": 6.4323,
|
|
"mean_token_accuracy": 0.13564323037862777,
|
|
"num_tokens": 3922148.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.398992824554443,
|
|
"epoch": 0.1863299187566231,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998915916944579,
|
|
"loss": 6.4129,
|
|
"mean_token_accuracy": 0.13676537424325944,
|
|
"num_tokens": 3931333.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.4503209590911865,
|
|
"epoch": 0.18677145884846344,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998906120381568,
|
|
"loss": 6.374,
|
|
"mean_token_accuracy": 0.14027760475873946,
|
|
"num_tokens": 3941061.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.508623027801514,
|
|
"epoch": 0.18721299894030377,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998896279763766,
|
|
"loss": 6.4722,
|
|
"mean_token_accuracy": 0.13354425132274628,
|
|
"num_tokens": 3950075.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.476150178909302,
|
|
"epoch": 0.18765453903214413,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998886395091365,
|
|
"loss": 6.3316,
|
|
"mean_token_accuracy": 0.13695699274539946,
|
|
"num_tokens": 3958885.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.3462131977081295,
|
|
"epoch": 0.18809607912398446,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998876466364559,
|
|
"loss": 6.4256,
|
|
"mean_token_accuracy": 0.13821618929505347,
|
|
"num_tokens": 3968218.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.443255281448364,
|
|
"epoch": 0.1885376192158248,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004998866493583541,
|
|
"loss": 6.3606,
|
|
"mean_token_accuracy": 0.1383839502930641,
|
|
"num_tokens": 3977435.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.356647872924805,
|
|
"epoch": 0.18897915930766512,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998856476748509,
|
|
"loss": 6.3498,
|
|
"mean_token_accuracy": 0.1412102162837982,
|
|
"num_tokens": 3986608.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.431683588027954,
|
|
"epoch": 0.18942069939950548,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998846415859656,
|
|
"loss": 6.3585,
|
|
"mean_token_accuracy": 0.14042109996080399,
|
|
"num_tokens": 3996087.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.444839954376221,
|
|
"epoch": 0.1898622394913458,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998836310917182,
|
|
"loss": 6.4002,
|
|
"mean_token_accuracy": 0.13422135785222053,
|
|
"num_tokens": 4006257.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.52928204536438,
|
|
"epoch": 0.19030377958318614,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998826161921282,
|
|
"loss": 6.4156,
|
|
"mean_token_accuracy": 0.1421043999493122,
|
|
"num_tokens": 4015904.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.383703994750976,
|
|
"epoch": 0.1907453196750265,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998815968872157,
|
|
"loss": 6.4115,
|
|
"mean_token_accuracy": 0.13542449548840524,
|
|
"num_tokens": 4025417.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.363600587844848,
|
|
"epoch": 0.19118685976686683,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998805731770007,
|
|
"loss": 6.2801,
|
|
"mean_token_accuracy": 0.15531659871339798,
|
|
"num_tokens": 4035181.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.47420859336853,
|
|
"epoch": 0.19162839985870717,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499879545061503,
|
|
"loss": 6.4466,
|
|
"mean_token_accuracy": 0.13034741580486298,
|
|
"num_tokens": 4045112.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.513006162643433,
|
|
"epoch": 0.1920699399505475,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998785125407432,
|
|
"loss": 6.5509,
|
|
"mean_token_accuracy": 0.12224270775914192,
|
|
"num_tokens": 4054566.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.533349084854126,
|
|
"epoch": 0.19251148004238786,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499877475614741,
|
|
"loss": 6.3601,
|
|
"mean_token_accuracy": 0.1362767845392227,
|
|
"num_tokens": 4063960.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.30982141494751,
|
|
"epoch": 0.1929530201342282,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998764342835169,
|
|
"loss": 6.3077,
|
|
"mean_token_accuracy": 0.14436172991991042,
|
|
"num_tokens": 4072620.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.325711584091186,
|
|
"epoch": 0.19339456022606852,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998753885470915,
|
|
"loss": 6.2856,
|
|
"mean_token_accuracy": 0.1410072512924671,
|
|
"num_tokens": 4081590.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.443258333206177,
|
|
"epoch": 0.19383610031790888,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998743384054851,
|
|
"loss": 6.4024,
|
|
"mean_token_accuracy": 0.14100263118743897,
|
|
"num_tokens": 4090758.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.343490505218506,
|
|
"epoch": 0.1942776404097492,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998732838587183,
|
|
"loss": 6.2339,
|
|
"mean_token_accuracy": 0.14560552909970284,
|
|
"num_tokens": 4099281.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.428974056243897,
|
|
"epoch": 0.19471918050158954,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998722249068118,
|
|
"loss": 6.3508,
|
|
"mean_token_accuracy": 0.13685052916407586,
|
|
"num_tokens": 4108953.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.399385118484497,
|
|
"epoch": 0.19516072059342987,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998711615497863,
|
|
"loss": 6.453,
|
|
"mean_token_accuracy": 0.13157615587115287,
|
|
"num_tokens": 4118799.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.466229009628296,
|
|
"epoch": 0.19560226068527023,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998700937876626,
|
|
"loss": 6.4178,
|
|
"mean_token_accuracy": 0.13792671710252763,
|
|
"num_tokens": 4127862.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.433918428421021,
|
|
"epoch": 0.19604380077711056,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998690216204615,
|
|
"loss": 6.4809,
|
|
"mean_token_accuracy": 0.12709898576140405,
|
|
"num_tokens": 4139029.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.472035026550293,
|
|
"epoch": 0.1964853408689509,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998679450482043,
|
|
"loss": 6.3062,
|
|
"mean_token_accuracy": 0.14617246389389038,
|
|
"num_tokens": 4148257.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.365042209625244,
|
|
"epoch": 0.19692688096079125,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499866864070912,
|
|
"loss": 6.2528,
|
|
"mean_token_accuracy": 0.13704810217022895,
|
|
"num_tokens": 4157626.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.411515808105468,
|
|
"epoch": 0.19736842105263158,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998657786886056,
|
|
"loss": 6.4857,
|
|
"mean_token_accuracy": 0.13319113329052926,
|
|
"num_tokens": 4166804.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.512064790725708,
|
|
"epoch": 0.1978099611444719,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998646889013066,
|
|
"loss": 6.3594,
|
|
"mean_token_accuracy": 0.1408704087138176,
|
|
"num_tokens": 4175701.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.454934692382812,
|
|
"epoch": 0.19825150123631224,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998635947090362,
|
|
"loss": 6.4385,
|
|
"mean_token_accuracy": 0.13829853162169456,
|
|
"num_tokens": 4184711.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.43667516708374,
|
|
"epoch": 0.1986930413281526,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004998624961118158,
|
|
"loss": 6.4075,
|
|
"mean_token_accuracy": 0.1378791533410549,
|
|
"num_tokens": 4193931.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.431211376190186,
|
|
"epoch": 0.19913458141999293,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499861393109667,
|
|
"loss": 6.2785,
|
|
"mean_token_accuracy": 0.1496691472828388,
|
|
"num_tokens": 4203000.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.271931219100952,
|
|
"epoch": 0.19957612151183327,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998602857026114,
|
|
"loss": 6.3027,
|
|
"mean_token_accuracy": 0.14634378403425216,
|
|
"num_tokens": 4211977.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.447495746612549,
|
|
"epoch": 0.20001766160367362,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998591738906708,
|
|
"loss": 6.3856,
|
|
"mean_token_accuracy": 0.14466411918401717,
|
|
"num_tokens": 4220375.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.450564670562744,
|
|
"epoch": 0.20045920169551396,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998580576738668,
|
|
"loss": 6.3736,
|
|
"mean_token_accuracy": 0.13617569133639335,
|
|
"num_tokens": 4230506.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.417342710494995,
|
|
"epoch": 0.2009007417873543,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998569370522213,
|
|
"loss": 6.3418,
|
|
"mean_token_accuracy": 0.1376570299267769,
|
|
"num_tokens": 4240235.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.406169795989991,
|
|
"epoch": 0.20134228187919462,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998558120257563,
|
|
"loss": 6.3963,
|
|
"mean_token_accuracy": 0.1375265248119831,
|
|
"num_tokens": 4249775.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.394922590255737,
|
|
"epoch": 0.20178382197103498,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998546825944938,
|
|
"loss": 6.3459,
|
|
"mean_token_accuracy": 0.13671498522162437,
|
|
"num_tokens": 4258523.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.455841398239135,
|
|
"epoch": 0.2022253620628753,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000499853548758456,
|
|
"loss": 6.348,
|
|
"mean_token_accuracy": 0.13951429352164268,
|
|
"num_tokens": 4267683.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.456370782852173,
|
|
"epoch": 0.20266690215471564,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000499852410517665,
|
|
"loss": 6.4251,
|
|
"mean_token_accuracy": 0.13333277478814126,
|
|
"num_tokens": 4276903.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.424126148223877,
|
|
"epoch": 0.203108442246556,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004998512678721431,
|
|
"loss": 6.3582,
|
|
"mean_token_accuracy": 0.13599715530872344,
|
|
"num_tokens": 4287257.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.428092765808105,
|
|
"epoch": 0.20354998233839633,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499850120821913,
|
|
"loss": 6.3942,
|
|
"mean_token_accuracy": 0.1357932858169079,
|
|
"num_tokens": 4297345.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.505991411209107,
|
|
"epoch": 0.20399152243023666,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004998489693669967,
|
|
"loss": 6.3373,
|
|
"mean_token_accuracy": 0.1377499908208847,
|
|
"num_tokens": 4306533.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.375578260421753,
|
|
"epoch": 0.204433062522077,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000499847813507417,
|
|
"loss": 6.404,
|
|
"mean_token_accuracy": 0.1373721919953823,
|
|
"num_tokens": 4316338.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.3910376071929935,
|
|
"epoch": 0.20487460261391735,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998466532431966,
|
|
"loss": 6.3591,
|
|
"mean_token_accuracy": 0.13833511546254157,
|
|
"num_tokens": 4326585.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.526772165298462,
|
|
"epoch": 0.20531614270575768,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998454885743581,
|
|
"loss": 6.4797,
|
|
"mean_token_accuracy": 0.13367427811026572,
|
|
"num_tokens": 4336490.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.390794658660889,
|
|
"epoch": 0.205757682797598,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998443195009242,
|
|
"loss": 6.3557,
|
|
"mean_token_accuracy": 0.13690442964434624,
|
|
"num_tokens": 4346264.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.416228103637695,
|
|
"epoch": 0.20619922288943837,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004998431460229182,
|
|
"loss": 6.3522,
|
|
"mean_token_accuracy": 0.14161566868424416,
|
|
"num_tokens": 4355102.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.50619740486145,
|
|
"epoch": 0.2066407629812787,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004998419681403627,
|
|
"loss": 6.5122,
|
|
"mean_token_accuracy": 0.13018345609307289,
|
|
"num_tokens": 4365569.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.376383638381958,
|
|
"epoch": 0.20708230307311903,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998407858532809,
|
|
"loss": 6.3474,
|
|
"mean_token_accuracy": 0.14445040747523308,
|
|
"num_tokens": 4375437.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.404212570190429,
|
|
"epoch": 0.20752384316495937,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000499839599161696,
|
|
"loss": 6.342,
|
|
"mean_token_accuracy": 0.14488178491592407,
|
|
"num_tokens": 4384420.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.3863624095916744,
|
|
"epoch": 0.20796538325679972,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998384080656314,
|
|
"loss": 6.3048,
|
|
"mean_token_accuracy": 0.14010295867919922,
|
|
"num_tokens": 4393730.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.41755633354187,
|
|
"epoch": 0.20840692334864006,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00049983721256511,
|
|
"loss": 6.2839,
|
|
"mean_token_accuracy": 0.14398279786109924,
|
|
"num_tokens": 4402731.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.356699991226196,
|
|
"epoch": 0.2088484634404804,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998360126601556,
|
|
"loss": 6.3451,
|
|
"mean_token_accuracy": 0.13590056598186492,
|
|
"num_tokens": 4411606.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.3600341320037845,
|
|
"epoch": 0.20929000353232075,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998348083507916,
|
|
"loss": 6.4137,
|
|
"mean_token_accuracy": 0.14537313282489778,
|
|
"num_tokens": 4421685.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.469741773605347,
|
|
"epoch": 0.20973154362416108,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998335996370416,
|
|
"loss": 6.3762,
|
|
"mean_token_accuracy": 0.14291262701153756,
|
|
"num_tokens": 4431765.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.3404137134552006,
|
|
"epoch": 0.2101730837160014,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998323865189291,
|
|
"loss": 6.3491,
|
|
"mean_token_accuracy": 0.13587618842720986,
|
|
"num_tokens": 4441191.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.441226148605347,
|
|
"epoch": 0.21061462380784174,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004998311689964781,
|
|
"loss": 6.4262,
|
|
"mean_token_accuracy": 0.13476165607571602,
|
|
"num_tokens": 4450156.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.482025623321533,
|
|
"epoch": 0.2110561638996821,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004998299470697125,
|
|
"loss": 6.4254,
|
|
"mean_token_accuracy": 0.13929163217544555,
|
|
"num_tokens": 4459466.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.4643927097320555,
|
|
"epoch": 0.21149770399152243,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998287207386559,
|
|
"loss": 6.4186,
|
|
"mean_token_accuracy": 0.13592114597558974,
|
|
"num_tokens": 4468539.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.458880662918091,
|
|
"epoch": 0.21193924408336276,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004998274900033326,
|
|
"loss": 6.3128,
|
|
"mean_token_accuracy": 0.1451781891286373,
|
|
"num_tokens": 4477579.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.237515068054199,
|
|
"epoch": 0.21238078417520312,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998262548637667,
|
|
"loss": 6.2987,
|
|
"mean_token_accuracy": 0.15145175904035568,
|
|
"num_tokens": 4486800.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.3636726379394535,
|
|
"epoch": 0.21282232426704345,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004998250153199822,
|
|
"loss": 6.2466,
|
|
"mean_token_accuracy": 0.14414403662085534,
|
|
"num_tokens": 4495985.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.3901232242584225,
|
|
"epoch": 0.21326386435888378,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998237713720036,
|
|
"loss": 6.4023,
|
|
"mean_token_accuracy": 0.13772075325250627,
|
|
"num_tokens": 4504944.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.378066778182983,
|
|
"epoch": 0.2137054044507241,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998225230198552,
|
|
"loss": 6.2988,
|
|
"mean_token_accuracy": 0.14631899818778038,
|
|
"num_tokens": 4515402.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.370953845977783,
|
|
"epoch": 0.21414694454256447,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998212702635614,
|
|
"loss": 6.3772,
|
|
"mean_token_accuracy": 0.13616864681243895,
|
|
"num_tokens": 4525009.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.461081838607788,
|
|
"epoch": 0.2145884846344048,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004998200131031469,
|
|
"loss": 6.4151,
|
|
"mean_token_accuracy": 0.1347133368253708,
|
|
"num_tokens": 4534460.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.410534858703613,
|
|
"epoch": 0.21503002472624513,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998187515386361,
|
|
"loss": 6.2343,
|
|
"mean_token_accuracy": 0.15018296167254447,
|
|
"num_tokens": 4543748.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.339591979980469,
|
|
"epoch": 0.2154715648180855,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998174855700538,
|
|
"loss": 6.3688,
|
|
"mean_token_accuracy": 0.1432553306221962,
|
|
"num_tokens": 4552722.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.41485447883606,
|
|
"epoch": 0.21591310490992582,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004998162151974248,
|
|
"loss": 6.2185,
|
|
"mean_token_accuracy": 0.14427638724446296,
|
|
"num_tokens": 4561607.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.3874904155731205,
|
|
"epoch": 0.21635464500176615,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499814940420774,
|
|
"loss": 6.4909,
|
|
"mean_token_accuracy": 0.12949233055114745,
|
|
"num_tokens": 4572524.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.417895221710205,
|
|
"epoch": 0.21679618509360649,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998136612401266,
|
|
"loss": 6.3014,
|
|
"mean_token_accuracy": 0.14473507031798363,
|
|
"num_tokens": 4581601.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.422118234634399,
|
|
"epoch": 0.21723772518544684,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004998123776555071,
|
|
"loss": 6.3707,
|
|
"mean_token_accuracy": 0.13594872877001762,
|
|
"num_tokens": 4591795.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.37589282989502,
|
|
"epoch": 0.21767926527728718,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998110896669412,
|
|
"loss": 6.3263,
|
|
"mean_token_accuracy": 0.14324304163455964,
|
|
"num_tokens": 4600745.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.44620714187622,
|
|
"epoch": 0.2181208053691275,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998097972744539,
|
|
"loss": 6.3737,
|
|
"mean_token_accuracy": 0.13417462706565858,
|
|
"num_tokens": 4610490.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.308990859985352,
|
|
"epoch": 0.21856234546096787,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998085004780705,
|
|
"loss": 6.3033,
|
|
"mean_token_accuracy": 0.14380738139152527,
|
|
"num_tokens": 4619511.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.418844270706177,
|
|
"epoch": 0.2190038855528082,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004998071992778164,
|
|
"loss": 6.3332,
|
|
"mean_token_accuracy": 0.1404774770140648,
|
|
"num_tokens": 4628186.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.4331684589385985,
|
|
"epoch": 0.21944542564464853,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499805893673717,
|
|
"loss": 6.3698,
|
|
"mean_token_accuracy": 0.13764599412679673,
|
|
"num_tokens": 4637431.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.297756719589233,
|
|
"epoch": 0.21988696573648886,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998045836657982,
|
|
"loss": 6.2293,
|
|
"mean_token_accuracy": 0.14262612611055375,
|
|
"num_tokens": 4646627.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.3736780166625975,
|
|
"epoch": 0.22032850582832922,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998032692540853,
|
|
"loss": 6.3126,
|
|
"mean_token_accuracy": 0.14131407141685487,
|
|
"num_tokens": 4656095.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.4055116176605225,
|
|
"epoch": 0.22077004592016955,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998019504386044,
|
|
"loss": 6.3932,
|
|
"mean_token_accuracy": 0.1364021860063076,
|
|
"num_tokens": 4665807.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.426360702514648,
|
|
"epoch": 0.22121158601200988,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004998006272193809,
|
|
"loss": 6.3543,
|
|
"mean_token_accuracy": 0.13680973649024963,
|
|
"num_tokens": 4674459.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.394381141662597,
|
|
"epoch": 0.22165312610385024,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997992995964412,
|
|
"loss": 6.4516,
|
|
"mean_token_accuracy": 0.13281818181276323,
|
|
"num_tokens": 4684063.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.355847644805908,
|
|
"epoch": 0.22209466619569057,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004997979675698109,
|
|
"loss": 6.3149,
|
|
"mean_token_accuracy": 0.1413261540234089,
|
|
"num_tokens": 4692807.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.509682703018188,
|
|
"epoch": 0.2225362062875309,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997966311395164,
|
|
"loss": 6.2561,
|
|
"mean_token_accuracy": 0.14682135581970215,
|
|
"num_tokens": 4701100.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.257225751876831,
|
|
"epoch": 0.22297774637937123,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997952903055836,
|
|
"loss": 6.313,
|
|
"mean_token_accuracy": 0.13383381441235542,
|
|
"num_tokens": 4710697.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.414626836776733,
|
|
"epoch": 0.2234192864712116,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000499793945068039,
|
|
"loss": 6.2857,
|
|
"mean_token_accuracy": 0.14284081608057023,
|
|
"num_tokens": 4718745.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.376047897338867,
|
|
"epoch": 0.22386082656305192,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997925954269088,
|
|
"loss": 6.2546,
|
|
"mean_token_accuracy": 0.1423765517771244,
|
|
"num_tokens": 4728056.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.395099210739136,
|
|
"epoch": 0.22430236665489225,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997912413822196,
|
|
"loss": 6.4015,
|
|
"mean_token_accuracy": 0.13739172071218492,
|
|
"num_tokens": 4737605.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.362033176422119,
|
|
"epoch": 0.2247439067467326,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997898829339979,
|
|
"loss": 6.2078,
|
|
"mean_token_accuracy": 0.1501716986298561,
|
|
"num_tokens": 4746168.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.374505853652954,
|
|
"epoch": 0.22518544683857294,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00049978852008227,
|
|
"loss": 6.303,
|
|
"mean_token_accuracy": 0.14293254241347314,
|
|
"num_tokens": 4755072.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.365287399291992,
|
|
"epoch": 0.22562698693041328,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499787152827063,
|
|
"loss": 6.3115,
|
|
"mean_token_accuracy": 0.14703276976943017,
|
|
"num_tokens": 4764580.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.290199661254883,
|
|
"epoch": 0.2260685270222536,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997857811684035,
|
|
"loss": 6.311,
|
|
"mean_token_accuracy": 0.14540447145700455,
|
|
"num_tokens": 4774135.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.446982669830322,
|
|
"epoch": 0.22651006711409397,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997844051063183,
|
|
"loss": 6.4022,
|
|
"mean_token_accuracy": 0.13750494867563248,
|
|
"num_tokens": 4784733.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.385699224472046,
|
|
"epoch": 0.2269516072059343,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997830246408346,
|
|
"loss": 6.3325,
|
|
"mean_token_accuracy": 0.14618165343999862,
|
|
"num_tokens": 4795327.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.268747615814209,
|
|
"epoch": 0.22739314729777463,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997816397719791,
|
|
"loss": 6.2983,
|
|
"mean_token_accuracy": 0.14008190780878066,
|
|
"num_tokens": 4804314.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.464039659500122,
|
|
"epoch": 0.227834687389615,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997802504997792,
|
|
"loss": 6.3878,
|
|
"mean_token_accuracy": 0.135692573338747,
|
|
"num_tokens": 4813637.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.395627689361572,
|
|
"epoch": 0.22827622748145532,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997788568242621,
|
|
"loss": 6.279,
|
|
"mean_token_accuracy": 0.14184108525514602,
|
|
"num_tokens": 4823094.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.306736660003662,
|
|
"epoch": 0.22871776757329565,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499777458745455,
|
|
"loss": 6.206,
|
|
"mean_token_accuracy": 0.14509037733078003,
|
|
"num_tokens": 4833199.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.366773796081543,
|
|
"epoch": 0.22915930766513598,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997760562633853,
|
|
"loss": 6.3026,
|
|
"mean_token_accuracy": 0.13990220725536345,
|
|
"num_tokens": 4842970.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.40316481590271,
|
|
"epoch": 0.22960084775697634,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997746493780804,
|
|
"loss": 6.3996,
|
|
"mean_token_accuracy": 0.13560035228729247,
|
|
"num_tokens": 4852043.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.3404639720916744,
|
|
"epoch": 0.23004238784881667,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000499773238089568,
|
|
"loss": 6.2733,
|
|
"mean_token_accuracy": 0.1323003001511097,
|
|
"num_tokens": 4862232.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.368188381195068,
|
|
"epoch": 0.230483927940657,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997718223978758,
|
|
"loss": 6.2331,
|
|
"mean_token_accuracy": 0.141592987626791,
|
|
"num_tokens": 4871186.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.359707069396973,
|
|
"epoch": 0.23092546803249736,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997704023030315,
|
|
"loss": 6.3149,
|
|
"mean_token_accuracy": 0.1497406058013439,
|
|
"num_tokens": 4879974.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.361460208892822,
|
|
"epoch": 0.2313670081243377,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997689778050627,
|
|
"loss": 6.3786,
|
|
"mean_token_accuracy": 0.13907945528626442,
|
|
"num_tokens": 4890300.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.382821655273437,
|
|
"epoch": 0.23180854821617802,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997675489039975,
|
|
"loss": 6.3479,
|
|
"mean_token_accuracy": 0.1409486159682274,
|
|
"num_tokens": 4900428.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.362385368347168,
|
|
"epoch": 0.23225008830801835,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997661155998638,
|
|
"loss": 6.3464,
|
|
"mean_token_accuracy": 0.14176245480775834,
|
|
"num_tokens": 4910092.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.404236078262329,
|
|
"epoch": 0.2326916283998587,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997646778926898,
|
|
"loss": 6.3387,
|
|
"mean_token_accuracy": 0.1374544769525528,
|
|
"num_tokens": 4919593.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.340755891799927,
|
|
"epoch": 0.23313316849169904,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997632357825035,
|
|
"loss": 6.3179,
|
|
"mean_token_accuracy": 0.13816252574324608,
|
|
"num_tokens": 4929098.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.456451320648194,
|
|
"epoch": 0.23357470858353938,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997617892693333,
|
|
"loss": 6.35,
|
|
"mean_token_accuracy": 0.14228551536798478,
|
|
"num_tokens": 4938265.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.420327758789062,
|
|
"epoch": 0.23401624867537973,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997603383532075,
|
|
"loss": 6.3003,
|
|
"mean_token_accuracy": 0.14199153259396552,
|
|
"num_tokens": 4946694.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.29103798866272,
|
|
"epoch": 0.23445778876722007,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004997588830341545,
|
|
"loss": 6.3152,
|
|
"mean_token_accuracy": 0.14269881397485734,
|
|
"num_tokens": 4955296.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.337425231933594,
|
|
"epoch": 0.2348993288590604,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997574233122028,
|
|
"loss": 6.3015,
|
|
"mean_token_accuracy": 0.147652330994606,
|
|
"num_tokens": 4964409.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.348830032348633,
|
|
"epoch": 0.23534086895090076,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997559591873809,
|
|
"loss": 6.2893,
|
|
"mean_token_accuracy": 0.14996645748615264,
|
|
"num_tokens": 4973449.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.3320433616638185,
|
|
"epoch": 0.2357824090427411,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004997544906597178,
|
|
"loss": 6.2838,
|
|
"mean_token_accuracy": 0.14481322914361955,
|
|
"num_tokens": 4983057.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.325637483596802,
|
|
"epoch": 0.23622394913458142,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004997530177292418,
|
|
"loss": 6.3494,
|
|
"mean_token_accuracy": 0.13466070592403412,
|
|
"num_tokens": 4991950.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.43730731010437,
|
|
"epoch": 0.23666548922642175,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997515403959823,
|
|
"loss": 6.2955,
|
|
"mean_token_accuracy": 0.14283260256052016,
|
|
"num_tokens": 5001042.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.396935892105103,
|
|
"epoch": 0.2371070293182621,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004997500586599677,
|
|
"loss": 6.2485,
|
|
"mean_token_accuracy": 0.14456894770264625,
|
|
"num_tokens": 5009827.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.2672782897949215,
|
|
"epoch": 0.23754856941010244,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997485725212274,
|
|
"loss": 6.2885,
|
|
"mean_token_accuracy": 0.14449408054351806,
|
|
"num_tokens": 5018708.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.272865295410156,
|
|
"epoch": 0.23799010950194277,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997470819797903,
|
|
"loss": 6.1679,
|
|
"mean_token_accuracy": 0.1527360998094082,
|
|
"num_tokens": 5027522.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.284482479095459,
|
|
"epoch": 0.23843164959378313,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997455870356857,
|
|
"loss": 6.2987,
|
|
"mean_token_accuracy": 0.1458326295018196,
|
|
"num_tokens": 5035755.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.35535831451416,
|
|
"epoch": 0.23887318968562346,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997440876889429,
|
|
"loss": 6.2588,
|
|
"mean_token_accuracy": 0.14414956122636796,
|
|
"num_tokens": 5045289.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.262376403808593,
|
|
"epoch": 0.2393147297774638,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997425839395913,
|
|
"loss": 6.2788,
|
|
"mean_token_accuracy": 0.14706139862537385,
|
|
"num_tokens": 5053774.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.403537845611572,
|
|
"epoch": 0.23975626986930412,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997410757876602,
|
|
"loss": 6.3583,
|
|
"mean_token_accuracy": 0.1355147533118725,
|
|
"num_tokens": 5062911.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.373221158981323,
|
|
"epoch": 0.24019780996114448,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997395632331793,
|
|
"loss": 6.223,
|
|
"mean_token_accuracy": 0.1466083750128746,
|
|
"num_tokens": 5072107.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.251181316375733,
|
|
"epoch": 0.2406393500529848,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004997380462761781,
|
|
"loss": 6.181,
|
|
"mean_token_accuracy": 0.1502533346414566,
|
|
"num_tokens": 5080588.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.353159761428833,
|
|
"epoch": 0.24108089014482514,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997365249166864,
|
|
"loss": 6.3864,
|
|
"mean_token_accuracy": 0.14094797894358635,
|
|
"num_tokens": 5090262.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.381338834762573,
|
|
"epoch": 0.2415224302366655,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997349991547342,
|
|
"loss": 6.2962,
|
|
"mean_token_accuracy": 0.14725386276841163,
|
|
"num_tokens": 5099285.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.368382978439331,
|
|
"epoch": 0.24196397032850583,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997334689903509,
|
|
"loss": 6.3307,
|
|
"mean_token_accuracy": 0.14288511276245117,
|
|
"num_tokens": 5109115.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.4272418975830075,
|
|
"epoch": 0.24240551042034617,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004997319344235668,
|
|
"loss": 6.3721,
|
|
"mean_token_accuracy": 0.1368812806904316,
|
|
"num_tokens": 5117977.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.413669633865356,
|
|
"epoch": 0.2428470505121865,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000499730395454412,
|
|
"loss": 6.3185,
|
|
"mean_token_accuracy": 0.14243241772055626,
|
|
"num_tokens": 5127457.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.2972547054290775,
|
|
"epoch": 0.24328859060402686,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997288520829166,
|
|
"loss": 6.3606,
|
|
"mean_token_accuracy": 0.13738505616784097,
|
|
"num_tokens": 5137310.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.390837097167969,
|
|
"epoch": 0.2437301306958672,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997273043091107,
|
|
"loss": 6.2857,
|
|
"mean_token_accuracy": 0.1405642569065094,
|
|
"num_tokens": 5146963.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.359052276611328,
|
|
"epoch": 0.24417167078770752,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004997257521330248,
|
|
"loss": 6.2501,
|
|
"mean_token_accuracy": 0.14132534712553024,
|
|
"num_tokens": 5155521.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.37628755569458,
|
|
"epoch": 0.24461321087954788,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004997241955546892,
|
|
"loss": 6.2586,
|
|
"mean_token_accuracy": 0.13957886546850204,
|
|
"num_tokens": 5165182.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.32409896850586,
|
|
"epoch": 0.2450547509713882,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997226345741343,
|
|
"loss": 6.2833,
|
|
"mean_token_accuracy": 0.13664216697216033,
|
|
"num_tokens": 5175511.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.361440944671631,
|
|
"epoch": 0.24549629106322854,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000499721069191391,
|
|
"loss": 6.2283,
|
|
"mean_token_accuracy": 0.14851997345685958,
|
|
"num_tokens": 5184772.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.256292247772217,
|
|
"epoch": 0.24593783115506887,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997194994064896,
|
|
"loss": 6.3103,
|
|
"mean_token_accuracy": 0.14024864211678506,
|
|
"num_tokens": 5195136.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.3915492534637455,
|
|
"epoch": 0.24637937124690923,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000499717925219461,
|
|
"loss": 6.249,
|
|
"mean_token_accuracy": 0.14889259859919549,
|
|
"num_tokens": 5203163.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.315735578536987,
|
|
"epoch": 0.24682091133874956,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004997163466303362,
|
|
"loss": 6.3376,
|
|
"mean_token_accuracy": 0.13842562362551689,
|
|
"num_tokens": 5213233.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.385308742523193,
|
|
"epoch": 0.2472624514305899,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499714763639146,
|
|
"loss": 6.2659,
|
|
"mean_token_accuracy": 0.1395539350807667,
|
|
"num_tokens": 5222940.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.331099176406861,
|
|
"epoch": 0.24770399152243025,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004997131762459211,
|
|
"loss": 6.2917,
|
|
"mean_token_accuracy": 0.13852713331580163,
|
|
"num_tokens": 5232263.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.337193584442138,
|
|
"epoch": 0.24814553161427058,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004997115844506932,
|
|
"loss": 6.2739,
|
|
"mean_token_accuracy": 0.13856021910905839,
|
|
"num_tokens": 5241536.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.376536083221436,
|
|
"epoch": 0.2485870717061109,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004997099882534929,
|
|
"loss": 6.2841,
|
|
"mean_token_accuracy": 0.13798881992697715,
|
|
"num_tokens": 5250702.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.387785339355469,
|
|
"epoch": 0.24902861179795124,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997083876543519,
|
|
"loss": 6.2968,
|
|
"mean_token_accuracy": 0.1387106366455555,
|
|
"num_tokens": 5259811.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.46270055770874,
|
|
"epoch": 0.2494701518897916,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997067826533014,
|
|
"loss": 6.3787,
|
|
"mean_token_accuracy": 0.13431487306952478,
|
|
"num_tokens": 5270518.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.343976306915283,
|
|
"epoch": 0.24991169198163193,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004997051732503726,
|
|
"loss": 6.2692,
|
|
"mean_token_accuracy": 0.14104484394192696,
|
|
"num_tokens": 5279538.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.288498306274414,
|
|
"epoch": 0.25035323207347226,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004997035594455975,
|
|
"loss": 6.2935,
|
|
"mean_token_accuracy": 0.13851248025894164,
|
|
"num_tokens": 5289633.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.403953742980957,
|
|
"epoch": 0.2507947721653126,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004997019412390074,
|
|
"loss": 6.3908,
|
|
"mean_token_accuracy": 0.14274223148822784,
|
|
"num_tokens": 5299148.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.318170356750488,
|
|
"epoch": 0.2512363122571529,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000499700318630634,
|
|
"loss": 6.2601,
|
|
"mean_token_accuracy": 0.14186998903751374,
|
|
"num_tokens": 5309090.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.378513526916504,
|
|
"epoch": 0.2516778523489933,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004996986916205092,
|
|
"loss": 6.3497,
|
|
"mean_token_accuracy": 0.13793296962976456,
|
|
"num_tokens": 5318798.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.330926847457886,
|
|
"epoch": 0.25211939244083365,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004996970602086648,
|
|
"loss": 6.2201,
|
|
"mean_token_accuracy": 0.14618514329195023,
|
|
"num_tokens": 5327915.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.227778148651123,
|
|
"epoch": 0.252560932532674,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004996954243951327,
|
|
"loss": 6.2396,
|
|
"mean_token_accuracy": 0.15025153011083603,
|
|
"num_tokens": 5336970.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.362396907806397,
|
|
"epoch": 0.2530024726245143,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004996937841799451,
|
|
"loss": 6.2156,
|
|
"mean_token_accuracy": 0.14974839985370636,
|
|
"num_tokens": 5345167.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.2248279571533205,
|
|
"epoch": 0.25344401271635464,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004996921395631342,
|
|
"loss": 6.1855,
|
|
"mean_token_accuracy": 0.14353529885411262,
|
|
"num_tokens": 5353399.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.3493123054504395,
|
|
"epoch": 0.25388555280819497,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000499690490544732,
|
|
"loss": 6.2894,
|
|
"mean_token_accuracy": 0.14165428131818772,
|
|
"num_tokens": 5363224.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.406820631027221,
|
|
"epoch": 0.2543270929000353,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004996888371247707,
|
|
"loss": 6.3181,
|
|
"mean_token_accuracy": 0.14098790735006334,
|
|
"num_tokens": 5372274.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.34662184715271,
|
|
"epoch": 0.2547686329918757,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000499687179303283,
|
|
"loss": 6.2874,
|
|
"mean_token_accuracy": 0.14797318130731582,
|
|
"num_tokens": 5380240.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.25228362083435,
|
|
"epoch": 0.255210173083716,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996855170803012,
|
|
"loss": 6.1607,
|
|
"mean_token_accuracy": 0.15010163933038712,
|
|
"num_tokens": 5389390.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.337203788757324,
|
|
"epoch": 0.25565171317555635,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004996838504558581,
|
|
"loss": 6.3114,
|
|
"mean_token_accuracy": 0.14117275848984717,
|
|
"num_tokens": 5399425.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.378282165527343,
|
|
"epoch": 0.2560932532673967,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000499682179429986,
|
|
"loss": 6.3199,
|
|
"mean_token_accuracy": 0.14286198765039443,
|
|
"num_tokens": 5408717.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.319883155822754,
|
|
"epoch": 0.256534793359237,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996805040027178,
|
|
"loss": 6.2533,
|
|
"mean_token_accuracy": 0.13577283322811126,
|
|
"num_tokens": 5418475.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.353005409240723,
|
|
"epoch": 0.25697633345107734,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996788241740863,
|
|
"loss": 6.3058,
|
|
"mean_token_accuracy": 0.14230425655841827,
|
|
"num_tokens": 5428403.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.345854663848877,
|
|
"epoch": 0.2574178735429177,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004996771399441243,
|
|
"loss": 6.3449,
|
|
"mean_token_accuracy": 0.13902240470051766,
|
|
"num_tokens": 5437347.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.451116418838501,
|
|
"epoch": 0.25785941363475806,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004996754513128652,
|
|
"loss": 6.2456,
|
|
"mean_token_accuracy": 0.14855796918272973,
|
|
"num_tokens": 5446804.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.2477256774902346,
|
|
"epoch": 0.2583009537265984,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004996737582803416,
|
|
"loss": 6.2091,
|
|
"mean_token_accuracy": 0.1402647577226162,
|
|
"num_tokens": 5455888.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.36528582572937,
|
|
"epoch": 0.2587424938184387,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004996720608465868,
|
|
"loss": 6.1781,
|
|
"mean_token_accuracy": 0.14944826811552048,
|
|
"num_tokens": 5463977.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.307216548919678,
|
|
"epoch": 0.25918403391027905,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0004996703590116342,
|
|
"loss": 6.3079,
|
|
"mean_token_accuracy": 0.14276890456676483,
|
|
"num_tokens": 5473780.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.31512484550476,
|
|
"epoch": 0.2596255740021194,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004996686527755171,
|
|
"loss": 6.1977,
|
|
"mean_token_accuracy": 0.1501262977719307,
|
|
"num_tokens": 5482151.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.3214874267578125,
|
|
"epoch": 0.2600671140939597,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004996669421382687,
|
|
"loss": 6.2257,
|
|
"mean_token_accuracy": 0.1505381152033806,
|
|
"num_tokens": 5491103.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.25068473815918,
|
|
"epoch": 0.26050865418580005,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.0004996652270999228,
|
|
"loss": 6.2349,
|
|
"mean_token_accuracy": 0.15028091371059418,
|
|
"num_tokens": 5500367.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.329337501525879,
|
|
"epoch": 0.26095019427764043,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004996635076605128,
|
|
"loss": 6.2539,
|
|
"mean_token_accuracy": 0.15224550366401673,
|
|
"num_tokens": 5509631.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.459008836746216,
|
|
"epoch": 0.26139173436948077,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0004996617838200725,
|
|
"loss": 6.289,
|
|
"mean_token_accuracy": 0.1362527571618557,
|
|
"num_tokens": 5518635.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.2465019702911375,
|
|
"epoch": 0.2618332744613211,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004996600555786357,
|
|
"loss": 6.2431,
|
|
"mean_token_accuracy": 0.14100067913532258,
|
|
"num_tokens": 5527696.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.315958070755005,
|
|
"epoch": 0.26227481455316143,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004996583229362362,
|
|
"loss": 6.2046,
|
|
"mean_token_accuracy": 0.14614154621958733,
|
|
"num_tokens": 5536632.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.385778141021729,
|
|
"epoch": 0.26271635464500176,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004996565858929078,
|
|
"loss": 6.2772,
|
|
"mean_token_accuracy": 0.14362908750772477,
|
|
"num_tokens": 5545825.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.284611654281616,
|
|
"epoch": 0.2631578947368421,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004996548444486847,
|
|
"loss": 6.166,
|
|
"mean_token_accuracy": 0.14970893263816834,
|
|
"num_tokens": 5555158.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.209372615814209,
|
|
"epoch": 0.2635994348286824,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004996530986036008,
|
|
"loss": 6.0922,
|
|
"mean_token_accuracy": 0.15141627043485642,
|
|
"num_tokens": 5564218.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.198943185806274,
|
|
"epoch": 0.2640409749205228,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004996513483576907,
|
|
"loss": 6.2522,
|
|
"mean_token_accuracy": 0.14398482963442802,
|
|
"num_tokens": 5572760.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.441630268096924,
|
|
"epoch": 0.26448251501236314,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.0004996495937109884,
|
|
"loss": 6.3173,
|
|
"mean_token_accuracy": 0.14210836142301558,
|
|
"num_tokens": 5581660.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.330695295333863,
|
|
"epoch": 0.26492405510420347,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004996478346635283,
|
|
"loss": 6.2178,
|
|
"mean_token_accuracy": 0.14186472147703172,
|
|
"num_tokens": 5590664.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.26492405510420347,
|
|
"eval_entropy": 6.125705798268993,
|
|
"eval_loss": 6.292537212371826,
|
|
"eval_mean_token_accuracy": 0.14802020881356648,
|
|
"eval_num_tokens": 5590664.0,
|
|
"eval_runtime": 26.6031,
|
|
"eval_samples_per_second": 1327.479,
|
|
"eval_steps_per_second": 165.958,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 113230,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 3000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 8221684432896000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|