3035 lines
82 KiB
JSON
3035 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.10645470352365069,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691987895965577,
|
|
"epoch": 0.00035484901174550227,
|
|
"grad_norm": 11.625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8366,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 11280.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691946220397949,
|
|
"epoch": 0.0007096980234910045,
|
|
"grad_norm": 11.5625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7735,
|
|
"mean_token_accuracy": 0.00010460250778123736,
|
|
"num_tokens": 21887.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.69083604812622,
|
|
"epoch": 0.001064547035236507,
|
|
"grad_norm": 9.6875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5654,
|
|
"mean_token_accuracy": 0.02854402889497578,
|
|
"num_tokens": 32911.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.66428632736206,
|
|
"epoch": 0.001419396046982009,
|
|
"grad_norm": 6.0625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.3144,
|
|
"mean_token_accuracy": 0.04550958257168532,
|
|
"num_tokens": 43994.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.494576263427735,
|
|
"epoch": 0.0017742450587275114,
|
|
"grad_norm": 4.28125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.105,
|
|
"mean_token_accuracy": 0.041560862213373184,
|
|
"num_tokens": 54342.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.533498859405517,
|
|
"epoch": 0.002129094070473014,
|
|
"grad_norm": 2.921875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.9509,
|
|
"mean_token_accuracy": 0.044634624011814596,
|
|
"num_tokens": 64768.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.53185338973999,
|
|
"epoch": 0.002483943082218516,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.9269,
|
|
"mean_token_accuracy": 0.04376727268099785,
|
|
"num_tokens": 75530.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.487653255462646,
|
|
"epoch": 0.002838792093964018,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.8511,
|
|
"mean_token_accuracy": 0.045070053450763224,
|
|
"num_tokens": 85983.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.522008514404297,
|
|
"epoch": 0.0031936411057095207,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8344,
|
|
"mean_token_accuracy": 0.04323017336428166,
|
|
"num_tokens": 96168.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.482469272613525,
|
|
"epoch": 0.0035484901174550228,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.7891,
|
|
"mean_token_accuracy": 0.05232359655201435,
|
|
"num_tokens": 107899.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.414291763305664,
|
|
"epoch": 0.0039033391292005253,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.7448,
|
|
"mean_token_accuracy": 0.05363052934408188,
|
|
"num_tokens": 119073.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.483110237121583,
|
|
"epoch": 0.004258188140946028,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.7062,
|
|
"mean_token_accuracy": 0.05208716280758381,
|
|
"num_tokens": 129149.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.321412563323975,
|
|
"epoch": 0.00461303715269153,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.5992,
|
|
"mean_token_accuracy": 0.05521074235439301,
|
|
"num_tokens": 139045.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.363921737670898,
|
|
"epoch": 0.004967886164437032,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.6115,
|
|
"mean_token_accuracy": 0.054533609002828595,
|
|
"num_tokens": 150251.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.423143291473389,
|
|
"epoch": 0.005322735176182534,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.5269,
|
|
"mean_token_accuracy": 0.05440324060618877,
|
|
"num_tokens": 160230.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.341320323944093,
|
|
"epoch": 0.005677584187928036,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.478,
|
|
"mean_token_accuracy": 0.05366469696164131,
|
|
"num_tokens": 170692.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.290056896209716,
|
|
"epoch": 0.006032433199673539,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.3959,
|
|
"mean_token_accuracy": 0.05197730958461762,
|
|
"num_tokens": 180616.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.320787811279297,
|
|
"epoch": 0.006387282211419041,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.3057,
|
|
"mean_token_accuracy": 0.05711167939007282,
|
|
"num_tokens": 191016.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.258602237701416,
|
|
"epoch": 0.0067421312231645434,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.2578,
|
|
"mean_token_accuracy": 0.05482340231537819,
|
|
"num_tokens": 202433.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.235629463195801,
|
|
"epoch": 0.0070969802349100455,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2401,
|
|
"mean_token_accuracy": 0.05469023734331131,
|
|
"num_tokens": 214586.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.223749732971191,
|
|
"epoch": 0.007451829246655548,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.0977,
|
|
"mean_token_accuracy": 0.05747384466230869,
|
|
"num_tokens": 225155.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.136597728729248,
|
|
"epoch": 0.007806678258401051,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.9983,
|
|
"mean_token_accuracy": 0.05849210806190967,
|
|
"num_tokens": 236050.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.020172691345214,
|
|
"epoch": 0.008161527270146553,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.8941,
|
|
"mean_token_accuracy": 0.05953489430248737,
|
|
"num_tokens": 245825.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.979390907287598,
|
|
"epoch": 0.008516376281892056,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8174,
|
|
"mean_token_accuracy": 0.06152731701731682,
|
|
"num_tokens": 256752.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.866498374938965,
|
|
"epoch": 0.008871225293637557,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.802,
|
|
"mean_token_accuracy": 0.05181486271321774,
|
|
"num_tokens": 267603.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.726642417907716,
|
|
"epoch": 0.00922607430538306,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.7071,
|
|
"mean_token_accuracy": 0.05791006051003933,
|
|
"num_tokens": 277478.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.658979320526123,
|
|
"epoch": 0.009580923317128561,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.66,
|
|
"mean_token_accuracy": 0.05634133592247963,
|
|
"num_tokens": 289710.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.384044075012207,
|
|
"epoch": 0.009935772328874064,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.5346,
|
|
"mean_token_accuracy": 0.05907244272530079,
|
|
"num_tokens": 299295.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.350571346282958,
|
|
"epoch": 0.010290621340619567,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.5165,
|
|
"mean_token_accuracy": 0.05372942052781582,
|
|
"num_tokens": 310544.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.217219734191895,
|
|
"epoch": 0.010645470352365068,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.4485,
|
|
"mean_token_accuracy": 0.05611223764717579,
|
|
"num_tokens": 321558.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.077949619293213,
|
|
"epoch": 0.011000319364110571,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.44,
|
|
"mean_token_accuracy": 0.05572409965097904,
|
|
"num_tokens": 333265.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.919435977935791,
|
|
"epoch": 0.011355168375856073,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.3707,
|
|
"mean_token_accuracy": 0.060116075351834296,
|
|
"num_tokens": 343406.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.981222343444824,
|
|
"epoch": 0.011710017387601575,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.4388,
|
|
"mean_token_accuracy": 0.05228844713419676,
|
|
"num_tokens": 354934.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.788821792602539,
|
|
"epoch": 0.012064866399347078,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.3627,
|
|
"mean_token_accuracy": 0.05599985755980015,
|
|
"num_tokens": 364991.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.838678550720214,
|
|
"epoch": 0.01241971541109258,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.3174,
|
|
"mean_token_accuracy": 0.05698671489953995,
|
|
"num_tokens": 375362.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.669477367401123,
|
|
"epoch": 0.012774564422838083,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.2929,
|
|
"mean_token_accuracy": 0.06303150560706854,
|
|
"num_tokens": 386758.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.66919822692871,
|
|
"epoch": 0.013129413434583584,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.304,
|
|
"mean_token_accuracy": 0.061275123804807666,
|
|
"num_tokens": 397073.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.68841199874878,
|
|
"epoch": 0.013484262446329087,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.4007,
|
|
"mean_token_accuracy": 0.05668871849775314,
|
|
"num_tokens": 408098.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.577757740020752,
|
|
"epoch": 0.01383911145807459,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.2425,
|
|
"mean_token_accuracy": 0.06331006735563278,
|
|
"num_tokens": 418910.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.616193675994873,
|
|
"epoch": 0.014193960469820091,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.3266,
|
|
"mean_token_accuracy": 0.059384917840361595,
|
|
"num_tokens": 429328.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.564495944976807,
|
|
"epoch": 0.014548809481565594,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.286,
|
|
"mean_token_accuracy": 0.058925506100058554,
|
|
"num_tokens": 440293.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.546427249908447,
|
|
"epoch": 0.014903658493311095,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.2513,
|
|
"mean_token_accuracy": 0.057647600024938586,
|
|
"num_tokens": 451509.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.532197761535645,
|
|
"epoch": 0.015258507505056598,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.258,
|
|
"mean_token_accuracy": 0.058545221015810965,
|
|
"num_tokens": 461660.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.52504529953003,
|
|
"epoch": 0.015613356516802101,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.2467,
|
|
"mean_token_accuracy": 0.0674739558249712,
|
|
"num_tokens": 471362.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.518447113037109,
|
|
"epoch": 0.015968205528547604,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.1115,
|
|
"mean_token_accuracy": 0.07247221358120441,
|
|
"num_tokens": 482036.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.392018604278565,
|
|
"epoch": 0.016323054540293105,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.2038,
|
|
"mean_token_accuracy": 0.06354649700224399,
|
|
"num_tokens": 492633.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.478435707092284,
|
|
"epoch": 0.016677903552038607,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.2366,
|
|
"mean_token_accuracy": 0.06092475391924381,
|
|
"num_tokens": 502613.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.442237091064452,
|
|
"epoch": 0.01703275256378411,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.26,
|
|
"mean_token_accuracy": 0.06470727063715458,
|
|
"num_tokens": 513256.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.596701526641846,
|
|
"epoch": 0.017387601575529613,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.2988,
|
|
"mean_token_accuracy": 0.061711058393120764,
|
|
"num_tokens": 523355.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.451476860046387,
|
|
"epoch": 0.017742450587275114,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.2216,
|
|
"mean_token_accuracy": 0.06455190740525722,
|
|
"num_tokens": 535195.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.469265937805176,
|
|
"epoch": 0.018097299599020615,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.1852,
|
|
"mean_token_accuracy": 0.06743233427405357,
|
|
"num_tokens": 544831.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.445653438568115,
|
|
"epoch": 0.01845214861076612,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.2694,
|
|
"mean_token_accuracy": 0.059878384508192536,
|
|
"num_tokens": 555835.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.478738117218018,
|
|
"epoch": 0.01880699762251162,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.2247,
|
|
"mean_token_accuracy": 0.06131057888269424,
|
|
"num_tokens": 566875.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.375457763671875,
|
|
"epoch": 0.019161846634257122,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.237,
|
|
"mean_token_accuracy": 0.06333270035684109,
|
|
"num_tokens": 577462.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.501999855041504,
|
|
"epoch": 0.019516695646002627,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.1721,
|
|
"mean_token_accuracy": 0.06697900146245957,
|
|
"num_tokens": 587569.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.420077037811279,
|
|
"epoch": 0.019871544657748128,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.2355,
|
|
"mean_token_accuracy": 0.05800015553832054,
|
|
"num_tokens": 599322.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.452305221557618,
|
|
"epoch": 0.02022639366949363,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.2397,
|
|
"mean_token_accuracy": 0.06376027651131153,
|
|
"num_tokens": 610060.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.481741237640382,
|
|
"epoch": 0.020581242681239134,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.295,
|
|
"mean_token_accuracy": 0.06520596966147423,
|
|
"num_tokens": 620091.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.525186729431152,
|
|
"epoch": 0.020936091692984635,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.12,
|
|
"mean_token_accuracy": 0.0748389147222042,
|
|
"num_tokens": 631440.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.37361249923706,
|
|
"epoch": 0.021290940704730137,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.0801,
|
|
"mean_token_accuracy": 0.06787928678095341,
|
|
"num_tokens": 641374.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.305605411529541,
|
|
"epoch": 0.02164578971647564,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.1269,
|
|
"mean_token_accuracy": 0.06664356999099255,
|
|
"num_tokens": 652772.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.341912460327148,
|
|
"epoch": 0.022000638728221143,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.1348,
|
|
"mean_token_accuracy": 0.07087188474833965,
|
|
"num_tokens": 664052.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.363450813293458,
|
|
"epoch": 0.022355487739966644,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.1412,
|
|
"mean_token_accuracy": 0.07019479162991046,
|
|
"num_tokens": 675937.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.314255905151366,
|
|
"epoch": 0.022710336751712145,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.065,
|
|
"mean_token_accuracy": 0.07636438310146332,
|
|
"num_tokens": 687149.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.342943286895752,
|
|
"epoch": 0.02306518576345765,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.0868,
|
|
"mean_token_accuracy": 0.06923852488398552,
|
|
"num_tokens": 697646.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.419994449615478,
|
|
"epoch": 0.02342003477520315,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.2002,
|
|
"mean_token_accuracy": 0.06634289510548115,
|
|
"num_tokens": 708612.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.354696655273438,
|
|
"epoch": 0.023774883786948652,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.1205,
|
|
"mean_token_accuracy": 0.07153847180306912,
|
|
"num_tokens": 719705.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.35772638320923,
|
|
"epoch": 0.024129732798694157,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.0862,
|
|
"mean_token_accuracy": 0.06683175601065158,
|
|
"num_tokens": 730552.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.334082317352294,
|
|
"epoch": 0.024484581810439658,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.0935,
|
|
"mean_token_accuracy": 0.06866642013192177,
|
|
"num_tokens": 741651.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.407394886016846,
|
|
"epoch": 0.02483943082218516,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.1023,
|
|
"mean_token_accuracy": 0.06832008063793182,
|
|
"num_tokens": 751820.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.351481819152832,
|
|
"epoch": 0.025194279833930664,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.1154,
|
|
"mean_token_accuracy": 0.06219382137060166,
|
|
"num_tokens": 762125.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.352156829833984,
|
|
"epoch": 0.025549128845676165,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.0332,
|
|
"mean_token_accuracy": 0.07175925448536873,
|
|
"num_tokens": 773137.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.26557502746582,
|
|
"epoch": 0.025903977857421667,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.117,
|
|
"mean_token_accuracy": 0.06638101302087307,
|
|
"num_tokens": 784702.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.30114049911499,
|
|
"epoch": 0.026258826869167168,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.0759,
|
|
"mean_token_accuracy": 0.06737305559217929,
|
|
"num_tokens": 795806.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.324819755554199,
|
|
"epoch": 0.026613675880912673,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.0175,
|
|
"mean_token_accuracy": 0.06811637170612812,
|
|
"num_tokens": 806006.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.294619274139404,
|
|
"epoch": 0.026968524892658174,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.0306,
|
|
"mean_token_accuracy": 0.0675454068928957,
|
|
"num_tokens": 816345.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.203885173797607,
|
|
"epoch": 0.027323373904403675,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.0311,
|
|
"mean_token_accuracy": 0.06778408177196979,
|
|
"num_tokens": 827038.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.261708545684815,
|
|
"epoch": 0.02767822291614918,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.0325,
|
|
"mean_token_accuracy": 0.07145289778709411,
|
|
"num_tokens": 837465.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.322563171386719,
|
|
"epoch": 0.02803307192789468,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.0101,
|
|
"mean_token_accuracy": 0.0660801563411951,
|
|
"num_tokens": 848461.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.286761474609374,
|
|
"epoch": 0.028387920939640182,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.0912,
|
|
"mean_token_accuracy": 0.06686392053961754,
|
|
"num_tokens": 859215.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.153140926361084,
|
|
"epoch": 0.028742769951385687,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.9714,
|
|
"mean_token_accuracy": 0.06791464723646641,
|
|
"num_tokens": 869165.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.278606796264649,
|
|
"epoch": 0.029097618963131188,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.0299,
|
|
"mean_token_accuracy": 0.06314935013651848,
|
|
"num_tokens": 879737.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.317267990112304,
|
|
"epoch": 0.02945246797487669,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.0687,
|
|
"mean_token_accuracy": 0.06716468073427677,
|
|
"num_tokens": 889960.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.284847354888916,
|
|
"epoch": 0.02980731698662219,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.0315,
|
|
"mean_token_accuracy": 0.06850776262581348,
|
|
"num_tokens": 902004.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.195465469360352,
|
|
"epoch": 0.030162165998367695,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.9605,
|
|
"mean_token_accuracy": 0.06964701861143112,
|
|
"num_tokens": 912448.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.246010208129883,
|
|
"epoch": 0.030517015010113197,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.0425,
|
|
"mean_token_accuracy": 0.06564139947295189,
|
|
"num_tokens": 922295.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.212500381469727,
|
|
"epoch": 0.030871864021858698,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.9675,
|
|
"mean_token_accuracy": 0.07113610915839672,
|
|
"num_tokens": 932849.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.252998352050781,
|
|
"epoch": 0.031226713033604202,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.9716,
|
|
"mean_token_accuracy": 0.06684075817465782,
|
|
"num_tokens": 943570.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.093912363052368,
|
|
"epoch": 0.031581562045349704,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.0064,
|
|
"mean_token_accuracy": 0.06446249783039093,
|
|
"num_tokens": 953681.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.259466552734375,
|
|
"epoch": 0.03193641105709521,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.0026,
|
|
"mean_token_accuracy": 0.06529812254011631,
|
|
"num_tokens": 965027.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.120300006866454,
|
|
"epoch": 0.032291260068840706,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.9335,
|
|
"mean_token_accuracy": 0.06843561045825482,
|
|
"num_tokens": 975757.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.253607749938965,
|
|
"epoch": 0.03264610908058621,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.975,
|
|
"mean_token_accuracy": 0.07162793055176735,
|
|
"num_tokens": 987354.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.038683032989502,
|
|
"epoch": 0.033000958092331716,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.9604,
|
|
"mean_token_accuracy": 0.06958170719444752,
|
|
"num_tokens": 997844.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.277831172943115,
|
|
"epoch": 0.03335580710407721,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.0444,
|
|
"mean_token_accuracy": 0.07072999998927117,
|
|
"num_tokens": 1008748.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.073792266845704,
|
|
"epoch": 0.03371065611582272,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.8659,
|
|
"mean_token_accuracy": 0.0738751970231533,
|
|
"num_tokens": 1018612.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.137950706481934,
|
|
"epoch": 0.03406550512756822,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.9242,
|
|
"mean_token_accuracy": 0.07282396517693997,
|
|
"num_tokens": 1029431.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.098453617095947,
|
|
"epoch": 0.03442035413931372,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.9572,
|
|
"mean_token_accuracy": 0.06277543865144253,
|
|
"num_tokens": 1041255.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.044496870040893,
|
|
"epoch": 0.034775203151059225,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.7917,
|
|
"mean_token_accuracy": 0.08647977113723755,
|
|
"num_tokens": 1052416.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.117148971557617,
|
|
"epoch": 0.03513005216280473,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.9132,
|
|
"mean_token_accuracy": 0.07025011256337166,
|
|
"num_tokens": 1063801.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.052232360839843,
|
|
"epoch": 0.03548490117455023,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.9351,
|
|
"mean_token_accuracy": 0.06856431551277638,
|
|
"num_tokens": 1074984.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.149673557281494,
|
|
"epoch": 0.03583975018629573,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.8893,
|
|
"mean_token_accuracy": 0.07065611332654953,
|
|
"num_tokens": 1086306.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.180394887924194,
|
|
"epoch": 0.03619459919804123,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.8803,
|
|
"mean_token_accuracy": 0.06889106482267379,
|
|
"num_tokens": 1096569.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.032092428207397,
|
|
"epoch": 0.036549448209786735,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.9032,
|
|
"mean_token_accuracy": 0.0709243293851614,
|
|
"num_tokens": 1106599.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.193197917938232,
|
|
"epoch": 0.03690429722153224,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.9409,
|
|
"mean_token_accuracy": 0.06765258684754372,
|
|
"num_tokens": 1118337.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.961832189559937,
|
|
"epoch": 0.03725914623327774,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.9391,
|
|
"mean_token_accuracy": 0.07329438030719757,
|
|
"num_tokens": 1129685.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.117317533493042,
|
|
"epoch": 0.03761399524502324,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.8971,
|
|
"mean_token_accuracy": 0.07165143899619579,
|
|
"num_tokens": 1141055.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.06903042793274,
|
|
"epoch": 0.03796884425676875,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.9105,
|
|
"mean_token_accuracy": 0.0673241663724184,
|
|
"num_tokens": 1153130.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.082281255722046,
|
|
"epoch": 0.038323693268514245,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.8957,
|
|
"mean_token_accuracy": 0.06771765686571599,
|
|
"num_tokens": 1163764.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.043478631973267,
|
|
"epoch": 0.03867854228025975,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.8448,
|
|
"mean_token_accuracy": 0.07466785535216332,
|
|
"num_tokens": 1173902.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.998564767837524,
|
|
"epoch": 0.039033391292005254,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.7856,
|
|
"mean_token_accuracy": 0.07209023796021938,
|
|
"num_tokens": 1185055.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.057881355285645,
|
|
"epoch": 0.03938824030375075,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.8076,
|
|
"mean_token_accuracy": 0.07696683742105961,
|
|
"num_tokens": 1195695.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.934873914718628,
|
|
"epoch": 0.039743089315496256,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.8796,
|
|
"mean_token_accuracy": 0.07105770707130432,
|
|
"num_tokens": 1207459.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.985773086547852,
|
|
"epoch": 0.04009793832724176,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.8242,
|
|
"mean_token_accuracy": 0.07130814269185067,
|
|
"num_tokens": 1217533.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.973219776153565,
|
|
"epoch": 0.04045278733898726,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8496,
|
|
"mean_token_accuracy": 0.07043180353939534,
|
|
"num_tokens": 1231146.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.945406341552735,
|
|
"epoch": 0.040807636350732764,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.8843,
|
|
"mean_token_accuracy": 0.07025334909558296,
|
|
"num_tokens": 1243120.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.011637639999389,
|
|
"epoch": 0.04116248536247827,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.9156,
|
|
"mean_token_accuracy": 0.06993781737983226,
|
|
"num_tokens": 1255180.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.02280888557434,
|
|
"epoch": 0.041517334374223766,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.8431,
|
|
"mean_token_accuracy": 0.07286351397633553,
|
|
"num_tokens": 1266382.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.03951153755188,
|
|
"epoch": 0.04187218338596927,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.805,
|
|
"mean_token_accuracy": 0.07100504785776138,
|
|
"num_tokens": 1278184.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.95987548828125,
|
|
"epoch": 0.042227032397714775,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.7979,
|
|
"mean_token_accuracy": 0.07571852058172227,
|
|
"num_tokens": 1289273.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.960937976837158,
|
|
"epoch": 0.04258188140946027,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.7835,
|
|
"mean_token_accuracy": 0.07141958326101303,
|
|
"num_tokens": 1299684.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.025558280944825,
|
|
"epoch": 0.04293673042120578,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.8675,
|
|
"mean_token_accuracy": 0.06696930788457393,
|
|
"num_tokens": 1310859.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.910726022720337,
|
|
"epoch": 0.04329157943295128,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.7781,
|
|
"mean_token_accuracy": 0.07068931050598622,
|
|
"num_tokens": 1322011.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.939458894729614,
|
|
"epoch": 0.04364642844469678,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.7663,
|
|
"mean_token_accuracy": 0.07447356693446636,
|
|
"num_tokens": 1332187.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.852445602416992,
|
|
"epoch": 0.044001277456442285,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.6762,
|
|
"mean_token_accuracy": 0.07490805983543396,
|
|
"num_tokens": 1342678.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.9485992908477785,
|
|
"epoch": 0.04435612646818778,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.8034,
|
|
"mean_token_accuracy": 0.0721922166645527,
|
|
"num_tokens": 1352563.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.916344928741455,
|
|
"epoch": 0.04471097547993329,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.8087,
|
|
"mean_token_accuracy": 0.0754479069262743,
|
|
"num_tokens": 1363514.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.988961410522461,
|
|
"epoch": 0.04506582449167879,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.9224,
|
|
"mean_token_accuracy": 0.06770290434360504,
|
|
"num_tokens": 1374914.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.941910457611084,
|
|
"epoch": 0.04542067350342429,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.7749,
|
|
"mean_token_accuracy": 0.07321363165974618,
|
|
"num_tokens": 1385757.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.951854848861695,
|
|
"epoch": 0.045775522515169795,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.7757,
|
|
"mean_token_accuracy": 0.07017956748604774,
|
|
"num_tokens": 1397376.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.916997385025025,
|
|
"epoch": 0.0461303715269153,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.7839,
|
|
"mean_token_accuracy": 0.07471574619412422,
|
|
"num_tokens": 1408491.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.876117801666259,
|
|
"epoch": 0.0464852205386608,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.704,
|
|
"mean_token_accuracy": 0.09316127598285676,
|
|
"num_tokens": 1417643.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.789826583862305,
|
|
"epoch": 0.0468400695504063,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.7654,
|
|
"mean_token_accuracy": 0.0712949451059103,
|
|
"num_tokens": 1427446.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.888260316848755,
|
|
"epoch": 0.04719491856215181,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.7251,
|
|
"mean_token_accuracy": 0.0718452412635088,
|
|
"num_tokens": 1437782.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.854969501495361,
|
|
"epoch": 0.047549767573897304,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.7738,
|
|
"mean_token_accuracy": 0.07261287756264209,
|
|
"num_tokens": 1448778.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.974173069000244,
|
|
"epoch": 0.04790461658564281,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.7582,
|
|
"mean_token_accuracy": 0.07418306656181813,
|
|
"num_tokens": 1460343.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.7097618103027346,
|
|
"epoch": 0.048259465597388314,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.6886,
|
|
"mean_token_accuracy": 0.07660397589206695,
|
|
"num_tokens": 1471728.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.960760021209717,
|
|
"epoch": 0.04861431460913381,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.7778,
|
|
"mean_token_accuracy": 0.0769094929099083,
|
|
"num_tokens": 1482946.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.7662310123443605,
|
|
"epoch": 0.048969163620879316,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.6265,
|
|
"mean_token_accuracy": 0.0757428240031004,
|
|
"num_tokens": 1494028.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.854742527008057,
|
|
"epoch": 0.04932401263262482,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.7516,
|
|
"mean_token_accuracy": 0.07619248703122139,
|
|
"num_tokens": 1503763.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.773598098754883,
|
|
"epoch": 0.04967886164437032,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.6727,
|
|
"mean_token_accuracy": 0.07600370906293392,
|
|
"num_tokens": 1514488.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.798940801620484,
|
|
"epoch": 0.05003371065611582,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.6922,
|
|
"mean_token_accuracy": 0.07355734147131443,
|
|
"num_tokens": 1524489.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.838931655883789,
|
|
"epoch": 0.05038855966786133,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.6913,
|
|
"mean_token_accuracy": 0.07631286308169365,
|
|
"num_tokens": 1535005.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.7826978206634525,
|
|
"epoch": 0.050743408679606826,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.6585,
|
|
"mean_token_accuracy": 0.07577394507825375,
|
|
"num_tokens": 1546049.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.830225944519043,
|
|
"epoch": 0.05109825769135233,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.6845,
|
|
"mean_token_accuracy": 0.0732595931738615,
|
|
"num_tokens": 1556468.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.741913080215454,
|
|
"epoch": 0.05145310670309783,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.6607,
|
|
"mean_token_accuracy": 0.08094885796308518,
|
|
"num_tokens": 1567341.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.8365312099456785,
|
|
"epoch": 0.05180795571484333,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.7157,
|
|
"mean_token_accuracy": 0.07318115755915641,
|
|
"num_tokens": 1578492.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.815832948684692,
|
|
"epoch": 0.05216280472658884,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.6973,
|
|
"mean_token_accuracy": 0.07630126550793648,
|
|
"num_tokens": 1589147.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.671171092987061,
|
|
"epoch": 0.052517653738334336,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.5389,
|
|
"mean_token_accuracy": 0.08950962349772454,
|
|
"num_tokens": 1598725.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.733880805969238,
|
|
"epoch": 0.05287250275007984,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.6096,
|
|
"mean_token_accuracy": 0.07969549223780632,
|
|
"num_tokens": 1608913.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.702016448974609,
|
|
"epoch": 0.053227351761825345,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.6984,
|
|
"mean_token_accuracy": 0.07809793129563332,
|
|
"num_tokens": 1619138.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.733037090301513,
|
|
"epoch": 0.05358220077357084,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.5857,
|
|
"mean_token_accuracy": 0.07457101494073867,
|
|
"num_tokens": 1631217.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.7332252025604244,
|
|
"epoch": 0.05393704978531635,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.66,
|
|
"mean_token_accuracy": 0.07894963994622231,
|
|
"num_tokens": 1642552.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.6441277980804445,
|
|
"epoch": 0.05429189879706185,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6736,
|
|
"mean_token_accuracy": 0.07516682222485542,
|
|
"num_tokens": 1653422.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.763870811462402,
|
|
"epoch": 0.05464674780880735,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.6713,
|
|
"mean_token_accuracy": 0.07678476013243199,
|
|
"num_tokens": 1664709.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.717665433883667,
|
|
"epoch": 0.055001596820552855,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.6484,
|
|
"mean_token_accuracy": 0.07786873653531075,
|
|
"num_tokens": 1676114.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.750670051574707,
|
|
"epoch": 0.05535644583229836,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.7339,
|
|
"mean_token_accuracy": 0.07396974675357341,
|
|
"num_tokens": 1687558.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.702802991867065,
|
|
"epoch": 0.05571129484404386,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.6509,
|
|
"mean_token_accuracy": 0.07516442388296127,
|
|
"num_tokens": 1699046.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.7537883758544925,
|
|
"epoch": 0.05606614385578936,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.6358,
|
|
"mean_token_accuracy": 0.07771104238927365,
|
|
"num_tokens": 1709274.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.673342323303222,
|
|
"epoch": 0.05642099286753487,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.5738,
|
|
"mean_token_accuracy": 0.07565616965293884,
|
|
"num_tokens": 1720358.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.7894445896148685,
|
|
"epoch": 0.056775841879280364,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.7079,
|
|
"mean_token_accuracy": 0.08025034442543984,
|
|
"num_tokens": 1731684.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.651151466369629,
|
|
"epoch": 0.05713069089102587,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.6573,
|
|
"mean_token_accuracy": 0.07534100264310836,
|
|
"num_tokens": 1741862.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.712702083587646,
|
|
"epoch": 0.057485539902771374,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.6288,
|
|
"mean_token_accuracy": 0.07567652054131031,
|
|
"num_tokens": 1753166.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.656327629089356,
|
|
"epoch": 0.05784038891451687,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.6233,
|
|
"mean_token_accuracy": 0.08203103095293045,
|
|
"num_tokens": 1764166.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.6826342105865475,
|
|
"epoch": 0.058195237926262376,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.5671,
|
|
"mean_token_accuracy": 0.07672160528600216,
|
|
"num_tokens": 1774894.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.5881028175354,
|
|
"epoch": 0.05855008693800788,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.5501,
|
|
"mean_token_accuracy": 0.08005938455462455,
|
|
"num_tokens": 1785038.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.658244895935058,
|
|
"epoch": 0.05890493594975338,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.6152,
|
|
"mean_token_accuracy": 0.07943779826164246,
|
|
"num_tokens": 1795837.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.661693716049195,
|
|
"epoch": 0.05925978496149888,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.608,
|
|
"mean_token_accuracy": 0.07967479974031448,
|
|
"num_tokens": 1806885.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.671980381011963,
|
|
"epoch": 0.05961463397324438,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.6108,
|
|
"mean_token_accuracy": 0.08177201226353645,
|
|
"num_tokens": 1817403.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.65054988861084,
|
|
"epoch": 0.059969482984989886,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.5622,
|
|
"mean_token_accuracy": 0.08333921730518341,
|
|
"num_tokens": 1828929.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.580755519866943,
|
|
"epoch": 0.06032433199673539,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.5877,
|
|
"mean_token_accuracy": 0.08306344002485275,
|
|
"num_tokens": 1839387.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.6390235900878904,
|
|
"epoch": 0.06067918100848089,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.5976,
|
|
"mean_token_accuracy": 0.07957985401153564,
|
|
"num_tokens": 1850325.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.6384642124176025,
|
|
"epoch": 0.06103403002022639,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.591,
|
|
"mean_token_accuracy": 0.07911092862486839,
|
|
"num_tokens": 1861028.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.572419548034668,
|
|
"epoch": 0.0613888790319719,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.5673,
|
|
"mean_token_accuracy": 0.08450285941362382,
|
|
"num_tokens": 1872676.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.608766698837281,
|
|
"epoch": 0.061743728043717395,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.6249,
|
|
"mean_token_accuracy": 0.08118002340197564,
|
|
"num_tokens": 1883839.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.498948860168457,
|
|
"epoch": 0.0620985770554629,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.473,
|
|
"mean_token_accuracy": 0.08832938000559806,
|
|
"num_tokens": 1894805.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.565404224395752,
|
|
"epoch": 0.062453426067208405,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.4774,
|
|
"mean_token_accuracy": 0.08059397265315056,
|
|
"num_tokens": 1904770.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.594593524932861,
|
|
"epoch": 0.06280827507895391,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.534,
|
|
"mean_token_accuracy": 0.08095177710056305,
|
|
"num_tokens": 1916190.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.64970817565918,
|
|
"epoch": 0.06316312409069941,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.5897,
|
|
"mean_token_accuracy": 0.07850743383169174,
|
|
"num_tokens": 1926519.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.572882080078125,
|
|
"epoch": 0.0635179731024449,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.5959,
|
|
"mean_token_accuracy": 0.08198425024747849,
|
|
"num_tokens": 1936625.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.570544862747193,
|
|
"epoch": 0.06387282211419042,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.552,
|
|
"mean_token_accuracy": 0.08835628777742385,
|
|
"num_tokens": 1946892.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.569087743759155,
|
|
"epoch": 0.06422767112593591,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.5459,
|
|
"mean_token_accuracy": 0.0858149766921997,
|
|
"num_tokens": 1958464.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.445690727233886,
|
|
"epoch": 0.06458252013768141,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.4842,
|
|
"mean_token_accuracy": 0.08828370273113251,
|
|
"num_tokens": 1968717.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.524918699264527,
|
|
"epoch": 0.06493736914942692,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.5477,
|
|
"mean_token_accuracy": 0.08735504001379013,
|
|
"num_tokens": 1979054.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.48834342956543,
|
|
"epoch": 0.06529221816117242,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.51,
|
|
"mean_token_accuracy": 0.07926352620124817,
|
|
"num_tokens": 1991069.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.58330545425415,
|
|
"epoch": 0.06564706717291792,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.5667,
|
|
"mean_token_accuracy": 0.08001754432916641,
|
|
"num_tokens": 2002182.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.495837259292602,
|
|
"epoch": 0.06600191618466343,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.4959,
|
|
"mean_token_accuracy": 0.08842689022421837,
|
|
"num_tokens": 2012165.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.61547622680664,
|
|
"epoch": 0.06635676519640893,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.5157,
|
|
"mean_token_accuracy": 0.08534864895045757,
|
|
"num_tokens": 2022379.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.350019550323486,
|
|
"epoch": 0.06671161420815443,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.469,
|
|
"mean_token_accuracy": 0.08241500668227672,
|
|
"num_tokens": 2033497.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.599674367904663,
|
|
"epoch": 0.06706646321989994,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.5326,
|
|
"mean_token_accuracy": 0.08250435441732407,
|
|
"num_tokens": 2044914.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.5432432174682615,
|
|
"epoch": 0.06742131223164544,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.5577,
|
|
"mean_token_accuracy": 0.07833333164453507,
|
|
"num_tokens": 2056450.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.4970087051391605,
|
|
"epoch": 0.06777616124339093,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.5441,
|
|
"mean_token_accuracy": 0.08571040853857995,
|
|
"num_tokens": 2067132.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.541100406646729,
|
|
"epoch": 0.06813101025513645,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.5381,
|
|
"mean_token_accuracy": 0.084527288377285,
|
|
"num_tokens": 2077268.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.537912464141845,
|
|
"epoch": 0.06848585926688194,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.5354,
|
|
"mean_token_accuracy": 0.08061212226748467,
|
|
"num_tokens": 2088550.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.5197999477386475,
|
|
"epoch": 0.06884070827862744,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.5101,
|
|
"mean_token_accuracy": 0.08455962091684341,
|
|
"num_tokens": 2099937.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.416968822479248,
|
|
"epoch": 0.06919555729037295,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.4727,
|
|
"mean_token_accuracy": 0.08963474035263061,
|
|
"num_tokens": 2110257.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.585858726501465,
|
|
"epoch": 0.06955040630211845,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.5953,
|
|
"mean_token_accuracy": 0.07949252240359783,
|
|
"num_tokens": 2122537.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.555169153213501,
|
|
"epoch": 0.06990525531386395,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.5226,
|
|
"mean_token_accuracy": 0.08293924927711487,
|
|
"num_tokens": 2132892.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.499892282485962,
|
|
"epoch": 0.07026010432560946,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.4475,
|
|
"mean_token_accuracy": 0.08585060015320778,
|
|
"num_tokens": 2143465.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.430732774734497,
|
|
"epoch": 0.07061495333735496,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.4201,
|
|
"mean_token_accuracy": 0.08321849703788757,
|
|
"num_tokens": 2154439.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.451830673217773,
|
|
"epoch": 0.07096980234910046,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.4062,
|
|
"mean_token_accuracy": 0.08034207150340081,
|
|
"num_tokens": 2164849.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.4653853416442875,
|
|
"epoch": 0.07132465136084595,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 7.5331,
|
|
"mean_token_accuracy": 0.08420314416289329,
|
|
"num_tokens": 2174889.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.4106261253356935,
|
|
"epoch": 0.07167950037259146,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 7.443,
|
|
"mean_token_accuracy": 0.08532139733433723,
|
|
"num_tokens": 2184969.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.500397348403931,
|
|
"epoch": 0.07203434938433696,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 7.4641,
|
|
"mean_token_accuracy": 0.08998854607343673,
|
|
"num_tokens": 2194549.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.426589250564575,
|
|
"epoch": 0.07238919839608246,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 7.4863,
|
|
"mean_token_accuracy": 0.08732765316963195,
|
|
"num_tokens": 2206300.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.3821296215057375,
|
|
"epoch": 0.07274404740782797,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 7.4323,
|
|
"mean_token_accuracy": 0.08991043567657471,
|
|
"num_tokens": 2216717.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.46541018486023,
|
|
"epoch": 0.07309889641957347,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 7.4952,
|
|
"mean_token_accuracy": 0.08183322325348855,
|
|
"num_tokens": 2227047.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.472234773635864,
|
|
"epoch": 0.07345374543131897,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 7.3876,
|
|
"mean_token_accuracy": 0.08948077708482742,
|
|
"num_tokens": 2236811.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.367137098312378,
|
|
"epoch": 0.07380859444306448,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 7.4732,
|
|
"mean_token_accuracy": 0.08651004880666732,
|
|
"num_tokens": 2248949.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.470364713668824,
|
|
"epoch": 0.07416344345480998,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 7.4892,
|
|
"mean_token_accuracy": 0.09045022875070571,
|
|
"num_tokens": 2260720.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.41192946434021,
|
|
"epoch": 0.07451829246655547,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 7.4659,
|
|
"mean_token_accuracy": 0.08553651496767997,
|
|
"num_tokens": 2271855.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.329938554763794,
|
|
"epoch": 0.07487314147830099,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 7.4186,
|
|
"mean_token_accuracy": 0.08984486237168313,
|
|
"num_tokens": 2283203.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.525700807571411,
|
|
"epoch": 0.07522799049004648,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 7.4772,
|
|
"mean_token_accuracy": 0.0798021551221609,
|
|
"num_tokens": 2295881.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.448550653457642,
|
|
"epoch": 0.07558283950179198,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 7.4512,
|
|
"mean_token_accuracy": 0.08743728250265122,
|
|
"num_tokens": 2306411.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.442822980880737,
|
|
"epoch": 0.0759376885135375,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 7.4075,
|
|
"mean_token_accuracy": 0.08435096815228463,
|
|
"num_tokens": 2317519.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.448988389968872,
|
|
"epoch": 0.07629253752528299,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 7.4994,
|
|
"mean_token_accuracy": 0.08249662891030311,
|
|
"num_tokens": 2327921.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.324292278289795,
|
|
"epoch": 0.07664738653702849,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 7.3506,
|
|
"mean_token_accuracy": 0.09711327776312828,
|
|
"num_tokens": 2338925.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.464797782897949,
|
|
"epoch": 0.077002235548774,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 7.4976,
|
|
"mean_token_accuracy": 0.08892218843102455,
|
|
"num_tokens": 2349576.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.314234113693237,
|
|
"epoch": 0.0773570845605195,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 7.3705,
|
|
"mean_token_accuracy": 0.08954303413629532,
|
|
"num_tokens": 2359671.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.498252105712891,
|
|
"epoch": 0.077711933572265,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 7.4653,
|
|
"mean_token_accuracy": 0.08781479671597481,
|
|
"num_tokens": 2370173.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.377524137496948,
|
|
"epoch": 0.07806678258401051,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 7.3908,
|
|
"mean_token_accuracy": 0.08205282092094421,
|
|
"num_tokens": 2381330.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.393895387649536,
|
|
"epoch": 0.078421631595756,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 7.4112,
|
|
"mean_token_accuracy": 0.08845710456371307,
|
|
"num_tokens": 2392487.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.278064727783203,
|
|
"epoch": 0.0787764806075015,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 7.3277,
|
|
"mean_token_accuracy": 0.09402666017413139,
|
|
"num_tokens": 2404083.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.404061222076416,
|
|
"epoch": 0.07913132961924702,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 7.4268,
|
|
"mean_token_accuracy": 0.08385513573884965,
|
|
"num_tokens": 2414646.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.46454119682312,
|
|
"epoch": 0.07948617863099251,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 7.4392,
|
|
"mean_token_accuracy": 0.08458093479275704,
|
|
"num_tokens": 2425823.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.345199346542358,
|
|
"epoch": 0.07984102764273801,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 7.4252,
|
|
"mean_token_accuracy": 0.08571918532252312,
|
|
"num_tokens": 2436999.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.397942304611206,
|
|
"epoch": 0.08019587665448352,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 7.3874,
|
|
"mean_token_accuracy": 0.08777525499463082,
|
|
"num_tokens": 2448166.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.41080527305603,
|
|
"epoch": 0.08055072566622902,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 7.414,
|
|
"mean_token_accuracy": 0.08722149506211281,
|
|
"num_tokens": 2459018.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.315374708175659,
|
|
"epoch": 0.08090557467797452,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 7.3751,
|
|
"mean_token_accuracy": 0.08800306916236877,
|
|
"num_tokens": 2469113.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.370522165298462,
|
|
"epoch": 0.08126042368972003,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 7.3597,
|
|
"mean_token_accuracy": 0.09009913429617881,
|
|
"num_tokens": 2481506.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.3977306365966795,
|
|
"epoch": 0.08161527270146553,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 7.3996,
|
|
"mean_token_accuracy": 0.09371651038527488,
|
|
"num_tokens": 2491924.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.312761831283569,
|
|
"epoch": 0.08197012171321102,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 7.3336,
|
|
"mean_token_accuracy": 0.08466937094926834,
|
|
"num_tokens": 2503920.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.372441291809082,
|
|
"epoch": 0.08232497072495654,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 7.3572,
|
|
"mean_token_accuracy": 0.08929058611392975,
|
|
"num_tokens": 2514366.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.331275081634521,
|
|
"epoch": 0.08267981973670203,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 7.4348,
|
|
"mean_token_accuracy": 0.08657754585146904,
|
|
"num_tokens": 2524749.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.35433521270752,
|
|
"epoch": 0.08303466874844753,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 7.4169,
|
|
"mean_token_accuracy": 0.09042649567127228,
|
|
"num_tokens": 2536202.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.385815572738648,
|
|
"epoch": 0.08338951776019304,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 7.3481,
|
|
"mean_token_accuracy": 0.09155565276741981,
|
|
"num_tokens": 2545573.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.3866798877716064,
|
|
"epoch": 0.08374436677193854,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 7.3933,
|
|
"mean_token_accuracy": 0.09010633900761604,
|
|
"num_tokens": 2556627.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.227568435668945,
|
|
"epoch": 0.08409921578368404,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 7.3546,
|
|
"mean_token_accuracy": 0.08734624981880187,
|
|
"num_tokens": 2567064.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.4254954814910885,
|
|
"epoch": 0.08445406479542955,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 7.3938,
|
|
"mean_token_accuracy": 0.08469272255897523,
|
|
"num_tokens": 2577772.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.222232055664063,
|
|
"epoch": 0.08480891380717505,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 7.3055,
|
|
"mean_token_accuracy": 0.08891206458210946,
|
|
"num_tokens": 2588880.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.418983125686646,
|
|
"epoch": 0.08516376281892055,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 7.3926,
|
|
"mean_token_accuracy": 0.08361146301031112,
|
|
"num_tokens": 2600146.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.26065993309021,
|
|
"epoch": 0.08551861183066606,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 7.3419,
|
|
"mean_token_accuracy": 0.08811083883047104,
|
|
"num_tokens": 2610305.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.385571146011353,
|
|
"epoch": 0.08587346084241156,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 7.351,
|
|
"mean_token_accuracy": 0.08472810387611389,
|
|
"num_tokens": 2621218.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.362215375900268,
|
|
"epoch": 0.08622830985415705,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 7.4032,
|
|
"mean_token_accuracy": 0.07960698269307613,
|
|
"num_tokens": 2632399.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.4094336986541744,
|
|
"epoch": 0.08658315886590257,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 7.2759,
|
|
"mean_token_accuracy": 0.09264603853225709,
|
|
"num_tokens": 2643043.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.263467264175415,
|
|
"epoch": 0.08693800787764806,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 7.3037,
|
|
"mean_token_accuracy": 0.08933228254318237,
|
|
"num_tokens": 2653336.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.435257434844971,
|
|
"epoch": 0.08729285688939356,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 7.3784,
|
|
"mean_token_accuracy": 0.08661267906427383,
|
|
"num_tokens": 2664414.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.223228883743286,
|
|
"epoch": 0.08764770590113906,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 7.3525,
|
|
"mean_token_accuracy": 0.09211425483226776,
|
|
"num_tokens": 2674338.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.366180896759033,
|
|
"epoch": 0.08800255491288457,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 7.3397,
|
|
"mean_token_accuracy": 0.08799645379185676,
|
|
"num_tokens": 2686022.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.351464748382568,
|
|
"epoch": 0.08835740392463007,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 7.304,
|
|
"mean_token_accuracy": 0.08788969665765763,
|
|
"num_tokens": 2696746.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.238297367095948,
|
|
"epoch": 0.08871225293637557,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 7.3415,
|
|
"mean_token_accuracy": 0.08992047309875488,
|
|
"num_tokens": 2706737.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.3518078327178955,
|
|
"epoch": 0.08906710194812108,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 7.3486,
|
|
"mean_token_accuracy": 0.08713155873119831,
|
|
"num_tokens": 2717757.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.33965802192688,
|
|
"epoch": 0.08942195095986658,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 7.2968,
|
|
"mean_token_accuracy": 0.09177965298295021,
|
|
"num_tokens": 2728714.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.179767322540283,
|
|
"epoch": 0.08977679997161207,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 7.2606,
|
|
"mean_token_accuracy": 0.09344343543052673,
|
|
"num_tokens": 2739892.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.360199499130249,
|
|
"epoch": 0.09013164898335758,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 7.3165,
|
|
"mean_token_accuracy": 0.09559072181582451,
|
|
"num_tokens": 2750164.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.324847936630249,
|
|
"epoch": 0.09048649799510308,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 7.3193,
|
|
"mean_token_accuracy": 0.08874515891075134,
|
|
"num_tokens": 2760865.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.346493816375732,
|
|
"epoch": 0.09084134700684858,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 7.3471,
|
|
"mean_token_accuracy": 0.09163894280791282,
|
|
"num_tokens": 2772993.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.259117269515992,
|
|
"epoch": 0.09119619601859409,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 7.2786,
|
|
"mean_token_accuracy": 0.09321993291378021,
|
|
"num_tokens": 2782573.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.327762031555176,
|
|
"epoch": 0.09155104503033959,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 7.3555,
|
|
"mean_token_accuracy": 0.08510730862617492,
|
|
"num_tokens": 2793372.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.372673606872558,
|
|
"epoch": 0.09190589404208509,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 7.3335,
|
|
"mean_token_accuracy": 0.08849020972847939,
|
|
"num_tokens": 2805067.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.204169082641601,
|
|
"epoch": 0.0922607430538306,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 7.2726,
|
|
"mean_token_accuracy": 0.09518759399652481,
|
|
"num_tokens": 2815029.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.380017948150635,
|
|
"epoch": 0.0926155920655761,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 7.3445,
|
|
"mean_token_accuracy": 0.09246201291680337,
|
|
"num_tokens": 2825206.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.238814830780029,
|
|
"epoch": 0.0929704410773216,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 7.3075,
|
|
"mean_token_accuracy": 0.09198266863822938,
|
|
"num_tokens": 2837134.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.282371139526367,
|
|
"epoch": 0.0933252900890671,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 7.261,
|
|
"mean_token_accuracy": 0.09279094189405442,
|
|
"num_tokens": 2848085.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.297031307220459,
|
|
"epoch": 0.0936801391008126,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 7.2988,
|
|
"mean_token_accuracy": 0.092452073097229,
|
|
"num_tokens": 2858270.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.190067005157471,
|
|
"epoch": 0.0940349881125581,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 7.1873,
|
|
"mean_token_accuracy": 0.09650815352797508,
|
|
"num_tokens": 2868319.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.2879091739654545,
|
|
"epoch": 0.09438983712430361,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 7.2142,
|
|
"mean_token_accuracy": 0.09751713499426842,
|
|
"num_tokens": 2879042.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.145177888870239,
|
|
"epoch": 0.09474468613604911,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 7.2146,
|
|
"mean_token_accuracy": 0.09119046702980996,
|
|
"num_tokens": 2890687.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.305484199523926,
|
|
"epoch": 0.09509953514779461,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 7.2883,
|
|
"mean_token_accuracy": 0.0910595864057541,
|
|
"num_tokens": 2900901.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.299550008773804,
|
|
"epoch": 0.09545438415954012,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 7.2978,
|
|
"mean_token_accuracy": 0.09110193699598312,
|
|
"num_tokens": 2912233.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.180187082290649,
|
|
"epoch": 0.09580923317128562,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 7.1896,
|
|
"mean_token_accuracy": 0.09826849699020386,
|
|
"num_tokens": 2923002.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.222209358215332,
|
|
"epoch": 0.09616408218303112,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 7.2904,
|
|
"mean_token_accuracy": 0.09233295768499375,
|
|
"num_tokens": 2934139.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.296793508529663,
|
|
"epoch": 0.09651893119477663,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 7.264,
|
|
"mean_token_accuracy": 0.09283363372087479,
|
|
"num_tokens": 2945229.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.259587001800537,
|
|
"epoch": 0.09687378020652213,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 7.2666,
|
|
"mean_token_accuracy": 0.09352016374468804,
|
|
"num_tokens": 2956118.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.230564737319947,
|
|
"epoch": 0.09722862921826762,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 7.1702,
|
|
"mean_token_accuracy": 0.09822328612208367,
|
|
"num_tokens": 2967120.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.109541511535644,
|
|
"epoch": 0.09758347823001313,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 7.2177,
|
|
"mean_token_accuracy": 0.0937571682035923,
|
|
"num_tokens": 2977852.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.361109018325806,
|
|
"epoch": 0.09793832724175863,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 7.3398,
|
|
"mean_token_accuracy": 0.08829166814684868,
|
|
"num_tokens": 2988076.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.184476566314697,
|
|
"epoch": 0.09829317625350413,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 7.1971,
|
|
"mean_token_accuracy": 0.09840201437473298,
|
|
"num_tokens": 2999159.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.314798021316529,
|
|
"epoch": 0.09864802526524964,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 7.2446,
|
|
"mean_token_accuracy": 0.09077045172452927,
|
|
"num_tokens": 3009419.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.1688261985778805,
|
|
"epoch": 0.09900287427699514,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 7.2273,
|
|
"mean_token_accuracy": 0.09468511268496513,
|
|
"num_tokens": 3020261.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.288452386856079,
|
|
"epoch": 0.09935772328874064,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 7.264,
|
|
"mean_token_accuracy": 0.0897028960287571,
|
|
"num_tokens": 3031118.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.270044612884521,
|
|
"epoch": 0.09971257230048615,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 7.303,
|
|
"mean_token_accuracy": 0.09120242893695832,
|
|
"num_tokens": 3041921.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.301294279098511,
|
|
"epoch": 0.10006742131223165,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 7.2724,
|
|
"mean_token_accuracy": 0.0923406146466732,
|
|
"num_tokens": 3053935.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.298661994934082,
|
|
"epoch": 0.10042227032397714,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 7.2681,
|
|
"mean_token_accuracy": 0.09436995908617973,
|
|
"num_tokens": 3065260.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.141231632232666,
|
|
"epoch": 0.10077711933572266,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 7.212,
|
|
"mean_token_accuracy": 0.09240631237626076,
|
|
"num_tokens": 3076012.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.3381248950958256,
|
|
"epoch": 0.10113196834746815,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 7.2001,
|
|
"mean_token_accuracy": 0.09209805354475975,
|
|
"num_tokens": 3086517.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.143636035919189,
|
|
"epoch": 0.10148681735921365,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 7.2568,
|
|
"mean_token_accuracy": 0.09073155596852303,
|
|
"num_tokens": 3097511.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.358057594299316,
|
|
"epoch": 0.10184166637095916,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 7.2242,
|
|
"mean_token_accuracy": 0.08778738155961037,
|
|
"num_tokens": 3108482.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.192478942871094,
|
|
"epoch": 0.10219651538270466,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 7.2047,
|
|
"mean_token_accuracy": 0.09349827691912652,
|
|
"num_tokens": 3119630.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.235249710083008,
|
|
"epoch": 0.10255136439445016,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 7.2685,
|
|
"mean_token_accuracy": 0.08921699076890946,
|
|
"num_tokens": 3131050.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.221825456619262,
|
|
"epoch": 0.10290621340619566,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 7.2587,
|
|
"mean_token_accuracy": 0.09253018423914909,
|
|
"num_tokens": 3141956.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.341269016265869,
|
|
"epoch": 0.10326106241794117,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 7.2215,
|
|
"mean_token_accuracy": 0.09548219069838523,
|
|
"num_tokens": 3152693.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.267981004714966,
|
|
"epoch": 0.10361591142968667,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 7.2577,
|
|
"mean_token_accuracy": 0.0954539954662323,
|
|
"num_tokens": 3163736.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.165822839736938,
|
|
"epoch": 0.10397076044143216,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 7.2569,
|
|
"mean_token_accuracy": 0.08883126974105834,
|
|
"num_tokens": 3175537.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.244705867767334,
|
|
"epoch": 0.10432560945317768,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 7.1729,
|
|
"mean_token_accuracy": 0.09741863459348679,
|
|
"num_tokens": 3185659.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.163289546966553,
|
|
"epoch": 0.10468045846492317,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 7.0888,
|
|
"mean_token_accuracy": 0.10371233746409417,
|
|
"num_tokens": 3196548.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.239425230026245,
|
|
"epoch": 0.10503530747666867,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 7.1853,
|
|
"mean_token_accuracy": 0.09878996461629867,
|
|
"num_tokens": 3207516.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.1664224624633786,
|
|
"epoch": 0.10539015648841418,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 7.2059,
|
|
"mean_token_accuracy": 0.09172281622886658,
|
|
"num_tokens": 3219171.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.128496742248535,
|
|
"epoch": 0.10574500550015968,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 7.1106,
|
|
"mean_token_accuracy": 0.10128247514367103,
|
|
"num_tokens": 3230911.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.15082368850708,
|
|
"epoch": 0.10609985451190518,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 7.2156,
|
|
"mean_token_accuracy": 0.09450112357735634,
|
|
"num_tokens": 3240977.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.310815048217774,
|
|
"epoch": 0.10645470352365069,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 7.2455,
|
|
"mean_token_accuracy": 0.09728037863969803,
|
|
"num_tokens": 3251714.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 4757686972416000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|