6035 lines
164 KiB
JSON
6035 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.21290940704730138,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 7.228920888900757,
|
|
"epoch": 0.10680955253539619,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 7.2048,
|
|
"mean_token_accuracy": 0.09683412909507752,
|
|
"num_tokens": 3262558.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.222548103332519,
|
|
"epoch": 0.10716440154714169,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 7.1702,
|
|
"mean_token_accuracy": 0.09433976039290429,
|
|
"num_tokens": 3273473.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.1861412525177,
|
|
"epoch": 0.1075192505588872,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 7.2912,
|
|
"mean_token_accuracy": 0.09852436482906342,
|
|
"num_tokens": 3284238.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.240938949584961,
|
|
"epoch": 0.1078740995706327,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 7.1773,
|
|
"mean_token_accuracy": 0.10066472664475441,
|
|
"num_tokens": 3294848.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.1551109790802006,
|
|
"epoch": 0.10822894858237819,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 7.1802,
|
|
"mean_token_accuracy": 0.09828831255435944,
|
|
"num_tokens": 3305487.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.182540702819824,
|
|
"epoch": 0.1085837975941237,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 7.1249,
|
|
"mean_token_accuracy": 0.09082185253500938,
|
|
"num_tokens": 3317032.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.226040554046631,
|
|
"epoch": 0.1089386466058692,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 7.1868,
|
|
"mean_token_accuracy": 0.09265627264976502,
|
|
"num_tokens": 3327695.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.1324906826019285,
|
|
"epoch": 0.1092934956176147,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 7.1214,
|
|
"mean_token_accuracy": 0.09917151257395744,
|
|
"num_tokens": 3337445.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.162045288085937,
|
|
"epoch": 0.10964834462936021,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 7.1469,
|
|
"mean_token_accuracy": 0.09252285063266755,
|
|
"num_tokens": 3347958.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.268742990493775,
|
|
"epoch": 0.11000319364110571,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 7.103,
|
|
"mean_token_accuracy": 0.09710876047611236,
|
|
"num_tokens": 3357974.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.088126087188721,
|
|
"epoch": 0.11035804265285121,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 7.1715,
|
|
"mean_token_accuracy": 0.09461688697338104,
|
|
"num_tokens": 3368901.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.292183065414429,
|
|
"epoch": 0.11071289166459672,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 7.1501,
|
|
"mean_token_accuracy": 0.09700253531336785,
|
|
"num_tokens": 3379028.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.114231586456299,
|
|
"epoch": 0.11106774067634222,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 7.2125,
|
|
"mean_token_accuracy": 0.0952233262360096,
|
|
"num_tokens": 3389688.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.312720584869385,
|
|
"epoch": 0.11142258968808771,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 7.2162,
|
|
"mean_token_accuracy": 0.0928651139140129,
|
|
"num_tokens": 3400782.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.103663063049316,
|
|
"epoch": 0.11177743869983323,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 7.1171,
|
|
"mean_token_accuracy": 0.0949582502245903,
|
|
"num_tokens": 3412404.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.07365198135376,
|
|
"epoch": 0.11213228771157872,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 7.1578,
|
|
"mean_token_accuracy": 0.09038265198469161,
|
|
"num_tokens": 3424129.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.2252601146697994,
|
|
"epoch": 0.11248713672332422,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 7.1011,
|
|
"mean_token_accuracy": 0.09262050092220306,
|
|
"num_tokens": 3436570.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.137156534194946,
|
|
"epoch": 0.11284198573506973,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 7.1351,
|
|
"mean_token_accuracy": 0.09676287397742271,
|
|
"num_tokens": 3447222.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.170758247375488,
|
|
"epoch": 0.11319683474681523,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 7.1081,
|
|
"mean_token_accuracy": 0.096658343821764,
|
|
"num_tokens": 3458921.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.246716260910034,
|
|
"epoch": 0.11355168375856073,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 7.2579,
|
|
"mean_token_accuracy": 0.08915452212095261,
|
|
"num_tokens": 3470065.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.13914909362793,
|
|
"epoch": 0.11390653277030624,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 7.0704,
|
|
"mean_token_accuracy": 0.10325511917471886,
|
|
"num_tokens": 3481062.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.162566661834717,
|
|
"epoch": 0.11426138178205174,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 7.0767,
|
|
"mean_token_accuracy": 0.09748337939381599,
|
|
"num_tokens": 3491586.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.1580146789550785,
|
|
"epoch": 0.11461623079379724,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 7.1824,
|
|
"mean_token_accuracy": 0.0963168665766716,
|
|
"num_tokens": 3502367.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.106345272064209,
|
|
"epoch": 0.11497107980554275,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 7.0516,
|
|
"mean_token_accuracy": 0.10276207253336907,
|
|
"num_tokens": 3512114.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.237199068069458,
|
|
"epoch": 0.11532592881728825,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 7.2058,
|
|
"mean_token_accuracy": 0.09948946833610535,
|
|
"num_tokens": 3522954.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.1651527881622314,
|
|
"epoch": 0.11568077782903374,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 7.1538,
|
|
"mean_token_accuracy": 0.09592896923422814,
|
|
"num_tokens": 3533205.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.125106525421143,
|
|
"epoch": 0.11603562684077925,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 7.058,
|
|
"mean_token_accuracy": 0.0978856511414051,
|
|
"num_tokens": 3542901.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.144106674194336,
|
|
"epoch": 0.11639047585252475,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 7.1462,
|
|
"mean_token_accuracy": 0.09332720711827278,
|
|
"num_tokens": 3554141.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.1256311416625975,
|
|
"epoch": 0.11674532486427025,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 7.0938,
|
|
"mean_token_accuracy": 0.10101081579923629,
|
|
"num_tokens": 3566074.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.188183689117432,
|
|
"epoch": 0.11710017387601576,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 7.1298,
|
|
"mean_token_accuracy": 0.09676677137613296,
|
|
"num_tokens": 3576939.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.127859163284302,
|
|
"epoch": 0.11745502288776126,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 7.1605,
|
|
"mean_token_accuracy": 0.09790143445134163,
|
|
"num_tokens": 3587871.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.06299614906311,
|
|
"epoch": 0.11780987189950676,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 7.0346,
|
|
"mean_token_accuracy": 0.10373768284916877,
|
|
"num_tokens": 3597597.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.241457796096801,
|
|
"epoch": 0.11816472091125227,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 7.1148,
|
|
"mean_token_accuracy": 0.09652233868837357,
|
|
"num_tokens": 3607646.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.178923034667969,
|
|
"epoch": 0.11851956992299777,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 7.1293,
|
|
"mean_token_accuracy": 0.09854810833930969,
|
|
"num_tokens": 3618795.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.151714897155761,
|
|
"epoch": 0.11887441893474326,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 7.1202,
|
|
"mean_token_accuracy": 0.09609093591570854,
|
|
"num_tokens": 3629347.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.126868963241577,
|
|
"epoch": 0.11922926794648876,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 7.0511,
|
|
"mean_token_accuracy": 0.09984703361988068,
|
|
"num_tokens": 3638849.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.176082944869995,
|
|
"epoch": 0.11958411695823427,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 7.1147,
|
|
"mean_token_accuracy": 0.10202176421880722,
|
|
"num_tokens": 3649630.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.105949640274048,
|
|
"epoch": 0.11993896596997977,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 7.165,
|
|
"mean_token_accuracy": 0.09765942394733429,
|
|
"num_tokens": 3660885.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.150793170928955,
|
|
"epoch": 0.12029381498172527,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 7.1008,
|
|
"mean_token_accuracy": 0.09718259535729885,
|
|
"num_tokens": 3672714.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.211589431762695,
|
|
"epoch": 0.12064866399347078,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 7.0464,
|
|
"mean_token_accuracy": 0.09509608298540115,
|
|
"num_tokens": 3683393.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 7.014157676696778,
|
|
"epoch": 0.12100351300521628,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 7.1575,
|
|
"mean_token_accuracy": 0.09731999039649963,
|
|
"num_tokens": 3694158.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.235012865066528,
|
|
"epoch": 0.12135836201696178,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 7.0723,
|
|
"mean_token_accuracy": 0.10333672091364861,
|
|
"num_tokens": 3704334.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.064513492584228,
|
|
"epoch": 0.12171321102870729,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 7.0853,
|
|
"mean_token_accuracy": 0.10058822184801101,
|
|
"num_tokens": 3714492.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.091410732269287,
|
|
"epoch": 0.12206806004045279,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 7.0897,
|
|
"mean_token_accuracy": 0.08864482194185257,
|
|
"num_tokens": 3726138.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.256516599655152,
|
|
"epoch": 0.12242290905219828,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 7.1157,
|
|
"mean_token_accuracy": 0.09941046983003617,
|
|
"num_tokens": 3735382.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.111510515213013,
|
|
"epoch": 0.1227777580639438,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 7.0708,
|
|
"mean_token_accuracy": 0.0973458357155323,
|
|
"num_tokens": 3746160.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.065353298187256,
|
|
"epoch": 0.1231326070756893,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 7.0785,
|
|
"mean_token_accuracy": 0.09655671492218972,
|
|
"num_tokens": 3757679.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.209886121749878,
|
|
"epoch": 0.12348745608743479,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 7.084,
|
|
"mean_token_accuracy": 0.09250145927071571,
|
|
"num_tokens": 3767514.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.113470029830933,
|
|
"epoch": 0.1238423050991803,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 7.0898,
|
|
"mean_token_accuracy": 0.09914448335766793,
|
|
"num_tokens": 3778396.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.160091066360474,
|
|
"epoch": 0.1241971541109258,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 7.1196,
|
|
"mean_token_accuracy": 0.09739524349570275,
|
|
"num_tokens": 3789351.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.046756172180176,
|
|
"epoch": 0.1245520031226713,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 7.0865,
|
|
"mean_token_accuracy": 0.09295786619186401,
|
|
"num_tokens": 3799709.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.234890604019165,
|
|
"epoch": 0.12490685213441681,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 7.0903,
|
|
"mean_token_accuracy": 0.09519705399870873,
|
|
"num_tokens": 3810633.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.040793085098267,
|
|
"epoch": 0.1252617011461623,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 7.1194,
|
|
"mean_token_accuracy": 0.09463633596897125,
|
|
"num_tokens": 3821984.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.0925640106201175,
|
|
"epoch": 0.12561655015790782,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 7.0725,
|
|
"mean_token_accuracy": 0.10138425305485725,
|
|
"num_tokens": 3832488.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 7.049199962615967,
|
|
"epoch": 0.1259713991696533,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.9459,
|
|
"mean_token_accuracy": 0.11415167003870011,
|
|
"num_tokens": 3842329.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.090542364120483,
|
|
"epoch": 0.12632624818139881,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 7.0909,
|
|
"mean_token_accuracy": 0.10182333439588546,
|
|
"num_tokens": 3853585.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.182557058334351,
|
|
"epoch": 0.12668109719314433,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 7.0519,
|
|
"mean_token_accuracy": 0.1040547601878643,
|
|
"num_tokens": 3863538.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 7.089897584915161,
|
|
"epoch": 0.1270359462048898,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 7.1704,
|
|
"mean_token_accuracy": 0.09324254468083382,
|
|
"num_tokens": 3874277.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.142627429962158,
|
|
"epoch": 0.12739079521663532,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 7.0251,
|
|
"mean_token_accuracy": 0.10115873888134956,
|
|
"num_tokens": 3884591.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 7.1000548839569095,
|
|
"epoch": 0.12774564422838083,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 7.1397,
|
|
"mean_token_accuracy": 0.09500750824809075,
|
|
"num_tokens": 3895199.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.094432258605957,
|
|
"epoch": 0.12810049324012632,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.9529,
|
|
"mean_token_accuracy": 0.10029699578881264,
|
|
"num_tokens": 3907047.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 7.037583446502685,
|
|
"epoch": 0.12845534225187183,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 7.0352,
|
|
"mean_token_accuracy": 0.10667306259274482,
|
|
"num_tokens": 3918385.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.121558809280396,
|
|
"epoch": 0.12881019126361734,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 7.0685,
|
|
"mean_token_accuracy": 0.10168422237038613,
|
|
"num_tokens": 3928468.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 7.143169403076172,
|
|
"epoch": 0.12916504027536282,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 7.0534,
|
|
"mean_token_accuracy": 0.09920291230082512,
|
|
"num_tokens": 3939530.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.0373697757720945,
|
|
"epoch": 0.12951988928710834,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 7.0355,
|
|
"mean_token_accuracy": 0.0987101249396801,
|
|
"num_tokens": 3950146.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.019313764572144,
|
|
"epoch": 0.12987473829885385,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 7.0325,
|
|
"mean_token_accuracy": 0.0973743423819542,
|
|
"num_tokens": 3961499.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.121405696868896,
|
|
"epoch": 0.13022958731059933,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 7.0472,
|
|
"mean_token_accuracy": 0.10192081406712532,
|
|
"num_tokens": 3972724.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.079566621780396,
|
|
"epoch": 0.13058443632234484,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 7.0077,
|
|
"mean_token_accuracy": 0.1040501557290554,
|
|
"num_tokens": 3983457.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 7.119902420043945,
|
|
"epoch": 0.13093928533409036,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 7.0384,
|
|
"mean_token_accuracy": 0.09878516718745231,
|
|
"num_tokens": 3994581.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.0431475162506105,
|
|
"epoch": 0.13129413434583584,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 7.054,
|
|
"mean_token_accuracy": 0.10154439359903336,
|
|
"num_tokens": 4005940.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.145562553405762,
|
|
"epoch": 0.13164898335758135,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 7.0096,
|
|
"mean_token_accuracy": 0.109462571144104,
|
|
"num_tokens": 4015961.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 7.093667459487915,
|
|
"epoch": 0.13200383236932686,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 7.0528,
|
|
"mean_token_accuracy": 0.10200647190213204,
|
|
"num_tokens": 4025498.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.101914739608764,
|
|
"epoch": 0.13235868138107235,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 7.021,
|
|
"mean_token_accuracy": 0.09907697066664696,
|
|
"num_tokens": 4036123.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 7.147734260559082,
|
|
"epoch": 0.13271353039281786,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 7.0086,
|
|
"mean_token_accuracy": 0.10064524784684181,
|
|
"num_tokens": 4046915.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 7.052229022979736,
|
|
"epoch": 0.13306837940456337,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 7.0852,
|
|
"mean_token_accuracy": 0.09829342514276504,
|
|
"num_tokens": 4058050.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.142696905136108,
|
|
"epoch": 0.13342322841630885,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 7.0328,
|
|
"mean_token_accuracy": 0.1022889830172062,
|
|
"num_tokens": 4069184.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.088444232940674,
|
|
"epoch": 0.13377807742805436,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 7.0415,
|
|
"mean_token_accuracy": 0.10254081562161446,
|
|
"num_tokens": 4079140.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 7.028472805023194,
|
|
"epoch": 0.13413292643979988,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.973,
|
|
"mean_token_accuracy": 0.10482171401381493,
|
|
"num_tokens": 4088905.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 7.023287677764893,
|
|
"epoch": 0.13448777545154536,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 7.0046,
|
|
"mean_token_accuracy": 0.10013890415430068,
|
|
"num_tokens": 4099425.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 7.070297384262085,
|
|
"epoch": 0.13484262446329087,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 6.9933,
|
|
"mean_token_accuracy": 0.10154245495796203,
|
|
"num_tokens": 4109313.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 7.08957085609436,
|
|
"epoch": 0.13519747347503638,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 7.0308,
|
|
"mean_token_accuracy": 0.1028820551931858,
|
|
"num_tokens": 4120326.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 7.095383787155152,
|
|
"epoch": 0.13555232248678187,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 7.0657,
|
|
"mean_token_accuracy": 0.10205548778176307,
|
|
"num_tokens": 4131001.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 7.074268007278443,
|
|
"epoch": 0.13590717149852738,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 7.0221,
|
|
"mean_token_accuracy": 0.10284601524472237,
|
|
"num_tokens": 4142565.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 7.059752988815307,
|
|
"epoch": 0.1362620205102729,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 7.0084,
|
|
"mean_token_accuracy": 0.09651347175240517,
|
|
"num_tokens": 4153453.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 7.082843971252442,
|
|
"epoch": 0.13661686952201837,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 7.0471,
|
|
"mean_token_accuracy": 0.10347581580281258,
|
|
"num_tokens": 4164445.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 7.052491092681885,
|
|
"epoch": 0.1369717185337639,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.9734,
|
|
"mean_token_accuracy": 0.10101649984717369,
|
|
"num_tokens": 4175221.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 7.1004551410675045,
|
|
"epoch": 0.1373265675455094,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 6.9814,
|
|
"mean_token_accuracy": 0.09965018033981324,
|
|
"num_tokens": 4185686.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.955038690567017,
|
|
"epoch": 0.13768141655725488,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.9261,
|
|
"mean_token_accuracy": 0.10875649526715278,
|
|
"num_tokens": 4196217.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 7.081155776977539,
|
|
"epoch": 0.1380362655690004,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 7.064,
|
|
"mean_token_accuracy": 0.10312055200338363,
|
|
"num_tokens": 4206820.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 7.1204061031341555,
|
|
"epoch": 0.1383911145807459,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 7.0435,
|
|
"mean_token_accuracy": 0.10409965664148331,
|
|
"num_tokens": 4217855.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 7.0895867347717285,
|
|
"epoch": 0.1387459635924914,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 7.0694,
|
|
"mean_token_accuracy": 0.09884442016482353,
|
|
"num_tokens": 4229654.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 7.1512257099151615,
|
|
"epoch": 0.1391008126042369,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.9793,
|
|
"mean_token_accuracy": 0.10423392802476883,
|
|
"num_tokens": 4240929.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 7.06306209564209,
|
|
"epoch": 0.1394556616159824,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 7.0795,
|
|
"mean_token_accuracy": 0.10174657627940178,
|
|
"num_tokens": 4252605.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 7.071248960494995,
|
|
"epoch": 0.1398105106277279,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 6.9608,
|
|
"mean_token_accuracy": 0.10477321967482567,
|
|
"num_tokens": 4263690.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 7.039036417007447,
|
|
"epoch": 0.1401653596394734,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 6.986,
|
|
"mean_token_accuracy": 0.09907544031739235,
|
|
"num_tokens": 4274662.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 7.023434638977051,
|
|
"epoch": 0.14052020865121892,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 6.9599,
|
|
"mean_token_accuracy": 0.10383629649877549,
|
|
"num_tokens": 4286231.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 7.086272573471069,
|
|
"epoch": 0.1408750576629644,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.9781,
|
|
"mean_token_accuracy": 0.1037204198539257,
|
|
"num_tokens": 4297489.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 7.031176328659058,
|
|
"epoch": 0.14122990667470992,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 7.0469,
|
|
"mean_token_accuracy": 0.10251927375793457,
|
|
"num_tokens": 4308164.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 7.13521671295166,
|
|
"epoch": 0.14158475568645543,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 7.0006,
|
|
"mean_token_accuracy": 0.10318920761346817,
|
|
"num_tokens": 4319209.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.968696165084839,
|
|
"epoch": 0.1419396046982009,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 6.9492,
|
|
"mean_token_accuracy": 0.10085738375782967,
|
|
"num_tokens": 4329816.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 7.155592060089111,
|
|
"epoch": 0.14229445370994642,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 7.0195,
|
|
"mean_token_accuracy": 0.10258031487464905,
|
|
"num_tokens": 4340752.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 7.004485082626343,
|
|
"epoch": 0.1426493027216919,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 6.9452,
|
|
"mean_token_accuracy": 0.10718585550785065,
|
|
"num_tokens": 4352299.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 7.016558504104614,
|
|
"epoch": 0.14300415173343742,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 6.9894,
|
|
"mean_token_accuracy": 0.10042356625199318,
|
|
"num_tokens": 4363314.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 7.06671667098999,
|
|
"epoch": 0.14335900074518293,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 6.9807,
|
|
"mean_token_accuracy": 0.094207713752985,
|
|
"num_tokens": 4374741.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 7.052701568603515,
|
|
"epoch": 0.1437138497569284,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 6.9085,
|
|
"mean_token_accuracy": 0.10400079190731049,
|
|
"num_tokens": 4385496.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.985438394546509,
|
|
"epoch": 0.14406869876867393,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 7.0456,
|
|
"mean_token_accuracy": 0.0999737560749054,
|
|
"num_tokens": 4396411.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 7.02910966873169,
|
|
"epoch": 0.14442354778041944,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 6.931,
|
|
"mean_token_accuracy": 0.10462129339575768,
|
|
"num_tokens": 4406299.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.981289100646973,
|
|
"epoch": 0.14477839679216492,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 6.9247,
|
|
"mean_token_accuracy": 0.10906698405742646,
|
|
"num_tokens": 4417843.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 7.01538462638855,
|
|
"epoch": 0.14513324580391043,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 6.9303,
|
|
"mean_token_accuracy": 0.10332909226417542,
|
|
"num_tokens": 4428402.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.98654146194458,
|
|
"epoch": 0.14548809481565594,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 6.9306,
|
|
"mean_token_accuracy": 0.10650690644979477,
|
|
"num_tokens": 4439179.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 7.098060894012451,
|
|
"epoch": 0.14584294382740143,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 7.0222,
|
|
"mean_token_accuracy": 0.10229001268744468,
|
|
"num_tokens": 4450220.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 7.02388334274292,
|
|
"epoch": 0.14619779283914694,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 6.9459,
|
|
"mean_token_accuracy": 0.09994500055909157,
|
|
"num_tokens": 4460861.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 7.0822837352752686,
|
|
"epoch": 0.14655264185089245,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 6.9768,
|
|
"mean_token_accuracy": 0.10057543143630028,
|
|
"num_tokens": 4472241.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 7.131308603286743,
|
|
"epoch": 0.14690749086263793,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 7.0109,
|
|
"mean_token_accuracy": 0.099363424628973,
|
|
"num_tokens": 4483682.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 7.054377174377441,
|
|
"epoch": 0.14726233987438345,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 6.9847,
|
|
"mean_token_accuracy": 0.10540137887001037,
|
|
"num_tokens": 4494175.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.972439384460449,
|
|
"epoch": 0.14761718888612896,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 6.7905,
|
|
"mean_token_accuracy": 0.09852890223264694,
|
|
"num_tokens": 4504781.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 7.004600477218628,
|
|
"epoch": 0.14797203789787444,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 6.9715,
|
|
"mean_token_accuracy": 0.09647675678133964,
|
|
"num_tokens": 4516246.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 7.0212174415588375,
|
|
"epoch": 0.14832688690961995,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 6.8931,
|
|
"mean_token_accuracy": 0.10757157430052758,
|
|
"num_tokens": 4526678.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.9830567836761475,
|
|
"epoch": 0.14868173592136547,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 6.9876,
|
|
"mean_token_accuracy": 0.10509380027651787,
|
|
"num_tokens": 4537554.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.98748779296875,
|
|
"epoch": 0.14903658493311095,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 6.9897,
|
|
"mean_token_accuracy": 0.09809067100286484,
|
|
"num_tokens": 4548226.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 7.235601758956909,
|
|
"epoch": 0.14939143394485646,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 7.0464,
|
|
"mean_token_accuracy": 0.09804995954036713,
|
|
"num_tokens": 4560836.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 7.032708978652954,
|
|
"epoch": 0.14974628295660197,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 6.9384,
|
|
"mean_token_accuracy": 0.10675909146666526,
|
|
"num_tokens": 4572473.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.9664177894592285,
|
|
"epoch": 0.15010113196834746,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 6.995,
|
|
"mean_token_accuracy": 0.09995530992746353,
|
|
"num_tokens": 4584029.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 7.070379304885864,
|
|
"epoch": 0.15045598098009297,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 6.9981,
|
|
"mean_token_accuracy": 0.09843715131282807,
|
|
"num_tokens": 4595539.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 7.080266904830933,
|
|
"epoch": 0.15081082999183848,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 6.9609,
|
|
"mean_token_accuracy": 0.10313019677996635,
|
|
"num_tokens": 4606914.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 7.061946630477905,
|
|
"epoch": 0.15116567900358396,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 6.9116,
|
|
"mean_token_accuracy": 0.10261239334940911,
|
|
"num_tokens": 4618100.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.969537210464478,
|
|
"epoch": 0.15152052801532948,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 6.925,
|
|
"mean_token_accuracy": 0.1127280555665493,
|
|
"num_tokens": 4628896.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 7.063964605331421,
|
|
"epoch": 0.151875377027075,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 6.9457,
|
|
"mean_token_accuracy": 0.10601773262023925,
|
|
"num_tokens": 4639280.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.949308109283447,
|
|
"epoch": 0.15223022603882047,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 6.8887,
|
|
"mean_token_accuracy": 0.10163175389170646,
|
|
"num_tokens": 4649687.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 7.031176376342773,
|
|
"epoch": 0.15258507505056598,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 6.9478,
|
|
"mean_token_accuracy": 0.10624543651938438,
|
|
"num_tokens": 4661721.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 7.0959258556365965,
|
|
"epoch": 0.1529399240623115,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 6.9916,
|
|
"mean_token_accuracy": 0.09919418841600418,
|
|
"num_tokens": 4673648.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 7.012122440338135,
|
|
"epoch": 0.15329477307405698,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 6.9773,
|
|
"mean_token_accuracy": 0.10032318979501724,
|
|
"num_tokens": 4684786.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 7.042158746719361,
|
|
"epoch": 0.1536496220858025,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 6.9531,
|
|
"mean_token_accuracy": 0.09853310063481331,
|
|
"num_tokens": 4695353.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.978724813461303,
|
|
"epoch": 0.154004471097548,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 6.8553,
|
|
"mean_token_accuracy": 0.11324087306857109,
|
|
"num_tokens": 4705719.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 7.0046632289886475,
|
|
"epoch": 0.15435932010929349,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 7.0269,
|
|
"mean_token_accuracy": 0.09866114109754562,
|
|
"num_tokens": 4718978.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 7.1116374969482425,
|
|
"epoch": 0.154714169121039,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 6.88,
|
|
"mean_token_accuracy": 0.11316526606678963,
|
|
"num_tokens": 4729099.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.8555004596710205,
|
|
"epoch": 0.1550690181327845,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 6.9688,
|
|
"mean_token_accuracy": 0.09804181829094887,
|
|
"num_tokens": 4740508.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 7.0752723693847654,
|
|
"epoch": 0.15542386714453,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 6.8949,
|
|
"mean_token_accuracy": 0.11416682377457618,
|
|
"num_tokens": 4752142.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 7.102331590652466,
|
|
"epoch": 0.1557787161562755,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 6.9794,
|
|
"mean_token_accuracy": 0.10389717966318131,
|
|
"num_tokens": 4762708.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.950309753417969,
|
|
"epoch": 0.15613356516802102,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 6.8937,
|
|
"mean_token_accuracy": 0.11124820932745934,
|
|
"num_tokens": 4773239.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.958485841751099,
|
|
"epoch": 0.1564884141797665,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 6.8641,
|
|
"mean_token_accuracy": 0.10944292694330215,
|
|
"num_tokens": 4783489.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.992775583267212,
|
|
"epoch": 0.156843263191512,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 6.8619,
|
|
"mean_token_accuracy": 0.11418202966451645,
|
|
"num_tokens": 4794425.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 7.0300380229949955,
|
|
"epoch": 0.15719811220325752,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 6.9179,
|
|
"mean_token_accuracy": 0.10028508976101876,
|
|
"num_tokens": 4806391.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.990481758117676,
|
|
"epoch": 0.157552961215003,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 6.8839,
|
|
"mean_token_accuracy": 0.10272055044770241,
|
|
"num_tokens": 4816689.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 7.015410566329956,
|
|
"epoch": 0.15790781022674852,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 6.9709,
|
|
"mean_token_accuracy": 0.09755917191505432,
|
|
"num_tokens": 4827892.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 7.008429431915284,
|
|
"epoch": 0.15826265923849403,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 6.8583,
|
|
"mean_token_accuracy": 0.11253580972552299,
|
|
"num_tokens": 4838174.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 7.028060054779052,
|
|
"epoch": 0.15861750825023951,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 6.9501,
|
|
"mean_token_accuracy": 0.10639444887638091,
|
|
"num_tokens": 4849390.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 7.012446880340576,
|
|
"epoch": 0.15897235726198503,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 6.9322,
|
|
"mean_token_accuracy": 0.1081770859658718,
|
|
"num_tokens": 4859130.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.958835411071777,
|
|
"epoch": 0.15932720627373054,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 6.7842,
|
|
"mean_token_accuracy": 0.1121189422905445,
|
|
"num_tokens": 4869992.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.9395585536956785,
|
|
"epoch": 0.15968205528547602,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 6.8761,
|
|
"mean_token_accuracy": 0.10806992277503014,
|
|
"num_tokens": 4880178.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 7.033035564422607,
|
|
"epoch": 0.16003690429722153,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 6.9967,
|
|
"mean_token_accuracy": 0.10657852441072464,
|
|
"num_tokens": 4892521.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 7.062964677810669,
|
|
"epoch": 0.16039175330896704,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 6.8279,
|
|
"mean_token_accuracy": 0.10873689278960227,
|
|
"num_tokens": 4904015.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 7.0085608005523685,
|
|
"epoch": 0.16074660232071253,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 6.9017,
|
|
"mean_token_accuracy": 0.11354547590017319,
|
|
"num_tokens": 4914806.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.913502645492554,
|
|
"epoch": 0.16110145133245804,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 6.9735,
|
|
"mean_token_accuracy": 0.10321487337350846,
|
|
"num_tokens": 4926590.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 7.050243616104126,
|
|
"epoch": 0.16145630034420355,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 6.8725,
|
|
"mean_token_accuracy": 0.10716884061694146,
|
|
"num_tokens": 4937212.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 7.00547342300415,
|
|
"epoch": 0.16181114935594904,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 6.9122,
|
|
"mean_token_accuracy": 0.10085045397281647,
|
|
"num_tokens": 4948059.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 7.020096683502198,
|
|
"epoch": 0.16216599836769455,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 6.9249,
|
|
"mean_token_accuracy": 0.10465004742145538,
|
|
"num_tokens": 4958962.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.947905540466309,
|
|
"epoch": 0.16252084737944006,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 6.7898,
|
|
"mean_token_accuracy": 0.1130453810095787,
|
|
"num_tokens": 4969480.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.939700412750244,
|
|
"epoch": 0.16287569639118554,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 6.8669,
|
|
"mean_token_accuracy": 0.10693132355809212,
|
|
"num_tokens": 4979784.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.982529783248902,
|
|
"epoch": 0.16323054540293105,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 6.9363,
|
|
"mean_token_accuracy": 0.11092625781893731,
|
|
"num_tokens": 4990308.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 7.040096569061279,
|
|
"epoch": 0.16358539441467657,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 6.9497,
|
|
"mean_token_accuracy": 0.10680189654231072,
|
|
"num_tokens": 5002068.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 7.060698699951172,
|
|
"epoch": 0.16394024342642205,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 6.8228,
|
|
"mean_token_accuracy": 0.11382793858647347,
|
|
"num_tokens": 5012567.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.907808923721314,
|
|
"epoch": 0.16429509243816756,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 6.8369,
|
|
"mean_token_accuracy": 0.11393960565328598,
|
|
"num_tokens": 5022899.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.992615556716919,
|
|
"epoch": 0.16464994144991307,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 6.7848,
|
|
"mean_token_accuracy": 0.11245036497712135,
|
|
"num_tokens": 5033221.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 7.0934224128723145,
|
|
"epoch": 0.16500479046165856,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 6.9766,
|
|
"mean_token_accuracy": 0.10262498036026954,
|
|
"num_tokens": 5045265.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.954192686080932,
|
|
"epoch": 0.16535963947340407,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 6.8155,
|
|
"mean_token_accuracy": 0.10801345705986024,
|
|
"num_tokens": 5055735.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.902051687240601,
|
|
"epoch": 0.16571448848514958,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 6.8765,
|
|
"mean_token_accuracy": 0.10825056955218315,
|
|
"num_tokens": 5066258.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 7.005957317352295,
|
|
"epoch": 0.16606933749689506,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 6.9255,
|
|
"mean_token_accuracy": 0.1042802594602108,
|
|
"num_tokens": 5077202.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 7.0072572231292725,
|
|
"epoch": 0.16642418650864058,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 6.8858,
|
|
"mean_token_accuracy": 0.10783494263887405,
|
|
"num_tokens": 5088367.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 7.012457418441772,
|
|
"epoch": 0.1667790355203861,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 6.8797,
|
|
"mean_token_accuracy": 0.10552211254835128,
|
|
"num_tokens": 5098724.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.929810476303101,
|
|
"epoch": 0.16713388453213157,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 6.7804,
|
|
"mean_token_accuracy": 0.11827445849776268,
|
|
"num_tokens": 5109996.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.902822065353393,
|
|
"epoch": 0.16748873354387708,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 6.9349,
|
|
"mean_token_accuracy": 0.11100076511502266,
|
|
"num_tokens": 5121137.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 7.053985357284546,
|
|
"epoch": 0.1678435825556226,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 6.9059,
|
|
"mean_token_accuracy": 0.10487226918339729,
|
|
"num_tokens": 5132369.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.991532182693481,
|
|
"epoch": 0.16819843156736808,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 6.7483,
|
|
"mean_token_accuracy": 0.11153419762849807,
|
|
"num_tokens": 5143360.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.964412689208984,
|
|
"epoch": 0.1685532805791136,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 6.8533,
|
|
"mean_token_accuracy": 0.10653385668992996,
|
|
"num_tokens": 5153975.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 7.016902065277099,
|
|
"epoch": 0.1689081295908591,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 6.8784,
|
|
"mean_token_accuracy": 0.10774541646242142,
|
|
"num_tokens": 5165037.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.897519445419311,
|
|
"epoch": 0.16926297860260459,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 6.7927,
|
|
"mean_token_accuracy": 0.1141560547053814,
|
|
"num_tokens": 5175316.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.936920261383056,
|
|
"epoch": 0.1696178276143501,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 6.8653,
|
|
"mean_token_accuracy": 0.1071905441582203,
|
|
"num_tokens": 5187567.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 7.035586547851563,
|
|
"epoch": 0.1699726766260956,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 6.8021,
|
|
"mean_token_accuracy": 0.11246755048632621,
|
|
"num_tokens": 5198046.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.898006439208984,
|
|
"epoch": 0.1703275256378411,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 6.8817,
|
|
"mean_token_accuracy": 0.105026775598526,
|
|
"num_tokens": 5207954.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.970842695236206,
|
|
"epoch": 0.1706823746495866,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 6.8182,
|
|
"mean_token_accuracy": 0.11151416301727295,
|
|
"num_tokens": 5218524.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.9917010307312015,
|
|
"epoch": 0.17103722366133212,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 6.9518,
|
|
"mean_token_accuracy": 0.1023468978703022,
|
|
"num_tokens": 5230375.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 7.060526180267334,
|
|
"epoch": 0.1713920726730776,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 6.9012,
|
|
"mean_token_accuracy": 0.11125127226114273,
|
|
"num_tokens": 5240425.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 7.019340848922729,
|
|
"epoch": 0.1717469216848231,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 6.825,
|
|
"mean_token_accuracy": 0.10501738041639327,
|
|
"num_tokens": 5251713.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.970897150039673,
|
|
"epoch": 0.17210177069656862,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 6.8769,
|
|
"mean_token_accuracy": 0.10659899786114693,
|
|
"num_tokens": 5262639.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.917237234115601,
|
|
"epoch": 0.1724566197083141,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 6.8396,
|
|
"mean_token_accuracy": 0.116180519759655,
|
|
"num_tokens": 5273609.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 7.0096338272094725,
|
|
"epoch": 0.17281146872005962,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 6.8977,
|
|
"mean_token_accuracy": 0.10368822440505028,
|
|
"num_tokens": 5284090.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 7.017961835861206,
|
|
"epoch": 0.17316631773180513,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 6.8516,
|
|
"mean_token_accuracy": 0.10498897284269333,
|
|
"num_tokens": 5294537.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.940798044204712,
|
|
"epoch": 0.17352116674355061,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 6.8249,
|
|
"mean_token_accuracy": 0.1162478543817997,
|
|
"num_tokens": 5305524.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.935272312164306,
|
|
"epoch": 0.17387601575529613,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 6.8247,
|
|
"mean_token_accuracy": 0.10882443636655807,
|
|
"num_tokens": 5317408.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.978143835067749,
|
|
"epoch": 0.1742308647670416,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 6.871,
|
|
"mean_token_accuracy": 0.10791338384151458,
|
|
"num_tokens": 5328275.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.904288864135742,
|
|
"epoch": 0.17458571377878712,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 6.8864,
|
|
"mean_token_accuracy": 0.11189721897244453,
|
|
"num_tokens": 5338990.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 7.034686183929443,
|
|
"epoch": 0.17494056279053263,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 6.851,
|
|
"mean_token_accuracy": 0.1080154150724411,
|
|
"num_tokens": 5349826.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.9757061958312985,
|
|
"epoch": 0.17529541180227812,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 6.8879,
|
|
"mean_token_accuracy": 0.11000979393720627,
|
|
"num_tokens": 5361354.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.949403190612793,
|
|
"epoch": 0.17565026081402363,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 6.885,
|
|
"mean_token_accuracy": 0.1100533053278923,
|
|
"num_tokens": 5371079.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 7.012681579589843,
|
|
"epoch": 0.17600510982576914,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 6.8556,
|
|
"mean_token_accuracy": 0.10374706834554673,
|
|
"num_tokens": 5381351.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.982795715332031,
|
|
"epoch": 0.17635995883751462,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 6.7851,
|
|
"mean_token_accuracy": 0.11224160492420196,
|
|
"num_tokens": 5391638.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.904779100418091,
|
|
"epoch": 0.17671480784926014,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 6.8322,
|
|
"mean_token_accuracy": 0.10784663334488868,
|
|
"num_tokens": 5402882.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 7.070130634307861,
|
|
"epoch": 0.17706965686100565,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 6.818,
|
|
"mean_token_accuracy": 0.11323273256421089,
|
|
"num_tokens": 5412563.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.9515509605407715,
|
|
"epoch": 0.17742450587275113,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 6.8012,
|
|
"mean_token_accuracy": 0.10688069686293603,
|
|
"num_tokens": 5423980.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.983108234405518,
|
|
"epoch": 0.17777935488449664,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002740575249600342,
|
|
"loss": 6.8567,
|
|
"mean_token_accuracy": 0.11107314750552177,
|
|
"num_tokens": 5434179.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.975622844696045,
|
|
"epoch": 0.17813420389624215,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00027287945635254263,
|
|
"loss": 6.8159,
|
|
"mean_token_accuracy": 0.11661912426352501,
|
|
"num_tokens": 5444673.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.953983354568481,
|
|
"epoch": 0.17848905290798764,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00027170144588082635,
|
|
"loss": 6.9607,
|
|
"mean_token_accuracy": 0.10692217573523521,
|
|
"num_tokens": 5456436.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.990560388565063,
|
|
"epoch": 0.17884390191973315,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00027052352584063763,
|
|
"loss": 6.829,
|
|
"mean_token_accuracy": 0.10766687542200089,
|
|
"num_tokens": 5467722.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.96366696357727,
|
|
"epoch": 0.17919875093147866,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 6.8025,
|
|
"mean_token_accuracy": 0.10859449803829194,
|
|
"num_tokens": 5478696.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.942538738250732,
|
|
"epoch": 0.17955359994322415,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00026816808622456937,
|
|
"loss": 6.782,
|
|
"mean_token_accuracy": 0.10927296131849289,
|
|
"num_tokens": 5490355.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.861541128158569,
|
|
"epoch": 0.17990844895496966,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002669906312242569,
|
|
"loss": 6.7646,
|
|
"mean_token_accuracy": 0.1071113832294941,
|
|
"num_tokens": 5501543.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 7.008576345443726,
|
|
"epoch": 0.18026329796671517,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026581339580483525,
|
|
"loss": 6.7959,
|
|
"mean_token_accuracy": 0.1064346343278885,
|
|
"num_tokens": 5513374.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.9608423709869385,
|
|
"epoch": 0.18061814697846065,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002646364122408082,
|
|
"loss": 6.8561,
|
|
"mean_token_accuracy": 0.1085042454302311,
|
|
"num_tokens": 5523608.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.885600328445435,
|
|
"epoch": 0.18097299599020616,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002634597127997749,
|
|
"loss": 6.8051,
|
|
"mean_token_accuracy": 0.11450423449277877,
|
|
"num_tokens": 5535108.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.912053108215332,
|
|
"epoch": 0.18132784500195168,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002622833297415445,
|
|
"loss": 6.7962,
|
|
"mean_token_accuracy": 0.11635999530553817,
|
|
"num_tokens": 5545927.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 7.005493640899658,
|
|
"epoch": 0.18168269401369716,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0002611072953172531,
|
|
"loss": 6.8012,
|
|
"mean_token_accuracy": 0.1119636908173561,
|
|
"num_tokens": 5556337.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.92455563545227,
|
|
"epoch": 0.18203754302544267,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00025993164176847845,
|
|
"loss": 6.7837,
|
|
"mean_token_accuracy": 0.1091103196144104,
|
|
"num_tokens": 5567024.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.922195863723755,
|
|
"epoch": 0.18239239203718818,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002587564013263564,
|
|
"loss": 6.8145,
|
|
"mean_token_accuracy": 0.10922716110944748,
|
|
"num_tokens": 5578168.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.919791984558105,
|
|
"epoch": 0.18274724104893367,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0002575816062106974,
|
|
"loss": 6.8679,
|
|
"mean_token_accuracy": 0.11193211972713471,
|
|
"num_tokens": 5589000.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.992647314071656,
|
|
"epoch": 0.18310209006067918,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00025640728862910293,
|
|
"loss": 6.8424,
|
|
"mean_token_accuracy": 0.107405274361372,
|
|
"num_tokens": 5599019.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.995386791229248,
|
|
"epoch": 0.1834569390724247,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 6.8686,
|
|
"mean_token_accuracy": 0.10983893051743507,
|
|
"num_tokens": 5610169.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.944970226287841,
|
|
"epoch": 0.18381178808417017,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00025406021483217225,
|
|
"loss": 6.8337,
|
|
"mean_token_accuracy": 0.11445480808615685,
|
|
"num_tokens": 5621471.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.932586908340454,
|
|
"epoch": 0.1841666370959157,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00025288752296304963,
|
|
"loss": 6.8451,
|
|
"mean_token_accuracy": 0.10618517324328422,
|
|
"num_tokens": 5632708.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.991014862060547,
|
|
"epoch": 0.1845214861076612,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000251715437318655,
|
|
"loss": 6.8711,
|
|
"mean_token_accuracy": 0.11433243229985238,
|
|
"num_tokens": 5643057.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 7.0343590259552,
|
|
"epoch": 0.18487633511940668,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0002505439900323084,
|
|
"loss": 6.8464,
|
|
"mean_token_accuracy": 0.11091703474521637,
|
|
"num_tokens": 5654928.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.969445991516113,
|
|
"epoch": 0.1852311841311522,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00024937321321982894,
|
|
"loss": 6.7362,
|
|
"mean_token_accuracy": 0.11556778773665428,
|
|
"num_tokens": 5665135.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.9363429069519045,
|
|
"epoch": 0.1855860331428977,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00024820313897865433,
|
|
"loss": 6.7503,
|
|
"mean_token_accuracy": 0.11977223679423332,
|
|
"num_tokens": 5676044.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.928713369369507,
|
|
"epoch": 0.1859408821546432,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00024703379938696105,
|
|
"loss": 6.8882,
|
|
"mean_token_accuracy": 0.10680016577243805,
|
|
"num_tokens": 5688329.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.944188451766967,
|
|
"epoch": 0.1862957311663887,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00024586522650278447,
|
|
"loss": 6.7589,
|
|
"mean_token_accuracy": 0.11397264674305915,
|
|
"num_tokens": 5698762.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.975194931030273,
|
|
"epoch": 0.1866505801781342,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00024469745236314064,
|
|
"loss": 6.8735,
|
|
"mean_token_accuracy": 0.10839726105332374,
|
|
"num_tokens": 5709087.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.99052848815918,
|
|
"epoch": 0.1870054291898797,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00024353050898314767,
|
|
"loss": 6.8089,
|
|
"mean_token_accuracy": 0.1089420735836029,
|
|
"num_tokens": 5719633.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.9208982467651365,
|
|
"epoch": 0.1873602782016252,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00024236442835514743,
|
|
"loss": 6.8141,
|
|
"mean_token_accuracy": 0.11238551586866379,
|
|
"num_tokens": 5730234.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.96273512840271,
|
|
"epoch": 0.18771512721337072,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 6.8139,
|
|
"mean_token_accuracy": 0.11327897533774375,
|
|
"num_tokens": 5741279.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.967271327972412,
|
|
"epoch": 0.1880699762251162,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00024003498320535462,
|
|
"loss": 6.8053,
|
|
"mean_token_accuracy": 0.11368138119578361,
|
|
"num_tokens": 5751699.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.979606866836548,
|
|
"epoch": 0.18842482523686171,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00023887168254647727,
|
|
"loss": 6.8163,
|
|
"mean_token_accuracy": 0.11396170631051064,
|
|
"num_tokens": 5761527.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.991921281814575,
|
|
"epoch": 0.18877967424860723,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00023770937236367308,
|
|
"loss": 6.8199,
|
|
"mean_token_accuracy": 0.11399022936820984,
|
|
"num_tokens": 5771348.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.988916683197021,
|
|
"epoch": 0.1891345232603527,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00023654808452226278,
|
|
"loss": 6.8105,
|
|
"mean_token_accuracy": 0.11595386117696763,
|
|
"num_tokens": 5780350.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.9241063594818115,
|
|
"epoch": 0.18948937227209822,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00023538785085953912,
|
|
"loss": 6.8008,
|
|
"mean_token_accuracy": 0.11144680380821229,
|
|
"num_tokens": 5791743.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.939181470870972,
|
|
"epoch": 0.18984422128384373,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00023422870318389404,
|
|
"loss": 6.738,
|
|
"mean_token_accuracy": 0.11290448680520057,
|
|
"num_tokens": 5802958.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.964320516586303,
|
|
"epoch": 0.19019907029558922,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002330706732739468,
|
|
"loss": 6.8065,
|
|
"mean_token_accuracy": 0.106526218354702,
|
|
"num_tokens": 5814503.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.862625169754028,
|
|
"epoch": 0.19055391930733473,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00023191379287767211,
|
|
"loss": 6.6636,
|
|
"mean_token_accuracy": 0.12738021984696388,
|
|
"num_tokens": 5826182.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.933192300796509,
|
|
"epoch": 0.19090876831908024,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002307580937115305,
|
|
"loss": 6.7616,
|
|
"mean_token_accuracy": 0.11562723517417908,
|
|
"num_tokens": 5837426.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.910529088973999,
|
|
"epoch": 0.19126361733082572,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00022960360745959846,
|
|
"loss": 6.8184,
|
|
"mean_token_accuracy": 0.10251531153917312,
|
|
"num_tokens": 5848344.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.921610736846924,
|
|
"epoch": 0.19161846634257124,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00022845036577269972,
|
|
"loss": 6.7971,
|
|
"mean_token_accuracy": 0.11794155836105347,
|
|
"num_tokens": 5859905.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.920235109329224,
|
|
"epoch": 0.19197331535431675,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 6.794,
|
|
"mean_token_accuracy": 0.1141629844903946,
|
|
"num_tokens": 5871121.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.976440477371216,
|
|
"epoch": 0.19232816436606223,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002261477425258287,
|
|
"loss": 6.7817,
|
|
"mean_token_accuracy": 0.11112578511238098,
|
|
"num_tokens": 5882211.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.9788895606994625,
|
|
"epoch": 0.19268301337780774,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002249984240934358,
|
|
"loss": 6.7742,
|
|
"mean_token_accuracy": 0.11779958754777908,
|
|
"num_tokens": 5892702.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.8424193382263185,
|
|
"epoch": 0.19303786238955326,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00022385047647950464,
|
|
"loss": 6.7369,
|
|
"mean_token_accuracy": 0.11136523112654687,
|
|
"num_tokens": 5903655.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.986451768875122,
|
|
"epoch": 0.19339271140129874,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002227039311555986,
|
|
"loss": 6.8988,
|
|
"mean_token_accuracy": 0.11267425641417503,
|
|
"num_tokens": 5915387.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 7.007633018493652,
|
|
"epoch": 0.19374756041304425,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002215588195548372,
|
|
"loss": 6.7527,
|
|
"mean_token_accuracy": 0.11794544607400895,
|
|
"num_tokens": 5926827.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.983679103851318,
|
|
"epoch": 0.19410240942478976,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022041517307103337,
|
|
"loss": 6.834,
|
|
"mean_token_accuracy": 0.10779970809817314,
|
|
"num_tokens": 5937459.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.914843559265137,
|
|
"epoch": 0.19445725843653525,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002192730230578331,
|
|
"loss": 6.7607,
|
|
"mean_token_accuracy": 0.11893753930926323,
|
|
"num_tokens": 5949116.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.978890895843506,
|
|
"epoch": 0.19481210744828076,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002181324008278559,
|
|
"loss": 6.8364,
|
|
"mean_token_accuracy": 0.11154205128550529,
|
|
"num_tokens": 5959447.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.891154098510742,
|
|
"epoch": 0.19516695646002627,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00021699333765183655,
|
|
"loss": 6.7122,
|
|
"mean_token_accuracy": 0.1195847325026989,
|
|
"num_tokens": 5969754.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.881964254379272,
|
|
"epoch": 0.19552180547177175,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0002158558647577673,
|
|
"loss": 6.8052,
|
|
"mean_token_accuracy": 0.1127731017768383,
|
|
"num_tokens": 5981395.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 7.011906147003174,
|
|
"epoch": 0.19587665448351727,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00021472001333004215,
|
|
"loss": 6.8374,
|
|
"mean_token_accuracy": 0.11157171055674553,
|
|
"num_tokens": 5992177.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.933767938613892,
|
|
"epoch": 0.19623150349526278,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 6.7019,
|
|
"mean_token_accuracy": 0.11525828316807747,
|
|
"num_tokens": 6001683.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.925135231018066,
|
|
"epoch": 0.19658635250700826,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002124532993880799,
|
|
"loss": 6.8463,
|
|
"mean_token_accuracy": 0.11589981392025947,
|
|
"num_tokens": 6013278.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.982972240447998,
|
|
"epoch": 0.19694120151875377,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00021132249901695044,
|
|
"loss": 6.7965,
|
|
"mean_token_accuracy": 0.10612709373235703,
|
|
"num_tokens": 6024083.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.957027196884155,
|
|
"epoch": 0.19729605053049928,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00021019344439667705,
|
|
"loss": 6.7964,
|
|
"mean_token_accuracy": 0.11672135517001152,
|
|
"num_tokens": 6034445.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.9106330394744875,
|
|
"epoch": 0.19765089954224477,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00020906616648086213,
|
|
"loss": 6.8121,
|
|
"mean_token_accuracy": 0.11197659894824027,
|
|
"num_tokens": 6044698.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.947172451019287,
|
|
"epoch": 0.19800574855399028,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00020794069617439942,
|
|
"loss": 6.8269,
|
|
"mean_token_accuracy": 0.11555901169776917,
|
|
"num_tokens": 6054921.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.985541725158692,
|
|
"epoch": 0.1983605975657358,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00020681706433262593,
|
|
"loss": 6.6881,
|
|
"mean_token_accuracy": 0.11160681322216988,
|
|
"num_tokens": 6065086.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.9685831546783445,
|
|
"epoch": 0.19871544657748128,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00020569530176047602,
|
|
"loss": 6.8081,
|
|
"mean_token_accuracy": 0.11681672781705857,
|
|
"num_tokens": 6076179.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.895740032196045,
|
|
"epoch": 0.1990702955892268,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002045754392116374,
|
|
"loss": 6.76,
|
|
"mean_token_accuracy": 0.1167929820716381,
|
|
"num_tokens": 6086643.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.869204473495484,
|
|
"epoch": 0.1994251446009723,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00020345750738770757,
|
|
"loss": 6.7624,
|
|
"mean_token_accuracy": 0.10626718774437904,
|
|
"num_tokens": 6097482.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.947177410125732,
|
|
"epoch": 0.19977999361271778,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00020234153693735214,
|
|
"loss": 6.74,
|
|
"mean_token_accuracy": 0.11464135870337486,
|
|
"num_tokens": 6107718.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.9241992950439455,
|
|
"epoch": 0.2001348426244633,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002012275584554647,
|
|
"loss": 6.7492,
|
|
"mean_token_accuracy": 0.10922236144542694,
|
|
"num_tokens": 6117304.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.93255729675293,
|
|
"epoch": 0.2004896916362088,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 6.6845,
|
|
"mean_token_accuracy": 0.11656132563948632,
|
|
"num_tokens": 6127383.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.9440559387207035,
|
|
"epoch": 0.2008445406479543,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00019900569950277692,
|
|
"loss": 6.7635,
|
|
"mean_token_accuracy": 0.11715753227472306,
|
|
"num_tokens": 6138163.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.8963623046875,
|
|
"epoch": 0.2011993896596998,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00019789787994536228,
|
|
"loss": 6.7243,
|
|
"mean_token_accuracy": 0.11367726176977158,
|
|
"num_tokens": 6149519.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.944343852996826,
|
|
"epoch": 0.2015542386714453,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00019679217418151667,
|
|
"loss": 6.7561,
|
|
"mean_token_accuracy": 0.11886984705924988,
|
|
"num_tokens": 6160234.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.953767442703247,
|
|
"epoch": 0.2019090876831908,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00019568861252472236,
|
|
"loss": 6.7664,
|
|
"mean_token_accuracy": 0.11626882553100586,
|
|
"num_tokens": 6170589.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.919698905944824,
|
|
"epoch": 0.2022639366949363,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00019458722522967952,
|
|
"loss": 6.7532,
|
|
"mean_token_accuracy": 0.11209649443626404,
|
|
"num_tokens": 6182291.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.921873092651367,
|
|
"epoch": 0.20261878570668182,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00019348804249147723,
|
|
"loss": 6.7244,
|
|
"mean_token_accuracy": 0.11564923822879791,
|
|
"num_tokens": 6192461.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.8797821521759035,
|
|
"epoch": 0.2029736347184273,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001923910944447655,
|
|
"loss": 6.6885,
|
|
"mean_token_accuracy": 0.11495175957679749,
|
|
"num_tokens": 6203038.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.976866340637207,
|
|
"epoch": 0.20332848373017282,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00019129641116292928,
|
|
"loss": 6.7681,
|
|
"mean_token_accuracy": 0.10720009878277778,
|
|
"num_tokens": 6212772.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.932071924209595,
|
|
"epoch": 0.20368333274191833,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00019020402265726343,
|
|
"loss": 6.7421,
|
|
"mean_token_accuracy": 0.11611971035599708,
|
|
"num_tokens": 6224751.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.8816296577453615,
|
|
"epoch": 0.2040381817536638,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001891139588761509,
|
|
"loss": 6.7564,
|
|
"mean_token_accuracy": 0.11213191598653793,
|
|
"num_tokens": 6236112.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.855004405975341,
|
|
"epoch": 0.20439303076540932,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00018802624970424076,
|
|
"loss": 6.6789,
|
|
"mean_token_accuracy": 0.11986380815505981,
|
|
"num_tokens": 6247472.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.962516593933105,
|
|
"epoch": 0.20474787977715483,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 6.7785,
|
|
"mean_token_accuracy": 0.11515040174126626,
|
|
"num_tokens": 6257330.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.893075942993164,
|
|
"epoch": 0.20510272878890032,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00018585801440304306,
|
|
"loss": 6.7531,
|
|
"mean_token_accuracy": 0.11047643274068833,
|
|
"num_tokens": 6267879.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.980209064483643,
|
|
"epoch": 0.20545757780064583,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00018477754771702165,
|
|
"loss": 6.786,
|
|
"mean_token_accuracy": 0.1101107120513916,
|
|
"num_tokens": 6279252.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.962943887710571,
|
|
"epoch": 0.2058124268123913,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00018369955452510506,
|
|
"loss": 6.7178,
|
|
"mean_token_accuracy": 0.11665593087673187,
|
|
"num_tokens": 6288915.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.84703803062439,
|
|
"epoch": 0.20616727582413683,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001826240643810212,
|
|
"loss": 6.7232,
|
|
"mean_token_accuracy": 0.11100221052765846,
|
|
"num_tokens": 6300065.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.887630128860474,
|
|
"epoch": 0.20652212483588234,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001815511067698758,
|
|
"loss": 6.7618,
|
|
"mean_token_accuracy": 0.1111633986234665,
|
|
"num_tokens": 6311185.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.956713914871216,
|
|
"epoch": 0.20687697384762782,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0001804807111073436,
|
|
"loss": 6.8147,
|
|
"mean_token_accuracy": 0.10564543157815934,
|
|
"num_tokens": 6324160.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.998915100097657,
|
|
"epoch": 0.20723182285937333,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001794129067388625,
|
|
"loss": 6.7176,
|
|
"mean_token_accuracy": 0.1167424201965332,
|
|
"num_tokens": 6334212.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.9070930004119875,
|
|
"epoch": 0.20758667187111884,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00017834772293882868,
|
|
"loss": 6.7137,
|
|
"mean_token_accuracy": 0.1089239127933979,
|
|
"num_tokens": 6345898.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.909444189071655,
|
|
"epoch": 0.20794152088286433,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000177285188909794,
|
|
"loss": 6.7742,
|
|
"mean_token_accuracy": 0.11567335724830627,
|
|
"num_tokens": 6356287.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.9139430046081545,
|
|
"epoch": 0.20829636989460984,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001762253337816656,
|
|
"loss": 6.7081,
|
|
"mean_token_accuracy": 0.11572626531124115,
|
|
"num_tokens": 6367415.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.938155174255371,
|
|
"epoch": 0.20865121890635535,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017516818661090738,
|
|
"loss": 6.78,
|
|
"mean_token_accuracy": 0.11153295859694481,
|
|
"num_tokens": 6378463.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.928212881088257,
|
|
"epoch": 0.20900606791810084,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 6.7762,
|
|
"mean_token_accuracy": 0.1111447274684906,
|
|
"num_tokens": 6389075.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.880980730056763,
|
|
"epoch": 0.20936091692984635,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00017306213199536115,
|
|
"loss": 6.6766,
|
|
"mean_token_accuracy": 0.1157895028591156,
|
|
"num_tokens": 6398872.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.899291181564331,
|
|
"epoch": 0.20971576594159186,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001720132822891243,
|
|
"loss": 6.6369,
|
|
"mean_token_accuracy": 0.11267390474677086,
|
|
"num_tokens": 6408557.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.963399028778076,
|
|
"epoch": 0.21007061495333734,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001709672560157769,
|
|
"loss": 6.6999,
|
|
"mean_token_accuracy": 0.1197504609823227,
|
|
"num_tokens": 6420191.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.916766119003296,
|
|
"epoch": 0.21042546396508285,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00016992408185265758,
|
|
"loss": 6.8834,
|
|
"mean_token_accuracy": 0.10643507018685341,
|
|
"num_tokens": 6432277.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.881908321380616,
|
|
"epoch": 0.21078031297682837,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00016888378839891298,
|
|
"loss": 6.6181,
|
|
"mean_token_accuracy": 0.11972883269190789,
|
|
"num_tokens": 6442321.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.883301639556885,
|
|
"epoch": 0.21113516198857385,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001678464041747137,
|
|
"loss": 6.6313,
|
|
"mean_token_accuracy": 0.11854083985090255,
|
|
"num_tokens": 6452545.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.903544282913208,
|
|
"epoch": 0.21149001100031936,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00016681195762047223,
|
|
"loss": 6.7528,
|
|
"mean_token_accuracy": 0.1161648727953434,
|
|
"num_tokens": 6463188.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.933883905410767,
|
|
"epoch": 0.21184486001206487,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00016578047709606337,
|
|
"loss": 6.7657,
|
|
"mean_token_accuracy": 0.11609257981181145,
|
|
"num_tokens": 6473847.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.880580806732178,
|
|
"epoch": 0.21219970902381036,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00016475199088004678,
|
|
"loss": 6.765,
|
|
"mean_token_accuracy": 0.11903324723243713,
|
|
"num_tokens": 6485178.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.987133598327636,
|
|
"epoch": 0.21255455803555587,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016372652716889163,
|
|
"loss": 6.7447,
|
|
"mean_token_accuracy": 0.11706713512539864,
|
|
"num_tokens": 6495102.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.990228223800659,
|
|
"epoch": 0.21290940704730138,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001627041140762035,
|
|
"loss": 6.7779,
|
|
"mean_token_accuracy": 0.11222531944513321,
|
|
"num_tokens": 6505829.0,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"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": 9483030890496000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|