12167 lines
332 KiB
JSON
12167 lines
332 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 7.017543859649122,
|
|
"eval_steps": 500,
|
|
"global_step": 6000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691974544525147,
|
|
"epoch": 0.005847953216374269,
|
|
"grad_norm": 13.5,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8755,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 13999.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691928291320801,
|
|
"epoch": 0.011695906432748537,
|
|
"grad_norm": 13.1875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7881,
|
|
"mean_token_accuracy": 8.032128680497408e-05,
|
|
"num_tokens": 25888.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690763854980469,
|
|
"epoch": 0.017543859649122806,
|
|
"grad_norm": 10.5625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5299,
|
|
"mean_token_accuracy": 0.029781045974232256,
|
|
"num_tokens": 37699.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.669985389709472,
|
|
"epoch": 0.023391812865497075,
|
|
"grad_norm": 5.9375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2235,
|
|
"mean_token_accuracy": 0.05048878975212574,
|
|
"num_tokens": 49648.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.5446457862854,
|
|
"epoch": 0.029239766081871343,
|
|
"grad_norm": 4.0,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9371,
|
|
"mean_token_accuracy": 0.049999012425541875,
|
|
"num_tokens": 62255.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.506792068481445,
|
|
"epoch": 0.03508771929824561,
|
|
"grad_norm": 3.4375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.8635,
|
|
"mean_token_accuracy": 0.048928847163915636,
|
|
"num_tokens": 76003.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.482833862304688,
|
|
"epoch": 0.04093567251461988,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.7065,
|
|
"mean_token_accuracy": 0.06288735270500183,
|
|
"num_tokens": 87585.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.464082145690918,
|
|
"epoch": 0.04678362573099415,
|
|
"grad_norm": 2.984375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.6752,
|
|
"mean_token_accuracy": 0.061649443581700325,
|
|
"num_tokens": 100404.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.373864269256591,
|
|
"epoch": 0.05263157894736842,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6279,
|
|
"mean_token_accuracy": 0.0631986454129219,
|
|
"num_tokens": 113130.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.265812587738036,
|
|
"epoch": 0.05847953216374269,
|
|
"grad_norm": 3.265625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.4722,
|
|
"mean_token_accuracy": 0.0745375208556652,
|
|
"num_tokens": 125430.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.30671148300171,
|
|
"epoch": 0.06432748538011696,
|
|
"grad_norm": 2.703125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5423,
|
|
"mean_token_accuracy": 0.06626624949276447,
|
|
"num_tokens": 137550.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.23580093383789,
|
|
"epoch": 0.07017543859649122,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4244,
|
|
"mean_token_accuracy": 0.08311915919184684,
|
|
"num_tokens": 149530.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.123470211029053,
|
|
"epoch": 0.07602339181286549,
|
|
"grad_norm": 2.828125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4496,
|
|
"mean_token_accuracy": 0.07575741671025753,
|
|
"num_tokens": 161671.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.18219289779663,
|
|
"epoch": 0.08187134502923976,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.3854,
|
|
"mean_token_accuracy": 0.07742468379437924,
|
|
"num_tokens": 174160.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.220508003234864,
|
|
"epoch": 0.08771929824561403,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.3404,
|
|
"mean_token_accuracy": 0.07739944010972977,
|
|
"num_tokens": 186969.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.152502727508544,
|
|
"epoch": 0.0935672514619883,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.2504,
|
|
"mean_token_accuracy": 0.08422459289431572,
|
|
"num_tokens": 199470.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.011490631103516,
|
|
"epoch": 0.09941520467836257,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1723,
|
|
"mean_token_accuracy": 0.09044785127043724,
|
|
"num_tokens": 211306.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.162246799468994,
|
|
"epoch": 0.10526315789473684,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.1514,
|
|
"mean_token_accuracy": 0.08413446620106697,
|
|
"num_tokens": 223323.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 9.993253993988038,
|
|
"epoch": 0.1111111111111111,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.0369,
|
|
"mean_token_accuracy": 0.09222856536507607,
|
|
"num_tokens": 235896.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.006341552734375,
|
|
"epoch": 0.11695906432748537,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.0269,
|
|
"mean_token_accuracy": 0.08629260584712029,
|
|
"num_tokens": 247879.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.960891246795654,
|
|
"epoch": 0.12280701754385964,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.8842,
|
|
"mean_token_accuracy": 0.08838528990745545,
|
|
"num_tokens": 259714.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.847960281372071,
|
|
"epoch": 0.1286549707602339,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.8575,
|
|
"mean_token_accuracy": 0.09311703070998192,
|
|
"num_tokens": 272429.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.824352073669434,
|
|
"epoch": 0.13450292397660818,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.834,
|
|
"mean_token_accuracy": 0.08550347462296486,
|
|
"num_tokens": 285924.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.759407234191894,
|
|
"epoch": 0.14035087719298245,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.6848,
|
|
"mean_token_accuracy": 0.09789064452052117,
|
|
"num_tokens": 299453.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.61653537750244,
|
|
"epoch": 0.14619883040935672,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.6997,
|
|
"mean_token_accuracy": 0.09431982859969139,
|
|
"num_tokens": 312295.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.595601081848145,
|
|
"epoch": 0.15204678362573099,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.5501,
|
|
"mean_token_accuracy": 0.09463789388537407,
|
|
"num_tokens": 324730.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.421235179901123,
|
|
"epoch": 0.15789473684210525,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.4792,
|
|
"mean_token_accuracy": 0.09828749150037766,
|
|
"num_tokens": 336481.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.258519268035888,
|
|
"epoch": 0.16374269005847952,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.4926,
|
|
"mean_token_accuracy": 0.0937332771718502,
|
|
"num_tokens": 348987.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.255396842956543,
|
|
"epoch": 0.1695906432748538,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.4024,
|
|
"mean_token_accuracy": 0.09496198520064354,
|
|
"num_tokens": 360390.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.080035209655762,
|
|
"epoch": 0.17543859649122806,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.3663,
|
|
"mean_token_accuracy": 0.0986220009624958,
|
|
"num_tokens": 372456.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.131995296478271,
|
|
"epoch": 0.18128654970760233,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.3326,
|
|
"mean_token_accuracy": 0.09041213691234588,
|
|
"num_tokens": 384185.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.906043529510498,
|
|
"epoch": 0.1871345029239766,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.2872,
|
|
"mean_token_accuracy": 0.0966079480946064,
|
|
"num_tokens": 395399.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.800198173522949,
|
|
"epoch": 0.19298245614035087,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.2863,
|
|
"mean_token_accuracy": 0.09909554198384285,
|
|
"num_tokens": 408676.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.711225223541259,
|
|
"epoch": 0.19883040935672514,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.2698,
|
|
"mean_token_accuracy": 0.09550312235951423,
|
|
"num_tokens": 421861.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.798204135894775,
|
|
"epoch": 0.2046783625730994,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.3183,
|
|
"mean_token_accuracy": 0.09797023683786392,
|
|
"num_tokens": 434852.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.719051837921143,
|
|
"epoch": 0.21052631578947367,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.2497,
|
|
"mean_token_accuracy": 0.09236074090003968,
|
|
"num_tokens": 447313.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.575885391235351,
|
|
"epoch": 0.21637426900584794,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.1896,
|
|
"mean_token_accuracy": 0.10111410841345787,
|
|
"num_tokens": 460271.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.562645149230956,
|
|
"epoch": 0.2222222222222222,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.1582,
|
|
"mean_token_accuracy": 0.1076015666127205,
|
|
"num_tokens": 473006.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.600567626953126,
|
|
"epoch": 0.22807017543859648,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.1814,
|
|
"mean_token_accuracy": 0.09954929873347282,
|
|
"num_tokens": 486100.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.427424430847168,
|
|
"epoch": 0.23391812865497075,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.0686,
|
|
"mean_token_accuracy": 0.108884796500206,
|
|
"num_tokens": 499460.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.508369827270508,
|
|
"epoch": 0.23976608187134502,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.0849,
|
|
"mean_token_accuracy": 0.10188826620578766,
|
|
"num_tokens": 511795.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.331426906585694,
|
|
"epoch": 0.24561403508771928,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.1497,
|
|
"mean_token_accuracy": 0.10387411639094353,
|
|
"num_tokens": 524897.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.393550872802734,
|
|
"epoch": 0.25146198830409355,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.0901,
|
|
"mean_token_accuracy": 0.10539199337363243,
|
|
"num_tokens": 538325.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.538016510009765,
|
|
"epoch": 0.2573099415204678,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.1375,
|
|
"mean_token_accuracy": 0.09665202274918556,
|
|
"num_tokens": 550979.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.389378452301026,
|
|
"epoch": 0.2631578947368421,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.1055,
|
|
"mean_token_accuracy": 0.10522033721208572,
|
|
"num_tokens": 563001.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.339937973022462,
|
|
"epoch": 0.26900584795321636,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.0235,
|
|
"mean_token_accuracy": 0.1068884126842022,
|
|
"num_tokens": 573600.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.336192607879639,
|
|
"epoch": 0.27485380116959063,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.0318,
|
|
"mean_token_accuracy": 0.10861181318759919,
|
|
"num_tokens": 585287.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.343888473510741,
|
|
"epoch": 0.2807017543859649,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.0583,
|
|
"mean_token_accuracy": 0.10137449800968171,
|
|
"num_tokens": 597386.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.288441467285157,
|
|
"epoch": 0.28654970760233917,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.0481,
|
|
"mean_token_accuracy": 0.10464474633336067,
|
|
"num_tokens": 609256.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.411810398101807,
|
|
"epoch": 0.29239766081871343,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.0474,
|
|
"mean_token_accuracy": 0.10834256708621978,
|
|
"num_tokens": 621506.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.100932836532593,
|
|
"epoch": 0.2982456140350877,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.9646,
|
|
"mean_token_accuracy": 0.1123759962618351,
|
|
"num_tokens": 633512.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.300220108032226,
|
|
"epoch": 0.30409356725146197,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.0895,
|
|
"mean_token_accuracy": 0.10339951142668724,
|
|
"num_tokens": 647098.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.343167591094971,
|
|
"epoch": 0.30994152046783624,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.0972,
|
|
"mean_token_accuracy": 0.09793601483106613,
|
|
"num_tokens": 659559.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.258354759216308,
|
|
"epoch": 0.3157894736842105,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.0199,
|
|
"mean_token_accuracy": 0.10353473275899887,
|
|
"num_tokens": 672621.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.298446559906006,
|
|
"epoch": 0.3216374269005848,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.9906,
|
|
"mean_token_accuracy": 0.10663355737924576,
|
|
"num_tokens": 685652.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.19245777130127,
|
|
"epoch": 0.32748538011695905,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.04,
|
|
"mean_token_accuracy": 0.11108417361974716,
|
|
"num_tokens": 696910.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.219241619110107,
|
|
"epoch": 0.3333333333333333,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.9193,
|
|
"mean_token_accuracy": 0.11071920469403267,
|
|
"num_tokens": 709054.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.26666088104248,
|
|
"epoch": 0.3391812865497076,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.9479,
|
|
"mean_token_accuracy": 0.11016275808215141,
|
|
"num_tokens": 720902.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.227052402496337,
|
|
"epoch": 0.34502923976608185,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.0106,
|
|
"mean_token_accuracy": 0.10988251566886902,
|
|
"num_tokens": 733319.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.17465419769287,
|
|
"epoch": 0.3508771929824561,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.8917,
|
|
"mean_token_accuracy": 0.11573786139488221,
|
|
"num_tokens": 744817.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.232967758178711,
|
|
"epoch": 0.3567251461988304,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.9535,
|
|
"mean_token_accuracy": 0.10650328993797302,
|
|
"num_tokens": 756582.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.144847440719605,
|
|
"epoch": 0.36257309941520466,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.8627,
|
|
"mean_token_accuracy": 0.12130056619644165,
|
|
"num_tokens": 769273.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.138186359405518,
|
|
"epoch": 0.3684210526315789,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.9351,
|
|
"mean_token_accuracy": 0.11499258056282997,
|
|
"num_tokens": 782001.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.08138952255249,
|
|
"epoch": 0.3742690058479532,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.926,
|
|
"mean_token_accuracy": 0.11798002570867538,
|
|
"num_tokens": 794575.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.198786306381226,
|
|
"epoch": 0.38011695906432746,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.9014,
|
|
"mean_token_accuracy": 0.11587951257824898,
|
|
"num_tokens": 805583.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.112969303131104,
|
|
"epoch": 0.38596491228070173,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.9467,
|
|
"mean_token_accuracy": 0.1160537801682949,
|
|
"num_tokens": 817617.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.17340841293335,
|
|
"epoch": 0.391812865497076,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.8426,
|
|
"mean_token_accuracy": 0.11603786572813987,
|
|
"num_tokens": 830043.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.051109790802002,
|
|
"epoch": 0.39766081871345027,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.8356,
|
|
"mean_token_accuracy": 0.12449322417378425,
|
|
"num_tokens": 843200.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.152585649490357,
|
|
"epoch": 0.40350877192982454,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.8698,
|
|
"mean_token_accuracy": 0.12035520151257514,
|
|
"num_tokens": 856031.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.055157375335693,
|
|
"epoch": 0.4093567251461988,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.8494,
|
|
"mean_token_accuracy": 0.1174808993935585,
|
|
"num_tokens": 869119.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.974756145477295,
|
|
"epoch": 0.4152046783625731,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.8566,
|
|
"mean_token_accuracy": 0.12319247648119927,
|
|
"num_tokens": 880802.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.131376028060913,
|
|
"epoch": 0.42105263157894735,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.8644,
|
|
"mean_token_accuracy": 0.11949887052178383,
|
|
"num_tokens": 894237.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.112016677856445,
|
|
"epoch": 0.4269005847953216,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.8648,
|
|
"mean_token_accuracy": 0.12209260165691375,
|
|
"num_tokens": 906695.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.052251434326172,
|
|
"epoch": 0.4327485380116959,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.8327,
|
|
"mean_token_accuracy": 0.1234503522515297,
|
|
"num_tokens": 918955.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.96677598953247,
|
|
"epoch": 0.43859649122807015,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.8616,
|
|
"mean_token_accuracy": 0.12289973944425583,
|
|
"num_tokens": 931812.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.08064661026001,
|
|
"epoch": 0.4444444444444444,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.8693,
|
|
"mean_token_accuracy": 0.11234893426299095,
|
|
"num_tokens": 945261.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.010916709899902,
|
|
"epoch": 0.4502923976608187,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.781,
|
|
"mean_token_accuracy": 0.12615325823426246,
|
|
"num_tokens": 958376.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.934221315383911,
|
|
"epoch": 0.45614035087719296,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.7977,
|
|
"mean_token_accuracy": 0.12733107730746268,
|
|
"num_tokens": 971709.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.095316505432129,
|
|
"epoch": 0.4619883040935672,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.772,
|
|
"mean_token_accuracy": 0.12458922639489174,
|
|
"num_tokens": 983847.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.061276531219482,
|
|
"epoch": 0.4678362573099415,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.8479,
|
|
"mean_token_accuracy": 0.11282772347331047,
|
|
"num_tokens": 997034.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.900276470184326,
|
|
"epoch": 0.47368421052631576,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.8811,
|
|
"mean_token_accuracy": 0.11630931496620178,
|
|
"num_tokens": 1010213.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.164872455596925,
|
|
"epoch": 0.47953216374269003,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.7587,
|
|
"mean_token_accuracy": 0.12265534400939941,
|
|
"num_tokens": 1021618.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.90818042755127,
|
|
"epoch": 0.4853801169590643,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.7779,
|
|
"mean_token_accuracy": 0.119289730489254,
|
|
"num_tokens": 1033809.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.035586547851562,
|
|
"epoch": 0.49122807017543857,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.7602,
|
|
"mean_token_accuracy": 0.122083730250597,
|
|
"num_tokens": 1045291.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.94893217086792,
|
|
"epoch": 0.49707602339181284,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.7345,
|
|
"mean_token_accuracy": 0.12310368940234184,
|
|
"num_tokens": 1057516.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.834074544906616,
|
|
"epoch": 0.5029239766081871,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.7166,
|
|
"mean_token_accuracy": 0.13022998943924904,
|
|
"num_tokens": 1070675.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.962926816940308,
|
|
"epoch": 0.5087719298245614,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.7871,
|
|
"mean_token_accuracy": 0.12778857946395875,
|
|
"num_tokens": 1082965.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.948134422302246,
|
|
"epoch": 0.5146198830409356,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.767,
|
|
"mean_token_accuracy": 0.11752260848879814,
|
|
"num_tokens": 1095922.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.899146366119385,
|
|
"epoch": 0.52046783625731,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.7519,
|
|
"mean_token_accuracy": 0.1246542438864708,
|
|
"num_tokens": 1109002.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.014074802398682,
|
|
"epoch": 0.5263157894736842,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.7662,
|
|
"mean_token_accuracy": 0.11872777864336967,
|
|
"num_tokens": 1121199.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.865769338607788,
|
|
"epoch": 0.5321637426900585,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.7036,
|
|
"mean_token_accuracy": 0.12603166699409485,
|
|
"num_tokens": 1134721.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.8770805358886715,
|
|
"epoch": 0.5380116959064327,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.788,
|
|
"mean_token_accuracy": 0.12093541622161866,
|
|
"num_tokens": 1147954.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.915914726257324,
|
|
"epoch": 0.543859649122807,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.7121,
|
|
"mean_token_accuracy": 0.12637170255184174,
|
|
"num_tokens": 1161208.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.918968534469604,
|
|
"epoch": 0.5497076023391813,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.8051,
|
|
"mean_token_accuracy": 0.11931493803858757,
|
|
"num_tokens": 1173789.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.964196968078613,
|
|
"epoch": 0.5555555555555556,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.7054,
|
|
"mean_token_accuracy": 0.12410269007086754,
|
|
"num_tokens": 1185765.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.763233613967896,
|
|
"epoch": 0.5614035087719298,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.7242,
|
|
"mean_token_accuracy": 0.12448163107037544,
|
|
"num_tokens": 1197954.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.995425701141357,
|
|
"epoch": 0.5672514619883041,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.6458,
|
|
"mean_token_accuracy": 0.12746693715453147,
|
|
"num_tokens": 1211285.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.84964427947998,
|
|
"epoch": 0.5730994152046783,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.7247,
|
|
"mean_token_accuracy": 0.12084417417645454,
|
|
"num_tokens": 1223197.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.825186395645142,
|
|
"epoch": 0.5789473684210527,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.6581,
|
|
"mean_token_accuracy": 0.12425737157464027,
|
|
"num_tokens": 1234817.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.8655126094818115,
|
|
"epoch": 0.5847953216374269,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.7119,
|
|
"mean_token_accuracy": 0.13079003244638443,
|
|
"num_tokens": 1249305.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5847953216374269,
|
|
"eval_entropy": 7.929424790857816,
|
|
"eval_loss": 7.730849742889404,
|
|
"eval_mean_token_accuracy": 0.11598945001640239,
|
|
"eval_num_tokens": 1249305.0,
|
|
"eval_runtime": 2.3042,
|
|
"eval_samples_per_second": 1174.802,
|
|
"eval_steps_per_second": 147.121,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.852921438217163,
|
|
"epoch": 0.5906432748538012,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.6822,
|
|
"mean_token_accuracy": 0.12737054675817489,
|
|
"num_tokens": 1262182.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.7925210952758786,
|
|
"epoch": 0.5964912280701754,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.7134,
|
|
"mean_token_accuracy": 0.13160952627658845,
|
|
"num_tokens": 1274616.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.7616315364837645,
|
|
"epoch": 0.6023391812865497,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.7244,
|
|
"mean_token_accuracy": 0.11796518042683601,
|
|
"num_tokens": 1287306.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.773230743408203,
|
|
"epoch": 0.6081871345029239,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.5783,
|
|
"mean_token_accuracy": 0.13205838426947594,
|
|
"num_tokens": 1299472.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.825972223281861,
|
|
"epoch": 0.6140350877192983,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.5578,
|
|
"mean_token_accuracy": 0.13092489391565323,
|
|
"num_tokens": 1312636.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.714103269577026,
|
|
"epoch": 0.6198830409356725,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.5719,
|
|
"mean_token_accuracy": 0.1383853517472744,
|
|
"num_tokens": 1324058.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.7577794075012205,
|
|
"epoch": 0.6257309941520468,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.6831,
|
|
"mean_token_accuracy": 0.12188005968928337,
|
|
"num_tokens": 1337605.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.853874206542969,
|
|
"epoch": 0.631578947368421,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.6867,
|
|
"mean_token_accuracy": 0.12546794787049292,
|
|
"num_tokens": 1350658.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.805501413345337,
|
|
"epoch": 0.6374269005847953,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.6069,
|
|
"mean_token_accuracy": 0.12714534625411034,
|
|
"num_tokens": 1362091.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.723537731170654,
|
|
"epoch": 0.6432748538011696,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.6142,
|
|
"mean_token_accuracy": 0.1300032578408718,
|
|
"num_tokens": 1374055.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.724965763092041,
|
|
"epoch": 0.6491228070175439,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.5059,
|
|
"mean_token_accuracy": 0.12956581264734268,
|
|
"num_tokens": 1385269.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.69310941696167,
|
|
"epoch": 0.6549707602339181,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.6122,
|
|
"mean_token_accuracy": 0.13100477755069734,
|
|
"num_tokens": 1396559.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.651605224609375,
|
|
"epoch": 0.6608187134502924,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.6413,
|
|
"mean_token_accuracy": 0.13066264986991882,
|
|
"num_tokens": 1408838.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.753659629821778,
|
|
"epoch": 0.6666666666666666,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.5345,
|
|
"mean_token_accuracy": 0.1306803986430168,
|
|
"num_tokens": 1421023.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.645546102523804,
|
|
"epoch": 0.672514619883041,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.5958,
|
|
"mean_token_accuracy": 0.13120171204209327,
|
|
"num_tokens": 1433912.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.768090343475341,
|
|
"epoch": 0.6783625730994152,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.6014,
|
|
"mean_token_accuracy": 0.12672622352838517,
|
|
"num_tokens": 1445262.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.660789060592651,
|
|
"epoch": 0.6842105263157895,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.5561,
|
|
"mean_token_accuracy": 0.1315220259130001,
|
|
"num_tokens": 1457927.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.648988533020019,
|
|
"epoch": 0.6900584795321637,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.6251,
|
|
"mean_token_accuracy": 0.12571160346269608,
|
|
"num_tokens": 1470116.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.753493022918701,
|
|
"epoch": 0.695906432748538,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.5099,
|
|
"mean_token_accuracy": 0.1337502546608448,
|
|
"num_tokens": 1482067.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.677929210662842,
|
|
"epoch": 0.7017543859649122,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.598,
|
|
"mean_token_accuracy": 0.12390129417181014,
|
|
"num_tokens": 1493902.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.769023275375366,
|
|
"epoch": 0.7076023391812866,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.5527,
|
|
"mean_token_accuracy": 0.12497177794575691,
|
|
"num_tokens": 1505689.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.566888380050659,
|
|
"epoch": 0.7134502923976608,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.5007,
|
|
"mean_token_accuracy": 0.14484675377607345,
|
|
"num_tokens": 1518002.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.677926254272461,
|
|
"epoch": 0.7192982456140351,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.5296,
|
|
"mean_token_accuracy": 0.12993534058332443,
|
|
"num_tokens": 1529307.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.742590284347534,
|
|
"epoch": 0.7251461988304093,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.6741,
|
|
"mean_token_accuracy": 0.11497443467378617,
|
|
"num_tokens": 1542351.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.50280613899231,
|
|
"epoch": 0.7309941520467836,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.5082,
|
|
"mean_token_accuracy": 0.13412842527031898,
|
|
"num_tokens": 1556263.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.664288806915283,
|
|
"epoch": 0.7368421052631579,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.5242,
|
|
"mean_token_accuracy": 0.1329628087580204,
|
|
"num_tokens": 1568906.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.606091737747192,
|
|
"epoch": 0.7426900584795322,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.5109,
|
|
"mean_token_accuracy": 0.13439518883824347,
|
|
"num_tokens": 1580631.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.604536008834839,
|
|
"epoch": 0.7485380116959064,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.5301,
|
|
"mean_token_accuracy": 0.1288958840072155,
|
|
"num_tokens": 1592369.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.587004566192627,
|
|
"epoch": 0.7543859649122807,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.5573,
|
|
"mean_token_accuracy": 0.1358466237783432,
|
|
"num_tokens": 1604888.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.692623138427734,
|
|
"epoch": 0.7602339181286549,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.4648,
|
|
"mean_token_accuracy": 0.13601355329155923,
|
|
"num_tokens": 1617128.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.592813205718994,
|
|
"epoch": 0.7660818713450293,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.5261,
|
|
"mean_token_accuracy": 0.12374127358198166,
|
|
"num_tokens": 1629350.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.661840629577637,
|
|
"epoch": 0.7719298245614035,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.5617,
|
|
"mean_token_accuracy": 0.1267032027244568,
|
|
"num_tokens": 1643082.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.53505277633667,
|
|
"epoch": 0.7777777777777778,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.5229,
|
|
"mean_token_accuracy": 0.13297795206308366,
|
|
"num_tokens": 1654785.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.66111946105957,
|
|
"epoch": 0.783625730994152,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.565,
|
|
"mean_token_accuracy": 0.12604237273335456,
|
|
"num_tokens": 1667704.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.700377321243286,
|
|
"epoch": 0.7894736842105263,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.5012,
|
|
"mean_token_accuracy": 0.12841907143592834,
|
|
"num_tokens": 1680459.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.515329790115357,
|
|
"epoch": 0.7953216374269005,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.4115,
|
|
"mean_token_accuracy": 0.1341998517513275,
|
|
"num_tokens": 1692912.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.572557640075684,
|
|
"epoch": 0.8011695906432749,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.4348,
|
|
"mean_token_accuracy": 0.13281576186418534,
|
|
"num_tokens": 1705329.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.603346347808838,
|
|
"epoch": 0.8070175438596491,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.5435,
|
|
"mean_token_accuracy": 0.12343821972608567,
|
|
"num_tokens": 1717283.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.513457489013672,
|
|
"epoch": 0.8128654970760234,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.483,
|
|
"mean_token_accuracy": 0.1284686468541622,
|
|
"num_tokens": 1730308.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.676495838165283,
|
|
"epoch": 0.8187134502923976,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.5498,
|
|
"mean_token_accuracy": 0.1264888271689415,
|
|
"num_tokens": 1742716.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.500397348403931,
|
|
"epoch": 0.8245614035087719,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.4877,
|
|
"mean_token_accuracy": 0.13529630452394487,
|
|
"num_tokens": 1754183.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.494405746459961,
|
|
"epoch": 0.8304093567251462,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.4888,
|
|
"mean_token_accuracy": 0.1257370926439762,
|
|
"num_tokens": 1766601.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.553955984115601,
|
|
"epoch": 0.8362573099415205,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.3649,
|
|
"mean_token_accuracy": 0.14501583501696585,
|
|
"num_tokens": 1778756.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.401547574996949,
|
|
"epoch": 0.8421052631578947,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.4454,
|
|
"mean_token_accuracy": 0.13954100236296654,
|
|
"num_tokens": 1791485.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.540654993057251,
|
|
"epoch": 0.847953216374269,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.4696,
|
|
"mean_token_accuracy": 0.12822862416505815,
|
|
"num_tokens": 1803712.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.457326412200928,
|
|
"epoch": 0.8538011695906432,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.4343,
|
|
"mean_token_accuracy": 0.12846877947449684,
|
|
"num_tokens": 1818387.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.5405213832855225,
|
|
"epoch": 0.8596491228070176,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.4252,
|
|
"mean_token_accuracy": 0.1321238525211811,
|
|
"num_tokens": 1830113.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.537390804290771,
|
|
"epoch": 0.8654970760233918,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.5229,
|
|
"mean_token_accuracy": 0.13313855454325677,
|
|
"num_tokens": 1841789.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.509795761108398,
|
|
"epoch": 0.8713450292397661,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.389,
|
|
"mean_token_accuracy": 0.13464695289731027,
|
|
"num_tokens": 1853669.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.390686273574829,
|
|
"epoch": 0.8771929824561403,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.456,
|
|
"mean_token_accuracy": 0.1308678649365902,
|
|
"num_tokens": 1865322.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.4917685985565186,
|
|
"epoch": 0.8830409356725146,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.3569,
|
|
"mean_token_accuracy": 0.13735375180840492,
|
|
"num_tokens": 1879039.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.56513204574585,
|
|
"epoch": 0.8888888888888888,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.499,
|
|
"mean_token_accuracy": 0.12612639144062995,
|
|
"num_tokens": 1893078.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.541068363189697,
|
|
"epoch": 0.8947368421052632,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.5139,
|
|
"mean_token_accuracy": 0.12583210840821266,
|
|
"num_tokens": 1907399.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.450762844085693,
|
|
"epoch": 0.9005847953216374,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.4699,
|
|
"mean_token_accuracy": 0.1259562186896801,
|
|
"num_tokens": 1919290.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.545792961120606,
|
|
"epoch": 0.9064327485380117,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.4182,
|
|
"mean_token_accuracy": 0.13090129122138022,
|
|
"num_tokens": 1931658.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.500565242767334,
|
|
"epoch": 0.9122807017543859,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.4377,
|
|
"mean_token_accuracy": 0.12877111062407492,
|
|
"num_tokens": 1943881.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.516105222702026,
|
|
"epoch": 0.9181286549707602,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.3502,
|
|
"mean_token_accuracy": 0.13375887349247934,
|
|
"num_tokens": 1956288.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.317897605895996,
|
|
"epoch": 0.9239766081871345,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.2505,
|
|
"mean_token_accuracy": 0.1360914021730423,
|
|
"num_tokens": 1967265.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.435814285278321,
|
|
"epoch": 0.9298245614035088,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.4049,
|
|
"mean_token_accuracy": 0.13047077506780624,
|
|
"num_tokens": 1980039.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.459825086593628,
|
|
"epoch": 0.935672514619883,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.361,
|
|
"mean_token_accuracy": 0.13110343888401985,
|
|
"num_tokens": 1992735.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.465135145187378,
|
|
"epoch": 0.9415204678362573,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.3883,
|
|
"mean_token_accuracy": 0.13711876720190047,
|
|
"num_tokens": 2004545.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.346128177642822,
|
|
"epoch": 0.9473684210526315,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.3759,
|
|
"mean_token_accuracy": 0.13430287092924117,
|
|
"num_tokens": 2017100.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.417292737960816,
|
|
"epoch": 0.9532163742690059,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.3286,
|
|
"mean_token_accuracy": 0.14560126513242722,
|
|
"num_tokens": 2029868.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.466277170181274,
|
|
"epoch": 0.9590643274853801,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.3827,
|
|
"mean_token_accuracy": 0.12977704107761384,
|
|
"num_tokens": 2041334.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.440905380249023,
|
|
"epoch": 0.9649122807017544,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.448,
|
|
"mean_token_accuracy": 0.1350324772298336,
|
|
"num_tokens": 2054797.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.432639074325562,
|
|
"epoch": 0.9707602339181286,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.3314,
|
|
"mean_token_accuracy": 0.13245931789278984,
|
|
"num_tokens": 2068337.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.450676774978637,
|
|
"epoch": 0.9766081871345029,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.363,
|
|
"mean_token_accuracy": 0.1392313800752163,
|
|
"num_tokens": 2079984.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.329059696197509,
|
|
"epoch": 0.9824561403508771,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.3717,
|
|
"mean_token_accuracy": 0.1394517168402672,
|
|
"num_tokens": 2093069.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.384470462799072,
|
|
"epoch": 0.9883040935672515,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.3291,
|
|
"mean_token_accuracy": 0.13665454760193824,
|
|
"num_tokens": 2104753.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.348030519485474,
|
|
"epoch": 0.9941520467836257,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.3418,
|
|
"mean_token_accuracy": 0.1328301377594471,
|
|
"num_tokens": 2116576.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.415113210678101,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.3202,
|
|
"mean_token_accuracy": 0.1307046242058277,
|
|
"num_tokens": 2129484.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.309794139862061,
|
|
"epoch": 1.0058479532163742,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.1455,
|
|
"mean_token_accuracy": 0.14334668442606927,
|
|
"num_tokens": 2142000.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.259340381622314,
|
|
"epoch": 1.0116959064327484,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.1,
|
|
"mean_token_accuracy": 0.1472266711294651,
|
|
"num_tokens": 2152965.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.345311832427979,
|
|
"epoch": 1.0175438596491229,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.1904,
|
|
"mean_token_accuracy": 0.13219141513109206,
|
|
"num_tokens": 2165078.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.180283641815185,
|
|
"epoch": 1.023391812865497,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.1359,
|
|
"mean_token_accuracy": 0.14372573569417,
|
|
"num_tokens": 2176706.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.374202919006348,
|
|
"epoch": 1.0292397660818713,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.1526,
|
|
"mean_token_accuracy": 0.13850091844797136,
|
|
"num_tokens": 2190157.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.3007509231567385,
|
|
"epoch": 1.0350877192982457,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.1248,
|
|
"mean_token_accuracy": 0.14188719019293786,
|
|
"num_tokens": 2201821.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.164031553268432,
|
|
"epoch": 1.04093567251462,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.0664,
|
|
"mean_token_accuracy": 0.1440157599747181,
|
|
"num_tokens": 2214504.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.243292570114136,
|
|
"epoch": 1.0467836257309941,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.1924,
|
|
"mean_token_accuracy": 0.13835916370153428,
|
|
"num_tokens": 2226318.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.315338039398194,
|
|
"epoch": 1.0526315789473684,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.2242,
|
|
"mean_token_accuracy": 0.13328551575541497,
|
|
"num_tokens": 2237892.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.251867389678955,
|
|
"epoch": 1.0584795321637426,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.1719,
|
|
"mean_token_accuracy": 0.1382812075316906,
|
|
"num_tokens": 2250656.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.2664087295532225,
|
|
"epoch": 1.064327485380117,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.1879,
|
|
"mean_token_accuracy": 0.13563519567251206,
|
|
"num_tokens": 2263685.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.246600580215454,
|
|
"epoch": 1.0701754385964912,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.1603,
|
|
"mean_token_accuracy": 0.13542255833745004,
|
|
"num_tokens": 2275413.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.257514142990113,
|
|
"epoch": 1.0760233918128654,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.1339,
|
|
"mean_token_accuracy": 0.13871516659855843,
|
|
"num_tokens": 2288165.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.2856926918029785,
|
|
"epoch": 1.0818713450292399,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.1488,
|
|
"mean_token_accuracy": 0.1338664099574089,
|
|
"num_tokens": 2300574.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.196798849105835,
|
|
"epoch": 1.087719298245614,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.148,
|
|
"mean_token_accuracy": 0.1378339521586895,
|
|
"num_tokens": 2311631.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.23977427482605,
|
|
"epoch": 1.0935672514619883,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.169,
|
|
"mean_token_accuracy": 0.1378619410097599,
|
|
"num_tokens": 2324312.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.282365465164185,
|
|
"epoch": 1.0994152046783625,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.2368,
|
|
"mean_token_accuracy": 0.1315642088651657,
|
|
"num_tokens": 2337747.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.191779136657715,
|
|
"epoch": 1.1052631578947367,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.1814,
|
|
"mean_token_accuracy": 0.13839799612760545,
|
|
"num_tokens": 2349830.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.2542595863342285,
|
|
"epoch": 1.1111111111111112,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.1209,
|
|
"mean_token_accuracy": 0.14533643126487733,
|
|
"num_tokens": 2361825.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.1857686042785645,
|
|
"epoch": 1.1169590643274854,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.0716,
|
|
"mean_token_accuracy": 0.14158036708831787,
|
|
"num_tokens": 2373963.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.213093662261963,
|
|
"epoch": 1.1228070175438596,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.2192,
|
|
"mean_token_accuracy": 0.13446253091096877,
|
|
"num_tokens": 2386348.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.232397890090942,
|
|
"epoch": 1.128654970760234,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.1337,
|
|
"mean_token_accuracy": 0.13307930007576943,
|
|
"num_tokens": 2398338.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.1732768535614015,
|
|
"epoch": 1.1345029239766082,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.0945,
|
|
"mean_token_accuracy": 0.1407437168061733,
|
|
"num_tokens": 2411855.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.160850143432617,
|
|
"epoch": 1.1403508771929824,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.1619,
|
|
"mean_token_accuracy": 0.14437102526426315,
|
|
"num_tokens": 2424271.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.1891566753387455,
|
|
"epoch": 1.1461988304093567,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.1708,
|
|
"mean_token_accuracy": 0.1375895693898201,
|
|
"num_tokens": 2436668.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.174351406097412,
|
|
"epoch": 1.1520467836257309,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.1176,
|
|
"mean_token_accuracy": 0.13666361346840858,
|
|
"num_tokens": 2449451.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.188272762298584,
|
|
"epoch": 1.1578947368421053,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.2191,
|
|
"mean_token_accuracy": 0.13468535766005515,
|
|
"num_tokens": 2463107.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.275109195709229,
|
|
"epoch": 1.1637426900584795,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.1589,
|
|
"mean_token_accuracy": 0.13581722751259803,
|
|
"num_tokens": 2475443.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.135958623886109,
|
|
"epoch": 1.1695906432748537,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.0839,
|
|
"mean_token_accuracy": 0.1393029898405075,
|
|
"num_tokens": 2486884.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.1695906432748537,
|
|
"eval_entropy": 7.071001939013996,
|
|
"eval_loss": 7.3009033203125,
|
|
"eval_mean_token_accuracy": 0.12748390563920273,
|
|
"eval_num_tokens": 2486884.0,
|
|
"eval_runtime": 2.2417,
|
|
"eval_samples_per_second": 1207.555,
|
|
"eval_steps_per_second": 151.223,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.080128574371338,
|
|
"epoch": 1.1754385964912282,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999996875147963,
|
|
"loss": 7.0353,
|
|
"mean_token_accuracy": 0.14896628335118295,
|
|
"num_tokens": 2499310.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.226775979995727,
|
|
"epoch": 1.1812865497076024,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999984180451436,
|
|
"loss": 7.1649,
|
|
"mean_token_accuracy": 0.1328574500977993,
|
|
"num_tokens": 2512230.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.102473497390747,
|
|
"epoch": 1.1871345029239766,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999961720662221,
|
|
"loss": 7.2272,
|
|
"mean_token_accuracy": 0.13584866672754287,
|
|
"num_tokens": 2525255.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.121642780303955,
|
|
"epoch": 1.1929824561403508,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999929495877796,
|
|
"loss": 7.118,
|
|
"mean_token_accuracy": 0.13906984254717827,
|
|
"num_tokens": 2537672.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.139839220046997,
|
|
"epoch": 1.198830409356725,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999887506238018,
|
|
"loss": 7.1251,
|
|
"mean_token_accuracy": 0.13929466605186464,
|
|
"num_tokens": 2550371.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.115478038787842,
|
|
"epoch": 1.2046783625730995,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999835751925124,
|
|
"loss": 7.1417,
|
|
"mean_token_accuracy": 0.13262443989515305,
|
|
"num_tokens": 2563708.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.243052911758423,
|
|
"epoch": 1.2105263157894737,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999774233163734,
|
|
"loss": 7.1046,
|
|
"mean_token_accuracy": 0.13376289308071138,
|
|
"num_tokens": 2576529.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.117304182052612,
|
|
"epoch": 1.2163742690058479,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999702950220841,
|
|
"loss": 7.1329,
|
|
"mean_token_accuracy": 0.13227491527795793,
|
|
"num_tokens": 2588970.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.104359102249146,
|
|
"epoch": 1.2222222222222223,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999621903405819,
|
|
"loss": 7.0833,
|
|
"mean_token_accuracy": 0.136895339936018,
|
|
"num_tokens": 2601134.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.140307426452637,
|
|
"epoch": 1.2280701754385965,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000499953109307042,
|
|
"loss": 7.1338,
|
|
"mean_token_accuracy": 0.13351348638534546,
|
|
"num_tokens": 2612670.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.080586051940918,
|
|
"epoch": 1.2339181286549707,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999430519608762,
|
|
"loss": 7.1057,
|
|
"mean_token_accuracy": 0.14540162831544876,
|
|
"num_tokens": 2625157.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.201329231262207,
|
|
"epoch": 1.239766081871345,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999320183457346,
|
|
"loss": 7.1085,
|
|
"mean_token_accuracy": 0.13505723252892493,
|
|
"num_tokens": 2638318.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.053661489486695,
|
|
"epoch": 1.2456140350877192,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999200085095035,
|
|
"loss": 7.074,
|
|
"mean_token_accuracy": 0.13898903355002404,
|
|
"num_tokens": 2651024.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.962784957885742,
|
|
"epoch": 1.2514619883040936,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999070225043068,
|
|
"loss": 7.0091,
|
|
"mean_token_accuracy": 0.14594187438488007,
|
|
"num_tokens": 2665568.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.270829439163208,
|
|
"epoch": 1.2573099415204678,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998930603865042,
|
|
"loss": 7.14,
|
|
"mean_token_accuracy": 0.13617520555853843,
|
|
"num_tokens": 2678748.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.042156457901001,
|
|
"epoch": 1.263157894736842,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998781222166929,
|
|
"loss": 7.0922,
|
|
"mean_token_accuracy": 0.13991621881723404,
|
|
"num_tokens": 2690682.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.046918058395386,
|
|
"epoch": 1.2690058479532165,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998622080597052,
|
|
"loss": 7.1164,
|
|
"mean_token_accuracy": 0.1319088101387024,
|
|
"num_tokens": 2703077.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.098138236999512,
|
|
"epoch": 1.2748538011695907,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998453179846098,
|
|
"loss": 7.0887,
|
|
"mean_token_accuracy": 0.13736322820186614,
|
|
"num_tokens": 2714609.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.1358263969421385,
|
|
"epoch": 1.280701754385965,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998274520647109,
|
|
"loss": 7.0558,
|
|
"mean_token_accuracy": 0.14192912206053734,
|
|
"num_tokens": 2726907.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.00035605430603,
|
|
"epoch": 1.286549707602339,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998086103775478,
|
|
"loss": 7.0525,
|
|
"mean_token_accuracy": 0.1388276733458042,
|
|
"num_tokens": 2739248.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.098528242111206,
|
|
"epoch": 1.2923976608187133,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004997887930048947,
|
|
"loss": 7.0155,
|
|
"mean_token_accuracy": 0.14455852061510086,
|
|
"num_tokens": 2753842.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.9865405559539795,
|
|
"epoch": 1.2982456140350878,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997680000327607,
|
|
"loss": 7.0665,
|
|
"mean_token_accuracy": 0.13506068885326386,
|
|
"num_tokens": 2767249.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.14464693069458,
|
|
"epoch": 1.304093567251462,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997462315513884,
|
|
"loss": 7.0602,
|
|
"mean_token_accuracy": 0.1384397678077221,
|
|
"num_tokens": 2779657.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.082661390304565,
|
|
"epoch": 1.3099415204678362,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.000499723487655255,
|
|
"loss": 6.9763,
|
|
"mean_token_accuracy": 0.14285208955407142,
|
|
"num_tokens": 2792760.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.954230690002442,
|
|
"epoch": 1.3157894736842106,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996997684430703,
|
|
"loss": 7.0164,
|
|
"mean_token_accuracy": 0.1394075021147728,
|
|
"num_tokens": 2805277.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.065880632400512,
|
|
"epoch": 1.3216374269005848,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996750740177775,
|
|
"loss": 7.1166,
|
|
"mean_token_accuracy": 0.13852448090910913,
|
|
"num_tokens": 2818341.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.145817613601684,
|
|
"epoch": 1.327485380116959,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996494044865521,
|
|
"loss": 7.0604,
|
|
"mean_token_accuracy": 0.13684254363179207,
|
|
"num_tokens": 2830199.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.950229167938232,
|
|
"epoch": 1.3333333333333333,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996227599608018,
|
|
"loss": 6.9667,
|
|
"mean_token_accuracy": 0.14581984430551528,
|
|
"num_tokens": 2842141.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.08269419670105,
|
|
"epoch": 1.3391812865497075,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995951405561658,
|
|
"loss": 7.0522,
|
|
"mean_token_accuracy": 0.13312725201249123,
|
|
"num_tokens": 2854929.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.987738466262817,
|
|
"epoch": 1.345029239766082,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995665463925142,
|
|
"loss": 7.0516,
|
|
"mean_token_accuracy": 0.14139380380511285,
|
|
"num_tokens": 2867034.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.075216960906983,
|
|
"epoch": 1.3508771929824561,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004995369775939478,
|
|
"loss": 6.9576,
|
|
"mean_token_accuracy": 0.14617282450199126,
|
|
"num_tokens": 2881254.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.949263334274292,
|
|
"epoch": 1.3567251461988303,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004995064342887972,
|
|
"loss": 7.0713,
|
|
"mean_token_accuracy": 0.14057645201683044,
|
|
"num_tokens": 2894592.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.0024169921875,
|
|
"epoch": 1.3625730994152048,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004994749166096226,
|
|
"loss": 6.9747,
|
|
"mean_token_accuracy": 0.14208125099539756,
|
|
"num_tokens": 2907196.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.013747358322144,
|
|
"epoch": 1.368421052631579,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004994424246932129,
|
|
"loss": 7.0602,
|
|
"mean_token_accuracy": 0.13577224165201188,
|
|
"num_tokens": 2920398.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.958837985992432,
|
|
"epoch": 1.3742690058479532,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004994089586805856,
|
|
"loss": 6.9887,
|
|
"mean_token_accuracy": 0.1430194616317749,
|
|
"num_tokens": 2932052.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.202078723907471,
|
|
"epoch": 1.3801169590643274,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004993745187169852,
|
|
"loss": 7.054,
|
|
"mean_token_accuracy": 0.1366309255361557,
|
|
"num_tokens": 2943620.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.892934131622314,
|
|
"epoch": 1.3859649122807016,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499339104951884,
|
|
"loss": 7.0283,
|
|
"mean_token_accuracy": 0.14119249433279038,
|
|
"num_tokens": 2956159.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.062256336212158,
|
|
"epoch": 1.391812865497076,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00049930271753898,
|
|
"loss": 7.0003,
|
|
"mean_token_accuracy": 0.13841674402356147,
|
|
"num_tokens": 2967332.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.983926677703858,
|
|
"epoch": 1.3976608187134503,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004992653566361974,
|
|
"loss": 7.0041,
|
|
"mean_token_accuracy": 0.13496125042438506,
|
|
"num_tokens": 2980168.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.037136030197144,
|
|
"epoch": 1.4035087719298245,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004992270224056849,
|
|
"loss": 7.0935,
|
|
"mean_token_accuracy": 0.1383035495877266,
|
|
"num_tokens": 2993509.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.023516464233398,
|
|
"epoch": 1.409356725146199,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004991877150138162,
|
|
"loss": 7.0183,
|
|
"mean_token_accuracy": 0.1392554260790348,
|
|
"num_tokens": 3005827.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.972181892395019,
|
|
"epoch": 1.4152046783625731,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004991474346311879,
|
|
"loss": 6.9922,
|
|
"mean_token_accuracy": 0.14359172955155372,
|
|
"num_tokens": 3018968.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.086374950408936,
|
|
"epoch": 1.4210526315789473,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991061814326198,
|
|
"loss": 7.0242,
|
|
"mean_token_accuracy": 0.13482476994395257,
|
|
"num_tokens": 3031892.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.958857440948487,
|
|
"epoch": 1.4269005847953216,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 0.0004990639555971539,
|
|
"loss": 6.9632,
|
|
"mean_token_accuracy": 0.14786728024482726,
|
|
"num_tokens": 3045994.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.935246562957763,
|
|
"epoch": 1.4327485380116958,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004990207573080533,
|
|
"loss": 6.9662,
|
|
"mean_token_accuracy": 0.14441566094756125,
|
|
"num_tokens": 3058304.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.971540212631226,
|
|
"epoch": 1.4385964912280702,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004989765867528014,
|
|
"loss": 6.9762,
|
|
"mean_token_accuracy": 0.1433918684720993,
|
|
"num_tokens": 3071102.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.901489162445069,
|
|
"epoch": 1.4444444444444444,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004989314441231019,
|
|
"loss": 7.0027,
|
|
"mean_token_accuracy": 0.13700876906514167,
|
|
"num_tokens": 3084653.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.029861402511597,
|
|
"epoch": 1.4502923976608186,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000498885329614877,
|
|
"loss": 7.0022,
|
|
"mean_token_accuracy": 0.14563447535037993,
|
|
"num_tokens": 3097425.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.994398736953736,
|
|
"epoch": 1.456140350877193,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004988382434282668,
|
|
"loss": 7.0236,
|
|
"mean_token_accuracy": 0.13943543583154677,
|
|
"num_tokens": 3111898.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.006692504882812,
|
|
"epoch": 1.4619883040935673,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000498790185767629,
|
|
"loss": 6.9668,
|
|
"mean_token_accuracy": 0.14214081391692163,
|
|
"num_tokens": 3124271.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.997828435897827,
|
|
"epoch": 1.4678362573099415,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000498741156841537,
|
|
"loss": 7.0595,
|
|
"mean_token_accuracy": 0.14109789729118347,
|
|
"num_tokens": 3136344.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.945074844360351,
|
|
"epoch": 1.4736842105263157,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00049869115686278,
|
|
"loss": 6.9611,
|
|
"mean_token_accuracy": 0.14131287783384322,
|
|
"num_tokens": 3148299.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.050410270690918,
|
|
"epoch": 1.47953216374269,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004986401860483615,
|
|
"loss": 7.0176,
|
|
"mean_token_accuracy": 0.13987123295664788,
|
|
"num_tokens": 3161537.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.907403945922852,
|
|
"epoch": 1.4853801169590644,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004985882446194984,
|
|
"loss": 6.9717,
|
|
"mean_token_accuracy": 0.13868435025215148,
|
|
"num_tokens": 3173676.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.050081634521485,
|
|
"epoch": 1.4912280701754386,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 0.0004985353328016204,
|
|
"loss": 6.9217,
|
|
"mean_token_accuracy": 0.15403224006295205,
|
|
"num_tokens": 3186542.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.8447894096374515,
|
|
"epoch": 1.4970760233918128,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004984814508243685,
|
|
"loss": 6.9633,
|
|
"mean_token_accuracy": 0.1440097950398922,
|
|
"num_tokens": 3197867.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.016148805618286,
|
|
"epoch": 1.5029239766081872,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004984265989215942,
|
|
"loss": 6.919,
|
|
"mean_token_accuracy": 0.14048407971858978,
|
|
"num_tokens": 3209671.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.8506049633026125,
|
|
"epoch": 1.5087719298245614,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004983707773313589,
|
|
"loss": 6.9187,
|
|
"mean_token_accuracy": 0.14848409295082093,
|
|
"num_tokens": 3220707.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.872060537338257,
|
|
"epoch": 1.5146198830409356,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004983139862959322,
|
|
"loss": 6.8841,
|
|
"mean_token_accuracy": 0.1535121612250805,
|
|
"num_tokens": 3235122.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.986389875411987,
|
|
"epoch": 1.52046783625731,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004982562260617912,
|
|
"loss": 6.8776,
|
|
"mean_token_accuracy": 0.13864699825644494,
|
|
"num_tokens": 3246833.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.929480314254761,
|
|
"epoch": 1.526315789473684,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004981974968796197,
|
|
"loss": 6.9988,
|
|
"mean_token_accuracy": 0.13592579811811448,
|
|
"num_tokens": 3260493.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.903130149841308,
|
|
"epoch": 1.5321637426900585,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004981377990043064,
|
|
"loss": 6.916,
|
|
"mean_token_accuracy": 0.14037951976060867,
|
|
"num_tokens": 3272445.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.899147319793701,
|
|
"epoch": 1.5380116959064327,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004980771326949442,
|
|
"loss": 6.9349,
|
|
"mean_token_accuracy": 0.1383832648396492,
|
|
"num_tokens": 3284360.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.97801103591919,
|
|
"epoch": 1.543859649122807,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004980154982148296,
|
|
"loss": 6.934,
|
|
"mean_token_accuracy": 0.14036118239164352,
|
|
"num_tokens": 3296841.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.864973831176758,
|
|
"epoch": 1.5497076023391814,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004979528958314604,
|
|
"loss": 6.9121,
|
|
"mean_token_accuracy": 0.13837594985961915,
|
|
"num_tokens": 3309393.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.999245882034302,
|
|
"epoch": 1.5555555555555556,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004978893258165354,
|
|
"loss": 6.9954,
|
|
"mean_token_accuracy": 0.14061468243598937,
|
|
"num_tokens": 3322094.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.905670976638794,
|
|
"epoch": 1.5614035087719298,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004978247884459532,
|
|
"loss": 6.9197,
|
|
"mean_token_accuracy": 0.14624207094311714,
|
|
"num_tokens": 3334468.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.932331085205078,
|
|
"epoch": 1.5672514619883042,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004977592839998105,
|
|
"loss": 6.9064,
|
|
"mean_token_accuracy": 0.1421583592891693,
|
|
"num_tokens": 3346367.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.8350423812866214,
|
|
"epoch": 1.5730994152046782,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004976928127624013,
|
|
"loss": 6.8967,
|
|
"mean_token_accuracy": 0.1457481265068054,
|
|
"num_tokens": 3357463.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.8896355628967285,
|
|
"epoch": 1.5789473684210527,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004976253750222157,
|
|
"loss": 6.8837,
|
|
"mean_token_accuracy": 0.1439684234559536,
|
|
"num_tokens": 3369465.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.872499275207519,
|
|
"epoch": 1.5847953216374269,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004975569710719382,
|
|
"loss": 7.0025,
|
|
"mean_token_accuracy": 0.13503989055752755,
|
|
"num_tokens": 3382859.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.969785070419311,
|
|
"epoch": 1.590643274853801,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004974876012084468,
|
|
"loss": 6.9562,
|
|
"mean_token_accuracy": 0.13781241104006767,
|
|
"num_tokens": 3395431.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.84591851234436,
|
|
"epoch": 1.5964912280701755,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004974172657328116,
|
|
"loss": 6.8481,
|
|
"mean_token_accuracy": 0.1415877401828766,
|
|
"num_tokens": 3408388.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.912861585617065,
|
|
"epoch": 1.6023391812865497,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004973459649502939,
|
|
"loss": 6.864,
|
|
"mean_token_accuracy": 0.14842946082353592,
|
|
"num_tokens": 3421275.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.861267900466919,
|
|
"epoch": 1.608187134502924,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004972736991703439,
|
|
"loss": 6.9159,
|
|
"mean_token_accuracy": 0.143263141810894,
|
|
"num_tokens": 3433537.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.905822086334228,
|
|
"epoch": 1.6140350877192984,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004972004687066005,
|
|
"loss": 6.9023,
|
|
"mean_token_accuracy": 0.14614611566066743,
|
|
"num_tokens": 3445167.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.768242454528808,
|
|
"epoch": 1.6198830409356724,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000497126273876889,
|
|
"loss": 6.91,
|
|
"mean_token_accuracy": 0.1506970815360546,
|
|
"num_tokens": 3458529.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.879518270492554,
|
|
"epoch": 1.6257309941520468,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004970511150032205,
|
|
"loss": 6.9002,
|
|
"mean_token_accuracy": 0.14469428062438966,
|
|
"num_tokens": 3470315.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.876277732849121,
|
|
"epoch": 1.631578947368421,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004969749924117896,
|
|
"loss": 6.7924,
|
|
"mean_token_accuracy": 0.14415119141340255,
|
|
"num_tokens": 3481899.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.8515345573425295,
|
|
"epoch": 1.6374269005847952,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004968979064329741,
|
|
"loss": 6.9329,
|
|
"mean_token_accuracy": 0.1479893445968628,
|
|
"num_tokens": 3494076.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.007955503463745,
|
|
"epoch": 1.6432748538011697,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004968198574013323,
|
|
"loss": 6.934,
|
|
"mean_token_accuracy": 0.13703711330890656,
|
|
"num_tokens": 3507670.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.893377113342285,
|
|
"epoch": 1.6491228070175439,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000496740845655603,
|
|
"loss": 6.8404,
|
|
"mean_token_accuracy": 0.14385494217276573,
|
|
"num_tokens": 3519704.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.873383617401123,
|
|
"epoch": 1.654970760233918,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004966608715387029,
|
|
"loss": 6.8984,
|
|
"mean_token_accuracy": 0.14520218521356582,
|
|
"num_tokens": 3531665.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.817680501937867,
|
|
"epoch": 1.6608187134502925,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004965799353977248,
|
|
"loss": 6.9069,
|
|
"mean_token_accuracy": 0.14890925586223602,
|
|
"num_tokens": 3543876.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.816280269622803,
|
|
"epoch": 1.6666666666666665,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004964980375839379,
|
|
"loss": 6.8107,
|
|
"mean_token_accuracy": 0.15025360584259034,
|
|
"num_tokens": 3556517.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.824383068084717,
|
|
"epoch": 1.672514619883041,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004964151784527844,
|
|
"loss": 6.8378,
|
|
"mean_token_accuracy": 0.15114935636520385,
|
|
"num_tokens": 3569378.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.922686195373535,
|
|
"epoch": 1.6783625730994152,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000496331358363879,
|
|
"loss": 6.833,
|
|
"mean_token_accuracy": 0.14508106857538222,
|
|
"num_tokens": 3581684.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.764388227462769,
|
|
"epoch": 1.6842105263157894,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004962465776810066,
|
|
"loss": 6.7812,
|
|
"mean_token_accuracy": 0.15873703435063363,
|
|
"num_tokens": 3594961.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.858131456375122,
|
|
"epoch": 1.6900584795321638,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004961608367721217,
|
|
"loss": 6.8265,
|
|
"mean_token_accuracy": 0.14417641311883928,
|
|
"num_tokens": 3607197.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.812173652648926,
|
|
"epoch": 1.695906432748538,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004960741360093459,
|
|
"loss": 6.8627,
|
|
"mean_token_accuracy": 0.1441764861345291,
|
|
"num_tokens": 3618728.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.915625095367432,
|
|
"epoch": 1.7017543859649122,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004959864757689669,
|
|
"loss": 6.9563,
|
|
"mean_token_accuracy": 0.13298699259757996,
|
|
"num_tokens": 3630872.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.860816144943238,
|
|
"epoch": 1.7076023391812867,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004958978564314361,
|
|
"loss": 6.8413,
|
|
"mean_token_accuracy": 0.14621406346559523,
|
|
"num_tokens": 3642657.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.870521593093872,
|
|
"epoch": 1.7134502923976607,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000495808278381368,
|
|
"loss": 6.864,
|
|
"mean_token_accuracy": 0.144855272769928,
|
|
"num_tokens": 3654505.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.7053485870361325,
|
|
"epoch": 1.719298245614035,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004957177420075377,
|
|
"loss": 6.7807,
|
|
"mean_token_accuracy": 0.15822129100561141,
|
|
"num_tokens": 3668554.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.885648488998413,
|
|
"epoch": 1.7251461988304093,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004956262477028797,
|
|
"loss": 6.936,
|
|
"mean_token_accuracy": 0.1396143242716789,
|
|
"num_tokens": 3681448.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.84114933013916,
|
|
"epoch": 1.7309941520467835,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004955337958644857,
|
|
"loss": 6.8637,
|
|
"mean_token_accuracy": 0.13967747539281844,
|
|
"num_tokens": 3693677.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.7674092769622805,
|
|
"epoch": 1.736842105263158,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004954403868936032,
|
|
"loss": 6.8596,
|
|
"mean_token_accuracy": 0.14602801874279975,
|
|
"num_tokens": 3706204.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.906683778762817,
|
|
"epoch": 1.7426900584795322,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000495346021195634,
|
|
"loss": 6.8156,
|
|
"mean_token_accuracy": 0.1473531424999237,
|
|
"num_tokens": 3718255.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.838267850875854,
|
|
"epoch": 1.7485380116959064,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004952506991801319,
|
|
"loss": 6.8289,
|
|
"mean_token_accuracy": 0.14677032232284545,
|
|
"num_tokens": 3730623.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.830192804336548,
|
|
"epoch": 1.7543859649122808,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004951544212608015,
|
|
"loss": 6.8732,
|
|
"mean_token_accuracy": 0.14355233311653137,
|
|
"num_tokens": 3743636.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.7543859649122808,
|
|
"eval_entropy": 6.78361423880653,
|
|
"eval_loss": 6.987274169921875,
|
|
"eval_mean_token_accuracy": 0.1345269678072636,
|
|
"eval_num_tokens": 3743636.0,
|
|
"eval_runtime": 2.238,
|
|
"eval_samples_per_second": 1209.556,
|
|
"eval_steps_per_second": 151.474,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.85185809135437,
|
|
"epoch": 1.7602339181286548,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004950571878554956,
|
|
"loss": 6.9139,
|
|
"mean_token_accuracy": 0.13569407239556314,
|
|
"num_tokens": 3755608.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.905317783355713,
|
|
"epoch": 1.7660818713450293,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004949589993862143,
|
|
"loss": 6.8189,
|
|
"mean_token_accuracy": 0.14117416441440583,
|
|
"num_tokens": 3768208.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.804564189910889,
|
|
"epoch": 1.7719298245614035,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004948598562791025,
|
|
"loss": 6.8354,
|
|
"mean_token_accuracy": 0.14516011774539947,
|
|
"num_tokens": 3780550.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.734309101104737,
|
|
"epoch": 1.7777777777777777,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004947597589644484,
|
|
"loss": 6.817,
|
|
"mean_token_accuracy": 0.14395568445324897,
|
|
"num_tokens": 3792417.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.883447170257568,
|
|
"epoch": 1.7836257309941521,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004946587078766818,
|
|
"loss": 6.7954,
|
|
"mean_token_accuracy": 0.14844498932361602,
|
|
"num_tokens": 3803887.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.835381889343262,
|
|
"epoch": 1.7894736842105263,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004945567034543712,
|
|
"loss": 6.9313,
|
|
"mean_token_accuracy": 0.13484521508216857,
|
|
"num_tokens": 3816534.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.8271290302276615,
|
|
"epoch": 1.7953216374269005,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004944537461402234,
|
|
"loss": 6.8922,
|
|
"mean_token_accuracy": 0.14591248333454132,
|
|
"num_tokens": 3827790.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.8720885753631595,
|
|
"epoch": 1.801169590643275,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004943498363810807,
|
|
"loss": 6.8323,
|
|
"mean_token_accuracy": 0.1483864262700081,
|
|
"num_tokens": 3839800.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.697543144226074,
|
|
"epoch": 1.807017543859649,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004942449746279184,
|
|
"loss": 6.7608,
|
|
"mean_token_accuracy": 0.15420471429824828,
|
|
"num_tokens": 3852621.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.82108359336853,
|
|
"epoch": 1.8128654970760234,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004941391613358444,
|
|
"loss": 6.8153,
|
|
"mean_token_accuracy": 0.14005157276988028,
|
|
"num_tokens": 3864184.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.835338401794433,
|
|
"epoch": 1.8187134502923976,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000494032396964096,
|
|
"loss": 6.7941,
|
|
"mean_token_accuracy": 0.14192988201975823,
|
|
"num_tokens": 3876750.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.804272985458374,
|
|
"epoch": 1.8245614035087718,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000493924681976038,
|
|
"loss": 6.8399,
|
|
"mean_token_accuracy": 0.1471085824072361,
|
|
"num_tokens": 3888788.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.753233766555786,
|
|
"epoch": 1.8304093567251463,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004938160168391614,
|
|
"loss": 6.8905,
|
|
"mean_token_accuracy": 0.13889118731021882,
|
|
"num_tokens": 3901816.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.806808280944824,
|
|
"epoch": 1.8362573099415205,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004937064020250804,
|
|
"loss": 6.7818,
|
|
"mean_token_accuracy": 0.15054665878415108,
|
|
"num_tokens": 3913127.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.792178058624268,
|
|
"epoch": 1.8421052631578947,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004935958380095316,
|
|
"loss": 6.7941,
|
|
"mean_token_accuracy": 0.14823683351278305,
|
|
"num_tokens": 3924878.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.834120273590088,
|
|
"epoch": 1.8479532163742691,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004934843252723702,
|
|
"loss": 6.8727,
|
|
"mean_token_accuracy": 0.1455922856926918,
|
|
"num_tokens": 3938334.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.78785252571106,
|
|
"epoch": 1.8538011695906431,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00049337186429757,
|
|
"loss": 6.8358,
|
|
"mean_token_accuracy": 0.14139243289828302,
|
|
"num_tokens": 3949733.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.91608624458313,
|
|
"epoch": 1.8596491228070176,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004932584555732192,
|
|
"loss": 6.8612,
|
|
"mean_token_accuracy": 0.14494034871459008,
|
|
"num_tokens": 3963109.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.812306976318359,
|
|
"epoch": 1.8654970760233918,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004931440995915203,
|
|
"loss": 6.8472,
|
|
"mean_token_accuracy": 0.1428239457309246,
|
|
"num_tokens": 3975477.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.742663955688476,
|
|
"epoch": 1.871345029239766,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000493028796848786,
|
|
"loss": 6.8052,
|
|
"mean_token_accuracy": 0.15210191905498505,
|
|
"num_tokens": 3987757.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.828237295150757,
|
|
"epoch": 1.8771929824561404,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004929125478454389,
|
|
"loss": 6.6838,
|
|
"mean_token_accuracy": 0.14859880805015563,
|
|
"num_tokens": 3998807.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.669670343399048,
|
|
"epoch": 1.8830409356725146,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004927953530860077,
|
|
"loss": 6.7556,
|
|
"mean_token_accuracy": 0.14744624197483064,
|
|
"num_tokens": 4010320.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.863097715377807,
|
|
"epoch": 1.8888888888888888,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004926772130791261,
|
|
"loss": 6.8995,
|
|
"mean_token_accuracy": 0.13792884200811387,
|
|
"num_tokens": 4023432.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.776876592636109,
|
|
"epoch": 1.8947368421052633,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004925581283375303,
|
|
"loss": 6.8713,
|
|
"mean_token_accuracy": 0.13947884514927864,
|
|
"num_tokens": 4036282.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.789672327041626,
|
|
"epoch": 1.9005847953216373,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004924380993780566,
|
|
"loss": 6.7491,
|
|
"mean_token_accuracy": 0.15059926509857177,
|
|
"num_tokens": 4048164.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.733688116073608,
|
|
"epoch": 1.9064327485380117,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004923171267216392,
|
|
"loss": 6.7895,
|
|
"mean_token_accuracy": 0.14438801854848862,
|
|
"num_tokens": 4059287.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.773806571960449,
|
|
"epoch": 1.912280701754386,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004921952108933083,
|
|
"loss": 6.7883,
|
|
"mean_token_accuracy": 0.14180546924471854,
|
|
"num_tokens": 4071976.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.726539087295532,
|
|
"epoch": 1.9181286549707601,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004920723524221871,
|
|
"loss": 6.8096,
|
|
"mean_token_accuracy": 0.15039053112268447,
|
|
"num_tokens": 4085116.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.895672273635864,
|
|
"epoch": 1.9239766081871346,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004919485518414902,
|
|
"loss": 6.8046,
|
|
"mean_token_accuracy": 0.140822733938694,
|
|
"num_tokens": 4096778.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.680618095397949,
|
|
"epoch": 1.9298245614035088,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004918238096885213,
|
|
"loss": 6.7161,
|
|
"mean_token_accuracy": 0.14424919709563255,
|
|
"num_tokens": 4109490.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.785288667678833,
|
|
"epoch": 1.935672514619883,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00049169812650467,
|
|
"loss": 6.7552,
|
|
"mean_token_accuracy": 0.14924068748950958,
|
|
"num_tokens": 4122832.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.841014194488525,
|
|
"epoch": 1.9415204678362574,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004915715028354102,
|
|
"loss": 6.783,
|
|
"mean_token_accuracy": 0.14009461998939515,
|
|
"num_tokens": 4134585.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.72946720123291,
|
|
"epoch": 1.9473684210526314,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004914439392302981,
|
|
"loss": 6.7209,
|
|
"mean_token_accuracy": 0.14602409675717354,
|
|
"num_tokens": 4145909.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.716705989837647,
|
|
"epoch": 1.9532163742690059,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004913154362429686,
|
|
"loss": 6.6342,
|
|
"mean_token_accuracy": 0.15917537361383438,
|
|
"num_tokens": 4157360.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.692678165435791,
|
|
"epoch": 1.95906432748538,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004911859944311341,
|
|
"loss": 6.8247,
|
|
"mean_token_accuracy": 0.1407847747206688,
|
|
"num_tokens": 4169789.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.856312084197998,
|
|
"epoch": 1.9649122807017543,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004910556143565812,
|
|
"loss": 6.8312,
|
|
"mean_token_accuracy": 0.14292604625225067,
|
|
"num_tokens": 4183131.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.738270616531372,
|
|
"epoch": 1.9707602339181287,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004909242965851688,
|
|
"loss": 6.7986,
|
|
"mean_token_accuracy": 0.14529500007629395,
|
|
"num_tokens": 4194865.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.79417314529419,
|
|
"epoch": 1.976608187134503,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004907920416868253,
|
|
"loss": 6.819,
|
|
"mean_token_accuracy": 0.1439730167388916,
|
|
"num_tokens": 4207406.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.730139255523682,
|
|
"epoch": 1.9824561403508771,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004906588502355467,
|
|
"loss": 6.7533,
|
|
"mean_token_accuracy": 0.14599005728960038,
|
|
"num_tokens": 4221623.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.724703598022461,
|
|
"epoch": 1.9883040935672516,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004905247228093933,
|
|
"loss": 6.6975,
|
|
"mean_token_accuracy": 0.15941915661096573,
|
|
"num_tokens": 4233699.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.703160762786865,
|
|
"epoch": 1.9941520467836256,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004903896599904878,
|
|
"loss": 6.7079,
|
|
"mean_token_accuracy": 0.1484083577990532,
|
|
"num_tokens": 4247669.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.775749015808105,
|
|
"epoch": 2.0,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004902536623650125,
|
|
"loss": 6.7558,
|
|
"mean_token_accuracy": 0.1439180076122284,
|
|
"num_tokens": 4258968.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.73444766998291,
|
|
"epoch": 2.0058479532163744,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004901167305232068,
|
|
"loss": 6.4691,
|
|
"mean_token_accuracy": 0.15124359130859374,
|
|
"num_tokens": 4270684.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.663851499557495,
|
|
"epoch": 2.0116959064327484,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004899788650593648,
|
|
"loss": 6.3956,
|
|
"mean_token_accuracy": 0.1556681588292122,
|
|
"num_tokens": 4282567.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.703156185150147,
|
|
"epoch": 2.017543859649123,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004898400665718325,
|
|
"loss": 6.4372,
|
|
"mean_token_accuracy": 0.15561432689428328,
|
|
"num_tokens": 4294803.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.656558609008789,
|
|
"epoch": 2.023391812865497,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004897003356630052,
|
|
"loss": 6.4204,
|
|
"mean_token_accuracy": 0.15574801713228226,
|
|
"num_tokens": 4305898.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.6269762992858885,
|
|
"epoch": 2.0292397660818713,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004895596729393252,
|
|
"loss": 6.3562,
|
|
"mean_token_accuracy": 0.15742129534482957,
|
|
"num_tokens": 4317985.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.54496283531189,
|
|
"epoch": 2.0350877192982457,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004894180790112787,
|
|
"loss": 6.4188,
|
|
"mean_token_accuracy": 0.15664637386798858,
|
|
"num_tokens": 4330482.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.694279479980469,
|
|
"epoch": 2.0409356725146197,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004892755544933934,
|
|
"loss": 6.4747,
|
|
"mean_token_accuracy": 0.1461329534649849,
|
|
"num_tokens": 4342782.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.626242208480835,
|
|
"epoch": 2.046783625730994,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004891321000042361,
|
|
"loss": 6.4792,
|
|
"mean_token_accuracy": 0.1464425027370453,
|
|
"num_tokens": 4356204.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.624731397628784,
|
|
"epoch": 2.0526315789473686,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 6.3913,
|
|
"mean_token_accuracy": 0.1517579272389412,
|
|
"num_tokens": 4370003.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.635585975646973,
|
|
"epoch": 2.0584795321637426,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048884240360655,
|
|
"loss": 6.45,
|
|
"mean_token_accuracy": 0.15283022224903106,
|
|
"num_tokens": 4381972.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.579027986526489,
|
|
"epoch": 2.064327485380117,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004886961629553242,
|
|
"loss": 6.3762,
|
|
"mean_token_accuracy": 0.15750732719898225,
|
|
"num_tokens": 4394033.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.599072217941284,
|
|
"epoch": 2.0701754385964914,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004885489948474272,
|
|
"loss": 6.4731,
|
|
"mean_token_accuracy": 0.1535159945487976,
|
|
"num_tokens": 4406799.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.641034030914307,
|
|
"epoch": 2.0760233918128654,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004884008999215792,
|
|
"loss": 6.4766,
|
|
"mean_token_accuracy": 0.148863784968853,
|
|
"num_tokens": 4420182.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.611565637588501,
|
|
"epoch": 2.08187134502924,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004882518788205225,
|
|
"loss": 6.4328,
|
|
"mean_token_accuracy": 0.15162279605865478,
|
|
"num_tokens": 4432014.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.6640540599823,
|
|
"epoch": 2.087719298245614,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004881019321910198,
|
|
"loss": 6.4501,
|
|
"mean_token_accuracy": 0.15250293016433716,
|
|
"num_tokens": 4444755.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.577320575714111,
|
|
"epoch": 2.0935672514619883,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048795106068384985,
|
|
"loss": 6.5076,
|
|
"mean_token_accuracy": 0.1541823074221611,
|
|
"num_tokens": 4456390.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.621748447418213,
|
|
"epoch": 2.0994152046783627,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000487799264953806,
|
|
"loss": 6.4261,
|
|
"mean_token_accuracy": 0.15415703356266022,
|
|
"num_tokens": 4468567.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.576422262191772,
|
|
"epoch": 2.1052631578947367,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004876465456596926,
|
|
"loss": 6.3937,
|
|
"mean_token_accuracy": 0.16213200390338897,
|
|
"num_tokens": 4479553.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.578837108612061,
|
|
"epoch": 2.111111111111111,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004874929034643223,
|
|
"loss": 6.4095,
|
|
"mean_token_accuracy": 0.1652977555990219,
|
|
"num_tokens": 4491505.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.571292304992676,
|
|
"epoch": 2.116959064327485,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004873383390345132,
|
|
"loss": 6.4702,
|
|
"mean_token_accuracy": 0.15317193418741226,
|
|
"num_tokens": 4504896.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.558498668670654,
|
|
"epoch": 2.1228070175438596,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048718285304108613,
|
|
"loss": 6.4574,
|
|
"mean_token_accuracy": 0.1517552226781845,
|
|
"num_tokens": 4516319.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.611506986618042,
|
|
"epoch": 2.128654970760234,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048702644615886137,
|
|
"loss": 6.4825,
|
|
"mean_token_accuracy": 0.14925179556012153,
|
|
"num_tokens": 4529132.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.613365411758423,
|
|
"epoch": 2.134502923976608,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048686911906665597,
|
|
"loss": 6.4493,
|
|
"mean_token_accuracy": 0.14987133592367172,
|
|
"num_tokens": 4541080.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.573991918563843,
|
|
"epoch": 2.1403508771929824,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00048671087244728093,
|
|
"loss": 6.4614,
|
|
"mean_token_accuracy": 0.15233668237924575,
|
|
"num_tokens": 4553442.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.486274766921997,
|
|
"epoch": 2.146198830409357,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048655170698753774,
|
|
"loss": 6.3905,
|
|
"mean_token_accuracy": 0.16419739574193953,
|
|
"num_tokens": 4565617.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.626535081863404,
|
|
"epoch": 2.152046783625731,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000486391623378216,
|
|
"loss": 6.4744,
|
|
"mean_token_accuracy": 0.14740555733442307,
|
|
"num_tokens": 4577418.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.598245096206665,
|
|
"epoch": 2.1578947368421053,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004862306223140901,
|
|
"loss": 6.4,
|
|
"mean_token_accuracy": 0.1555834747850895,
|
|
"num_tokens": 4589763.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.601409673690796,
|
|
"epoch": 2.1637426900584797,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048606870449391604,
|
|
"loss": 6.5003,
|
|
"mean_token_accuracy": 0.15487492382526397,
|
|
"num_tokens": 4602522.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.6073479652404785,
|
|
"epoch": 2.1695906432748537,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048590587062042907,
|
|
"loss": 6.4661,
|
|
"mean_token_accuracy": 0.15708144158124923,
|
|
"num_tokens": 4614680.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.566403198242187,
|
|
"epoch": 2.175438596491228,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048574212140033965,
|
|
"loss": 6.4809,
|
|
"mean_token_accuracy": 0.15187129229307175,
|
|
"num_tokens": 4628166.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.573957204818726,
|
|
"epoch": 2.181286549707602,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004855774575443311,
|
|
"loss": 6.5018,
|
|
"mean_token_accuracy": 0.15662527829408646,
|
|
"num_tokens": 4640200.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.616477537155151,
|
|
"epoch": 2.1871345029239766,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048541187976705645,
|
|
"loss": 6.4426,
|
|
"mean_token_accuracy": 0.15040336698293685,
|
|
"num_tokens": 4652968.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.576403617858887,
|
|
"epoch": 2.192982456140351,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048524538878713517,
|
|
"loss": 6.4527,
|
|
"mean_token_accuracy": 0.14977433681488037,
|
|
"num_tokens": 4665741.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.52663288116455,
|
|
"epoch": 2.198830409356725,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00048507798532715005,
|
|
"loss": 6.4553,
|
|
"mean_token_accuracy": 0.15197168439626693,
|
|
"num_tokens": 4680074.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.625212478637695,
|
|
"epoch": 2.2046783625730995,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048490967011364394,
|
|
"loss": 6.4871,
|
|
"mean_token_accuracy": 0.15389255732297896,
|
|
"num_tokens": 4692340.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.529519033432007,
|
|
"epoch": 2.2105263157894735,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004847404438771172,
|
|
"loss": 6.4935,
|
|
"mean_token_accuracy": 0.15022647008299828,
|
|
"num_tokens": 4704470.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.648282909393311,
|
|
"epoch": 2.216374269005848,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004845703073520236,
|
|
"loss": 6.5072,
|
|
"mean_token_accuracy": 0.14370284005999565,
|
|
"num_tokens": 4716253.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.664510059356689,
|
|
"epoch": 2.2222222222222223,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048439926127676796,
|
|
"loss": 6.6323,
|
|
"mean_token_accuracy": 0.14200105741620064,
|
|
"num_tokens": 4730083.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.622454452514648,
|
|
"epoch": 2.2280701754385963,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00048422730639370256,
|
|
"loss": 6.3976,
|
|
"mean_token_accuracy": 0.1607450395822525,
|
|
"num_tokens": 4742770.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.517550230026245,
|
|
"epoch": 2.2339181286549707,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004840544434491238,
|
|
"loss": 6.5229,
|
|
"mean_token_accuracy": 0.15023039728403093,
|
|
"num_tokens": 4754299.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.603322982788086,
|
|
"epoch": 2.239766081871345,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048388067319326937,
|
|
"loss": 6.477,
|
|
"mean_token_accuracy": 0.14898111671209335,
|
|
"num_tokens": 4766964.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.506717586517334,
|
|
"epoch": 2.245614035087719,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048370599638031464,
|
|
"loss": 6.5044,
|
|
"mean_token_accuracy": 0.14812409728765488,
|
|
"num_tokens": 4779921.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.654232549667358,
|
|
"epoch": 2.2514619883040936,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048353041376836937,
|
|
"loss": 6.5058,
|
|
"mean_token_accuracy": 0.14764770194888116,
|
|
"num_tokens": 4792233.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.529701566696167,
|
|
"epoch": 2.257309941520468,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004833539261194749,
|
|
"loss": 6.4803,
|
|
"mean_token_accuracy": 0.15641626417636872,
|
|
"num_tokens": 4803348.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.634726905822754,
|
|
"epoch": 2.263157894736842,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048317653419960007,
|
|
"loss": 6.5588,
|
|
"mean_token_accuracy": 0.14596108347177505,
|
|
"num_tokens": 4816574.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.476966142654419,
|
|
"epoch": 2.2690058479532165,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048299823877863857,
|
|
"loss": 6.4772,
|
|
"mean_token_accuracy": 0.1571604609489441,
|
|
"num_tokens": 4828008.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.661566209793091,
|
|
"epoch": 2.2748538011695905,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004828190406304054,
|
|
"loss": 6.5338,
|
|
"mean_token_accuracy": 0.14808987975120544,
|
|
"num_tokens": 4840621.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.566416645050049,
|
|
"epoch": 2.280701754385965,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004826389405326334,
|
|
"loss": 6.4336,
|
|
"mean_token_accuracy": 0.1514001801609993,
|
|
"num_tokens": 4853503.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.560876512527466,
|
|
"epoch": 2.2865497076023393,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004824579392669698,
|
|
"loss": 6.4614,
|
|
"mean_token_accuracy": 0.1499699056148529,
|
|
"num_tokens": 4865925.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.561740493774414,
|
|
"epoch": 2.2923976608187133,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048227603761897316,
|
|
"loss": 6.527,
|
|
"mean_token_accuracy": 0.1484415277838707,
|
|
"num_tokens": 4878704.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.646590280532837,
|
|
"epoch": 2.2982456140350878,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004820932363781096,
|
|
"loss": 6.4919,
|
|
"mean_token_accuracy": 0.14958563446998596,
|
|
"num_tokens": 4891750.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.548250484466553,
|
|
"epoch": 2.3040935672514617,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004819095363377496,
|
|
"loss": 6.4752,
|
|
"mean_token_accuracy": 0.15743751972913742,
|
|
"num_tokens": 4904539.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.508108806610108,
|
|
"epoch": 2.309941520467836,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048172493829516465,
|
|
"loss": 6.3696,
|
|
"mean_token_accuracy": 0.15684462636709212,
|
|
"num_tokens": 4916210.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.562199449539184,
|
|
"epoch": 2.3157894736842106,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004815394430515234,
|
|
"loss": 6.4354,
|
|
"mean_token_accuracy": 0.15617517828941346,
|
|
"num_tokens": 4929011.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.486643314361572,
|
|
"epoch": 2.3216374269005846,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004813530514118884,
|
|
"loss": 6.4113,
|
|
"mean_token_accuracy": 0.15723690688610076,
|
|
"num_tokens": 4942484.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.480406904220581,
|
|
"epoch": 2.327485380116959,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004811657641852129,
|
|
"loss": 6.4178,
|
|
"mean_token_accuracy": 0.15218075662851333,
|
|
"num_tokens": 4955395.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.5748659610748295,
|
|
"epoch": 2.3333333333333335,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048097758218433687,
|
|
"loss": 6.4423,
|
|
"mean_token_accuracy": 0.15172057300806047,
|
|
"num_tokens": 4966853.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.3933188915252686,
|
|
"epoch": 2.3391812865497075,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048078850622598355,
|
|
"loss": 6.3266,
|
|
"mean_token_accuracy": 0.16453724801540376,
|
|
"num_tokens": 4978514.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.3391812865497075,
|
|
"eval_entropy": 6.446461832277191,
|
|
"eval_loss": 6.858452796936035,
|
|
"eval_mean_token_accuracy": 0.1379934970240018,
|
|
"eval_num_tokens": 4978514.0,
|
|
"eval_runtime": 2.2419,
|
|
"eval_samples_per_second": 1207.444,
|
|
"eval_steps_per_second": 151.209,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.5199840545654295,
|
|
"epoch": 2.345029239766082,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00048059853713075637,
|
|
"loss": 6.4721,
|
|
"mean_token_accuracy": 0.14687636196613313,
|
|
"num_tokens": 4991331.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.534999179840088,
|
|
"epoch": 2.3508771929824563,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004804076757231347,
|
|
"loss": 6.4485,
|
|
"mean_token_accuracy": 0.15023165941238403,
|
|
"num_tokens": 5004652.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.5891499519348145,
|
|
"epoch": 2.3567251461988303,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048021592283147073,
|
|
"loss": 6.4571,
|
|
"mean_token_accuracy": 0.16380735784769057,
|
|
"num_tokens": 5017204.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.457752466201782,
|
|
"epoch": 2.3625730994152048,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004800232792879858,
|
|
"loss": 6.4764,
|
|
"mean_token_accuracy": 0.1578833803534508,
|
|
"num_tokens": 5029667.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.487964344024658,
|
|
"epoch": 2.3684210526315788,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047982974592876665,
|
|
"loss": 6.3272,
|
|
"mean_token_accuracy": 0.16256631016731263,
|
|
"num_tokens": 5041490.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.499410390853882,
|
|
"epoch": 2.374269005847953,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000479635323593762,
|
|
"loss": 6.4818,
|
|
"mean_token_accuracy": 0.14881361797451972,
|
|
"num_tokens": 5054353.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.571859121322632,
|
|
"epoch": 2.3801169590643276,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047944001312677867,
|
|
"loss": 6.4422,
|
|
"mean_token_accuracy": 0.1490795999765396,
|
|
"num_tokens": 5067322.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.515582656860351,
|
|
"epoch": 2.3859649122807016,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004792438153754782,
|
|
"loss": 6.4851,
|
|
"mean_token_accuracy": 0.16214649230241776,
|
|
"num_tokens": 5079957.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.477277565002441,
|
|
"epoch": 2.391812865497076,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004790467311913727,
|
|
"loss": 6.4741,
|
|
"mean_token_accuracy": 0.15025717318058013,
|
|
"num_tokens": 5091975.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.5584574222564695,
|
|
"epoch": 2.39766081871345,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004788487614298219,
|
|
"loss": 6.4746,
|
|
"mean_token_accuracy": 0.1482400640845299,
|
|
"num_tokens": 5103847.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.500200176239014,
|
|
"epoch": 2.4035087719298245,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004786499069500286,
|
|
"loss": 6.4251,
|
|
"mean_token_accuracy": 0.15661314576864244,
|
|
"num_tokens": 5115153.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.421478891372681,
|
|
"epoch": 2.409356725146199,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047845016861503556,
|
|
"loss": 6.3594,
|
|
"mean_token_accuracy": 0.1641426756978035,
|
|
"num_tokens": 5128008.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.458701324462891,
|
|
"epoch": 2.415204678362573,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047824954729172136,
|
|
"loss": 6.3414,
|
|
"mean_token_accuracy": 0.16330295652151108,
|
|
"num_tokens": 5140023.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.482297706604004,
|
|
"epoch": 2.4210526315789473,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004780480438507972,
|
|
"loss": 6.3753,
|
|
"mean_token_accuracy": 0.1609407290816307,
|
|
"num_tokens": 5152660.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.416270637512207,
|
|
"epoch": 2.426900584795322,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047784565916680216,
|
|
"loss": 6.4002,
|
|
"mean_token_accuracy": 0.1591355249285698,
|
|
"num_tokens": 5164481.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.615703105926514,
|
|
"epoch": 2.4327485380116958,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004776423941181005,
|
|
"loss": 6.4505,
|
|
"mean_token_accuracy": 0.15818746015429497,
|
|
"num_tokens": 5177178.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.491850328445435,
|
|
"epoch": 2.43859649122807,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047743824958687694,
|
|
"loss": 6.3783,
|
|
"mean_token_accuracy": 0.15687938779592514,
|
|
"num_tokens": 5189385.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.503193140029907,
|
|
"epoch": 2.4444444444444446,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047723322645913354,
|
|
"loss": 6.4422,
|
|
"mean_token_accuracy": 0.15658749267458916,
|
|
"num_tokens": 5201839.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.529379796981812,
|
|
"epoch": 2.4502923976608186,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004770273256246853,
|
|
"loss": 6.4728,
|
|
"mean_token_accuracy": 0.15012200623750688,
|
|
"num_tokens": 5214565.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.527002191543579,
|
|
"epoch": 2.456140350877193,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004768205479771566,
|
|
"loss": 6.4523,
|
|
"mean_token_accuracy": 0.15789269655942917,
|
|
"num_tokens": 5227029.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.529928922653198,
|
|
"epoch": 2.461988304093567,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047661289441397724,
|
|
"loss": 6.4556,
|
|
"mean_token_accuracy": 0.15929979234933853,
|
|
"num_tokens": 5239208.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.441249036788941,
|
|
"epoch": 2.4678362573099415,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047640436583637865,
|
|
"loss": 6.4264,
|
|
"mean_token_accuracy": 0.162189257144928,
|
|
"num_tokens": 5252272.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.589903020858765,
|
|
"epoch": 2.473684210526316,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047619496314938977,
|
|
"loss": 6.5583,
|
|
"mean_token_accuracy": 0.14669708609580995,
|
|
"num_tokens": 5264387.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.540225839614868,
|
|
"epoch": 2.47953216374269,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004759846872618332,
|
|
"loss": 6.4985,
|
|
"mean_token_accuracy": 0.15226175636053085,
|
|
"num_tokens": 5276662.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.484582710266113,
|
|
"epoch": 2.4853801169590644,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047577353908632156,
|
|
"loss": 6.4453,
|
|
"mean_token_accuracy": 0.1597101256251335,
|
|
"num_tokens": 5290734.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.408129453659058,
|
|
"epoch": 2.4912280701754383,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047556151953925287,
|
|
"loss": 6.3124,
|
|
"mean_token_accuracy": 0.16183723360300065,
|
|
"num_tokens": 5302510.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.6220180034637455,
|
|
"epoch": 2.497076023391813,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004753486295408073,
|
|
"loss": 6.4016,
|
|
"mean_token_accuracy": 0.16100193858146666,
|
|
"num_tokens": 5313243.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.374112558364868,
|
|
"epoch": 2.502923976608187,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047513487001494257,
|
|
"loss": 6.4946,
|
|
"mean_token_accuracy": 0.1533479079604149,
|
|
"num_tokens": 5325485.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.6277507781982425,
|
|
"epoch": 2.5087719298245617,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00047492024188939056,
|
|
"loss": 6.5502,
|
|
"mean_token_accuracy": 0.14766086041927337,
|
|
"num_tokens": 5339335.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.510128116607666,
|
|
"epoch": 2.5146198830409356,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004747047460956525,
|
|
"loss": 6.5008,
|
|
"mean_token_accuracy": 0.14955624788999558,
|
|
"num_tokens": 5351195.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.476936531066895,
|
|
"epoch": 2.52046783625731,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004744883835689958,
|
|
"loss": 6.4442,
|
|
"mean_token_accuracy": 0.15402765274047853,
|
|
"num_tokens": 5362884.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.478754186630249,
|
|
"epoch": 2.526315789473684,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004742711552484492,
|
|
"loss": 6.4693,
|
|
"mean_token_accuracy": 0.15400809645652772,
|
|
"num_tokens": 5375770.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.530382347106934,
|
|
"epoch": 2.5321637426900585,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047405306207679926,
|
|
"loss": 6.4428,
|
|
"mean_token_accuracy": 0.15064893513917924,
|
|
"num_tokens": 5388673.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.315913486480713,
|
|
"epoch": 2.538011695906433,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004738341050005859,
|
|
"loss": 6.3049,
|
|
"mean_token_accuracy": 0.1622413605451584,
|
|
"num_tokens": 5400454.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.5142138481140135,
|
|
"epoch": 2.543859649122807,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004736142849700986,
|
|
"loss": 6.438,
|
|
"mean_token_accuracy": 0.15746838450431824,
|
|
"num_tokens": 5413740.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.5082385540008545,
|
|
"epoch": 2.5497076023391814,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004733936029393721,
|
|
"loss": 6.452,
|
|
"mean_token_accuracy": 0.15852005034685135,
|
|
"num_tokens": 5426089.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.516745948791504,
|
|
"epoch": 2.5555555555555554,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004731720598661821,
|
|
"loss": 6.4978,
|
|
"mean_token_accuracy": 0.14672086834907533,
|
|
"num_tokens": 5438835.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.409680557250977,
|
|
"epoch": 2.56140350877193,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004729496567120415,
|
|
"loss": 6.4642,
|
|
"mean_token_accuracy": 0.15743984282016754,
|
|
"num_tokens": 5452054.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.517363023757935,
|
|
"epoch": 2.5672514619883042,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000472726394442196,
|
|
"loss": 6.4131,
|
|
"mean_token_accuracy": 0.1488945320248604,
|
|
"num_tokens": 5466455.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.461007213592529,
|
|
"epoch": 2.573099415204678,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047250227402561976,
|
|
"loss": 6.418,
|
|
"mean_token_accuracy": 0.15548876225948333,
|
|
"num_tokens": 5479990.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.5425872802734375,
|
|
"epoch": 2.5789473684210527,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004722772964350115,
|
|
"loss": 6.4704,
|
|
"mean_token_accuracy": 0.1586227610707283,
|
|
"num_tokens": 5493035.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.412850666046142,
|
|
"epoch": 2.5847953216374266,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047205146264679015,
|
|
"loss": 6.498,
|
|
"mean_token_accuracy": 0.15221143364906312,
|
|
"num_tokens": 5506093.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.515600872039795,
|
|
"epoch": 2.590643274853801,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00047182477364109045,
|
|
"loss": 6.4558,
|
|
"mean_token_accuracy": 0.1604716882109642,
|
|
"num_tokens": 5518552.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.470510339736938,
|
|
"epoch": 2.5964912280701755,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047159723040175896,
|
|
"loss": 6.468,
|
|
"mean_token_accuracy": 0.15388281643390656,
|
|
"num_tokens": 5530123.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.586093997955322,
|
|
"epoch": 2.60233918128655,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00047136883391634966,
|
|
"loss": 6.4747,
|
|
"mean_token_accuracy": 0.14859360009431838,
|
|
"num_tokens": 5542769.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.316356563568116,
|
|
"epoch": 2.608187134502924,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004711395851761198,
|
|
"loss": 6.3187,
|
|
"mean_token_accuracy": 0.17227000445127488,
|
|
"num_tokens": 5555975.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.523240566253662,
|
|
"epoch": 2.6140350877192984,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004709094851760251,
|
|
"loss": 6.4142,
|
|
"mean_token_accuracy": 0.15561964437365533,
|
|
"num_tokens": 5569059.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.491022443771362,
|
|
"epoch": 2.6198830409356724,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0004706785349147163,
|
|
"loss": 6.4596,
|
|
"mean_token_accuracy": 0.15196931660175322,
|
|
"num_tokens": 5581358.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.428774309158325,
|
|
"epoch": 2.625730994152047,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000470446735394534,
|
|
"loss": 6.4231,
|
|
"mean_token_accuracy": 0.15621133744716645,
|
|
"num_tokens": 5594256.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.536397981643677,
|
|
"epoch": 2.6315789473684212,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004702140876215047,
|
|
"loss": 6.4229,
|
|
"mean_token_accuracy": 0.15465475767850875,
|
|
"num_tokens": 5607236.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.529411268234253,
|
|
"epoch": 2.6374269005847952,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004699805926053365,
|
|
"loss": 6.5271,
|
|
"mean_token_accuracy": 0.14624501764774323,
|
|
"num_tokens": 5619550.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.470804214477539,
|
|
"epoch": 2.6432748538011697,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00046974625135941445,
|
|
"loss": 6.3994,
|
|
"mean_token_accuracy": 0.1647377222776413,
|
|
"num_tokens": 5632721.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.336612796783447,
|
|
"epoch": 2.6491228070175437,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046951106490079636,
|
|
"loss": 6.2576,
|
|
"mean_token_accuracy": 0.15811193138360977,
|
|
"num_tokens": 5643672.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.481063795089722,
|
|
"epoch": 2.654970760233918,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046927503425020847,
|
|
"loss": 6.4026,
|
|
"mean_token_accuracy": 0.15492139309644698,
|
|
"num_tokens": 5656072.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.477350425720215,
|
|
"epoch": 2.6608187134502925,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046903816043204045,
|
|
"loss": 6.4076,
|
|
"mean_token_accuracy": 0.15917783379554748,
|
|
"num_tokens": 5668365.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.436186981201172,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004688004444743419,
|
|
"loss": 6.3806,
|
|
"mean_token_accuracy": 0.15927661657333375,
|
|
"num_tokens": 5680013.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.420562314987182,
|
|
"epoch": 2.672514619883041,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000468561887408817,
|
|
"loss": 6.3841,
|
|
"mean_token_accuracy": 0.16449666768312454,
|
|
"num_tokens": 5691884.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.448697948455811,
|
|
"epoch": 2.678362573099415,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004683224902708206,
|
|
"loss": 6.4672,
|
|
"mean_token_accuracy": 0.1509403944015503,
|
|
"num_tokens": 5704340.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.5656813621521,
|
|
"epoch": 2.6842105263157894,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004680822540993534,
|
|
"loss": 6.5131,
|
|
"mean_token_accuracy": 0.15306990891695021,
|
|
"num_tokens": 5717792.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.386775875091553,
|
|
"epoch": 2.690058479532164,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046784117993705753,
|
|
"loss": 6.3883,
|
|
"mean_token_accuracy": 0.15634631961584092,
|
|
"num_tokens": 5730237.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.417804336547851,
|
|
"epoch": 2.6959064327485383,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046759926883021234,
|
|
"loss": 6.3995,
|
|
"mean_token_accuracy": 0.15913025587797164,
|
|
"num_tokens": 5743953.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.40196533203125,
|
|
"epoch": 2.7017543859649122,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00046735652182872925,
|
|
"loss": 6.3454,
|
|
"mean_token_accuracy": 0.16076460778713225,
|
|
"num_tokens": 5756148.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.473367595672608,
|
|
"epoch": 2.7076023391812867,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004671129399861476,
|
|
"loss": 6.3831,
|
|
"mean_token_accuracy": 0.1630585789680481,
|
|
"num_tokens": 5769163.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.463471412658691,
|
|
"epoch": 2.7134502923976607,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046686852435963016,
|
|
"loss": 6.4201,
|
|
"mean_token_accuracy": 0.15539160817861558,
|
|
"num_tokens": 5781832.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.488919401168824,
|
|
"epoch": 2.719298245614035,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004666232760099583,
|
|
"loss": 6.4088,
|
|
"mean_token_accuracy": 0.15347570031881333,
|
|
"num_tokens": 5794803.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.401149463653565,
|
|
"epoch": 2.7251461988304095,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046637719600152737,
|
|
"loss": 6.438,
|
|
"mean_token_accuracy": 0.15484387502074243,
|
|
"num_tokens": 5806768.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.560527515411377,
|
|
"epoch": 2.7309941520467835,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046613028540234226,
|
|
"loss": 6.4406,
|
|
"mean_token_accuracy": 0.14922621697187424,
|
|
"num_tokens": 5819113.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.475566816329956,
|
|
"epoch": 2.736842105263158,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046588254528401264,
|
|
"loss": 6.4247,
|
|
"mean_token_accuracy": 0.15603845268487931,
|
|
"num_tokens": 5830067.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.481828498840332,
|
|
"epoch": 2.742690058479532,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004656339767217485,
|
|
"loss": 6.4547,
|
|
"mean_token_accuracy": 0.15912400484085082,
|
|
"num_tokens": 5842882.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.425919151306152,
|
|
"epoch": 2.7485380116959064,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046538458079435516,
|
|
"loss": 6.4218,
|
|
"mean_token_accuracy": 0.15698028802871705,
|
|
"num_tokens": 5854041.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.461370134353638,
|
|
"epoch": 2.754385964912281,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004651343585842287,
|
|
"loss": 6.4289,
|
|
"mean_token_accuracy": 0.15794608741998672,
|
|
"num_tokens": 5866027.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.4920432567596436,
|
|
"epoch": 2.760233918128655,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00046488331117735176,
|
|
"loss": 6.4735,
|
|
"mean_token_accuracy": 0.15814025700092316,
|
|
"num_tokens": 5877064.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.470999765396118,
|
|
"epoch": 2.7660818713450293,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004646314396632878,
|
|
"loss": 6.4234,
|
|
"mean_token_accuracy": 0.1507804736495018,
|
|
"num_tokens": 5889190.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.467544317245483,
|
|
"epoch": 2.7719298245614032,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004643787451351774,
|
|
"loss": 6.4068,
|
|
"mean_token_accuracy": 0.16288078129291533,
|
|
"num_tokens": 5901615.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.401583480834961,
|
|
"epoch": 2.7777777777777777,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00046412522868973297,
|
|
"loss": 6.3943,
|
|
"mean_token_accuracy": 0.16063528507947922,
|
|
"num_tokens": 5913797.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.556414985656739,
|
|
"epoch": 2.783625730994152,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046387089142723386,
|
|
"loss": 6.504,
|
|
"mean_token_accuracy": 0.15080462843179704,
|
|
"num_tokens": 5927607.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.362700653076172,
|
|
"epoch": 2.7894736842105265,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00046361573445152224,
|
|
"loss": 6.3209,
|
|
"mean_token_accuracy": 0.16778908371925355,
|
|
"num_tokens": 5940254.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.2848857879638675,
|
|
"epoch": 2.7953216374269005,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046335975886999754,
|
|
"loss": 6.1674,
|
|
"mean_token_accuracy": 0.1768309459090233,
|
|
"num_tokens": 5953341.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.351932191848755,
|
|
"epoch": 2.801169590643275,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046310296579361225,
|
|
"loss": 6.196,
|
|
"mean_token_accuracy": 0.174995157122612,
|
|
"num_tokens": 5965148.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.458129262924194,
|
|
"epoch": 2.807017543859649,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004628453563368666,
|
|
"loss": 6.494,
|
|
"mean_token_accuracy": 0.15046632140874863,
|
|
"num_tokens": 5979143.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.479505300521851,
|
|
"epoch": 2.8128654970760234,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004625869316178043,
|
|
"loss": 6.4556,
|
|
"mean_token_accuracy": 0.1548162132501602,
|
|
"num_tokens": 5992735.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.444118404388428,
|
|
"epoch": 2.818713450292398,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000462327692758007,
|
|
"loss": 6.5152,
|
|
"mean_token_accuracy": 0.14927757680416107,
|
|
"num_tokens": 6005033.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.471355819702149,
|
|
"epoch": 2.824561403508772,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046206764088259003,
|
|
"loss": 6.3743,
|
|
"mean_token_accuracy": 0.15734981149435043,
|
|
"num_tokens": 6016746.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.407990026473999,
|
|
"epoch": 2.8304093567251463,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00046180677712019707,
|
|
"loss": 6.4686,
|
|
"mean_token_accuracy": 0.15794578492641448,
|
|
"num_tokens": 6029774.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.508468055725098,
|
|
"epoch": 2.8362573099415203,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004615451026029957,
|
|
"loss": 6.5018,
|
|
"mean_token_accuracy": 0.1469054102897644,
|
|
"num_tokens": 6043773.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.470884323120117,
|
|
"epoch": 2.8421052631578947,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000461282618466672,
|
|
"loss": 6.4126,
|
|
"mean_token_accuracy": 0.15324558317661285,
|
|
"num_tokens": 6055840.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.442924118041992,
|
|
"epoch": 2.847953216374269,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046101932585042597,
|
|
"loss": 6.3295,
|
|
"mean_token_accuracy": 0.1638964146375656,
|
|
"num_tokens": 6069170.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.4292243957519535,
|
|
"epoch": 2.853801169590643,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046075522589696636,
|
|
"loss": 6.3788,
|
|
"mean_token_accuracy": 0.15840657502412797,
|
|
"num_tokens": 6080888.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.360028314590454,
|
|
"epoch": 2.8596491228070176,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000460490319752506,
|
|
"loss": 6.3307,
|
|
"mean_token_accuracy": 0.1591005802154541,
|
|
"num_tokens": 6093035.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.416272878646851,
|
|
"epoch": 2.8654970760233915,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00046022460856675663,
|
|
"loss": 6.395,
|
|
"mean_token_accuracy": 0.1552513599395752,
|
|
"num_tokens": 6105963.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.520850372314453,
|
|
"epoch": 2.871345029239766,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004599580934929237,
|
|
"loss": 6.4472,
|
|
"mean_token_accuracy": 0.15403655767440796,
|
|
"num_tokens": 6118541.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.339373826980591,
|
|
"epoch": 2.8771929824561404,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004596907756877018,
|
|
"loss": 6.3476,
|
|
"mean_token_accuracy": 0.15784837901592255,
|
|
"num_tokens": 6131317.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.456536865234375,
|
|
"epoch": 2.883040935672515,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045942265631126935,
|
|
"loss": 6.4016,
|
|
"mean_token_accuracy": 0.15526122897863387,
|
|
"num_tokens": 6143721.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.422988605499268,
|
|
"epoch": 2.888888888888889,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004591537365272837,
|
|
"loss": 6.394,
|
|
"mean_token_accuracy": 0.15404924750328064,
|
|
"num_tokens": 6156497.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.424336671829224,
|
|
"epoch": 2.8947368421052633,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000458884017502876,
|
|
"loss": 6.3849,
|
|
"mean_token_accuracy": 0.1570442482829094,
|
|
"num_tokens": 6168140.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.450780916213989,
|
|
"epoch": 2.9005847953216373,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004586135004086461,
|
|
"loss": 6.3707,
|
|
"mean_token_accuracy": 0.1558571368455887,
|
|
"num_tokens": 6181067.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.411061382293701,
|
|
"epoch": 2.9064327485380117,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004583421864186577,
|
|
"loss": 6.3539,
|
|
"mean_token_accuracy": 0.15411918610334396,
|
|
"num_tokens": 6194072.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.4376201152801515,
|
|
"epoch": 2.912280701754386,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004580700767104328,
|
|
"loss": 6.4233,
|
|
"mean_token_accuracy": 0.15615556836128236,
|
|
"num_tokens": 6206711.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.409209823608398,
|
|
"epoch": 2.91812865497076,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045779717246494724,
|
|
"loss": 6.3584,
|
|
"mean_token_accuracy": 0.15740323215723037,
|
|
"num_tokens": 6219903.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.458169221878052,
|
|
"epoch": 2.9239766081871346,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004575234748666249,
|
|
"loss": 6.3899,
|
|
"mean_token_accuracy": 0.15397413522005082,
|
|
"num_tokens": 6232947.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.9239766081871346,
|
|
"eval_entropy": 6.289016930402908,
|
|
"eval_loss": 6.731997013092041,
|
|
"eval_mean_token_accuracy": 0.1413759137902156,
|
|
"eval_num_tokens": 6232947.0,
|
|
"eval_runtime": 2.7534,
|
|
"eval_samples_per_second": 983.158,
|
|
"eval_steps_per_second": 123.122,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.419559049606323,
|
|
"epoch": 2.9298245614035086,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000457248985103333,
|
|
"loss": 6.3755,
|
|
"mean_token_accuracy": 0.1601724475622177,
|
|
"num_tokens": 6245597.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.351097059249878,
|
|
"epoch": 2.935672514619883,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004569737043663768,
|
|
"loss": 6.3083,
|
|
"mean_token_accuracy": 0.15499867647886276,
|
|
"num_tokens": 6257745.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.459050130844116,
|
|
"epoch": 2.9415204678362574,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004566976338504945,
|
|
"loss": 6.396,
|
|
"mean_token_accuracy": 0.15999313890933992,
|
|
"num_tokens": 6268871.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.3960713863372805,
|
|
"epoch": 2.9473684210526314,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00045642077475385186,
|
|
"loss": 6.3462,
|
|
"mean_token_accuracy": 0.15839040875434876,
|
|
"num_tokens": 6281487.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.3280322551727295,
|
|
"epoch": 2.953216374269006,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004561431282780372,
|
|
"loss": 6.3334,
|
|
"mean_token_accuracy": 0.158796264231205,
|
|
"num_tokens": 6292928.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.4512107372283936,
|
|
"epoch": 2.95906432748538,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045586469562805624,
|
|
"loss": 6.3276,
|
|
"mean_token_accuracy": 0.1728435277938843,
|
|
"num_tokens": 6304340.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.350674390792847,
|
|
"epoch": 2.9649122807017543,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004555854780123265,
|
|
"loss": 6.3894,
|
|
"mean_token_accuracy": 0.15501194298267365,
|
|
"num_tokens": 6315972.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.5018211841583256,
|
|
"epoch": 2.9707602339181287,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004553054766426725,
|
|
"loss": 6.3899,
|
|
"mean_token_accuracy": 0.15937701165676116,
|
|
"num_tokens": 6327482.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.388442945480347,
|
|
"epoch": 2.976608187134503,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004550246927343203,
|
|
"loss": 6.3807,
|
|
"mean_token_accuracy": 0.15886644572019576,
|
|
"num_tokens": 6338932.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.382300615310669,
|
|
"epoch": 2.982456140350877,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00045474312750589226,
|
|
"loss": 6.3963,
|
|
"mean_token_accuracy": 0.15714696645736695,
|
|
"num_tokens": 6350770.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.499816179275513,
|
|
"epoch": 2.9883040935672516,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00045446078217940155,
|
|
"loss": 6.4166,
|
|
"mean_token_accuracy": 0.1527273342013359,
|
|
"num_tokens": 6363472.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.4609089374542235,
|
|
"epoch": 2.9941520467836256,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045417765798024716,
|
|
"loss": 6.3877,
|
|
"mean_token_accuracy": 0.15418013632297517,
|
|
"num_tokens": 6377174.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.353840494155884,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004538937561372086,
|
|
"loss": 6.3529,
|
|
"mean_token_accuracy": 0.15303714722394943,
|
|
"num_tokens": 6388452.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.495630693435669,
|
|
"epoch": 3.0058479532163744,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004536090778824398,
|
|
"loss": 6.0336,
|
|
"mean_token_accuracy": 0.1663808062672615,
|
|
"num_tokens": 6401740.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.317800140380859,
|
|
"epoch": 3.0116959064327484,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004533236244514652,
|
|
"loss": 6.0088,
|
|
"mean_token_accuracy": 0.17506007999181747,
|
|
"num_tokens": 6414850.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.410755586624146,
|
|
"epoch": 3.017543859649123,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00045303739708317303,
|
|
"loss": 6.158,
|
|
"mean_token_accuracy": 0.1593886986374855,
|
|
"num_tokens": 6429250.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.34282546043396,
|
|
"epoch": 3.023391812865497,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004527503970198105,
|
|
"loss": 5.989,
|
|
"mean_token_accuracy": 0.16927866041660308,
|
|
"num_tokens": 6440583.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.287679862976074,
|
|
"epoch": 3.0292397660818713,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004524626255069786,
|
|
"loss": 6.002,
|
|
"mean_token_accuracy": 0.1686646655201912,
|
|
"num_tokens": 6452678.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.38556399345398,
|
|
"epoch": 3.0350877192982457,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045217408379362635,
|
|
"loss": 6.0535,
|
|
"mean_token_accuracy": 0.16365511119365692,
|
|
"num_tokens": 6464751.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.198170328140259,
|
|
"epoch": 3.0409356725146197,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045188477313204536,
|
|
"loss": 6.0347,
|
|
"mean_token_accuracy": 0.16861024498939514,
|
|
"num_tokens": 6476017.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.442261600494385,
|
|
"epoch": 3.046783625730994,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045159469477786457,
|
|
"loss": 6.0412,
|
|
"mean_token_accuracy": 0.16376086920499802,
|
|
"num_tokens": 6487404.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.303383779525757,
|
|
"epoch": 3.0526315789473686,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004513038499900451,
|
|
"loss": 6.0865,
|
|
"mean_token_accuracy": 0.15962110161781312,
|
|
"num_tokens": 6500027.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.294409847259521,
|
|
"epoch": 3.0584795321637426,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000451012240030874,
|
|
"loss": 6.0868,
|
|
"mean_token_accuracy": 0.1579651951789856,
|
|
"num_tokens": 6513067.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.345212364196778,
|
|
"epoch": 3.064327485380117,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045071986616595946,
|
|
"loss": 6.034,
|
|
"mean_token_accuracy": 0.16172813028097152,
|
|
"num_tokens": 6524932.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.2884471893310545,
|
|
"epoch": 3.0701754385964914,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045042672966422506,
|
|
"loss": 6.0786,
|
|
"mean_token_accuracy": 0.15962005853652955,
|
|
"num_tokens": 6536843.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.380095529556274,
|
|
"epoch": 3.0760233918128654,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004501328317979041,
|
|
"loss": 5.9917,
|
|
"mean_token_accuracy": 0.1719750076532364,
|
|
"num_tokens": 6549414.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.299890518188477,
|
|
"epoch": 3.08187134502924,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004498381738425346,
|
|
"loss": 6.0646,
|
|
"mean_token_accuracy": 0.16587070971727372,
|
|
"num_tokens": 6561492.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.214857149124145,
|
|
"epoch": 3.087719298245614,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00044954275707695304,
|
|
"loss": 5.9737,
|
|
"mean_token_accuracy": 0.17136145383119583,
|
|
"num_tokens": 6572873.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.31891450881958,
|
|
"epoch": 3.0935672514619883,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004492465827832894,
|
|
"loss": 6.1127,
|
|
"mean_token_accuracy": 0.16485580950975418,
|
|
"num_tokens": 6584902.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.386740827560425,
|
|
"epoch": 3.0994152046783627,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00044894965224696144,
|
|
"loss": 6.1431,
|
|
"mean_token_accuracy": 0.1622363805770874,
|
|
"num_tokens": 6596792.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.351836252212524,
|
|
"epoch": 3.1052631578947367,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044865196675666886,
|
|
"loss": 6.057,
|
|
"mean_token_accuracy": 0.16659751534461975,
|
|
"num_tokens": 6608559.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.2484596252441404,
|
|
"epoch": 3.111111111111111,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004483535276043881,
|
|
"loss": 6.1129,
|
|
"mean_token_accuracy": 0.1641818404197693,
|
|
"num_tokens": 6619938.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.349063205718994,
|
|
"epoch": 3.116959064327485,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004480543360853665,
|
|
"loss": 6.0912,
|
|
"mean_token_accuracy": 0.16634685248136521,
|
|
"num_tokens": 6632850.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.195001554489136,
|
|
"epoch": 3.1228070175438596,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004477543934981167,
|
|
"loss": 6.0179,
|
|
"mean_token_accuracy": 0.17676016837358474,
|
|
"num_tokens": 6645761.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.328713607788086,
|
|
"epoch": 3.128654970760234,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044745370114441103,
|
|
"loss": 6.0701,
|
|
"mean_token_accuracy": 0.1652401402592659,
|
|
"num_tokens": 6657861.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.304843997955322,
|
|
"epoch": 3.134502923976608,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044715226032927585,
|
|
"loss": 6.0416,
|
|
"mean_token_accuracy": 0.17727303504943848,
|
|
"num_tokens": 6670495.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.15926342010498,
|
|
"epoch": 3.1403508771929824,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004468500723609859,
|
|
"loss": 5.9805,
|
|
"mean_token_accuracy": 0.1696304053068161,
|
|
"num_tokens": 6683127.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.331301021575928,
|
|
"epoch": 3.146198830409357,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044654713855105864,
|
|
"loss": 6.0829,
|
|
"mean_token_accuracy": 0.15566323101520538,
|
|
"num_tokens": 6696159.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.283361864089966,
|
|
"epoch": 3.152046783625731,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004462434602142486,
|
|
"loss": 6.1562,
|
|
"mean_token_accuracy": 0.15491898804903032,
|
|
"num_tokens": 6708993.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.365117835998535,
|
|
"epoch": 3.1578947368421053,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004459390386685414,
|
|
"loss": 6.1536,
|
|
"mean_token_accuracy": 0.15937100797891618,
|
|
"num_tokens": 6722121.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.208102130889893,
|
|
"epoch": 3.1637426900584797,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00044563387523514854,
|
|
"loss": 5.9646,
|
|
"mean_token_accuracy": 0.17161332964897155,
|
|
"num_tokens": 6734167.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.2193512439727785,
|
|
"epoch": 3.1695906432748537,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004453279712385011,
|
|
"loss": 6.0525,
|
|
"mean_token_accuracy": 0.1608000710606575,
|
|
"num_tokens": 6745720.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.286792802810669,
|
|
"epoch": 3.175438596491228,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004450213280062444,
|
|
"loss": 6.0644,
|
|
"mean_token_accuracy": 0.16539445370435715,
|
|
"num_tokens": 6757523.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.3120691776275635,
|
|
"epoch": 3.181286549707602,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044471394686923206,
|
|
"loss": 6.0696,
|
|
"mean_token_accuracy": 0.16432759016752244,
|
|
"num_tokens": 6768775.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.256685209274292,
|
|
"epoch": 3.1871345029239766,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004444058291615202,
|
|
"loss": 6.0669,
|
|
"mean_token_accuracy": 0.1624320149421692,
|
|
"num_tokens": 6781547.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.295102643966675,
|
|
"epoch": 3.192982456140351,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004440969762203618,
|
|
"loss": 6.1075,
|
|
"mean_token_accuracy": 0.1577914759516716,
|
|
"num_tokens": 6794708.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.231550931930542,
|
|
"epoch": 3.198830409356725,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004437873893862008,
|
|
"loss": 6.0411,
|
|
"mean_token_accuracy": 0.17247709333896638,
|
|
"num_tokens": 6807738.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.293457794189453,
|
|
"epoch": 3.2046783625730995,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044347707000266627,
|
|
"loss": 6.0875,
|
|
"mean_token_accuracy": 0.16551728844642638,
|
|
"num_tokens": 6820851.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.298693513870239,
|
|
"epoch": 3.2105263157894735,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00044316601941656655,
|
|
"loss": 6.0564,
|
|
"mean_token_accuracy": 0.17486980706453323,
|
|
"num_tokens": 6833888.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.246747827529907,
|
|
"epoch": 3.216374269005848,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004428542389778834,
|
|
"loss": 6.086,
|
|
"mean_token_accuracy": 0.16642929166555404,
|
|
"num_tokens": 6846104.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.269458293914795,
|
|
"epoch": 3.2222222222222223,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004425417300397665,
|
|
"loss": 6.077,
|
|
"mean_token_accuracy": 0.1683942660689354,
|
|
"num_tokens": 6857458.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.240167617797852,
|
|
"epoch": 3.2280701754385963,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044222849395852683,
|
|
"loss": 6.0298,
|
|
"mean_token_accuracy": 0.17239536941051484,
|
|
"num_tokens": 6870455.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.322045087814331,
|
|
"epoch": 3.2339181286549707,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044191453209363165,
|
|
"loss": 6.0913,
|
|
"mean_token_accuracy": 0.1617986135184765,
|
|
"num_tokens": 6883359.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.182917976379395,
|
|
"epoch": 3.239766081871345,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004415998458076978,
|
|
"loss": 6.0312,
|
|
"mean_token_accuracy": 0.16926027834415436,
|
|
"num_tokens": 6894940.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.259634113311767,
|
|
"epoch": 3.245614035087719,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044128443646648634,
|
|
"loss": 6.0877,
|
|
"mean_token_accuracy": 0.16552008986473082,
|
|
"num_tokens": 6905422.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.302772569656372,
|
|
"epoch": 3.2514619883040936,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044096830543889653,
|
|
"loss": 6.0844,
|
|
"mean_token_accuracy": 0.16717624962329863,
|
|
"num_tokens": 6918106.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.306543445587158,
|
|
"epoch": 3.257309941520468,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004406514540969596,
|
|
"loss": 6.0633,
|
|
"mean_token_accuracy": 0.1664744108915329,
|
|
"num_tokens": 6930975.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.315606260299683,
|
|
"epoch": 3.263157894736842,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044033388381583324,
|
|
"loss": 6.1529,
|
|
"mean_token_accuracy": 0.15466532707214356,
|
|
"num_tokens": 6944436.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.268398380279541,
|
|
"epoch": 3.2690058479532165,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00044001559597379503,
|
|
"loss": 6.0996,
|
|
"mean_token_accuracy": 0.16065843403339386,
|
|
"num_tokens": 6957095.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.263225793838501,
|
|
"epoch": 3.2748538011695905,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004396965919522373,
|
|
"loss": 6.0538,
|
|
"mean_token_accuracy": 0.16990793496370316,
|
|
"num_tokens": 6969792.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.283797550201416,
|
|
"epoch": 3.280701754385965,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00043937687313565997,
|
|
"loss": 6.1099,
|
|
"mean_token_accuracy": 0.1642662450671196,
|
|
"num_tokens": 6982054.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.271491861343383,
|
|
"epoch": 3.2865497076023393,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004390564409116661,
|
|
"loss": 6.0949,
|
|
"mean_token_accuracy": 0.16641355007886888,
|
|
"num_tokens": 6994057.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.246029710769653,
|
|
"epoch": 3.2923976608187133,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00043873529667095423,
|
|
"loss": 6.0992,
|
|
"mean_token_accuracy": 0.16476769596338273,
|
|
"num_tokens": 7006417.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.247483491897583,
|
|
"epoch": 3.2982456140350878,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043841344180731346,
|
|
"loss": 6.1233,
|
|
"mean_token_accuracy": 0.15914071649312972,
|
|
"num_tokens": 7018331.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.235298728942871,
|
|
"epoch": 3.3040935672514617,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000438090877717617,
|
|
"loss": 6.0166,
|
|
"mean_token_accuracy": 0.17036083489656448,
|
|
"num_tokens": 7031654.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.3546202182769775,
|
|
"epoch": 3.309941520467836,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004377676058018161,
|
|
"loss": 6.0896,
|
|
"mean_token_accuracy": 0.16434545665979386,
|
|
"num_tokens": 7044921.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.177653026580811,
|
|
"epoch": 3.3157894736842106,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004374436274629341,
|
|
"loss": 6.0585,
|
|
"mean_token_accuracy": 0.16516947448253633,
|
|
"num_tokens": 7056700.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.232658576965332,
|
|
"epoch": 3.3216374269005846,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004371189441070602,
|
|
"loss": 5.9951,
|
|
"mean_token_accuracy": 0.16497577130794525,
|
|
"num_tokens": 7069196.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.1874377727508545,
|
|
"epoch": 3.327485380116959,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004367935571433433,
|
|
"loss": 6.0283,
|
|
"mean_token_accuracy": 0.1698434144258499,
|
|
"num_tokens": 7081528.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.318471622467041,
|
|
"epoch": 3.3333333333333335,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004364674679839864,
|
|
"loss": 6.1357,
|
|
"mean_token_accuracy": 0.1573349192738533,
|
|
"num_tokens": 7093474.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.203385353088379,
|
|
"epoch": 3.3391812865497075,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004361406780442398,
|
|
"loss": 6.0511,
|
|
"mean_token_accuracy": 0.1675581693649292,
|
|
"num_tokens": 7105508.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.196110630035401,
|
|
"epoch": 3.345029239766082,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00043581318874239504,
|
|
"loss": 6.0783,
|
|
"mean_token_accuracy": 0.1669802948832512,
|
|
"num_tokens": 7117952.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.265625381469727,
|
|
"epoch": 3.3508771929824563,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043548500149977946,
|
|
"loss": 6.0727,
|
|
"mean_token_accuracy": 0.1688978374004364,
|
|
"num_tokens": 7130666.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.239003849029541,
|
|
"epoch": 3.3567251461988303,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004351561177407491,
|
|
"loss": 6.0679,
|
|
"mean_token_accuracy": 0.16751710176467896,
|
|
"num_tokens": 7142717.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.285009670257568,
|
|
"epoch": 3.3625730994152048,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004348265388926831,
|
|
"loss": 6.1153,
|
|
"mean_token_accuracy": 0.16684373319149018,
|
|
"num_tokens": 7154794.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.278839159011841,
|
|
"epoch": 3.3684210526315788,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004344962663859771,
|
|
"loss": 6.1682,
|
|
"mean_token_accuracy": 0.16089486628770827,
|
|
"num_tokens": 7168907.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.202295017242432,
|
|
"epoch": 3.374269005847953,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00043416530165403777,
|
|
"loss": 6.0044,
|
|
"mean_token_accuracy": 0.17788932025432586,
|
|
"num_tokens": 7180828.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.200469064712524,
|
|
"epoch": 3.3801169590643276,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004338336461332756,
|
|
"loss": 6.0096,
|
|
"mean_token_accuracy": 0.17090374678373338,
|
|
"num_tokens": 7193493.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.251628875732422,
|
|
"epoch": 3.3859649122807016,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00043350130126309936,
|
|
"loss": 6.0162,
|
|
"mean_token_accuracy": 0.16642655581235885,
|
|
"num_tokens": 7206033.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.223085641860962,
|
|
"epoch": 3.391812865497076,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00043316826848590955,
|
|
"loss": 6.0699,
|
|
"mean_token_accuracy": 0.1658588320016861,
|
|
"num_tokens": 7218813.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.259242057800293,
|
|
"epoch": 3.39766081871345,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004328345492470924,
|
|
"loss": 6.0759,
|
|
"mean_token_accuracy": 0.16798142790794374,
|
|
"num_tokens": 7230698.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.223297882080078,
|
|
"epoch": 3.4035087719298245,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004325001449950134,
|
|
"loss": 6.0971,
|
|
"mean_token_accuracy": 0.15926420837640762,
|
|
"num_tokens": 7243354.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.15490870475769,
|
|
"epoch": 3.409356725146199,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004321650571810109,
|
|
"loss": 6.0188,
|
|
"mean_token_accuracy": 0.17210624665021895,
|
|
"num_tokens": 7255855.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.201034116744995,
|
|
"epoch": 3.415204678362573,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004318292872593899,
|
|
"loss": 6.1138,
|
|
"mean_token_accuracy": 0.16183559745550155,
|
|
"num_tokens": 7268921.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.290073728561401,
|
|
"epoch": 3.4210526315789473,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004314928366874162,
|
|
"loss": 6.1325,
|
|
"mean_token_accuracy": 0.1615605413913727,
|
|
"num_tokens": 7281467.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.242023468017578,
|
|
"epoch": 3.426900584795322,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004311557069253092,
|
|
"loss": 6.0404,
|
|
"mean_token_accuracy": 0.17091011106967927,
|
|
"num_tokens": 7294887.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.26421046257019,
|
|
"epoch": 3.4327485380116958,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004308178994362364,
|
|
"loss": 6.1567,
|
|
"mean_token_accuracy": 0.16541029065847396,
|
|
"num_tokens": 7307009.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.212188673019409,
|
|
"epoch": 3.43859649122807,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043047941568630644,
|
|
"loss": 6.0649,
|
|
"mean_token_accuracy": 0.16204213500022888,
|
|
"num_tokens": 7319751.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.319940519332886,
|
|
"epoch": 3.4444444444444446,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043014025714456305,
|
|
"loss": 6.1958,
|
|
"mean_token_accuracy": 0.1593628481030464,
|
|
"num_tokens": 7333952.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.182245302200317,
|
|
"epoch": 3.4502923976608186,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004298004252829787,
|
|
"loss": 5.9382,
|
|
"mean_token_accuracy": 0.17446714788675308,
|
|
"num_tokens": 7346834.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.26920804977417,
|
|
"epoch": 3.456140350877193,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000429459921576448,
|
|
"loss": 6.0949,
|
|
"mean_token_accuracy": 0.1647094652056694,
|
|
"num_tokens": 7359165.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.316441059112549,
|
|
"epoch": 3.461988304093567,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00042911874750278133,
|
|
"loss": 6.1696,
|
|
"mean_token_accuracy": 0.1595632717013359,
|
|
"num_tokens": 7371585.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.197934150695801,
|
|
"epoch": 3.4678362573099415,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00042877690454269884,
|
|
"loss": 6.0755,
|
|
"mean_token_accuracy": 0.1684444710612297,
|
|
"num_tokens": 7383455.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.219165515899658,
|
|
"epoch": 3.473684210526316,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00042843439417982344,
|
|
"loss": 6.1314,
|
|
"mean_token_accuracy": 0.1630835473537445,
|
|
"num_tokens": 7395918.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.299080467224121,
|
|
"epoch": 3.47953216374269,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004280912179006746,
|
|
"loss": 6.1119,
|
|
"mean_token_accuracy": 0.16668291240930558,
|
|
"num_tokens": 7408053.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.254216575622559,
|
|
"epoch": 3.4853801169590644,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000427747377194662,
|
|
"loss": 6.1399,
|
|
"mean_token_accuracy": 0.1587100461125374,
|
|
"num_tokens": 7419947.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.254539871215821,
|
|
"epoch": 3.4912280701754383,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004274028735540789,
|
|
"loss": 6.0343,
|
|
"mean_token_accuracy": 0.16398094445466996,
|
|
"num_tokens": 7433022.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.183673524856568,
|
|
"epoch": 3.497076023391813,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042705770847409596,
|
|
"loss": 6.0449,
|
|
"mean_token_accuracy": 0.17639647275209427,
|
|
"num_tokens": 7445280.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.2744606018066404,
|
|
"epoch": 3.502923976608187,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042671188345275423,
|
|
"loss": 6.0356,
|
|
"mean_token_accuracy": 0.16954765766859053,
|
|
"num_tokens": 7457324.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.177584266662597,
|
|
"epoch": 3.5087719298245617,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004263653999909593,
|
|
"loss": 6.0762,
|
|
"mean_token_accuracy": 0.16490701884031295,
|
|
"num_tokens": 7469070.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.5087719298245617,
|
|
"eval_entropy": 6.1530089884732675,
|
|
"eval_loss": 6.703362464904785,
|
|
"eval_mean_token_accuracy": 0.14372639493758524,
|
|
"eval_num_tokens": 7469070.0,
|
|
"eval_runtime": 2.2315,
|
|
"eval_samples_per_second": 1213.066,
|
|
"eval_steps_per_second": 151.913,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.2658124446868895,
|
|
"epoch": 3.5146198830409356,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000426018259592474,
|
|
"loss": 6.1492,
|
|
"mean_token_accuracy": 0.16011916399002074,
|
|
"num_tokens": 7480988.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.2107994079589846,
|
|
"epoch": 3.52046783625731,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004256704637639126,
|
|
"loss": 6.0581,
|
|
"mean_token_accuracy": 0.17161080688238145,
|
|
"num_tokens": 7493557.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.285580587387085,
|
|
"epoch": 3.526315789473684,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004253220140147341,
|
|
"loss": 6.1109,
|
|
"mean_token_accuracy": 0.16153184920549393,
|
|
"num_tokens": 7507446.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.202236461639404,
|
|
"epoch": 3.5321637426900585,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004249729118572352,
|
|
"loss": 6.0271,
|
|
"mean_token_accuracy": 0.16344058215618135,
|
|
"num_tokens": 7520054.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.16536774635315,
|
|
"epoch": 3.538011695906433,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004246231588065444,
|
|
"loss": 5.9461,
|
|
"mean_token_accuracy": 0.17561668008565903,
|
|
"num_tokens": 7533141.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.217787599563598,
|
|
"epoch": 3.543859649122807,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000424272756380615,
|
|
"loss": 6.1198,
|
|
"mean_token_accuracy": 0.16138767451047897,
|
|
"num_tokens": 7545544.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.226874303817749,
|
|
"epoch": 3.5497076023391814,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004239217061002188,
|
|
"loss": 6.1029,
|
|
"mean_token_accuracy": 0.1701584905385971,
|
|
"num_tokens": 7558055.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.196741962432862,
|
|
"epoch": 3.5555555555555554,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042357000948893894,
|
|
"loss": 6.0,
|
|
"mean_token_accuracy": 0.17685757279396058,
|
|
"num_tokens": 7569951.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.245180416107178,
|
|
"epoch": 3.56140350877193,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042321766807316405,
|
|
"loss": 6.1436,
|
|
"mean_token_accuracy": 0.1576438680291176,
|
|
"num_tokens": 7581810.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.308346462249756,
|
|
"epoch": 3.5672514619883042,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00042286468338208113,
|
|
"loss": 6.0905,
|
|
"mean_token_accuracy": 0.16553704142570497,
|
|
"num_tokens": 7594896.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.218596506118774,
|
|
"epoch": 3.573099415204678,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004225110569476691,
|
|
"loss": 6.1116,
|
|
"mean_token_accuracy": 0.16631555706262588,
|
|
"num_tokens": 7606620.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.17287974357605,
|
|
"epoch": 3.5789473684210527,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000422156790304692,
|
|
"loss": 5.9751,
|
|
"mean_token_accuracy": 0.17774234861135482,
|
|
"num_tokens": 7619493.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.143792819976807,
|
|
"epoch": 3.5847953216374266,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042180188499069246,
|
|
"loss": 5.9983,
|
|
"mean_token_accuracy": 0.16482216268777847,
|
|
"num_tokens": 7632711.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.242664766311646,
|
|
"epoch": 3.590643274853801,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000421446342545985,
|
|
"loss": 6.0608,
|
|
"mean_token_accuracy": 0.1681777149438858,
|
|
"num_tokens": 7645209.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.265076541900635,
|
|
"epoch": 3.5964912280701755,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042109016451364933,
|
|
"loss": 6.1559,
|
|
"mean_token_accuracy": 0.15856707096099854,
|
|
"num_tokens": 7657347.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.338294219970703,
|
|
"epoch": 3.60233918128655,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00042073335243952354,
|
|
"loss": 6.1071,
|
|
"mean_token_accuracy": 0.1555163413286209,
|
|
"num_tokens": 7669593.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.246266651153564,
|
|
"epoch": 3.608187134502924,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004203759078721979,
|
|
"loss": 6.1332,
|
|
"mean_token_accuracy": 0.15925390422344207,
|
|
"num_tokens": 7682548.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.222784805297851,
|
|
"epoch": 3.6140350877192984,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004200178323630072,
|
|
"loss": 6.1221,
|
|
"mean_token_accuracy": 0.1628882557153702,
|
|
"num_tokens": 7694554.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.283309412002564,
|
|
"epoch": 3.6198830409356724,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00041965912746602495,
|
|
"loss": 6.1623,
|
|
"mean_token_accuracy": 0.16299473643302917,
|
|
"num_tokens": 7707747.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.281281805038452,
|
|
"epoch": 3.625730994152047,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00041929979473805606,
|
|
"loss": 6.1332,
|
|
"mean_token_accuracy": 0.16667261868715286,
|
|
"num_tokens": 7719487.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.321485280990601,
|
|
"epoch": 3.6315789473684212,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041893983573863044,
|
|
"loss": 6.1539,
|
|
"mean_token_accuracy": 0.15928396433591843,
|
|
"num_tokens": 7732531.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.245786237716675,
|
|
"epoch": 3.6374269005847952,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00041857925202999586,
|
|
"loss": 6.0848,
|
|
"mean_token_accuracy": 0.17261341363191604,
|
|
"num_tokens": 7745034.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.224185276031494,
|
|
"epoch": 3.6432748538011697,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004182180451771115,
|
|
"loss": 6.0772,
|
|
"mean_token_accuracy": 0.16640402227640153,
|
|
"num_tokens": 7757442.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.238176536560059,
|
|
"epoch": 3.6491228070175437,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004178562167476409,
|
|
"loss": 6.1065,
|
|
"mean_token_accuracy": 0.16513924151659012,
|
|
"num_tokens": 7770358.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.223772621154785,
|
|
"epoch": 3.654970760233918,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00041749376831194565,
|
|
"loss": 6.0923,
|
|
"mean_token_accuracy": 0.16420244574546813,
|
|
"num_tokens": 7783611.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.174825382232666,
|
|
"epoch": 3.6608187134502925,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004171307014430777,
|
|
"loss": 6.0761,
|
|
"mean_token_accuracy": 0.16216820627450942,
|
|
"num_tokens": 7795522.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.278484153747558,
|
|
"epoch": 3.6666666666666665,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041676701771677343,
|
|
"loss": 6.0544,
|
|
"mean_token_accuracy": 0.17019673436880112,
|
|
"num_tokens": 7806801.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.257765865325927,
|
|
"epoch": 3.672514619883041,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004164027187114463,
|
|
"loss": 6.1552,
|
|
"mean_token_accuracy": 0.15927564352750778,
|
|
"num_tokens": 7821084.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.083193159103393,
|
|
"epoch": 3.678362573099415,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00041603780600818,
|
|
"loss": 5.8952,
|
|
"mean_token_accuracy": 0.18617731779813768,
|
|
"num_tokens": 7834065.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.209804201126099,
|
|
"epoch": 3.6842105263157894,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004156722811907219,
|
|
"loss": 6.0803,
|
|
"mean_token_accuracy": 0.17210820615291594,
|
|
"num_tokens": 7846378.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.183036470413208,
|
|
"epoch": 3.690058479532164,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041530614584547594,
|
|
"loss": 6.0977,
|
|
"mean_token_accuracy": 0.16261056661605836,
|
|
"num_tokens": 7858933.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.147948789596557,
|
|
"epoch": 3.6959064327485383,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00041493940156149577,
|
|
"loss": 5.9269,
|
|
"mean_token_accuracy": 0.17351364493370056,
|
|
"num_tokens": 7870408.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.293083620071411,
|
|
"epoch": 3.7017543859649122,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000414572049930478,
|
|
"loss": 6.112,
|
|
"mean_token_accuracy": 0.15937800854444503,
|
|
"num_tokens": 7882778.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.181996917724609,
|
|
"epoch": 3.7076023391812867,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00041420409254675465,
|
|
"loss": 6.0837,
|
|
"mean_token_accuracy": 0.16671226173639297,
|
|
"num_tokens": 7895169.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.217489767074585,
|
|
"epoch": 3.7134502923976607,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004138355310072876,
|
|
"loss": 6.1388,
|
|
"mean_token_accuracy": 0.16330468505620957,
|
|
"num_tokens": 7908607.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.270822429656983,
|
|
"epoch": 3.719298245614035,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00041346636691166015,
|
|
"loss": 6.0853,
|
|
"mean_token_accuracy": 0.16712178736925126,
|
|
"num_tokens": 7922269.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.1572980880737305,
|
|
"epoch": 3.7251461988304095,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00041309660186207093,
|
|
"loss": 6.0437,
|
|
"mean_token_accuracy": 0.17072458267211915,
|
|
"num_tokens": 7935205.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.230362844467163,
|
|
"epoch": 3.7309941520467835,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041272623746332687,
|
|
"loss": 6.0787,
|
|
"mean_token_accuracy": 0.16596648693084717,
|
|
"num_tokens": 7947970.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.037816143035888,
|
|
"epoch": 3.736842105263158,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041235527532283603,
|
|
"loss": 5.9341,
|
|
"mean_token_accuracy": 0.17992064952850342,
|
|
"num_tokens": 7961978.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.23446102142334,
|
|
"epoch": 3.742690058479532,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004119837170506005,
|
|
"loss": 6.1171,
|
|
"mean_token_accuracy": 0.16513372361660003,
|
|
"num_tokens": 7974708.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.2359343528747555,
|
|
"epoch": 3.7485380116959064,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041161156425921004,
|
|
"loss": 6.0806,
|
|
"mean_token_accuracy": 0.1685657650232315,
|
|
"num_tokens": 7986179.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.206500911712647,
|
|
"epoch": 3.754385964912281,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041123881856383436,
|
|
"loss": 6.1096,
|
|
"mean_token_accuracy": 0.1667914718389511,
|
|
"num_tokens": 7997816.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.222242307662964,
|
|
"epoch": 3.760233918128655,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004108654815822165,
|
|
"loss": 6.0382,
|
|
"mean_token_accuracy": 0.16846277415752411,
|
|
"num_tokens": 8009407.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.176695108413696,
|
|
"epoch": 3.7660818713450293,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004104915549346658,
|
|
"loss": 6.1161,
|
|
"mean_token_accuracy": 0.15902695059776306,
|
|
"num_tokens": 8021446.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.223764801025391,
|
|
"epoch": 3.7719298245614032,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041011704024405065,
|
|
"loss": 6.1519,
|
|
"mean_token_accuracy": 0.16430502235889435,
|
|
"num_tokens": 8035219.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.246078062057495,
|
|
"epoch": 3.7777777777777777,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040974193913579174,
|
|
"loss": 5.9445,
|
|
"mean_token_accuracy": 0.17976032495498656,
|
|
"num_tokens": 8047619.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.244170236587524,
|
|
"epoch": 3.783625730994152,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004093662532378547,
|
|
"loss": 6.143,
|
|
"mean_token_accuracy": 0.1656273901462555,
|
|
"num_tokens": 8059241.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.192588376998901,
|
|
"epoch": 3.7894736842105265,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004089899841807433,
|
|
"loss": 6.0797,
|
|
"mean_token_accuracy": 0.16690103113651275,
|
|
"num_tokens": 8071543.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.293214750289917,
|
|
"epoch": 3.7953216374269005,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00040861313359749217,
|
|
"loss": 6.1125,
|
|
"mean_token_accuracy": 0.16391814947128297,
|
|
"num_tokens": 8083515.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.105064439773559,
|
|
"epoch": 3.801169590643275,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00040823570312366005,
|
|
"loss": 6.0615,
|
|
"mean_token_accuracy": 0.16978096067905427,
|
|
"num_tokens": 8095703.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.282377290725708,
|
|
"epoch": 3.807017543859649,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000407857694397322,
|
|
"loss": 6.1816,
|
|
"mean_token_accuracy": 0.15659962445497513,
|
|
"num_tokens": 8108285.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.209602117538452,
|
|
"epoch": 3.8128654970760234,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004074791090590632,
|
|
"loss": 6.1095,
|
|
"mean_token_accuracy": 0.1648496612906456,
|
|
"num_tokens": 8120943.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.196795034408569,
|
|
"epoch": 3.818713450292398,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004070999487519711,
|
|
"loss": 6.1248,
|
|
"mean_token_accuracy": 0.161008557677269,
|
|
"num_tokens": 8133086.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.292535638809204,
|
|
"epoch": 3.824561403508772,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00040672021512162864,
|
|
"loss": 6.1419,
|
|
"mean_token_accuracy": 0.1632573276758194,
|
|
"num_tokens": 8147320.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.200564050674439,
|
|
"epoch": 3.8304093567251463,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000406339909816107,
|
|
"loss": 6.1158,
|
|
"mean_token_accuracy": 0.16135710775852202,
|
|
"num_tokens": 8160796.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.296334981918335,
|
|
"epoch": 3.8362573099415203,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004059590344859584,
|
|
"loss": 6.067,
|
|
"mean_token_accuracy": 0.16857662051916122,
|
|
"num_tokens": 8173087.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.142060327529907,
|
|
"epoch": 3.8421052631578947,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004055775907842092,
|
|
"loss": 6.032,
|
|
"mean_token_accuracy": 0.1658536523580551,
|
|
"num_tokens": 8185274.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.187477445602417,
|
|
"epoch": 3.847953216374269,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004051955803663523,
|
|
"loss": 6.0859,
|
|
"mean_token_accuracy": 0.17076018154621125,
|
|
"num_tokens": 8197753.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.182254028320313,
|
|
"epoch": 3.853801169590643,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004048130048903404,
|
|
"loss": 6.0706,
|
|
"mean_token_accuracy": 0.16768438667058944,
|
|
"num_tokens": 8211882.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.170766925811767,
|
|
"epoch": 3.8596491228070176,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004044298660165786,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.1804029017686844,
|
|
"num_tokens": 8225089.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.2010985851287845,
|
|
"epoch": 3.8654970760233915,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040404616540791683,
|
|
"loss": 6.1464,
|
|
"mean_token_accuracy": 0.1607299640774727,
|
|
"num_tokens": 8238808.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.15462851524353,
|
|
"epoch": 3.871345029239766,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004036619047296435,
|
|
"loss": 6.0634,
|
|
"mean_token_accuracy": 0.17139651924371718,
|
|
"num_tokens": 8252482.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.206887435913086,
|
|
"epoch": 3.8771929824561404,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004032770856494772,
|
|
"loss": 6.1119,
|
|
"mean_token_accuracy": 0.1642671838402748,
|
|
"num_tokens": 8264232.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.2487818717956545,
|
|
"epoch": 3.883040935672515,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040289170983756075,
|
|
"loss": 6.0903,
|
|
"mean_token_accuracy": 0.15881162732839585,
|
|
"num_tokens": 8276784.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.198899459838867,
|
|
"epoch": 3.888888888888889,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004025057789664524,
|
|
"loss": 6.0158,
|
|
"mean_token_accuracy": 0.16885033398866653,
|
|
"num_tokens": 8287887.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.086851978302002,
|
|
"epoch": 3.8947368421052633,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004021192947111201,
|
|
"loss": 5.8677,
|
|
"mean_token_accuracy": 0.18175480514764786,
|
|
"num_tokens": 8301258.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.184494256973267,
|
|
"epoch": 3.9005847953216373,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00040173225874893305,
|
|
"loss": 6.0902,
|
|
"mean_token_accuracy": 0.16973123103380203,
|
|
"num_tokens": 8313520.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.333401203155518,
|
|
"epoch": 3.9064327485380117,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040134467275965515,
|
|
"loss": 6.1393,
|
|
"mean_token_accuracy": 0.1629217892885208,
|
|
"num_tokens": 8325597.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.209193706512451,
|
|
"epoch": 3.912280701754386,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004009565384254374,
|
|
"loss": 6.1709,
|
|
"mean_token_accuracy": 0.16330498158931733,
|
|
"num_tokens": 8337449.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.22578992843628,
|
|
"epoch": 3.91812865497076,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040056785743081065,
|
|
"loss": 6.0856,
|
|
"mean_token_accuracy": 0.1670996814966202,
|
|
"num_tokens": 8348802.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.182878208160401,
|
|
"epoch": 3.9239766081871346,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004001786314626783,
|
|
"loss": 6.0733,
|
|
"mean_token_accuracy": 0.16238887161016463,
|
|
"num_tokens": 8360284.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.222674036026001,
|
|
"epoch": 3.9298245614035086,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003997888622103088,
|
|
"loss": 6.117,
|
|
"mean_token_accuracy": 0.16113592833280563,
|
|
"num_tokens": 8372080.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.247674417495728,
|
|
"epoch": 3.935672514619883,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003993985513653289,
|
|
"loss": 6.0655,
|
|
"mean_token_accuracy": 0.17076288014650345,
|
|
"num_tokens": 8385178.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.160210514068604,
|
|
"epoch": 3.9415204678362574,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003990077006217155,
|
|
"loss": 6.0378,
|
|
"mean_token_accuracy": 0.17264354377985,
|
|
"num_tokens": 8397735.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.226625967025757,
|
|
"epoch": 3.9473684210526314,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000398616311675789,
|
|
"loss": 6.1323,
|
|
"mean_token_accuracy": 0.1623571738600731,
|
|
"num_tokens": 8409916.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.2555866718292235,
|
|
"epoch": 3.953216374269006,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039822438622620523,
|
|
"loss": 6.1251,
|
|
"mean_token_accuracy": 0.16215181350708008,
|
|
"num_tokens": 8421883.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.142125844955444,
|
|
"epoch": 3.95906432748538,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039783192597394896,
|
|
"loss": 6.0027,
|
|
"mean_token_accuracy": 0.177200785279274,
|
|
"num_tokens": 8433610.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.129267454147339,
|
|
"epoch": 3.9649122807017543,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039743893262232584,
|
|
"loss": 6.0688,
|
|
"mean_token_accuracy": 0.16706801652908326,
|
|
"num_tokens": 8446194.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.246992301940918,
|
|
"epoch": 3.9707602339181287,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003970454078769551,
|
|
"loss": 6.0396,
|
|
"mean_token_accuracy": 0.17308892756700517,
|
|
"num_tokens": 8458846.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.203073358535766,
|
|
"epoch": 3.976608187134503,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003966513534457623,
|
|
"loss": 6.0539,
|
|
"mean_token_accuracy": 0.16625403314828874,
|
|
"num_tokens": 8470087.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.155765724182129,
|
|
"epoch": 3.982456140350877,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003962567710389721,
|
|
"loss": 6.0676,
|
|
"mean_token_accuracy": 0.16421020179986953,
|
|
"num_tokens": 8481293.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.286849880218506,
|
|
"epoch": 3.9883040935672516,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039586166236910033,
|
|
"loss": 6.1062,
|
|
"mean_token_accuracy": 0.1757156565785408,
|
|
"num_tokens": 8493376.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.084551620483398,
|
|
"epoch": 3.9941520467836256,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003954660291509469,
|
|
"loss": 5.9686,
|
|
"mean_token_accuracy": 0.16982612162828445,
|
|
"num_tokens": 8506132.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.217475366592407,
|
|
"epoch": 4.0,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039506987310158854,
|
|
"loss": 6.0605,
|
|
"mean_token_accuracy": 0.16575829982757567,
|
|
"num_tokens": 8517936.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.272924327850342,
|
|
"epoch": 4.005847953216374,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039467319594037076,
|
|
"loss": 5.7398,
|
|
"mean_token_accuracy": 0.17025531828403473,
|
|
"num_tokens": 8529863.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.1036382675170895,
|
|
"epoch": 4.011695906432749,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003942759993889009,
|
|
"loss": 5.7354,
|
|
"mean_token_accuracy": 0.17476040571928025,
|
|
"num_tokens": 8543621.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.13237361907959,
|
|
"epoch": 4.017543859649122,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039387828517104044,
|
|
"loss": 5.7042,
|
|
"mean_token_accuracy": 0.18571721911430358,
|
|
"num_tokens": 8557865.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.102181529998779,
|
|
"epoch": 4.023391812865497,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00039348005501289775,
|
|
"loss": 5.6357,
|
|
"mean_token_accuracy": 0.18461882770061494,
|
|
"num_tokens": 8570203.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.994093322753907,
|
|
"epoch": 4.029239766081871,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00039308131064282016,
|
|
"loss": 5.7319,
|
|
"mean_token_accuracy": 0.18554241359233856,
|
|
"num_tokens": 8582638.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.194820356369019,
|
|
"epoch": 4.035087719298246,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003926820537913868,
|
|
"loss": 5.8145,
|
|
"mean_token_accuracy": 0.17398348748683928,
|
|
"num_tokens": 8594973.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.109158754348755,
|
|
"epoch": 4.04093567251462,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003922822861914012,
|
|
"loss": 5.6624,
|
|
"mean_token_accuracy": 0.1860668420791626,
|
|
"num_tokens": 8607216.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.18676061630249,
|
|
"epoch": 4.046783625730994,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003918820095778832,
|
|
"loss": 5.8974,
|
|
"mean_token_accuracy": 0.16871323734521865,
|
|
"num_tokens": 8620833.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.125546407699585,
|
|
"epoch": 4.052631578947368,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00039148122568806215,
|
|
"loss": 5.7071,
|
|
"mean_token_accuracy": 0.18810465186834335,
|
|
"num_tokens": 8632931.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.061151218414307,
|
|
"epoch": 4.058479532163743,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003910799362613688,
|
|
"loss": 5.7823,
|
|
"mean_token_accuracy": 0.18191691637039184,
|
|
"num_tokens": 8644226.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.066484832763672,
|
|
"epoch": 4.064327485380117,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039067814303942806,
|
|
"loss": 5.6115,
|
|
"mean_token_accuracy": 0.1964157611131668,
|
|
"num_tokens": 8656569.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.951559829711914,
|
|
"epoch": 4.0701754385964914,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039027584776605136,
|
|
"loss": 5.6236,
|
|
"mean_token_accuracy": 0.18741821348667145,
|
|
"num_tokens": 8668077.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.1169427871704105,
|
|
"epoch": 4.076023391812866,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038987305218722904,
|
|
"loss": 5.7074,
|
|
"mean_token_accuracy": 0.17830572724342347,
|
|
"num_tokens": 8680227.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.156210470199585,
|
|
"epoch": 4.081871345029239,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003894697580511226,
|
|
"loss": 5.7995,
|
|
"mean_token_accuracy": 0.17887898534536362,
|
|
"num_tokens": 8692354.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.105235862731933,
|
|
"epoch": 4.087719298245614,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003890659671080579,
|
|
"loss": 5.7475,
|
|
"mean_token_accuracy": 0.16720127165317536,
|
|
"num_tokens": 8704634.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.133224058151245,
|
|
"epoch": 4.093567251461988,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003886616811105162,
|
|
"loss": 5.8135,
|
|
"mean_token_accuracy": 0.17422476261854172,
|
|
"num_tokens": 8717253.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 4.093567251461988,
|
|
"eval_entropy": 6.099606734813139,
|
|
"eval_loss": 6.661609172821045,
|
|
"eval_mean_token_accuracy": 0.14532060572289612,
|
|
"eval_num_tokens": 8717253.0,
|
|
"eval_runtime": 2.3775,
|
|
"eval_samples_per_second": 1138.578,
|
|
"eval_steps_per_second": 142.585,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.194639253616333,
|
|
"epoch": 4.099415204678363,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038825690181312793,
|
|
"loss": 5.7978,
|
|
"mean_token_accuracy": 0.1735048472881317,
|
|
"num_tokens": 8729840.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.065072298049927,
|
|
"epoch": 4.105263157894737,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038785163097266424,
|
|
"loss": 5.646,
|
|
"mean_token_accuracy": 0.18707735538482667,
|
|
"num_tokens": 8742065.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.129241514205932,
|
|
"epoch": 4.111111111111111,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003874458703480296,
|
|
"loss": 5.7997,
|
|
"mean_token_accuracy": 0.1744283378124237,
|
|
"num_tokens": 8754113.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.172555828094483,
|
|
"epoch": 4.116959064327485,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003870396217002542,
|
|
"loss": 5.8571,
|
|
"mean_token_accuracy": 0.16350731998682022,
|
|
"num_tokens": 8767255.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.102135562896729,
|
|
"epoch": 4.12280701754386,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003866328867924864,
|
|
"loss": 5.767,
|
|
"mean_token_accuracy": 0.16934611648321152,
|
|
"num_tokens": 8781010.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.197445344924927,
|
|
"epoch": 4.128654970760234,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003862256673899848,
|
|
"loss": 5.8026,
|
|
"mean_token_accuracy": 0.17756327688694,
|
|
"num_tokens": 8794393.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.150579023361206,
|
|
"epoch": 4.1345029239766085,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003858179652601108,
|
|
"loss": 5.8739,
|
|
"mean_token_accuracy": 0.16433500349521638,
|
|
"num_tokens": 8806496.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.085662031173706,
|
|
"epoch": 4.140350877192983,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038540978217232086,
|
|
"loss": 5.6766,
|
|
"mean_token_accuracy": 0.18490347713232042,
|
|
"num_tokens": 8818411.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.140443229675293,
|
|
"epoch": 4.146198830409356,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038500111989815894,
|
|
"loss": 5.7768,
|
|
"mean_token_accuracy": 0.17577040791511536,
|
|
"num_tokens": 8829714.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.062317705154419,
|
|
"epoch": 4.152046783625731,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038459198021124853,
|
|
"loss": 5.7428,
|
|
"mean_token_accuracy": 0.17809166610240937,
|
|
"num_tokens": 8842393.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.157629346847534,
|
|
"epoch": 4.157894736842105,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003841823648872852,
|
|
"loss": 5.7707,
|
|
"mean_token_accuracy": 0.17337929904460908,
|
|
"num_tokens": 8855263.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.087841987609863,
|
|
"epoch": 4.16374269005848,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003837722757040289,
|
|
"loss": 5.81,
|
|
"mean_token_accuracy": 0.1735988065600395,
|
|
"num_tokens": 8868220.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.10977029800415,
|
|
"epoch": 4.169590643274854,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003833617144412961,
|
|
"loss": 5.7494,
|
|
"mean_token_accuracy": 0.17875284403562547,
|
|
"num_tokens": 8879106.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.091527700424194,
|
|
"epoch": 4.175438596491228,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003829506828809521,
|
|
"loss": 5.835,
|
|
"mean_token_accuracy": 0.17815349847078324,
|
|
"num_tokens": 8891358.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.174573850631714,
|
|
"epoch": 4.181286549707602,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003825391828069032,
|
|
"loss": 5.7661,
|
|
"mean_token_accuracy": 0.17474008351564407,
|
|
"num_tokens": 8904327.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.047969579696655,
|
|
"epoch": 4.187134502923977,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003821272160050894,
|
|
"loss": 5.7317,
|
|
"mean_token_accuracy": 0.1748550921678543,
|
|
"num_tokens": 8916515.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.079746580123901,
|
|
"epoch": 4.192982456140351,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003817147842634762,
|
|
"loss": 5.8022,
|
|
"mean_token_accuracy": 0.17374901622533798,
|
|
"num_tokens": 8929482.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.18233323097229,
|
|
"epoch": 4.1988304093567255,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003813018893720468,
|
|
"loss": 5.854,
|
|
"mean_token_accuracy": 0.16606095135211946,
|
|
"num_tokens": 8942328.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.118831539154053,
|
|
"epoch": 4.204678362573099,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038088853312279473,
|
|
"loss": 5.7568,
|
|
"mean_token_accuracy": 0.17973800450563432,
|
|
"num_tokens": 8955116.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.1235450267791744,
|
|
"epoch": 4.2105263157894735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003804747173097157,
|
|
"loss": 5.8433,
|
|
"mean_token_accuracy": 0.1699218586087227,
|
|
"num_tokens": 8967459.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.077490043640137,
|
|
"epoch": 4.216374269005848,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038006044372880006,
|
|
"loss": 5.7597,
|
|
"mean_token_accuracy": 0.17578674703836442,
|
|
"num_tokens": 8980079.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.127010774612427,
|
|
"epoch": 4.222222222222222,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00037964571417802473,
|
|
"loss": 5.7786,
|
|
"mean_token_accuracy": 0.18156342655420304,
|
|
"num_tokens": 8992764.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.018903303146362,
|
|
"epoch": 4.228070175438597,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037923053045734595,
|
|
"loss": 5.6769,
|
|
"mean_token_accuracy": 0.17480476647615434,
|
|
"num_tokens": 9004904.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.200743103027344,
|
|
"epoch": 4.23391812865497,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00037881489436869054,
|
|
"loss": 5.8967,
|
|
"mean_token_accuracy": 0.16409194171428682,
|
|
"num_tokens": 9018411.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.100819253921509,
|
|
"epoch": 4.239766081871345,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003783988077159491,
|
|
"loss": 5.7252,
|
|
"mean_token_accuracy": 0.1818761184811592,
|
|
"num_tokens": 9030593.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.980781221389771,
|
|
"epoch": 4.245614035087719,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037798227230496754,
|
|
"loss": 5.7165,
|
|
"mean_token_accuracy": 0.18727464228868484,
|
|
"num_tokens": 9043715.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.015384769439697,
|
|
"epoch": 4.251461988304094,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00037756528994353945,
|
|
"loss": 5.6811,
|
|
"mean_token_accuracy": 0.18552859425544738,
|
|
"num_tokens": 9056384.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.103887605667114,
|
|
"epoch": 4.257309941520468,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000377147862441398,
|
|
"loss": 5.7871,
|
|
"mean_token_accuracy": 0.173315067589283,
|
|
"num_tokens": 9069444.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.084293794631958,
|
|
"epoch": 4.2631578947368425,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037672999161020873,
|
|
"loss": 5.7181,
|
|
"mean_token_accuracy": 0.18134546428918838,
|
|
"num_tokens": 9080933.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.082036828994751,
|
|
"epoch": 4.269005847953216,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00037631167926356087,
|
|
"loss": 5.7866,
|
|
"mean_token_accuracy": 0.17799059450626373,
|
|
"num_tokens": 9093405.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.048983716964722,
|
|
"epoch": 4.2748538011695905,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003758929272169599,
|
|
"loss": 5.7674,
|
|
"mean_token_accuracy": 0.18191103935241698,
|
|
"num_tokens": 9105707.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.160745286941529,
|
|
"epoch": 4.280701754385965,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00037547373728781996,
|
|
"loss": 5.7243,
|
|
"mean_token_accuracy": 0.18051233142614365,
|
|
"num_tokens": 9119476.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.0592817783355715,
|
|
"epoch": 4.286549707602339,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003750541112954552,
|
|
"loss": 5.6946,
|
|
"mean_token_accuracy": 0.18434519171714783,
|
|
"num_tokens": 9131788.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.096468114852906,
|
|
"epoch": 4.292397660818714,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003746340510610727,
|
|
"loss": 5.7914,
|
|
"mean_token_accuracy": 0.1767459839582443,
|
|
"num_tokens": 9144340.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.123143482208252,
|
|
"epoch": 4.298245614035087,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00037421355840776396,
|
|
"loss": 5.7641,
|
|
"mean_token_accuracy": 0.18209490031003953,
|
|
"num_tokens": 9156757.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.056727933883667,
|
|
"epoch": 4.304093567251462,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003737926351604971,
|
|
"loss": 5.7473,
|
|
"mean_token_accuracy": 0.17966152280569075,
|
|
"num_tokens": 9169975.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.129817771911621,
|
|
"epoch": 4.309941520467836,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037337128314610926,
|
|
"loss": 5.7639,
|
|
"mean_token_accuracy": 0.1814514145255089,
|
|
"num_tokens": 9181891.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.094612646102905,
|
|
"epoch": 4.315789473684211,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003729495041932984,
|
|
"loss": 5.8077,
|
|
"mean_token_accuracy": 0.1722259297966957,
|
|
"num_tokens": 9194068.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.126791858673096,
|
|
"epoch": 4.321637426900585,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003725273001326155,
|
|
"loss": 5.8169,
|
|
"mean_token_accuracy": 0.17457632720470428,
|
|
"num_tokens": 9205530.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.050809717178344,
|
|
"epoch": 4.3274853801169595,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003721046727964565,
|
|
"loss": 5.7272,
|
|
"mean_token_accuracy": 0.18317656368017196,
|
|
"num_tokens": 9218444.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 6.083014488220215,
|
|
"epoch": 4.333333333333333,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037168162401905416,
|
|
"loss": 5.8366,
|
|
"mean_token_accuracy": 0.18090108186006545,
|
|
"num_tokens": 9229507.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.161681842803955,
|
|
"epoch": 4.3391812865497075,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037125815563647063,
|
|
"loss": 5.8715,
|
|
"mean_token_accuracy": 0.1719798892736435,
|
|
"num_tokens": 9242128.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.0982666015625,
|
|
"epoch": 4.345029239766082,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00037083426948658903,
|
|
"loss": 5.8322,
|
|
"mean_token_accuracy": 0.17149512469768524,
|
|
"num_tokens": 9255598.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.107556438446045,
|
|
"epoch": 4.350877192982456,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037040996740910566,
|
|
"loss": 5.7734,
|
|
"mean_token_accuracy": 0.1799965977668762,
|
|
"num_tokens": 9267568.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.06874532699585,
|
|
"epoch": 4.356725146198831,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000369985251245522,
|
|
"loss": 5.8097,
|
|
"mean_token_accuracy": 0.1752200096845627,
|
|
"num_tokens": 9280223.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.120317602157593,
|
|
"epoch": 4.362573099415204,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00036956012283913655,
|
|
"loss": 5.8678,
|
|
"mean_token_accuracy": 0.17524777054786683,
|
|
"num_tokens": 9292933.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.08168215751648,
|
|
"epoch": 4.368421052631579,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00036913458403503715,
|
|
"loss": 5.811,
|
|
"mean_token_accuracy": 0.18154270499944686,
|
|
"num_tokens": 9306739.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.0938475131988525,
|
|
"epoch": 4.374269005847953,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003687086366800927,
|
|
"loss": 5.7634,
|
|
"mean_token_accuracy": 0.17334699034690856,
|
|
"num_tokens": 9319151.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.118067646026612,
|
|
"epoch": 4.380116959064328,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003682822826229453,
|
|
"loss": 5.8229,
|
|
"mean_token_accuracy": 0.17274134159088134,
|
|
"num_tokens": 9332035.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.122704982757568,
|
|
"epoch": 4.385964912280702,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036785552371400206,
|
|
"loss": 5.8743,
|
|
"mean_token_accuracy": 0.1715137928724289,
|
|
"num_tokens": 9345248.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.135756349563598,
|
|
"epoch": 4.391812865497076,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003674283618054273,
|
|
"loss": 5.822,
|
|
"mean_token_accuracy": 0.17474781423807145,
|
|
"num_tokens": 9357514.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.076525497436523,
|
|
"epoch": 4.39766081871345,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003670007987511344,
|
|
"loss": 5.8425,
|
|
"mean_token_accuracy": 0.16845586895942688,
|
|
"num_tokens": 9368627.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.0563342571258545,
|
|
"epoch": 4.4035087719298245,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003665728364067776,
|
|
"loss": 5.7594,
|
|
"mean_token_accuracy": 0.17922329902648926,
|
|
"num_tokens": 9380591.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.067640352249145,
|
|
"epoch": 4.409356725146199,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00036614447662974413,
|
|
"loss": 5.7557,
|
|
"mean_token_accuracy": 0.17874309718608855,
|
|
"num_tokens": 9392873.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.169782733917236,
|
|
"epoch": 4.415204678362573,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003657157212791462,
|
|
"loss": 5.8895,
|
|
"mean_token_accuracy": 0.1673583745956421,
|
|
"num_tokens": 9404252.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.059755420684814,
|
|
"epoch": 4.421052631578947,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003652865722158128,
|
|
"loss": 5.8135,
|
|
"mean_token_accuracy": 0.17929302603006364,
|
|
"num_tokens": 9416091.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.042658948898316,
|
|
"epoch": 4.426900584795321,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00036485703130228147,
|
|
"loss": 5.7612,
|
|
"mean_token_accuracy": 0.18140476197004318,
|
|
"num_tokens": 9429913.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.117117977142334,
|
|
"epoch": 4.432748538011696,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00036442710040279086,
|
|
"loss": 5.7385,
|
|
"mean_token_accuracy": 0.18508173376321793,
|
|
"num_tokens": 9442413.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.9100892543792725,
|
|
"epoch": 4.43859649122807,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003639967813832717,
|
|
"loss": 5.6658,
|
|
"mean_token_accuracy": 0.1868099421262741,
|
|
"num_tokens": 9453636.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.033320093154908,
|
|
"epoch": 4.444444444444445,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003635660761113394,
|
|
"loss": 5.7492,
|
|
"mean_token_accuracy": 0.1792163297533989,
|
|
"num_tokens": 9466577.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.062642621994018,
|
|
"epoch": 4.450292397660819,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003631349864562857,
|
|
"loss": 5.7619,
|
|
"mean_token_accuracy": 0.1764058515429497,
|
|
"num_tokens": 9479852.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.051758241653443,
|
|
"epoch": 4.456140350877193,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003627035142890708,
|
|
"loss": 5.7015,
|
|
"mean_token_accuracy": 0.18134466409683228,
|
|
"num_tokens": 9493373.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.073123598098755,
|
|
"epoch": 4.461988304093567,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003622716614823147,
|
|
"loss": 5.8023,
|
|
"mean_token_accuracy": 0.18308435380458832,
|
|
"num_tokens": 9505793.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.072046566009521,
|
|
"epoch": 4.4678362573099415,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003618394299102896,
|
|
"loss": 5.8353,
|
|
"mean_token_accuracy": 0.16938746869564056,
|
|
"num_tokens": 9517810.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.0684850215911865,
|
|
"epoch": 4.473684210526316,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036140682144891143,
|
|
"loss": 5.8066,
|
|
"mean_token_accuracy": 0.17093569040298462,
|
|
"num_tokens": 9529430.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.096002912521362,
|
|
"epoch": 4.47953216374269,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00036097383797573205,
|
|
"loss": 5.7934,
|
|
"mean_token_accuracy": 0.1816865473985672,
|
|
"num_tokens": 9542385.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.072030639648437,
|
|
"epoch": 4.485380116959064,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00036054048136993075,
|
|
"loss": 5.7689,
|
|
"mean_token_accuracy": 0.17942011803388597,
|
|
"num_tokens": 9553955.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.169800233840943,
|
|
"epoch": 4.491228070175438,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036010675351230616,
|
|
"loss": 5.8207,
|
|
"mean_token_accuracy": 0.17117204666137695,
|
|
"num_tokens": 9566494.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.982374906539917,
|
|
"epoch": 4.497076023391813,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003596726562852684,
|
|
"loss": 5.7373,
|
|
"mean_token_accuracy": 0.1911873549222946,
|
|
"num_tokens": 9579122.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 6.043915700912476,
|
|
"epoch": 4.502923976608187,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00035923819157283074,
|
|
"loss": 5.7627,
|
|
"mean_token_accuracy": 0.17686564475297928,
|
|
"num_tokens": 9592144.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.9978265285491945,
|
|
"epoch": 4.508771929824562,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035880336126060074,
|
|
"loss": 5.6672,
|
|
"mean_token_accuracy": 0.180244180560112,
|
|
"num_tokens": 9604533.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.066030693054199,
|
|
"epoch": 4.514619883040936,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00035836816723577354,
|
|
"loss": 5.8434,
|
|
"mean_token_accuracy": 0.17792066484689711,
|
|
"num_tokens": 9617349.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.1015894412994385,
|
|
"epoch": 4.52046783625731,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003579326113871222,
|
|
"loss": 5.7963,
|
|
"mean_token_accuracy": 0.17852200269699098,
|
|
"num_tokens": 9629599.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.0220743179321286,
|
|
"epoch": 4.526315789473684,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003574966956049904,
|
|
"loss": 5.7865,
|
|
"mean_token_accuracy": 0.17893554866313935,
|
|
"num_tokens": 9642041.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.061077451705932,
|
|
"epoch": 4.5321637426900585,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035706042178128397,
|
|
"loss": 5.792,
|
|
"mean_token_accuracy": 0.17634476870298385,
|
|
"num_tokens": 9654796.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.061177396774292,
|
|
"epoch": 4.538011695906433,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003566237918094625,
|
|
"loss": 5.8068,
|
|
"mean_token_accuracy": 0.18041525930166244,
|
|
"num_tokens": 9667211.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.02913646697998,
|
|
"epoch": 4.543859649122807,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003561868075845313,
|
|
"loss": 5.7583,
|
|
"mean_token_accuracy": 0.180086849629879,
|
|
"num_tokens": 9679526.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.147022294998169,
|
|
"epoch": 4.549707602339181,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035574947100303327,
|
|
"loss": 5.9238,
|
|
"mean_token_accuracy": 0.16855545341968536,
|
|
"num_tokens": 9691760.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.093625688552857,
|
|
"epoch": 4.555555555555555,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00035531178396304083,
|
|
"loss": 5.8378,
|
|
"mean_token_accuracy": 0.17761591225862502,
|
|
"num_tokens": 9704154.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.941799020767212,
|
|
"epoch": 4.56140350877193,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035487374836414676,
|
|
"loss": 5.6454,
|
|
"mean_token_accuracy": 0.18655768036842346,
|
|
"num_tokens": 9715850.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.113169145584107,
|
|
"epoch": 4.567251461988304,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00035443536610745707,
|
|
"loss": 5.8149,
|
|
"mean_token_accuracy": 0.17653881311416625,
|
|
"num_tokens": 9728202.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.1170214176177975,
|
|
"epoch": 4.573099415204679,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00035399663909558226,
|
|
"loss": 5.8144,
|
|
"mean_token_accuracy": 0.16914215087890624,
|
|
"num_tokens": 9741363.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.045348882675171,
|
|
"epoch": 4.578947368421053,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00035355756923262897,
|
|
"loss": 5.7934,
|
|
"mean_token_accuracy": 0.17347656488418578,
|
|
"num_tokens": 9755231.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.030009984970093,
|
|
"epoch": 4.584795321637427,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003531181584241919,
|
|
"loss": 5.6227,
|
|
"mean_token_accuracy": 0.19290953129529953,
|
|
"num_tokens": 9768972.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 6.101548862457276,
|
|
"epoch": 4.590643274853801,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003526784085773455,
|
|
"loss": 5.8448,
|
|
"mean_token_accuracy": 0.17159931510686874,
|
|
"num_tokens": 9781356.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.022581481933594,
|
|
"epoch": 4.5964912280701755,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003522383216006356,
|
|
"loss": 5.7995,
|
|
"mean_token_accuracy": 0.17175742238759995,
|
|
"num_tokens": 9793227.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.096387624740601,
|
|
"epoch": 4.60233918128655,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003517978994040713,
|
|
"loss": 5.7919,
|
|
"mean_token_accuracy": 0.1758534535765648,
|
|
"num_tokens": 9805968.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.078047847747802,
|
|
"epoch": 4.6081871345029235,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035135714389911637,
|
|
"loss": 5.7911,
|
|
"mean_token_accuracy": 0.1754264324903488,
|
|
"num_tokens": 9817864.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.047876167297363,
|
|
"epoch": 4.614035087719298,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003509160569986814,
|
|
"loss": 5.7742,
|
|
"mean_token_accuracy": 0.18175394535064698,
|
|
"num_tokens": 9831192.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 6.108513069152832,
|
|
"epoch": 4.619883040935672,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00035047464061711527,
|
|
"loss": 5.774,
|
|
"mean_token_accuracy": 0.17558089047670364,
|
|
"num_tokens": 9842813.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.088702011108398,
|
|
"epoch": 4.625730994152047,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00035003289667019655,
|
|
"loss": 5.8142,
|
|
"mean_token_accuracy": 0.1783566415309906,
|
|
"num_tokens": 9854500.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.0222697257995605,
|
|
"epoch": 4.631578947368421,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003495908270751257,
|
|
"loss": 5.7811,
|
|
"mean_token_accuracy": 0.17853581458330153,
|
|
"num_tokens": 9865729.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.165131139755249,
|
|
"epoch": 4.637426900584796,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003491484337505166,
|
|
"loss": 5.826,
|
|
"mean_token_accuracy": 0.17576233446598052,
|
|
"num_tokens": 9878306.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.027988481521606,
|
|
"epoch": 4.643274853801169,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00034870571861638785,
|
|
"loss": 5.8461,
|
|
"mean_token_accuracy": 0.17825262397527694,
|
|
"num_tokens": 9890384.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.0605988025665285,
|
|
"epoch": 4.649122807017544,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00034826268359415487,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.1767077624797821,
|
|
"num_tokens": 9902960.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.048948812484741,
|
|
"epoch": 4.654970760233918,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003478193306066215,
|
|
"loss": 5.7603,
|
|
"mean_token_accuracy": 0.18155244290828704,
|
|
"num_tokens": 9915143.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.101191139221191,
|
|
"epoch": 4.6608187134502925,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00034737566157797147,
|
|
"loss": 5.8763,
|
|
"mean_token_accuracy": 0.17006099820137024,
|
|
"num_tokens": 9928440.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.116497802734375,
|
|
"epoch": 4.666666666666667,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00034693167843376016,
|
|
"loss": 5.8805,
|
|
"mean_token_accuracy": 0.17083102613687515,
|
|
"num_tokens": 9940973.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.068368673324585,
|
|
"epoch": 4.6725146198830405,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00034648738310090634,
|
|
"loss": 5.7617,
|
|
"mean_token_accuracy": 0.18380540311336518,
|
|
"num_tokens": 9953433.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.956298542022705,
|
|
"epoch": 4.678362573099415,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003460427775076835,
|
|
"loss": 5.698,
|
|
"mean_token_accuracy": 0.18590198904275895,
|
|
"num_tokens": 9964641.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.678362573099415,
|
|
"eval_entropy": 5.94115327452488,
|
|
"eval_loss": 6.626513957977295,
|
|
"eval_mean_token_accuracy": 0.14774621837390559,
|
|
"eval_num_tokens": 9964641.0,
|
|
"eval_runtime": 2.3516,
|
|
"eval_samples_per_second": 1151.12,
|
|
"eval_steps_per_second": 144.156,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 6.093565511703491,
|
|
"epoch": 4.684210526315789,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00034559786358371187,
|
|
"loss": 5.8891,
|
|
"mean_token_accuracy": 0.16637808680534363,
|
|
"num_tokens": 9978625.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 6.1122472286224365,
|
|
"epoch": 4.690058479532164,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034515264325994975,
|
|
"loss": 5.7488,
|
|
"mean_token_accuracy": 0.1779404640197754,
|
|
"num_tokens": 9991758.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 6.094180727005005,
|
|
"epoch": 4.695906432748538,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00034470711846868525,
|
|
"loss": 5.8624,
|
|
"mean_token_accuracy": 0.17362448573112488,
|
|
"num_tokens": 10005496.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 6.0957451343536375,
|
|
"epoch": 4.701754385964913,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00034426129114352807,
|
|
"loss": 5.8823,
|
|
"mean_token_accuracy": 0.17267407178878785,
|
|
"num_tokens": 10019025.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.989615488052368,
|
|
"epoch": 4.707602339181286,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003438151632194006,
|
|
"loss": 5.7305,
|
|
"mean_token_accuracy": 0.18343682140111922,
|
|
"num_tokens": 10030741.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 6.119991254806519,
|
|
"epoch": 4.713450292397661,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034336873663252997,
|
|
"loss": 5.8482,
|
|
"mean_token_accuracy": 0.17282668352127076,
|
|
"num_tokens": 10043138.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 6.065843820571899,
|
|
"epoch": 4.719298245614035,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00034292201332043976,
|
|
"loss": 5.8057,
|
|
"mean_token_accuracy": 0.17359557300806044,
|
|
"num_tokens": 10055927.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 6.054896545410156,
|
|
"epoch": 4.7251461988304095,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003424749952219411,
|
|
"loss": 5.7196,
|
|
"mean_token_accuracy": 0.18297003954648972,
|
|
"num_tokens": 10067392.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 6.0323529720306395,
|
|
"epoch": 4.730994152046784,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00034202768427712456,
|
|
"loss": 5.7942,
|
|
"mean_token_accuracy": 0.1779397889971733,
|
|
"num_tokens": 10079245.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 6.024519205093384,
|
|
"epoch": 4.7368421052631575,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003415800824273518,
|
|
"loss": 5.6813,
|
|
"mean_token_accuracy": 0.18731204867362977,
|
|
"num_tokens": 10091923.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 6.060008335113525,
|
|
"epoch": 4.742690058479532,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003411321916152469,
|
|
"loss": 5.7616,
|
|
"mean_token_accuracy": 0.1817208930850029,
|
|
"num_tokens": 10105121.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 6.028836488723755,
|
|
"epoch": 4.748538011695906,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003406840137846881,
|
|
"loss": 5.8507,
|
|
"mean_token_accuracy": 0.1777327075600624,
|
|
"num_tokens": 10117005.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 6.001136636734008,
|
|
"epoch": 4.754385964912281,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00034023555088079924,
|
|
"loss": 5.7567,
|
|
"mean_token_accuracy": 0.1875651091337204,
|
|
"num_tokens": 10129958.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 6.056399726867676,
|
|
"epoch": 4.760233918128655,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003397868048499416,
|
|
"loss": 5.8034,
|
|
"mean_token_accuracy": 0.17545792907476426,
|
|
"num_tokens": 10141537.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 6.059792423248291,
|
|
"epoch": 4.76608187134503,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033933777763970506,
|
|
"loss": 5.8082,
|
|
"mean_token_accuracy": 0.18597352951765062,
|
|
"num_tokens": 10153882.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 6.082842302322388,
|
|
"epoch": 4.771929824561403,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003388884711989,
|
|
"loss": 5.8056,
|
|
"mean_token_accuracy": 0.17504796236753464,
|
|
"num_tokens": 10165147.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 6.077103996276856,
|
|
"epoch": 4.777777777777778,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003384388874775486,
|
|
"loss": 5.8369,
|
|
"mean_token_accuracy": 0.16800857782363893,
|
|
"num_tokens": 10177926.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 6.086846160888672,
|
|
"epoch": 4.783625730994152,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033798902842687627,
|
|
"loss": 5.7979,
|
|
"mean_token_accuracy": 0.17941107898950576,
|
|
"num_tokens": 10190343.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 6.10236349105835,
|
|
"epoch": 4.7894736842105265,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003375388959993037,
|
|
"loss": 5.883,
|
|
"mean_token_accuracy": 0.1706821620464325,
|
|
"num_tokens": 10201363.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.99293966293335,
|
|
"epoch": 4.7953216374269,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003370884921484379,
|
|
"loss": 5.7184,
|
|
"mean_token_accuracy": 0.18832051306962966,
|
|
"num_tokens": 10213976.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 6.023731803894043,
|
|
"epoch": 4.8011695906432745,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003366378188290638,
|
|
"loss": 5.77,
|
|
"mean_token_accuracy": 0.18166877925395966,
|
|
"num_tokens": 10225645.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 6.1180191993713375,
|
|
"epoch": 4.807017543859649,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003361868779971361,
|
|
"loss": 5.7937,
|
|
"mean_token_accuracy": 0.18563875555992126,
|
|
"num_tokens": 10237144.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 6.151350688934326,
|
|
"epoch": 4.812865497076023,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033573567160977017,
|
|
"loss": 5.9653,
|
|
"mean_token_accuracy": 0.16757765114307405,
|
|
"num_tokens": 10249215.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.973816823959351,
|
|
"epoch": 4.818713450292398,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000335284201625234,
|
|
"loss": 5.7712,
|
|
"mean_token_accuracy": 0.1773104801774025,
|
|
"num_tokens": 10262428.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.970614433288574,
|
|
"epoch": 4.824561403508772,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033483247000294,
|
|
"loss": 5.7863,
|
|
"mean_token_accuracy": 0.18311231434345246,
|
|
"num_tokens": 10275370.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 6.111313676834106,
|
|
"epoch": 4.830409356725146,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003343804787034357,
|
|
"loss": 5.8007,
|
|
"mean_token_accuracy": 0.18538742512464523,
|
|
"num_tokens": 10287378.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 6.094429540634155,
|
|
"epoch": 4.83625730994152,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003339282296883956,
|
|
"loss": 5.7993,
|
|
"mean_token_accuracy": 0.16928013414144516,
|
|
"num_tokens": 10300063.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 6.082474994659424,
|
|
"epoch": 4.842105263157895,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000333475724920613,
|
|
"loss": 5.7819,
|
|
"mean_token_accuracy": 0.18151704519987105,
|
|
"num_tokens": 10312095.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 6.037060165405274,
|
|
"epoch": 4.847953216374269,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003330229663639909,
|
|
"loss": 5.8279,
|
|
"mean_token_accuracy": 0.1785791337490082,
|
|
"num_tokens": 10324903.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 6.159916925430298,
|
|
"epoch": 4.853801169590644,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000332569955983534,
|
|
"loss": 5.8454,
|
|
"mean_token_accuracy": 0.17747473120689392,
|
|
"num_tokens": 10339253.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 6.033841753005982,
|
|
"epoch": 4.859649122807017,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003321166957453398,
|
|
"loss": 5.8507,
|
|
"mean_token_accuracy": 0.16852351725101472,
|
|
"num_tokens": 10350845.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 6.099295139312744,
|
|
"epoch": 4.8654970760233915,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003316631876165904,
|
|
"loss": 5.7719,
|
|
"mean_token_accuracy": 0.18086027055978776,
|
|
"num_tokens": 10363357.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 6.000420141220093,
|
|
"epoch": 4.871345029239766,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00033120943356554334,
|
|
"loss": 5.7115,
|
|
"mean_token_accuracy": 0.17826687246561052,
|
|
"num_tokens": 10375664.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 6.088650751113891,
|
|
"epoch": 4.87719298245614,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000330755435561524,
|
|
"loss": 5.7688,
|
|
"mean_token_accuracy": 0.17806624621152878,
|
|
"num_tokens": 10387412.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 6.016922330856323,
|
|
"epoch": 4.883040935672515,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033030119557491604,
|
|
"loss": 5.8184,
|
|
"mean_token_accuracy": 0.18123234808444977,
|
|
"num_tokens": 10399083.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 6.155450296401978,
|
|
"epoch": 4.888888888888889,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003298467155771537,
|
|
"loss": 5.8661,
|
|
"mean_token_accuracy": 0.168977852165699,
|
|
"num_tokens": 10412713.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 6.0134110927581785,
|
|
"epoch": 4.894736842105263,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00032939199754071274,
|
|
"loss": 5.718,
|
|
"mean_token_accuracy": 0.18433569222688675,
|
|
"num_tokens": 10426836.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.968796253204346,
|
|
"epoch": 4.900584795321637,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000328937043439102,
|
|
"loss": 5.8174,
|
|
"mean_token_accuracy": 0.17732980847358704,
|
|
"num_tokens": 10439817.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 6.094321298599243,
|
|
"epoch": 4.906432748538012,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00032848185524685504,
|
|
"loss": 5.7832,
|
|
"mean_token_accuracy": 0.17585162073373795,
|
|
"num_tokens": 10452509.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 6.105631875991821,
|
|
"epoch": 4.912280701754386,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032802643493952106,
|
|
"loss": 5.8393,
|
|
"mean_token_accuracy": 0.17505937516689302,
|
|
"num_tokens": 10464593.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 6.0437376499176025,
|
|
"epoch": 4.918128654970761,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000327570784493657,
|
|
"loss": 5.8472,
|
|
"mean_token_accuracy": 0.17379515022039413,
|
|
"num_tokens": 10476868.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 6.028283023834229,
|
|
"epoch": 4.923976608187134,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00032711490588681845,
|
|
"loss": 5.6782,
|
|
"mean_token_accuracy": 0.17952150106430054,
|
|
"num_tokens": 10489510.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 6.097183465957642,
|
|
"epoch": 4.9298245614035086,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003266588010975512,
|
|
"loss": 5.8453,
|
|
"mean_token_accuracy": 0.17056355327367784,
|
|
"num_tokens": 10501901.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 6.003234386444092,
|
|
"epoch": 4.935672514619883,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003262024721053827,
|
|
"loss": 5.755,
|
|
"mean_token_accuracy": 0.18276577889919282,
|
|
"num_tokens": 10513604.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 6.00689001083374,
|
|
"epoch": 4.941520467836257,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00032574592089081375,
|
|
"loss": 5.7735,
|
|
"mean_token_accuracy": 0.18512602597475053,
|
|
"num_tokens": 10527882.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 6.077584552764892,
|
|
"epoch": 4.947368421052632,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032528914943530905,
|
|
"loss": 5.7377,
|
|
"mean_token_accuracy": 0.1877928838133812,
|
|
"num_tokens": 10539514.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 6.044604396820068,
|
|
"epoch": 4.953216374269006,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003248321597212896,
|
|
"loss": 5.7826,
|
|
"mean_token_accuracy": 0.17872884571552278,
|
|
"num_tokens": 10551984.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 6.158230209350586,
|
|
"epoch": 4.95906432748538,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003243749537321236,
|
|
"loss": 5.8796,
|
|
"mean_token_accuracy": 0.1718206971883774,
|
|
"num_tokens": 10564915.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 6.024348974227905,
|
|
"epoch": 4.964912280701754,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032391753345211767,
|
|
"loss": 5.8411,
|
|
"mean_token_accuracy": 0.1818984031677246,
|
|
"num_tokens": 10576614.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 6.022074127197266,
|
|
"epoch": 4.970760233918129,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003234599008665088,
|
|
"loss": 5.7998,
|
|
"mean_token_accuracy": 0.18030415028333663,
|
|
"num_tokens": 10588746.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 6.033704853057861,
|
|
"epoch": 4.976608187134503,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003230020579614551,
|
|
"loss": 5.8075,
|
|
"mean_token_accuracy": 0.17157047092914582,
|
|
"num_tokens": 10600101.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 6.083580160140992,
|
|
"epoch": 4.982456140350877,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003225440067240276,
|
|
"loss": 5.7758,
|
|
"mean_token_accuracy": 0.17635996788740158,
|
|
"num_tokens": 10611879.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 6.073849010467529,
|
|
"epoch": 4.988304093567251,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032208574914220167,
|
|
"loss": 5.8558,
|
|
"mean_token_accuracy": 0.17456222027540208,
|
|
"num_tokens": 10624201.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 6.106013202667237,
|
|
"epoch": 4.994152046783626,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003216272872048479,
|
|
"loss": 5.7867,
|
|
"mean_token_accuracy": 0.17271582931280136,
|
|
"num_tokens": 10636612.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 6.033702611923218,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003211686229017241,
|
|
"loss": 5.7878,
|
|
"mean_token_accuracy": 0.18057709634304048,
|
|
"num_tokens": 10647420.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 6.027100229263306,
|
|
"epoch": 5.005847953216374,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003207097582234662,
|
|
"loss": 5.585,
|
|
"mean_token_accuracy": 0.1814449578523636,
|
|
"num_tokens": 10660096.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 6.034527444839478,
|
|
"epoch": 5.011695906432749,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003202506951615798,
|
|
"loss": 5.5165,
|
|
"mean_token_accuracy": 0.19347382336854935,
|
|
"num_tokens": 10672364.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 6.012624597549438,
|
|
"epoch": 5.017543859649122,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00031979143570843145,
|
|
"loss": 5.5127,
|
|
"mean_token_accuracy": 0.19000222086906432,
|
|
"num_tokens": 10685230.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 6.0932902812957765,
|
|
"epoch": 5.023391812865497,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00031933198185724016,
|
|
"loss": 5.6195,
|
|
"mean_token_accuracy": 0.18128057271242143,
|
|
"num_tokens": 10697837.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 6.037541055679322,
|
|
"epoch": 5.029239766081871,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00031887233560206864,
|
|
"loss": 5.4417,
|
|
"mean_token_accuracy": 0.1981600597500801,
|
|
"num_tokens": 10711549.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 6.102306175231933,
|
|
"epoch": 5.035087719298246,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00031841249893781447,
|
|
"loss": 5.609,
|
|
"mean_token_accuracy": 0.1839731901884079,
|
|
"num_tokens": 10724897.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 6.057007503509522,
|
|
"epoch": 5.04093567251462,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003179524738602019,
|
|
"loss": 5.5206,
|
|
"mean_token_accuracy": 0.18573260307312012,
|
|
"num_tokens": 10735808.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.988875961303711,
|
|
"epoch": 5.046783625730994,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00031749226236577277,
|
|
"loss": 5.472,
|
|
"mean_token_accuracy": 0.19190946370363235,
|
|
"num_tokens": 10749083.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 6.036866235733032,
|
|
"epoch": 5.052631578947368,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00031703186645187776,
|
|
"loss": 5.5058,
|
|
"mean_token_accuracy": 0.18597792387008666,
|
|
"num_tokens": 10761215.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.964029550552368,
|
|
"epoch": 5.058479532163743,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00031657128811666836,
|
|
"loss": 5.4225,
|
|
"mean_token_accuracy": 0.2076844811439514,
|
|
"num_tokens": 10772505.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 6.087826538085937,
|
|
"epoch": 5.064327485380117,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00031611052935908736,
|
|
"loss": 5.6356,
|
|
"mean_token_accuracy": 0.17894940078258514,
|
|
"num_tokens": 10785205.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 6.050455093383789,
|
|
"epoch": 5.0701754385964914,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000315649592178861,
|
|
"loss": 5.4564,
|
|
"mean_token_accuracy": 0.18949734270572663,
|
|
"num_tokens": 10796632.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 6.060620069503784,
|
|
"epoch": 5.076023391812866,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003151884785764897,
|
|
"loss": 5.5172,
|
|
"mean_token_accuracy": 0.18525703698396684,
|
|
"num_tokens": 10809466.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 6.013947534561157,
|
|
"epoch": 5.081871345029239,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003147271905532394,
|
|
"loss": 5.5835,
|
|
"mean_token_accuracy": 0.18749243915081024,
|
|
"num_tokens": 10821894.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 6.000489521026611,
|
|
"epoch": 5.087719298245614,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003142657301111333,
|
|
"loss": 5.4917,
|
|
"mean_token_accuracy": 0.19251226186752318,
|
|
"num_tokens": 10833612.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 6.0291407108306885,
|
|
"epoch": 5.093567251461988,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003138040992529428,
|
|
"loss": 5.5619,
|
|
"mean_token_accuracy": 0.18239215910434722,
|
|
"num_tokens": 10845146.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 6.000576972961426,
|
|
"epoch": 5.099415204678363,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000313342299982179,
|
|
"loss": 5.5615,
|
|
"mean_token_accuracy": 0.18556662499904633,
|
|
"num_tokens": 10857376.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 6.030778074264527,
|
|
"epoch": 5.105263157894737,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003128803343030836,
|
|
"loss": 5.5569,
|
|
"mean_token_accuracy": 0.18679306358098985,
|
|
"num_tokens": 10868778.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 6.005265712738037,
|
|
"epoch": 5.111111111111111,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031241820422062117,
|
|
"loss": 5.591,
|
|
"mean_token_accuracy": 0.1836424797773361,
|
|
"num_tokens": 10881050.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 6.0522665023803714,
|
|
"epoch": 5.116959064327485,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00031195591174046914,
|
|
"loss": 5.6004,
|
|
"mean_token_accuracy": 0.18612522333860398,
|
|
"num_tokens": 10894468.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 6.014405298233032,
|
|
"epoch": 5.12280701754386,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00031149345886901,
|
|
"loss": 5.5092,
|
|
"mean_token_accuracy": 0.19960740953683853,
|
|
"num_tokens": 10907039.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.970701074600219,
|
|
"epoch": 5.128654970760234,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003110308476133225,
|
|
"loss": 5.5217,
|
|
"mean_token_accuracy": 0.18747266829013826,
|
|
"num_tokens": 10918772.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.944737291336059,
|
|
"epoch": 5.1345029239766085,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031056807998117256,
|
|
"loss": 5.4575,
|
|
"mean_token_accuracy": 0.19456324726343155,
|
|
"num_tokens": 10931397.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 6.071500682830811,
|
|
"epoch": 5.140350877192983,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000310105157981005,
|
|
"loss": 5.546,
|
|
"mean_token_accuracy": 0.18476707935333253,
|
|
"num_tokens": 10942728.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 6.011225032806396,
|
|
"epoch": 5.146198830409356,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00030964208362193436,
|
|
"loss": 5.4873,
|
|
"mean_token_accuracy": 0.18788500726222992,
|
|
"num_tokens": 10954277.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.962978982925415,
|
|
"epoch": 5.152046783625731,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00030917885891373664,
|
|
"loss": 5.5394,
|
|
"mean_token_accuracy": 0.18400476723909379,
|
|
"num_tokens": 10966894.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.99399733543396,
|
|
"epoch": 5.157894736842105,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003087154858668404,
|
|
"loss": 5.5165,
|
|
"mean_token_accuracy": 0.18851179033517837,
|
|
"num_tokens": 10979229.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.982893943786621,
|
|
"epoch": 5.16374269005848,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003082519664923177,
|
|
"loss": 5.5851,
|
|
"mean_token_accuracy": 0.18724654763936996,
|
|
"num_tokens": 10991999.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 6.089335870742798,
|
|
"epoch": 5.169590643274854,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000307788302801876,
|
|
"loss": 5.5379,
|
|
"mean_token_accuracy": 0.18286632597446442,
|
|
"num_tokens": 11004408.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.954217720031738,
|
|
"epoch": 5.175438596491228,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00030732449680784886,
|
|
"loss": 5.6347,
|
|
"mean_token_accuracy": 0.1867969289422035,
|
|
"num_tokens": 11016322.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 6.02393069267273,
|
|
"epoch": 5.181286549707602,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003068605505231877,
|
|
"loss": 5.5917,
|
|
"mean_token_accuracy": 0.18348077386617662,
|
|
"num_tokens": 11028798.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.940709257125855,
|
|
"epoch": 5.187134502923977,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00030639646596145243,
|
|
"loss": 5.4176,
|
|
"mean_token_accuracy": 0.19733719080686568,
|
|
"num_tokens": 11039837.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.9875506401062015,
|
|
"epoch": 5.192982456140351,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00030593224513680357,
|
|
"loss": 5.4883,
|
|
"mean_token_accuracy": 0.19005788415670394,
|
|
"num_tokens": 11052943.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 6.096163415908814,
|
|
"epoch": 5.1988304093567255,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003054678900639927,
|
|
"loss": 5.5585,
|
|
"mean_token_accuracy": 0.1826662912964821,
|
|
"num_tokens": 11065483.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.989018678665161,
|
|
"epoch": 5.204678362573099,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00030500340275835405,
|
|
"loss": 5.5376,
|
|
"mean_token_accuracy": 0.1915007621049881,
|
|
"num_tokens": 11077743.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.990274238586426,
|
|
"epoch": 5.2105263157894735,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00030453878523579596,
|
|
"loss": 5.5475,
|
|
"mean_token_accuracy": 0.18578536212444305,
|
|
"num_tokens": 11089692.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.9152443408966064,
|
|
"epoch": 5.216374269005848,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00030407403951279174,
|
|
"loss": 5.3774,
|
|
"mean_token_accuracy": 0.20561654269695281,
|
|
"num_tokens": 11104220.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.9638536930084225,
|
|
"epoch": 5.222222222222222,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003036091676063711,
|
|
"loss": 5.5188,
|
|
"mean_token_accuracy": 0.1857563465833664,
|
|
"num_tokens": 11115599.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.979240560531617,
|
|
"epoch": 5.228070175438597,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003031441715341115,
|
|
"loss": 5.5064,
|
|
"mean_token_accuracy": 0.18795217275619508,
|
|
"num_tokens": 11128125.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.951366424560547,
|
|
"epoch": 5.23391812865497,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003026790533141293,
|
|
"loss": 5.4965,
|
|
"mean_token_accuracy": 0.1928900182247162,
|
|
"num_tokens": 11140105.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.974268007278442,
|
|
"epoch": 5.239766081871345,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003022138149650708,
|
|
"loss": 5.4964,
|
|
"mean_token_accuracy": 0.19691846817731856,
|
|
"num_tokens": 11152291.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.955184698104858,
|
|
"epoch": 5.245614035087719,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003017484585061039,
|
|
"loss": 5.5318,
|
|
"mean_token_accuracy": 0.19350652098655702,
|
|
"num_tokens": 11166330.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 6.017630386352539,
|
|
"epoch": 5.251461988304094,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00030128298595690905,
|
|
"loss": 5.4993,
|
|
"mean_token_accuracy": 0.19427543580532075,
|
|
"num_tokens": 11178737.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.982323217391968,
|
|
"epoch": 5.257309941520468,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00030081739933767034,
|
|
"loss": 5.551,
|
|
"mean_token_accuracy": 0.18491631001234055,
|
|
"num_tokens": 11191179.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.98020372390747,
|
|
"epoch": 5.2631578947368425,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003003517006690672,
|
|
"loss": 5.508,
|
|
"mean_token_accuracy": 0.18862734586000443,
|
|
"num_tokens": 11203890.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 5.2631578947368425,
|
|
"eval_entropy": 5.96874564235541,
|
|
"eval_loss": 6.60414981842041,
|
|
"eval_mean_token_accuracy": 0.14872372614159346,
|
|
"eval_num_tokens": 11203890.0,
|
|
"eval_runtime": 2.4314,
|
|
"eval_samples_per_second": 1113.335,
|
|
"eval_steps_per_second": 139.424,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.962557554244995,
|
|
"epoch": 5.269005847953216,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00029988589197226515,
|
|
"loss": 5.5267,
|
|
"mean_token_accuracy": 0.1900186538696289,
|
|
"num_tokens": 11216556.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 6.011793375015259,
|
|
"epoch": 5.2748538011695905,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00029941997526890745,
|
|
"loss": 5.515,
|
|
"mean_token_accuracy": 0.19090961664915085,
|
|
"num_tokens": 11228352.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.957869625091552,
|
|
"epoch": 5.280701754385965,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000298953952581106,
|
|
"loss": 5.5114,
|
|
"mean_token_accuracy": 0.19029652923345566,
|
|
"num_tokens": 11239423.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.931020784378052,
|
|
"epoch": 5.286549707602339,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00029848782593143246,
|
|
"loss": 5.4886,
|
|
"mean_token_accuracy": 0.1858043596148491,
|
|
"num_tokens": 11251117.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.986592769622803,
|
|
"epoch": 5.292397660818714,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00029802159734291015,
|
|
"loss": 5.544,
|
|
"mean_token_accuracy": 0.19186106324195862,
|
|
"num_tokens": 11264459.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.8925323486328125,
|
|
"epoch": 5.298245614035087,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002975552688390045,
|
|
"loss": 5.4416,
|
|
"mean_token_accuracy": 0.20553764700889587,
|
|
"num_tokens": 11276020.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 6.037499332427979,
|
|
"epoch": 5.304093567251462,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00029708884244361474,
|
|
"loss": 5.5015,
|
|
"mean_token_accuracy": 0.19072462916374205,
|
|
"num_tokens": 11287412.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.963116025924682,
|
|
"epoch": 5.309941520467836,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00029662232018106473,
|
|
"loss": 5.5869,
|
|
"mean_token_accuracy": 0.18666251450777055,
|
|
"num_tokens": 11298607.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 6.037617111206055,
|
|
"epoch": 5.315789473684211,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00029615570407609467,
|
|
"loss": 5.6339,
|
|
"mean_token_accuracy": 0.18118295520544053,
|
|
"num_tokens": 11310943.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.960480880737305,
|
|
"epoch": 5.321637426900585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00029568899615385174,
|
|
"loss": 5.4816,
|
|
"mean_token_accuracy": 0.19128619879484177,
|
|
"num_tokens": 11323684.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.868642330169678,
|
|
"epoch": 5.3274853801169595,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000295222198439882,
|
|
"loss": 5.4619,
|
|
"mean_token_accuracy": 0.19504098147153853,
|
|
"num_tokens": 11336817.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 6.038217782974243,
|
|
"epoch": 5.333333333333333,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002947553129601209,
|
|
"loss": 5.6306,
|
|
"mean_token_accuracy": 0.18371986746788024,
|
|
"num_tokens": 11349246.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.990368366241455,
|
|
"epoch": 5.3391812865497075,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002942883417408849,
|
|
"loss": 5.5203,
|
|
"mean_token_accuracy": 0.1891439750790596,
|
|
"num_tokens": 11360995.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.895724582672119,
|
|
"epoch": 5.345029239766082,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002938212868088626,
|
|
"loss": 5.4685,
|
|
"mean_token_accuracy": 0.19903143644332885,
|
|
"num_tokens": 11373713.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.98705563545227,
|
|
"epoch": 5.350877192982456,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00029335415019110593,
|
|
"loss": 5.5343,
|
|
"mean_token_accuracy": 0.18489537686109542,
|
|
"num_tokens": 11386532.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.9407679557800295,
|
|
"epoch": 5.356725146198831,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002928869339150212,
|
|
"loss": 5.4506,
|
|
"mean_token_accuracy": 0.19612859338521957,
|
|
"num_tokens": 11400535.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.9123039722442625,
|
|
"epoch": 5.362573099415204,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002924196400083607,
|
|
"loss": 5.5142,
|
|
"mean_token_accuracy": 0.19579964131116867,
|
|
"num_tokens": 11412577.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.967027807235718,
|
|
"epoch": 5.368421052631579,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00029195227049921336,
|
|
"loss": 5.5099,
|
|
"mean_token_accuracy": 0.18862133771181105,
|
|
"num_tokens": 11427142.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.998525905609131,
|
|
"epoch": 5.374269005847953,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00029148482741599634,
|
|
"loss": 5.5654,
|
|
"mean_token_accuracy": 0.19296870678663253,
|
|
"num_tokens": 11441319.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.945794725418091,
|
|
"epoch": 5.380116959064328,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00029101731278744636,
|
|
"loss": 5.4891,
|
|
"mean_token_accuracy": 0.1914825439453125,
|
|
"num_tokens": 11453947.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 6.017817592620849,
|
|
"epoch": 5.385964912280702,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002905497286426102,
|
|
"loss": 5.5054,
|
|
"mean_token_accuracy": 0.19210799932479858,
|
|
"num_tokens": 11465463.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 6.0063694477081295,
|
|
"epoch": 5.391812865497076,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00029008207701083677,
|
|
"loss": 5.5368,
|
|
"mean_token_accuracy": 0.19259862303733827,
|
|
"num_tokens": 11477249.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.998052978515625,
|
|
"epoch": 5.39766081871345,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002896143599217674,
|
|
"loss": 5.5976,
|
|
"mean_token_accuracy": 0.18772245198488235,
|
|
"num_tokens": 11490184.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.9521464824676515,
|
|
"epoch": 5.4035087719298245,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002891465794053282,
|
|
"loss": 5.6113,
|
|
"mean_token_accuracy": 0.18754082024097443,
|
|
"num_tokens": 11502935.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 6.0043864250183105,
|
|
"epoch": 5.409356725146199,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002886787374917198,
|
|
"loss": 5.5618,
|
|
"mean_token_accuracy": 0.1942692294716835,
|
|
"num_tokens": 11515385.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 6.027692127227783,
|
|
"epoch": 5.415204678362573,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00028821083621140986,
|
|
"loss": 5.63,
|
|
"mean_token_accuracy": 0.18261874169111253,
|
|
"num_tokens": 11528786.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.985819530487061,
|
|
"epoch": 5.421052631578947,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002877428775951234,
|
|
"loss": 5.6037,
|
|
"mean_token_accuracy": 0.18200570940971375,
|
|
"num_tokens": 11540401.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 6.032489538192749,
|
|
"epoch": 5.426900584795321,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002872748636738344,
|
|
"loss": 5.5725,
|
|
"mean_token_accuracy": 0.1851842522621155,
|
|
"num_tokens": 11552920.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 6.03400182723999,
|
|
"epoch": 5.432748538011696,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002868067964787568,
|
|
"loss": 5.6413,
|
|
"mean_token_accuracy": 0.18326327651739122,
|
|
"num_tokens": 11566237.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.9483922004699705,
|
|
"epoch": 5.43859649122807,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00028633867804133583,
|
|
"loss": 5.5305,
|
|
"mean_token_accuracy": 0.1880714401602745,
|
|
"num_tokens": 11577380.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 6.005774259567261,
|
|
"epoch": 5.444444444444445,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002858705103932391,
|
|
"loss": 5.5708,
|
|
"mean_token_accuracy": 0.18310502767562867,
|
|
"num_tokens": 11590708.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 6.05827145576477,
|
|
"epoch": 5.450292397660819,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00028540229556634785,
|
|
"loss": 5.6029,
|
|
"mean_token_accuracy": 0.18645296096801758,
|
|
"num_tokens": 11603351.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.876799917221069,
|
|
"epoch": 5.456140350877193,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002849340355927477,
|
|
"loss": 5.5271,
|
|
"mean_token_accuracy": 0.19114441126585008,
|
|
"num_tokens": 11615675.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 6.026376962661743,
|
|
"epoch": 5.461988304093567,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002844657325047208,
|
|
"loss": 5.5452,
|
|
"mean_token_accuracy": 0.1858105555176735,
|
|
"num_tokens": 11628184.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 6.022521877288819,
|
|
"epoch": 5.4678362573099415,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002839973883347361,
|
|
"loss": 5.5851,
|
|
"mean_token_accuracy": 0.18516751378774643,
|
|
"num_tokens": 11641058.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.824667978286743,
|
|
"epoch": 5.473684210526316,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00028352900511544076,
|
|
"loss": 5.4761,
|
|
"mean_token_accuracy": 0.19999025911092758,
|
|
"num_tokens": 11653959.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 6.013371992111206,
|
|
"epoch": 5.47953216374269,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00028306058487965156,
|
|
"loss": 5.5737,
|
|
"mean_token_accuracy": 0.18680061548948287,
|
|
"num_tokens": 11665241.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 6.05981593132019,
|
|
"epoch": 5.485380116959064,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002825921296603461,
|
|
"loss": 5.5942,
|
|
"mean_token_accuracy": 0.1876732051372528,
|
|
"num_tokens": 11678670.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.961554765701294,
|
|
"epoch": 5.491228070175438,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00028212364149065347,
|
|
"loss": 5.5071,
|
|
"mean_token_accuracy": 0.18966421186923982,
|
|
"num_tokens": 11689964.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 6.018433856964111,
|
|
"epoch": 5.497076023391813,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002816551224038458,
|
|
"loss": 5.5813,
|
|
"mean_token_accuracy": 0.19146395325660706,
|
|
"num_tokens": 11701895.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.991776180267334,
|
|
"epoch": 5.502923976608187,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00028118657443332984,
|
|
"loss": 5.5395,
|
|
"mean_token_accuracy": 0.18797546327114106,
|
|
"num_tokens": 11714335.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.951577281951904,
|
|
"epoch": 5.508771929824562,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002807179996126371,
|
|
"loss": 5.5164,
|
|
"mean_token_accuracy": 0.18807455897331238,
|
|
"num_tokens": 11726828.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.919211339950562,
|
|
"epoch": 5.514619883040936,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028024939997541606,
|
|
"loss": 5.4804,
|
|
"mean_token_accuracy": 0.19516902714967727,
|
|
"num_tokens": 11740729.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 6.0173440933227536,
|
|
"epoch": 5.52046783625731,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00027978077755542264,
|
|
"loss": 5.59,
|
|
"mean_token_accuracy": 0.18104726821184158,
|
|
"num_tokens": 11753017.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 6.013583517074585,
|
|
"epoch": 5.526315789473684,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002793121343865119,
|
|
"loss": 5.4824,
|
|
"mean_token_accuracy": 0.1929059535264969,
|
|
"num_tokens": 11766139.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.8062207221984865,
|
|
"epoch": 5.5321637426900585,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002788434725026286,
|
|
"loss": 5.3855,
|
|
"mean_token_accuracy": 0.20463382303714753,
|
|
"num_tokens": 11780315.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.984006023406982,
|
|
"epoch": 5.538011695906433,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000278374793937799,
|
|
"loss": 5.5056,
|
|
"mean_token_accuracy": 0.19316982924938203,
|
|
"num_tokens": 11792342.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.958705568313599,
|
|
"epoch": 5.543859649122807,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00027790610072612174,
|
|
"loss": 5.532,
|
|
"mean_token_accuracy": 0.19020303189754487,
|
|
"num_tokens": 11804587.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.994895505905151,
|
|
"epoch": 5.549707602339181,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00027743739490175903,
|
|
"loss": 5.5623,
|
|
"mean_token_accuracy": 0.18966150283813477,
|
|
"num_tokens": 11815884.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 6.049296236038208,
|
|
"epoch": 5.555555555555555,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002769686784989276,
|
|
"loss": 5.6162,
|
|
"mean_token_accuracy": 0.18544263988733292,
|
|
"num_tokens": 11827279.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.963166570663452,
|
|
"epoch": 5.56140350877193,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00027649995355189026,
|
|
"loss": 5.5092,
|
|
"mean_token_accuracy": 0.19190440326929092,
|
|
"num_tokens": 11838646.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 6.061287212371826,
|
|
"epoch": 5.567251461988304,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000276031222094947,
|
|
"loss": 5.5778,
|
|
"mean_token_accuracy": 0.19085145443677903,
|
|
"num_tokens": 11850565.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.8033280849456785,
|
|
"epoch": 5.573099415204679,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002755624861624259,
|
|
"loss": 5.402,
|
|
"mean_token_accuracy": 0.20996314138174058,
|
|
"num_tokens": 11863244.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.838659477233887,
|
|
"epoch": 5.578947368421053,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002750937477886745,
|
|
"loss": 5.5019,
|
|
"mean_token_accuracy": 0.1953998774290085,
|
|
"num_tokens": 11876884.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 6.020962858200074,
|
|
"epoch": 5.584795321637427,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000274625009008051,
|
|
"loss": 5.5108,
|
|
"mean_token_accuracy": 0.19839489907026292,
|
|
"num_tokens": 11889624.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.984988880157471,
|
|
"epoch": 5.590643274853801,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002741562718549152,
|
|
"loss": 5.5672,
|
|
"mean_token_accuracy": 0.18378231525421143,
|
|
"num_tokens": 11901304.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.818964672088623,
|
|
"epoch": 5.5964912280701755,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00027368753836362023,
|
|
"loss": 5.3836,
|
|
"mean_token_accuracy": 0.1986950233578682,
|
|
"num_tokens": 11913188.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 6.019370698928833,
|
|
"epoch": 5.60233918128655,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002732188105685027,
|
|
"loss": 5.6075,
|
|
"mean_token_accuracy": 0.18493279665708542,
|
|
"num_tokens": 11926771.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.977354431152344,
|
|
"epoch": 5.6081871345029235,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002727500905038752,
|
|
"loss": 5.587,
|
|
"mean_token_accuracy": 0.1873197540640831,
|
|
"num_tokens": 11939162.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 6.079027938842773,
|
|
"epoch": 5.614035087719298,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002722813802040161,
|
|
"loss": 5.6368,
|
|
"mean_token_accuracy": 0.17609017938375474,
|
|
"num_tokens": 11952376.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.93961181640625,
|
|
"epoch": 5.619883040935672,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00027181268170316193,
|
|
"loss": 5.4287,
|
|
"mean_token_accuracy": 0.19947983622550963,
|
|
"num_tokens": 11965154.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.911983633041382,
|
|
"epoch": 5.625730994152047,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002713439970354978,
|
|
"loss": 5.5032,
|
|
"mean_token_accuracy": 0.19836400747299193,
|
|
"num_tokens": 11978492.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.988540172576904,
|
|
"epoch": 5.631578947368421,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002708753282351486,
|
|
"loss": 5.5757,
|
|
"mean_token_accuracy": 0.18824120312929155,
|
|
"num_tokens": 11990603.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 6.000478887557984,
|
|
"epoch": 5.637426900584796,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002704066773361706,
|
|
"loss": 5.6084,
|
|
"mean_token_accuracy": 0.18332179188728331,
|
|
"num_tokens": 12002608.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 6.0086933135986325,
|
|
"epoch": 5.643274853801169,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002699380463725425,
|
|
"loss": 5.6119,
|
|
"mean_token_accuracy": 0.18649133294820786,
|
|
"num_tokens": 12016112.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 6.0149843215942385,
|
|
"epoch": 5.649122807017544,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000269469437378156,
|
|
"loss": 5.6197,
|
|
"mean_token_accuracy": 0.18180458694696428,
|
|
"num_tokens": 12027685.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.915473890304566,
|
|
"epoch": 5.654970760233918,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00026900085238680787,
|
|
"loss": 5.4574,
|
|
"mean_token_accuracy": 0.1948918178677559,
|
|
"num_tokens": 12040524.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.9021745204925535,
|
|
"epoch": 5.6608187134502925,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002685322934321907,
|
|
"loss": 5.5284,
|
|
"mean_token_accuracy": 0.1921526312828064,
|
|
"num_tokens": 12052841.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 6.037561273574829,
|
|
"epoch": 5.666666666666667,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002680637625478837,
|
|
"loss": 5.5727,
|
|
"mean_token_accuracy": 0.18483730405569077,
|
|
"num_tokens": 12065375.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 6.048633050918579,
|
|
"epoch": 5.6725146198830405,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002675952617673446,
|
|
"loss": 5.5977,
|
|
"mean_token_accuracy": 0.18980644196271895,
|
|
"num_tokens": 12078034.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.96263952255249,
|
|
"epoch": 5.678362573099415,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002671267931239006,
|
|
"loss": 5.6191,
|
|
"mean_token_accuracy": 0.18664494901895523,
|
|
"num_tokens": 12090218.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.974772691726685,
|
|
"epoch": 5.684210526315789,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026665835865073896,
|
|
"loss": 5.5588,
|
|
"mean_token_accuracy": 0.1896148592233658,
|
|
"num_tokens": 12101927.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 6.064923334121704,
|
|
"epoch": 5.690058479532164,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002661899603808989,
|
|
"loss": 5.5839,
|
|
"mean_token_accuracy": 0.1813189923763275,
|
|
"num_tokens": 12114887.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 6.022850084304809,
|
|
"epoch": 5.695906432748538,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002657216003472626,
|
|
"loss": 5.5758,
|
|
"mean_token_accuracy": 0.18997623771429062,
|
|
"num_tokens": 12127151.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.993904304504395,
|
|
"epoch": 5.701754385964913,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00026525328058254604,
|
|
"loss": 5.5956,
|
|
"mean_token_accuracy": 0.1962890550494194,
|
|
"num_tokens": 12141987.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.891600465774536,
|
|
"epoch": 5.707602339181286,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00026478500311929076,
|
|
"loss": 5.4774,
|
|
"mean_token_accuracy": 0.19275201857089996,
|
|
"num_tokens": 12154525.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.862962102890014,
|
|
"epoch": 5.713450292397661,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00026431676998985444,
|
|
"loss": 5.3915,
|
|
"mean_token_accuracy": 0.20468183159828185,
|
|
"num_tokens": 12166599.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 6.009985065460205,
|
|
"epoch": 5.719298245614035,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002638485832264024,
|
|
"loss": 5.6327,
|
|
"mean_token_accuracy": 0.18718489557504653,
|
|
"num_tokens": 12178658.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.91893367767334,
|
|
"epoch": 5.7251461988304095,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026338044486089895,
|
|
"loss": 5.4835,
|
|
"mean_token_accuracy": 0.1950855076313019,
|
|
"num_tokens": 12190311.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.976849222183228,
|
|
"epoch": 5.730994152046784,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026291235692509795,
|
|
"loss": 5.5828,
|
|
"mean_token_accuracy": 0.1811145231127739,
|
|
"num_tokens": 12202443.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 6.050861883163452,
|
|
"epoch": 5.7368421052631575,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00026244432145053467,
|
|
"loss": 5.6309,
|
|
"mean_token_accuracy": 0.18323461562395096,
|
|
"num_tokens": 12214245.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.944666242599487,
|
|
"epoch": 5.742690058479532,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002619763404685167,
|
|
"loss": 5.5353,
|
|
"mean_token_accuracy": 0.19292946606874467,
|
|
"num_tokens": 12226752.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.9917638301849365,
|
|
"epoch": 5.748538011695906,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00026150841601011495,
|
|
"loss": 5.6116,
|
|
"mean_token_accuracy": 0.18694002777338029,
|
|
"num_tokens": 12239043.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.9785418033599855,
|
|
"epoch": 5.754385964912281,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002610405501061552,
|
|
"loss": 5.5843,
|
|
"mean_token_accuracy": 0.18988247811794282,
|
|
"num_tokens": 12250799.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 6.023903703689575,
|
|
"epoch": 5.760233918128655,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0002605727447872091,
|
|
"loss": 5.6933,
|
|
"mean_token_accuracy": 0.17249799221754075,
|
|
"num_tokens": 12264503.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.976352500915527,
|
|
"epoch": 5.76608187134503,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002601050020835851,
|
|
"loss": 5.5813,
|
|
"mean_token_accuracy": 0.18607886880636215,
|
|
"num_tokens": 12276106.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.957596349716186,
|
|
"epoch": 5.771929824561403,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00025963732402532026,
|
|
"loss": 5.514,
|
|
"mean_token_accuracy": 0.19090104699134827,
|
|
"num_tokens": 12288322.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.965413188934326,
|
|
"epoch": 5.777777777777778,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002591697126421707,
|
|
"loss": 5.5418,
|
|
"mean_token_accuracy": 0.18858572393655776,
|
|
"num_tokens": 12301059.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 6.009155130386352,
|
|
"epoch": 5.783625730994152,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002587021699636035,
|
|
"loss": 5.5899,
|
|
"mean_token_accuracy": 0.18893082588911056,
|
|
"num_tokens": 12313206.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.970478868484497,
|
|
"epoch": 5.7894736842105265,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00025823469801878734,
|
|
"loss": 5.5152,
|
|
"mean_token_accuracy": 0.19645874947309494,
|
|
"num_tokens": 12325426.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.9093094825744625,
|
|
"epoch": 5.7953216374269,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00025776729883658406,
|
|
"loss": 5.5918,
|
|
"mean_token_accuracy": 0.1865328758955002,
|
|
"num_tokens": 12338683.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.97505202293396,
|
|
"epoch": 5.8011695906432745,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002572999744455395,
|
|
"loss": 5.5657,
|
|
"mean_token_accuracy": 0.19645072966814042,
|
|
"num_tokens": 12351139.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.955046463012695,
|
|
"epoch": 5.807017543859649,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00025683272687387513,
|
|
"loss": 5.4776,
|
|
"mean_token_accuracy": 0.19433395713567733,
|
|
"num_tokens": 12362804.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 6.045036268234253,
|
|
"epoch": 5.812865497076023,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002563655581494789,
|
|
"loss": 5.6561,
|
|
"mean_token_accuracy": 0.18500697910785674,
|
|
"num_tokens": 12375069.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.976785039901733,
|
|
"epoch": 5.818713450292398,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00025589847029989653,
|
|
"loss": 5.5473,
|
|
"mean_token_accuracy": 0.18888332545757294,
|
|
"num_tokens": 12387434.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.942727899551391,
|
|
"epoch": 5.824561403508772,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00025543146535232295,
|
|
"loss": 5.5893,
|
|
"mean_token_accuracy": 0.18724985867738725,
|
|
"num_tokens": 12400141.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.952801132202149,
|
|
"epoch": 5.830409356725146,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00025496454533359284,
|
|
"loss": 5.5445,
|
|
"mean_token_accuracy": 0.18958138674497604,
|
|
"num_tokens": 12412619.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.966850900650025,
|
|
"epoch": 5.83625730994152,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00025449771227017265,
|
|
"loss": 5.5374,
|
|
"mean_token_accuracy": 0.18733586221933365,
|
|
"num_tokens": 12425312.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.994867658615112,
|
|
"epoch": 5.842105263157895,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002540309681881516,
|
|
"loss": 5.5773,
|
|
"mean_token_accuracy": 0.18954136222600937,
|
|
"num_tokens": 12436972.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 6.079035043716431,
|
|
"epoch": 5.847953216374269,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002535643151132321,
|
|
"loss": 5.6085,
|
|
"mean_token_accuracy": 0.17920071333646775,
|
|
"num_tokens": 12448543.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.847953216374269,
|
|
"eval_entropy": 5.9722415479586894,
|
|
"eval_loss": 6.567556381225586,
|
|
"eval_mean_token_accuracy": 0.1501677834640386,
|
|
"eval_num_tokens": 12448543.0,
|
|
"eval_runtime": 2.6103,
|
|
"eval_samples_per_second": 1037.046,
|
|
"eval_steps_per_second": 129.87,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.969514656066894,
|
|
"epoch": 5.853801169590644,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00025309775507072234,
|
|
"loss": 5.5338,
|
|
"mean_token_accuracy": 0.1864420548081398,
|
|
"num_tokens": 12462292.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 6.006212139129639,
|
|
"epoch": 5.859649122807017,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002526312900855261,
|
|
"loss": 5.6498,
|
|
"mean_token_accuracy": 0.1812846451997757,
|
|
"num_tokens": 12475930.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 6.044881057739258,
|
|
"epoch": 5.8654970760233915,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002521649221821349,
|
|
"loss": 5.6468,
|
|
"mean_token_accuracy": 0.18115244805812836,
|
|
"num_tokens": 12487769.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.9316086769104,
|
|
"epoch": 5.871345029239766,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00025169865338461894,
|
|
"loss": 5.5671,
|
|
"mean_token_accuracy": 0.19364660531282424,
|
|
"num_tokens": 12500116.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.901181077957153,
|
|
"epoch": 5.87719298245614,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002512324857166182,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.19772685766220094,
|
|
"num_tokens": 12511896.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.9332112789154055,
|
|
"epoch": 5.883040935672515,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002507664212013337,
|
|
"loss": 5.5304,
|
|
"mean_token_accuracy": 0.18781700283288955,
|
|
"num_tokens": 12524435.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.996443367004394,
|
|
"epoch": 5.888888888888889,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00025030046186151915,
|
|
"loss": 5.5717,
|
|
"mean_token_accuracy": 0.18449462801218033,
|
|
"num_tokens": 12536315.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 6.009551906585694,
|
|
"epoch": 5.894736842105263,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002498346097194712,
|
|
"loss": 5.609,
|
|
"mean_token_accuracy": 0.18744842261075972,
|
|
"num_tokens": 12548982.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 6.056502437591552,
|
|
"epoch": 5.900584795321637,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00024936886679702167,
|
|
"loss": 5.719,
|
|
"mean_token_accuracy": 0.1785753309726715,
|
|
"num_tokens": 12562746.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 6.039142608642578,
|
|
"epoch": 5.906432748538012,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00024890323511552825,
|
|
"loss": 5.6096,
|
|
"mean_token_accuracy": 0.1867942690849304,
|
|
"num_tokens": 12575054.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 6.009220695495605,
|
|
"epoch": 5.912280701754386,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00024843771669586575,
|
|
"loss": 5.6345,
|
|
"mean_token_accuracy": 0.18174097836017608,
|
|
"num_tokens": 12588636.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.972481393814087,
|
|
"epoch": 5.918128654970761,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002479723135584176,
|
|
"loss": 5.5659,
|
|
"mean_token_accuracy": 0.18365766555070878,
|
|
"num_tokens": 12602522.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.999080801010132,
|
|
"epoch": 5.923976608187134,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002475070277230668,
|
|
"loss": 5.5824,
|
|
"mean_token_accuracy": 0.18348153829574584,
|
|
"num_tokens": 12616367.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.944814205169678,
|
|
"epoch": 5.9298245614035086,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002470418612091871,
|
|
"loss": 5.5124,
|
|
"mean_token_accuracy": 0.18980114459991454,
|
|
"num_tokens": 12628469.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.969945287704467,
|
|
"epoch": 5.935672514619883,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002465768160356347,
|
|
"loss": 5.4452,
|
|
"mean_token_accuracy": 0.19527565240859984,
|
|
"num_tokens": 12640506.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.979775428771973,
|
|
"epoch": 5.941520467836257,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002461118942207388,
|
|
"loss": 5.6076,
|
|
"mean_token_accuracy": 0.1879318028688431,
|
|
"num_tokens": 12652862.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.925210571289062,
|
|
"epoch": 5.947368421052632,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00024564709778229353,
|
|
"loss": 5.5079,
|
|
"mean_token_accuracy": 0.1902143895626068,
|
|
"num_tokens": 12664390.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.941389274597168,
|
|
"epoch": 5.953216374269006,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00024518242873754887,
|
|
"loss": 5.4747,
|
|
"mean_token_accuracy": 0.19881088882684708,
|
|
"num_tokens": 12676489.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.901304817199707,
|
|
"epoch": 5.95906432748538,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00024471788910320163,
|
|
"loss": 5.4674,
|
|
"mean_token_accuracy": 0.1936710298061371,
|
|
"num_tokens": 12688666.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.977006292343139,
|
|
"epoch": 5.964912280701754,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00024425348089538714,
|
|
"loss": 5.5779,
|
|
"mean_token_accuracy": 0.187896828353405,
|
|
"num_tokens": 12702159.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 6.008750534057617,
|
|
"epoch": 5.970760233918129,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002437892061296705,
|
|
"loss": 5.6195,
|
|
"mean_token_accuracy": 0.18383994847536086,
|
|
"num_tokens": 12715528.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 6.032082557678223,
|
|
"epoch": 5.976608187134503,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00024332506682103722,
|
|
"loss": 5.6665,
|
|
"mean_token_accuracy": 0.17827052921056746,
|
|
"num_tokens": 12727485.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.930669069290161,
|
|
"epoch": 5.982456140350877,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00024286106498388533,
|
|
"loss": 5.5436,
|
|
"mean_token_accuracy": 0.19424745738506316,
|
|
"num_tokens": 12740602.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.987565326690674,
|
|
"epoch": 5.988304093567251,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002423972026320162,
|
|
"loss": 5.5953,
|
|
"mean_token_accuracy": 0.1864088237285614,
|
|
"num_tokens": 12752599.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 6.034543228149414,
|
|
"epoch": 5.994152046783626,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00024193348177862551,
|
|
"loss": 5.5987,
|
|
"mean_token_accuracy": 0.1874092698097229,
|
|
"num_tokens": 12764515.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.955098104476929,
|
|
"epoch": 6.0,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00024146990443629508,
|
|
"loss": 5.5275,
|
|
"mean_token_accuracy": 0.19177109450101854,
|
|
"num_tokens": 12776904.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 6.064900875091553,
|
|
"epoch": 6.005847953216374,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002410064726169838,
|
|
"loss": 5.4447,
|
|
"mean_token_accuracy": 0.19268578439950942,
|
|
"num_tokens": 12789313.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.921218919754028,
|
|
"epoch": 6.011695906432749,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00024054318833201894,
|
|
"loss": 5.2963,
|
|
"mean_token_accuracy": 0.20243059992790222,
|
|
"num_tokens": 12801932.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.959689950942993,
|
|
"epoch": 6.017543859649122,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00024008005359208774,
|
|
"loss": 5.3384,
|
|
"mean_token_accuracy": 0.20557017475366593,
|
|
"num_tokens": 12814530.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 6.042625617980957,
|
|
"epoch": 6.023391812865497,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023961707040722792,
|
|
"loss": 5.4476,
|
|
"mean_token_accuracy": 0.19001543819904326,
|
|
"num_tokens": 12827053.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.990024089813232,
|
|
"epoch": 6.029239766081871,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023915424078681987,
|
|
"loss": 5.3054,
|
|
"mean_token_accuracy": 0.19994196891784669,
|
|
"num_tokens": 12838875.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.950985097885132,
|
|
"epoch": 6.035087719298246,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00023869156673957733,
|
|
"loss": 5.4092,
|
|
"mean_token_accuracy": 0.1907335713505745,
|
|
"num_tokens": 12851434.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 6.004544830322265,
|
|
"epoch": 6.04093567251462,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00023822905027353863,
|
|
"loss": 5.3979,
|
|
"mean_token_accuracy": 0.19422474652528762,
|
|
"num_tokens": 12864275.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.93700590133667,
|
|
"epoch": 6.046783625730994,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002377666933960586,
|
|
"loss": 5.3523,
|
|
"mean_token_accuracy": 0.19970278143882753,
|
|
"num_tokens": 12876675.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.870529985427856,
|
|
"epoch": 6.052631578947368,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002373044981137994,
|
|
"loss": 5.2417,
|
|
"mean_token_accuracy": 0.21245940923690795,
|
|
"num_tokens": 12888482.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.922803497314453,
|
|
"epoch": 6.058479532163743,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00023684246643272132,
|
|
"loss": 5.2629,
|
|
"mean_token_accuracy": 0.1989670768380165,
|
|
"num_tokens": 12901631.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 6.039050006866455,
|
|
"epoch": 6.064327485380117,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002363806003580752,
|
|
"loss": 5.4273,
|
|
"mean_token_accuracy": 0.18946568220853804,
|
|
"num_tokens": 12913210.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.899736976623535,
|
|
"epoch": 6.0701754385964914,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023591890189439296,
|
|
"loss": 5.2486,
|
|
"mean_token_accuracy": 0.2069963440299034,
|
|
"num_tokens": 12924964.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.921000719070435,
|
|
"epoch": 6.076023391812866,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023545737304547893,
|
|
"loss": 5.3517,
|
|
"mean_token_accuracy": 0.19325063079595567,
|
|
"num_tokens": 12936986.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 6.0719140529632565,
|
|
"epoch": 6.081871345029239,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002349960158144015,
|
|
"loss": 5.4246,
|
|
"mean_token_accuracy": 0.19281788915395737,
|
|
"num_tokens": 12949842.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 6.023975563049317,
|
|
"epoch": 6.087719298245614,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000234534832203484,
|
|
"loss": 5.4139,
|
|
"mean_token_accuracy": 0.19845283329486846,
|
|
"num_tokens": 12962112.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.906809329986572,
|
|
"epoch": 6.093567251461988,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002340738242142965,
|
|
"loss": 5.3077,
|
|
"mean_token_accuracy": 0.2061280533671379,
|
|
"num_tokens": 12973599.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.845286750793457,
|
|
"epoch": 6.099415204678363,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00023361299384764677,
|
|
"loss": 5.2637,
|
|
"mean_token_accuracy": 0.2109714478254318,
|
|
"num_tokens": 12987402.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.854632568359375,
|
|
"epoch": 6.105263157894737,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023315234310357147,
|
|
"loss": 5.2643,
|
|
"mean_token_accuracy": 0.20496350526809692,
|
|
"num_tokens": 12999299.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.986026430130005,
|
|
"epoch": 6.111111111111111,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023269187398132802,
|
|
"loss": 5.3843,
|
|
"mean_token_accuracy": 0.19448399692773818,
|
|
"num_tokens": 13011911.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.954691076278687,
|
|
"epoch": 6.116959064327485,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023223158847938543,
|
|
"loss": 5.3663,
|
|
"mean_token_accuracy": 0.1976665213704109,
|
|
"num_tokens": 13023493.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.924326324462891,
|
|
"epoch": 6.12280701754386,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00023177148859541565,
|
|
"loss": 5.3374,
|
|
"mean_token_accuracy": 0.2014690324664116,
|
|
"num_tokens": 13036364.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.92272572517395,
|
|
"epoch": 6.128654970760234,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023131157632628525,
|
|
"loss": 5.3868,
|
|
"mean_token_accuracy": 0.19727135747671126,
|
|
"num_tokens": 13048809.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.97372875213623,
|
|
"epoch": 6.1345029239766085,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023085185366804656,
|
|
"loss": 5.2896,
|
|
"mean_token_accuracy": 0.2047206863760948,
|
|
"num_tokens": 13061811.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.979695081710815,
|
|
"epoch": 6.140350877192983,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00023039232261592868,
|
|
"loss": 5.4164,
|
|
"mean_token_accuracy": 0.1925053507089615,
|
|
"num_tokens": 13074331.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.958146619796753,
|
|
"epoch": 6.146198830409356,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022993298516432937,
|
|
"loss": 5.3491,
|
|
"mean_token_accuracy": 0.20801514387130737,
|
|
"num_tokens": 13087642.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 6.013497686386108,
|
|
"epoch": 6.152046783625731,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00022947384330680636,
|
|
"loss": 5.4525,
|
|
"mean_token_accuracy": 0.18894832730293273,
|
|
"num_tokens": 13099453.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.975533056259155,
|
|
"epoch": 6.157894736842105,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00022901489903606794,
|
|
"loss": 5.3375,
|
|
"mean_token_accuracy": 0.20254694074392318,
|
|
"num_tokens": 13112092.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.887092351913452,
|
|
"epoch": 6.16374269005848,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022855615434396527,
|
|
"loss": 5.257,
|
|
"mean_token_accuracy": 0.20957382172346115,
|
|
"num_tokens": 13125023.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.938323593139648,
|
|
"epoch": 6.169590643274854,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002280976112214832,
|
|
"loss": 5.4114,
|
|
"mean_token_accuracy": 0.19596203714609145,
|
|
"num_tokens": 13137902.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.886929130554199,
|
|
"epoch": 6.175438596491228,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022763927165873177,
|
|
"loss": 5.318,
|
|
"mean_token_accuracy": 0.20674031227827072,
|
|
"num_tokens": 13151245.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.818854856491089,
|
|
"epoch": 6.181286549707602,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002271811376449376,
|
|
"loss": 5.1817,
|
|
"mean_token_accuracy": 0.21706302762031554,
|
|
"num_tokens": 13163916.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.936476802825927,
|
|
"epoch": 6.187134502923977,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022672321116843497,
|
|
"loss": 5.3566,
|
|
"mean_token_accuracy": 0.20504155158996581,
|
|
"num_tokens": 13176265.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.976873588562012,
|
|
"epoch": 6.192982456140351,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00022626549421665788,
|
|
"loss": 5.3993,
|
|
"mean_token_accuracy": 0.19733954668045045,
|
|
"num_tokens": 13188916.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.90433259010315,
|
|
"epoch": 6.1988304093567255,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00022580798877613062,
|
|
"loss": 5.3107,
|
|
"mean_token_accuracy": 0.20384003967046738,
|
|
"num_tokens": 13201444.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.985468149185181,
|
|
"epoch": 6.204678362573099,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00022535069683245952,
|
|
"loss": 5.3691,
|
|
"mean_token_accuracy": 0.199996118247509,
|
|
"num_tokens": 13213390.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.887739181518555,
|
|
"epoch": 6.2105263157894735,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022489362037032446,
|
|
"loss": 5.2661,
|
|
"mean_token_accuracy": 0.20353585481643677,
|
|
"num_tokens": 13225321.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.931001091003418,
|
|
"epoch": 6.216374269005848,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002244367613734702,
|
|
"loss": 5.3343,
|
|
"mean_token_accuracy": 0.19876824766397477,
|
|
"num_tokens": 13237737.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 6.002532243728638,
|
|
"epoch": 6.222222222222222,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00022398012182469734,
|
|
"loss": 5.4409,
|
|
"mean_token_accuracy": 0.19633575826883315,
|
|
"num_tokens": 13250235.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.9672768115997314,
|
|
"epoch": 6.228070175438597,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002235237037058544,
|
|
"loss": 5.4268,
|
|
"mean_token_accuracy": 0.18997211903333663,
|
|
"num_tokens": 13263537.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.96286301612854,
|
|
"epoch": 6.23391812865497,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022306750899782875,
|
|
"loss": 5.3843,
|
|
"mean_token_accuracy": 0.20043936967849732,
|
|
"num_tokens": 13275436.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.986579704284668,
|
|
"epoch": 6.239766081871345,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000222611539680538,
|
|
"loss": 5.421,
|
|
"mean_token_accuracy": 0.1945247769355774,
|
|
"num_tokens": 13287346.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.877496337890625,
|
|
"epoch": 6.245614035087719,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022215579773292188,
|
|
"loss": 5.296,
|
|
"mean_token_accuracy": 0.20808068364858628,
|
|
"num_tokens": 13299145.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.927312660217285,
|
|
"epoch": 6.251461988304094,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00022170028513293288,
|
|
"loss": 5.3615,
|
|
"mean_token_accuracy": 0.2018896222114563,
|
|
"num_tokens": 13311709.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.9335521221160885,
|
|
"epoch": 6.257309941520468,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002212450038575285,
|
|
"loss": 5.3439,
|
|
"mean_token_accuracy": 0.20503126829862595,
|
|
"num_tokens": 13324441.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.946239805221557,
|
|
"epoch": 6.2631578947368425,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00022078995588266227,
|
|
"loss": 5.2862,
|
|
"mean_token_accuracy": 0.20757026374340057,
|
|
"num_tokens": 13337276.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.904503965377808,
|
|
"epoch": 6.269005847953216,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002203351431832748,
|
|
"loss": 5.3468,
|
|
"mean_token_accuracy": 0.1951109379529953,
|
|
"num_tokens": 13350254.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.963783025741577,
|
|
"epoch": 6.2748538011695905,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002198805677332858,
|
|
"loss": 5.4479,
|
|
"mean_token_accuracy": 0.19651356637477874,
|
|
"num_tokens": 13363430.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.935182285308838,
|
|
"epoch": 6.280701754385965,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021942623150558556,
|
|
"loss": 5.3386,
|
|
"mean_token_accuracy": 0.19978471994400024,
|
|
"num_tokens": 13375350.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.9728387832641605,
|
|
"epoch": 6.286549707602339,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00021897213647202558,
|
|
"loss": 5.4017,
|
|
"mean_token_accuracy": 0.19889695793390275,
|
|
"num_tokens": 13388339.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.853464412689209,
|
|
"epoch": 6.292397660818714,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002185182846034111,
|
|
"loss": 5.2561,
|
|
"mean_token_accuracy": 0.2158266693353653,
|
|
"num_tokens": 13401122.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.965099954605103,
|
|
"epoch": 6.298245614035087,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002180646778694918,
|
|
"loss": 5.338,
|
|
"mean_token_accuracy": 0.20385863333940507,
|
|
"num_tokens": 13413428.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.9216712474822994,
|
|
"epoch": 6.304093567251462,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00021761131823895324,
|
|
"loss": 5.3793,
|
|
"mean_token_accuracy": 0.2030568614602089,
|
|
"num_tokens": 13427045.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.948976755142212,
|
|
"epoch": 6.309941520467836,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002171582076794088,
|
|
"loss": 5.3574,
|
|
"mean_token_accuracy": 0.19904676228761672,
|
|
"num_tokens": 13439524.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 6.009565591812134,
|
|
"epoch": 6.315789473684211,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002167053481573909,
|
|
"loss": 5.4119,
|
|
"mean_token_accuracy": 0.19538529217243195,
|
|
"num_tokens": 13453705.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.867125988006592,
|
|
"epoch": 6.321637426900585,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00021625274163834224,
|
|
"loss": 5.3029,
|
|
"mean_token_accuracy": 0.21245252043008805,
|
|
"num_tokens": 13466129.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.962126636505127,
|
|
"epoch": 6.3274853801169595,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00021580039008660762,
|
|
"loss": 5.3956,
|
|
"mean_token_accuracy": 0.19962157309055328,
|
|
"num_tokens": 13479522.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.917705297470093,
|
|
"epoch": 6.333333333333333,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00021534829546542518,
|
|
"loss": 5.289,
|
|
"mean_token_accuracy": 0.19926925599575043,
|
|
"num_tokens": 13492620.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.909750938415527,
|
|
"epoch": 6.3391812865497075,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002148964597369179,
|
|
"loss": 5.2935,
|
|
"mean_token_accuracy": 0.2055873990058899,
|
|
"num_tokens": 13503573.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.894689464569092,
|
|
"epoch": 6.345029239766082,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00021444488486208545,
|
|
"loss": 5.3544,
|
|
"mean_token_accuracy": 0.20585819631814956,
|
|
"num_tokens": 13515999.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.94100980758667,
|
|
"epoch": 6.350877192982456,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00021399357280079502,
|
|
"loss": 5.3497,
|
|
"mean_token_accuracy": 0.20052539855241774,
|
|
"num_tokens": 13528480.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.937503290176392,
|
|
"epoch": 6.356725146198831,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00021354252551177333,
|
|
"loss": 5.3972,
|
|
"mean_token_accuracy": 0.200355464220047,
|
|
"num_tokens": 13540606.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 6.004737377166748,
|
|
"epoch": 6.362573099415204,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00021309174495259814,
|
|
"loss": 5.3854,
|
|
"mean_token_accuracy": 0.1953230008482933,
|
|
"num_tokens": 13552947.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.937303447723389,
|
|
"epoch": 6.368421052631579,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00021264123307968913,
|
|
"loss": 5.3492,
|
|
"mean_token_accuracy": 0.2004652127623558,
|
|
"num_tokens": 13565499.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.830477046966553,
|
|
"epoch": 6.374269005847953,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002121909918483002,
|
|
"loss": 5.2269,
|
|
"mean_token_accuracy": 0.22254078686237336,
|
|
"num_tokens": 13578491.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.895497179031372,
|
|
"epoch": 6.380116959064328,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00021174102321251082,
|
|
"loss": 5.3202,
|
|
"mean_token_accuracy": 0.20375825613737106,
|
|
"num_tokens": 13590638.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.956038284301758,
|
|
"epoch": 6.385964912280702,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002112913291252168,
|
|
"loss": 5.4019,
|
|
"mean_token_accuracy": 0.19620103985071183,
|
|
"num_tokens": 13602198.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.9529869556427,
|
|
"epoch": 6.391812865497076,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000210841911538123,
|
|
"loss": 5.3564,
|
|
"mean_token_accuracy": 0.19753707200288773,
|
|
"num_tokens": 13615520.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.8862005233764645,
|
|
"epoch": 6.39766081871345,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00021039277240173372,
|
|
"loss": 5.374,
|
|
"mean_token_accuracy": 0.19987971037626268,
|
|
"num_tokens": 13629382.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.759228658676148,
|
|
"epoch": 6.4035087719298245,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002099439136653452,
|
|
"loss": 5.1355,
|
|
"mean_token_accuracy": 0.2163531959056854,
|
|
"num_tokens": 13642336.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.939532852172851,
|
|
"epoch": 6.409356725146199,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00020949533727703657,
|
|
"loss": 5.3689,
|
|
"mean_token_accuracy": 0.19888295829296113,
|
|
"num_tokens": 13654729.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.9622186660766605,
|
|
"epoch": 6.415204678362573,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00020904704518366136,
|
|
"loss": 5.321,
|
|
"mean_token_accuracy": 0.20493457317352295,
|
|
"num_tokens": 13667068.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.880156517028809,
|
|
"epoch": 6.421052631578947,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00020859903933083952,
|
|
"loss": 5.3575,
|
|
"mean_token_accuracy": 0.20183506608009338,
|
|
"num_tokens": 13680164.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.949753093719482,
|
|
"epoch": 6.426900584795321,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002081513216629486,
|
|
"loss": 5.4432,
|
|
"mean_token_accuracy": 0.19225913733243943,
|
|
"num_tokens": 13693510.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.958605098724365,
|
|
"epoch": 6.432748538011696,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002077038941231153,
|
|
"loss": 5.3261,
|
|
"mean_token_accuracy": 0.1985008180141449,
|
|
"num_tokens": 13705763.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 6.432748538011696,
|
|
"eval_entropy": 5.9271784996212755,
|
|
"eval_loss": 6.575543403625488,
|
|
"eval_mean_token_accuracy": 0.15067745491334822,
|
|
"eval_num_tokens": 13705763.0,
|
|
"eval_runtime": 2.2387,
|
|
"eval_samples_per_second": 1209.176,
|
|
"eval_steps_per_second": 151.426,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.960172319412232,
|
|
"epoch": 6.43859649122807,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00020725675865320721,
|
|
"loss": 5.3608,
|
|
"mean_token_accuracy": 0.1977481409907341,
|
|
"num_tokens": 13719223.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.906861019134522,
|
|
"epoch": 6.444444444444445,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00020680991719382446,
|
|
"loss": 5.328,
|
|
"mean_token_accuracy": 0.20375655889511107,
|
|
"num_tokens": 13731571.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.959193849563599,
|
|
"epoch": 6.450292397660819,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00020636337168429092,
|
|
"loss": 5.4623,
|
|
"mean_token_accuracy": 0.19494023621082307,
|
|
"num_tokens": 13744212.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.982366943359375,
|
|
"epoch": 6.456140350877193,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00020591712406264613,
|
|
"loss": 5.4488,
|
|
"mean_token_accuracy": 0.1927955836057663,
|
|
"num_tokens": 13756402.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.990514755249023,
|
|
"epoch": 6.461988304093567,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002054711762656369,
|
|
"loss": 5.3609,
|
|
"mean_token_accuracy": 0.20121330320835112,
|
|
"num_tokens": 13768386.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.8927123069763185,
|
|
"epoch": 6.4678362573099415,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002050255302287084,
|
|
"loss": 5.3056,
|
|
"mean_token_accuracy": 0.2077102079987526,
|
|
"num_tokens": 13778732.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.882137203216553,
|
|
"epoch": 6.473684210526316,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002045801878859967,
|
|
"loss": 5.3234,
|
|
"mean_token_accuracy": 0.20979188233613968,
|
|
"num_tokens": 13791708.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.92235107421875,
|
|
"epoch": 6.47953216374269,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00020413515117031917,
|
|
"loss": 5.3985,
|
|
"mean_token_accuracy": 0.19748836755752563,
|
|
"num_tokens": 13802526.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.887267398834228,
|
|
"epoch": 6.485380116959064,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00020369042201316722,
|
|
"loss": 5.374,
|
|
"mean_token_accuracy": 0.20154989510774612,
|
|
"num_tokens": 13815342.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.995775270462036,
|
|
"epoch": 6.491228070175438,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00020324600234469743,
|
|
"loss": 5.3762,
|
|
"mean_token_accuracy": 0.19823614060878753,
|
|
"num_tokens": 13827223.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.889878606796264,
|
|
"epoch": 6.497076023391813,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00020280189409372274,
|
|
"loss": 5.3328,
|
|
"mean_token_accuracy": 0.20451215505599976,
|
|
"num_tokens": 13840366.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.951476287841797,
|
|
"epoch": 6.502923976608187,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00020235809918770504,
|
|
"loss": 5.4612,
|
|
"mean_token_accuracy": 0.19124742448329926,
|
|
"num_tokens": 13853295.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.9772453784942625,
|
|
"epoch": 6.508771929824562,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00020191461955274597,
|
|
"loss": 5.4313,
|
|
"mean_token_accuracy": 0.20060124695301057,
|
|
"num_tokens": 13866551.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.965367031097412,
|
|
"epoch": 6.514619883040936,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002014714571135789,
|
|
"loss": 5.4068,
|
|
"mean_token_accuracy": 0.1984175130724907,
|
|
"num_tokens": 13878048.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.950995016098022,
|
|
"epoch": 6.52046783625731,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00020102861379356064,
|
|
"loss": 5.3884,
|
|
"mean_token_accuracy": 0.19899266809225083,
|
|
"num_tokens": 13890742.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.91658501625061,
|
|
"epoch": 6.526315789473684,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000200586091514663,
|
|
"loss": 5.3554,
|
|
"mean_token_accuracy": 0.1993677407503128,
|
|
"num_tokens": 13902943.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.873514080047608,
|
|
"epoch": 6.5321637426900585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00020014389219746438,
|
|
"loss": 5.2956,
|
|
"mean_token_accuracy": 0.19831887036561965,
|
|
"num_tokens": 13915732.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.8918133735656735,
|
|
"epoch": 6.538011695906433,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001997020177611414,
|
|
"loss": 5.3506,
|
|
"mean_token_accuracy": 0.1962844431400299,
|
|
"num_tokens": 13929173.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.9554249286651615,
|
|
"epoch": 6.543859649122807,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00019926047012346098,
|
|
"loss": 5.331,
|
|
"mean_token_accuracy": 0.20010444670915603,
|
|
"num_tokens": 13940288.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.86049256324768,
|
|
"epoch": 6.549707602339181,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00019881925120077143,
|
|
"loss": 5.2532,
|
|
"mean_token_accuracy": 0.21149016171693802,
|
|
"num_tokens": 13952633.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.818957042694092,
|
|
"epoch": 6.555555555555555,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00019837836290799454,
|
|
"loss": 5.28,
|
|
"mean_token_accuracy": 0.2099355936050415,
|
|
"num_tokens": 13964040.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.914199352264404,
|
|
"epoch": 6.56140350877193,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001979378071586171,
|
|
"loss": 5.2959,
|
|
"mean_token_accuracy": 0.2070560038089752,
|
|
"num_tokens": 13976156.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.962115526199341,
|
|
"epoch": 6.567251461988304,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001974975858646827,
|
|
"loss": 5.3115,
|
|
"mean_token_accuracy": 0.2090566262602806,
|
|
"num_tokens": 13987815.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.9349579334259035,
|
|
"epoch": 6.573099415204679,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019705770093678333,
|
|
"loss": 5.4121,
|
|
"mean_token_accuracy": 0.19032588005065917,
|
|
"num_tokens": 13999239.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.92317762374878,
|
|
"epoch": 6.578947368421053,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00019661815428405094,
|
|
"loss": 5.4419,
|
|
"mean_token_accuracy": 0.1879319116473198,
|
|
"num_tokens": 14012146.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.951567602157593,
|
|
"epoch": 6.584795321637427,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019617894781414968,
|
|
"loss": 5.3635,
|
|
"mean_token_accuracy": 0.20080339461565017,
|
|
"num_tokens": 14024292.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.914742708206177,
|
|
"epoch": 6.590643274853801,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00019574008343326726,
|
|
"loss": 5.3494,
|
|
"mean_token_accuracy": 0.19839009195566176,
|
|
"num_tokens": 14037011.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.960331153869629,
|
|
"epoch": 6.5964912280701755,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00019530156304610623,
|
|
"loss": 5.3282,
|
|
"mean_token_accuracy": 0.19663824886083603,
|
|
"num_tokens": 14049006.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.88726840019226,
|
|
"epoch": 6.60233918128655,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019486338855587683,
|
|
"loss": 5.3186,
|
|
"mean_token_accuracy": 0.2030755639076233,
|
|
"num_tokens": 14061594.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.947085905075073,
|
|
"epoch": 6.6081871345029235,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019442556186428762,
|
|
"loss": 5.4081,
|
|
"mean_token_accuracy": 0.19298507422208785,
|
|
"num_tokens": 14073244.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.911041355133056,
|
|
"epoch": 6.614035087719298,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001939880848715378,
|
|
"loss": 5.2835,
|
|
"mean_token_accuracy": 0.21012082546949387,
|
|
"num_tokens": 14086695.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.89512882232666,
|
|
"epoch": 6.619883040935672,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00019355095947630897,
|
|
"loss": 5.3691,
|
|
"mean_token_accuracy": 0.19647617191076278,
|
|
"num_tokens": 14099161.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.977403450012207,
|
|
"epoch": 6.625730994152047,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00019311418757575672,
|
|
"loss": 5.3433,
|
|
"mean_token_accuracy": 0.2004932165145874,
|
|
"num_tokens": 14110602.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.914765691757202,
|
|
"epoch": 6.631578947368421,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00019267777106550229,
|
|
"loss": 5.2943,
|
|
"mean_token_accuracy": 0.20131977200508117,
|
|
"num_tokens": 14122802.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.879426956176758,
|
|
"epoch": 6.637426900584796,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00019224171183962474,
|
|
"loss": 5.3502,
|
|
"mean_token_accuracy": 0.20228248238563537,
|
|
"num_tokens": 14135296.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 6.021072006225586,
|
|
"epoch": 6.643274853801169,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019180601179065232,
|
|
"loss": 5.4385,
|
|
"mean_token_accuracy": 0.19244106858968735,
|
|
"num_tokens": 14146825.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 6.022170448303223,
|
|
"epoch": 6.649122807017544,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001913706728095545,
|
|
"loss": 5.3972,
|
|
"mean_token_accuracy": 0.19312668591737747,
|
|
"num_tokens": 14159655.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.962015676498413,
|
|
"epoch": 6.654970760233918,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019093569678573375,
|
|
"loss": 5.4436,
|
|
"mean_token_accuracy": 0.19607775360345842,
|
|
"num_tokens": 14171421.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.86878342628479,
|
|
"epoch": 6.6608187134502925,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00019050108560701706,
|
|
"loss": 5.4037,
|
|
"mean_token_accuracy": 0.1935245230793953,
|
|
"num_tokens": 14183890.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.981724452972412,
|
|
"epoch": 6.666666666666667,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019006684115964828,
|
|
"loss": 5.4032,
|
|
"mean_token_accuracy": 0.19034560918807983,
|
|
"num_tokens": 14195715.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.992841577529907,
|
|
"epoch": 6.6725146198830405,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00018963296532827935,
|
|
"loss": 5.4207,
|
|
"mean_token_accuracy": 0.19373631626367568,
|
|
"num_tokens": 14207426.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.926816987991333,
|
|
"epoch": 6.678362573099415,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00018919945999596244,
|
|
"loss": 5.3985,
|
|
"mean_token_accuracy": 0.19680504202842714,
|
|
"num_tokens": 14219491.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.920701265335083,
|
|
"epoch": 6.684210526315789,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001887663270441418,
|
|
"loss": 5.3868,
|
|
"mean_token_accuracy": 0.2004100948572159,
|
|
"num_tokens": 14232130.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.997941398620606,
|
|
"epoch": 6.690058479532164,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00018833356835264576,
|
|
"loss": 5.3776,
|
|
"mean_token_accuracy": 0.19530023783445358,
|
|
"num_tokens": 14243323.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.949048948287964,
|
|
"epoch": 6.695906432748538,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00018790118579967762,
|
|
"loss": 5.4258,
|
|
"mean_token_accuracy": 0.19468031227588653,
|
|
"num_tokens": 14255338.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.970604944229126,
|
|
"epoch": 6.701754385964913,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000187469181261809,
|
|
"loss": 5.4226,
|
|
"mean_token_accuracy": 0.19823749959468842,
|
|
"num_tokens": 14268259.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.972260189056397,
|
|
"epoch": 6.707602339181286,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001870375566139706,
|
|
"loss": 5.4298,
|
|
"mean_token_accuracy": 0.19584240466356279,
|
|
"num_tokens": 14280208.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.86528697013855,
|
|
"epoch": 6.713450292397661,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00018660631372944403,
|
|
"loss": 5.2981,
|
|
"mean_token_accuracy": 0.20782756805419922,
|
|
"num_tokens": 14294018.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.922221755981445,
|
|
"epoch": 6.719298245614035,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00018617545447985461,
|
|
"loss": 5.3237,
|
|
"mean_token_accuracy": 0.20385898947715758,
|
|
"num_tokens": 14307023.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.930307245254516,
|
|
"epoch": 6.7251461988304095,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00018574498073516227,
|
|
"loss": 5.3927,
|
|
"mean_token_accuracy": 0.19293662756681443,
|
|
"num_tokens": 14319400.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 6.0053393840789795,
|
|
"epoch": 6.730994152046784,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00018531489436365395,
|
|
"loss": 5.4449,
|
|
"mean_token_accuracy": 0.1932617411017418,
|
|
"num_tokens": 14331288.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.889837026596069,
|
|
"epoch": 6.7368421052631575,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00018488519723193543,
|
|
"loss": 5.3029,
|
|
"mean_token_accuracy": 0.21191488057374955,
|
|
"num_tokens": 14344148.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.95243010520935,
|
|
"epoch": 6.742690058479532,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001844558912049229,
|
|
"loss": 5.3768,
|
|
"mean_token_accuracy": 0.19607398211956023,
|
|
"num_tokens": 14355906.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.975778245925904,
|
|
"epoch": 6.748538011695906,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00018402697814583535,
|
|
"loss": 5.4716,
|
|
"mean_token_accuracy": 0.1861868754029274,
|
|
"num_tokens": 14367028.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.933220434188843,
|
|
"epoch": 6.754385964912281,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001835984599161864,
|
|
"loss": 5.4645,
|
|
"mean_token_accuracy": 0.19147694408893584,
|
|
"num_tokens": 14380873.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.995866489410401,
|
|
"epoch": 6.760233918128655,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00018317033837577564,
|
|
"loss": 5.4274,
|
|
"mean_token_accuracy": 0.18774820268154144,
|
|
"num_tokens": 14394509.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.993253898620606,
|
|
"epoch": 6.76608187134503,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001827426153826814,
|
|
"loss": 5.3987,
|
|
"mean_token_accuracy": 0.1931233748793602,
|
|
"num_tokens": 14406330.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.988535642623901,
|
|
"epoch": 6.771929824561403,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001823152927932521,
|
|
"loss": 5.3917,
|
|
"mean_token_accuracy": 0.19790363907814026,
|
|
"num_tokens": 14418731.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.973641014099121,
|
|
"epoch": 6.777777777777778,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001818883724620985,
|
|
"loss": 5.3658,
|
|
"mean_token_accuracy": 0.20009478330612182,
|
|
"num_tokens": 14430038.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 6.021954154968261,
|
|
"epoch": 6.783625730994152,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00018146185624208527,
|
|
"loss": 5.4475,
|
|
"mean_token_accuracy": 0.19209431260824203,
|
|
"num_tokens": 14444212.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.954354000091553,
|
|
"epoch": 6.7894736842105265,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00018103574598432365,
|
|
"loss": 5.4418,
|
|
"mean_token_accuracy": 0.18894915282726288,
|
|
"num_tokens": 14457462.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.963011884689331,
|
|
"epoch": 6.7953216374269,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00018061004353816246,
|
|
"loss": 5.3673,
|
|
"mean_token_accuracy": 0.1932898297905922,
|
|
"num_tokens": 14469356.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.920188760757446,
|
|
"epoch": 6.8011695906432745,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000180184750751181,
|
|
"loss": 5.2955,
|
|
"mean_token_accuracy": 0.20596393048763276,
|
|
"num_tokens": 14482555.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.903044271469116,
|
|
"epoch": 6.807017543859649,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001797598694691804,
|
|
"loss": 5.4333,
|
|
"mean_token_accuracy": 0.194402377307415,
|
|
"num_tokens": 14495215.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.821519899368286,
|
|
"epoch": 6.812865497076023,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00017933540153617579,
|
|
"loss": 5.2686,
|
|
"mean_token_accuracy": 0.20770411491394042,
|
|
"num_tokens": 14506669.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.919344043731689,
|
|
"epoch": 6.818713450292398,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001789113487943888,
|
|
"loss": 5.3609,
|
|
"mean_token_accuracy": 0.19779593646526336,
|
|
"num_tokens": 14518578.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.94279842376709,
|
|
"epoch": 6.824561403508772,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00017848771308423835,
|
|
"loss": 5.4291,
|
|
"mean_token_accuracy": 0.189171701669693,
|
|
"num_tokens": 14531008.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 6.0453814506530765,
|
|
"epoch": 6.830409356725146,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00017806449624433384,
|
|
"loss": 5.477,
|
|
"mean_token_accuracy": 0.1865503594279289,
|
|
"num_tokens": 14543299.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.872481727600098,
|
|
"epoch": 6.83625730994152,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000177641700111467,
|
|
"loss": 5.3043,
|
|
"mean_token_accuracy": 0.20829666703939437,
|
|
"num_tokens": 14556102.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.972483444213867,
|
|
"epoch": 6.842105263157895,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000177219326520603,
|
|
"loss": 5.3904,
|
|
"mean_token_accuracy": 0.20064649730920792,
|
|
"num_tokens": 14569114.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.946630430221558,
|
|
"epoch": 6.847953216374269,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00017679737730487378,
|
|
"loss": 5.3532,
|
|
"mean_token_accuracy": 0.20192302912473678,
|
|
"num_tokens": 14582609.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.958604001998902,
|
|
"epoch": 6.853801169590644,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017637585429556912,
|
|
"loss": 5.3999,
|
|
"mean_token_accuracy": 0.19615365117788314,
|
|
"num_tokens": 14594590.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.886360740661621,
|
|
"epoch": 6.859649122807017,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00017595475932212908,
|
|
"loss": 5.3728,
|
|
"mean_token_accuracy": 0.1969984218478203,
|
|
"num_tokens": 14607475.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.882922840118408,
|
|
"epoch": 6.8654970760233915,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00017553409421213606,
|
|
"loss": 5.2758,
|
|
"mean_token_accuracy": 0.20855315625667573,
|
|
"num_tokens": 14620113.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.936597156524658,
|
|
"epoch": 6.871345029239766,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00017511386079130666,
|
|
"loss": 5.4089,
|
|
"mean_token_accuracy": 0.19885661602020263,
|
|
"num_tokens": 14632137.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.9268951416015625,
|
|
"epoch": 6.87719298245614,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017469406088348427,
|
|
"loss": 5.3572,
|
|
"mean_token_accuracy": 0.2003800556063652,
|
|
"num_tokens": 14643841.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.844141960144043,
|
|
"epoch": 6.883040935672515,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00017427469631063055,
|
|
"loss": 5.2927,
|
|
"mean_token_accuracy": 0.21007898449897766,
|
|
"num_tokens": 14655349.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.943072319030762,
|
|
"epoch": 6.888888888888889,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00017385576889281774,
|
|
"loss": 5.4141,
|
|
"mean_token_accuracy": 0.19330965727567673,
|
|
"num_tokens": 14668761.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.941866207122803,
|
|
"epoch": 6.894736842105263,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00017343728044822093,
|
|
"loss": 5.3829,
|
|
"mean_token_accuracy": 0.20177811980247498,
|
|
"num_tokens": 14680216.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.904266405105591,
|
|
"epoch": 6.900584795321637,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00017301923279310996,
|
|
"loss": 5.3251,
|
|
"mean_token_accuracy": 0.19742235243320466,
|
|
"num_tokens": 14691791.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.967219066619873,
|
|
"epoch": 6.906432748538012,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00017260162774184167,
|
|
"loss": 5.3985,
|
|
"mean_token_accuracy": 0.19997985661029816,
|
|
"num_tokens": 14704591.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.906804370880127,
|
|
"epoch": 6.912280701754386,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017218446710685176,
|
|
"loss": 5.2935,
|
|
"mean_token_accuracy": 0.20586687177419663,
|
|
"num_tokens": 14717469.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.9487861633300785,
|
|
"epoch": 6.918128654970761,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017176775269864763,
|
|
"loss": 5.3779,
|
|
"mean_token_accuracy": 0.19149671196937562,
|
|
"num_tokens": 14730121.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.925993871688843,
|
|
"epoch": 6.923976608187134,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001713514863257994,
|
|
"loss": 5.4003,
|
|
"mean_token_accuracy": 0.2024425521492958,
|
|
"num_tokens": 14743443.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.969432544708252,
|
|
"epoch": 6.9298245614035086,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017093566979493302,
|
|
"loss": 5.4297,
|
|
"mean_token_accuracy": 0.1959313780069351,
|
|
"num_tokens": 14757208.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.890073680877686,
|
|
"epoch": 6.935672514619883,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001705203049107223,
|
|
"loss": 5.3153,
|
|
"mean_token_accuracy": 0.19599432945251466,
|
|
"num_tokens": 14770964.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.914359855651855,
|
|
"epoch": 6.941520467836257,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00017010539347588023,
|
|
"loss": 5.4259,
|
|
"mean_token_accuracy": 0.19835536032915116,
|
|
"num_tokens": 14784042.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.860774278640747,
|
|
"epoch": 6.947368421052632,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016969093729115253,
|
|
"loss": 5.2827,
|
|
"mean_token_accuracy": 0.21147658228874205,
|
|
"num_tokens": 14797438.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.917108058929443,
|
|
"epoch": 6.953216374269006,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00016927693815530854,
|
|
"loss": 5.3596,
|
|
"mean_token_accuracy": 0.19992436617612838,
|
|
"num_tokens": 14808451.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.87166657447815,
|
|
"epoch": 6.95906432748538,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016886339786513428,
|
|
"loss": 5.3503,
|
|
"mean_token_accuracy": 0.20429475456476212,
|
|
"num_tokens": 14821479.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 6.020088481903076,
|
|
"epoch": 6.964912280701754,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001684503182154244,
|
|
"loss": 5.4043,
|
|
"mean_token_accuracy": 0.197114397585392,
|
|
"num_tokens": 14835394.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.858371877670288,
|
|
"epoch": 6.970760233918129,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016803770099897385,
|
|
"loss": 5.3635,
|
|
"mean_token_accuracy": 0.20254897475242614,
|
|
"num_tokens": 14847103.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.834583520889282,
|
|
"epoch": 6.976608187134503,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016762554800657132,
|
|
"loss": 5.2994,
|
|
"mean_token_accuracy": 0.20845072567462922,
|
|
"num_tokens": 14858950.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.9292731285095215,
|
|
"epoch": 6.982456140350877,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001672138610269901,
|
|
"loss": 5.3561,
|
|
"mean_token_accuracy": 0.20113305151462554,
|
|
"num_tokens": 14871272.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.918289613723755,
|
|
"epoch": 6.988304093567251,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00016680264184698135,
|
|
"loss": 5.3384,
|
|
"mean_token_accuracy": 0.2011785700917244,
|
|
"num_tokens": 14883260.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.901133489608765,
|
|
"epoch": 6.994152046783626,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016639189225126574,
|
|
"loss": 5.3509,
|
|
"mean_token_accuracy": 0.20175050497055053,
|
|
"num_tokens": 14895235.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.909830760955811,
|
|
"epoch": 7.0,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016598161402252596,
|
|
"loss": 5.4428,
|
|
"mean_token_accuracy": 0.18996424227952957,
|
|
"num_tokens": 14906388.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.987895488739014,
|
|
"epoch": 7.005847953216374,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016557180894139888,
|
|
"loss": 5.211,
|
|
"mean_token_accuracy": 0.20715796202421188,
|
|
"num_tokens": 14918467.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.853253889083862,
|
|
"epoch": 7.011695906432749,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016516247878646805,
|
|
"loss": 5.1273,
|
|
"mean_token_accuracy": 0.21750515103340148,
|
|
"num_tokens": 14931457.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.924554347991943,
|
|
"epoch": 7.017543859649122,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016475362533425574,
|
|
"loss": 5.2204,
|
|
"mean_token_accuracy": 0.20456665009260178,
|
|
"num_tokens": 14944639.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 7.017543859649122,
|
|
"eval_entropy": 5.902082860997293,
|
|
"eval_loss": 6.555347442626953,
|
|
"eval_mean_token_accuracy": 0.15144179733937116,
|
|
"eval_num_tokens": 14944639.0,
|
|
"eval_runtime": 2.2493,
|
|
"eval_samples_per_second": 1203.46,
|
|
"eval_steps_per_second": 150.71,
|
|
"step": 6000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 8540,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2.1642928222464e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|