3035 lines
82 KiB
JSON
3035 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.1971349717439874,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691970920562744,
|
|
"epoch": 0.000657116572479958,
|
|
"grad_norm": 11.375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.9102,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 18160.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69195909500122,
|
|
"epoch": 0.001314233144959916,
|
|
"grad_norm": 11.8125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.8426,
|
|
"mean_token_accuracy": 0.00034901105682365595,
|
|
"num_tokens": 34700.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.689266014099122,
|
|
"epoch": 0.001971349717439874,
|
|
"grad_norm": 8.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5846,
|
|
"mean_token_accuracy": 0.0499508892185986,
|
|
"num_tokens": 52678.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.562198734283447,
|
|
"epoch": 0.002628466289919832,
|
|
"grad_norm": 5.40625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.3172,
|
|
"mean_token_accuracy": 0.0605622336268425,
|
|
"num_tokens": 69607.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.345282936096192,
|
|
"epoch": 0.0032855828623997896,
|
|
"grad_norm": 3.71875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.0938,
|
|
"mean_token_accuracy": 0.06148592121899128,
|
|
"num_tokens": 88068.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.521661567687989,
|
|
"epoch": 0.003942699434879748,
|
|
"grad_norm": 3.171875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.9441,
|
|
"mean_token_accuracy": 0.06816631071269512,
|
|
"num_tokens": 105839.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.302499389648437,
|
|
"epoch": 0.004599816007359705,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.899,
|
|
"mean_token_accuracy": 0.0670669324696064,
|
|
"num_tokens": 125068.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.426428699493409,
|
|
"epoch": 0.005256932579839664,
|
|
"grad_norm": 2.765625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.8561,
|
|
"mean_token_accuracy": 0.06749101914465427,
|
|
"num_tokens": 143271.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.273355770111085,
|
|
"epoch": 0.0059140491523196215,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8657,
|
|
"mean_token_accuracy": 0.06502391099929809,
|
|
"num_tokens": 160886.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.348071479797364,
|
|
"epoch": 0.006571165724799579,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.7845,
|
|
"mean_token_accuracy": 0.06918984390795231,
|
|
"num_tokens": 179036.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.29029483795166,
|
|
"epoch": 0.007228282297279538,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.731,
|
|
"mean_token_accuracy": 0.07089065425097943,
|
|
"num_tokens": 195478.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.25726251602173,
|
|
"epoch": 0.007885398869759495,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.7119,
|
|
"mean_token_accuracy": 0.0690944753587246,
|
|
"num_tokens": 213003.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.302559661865235,
|
|
"epoch": 0.008542515442239454,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.6475,
|
|
"mean_token_accuracy": 0.07418905831873417,
|
|
"num_tokens": 230223.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.087755107879639,
|
|
"epoch": 0.00919963201471941,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.5049,
|
|
"mean_token_accuracy": 0.0840003713965416,
|
|
"num_tokens": 246555.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.160563468933105,
|
|
"epoch": 0.00985674858719937,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.4854,
|
|
"mean_token_accuracy": 0.07695499211549758,
|
|
"num_tokens": 262985.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.03568115234375,
|
|
"epoch": 0.010513865159679328,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.3528,
|
|
"mean_token_accuracy": 0.08797612711787224,
|
|
"num_tokens": 281701.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.050819873809814,
|
|
"epoch": 0.011170981732159285,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.3271,
|
|
"mean_token_accuracy": 0.08114923164248466,
|
|
"num_tokens": 298432.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.067263984680176,
|
|
"epoch": 0.011828098304639243,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.2657,
|
|
"mean_token_accuracy": 0.08178326860070229,
|
|
"num_tokens": 315559.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 9.996760272979737,
|
|
"epoch": 0.012485214877119202,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.174,
|
|
"mean_token_accuracy": 0.08735092431306839,
|
|
"num_tokens": 332959.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.032902812957763,
|
|
"epoch": 0.013142331449599158,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.1565,
|
|
"mean_token_accuracy": 0.07826860286295415,
|
|
"num_tokens": 349872.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.87295093536377,
|
|
"epoch": 0.013799448022079117,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.0779,
|
|
"mean_token_accuracy": 0.07845050990581512,
|
|
"num_tokens": 365952.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.94616346359253,
|
|
"epoch": 0.014456564594559075,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.9723,
|
|
"mean_token_accuracy": 0.0847774550318718,
|
|
"num_tokens": 382830.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.765446853637695,
|
|
"epoch": 0.015113681167039032,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.8959,
|
|
"mean_token_accuracy": 0.08348627090454101,
|
|
"num_tokens": 401729.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.696452522277832,
|
|
"epoch": 0.01577079773951899,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.7663,
|
|
"mean_token_accuracy": 0.0866998441517353,
|
|
"num_tokens": 419094.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.62384328842163,
|
|
"epoch": 0.01642791431199895,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7123,
|
|
"mean_token_accuracy": 0.08670503944158554,
|
|
"num_tokens": 438632.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.458528423309327,
|
|
"epoch": 0.017085030884478908,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.5991,
|
|
"mean_token_accuracy": 0.08925274014472961,
|
|
"num_tokens": 456025.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.395387935638428,
|
|
"epoch": 0.017742147456958863,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.5874,
|
|
"mean_token_accuracy": 0.08284339755773544,
|
|
"num_tokens": 473414.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.288764190673827,
|
|
"epoch": 0.01839926402943882,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.484,
|
|
"mean_token_accuracy": 0.08540036529302597,
|
|
"num_tokens": 490039.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.080249691009522,
|
|
"epoch": 0.01905638060191878,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.4185,
|
|
"mean_token_accuracy": 0.09069929495453835,
|
|
"num_tokens": 506896.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.990424156188965,
|
|
"epoch": 0.01971349717439874,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.3134,
|
|
"mean_token_accuracy": 0.0939207799732685,
|
|
"num_tokens": 523544.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.895142555236816,
|
|
"epoch": 0.020370613746878697,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.3193,
|
|
"mean_token_accuracy": 0.0861300639808178,
|
|
"num_tokens": 540240.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.84927806854248,
|
|
"epoch": 0.021027730319358656,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.3388,
|
|
"mean_token_accuracy": 0.08859094232320786,
|
|
"num_tokens": 558460.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.804749298095704,
|
|
"epoch": 0.02168484689183861,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.3132,
|
|
"mean_token_accuracy": 0.08978721722960473,
|
|
"num_tokens": 577356.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.755771827697753,
|
|
"epoch": 0.02234196346431857,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.3233,
|
|
"mean_token_accuracy": 0.08325318172574044,
|
|
"num_tokens": 594702.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.61918659210205,
|
|
"epoch": 0.022999080036798528,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.2927,
|
|
"mean_token_accuracy": 0.08967687636613846,
|
|
"num_tokens": 612005.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.692005062103272,
|
|
"epoch": 0.023656196609278486,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.2979,
|
|
"mean_token_accuracy": 0.08311463445425034,
|
|
"num_tokens": 629271.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.543895149230957,
|
|
"epoch": 0.024313313181758445,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.2711,
|
|
"mean_token_accuracy": 0.09236689656972885,
|
|
"num_tokens": 646273.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.623021125793457,
|
|
"epoch": 0.024970429754238403,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.2622,
|
|
"mean_token_accuracy": 0.08237618803977967,
|
|
"num_tokens": 663263.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.503989696502686,
|
|
"epoch": 0.02562754632671836,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.1725,
|
|
"mean_token_accuracy": 0.09226469621062279,
|
|
"num_tokens": 681032.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.496358585357665,
|
|
"epoch": 0.026284662899198317,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.2157,
|
|
"mean_token_accuracy": 0.08867634385824204,
|
|
"num_tokens": 699353.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.597747421264648,
|
|
"epoch": 0.026941779471678275,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.2941,
|
|
"mean_token_accuracy": 0.08325276263058186,
|
|
"num_tokens": 717777.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.493968677520751,
|
|
"epoch": 0.027598896044158234,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.2395,
|
|
"mean_token_accuracy": 0.08642884120345115,
|
|
"num_tokens": 733831.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.45910186767578,
|
|
"epoch": 0.028256012616638192,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.202,
|
|
"mean_token_accuracy": 0.0875705562531948,
|
|
"num_tokens": 750057.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.502654266357421,
|
|
"epoch": 0.02891312918911815,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.2335,
|
|
"mean_token_accuracy": 0.08846379294991494,
|
|
"num_tokens": 767590.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.471471691131592,
|
|
"epoch": 0.029570245761598106,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.1997,
|
|
"mean_token_accuracy": 0.0836844079196453,
|
|
"num_tokens": 784496.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.473236179351806,
|
|
"epoch": 0.030227362334078064,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.22,
|
|
"mean_token_accuracy": 0.08451615646481514,
|
|
"num_tokens": 802589.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.466453552246094,
|
|
"epoch": 0.030884478906558023,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.2459,
|
|
"mean_token_accuracy": 0.08304352685809135,
|
|
"num_tokens": 820375.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.500118160247803,
|
|
"epoch": 0.03154159547903798,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.1952,
|
|
"mean_token_accuracy": 0.08304070755839348,
|
|
"num_tokens": 838658.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.381299304962159,
|
|
"epoch": 0.03219871205151794,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.205,
|
|
"mean_token_accuracy": 0.0860319972038269,
|
|
"num_tokens": 856534.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.390718936920166,
|
|
"epoch": 0.0328558286239979,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.1696,
|
|
"mean_token_accuracy": 0.09368039518594742,
|
|
"num_tokens": 872813.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.36027283668518,
|
|
"epoch": 0.03351294519647786,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.1428,
|
|
"mean_token_accuracy": 0.09875724688172341,
|
|
"num_tokens": 891675.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.41018009185791,
|
|
"epoch": 0.034170061768957816,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.1972,
|
|
"mean_token_accuracy": 0.08137553408741952,
|
|
"num_tokens": 907440.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.448321056365966,
|
|
"epoch": 0.034827178341437774,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.2255,
|
|
"mean_token_accuracy": 0.08277432322502136,
|
|
"num_tokens": 923909.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.386664962768554,
|
|
"epoch": 0.035484294913917726,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.1077,
|
|
"mean_token_accuracy": 0.09150323122739792,
|
|
"num_tokens": 940658.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.34205665588379,
|
|
"epoch": 0.036141411486397684,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.1535,
|
|
"mean_token_accuracy": 0.08561521768569946,
|
|
"num_tokens": 959095.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.375339221954345,
|
|
"epoch": 0.03679852805887764,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.1774,
|
|
"mean_token_accuracy": 0.08340813145041466,
|
|
"num_tokens": 976656.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.368755626678468,
|
|
"epoch": 0.0374556446313576,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.2048,
|
|
"mean_token_accuracy": 0.08358296155929565,
|
|
"num_tokens": 994326.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.306324100494384,
|
|
"epoch": 0.03811276120383756,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.1147,
|
|
"mean_token_accuracy": 0.0932599276304245,
|
|
"num_tokens": 1011859.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.31154556274414,
|
|
"epoch": 0.03876987777631752,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.142,
|
|
"mean_token_accuracy": 0.08438249453902244,
|
|
"num_tokens": 1028329.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.38582706451416,
|
|
"epoch": 0.03942699434879748,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.1226,
|
|
"mean_token_accuracy": 0.08788742870092392,
|
|
"num_tokens": 1045519.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.42501277923584,
|
|
"epoch": 0.040084110921277435,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.1104,
|
|
"mean_token_accuracy": 0.09161722511053086,
|
|
"num_tokens": 1063832.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.180773544311524,
|
|
"epoch": 0.040741227493757394,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.0692,
|
|
"mean_token_accuracy": 0.09160381332039833,
|
|
"num_tokens": 1081062.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.319191455841064,
|
|
"epoch": 0.04139834406623735,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.1206,
|
|
"mean_token_accuracy": 0.08650392666459084,
|
|
"num_tokens": 1100139.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.339826774597167,
|
|
"epoch": 0.04205546063871731,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.0878,
|
|
"mean_token_accuracy": 0.08833744451403618,
|
|
"num_tokens": 1117100.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.358532238006593,
|
|
"epoch": 0.04271257721119727,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.1339,
|
|
"mean_token_accuracy": 0.08414132967591285,
|
|
"num_tokens": 1136607.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.360890769958496,
|
|
"epoch": 0.04336969378367722,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.1636,
|
|
"mean_token_accuracy": 0.08001711294054985,
|
|
"num_tokens": 1154463.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.302077484130859,
|
|
"epoch": 0.04402681035615718,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.0492,
|
|
"mean_token_accuracy": 0.08888905569911003,
|
|
"num_tokens": 1172540.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.284979152679444,
|
|
"epoch": 0.04468392692863714,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.089,
|
|
"mean_token_accuracy": 0.08480592146515846,
|
|
"num_tokens": 1189393.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.305260276794433,
|
|
"epoch": 0.0453410435011171,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.0804,
|
|
"mean_token_accuracy": 0.087804014980793,
|
|
"num_tokens": 1205232.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.235641956329346,
|
|
"epoch": 0.045998160073597055,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.0386,
|
|
"mean_token_accuracy": 0.08915610313415527,
|
|
"num_tokens": 1222020.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.296236038208008,
|
|
"epoch": 0.046655276646077014,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.0122,
|
|
"mean_token_accuracy": 0.09561690092086791,
|
|
"num_tokens": 1239201.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.312324523925781,
|
|
"epoch": 0.04731239321855697,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.0528,
|
|
"mean_token_accuracy": 0.09137414395809174,
|
|
"num_tokens": 1256715.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.276749324798583,
|
|
"epoch": 0.04796950979103693,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.0682,
|
|
"mean_token_accuracy": 0.08286751359701157,
|
|
"num_tokens": 1273800.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.162324810028077,
|
|
"epoch": 0.04862662636351689,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.0052,
|
|
"mean_token_accuracy": 0.09242228120565414,
|
|
"num_tokens": 1291181.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.235886669158935,
|
|
"epoch": 0.04928374293599685,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.088,
|
|
"mean_token_accuracy": 0.08434378504753112,
|
|
"num_tokens": 1309264.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.252655410766602,
|
|
"epoch": 0.049940859508476806,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.9823,
|
|
"mean_token_accuracy": 0.08949511088430881,
|
|
"num_tokens": 1327117.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.148046970367432,
|
|
"epoch": 0.050597976080956765,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.0288,
|
|
"mean_token_accuracy": 0.08982144668698311,
|
|
"num_tokens": 1343972.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.245464324951172,
|
|
"epoch": 0.05125509265343672,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.9572,
|
|
"mean_token_accuracy": 0.08928473889827729,
|
|
"num_tokens": 1361432.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.094884061813355,
|
|
"epoch": 0.051912209225916675,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.9669,
|
|
"mean_token_accuracy": 0.09419442862272262,
|
|
"num_tokens": 1378191.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.211234951019287,
|
|
"epoch": 0.052569325798396634,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.9813,
|
|
"mean_token_accuracy": 0.08906673565506935,
|
|
"num_tokens": 1395872.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.18825888633728,
|
|
"epoch": 0.05322644237087659,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.9578,
|
|
"mean_token_accuracy": 0.09145129919052124,
|
|
"num_tokens": 1412816.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.134449100494384,
|
|
"epoch": 0.05388355894335655,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.8935,
|
|
"mean_token_accuracy": 0.10044471621513366,
|
|
"num_tokens": 1430644.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.148541164398193,
|
|
"epoch": 0.05454067551583651,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.9432,
|
|
"mean_token_accuracy": 0.09160158559679984,
|
|
"num_tokens": 1448021.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.23169984817505,
|
|
"epoch": 0.05519779208831647,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.9585,
|
|
"mean_token_accuracy": 0.09313838183879852,
|
|
"num_tokens": 1466428.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.167637252807618,
|
|
"epoch": 0.055854908660796426,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.8798,
|
|
"mean_token_accuracy": 0.09942131265997886,
|
|
"num_tokens": 1484554.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.104038047790528,
|
|
"epoch": 0.056512025233276385,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.8905,
|
|
"mean_token_accuracy": 0.09655392467975617,
|
|
"num_tokens": 1500452.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.07980990409851,
|
|
"epoch": 0.05716914180575634,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.9718,
|
|
"mean_token_accuracy": 0.0888563945889473,
|
|
"num_tokens": 1518482.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.099183750152587,
|
|
"epoch": 0.0578262583782363,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.9304,
|
|
"mean_token_accuracy": 0.0907039225101471,
|
|
"num_tokens": 1536107.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.13019618988037,
|
|
"epoch": 0.05848337495071626,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.9209,
|
|
"mean_token_accuracy": 0.08752450793981552,
|
|
"num_tokens": 1553722.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.061606645584106,
|
|
"epoch": 0.05914049152319621,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.878,
|
|
"mean_token_accuracy": 0.09788650721311569,
|
|
"num_tokens": 1571959.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.092618274688721,
|
|
"epoch": 0.05979760809567617,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.965,
|
|
"mean_token_accuracy": 0.08797019943594933,
|
|
"num_tokens": 1588595.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.051505088806152,
|
|
"epoch": 0.06045472466815613,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.8329,
|
|
"mean_token_accuracy": 0.08995860144495964,
|
|
"num_tokens": 1605185.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.061400985717773,
|
|
"epoch": 0.06111184124063609,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.8334,
|
|
"mean_token_accuracy": 0.0945149227976799,
|
|
"num_tokens": 1622966.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.077755498886109,
|
|
"epoch": 0.061768957813116046,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.8267,
|
|
"mean_token_accuracy": 0.09398441277444362,
|
|
"num_tokens": 1639697.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.02926688194275,
|
|
"epoch": 0.062426074385596005,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.851,
|
|
"mean_token_accuracy": 0.09462198317050934,
|
|
"num_tokens": 1656342.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.030053997039795,
|
|
"epoch": 0.06308319095807596,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.9004,
|
|
"mean_token_accuracy": 0.09348834380507469,
|
|
"num_tokens": 1673829.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.142562198638917,
|
|
"epoch": 0.06374030753055591,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.9383,
|
|
"mean_token_accuracy": 0.08797306269407272,
|
|
"num_tokens": 1691938.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.987461423873901,
|
|
"epoch": 0.06439742410303588,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.8913,
|
|
"mean_token_accuracy": 0.09134467840194702,
|
|
"num_tokens": 1708545.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.951012325286865,
|
|
"epoch": 0.06505454067551583,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.8445,
|
|
"mean_token_accuracy": 0.09492667466402054,
|
|
"num_tokens": 1725199.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.079988193511962,
|
|
"epoch": 0.0657116572479958,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.8607,
|
|
"mean_token_accuracy": 0.09226195812225342,
|
|
"num_tokens": 1742893.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.04035177230835,
|
|
"epoch": 0.06636877382047575,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.8558,
|
|
"mean_token_accuracy": 0.09375119879841805,
|
|
"num_tokens": 1759540.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.108158254623413,
|
|
"epoch": 0.06702589039295571,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.8746,
|
|
"mean_token_accuracy": 0.0892824463546276,
|
|
"num_tokens": 1776617.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.932979726791382,
|
|
"epoch": 0.06768300696543567,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.8364,
|
|
"mean_token_accuracy": 0.0924128919839859,
|
|
"num_tokens": 1795065.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.049097728729247,
|
|
"epoch": 0.06834012353791563,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.8169,
|
|
"mean_token_accuracy": 0.08902051150798798,
|
|
"num_tokens": 1811477.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.031516742706298,
|
|
"epoch": 0.06899724011039558,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.8404,
|
|
"mean_token_accuracy": 0.09564107805490493,
|
|
"num_tokens": 1829272.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.003777980804443,
|
|
"epoch": 0.06965435668287555,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.8549,
|
|
"mean_token_accuracy": 0.08890713900327682,
|
|
"num_tokens": 1845882.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.024421834945679,
|
|
"epoch": 0.0703114732553555,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.8113,
|
|
"mean_token_accuracy": 0.09261564016342164,
|
|
"num_tokens": 1861000.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.050934314727783,
|
|
"epoch": 0.07096858982783545,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.8551,
|
|
"mean_token_accuracy": 0.08385444208979606,
|
|
"num_tokens": 1876979.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.997040605545044,
|
|
"epoch": 0.07162570640031542,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.8555,
|
|
"mean_token_accuracy": 0.08732549175620079,
|
|
"num_tokens": 1895191.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.934540605545044,
|
|
"epoch": 0.07228282297279537,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.7701,
|
|
"mean_token_accuracy": 0.08872183412313461,
|
|
"num_tokens": 1912466.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.969846534729004,
|
|
"epoch": 0.07293993954527533,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.7246,
|
|
"mean_token_accuracy": 0.09994208216667175,
|
|
"num_tokens": 1932285.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.008286952972412,
|
|
"epoch": 0.07359705611775529,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.7973,
|
|
"mean_token_accuracy": 0.08675110414624214,
|
|
"num_tokens": 1948437.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.939477729797363,
|
|
"epoch": 0.07425417269023525,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.8076,
|
|
"mean_token_accuracy": 0.08641846030950547,
|
|
"num_tokens": 1964579.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.910443115234375,
|
|
"epoch": 0.0749112892627152,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.7054,
|
|
"mean_token_accuracy": 0.10115818828344345,
|
|
"num_tokens": 1981247.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.956994104385376,
|
|
"epoch": 0.07556840583519517,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.7938,
|
|
"mean_token_accuracy": 0.09043643400073051,
|
|
"num_tokens": 1999031.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.981093311309815,
|
|
"epoch": 0.07622552240767512,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.7212,
|
|
"mean_token_accuracy": 0.09121812731027604,
|
|
"num_tokens": 2015715.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.911635208129883,
|
|
"epoch": 0.07688263898015509,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.7238,
|
|
"mean_token_accuracy": 0.09262361451983452,
|
|
"num_tokens": 2032235.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.910367107391357,
|
|
"epoch": 0.07753975555263504,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.7125,
|
|
"mean_token_accuracy": 0.09460373297333717,
|
|
"num_tokens": 2050378.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.8812439918518065,
|
|
"epoch": 0.078196872125115,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.7713,
|
|
"mean_token_accuracy": 0.08955961838364601,
|
|
"num_tokens": 2067713.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.799747514724731,
|
|
"epoch": 0.07885398869759495,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.6174,
|
|
"mean_token_accuracy": 0.09769630283117295,
|
|
"num_tokens": 2086602.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.92856650352478,
|
|
"epoch": 0.0795111052700749,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.5796,
|
|
"mean_token_accuracy": 0.09621628671884537,
|
|
"num_tokens": 2103142.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.727544260025025,
|
|
"epoch": 0.08016822184255487,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.5822,
|
|
"mean_token_accuracy": 0.09833648353815079,
|
|
"num_tokens": 2120792.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.851658725738526,
|
|
"epoch": 0.08082533841503482,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.6633,
|
|
"mean_token_accuracy": 0.1018780030310154,
|
|
"num_tokens": 2137688.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.81537652015686,
|
|
"epoch": 0.08148245498751479,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.7289,
|
|
"mean_token_accuracy": 0.0900080993771553,
|
|
"num_tokens": 2155353.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.8868913650512695,
|
|
"epoch": 0.08213957155999474,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.6782,
|
|
"mean_token_accuracy": 0.09196878299117088,
|
|
"num_tokens": 2173999.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.794561433792114,
|
|
"epoch": 0.0827966881324747,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.6572,
|
|
"mean_token_accuracy": 0.0915692389011383,
|
|
"num_tokens": 2190657.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.878007602691651,
|
|
"epoch": 0.08345380470495466,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.6938,
|
|
"mean_token_accuracy": 0.09030900672078132,
|
|
"num_tokens": 2207512.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.812237787246704,
|
|
"epoch": 0.08411092127743462,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.6791,
|
|
"mean_token_accuracy": 0.09167201295495034,
|
|
"num_tokens": 2224911.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.797879552841186,
|
|
"epoch": 0.08476803784991457,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.6568,
|
|
"mean_token_accuracy": 0.0889101818203926,
|
|
"num_tokens": 2244030.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.85651330947876,
|
|
"epoch": 0.08542515442239454,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.6423,
|
|
"mean_token_accuracy": 0.0900063842535019,
|
|
"num_tokens": 2260550.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.780899906158448,
|
|
"epoch": 0.08608227099487449,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.7007,
|
|
"mean_token_accuracy": 0.0889291986823082,
|
|
"num_tokens": 2277732.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.779685974121094,
|
|
"epoch": 0.08673938756735444,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.58,
|
|
"mean_token_accuracy": 0.09410673379898071,
|
|
"num_tokens": 2294074.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.794468116760254,
|
|
"epoch": 0.08739650413983441,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.6426,
|
|
"mean_token_accuracy": 0.09421529322862625,
|
|
"num_tokens": 2309118.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.813050365447998,
|
|
"epoch": 0.08805362071231436,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.6292,
|
|
"mean_token_accuracy": 0.09069699347019196,
|
|
"num_tokens": 2327193.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.743873262405396,
|
|
"epoch": 0.08871073728479432,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.5902,
|
|
"mean_token_accuracy": 0.09050673618912697,
|
|
"num_tokens": 2344357.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.736028623580933,
|
|
"epoch": 0.08936785385727428,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.5732,
|
|
"mean_token_accuracy": 0.09712009504437447,
|
|
"num_tokens": 2361933.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.702030611038208,
|
|
"epoch": 0.09002497042975424,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.5759,
|
|
"mean_token_accuracy": 0.09285529553890229,
|
|
"num_tokens": 2377774.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.651546764373779,
|
|
"epoch": 0.0906820870022342,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.4294,
|
|
"mean_token_accuracy": 0.11124403774738312,
|
|
"num_tokens": 2395363.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.507357358932495,
|
|
"epoch": 0.09133920357471416,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.4995,
|
|
"mean_token_accuracy": 0.1047697126865387,
|
|
"num_tokens": 2413270.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.713213491439819,
|
|
"epoch": 0.09199632014719411,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.5303,
|
|
"mean_token_accuracy": 0.09587173163890839,
|
|
"num_tokens": 2432077.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.674976873397827,
|
|
"epoch": 0.09265343671967408,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.6357,
|
|
"mean_token_accuracy": 0.09107815846800804,
|
|
"num_tokens": 2449541.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.67608757019043,
|
|
"epoch": 0.09331055329215403,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.5318,
|
|
"mean_token_accuracy": 0.09878497794270516,
|
|
"num_tokens": 2466375.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.702895450592041,
|
|
"epoch": 0.09396766986463398,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.6132,
|
|
"mean_token_accuracy": 0.0879031240940094,
|
|
"num_tokens": 2483793.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.7163526058197025,
|
|
"epoch": 0.09462478643711394,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.5686,
|
|
"mean_token_accuracy": 0.0924137331545353,
|
|
"num_tokens": 2500689.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.652881479263305,
|
|
"epoch": 0.0952819030095939,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.4627,
|
|
"mean_token_accuracy": 0.10279997587203979,
|
|
"num_tokens": 2517867.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.693321084976196,
|
|
"epoch": 0.09593901958207386,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.5954,
|
|
"mean_token_accuracy": 0.09244397953152657,
|
|
"num_tokens": 2536356.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.569602060317993,
|
|
"epoch": 0.09659613615455381,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.4796,
|
|
"mean_token_accuracy": 0.09813716039061546,
|
|
"num_tokens": 2552821.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.72858157157898,
|
|
"epoch": 0.09725325272703378,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.595,
|
|
"mean_token_accuracy": 0.09159181639552116,
|
|
"num_tokens": 2570098.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.5410655498504635,
|
|
"epoch": 0.09791036929951373,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.3465,
|
|
"mean_token_accuracy": 0.11115500703454018,
|
|
"num_tokens": 2587508.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.601456546783448,
|
|
"epoch": 0.0985674858719937,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.5464,
|
|
"mean_token_accuracy": 0.09635312631726264,
|
|
"num_tokens": 2604845.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.641898345947266,
|
|
"epoch": 0.09922460244447365,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.5835,
|
|
"mean_token_accuracy": 0.0863256387412548,
|
|
"num_tokens": 2623281.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.665192556381226,
|
|
"epoch": 0.09988171901695361,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.5832,
|
|
"mean_token_accuracy": 0.09262030571699142,
|
|
"num_tokens": 2641315.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.523155879974365,
|
|
"epoch": 0.10053883558943356,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.4307,
|
|
"mean_token_accuracy": 0.11434739604592323,
|
|
"num_tokens": 2658404.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.55363826751709,
|
|
"epoch": 0.10119595216191353,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.4513,
|
|
"mean_token_accuracy": 0.10143931359052658,
|
|
"num_tokens": 2675980.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.607959318161011,
|
|
"epoch": 0.10185306873439348,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.4692,
|
|
"mean_token_accuracy": 0.10197424218058586,
|
|
"num_tokens": 2694290.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.581793880462646,
|
|
"epoch": 0.10251018530687343,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.5235,
|
|
"mean_token_accuracy": 0.09472638443112373,
|
|
"num_tokens": 2710661.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.651349306106567,
|
|
"epoch": 0.1031673018793534,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.523,
|
|
"mean_token_accuracy": 0.09317268952727317,
|
|
"num_tokens": 2728660.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.531972074508667,
|
|
"epoch": 0.10382441845183335,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.4728,
|
|
"mean_token_accuracy": 0.09988932237029076,
|
|
"num_tokens": 2745130.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.562900590896606,
|
|
"epoch": 0.10448153502431332,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.4529,
|
|
"mean_token_accuracy": 0.0973394624888897,
|
|
"num_tokens": 2762001.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.580878829956054,
|
|
"epoch": 0.10513865159679327,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.4957,
|
|
"mean_token_accuracy": 0.09426558315753937,
|
|
"num_tokens": 2778634.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.577242612838745,
|
|
"epoch": 0.10579576816927323,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.5104,
|
|
"mean_token_accuracy": 0.0927124135196209,
|
|
"num_tokens": 2795207.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.537419080734253,
|
|
"epoch": 0.10645288474175318,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.3978,
|
|
"mean_token_accuracy": 0.10554074496030807,
|
|
"num_tokens": 2813506.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.516330528259277,
|
|
"epoch": 0.10711000131423315,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.4253,
|
|
"mean_token_accuracy": 0.0940089389681816,
|
|
"num_tokens": 2829732.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.539044332504273,
|
|
"epoch": 0.1077671178867131,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.46,
|
|
"mean_token_accuracy": 0.09880606159567833,
|
|
"num_tokens": 2846436.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.575464391708374,
|
|
"epoch": 0.10842423445919307,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.491,
|
|
"mean_token_accuracy": 0.09248777329921723,
|
|
"num_tokens": 2864719.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.577031135559082,
|
|
"epoch": 0.10908135103167302,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.4707,
|
|
"mean_token_accuracy": 0.09447416216135025,
|
|
"num_tokens": 2881566.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.481828498840332,
|
|
"epoch": 0.10973846760415297,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.3937,
|
|
"mean_token_accuracy": 0.09647827595472336,
|
|
"num_tokens": 2899034.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.493137073516846,
|
|
"epoch": 0.11039558417663294,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.4897,
|
|
"mean_token_accuracy": 0.09726456999778747,
|
|
"num_tokens": 2915773.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.474357795715332,
|
|
"epoch": 0.11105270074911289,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.4534,
|
|
"mean_token_accuracy": 0.08885478526353836,
|
|
"num_tokens": 2933718.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.565765857696533,
|
|
"epoch": 0.11170981732159285,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.4222,
|
|
"mean_token_accuracy": 0.09584522470831872,
|
|
"num_tokens": 2951690.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.415678358078003,
|
|
"epoch": 0.1123669338940728,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.4284,
|
|
"mean_token_accuracy": 0.10300374701619149,
|
|
"num_tokens": 2969331.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.552033138275147,
|
|
"epoch": 0.11302405046655277,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.4741,
|
|
"mean_token_accuracy": 0.09474552050232887,
|
|
"num_tokens": 2985229.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.461298561096191,
|
|
"epoch": 0.11368116703903272,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.4336,
|
|
"mean_token_accuracy": 0.08998932093381881,
|
|
"num_tokens": 3001075.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.452679061889649,
|
|
"epoch": 0.11433828361151269,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.3945,
|
|
"mean_token_accuracy": 0.09610669687390327,
|
|
"num_tokens": 3018095.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.448093366622925,
|
|
"epoch": 0.11499540018399264,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.3951,
|
|
"mean_token_accuracy": 0.1022491253912449,
|
|
"num_tokens": 3035726.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.428170299530029,
|
|
"epoch": 0.1156525167564726,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.3273,
|
|
"mean_token_accuracy": 0.10195292830467224,
|
|
"num_tokens": 3052054.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.476611471176147,
|
|
"epoch": 0.11630963332895256,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.4027,
|
|
"mean_token_accuracy": 0.09874287694692611,
|
|
"num_tokens": 3069350.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.442331171035766,
|
|
"epoch": 0.11696674990143252,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.3826,
|
|
"mean_token_accuracy": 0.09821617379784583,
|
|
"num_tokens": 3085891.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.453686094284057,
|
|
"epoch": 0.11762386647391247,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.4196,
|
|
"mean_token_accuracy": 0.09896242916584015,
|
|
"num_tokens": 3103571.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.417425680160522,
|
|
"epoch": 0.11828098304639242,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.3775,
|
|
"mean_token_accuracy": 0.1023717314004898,
|
|
"num_tokens": 3120883.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.509283113479614,
|
|
"epoch": 0.11893809961887239,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.4598,
|
|
"mean_token_accuracy": 0.09272672832012177,
|
|
"num_tokens": 3140137.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.376467370986939,
|
|
"epoch": 0.11959521619135234,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.3768,
|
|
"mean_token_accuracy": 0.09941514804959298,
|
|
"num_tokens": 3157118.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.486584043502807,
|
|
"epoch": 0.1202523327638323,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.4627,
|
|
"mean_token_accuracy": 0.089472546428442,
|
|
"num_tokens": 3174967.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.364890909194946,
|
|
"epoch": 0.12090944933631226,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.3251,
|
|
"mean_token_accuracy": 0.10790883004665375,
|
|
"num_tokens": 3193791.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.424156236648559,
|
|
"epoch": 0.12156656590879222,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.3888,
|
|
"mean_token_accuracy": 0.10138912200927734,
|
|
"num_tokens": 3212807.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.483264064788818,
|
|
"epoch": 0.12222368248127218,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.4603,
|
|
"mean_token_accuracy": 0.09148271009325981,
|
|
"num_tokens": 3229945.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.443746614456177,
|
|
"epoch": 0.12288079905375214,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.4491,
|
|
"mean_token_accuracy": 0.0934921719133854,
|
|
"num_tokens": 3248333.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.459245109558106,
|
|
"epoch": 0.12353791562623209,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.3967,
|
|
"mean_token_accuracy": 0.09736947789788246,
|
|
"num_tokens": 3266344.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.429486465454102,
|
|
"epoch": 0.12419503219871206,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.4148,
|
|
"mean_token_accuracy": 0.10019433051347733,
|
|
"num_tokens": 3283346.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.285105895996094,
|
|
"epoch": 0.12485214877119201,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.1709,
|
|
"mean_token_accuracy": 0.11596085354685784,
|
|
"num_tokens": 3301983.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.412484455108642,
|
|
"epoch": 0.12550926534367196,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.3552,
|
|
"mean_token_accuracy": 0.10020338520407676,
|
|
"num_tokens": 3320075.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.315499114990234,
|
|
"epoch": 0.12616638191615193,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.291,
|
|
"mean_token_accuracy": 0.09955885633826256,
|
|
"num_tokens": 3337782.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.315990972518921,
|
|
"epoch": 0.1268234984886319,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.3124,
|
|
"mean_token_accuracy": 0.09781972020864486,
|
|
"num_tokens": 3355114.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.252899885177612,
|
|
"epoch": 0.12748061506111183,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.1967,
|
|
"mean_token_accuracy": 0.10971798375248909,
|
|
"num_tokens": 3372664.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.372587108612061,
|
|
"epoch": 0.1281377316335918,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.2742,
|
|
"mean_token_accuracy": 0.10190480425953866,
|
|
"num_tokens": 3389687.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.19590539932251,
|
|
"epoch": 0.12879484820607176,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.2387,
|
|
"mean_token_accuracy": 0.10876442939043045,
|
|
"num_tokens": 3406613.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.346618175506592,
|
|
"epoch": 0.12945196477855173,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.4155,
|
|
"mean_token_accuracy": 0.09423026889562607,
|
|
"num_tokens": 3425530.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.372848224639893,
|
|
"epoch": 0.13010908135103166,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.3894,
|
|
"mean_token_accuracy": 0.09545382708311081,
|
|
"num_tokens": 3442988.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.38320107460022,
|
|
"epoch": 0.13076619792351163,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.3628,
|
|
"mean_token_accuracy": 0.09607246294617652,
|
|
"num_tokens": 3460763.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.417674016952515,
|
|
"epoch": 0.1314233144959916,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.3943,
|
|
"mean_token_accuracy": 0.09602649509906769,
|
|
"num_tokens": 3479425.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.343002462387085,
|
|
"epoch": 0.13208043106847156,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 7.3564,
|
|
"mean_token_accuracy": 0.09569800049066543,
|
|
"num_tokens": 3497403.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.324274969100952,
|
|
"epoch": 0.1327375476409515,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 7.2737,
|
|
"mean_token_accuracy": 0.09939010664820672,
|
|
"num_tokens": 3514376.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.334035348892212,
|
|
"epoch": 0.13339466421343146,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 7.2586,
|
|
"mean_token_accuracy": 0.10601741299033166,
|
|
"num_tokens": 3532288.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.304716682434082,
|
|
"epoch": 0.13405178078591143,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 7.3554,
|
|
"mean_token_accuracy": 0.09575305506587029,
|
|
"num_tokens": 3549354.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.30626745223999,
|
|
"epoch": 0.13470889735839137,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 7.3139,
|
|
"mean_token_accuracy": 0.09804008454084397,
|
|
"num_tokens": 3569161.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.337455797195434,
|
|
"epoch": 0.13536601393087133,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 7.385,
|
|
"mean_token_accuracy": 0.09783728495240211,
|
|
"num_tokens": 3586165.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.395480298995972,
|
|
"epoch": 0.1360231305033513,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 7.3912,
|
|
"mean_token_accuracy": 0.09173386469483376,
|
|
"num_tokens": 3604938.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.343952703475952,
|
|
"epoch": 0.13668024707583126,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 7.3583,
|
|
"mean_token_accuracy": 0.09979218170046807,
|
|
"num_tokens": 3621850.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.169143867492676,
|
|
"epoch": 0.1373373636483112,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 7.1837,
|
|
"mean_token_accuracy": 0.10983845517039299,
|
|
"num_tokens": 3641860.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.344118547439575,
|
|
"epoch": 0.13799448022079117,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 7.2699,
|
|
"mean_token_accuracy": 0.09669720306992531,
|
|
"num_tokens": 3659086.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.209481334686279,
|
|
"epoch": 0.13865159679327113,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 7.3641,
|
|
"mean_token_accuracy": 0.09310299232602119,
|
|
"num_tokens": 3676971.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.406137371063233,
|
|
"epoch": 0.1393087133657511,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 7.2983,
|
|
"mean_token_accuracy": 0.09514517113566398,
|
|
"num_tokens": 3696163.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.265346097946167,
|
|
"epoch": 0.13996582993823103,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 7.3231,
|
|
"mean_token_accuracy": 0.09595919102430343,
|
|
"num_tokens": 3713140.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.286369800567627,
|
|
"epoch": 0.140622946510711,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 7.2661,
|
|
"mean_token_accuracy": 0.09544462934136391,
|
|
"num_tokens": 3730600.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.2847730159759525,
|
|
"epoch": 0.14128006308319097,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 7.3234,
|
|
"mean_token_accuracy": 0.10120683461427689,
|
|
"num_tokens": 3747866.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.184893703460693,
|
|
"epoch": 0.1419371796556709,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 7.1738,
|
|
"mean_token_accuracy": 0.11132644340395928,
|
|
"num_tokens": 3765444.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.333232975006103,
|
|
"epoch": 0.14259429622815087,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 7.3202,
|
|
"mean_token_accuracy": 0.09391410201787949,
|
|
"num_tokens": 3781970.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.304718732833862,
|
|
"epoch": 0.14325141280063083,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 7.2165,
|
|
"mean_token_accuracy": 0.10574242919683456,
|
|
"num_tokens": 3799884.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.2470235347747805,
|
|
"epoch": 0.1439085293731108,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 7.261,
|
|
"mean_token_accuracy": 0.09376102611422539,
|
|
"num_tokens": 3816300.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.350685548782349,
|
|
"epoch": 0.14456564594559074,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 7.3215,
|
|
"mean_token_accuracy": 0.09178780242800713,
|
|
"num_tokens": 3835268.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.30710916519165,
|
|
"epoch": 0.1452227625180707,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 7.2716,
|
|
"mean_token_accuracy": 0.10289521887898445,
|
|
"num_tokens": 3852209.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.262703132629395,
|
|
"epoch": 0.14587987909055067,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 7.298,
|
|
"mean_token_accuracy": 0.09492710009217262,
|
|
"num_tokens": 3869024.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.169306993484497,
|
|
"epoch": 0.14653699566303063,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 7.1996,
|
|
"mean_token_accuracy": 0.1061355285346508,
|
|
"num_tokens": 3886897.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.180309152603149,
|
|
"epoch": 0.14719411223551057,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 7.1839,
|
|
"mean_token_accuracy": 0.10160853713750839,
|
|
"num_tokens": 3904397.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.284969186782837,
|
|
"epoch": 0.14785122880799054,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 7.2364,
|
|
"mean_token_accuracy": 0.09974497929215431,
|
|
"num_tokens": 3921792.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.175367546081543,
|
|
"epoch": 0.1485083453804705,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 7.1779,
|
|
"mean_token_accuracy": 0.10670020878314972,
|
|
"num_tokens": 3939896.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.243035173416137,
|
|
"epoch": 0.14916546195295044,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 7.2189,
|
|
"mean_token_accuracy": 0.09671530574560165,
|
|
"num_tokens": 3958019.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.291995239257813,
|
|
"epoch": 0.1498225785254304,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 7.3026,
|
|
"mean_token_accuracy": 0.1016198918223381,
|
|
"num_tokens": 3975880.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.189220190048218,
|
|
"epoch": 0.15047969509791037,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 7.1553,
|
|
"mean_token_accuracy": 0.10134812369942665,
|
|
"num_tokens": 3992649.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.181387281417846,
|
|
"epoch": 0.15113681167039034,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 7.1851,
|
|
"mean_token_accuracy": 0.10243008136749268,
|
|
"num_tokens": 4009641.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.194750642776489,
|
|
"epoch": 0.15179392824287027,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 7.2668,
|
|
"mean_token_accuracy": 0.09644793346524239,
|
|
"num_tokens": 4028324.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.255599069595337,
|
|
"epoch": 0.15245104481535024,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 7.2032,
|
|
"mean_token_accuracy": 0.10386807322502137,
|
|
"num_tokens": 4046547.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.242625427246094,
|
|
"epoch": 0.1531081613878302,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 7.2258,
|
|
"mean_token_accuracy": 0.1007157877087593,
|
|
"num_tokens": 4062227.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.163462209701538,
|
|
"epoch": 0.15376527796031017,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 7.1801,
|
|
"mean_token_accuracy": 0.10540101230144501,
|
|
"num_tokens": 4079046.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.251097249984741,
|
|
"epoch": 0.1544223945327901,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 7.244,
|
|
"mean_token_accuracy": 0.10471903905272484,
|
|
"num_tokens": 4096559.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.216784477233887,
|
|
"epoch": 0.15507951110527007,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 7.1556,
|
|
"mean_token_accuracy": 0.10767733827233314,
|
|
"num_tokens": 4113251.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.065131950378418,
|
|
"epoch": 0.15573662767775004,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 7.0617,
|
|
"mean_token_accuracy": 0.11634833589196206,
|
|
"num_tokens": 4133967.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.220524168014526,
|
|
"epoch": 0.15639374425023,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 7.1619,
|
|
"mean_token_accuracy": 0.1012740470468998,
|
|
"num_tokens": 4150801.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.286528825759888,
|
|
"epoch": 0.15705086082270994,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 7.2604,
|
|
"mean_token_accuracy": 0.0940374843776226,
|
|
"num_tokens": 4168727.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.180148553848267,
|
|
"epoch": 0.1577079773951899,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 7.1508,
|
|
"mean_token_accuracy": 0.10601523295044898,
|
|
"num_tokens": 4187639.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.135108709335327,
|
|
"epoch": 0.15836509396766987,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 7.1612,
|
|
"mean_token_accuracy": 0.10450763776898384,
|
|
"num_tokens": 4204266.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.2624317646026615,
|
|
"epoch": 0.1590222105401498,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 7.1404,
|
|
"mean_token_accuracy": 0.10236875712871552,
|
|
"num_tokens": 4221724.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.164175224304199,
|
|
"epoch": 0.15967932711262978,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 7.2273,
|
|
"mean_token_accuracy": 0.10266198962926865,
|
|
"num_tokens": 4240729.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.121058464050293,
|
|
"epoch": 0.16033644368510974,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 7.1063,
|
|
"mean_token_accuracy": 0.11425342187285423,
|
|
"num_tokens": 4260604.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.142035484313965,
|
|
"epoch": 0.1609935602575897,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 7.1992,
|
|
"mean_token_accuracy": 0.10155246257781983,
|
|
"num_tokens": 4277763.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.209809589385986,
|
|
"epoch": 0.16165067683006965,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 7.2427,
|
|
"mean_token_accuracy": 0.09955391138792039,
|
|
"num_tokens": 4294625.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.154326152801514,
|
|
"epoch": 0.1623077934025496,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 7.1676,
|
|
"mean_token_accuracy": 0.10906312316656112,
|
|
"num_tokens": 4312247.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.254214096069336,
|
|
"epoch": 0.16296490997502958,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 7.2685,
|
|
"mean_token_accuracy": 0.09215258359909058,
|
|
"num_tokens": 4329618.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.223868036270142,
|
|
"epoch": 0.16362202654750954,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 7.2086,
|
|
"mean_token_accuracy": 0.1000858373939991,
|
|
"num_tokens": 4347702.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.246126794815064,
|
|
"epoch": 0.16427914311998948,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 7.1858,
|
|
"mean_token_accuracy": 0.09782344475388527,
|
|
"num_tokens": 4365116.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.171083784103393,
|
|
"epoch": 0.16493625969246944,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 7.2512,
|
|
"mean_token_accuracy": 0.10115259513258934,
|
|
"num_tokens": 4382634.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.264097499847412,
|
|
"epoch": 0.1655933762649494,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 7.2555,
|
|
"mean_token_accuracy": 0.09854963570833206,
|
|
"num_tokens": 4400067.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.18475866317749,
|
|
"epoch": 0.16625049283742935,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 7.1363,
|
|
"mean_token_accuracy": 0.0983266532421112,
|
|
"num_tokens": 4415974.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.142167472839356,
|
|
"epoch": 0.1669076094099093,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 7.1565,
|
|
"mean_token_accuracy": 0.10286837294697762,
|
|
"num_tokens": 4433718.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.263294696807861,
|
|
"epoch": 0.16756472598238928,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 7.2295,
|
|
"mean_token_accuracy": 0.09759362787008286,
|
|
"num_tokens": 4452327.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.116361236572265,
|
|
"epoch": 0.16822184255486924,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 7.112,
|
|
"mean_token_accuracy": 0.10486611649394036,
|
|
"num_tokens": 4468025.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.023391008377075,
|
|
"epoch": 0.16887895912734918,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.9262,
|
|
"mean_token_accuracy": 0.12515356317162513,
|
|
"num_tokens": 4486104.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.159882354736328,
|
|
"epoch": 0.16953607569982915,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 7.2428,
|
|
"mean_token_accuracy": 0.09691546708345414,
|
|
"num_tokens": 4503776.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.169772386550903,
|
|
"epoch": 0.1701931922723091,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 7.2116,
|
|
"mean_token_accuracy": 0.1008364237844944,
|
|
"num_tokens": 4521574.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.2007242202758786,
|
|
"epoch": 0.17085030884478908,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 7.1505,
|
|
"mean_token_accuracy": 0.11244964003562927,
|
|
"num_tokens": 4539276.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.171060848236084,
|
|
"epoch": 0.17150742541726902,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 7.1326,
|
|
"mean_token_accuracy": 0.1008635438978672,
|
|
"num_tokens": 4557626.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.142626094818115,
|
|
"epoch": 0.17216454198974898,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 7.1549,
|
|
"mean_token_accuracy": 0.10123615637421608,
|
|
"num_tokens": 4574045.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.167723512649536,
|
|
"epoch": 0.17282165856222895,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 7.1759,
|
|
"mean_token_accuracy": 0.10515163168311119,
|
|
"num_tokens": 4590743.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.111159324645996,
|
|
"epoch": 0.17347877513470888,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 7.0989,
|
|
"mean_token_accuracy": 0.10333124399185181,
|
|
"num_tokens": 4607616.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.161529684066773,
|
|
"epoch": 0.17413589170718885,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 7.0922,
|
|
"mean_token_accuracy": 0.10455577373504639,
|
|
"num_tokens": 4624723.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.1527036190032955,
|
|
"epoch": 0.17479300827966882,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 7.1299,
|
|
"mean_token_accuracy": 0.09841537550091743,
|
|
"num_tokens": 4642535.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.17755012512207,
|
|
"epoch": 0.17545012485214878,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 7.1287,
|
|
"mean_token_accuracy": 0.09741237014532089,
|
|
"num_tokens": 4660068.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.029680061340332,
|
|
"epoch": 0.17610724142462872,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 7.0628,
|
|
"mean_token_accuracy": 0.11555436253547668,
|
|
"num_tokens": 4678985.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.155279445648193,
|
|
"epoch": 0.17676435799710868,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 7.1287,
|
|
"mean_token_accuracy": 0.09945940226316452,
|
|
"num_tokens": 4697827.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.107897138595581,
|
|
"epoch": 0.17742147456958865,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 7.1291,
|
|
"mean_token_accuracy": 0.09734978899359703,
|
|
"num_tokens": 4716437.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.036555910110474,
|
|
"epoch": 0.17807859114206862,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 7.0523,
|
|
"mean_token_accuracy": 0.10956270396709442,
|
|
"num_tokens": 4734528.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.1375220775604244,
|
|
"epoch": 0.17873570771454855,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 7.1495,
|
|
"mean_token_accuracy": 0.09941326081752777,
|
|
"num_tokens": 4752174.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.162973308563233,
|
|
"epoch": 0.17939282428702852,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 7.0365,
|
|
"mean_token_accuracy": 0.10331410691142082,
|
|
"num_tokens": 4769040.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.036287784576416,
|
|
"epoch": 0.18004994085950848,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 7.1211,
|
|
"mean_token_accuracy": 0.10569187849760056,
|
|
"num_tokens": 4787769.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.198102760314941,
|
|
"epoch": 0.18070705743198842,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 7.1446,
|
|
"mean_token_accuracy": 0.09686194062232971,
|
|
"num_tokens": 4805730.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.123643064498902,
|
|
"epoch": 0.1813641740044684,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 7.1348,
|
|
"mean_token_accuracy": 0.10434755310416222,
|
|
"num_tokens": 4823518.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.153635454177857,
|
|
"epoch": 0.18202129057694835,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 7.1404,
|
|
"mean_token_accuracy": 0.10507838502526283,
|
|
"num_tokens": 4840332.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.123753452301026,
|
|
"epoch": 0.18267840714942832,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 7.1231,
|
|
"mean_token_accuracy": 0.09961956813931465,
|
|
"num_tokens": 4857633.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.068319320678711,
|
|
"epoch": 0.18333552372190826,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 7.0578,
|
|
"mean_token_accuracy": 0.10383191257715225,
|
|
"num_tokens": 4874145.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.05531587600708,
|
|
"epoch": 0.18399264029438822,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.9817,
|
|
"mean_token_accuracy": 0.10680583864450455,
|
|
"num_tokens": 4892231.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.049816274642945,
|
|
"epoch": 0.1846497568668682,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 7.1012,
|
|
"mean_token_accuracy": 0.10223743543028832,
|
|
"num_tokens": 4908095.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.0817399501800535,
|
|
"epoch": 0.18530687343934815,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.9911,
|
|
"mean_token_accuracy": 0.11124311462044716,
|
|
"num_tokens": 4926608.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.9987890243530275,
|
|
"epoch": 0.1859639900118281,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.9371,
|
|
"mean_token_accuracy": 0.11885873302817344,
|
|
"num_tokens": 4943617.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.06851634979248,
|
|
"epoch": 0.18662110658430806,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 7.0327,
|
|
"mean_token_accuracy": 0.1135017842054367,
|
|
"num_tokens": 4962255.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.001637077331543,
|
|
"epoch": 0.18727822315678802,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.9265,
|
|
"mean_token_accuracy": 0.12256510108709336,
|
|
"num_tokens": 4980276.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.053653526306152,
|
|
"epoch": 0.18793533972926796,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 7.0544,
|
|
"mean_token_accuracy": 0.10385716408491134,
|
|
"num_tokens": 4996244.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.084101438522339,
|
|
"epoch": 0.18859245630174792,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 7.0743,
|
|
"mean_token_accuracy": 0.1009013146162033,
|
|
"num_tokens": 5016070.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.109147834777832,
|
|
"epoch": 0.1892495728742279,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 7.138,
|
|
"mean_token_accuracy": 0.0944010891020298,
|
|
"num_tokens": 5032302.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.157390308380127,
|
|
"epoch": 0.18990668944670785,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 7.0823,
|
|
"mean_token_accuracy": 0.10348261818289757,
|
|
"num_tokens": 5050474.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.962280035018921,
|
|
"epoch": 0.1905638060191878,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 7.0122,
|
|
"mean_token_accuracy": 0.10257056131958961,
|
|
"num_tokens": 5069626.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.163888025283813,
|
|
"epoch": 0.19122092259166776,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 7.1151,
|
|
"mean_token_accuracy": 0.0981297992169857,
|
|
"num_tokens": 5087126.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.0553618431091305,
|
|
"epoch": 0.19187803916414772,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 7.0958,
|
|
"mean_token_accuracy": 0.10250502526760101,
|
|
"num_tokens": 5105324.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.1029359817504885,
|
|
"epoch": 0.1925351557366277,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 7.1372,
|
|
"mean_token_accuracy": 0.1008988417685032,
|
|
"num_tokens": 5121832.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.041136693954468,
|
|
"epoch": 0.19319227230910763,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 7.0571,
|
|
"mean_token_accuracy": 0.1045589379966259,
|
|
"num_tokens": 5138072.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.143003797531128,
|
|
"epoch": 0.1938493888815876,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 7.0829,
|
|
"mean_token_accuracy": 0.10569359660148621,
|
|
"num_tokens": 5154486.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.041785955429077,
|
|
"epoch": 0.19450650545406756,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 7.0603,
|
|
"mean_token_accuracy": 0.10345918461680412,
|
|
"num_tokens": 5171995.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.947126388549805,
|
|
"epoch": 0.19516362202654752,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.8975,
|
|
"mean_token_accuracy": 0.12085305452346802,
|
|
"num_tokens": 5190708.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.049253225326538,
|
|
"epoch": 0.19582073859902746,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 7.0137,
|
|
"mean_token_accuracy": 0.10840069875121117,
|
|
"num_tokens": 5207518.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.979583120346069,
|
|
"epoch": 0.19647785517150743,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.9889,
|
|
"mean_token_accuracy": 0.11757338047027588,
|
|
"num_tokens": 5225043.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.06225094795227,
|
|
"epoch": 0.1971349717439874,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 7.0219,
|
|
"mean_token_accuracy": 0.1079759307205677,
|
|
"num_tokens": 5242101.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 6991815499776000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|