3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.0991375037176564,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.69155969619751,
|
|
"epoch": 0.0003304583457255213,
|
|
"grad_norm": 12.8125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8103,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10836.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69160442352295,
|
|
"epoch": 0.0006609166914510426,
|
|
"grad_norm": 10.9375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7287,
|
|
"mean_token_accuracy": 0.00010266940807923675,
|
|
"num_tokens": 20708.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691381645202636,
|
|
"epoch": 0.000991375037176564,
|
|
"grad_norm": 9.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.556,
|
|
"mean_token_accuracy": 0.012630514975171536,
|
|
"num_tokens": 32651.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.685187435150146,
|
|
"epoch": 0.0013218333829020852,
|
|
"grad_norm": 6.25,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2559,
|
|
"mean_token_accuracy": 0.04093300513923168,
|
|
"num_tokens": 43844.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.628204250335694,
|
|
"epoch": 0.0016522917286276066,
|
|
"grad_norm": 3.9375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.0415,
|
|
"mean_token_accuracy": 0.038726492412388326,
|
|
"num_tokens": 55019.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.554113483428955,
|
|
"epoch": 0.001982750074353128,
|
|
"grad_norm": 3.140625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.8257,
|
|
"mean_token_accuracy": 0.048476964607834815,
|
|
"num_tokens": 66746.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.590810203552246,
|
|
"epoch": 0.002313208420078649,
|
|
"grad_norm": 2.765625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.8306,
|
|
"mean_token_accuracy": 0.03930025473237038,
|
|
"num_tokens": 78821.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.605837535858154,
|
|
"epoch": 0.0026436667658041703,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.7593,
|
|
"mean_token_accuracy": 0.037112961150705816,
|
|
"num_tokens": 89051.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.587123107910156,
|
|
"epoch": 0.0029741251115296915,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6663,
|
|
"mean_token_accuracy": 0.040811435878276826,
|
|
"num_tokens": 99691.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.559133148193359,
|
|
"epoch": 0.003304583457255213,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.7075,
|
|
"mean_token_accuracy": 0.03913131728768349,
|
|
"num_tokens": 110614.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.571648788452148,
|
|
"epoch": 0.0036350418029807343,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.7135,
|
|
"mean_token_accuracy": 0.03680916838347912,
|
|
"num_tokens": 122915.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.579666042327881,
|
|
"epoch": 0.003965500148706256,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5367,
|
|
"mean_token_accuracy": 0.040818646922707556,
|
|
"num_tokens": 132627.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.55205478668213,
|
|
"epoch": 0.004295958494431777,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.5003,
|
|
"mean_token_accuracy": 0.044149795174598695,
|
|
"num_tokens": 144317.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.516563987731933,
|
|
"epoch": 0.004626416840157298,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.3204,
|
|
"mean_token_accuracy": 0.04889754429459572,
|
|
"num_tokens": 152798.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.421174716949462,
|
|
"epoch": 0.00495687518588282,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2668,
|
|
"mean_token_accuracy": 0.04544033259153366,
|
|
"num_tokens": 162656.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.396770095825195,
|
|
"epoch": 0.005287333531608341,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.226,
|
|
"mean_token_accuracy": 0.050432728976011275,
|
|
"num_tokens": 173784.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.419888401031494,
|
|
"epoch": 0.005617791877333862,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1258,
|
|
"mean_token_accuracy": 0.0568363256752491,
|
|
"num_tokens": 184681.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.358776760101318,
|
|
"epoch": 0.005948250223059383,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.0169,
|
|
"mean_token_accuracy": 0.053509436920285224,
|
|
"num_tokens": 195307.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.326674938201904,
|
|
"epoch": 0.006278708568784905,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8921,
|
|
"mean_token_accuracy": 0.060885720700025556,
|
|
"num_tokens": 206857.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.2583758354187,
|
|
"epoch": 0.006609166914510426,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8016,
|
|
"mean_token_accuracy": 0.05900029614567757,
|
|
"num_tokens": 217291.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.172372817993164,
|
|
"epoch": 0.006939625260235947,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.6847,
|
|
"mean_token_accuracy": 0.0601776622235775,
|
|
"num_tokens": 227184.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.142582893371582,
|
|
"epoch": 0.007270083605961469,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.5856,
|
|
"mean_token_accuracy": 0.062167299538850786,
|
|
"num_tokens": 237928.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.992514324188232,
|
|
"epoch": 0.00760054195168699,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.4541,
|
|
"mean_token_accuracy": 0.07033347450196743,
|
|
"num_tokens": 250073.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.925767707824708,
|
|
"epoch": 0.007931000297412512,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.3133,
|
|
"mean_token_accuracy": 0.07063820958137512,
|
|
"num_tokens": 261191.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.826898574829102,
|
|
"epoch": 0.008261458643138032,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.2501,
|
|
"mean_token_accuracy": 0.0650045171380043,
|
|
"num_tokens": 271226.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.688051795959472,
|
|
"epoch": 0.008591916988863553,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.1108,
|
|
"mean_token_accuracy": 0.07041079625487327,
|
|
"num_tokens": 282632.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.554755210876465,
|
|
"epoch": 0.008922375334589075,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.0812,
|
|
"mean_token_accuracy": 0.06935585699975491,
|
|
"num_tokens": 294119.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.3393235206604,
|
|
"epoch": 0.009252833680314597,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.9154,
|
|
"mean_token_accuracy": 0.06800433918833733,
|
|
"num_tokens": 304725.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.183171463012695,
|
|
"epoch": 0.009583292026040118,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.8272,
|
|
"mean_token_accuracy": 0.06738981902599335,
|
|
"num_tokens": 315260.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.023473262786865,
|
|
"epoch": 0.00991375037176564,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.816,
|
|
"mean_token_accuracy": 0.0658723335713148,
|
|
"num_tokens": 326228.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.789591312408447,
|
|
"epoch": 0.01024420871749116,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.6682,
|
|
"mean_token_accuracy": 0.07138085626065731,
|
|
"num_tokens": 336974.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.713123226165772,
|
|
"epoch": 0.010574667063216681,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.6937,
|
|
"mean_token_accuracy": 0.06671310998499394,
|
|
"num_tokens": 347380.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.564159774780274,
|
|
"epoch": 0.010905125408942203,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.5414,
|
|
"mean_token_accuracy": 0.06911698617041111,
|
|
"num_tokens": 358159.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.391338634490968,
|
|
"epoch": 0.011235583754667725,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.6154,
|
|
"mean_token_accuracy": 0.06779548972845077,
|
|
"num_tokens": 369224.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.317761039733886,
|
|
"epoch": 0.011566042100393246,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.5342,
|
|
"mean_token_accuracy": 0.0646525926887989,
|
|
"num_tokens": 380973.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.257954692840576,
|
|
"epoch": 0.011896500446118766,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.6128,
|
|
"mean_token_accuracy": 0.06305343359708786,
|
|
"num_tokens": 393367.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.156609535217285,
|
|
"epoch": 0.012226958791844288,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5075,
|
|
"mean_token_accuracy": 0.07255993783473969,
|
|
"num_tokens": 405104.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.167465686798096,
|
|
"epoch": 0.01255741713756981,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.5335,
|
|
"mean_token_accuracy": 0.06890368461608887,
|
|
"num_tokens": 415734.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.079139995574952,
|
|
"epoch": 0.01288787548329533,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.5476,
|
|
"mean_token_accuracy": 0.06830249056220054,
|
|
"num_tokens": 425949.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.103539848327637,
|
|
"epoch": 0.013218333829020852,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.3962,
|
|
"mean_token_accuracy": 0.06818307042121888,
|
|
"num_tokens": 436083.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.987107515335083,
|
|
"epoch": 0.013548792174746372,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4,
|
|
"mean_token_accuracy": 0.0744843065738678,
|
|
"num_tokens": 446731.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.954383039474488,
|
|
"epoch": 0.013879250520471894,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.392,
|
|
"mean_token_accuracy": 0.0729294192045927,
|
|
"num_tokens": 456630.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.964882946014404,
|
|
"epoch": 0.014209708866197416,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.406,
|
|
"mean_token_accuracy": 0.06909749209880829,
|
|
"num_tokens": 468804.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.789395570755005,
|
|
"epoch": 0.014540167211922937,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.3496,
|
|
"mean_token_accuracy": 0.07542757764458656,
|
|
"num_tokens": 479321.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.850273847579956,
|
|
"epoch": 0.014870625557648459,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3846,
|
|
"mean_token_accuracy": 0.0726801447570324,
|
|
"num_tokens": 490788.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.854997777938843,
|
|
"epoch": 0.01520108390337398,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.3583,
|
|
"mean_token_accuracy": 0.07561610527336597,
|
|
"num_tokens": 501992.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.8472926139831545,
|
|
"epoch": 0.0155315422490995,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2768,
|
|
"mean_token_accuracy": 0.08250004723668099,
|
|
"num_tokens": 511927.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.760110378265381,
|
|
"epoch": 0.015862000594825024,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.2478,
|
|
"mean_token_accuracy": 0.08096232078969479,
|
|
"num_tokens": 521874.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.749714708328247,
|
|
"epoch": 0.016192458940550544,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.3405,
|
|
"mean_token_accuracy": 0.07609539777040482,
|
|
"num_tokens": 532913.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.832424545288086,
|
|
"epoch": 0.016522917286276063,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2888,
|
|
"mean_token_accuracy": 0.07482444941997528,
|
|
"num_tokens": 544195.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.710200643539428,
|
|
"epoch": 0.016853375632001587,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.2822,
|
|
"mean_token_accuracy": 0.07511841356754304,
|
|
"num_tokens": 555555.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.723060989379883,
|
|
"epoch": 0.017183833977727107,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3017,
|
|
"mean_token_accuracy": 0.06918322369456291,
|
|
"num_tokens": 566446.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.735864305496216,
|
|
"epoch": 0.01751429232345263,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1435,
|
|
"mean_token_accuracy": 0.08523371145129204,
|
|
"num_tokens": 576138.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.7571961879730225,
|
|
"epoch": 0.01784475066917815,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.4068,
|
|
"mean_token_accuracy": 0.07323086895048618,
|
|
"num_tokens": 587559.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.637564659118652,
|
|
"epoch": 0.01817520901490367,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2492,
|
|
"mean_token_accuracy": 0.07899717763066291,
|
|
"num_tokens": 598704.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.6305924415588375,
|
|
"epoch": 0.018505667360629193,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2202,
|
|
"mean_token_accuracy": 0.08230236917734146,
|
|
"num_tokens": 609992.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.777820110321045,
|
|
"epoch": 0.018836125706354713,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.2906,
|
|
"mean_token_accuracy": 0.0766009196639061,
|
|
"num_tokens": 620756.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.612432479858398,
|
|
"epoch": 0.019166584052080236,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1494,
|
|
"mean_token_accuracy": 0.08179529085755348,
|
|
"num_tokens": 630335.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.54581618309021,
|
|
"epoch": 0.019497042397805756,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.2277,
|
|
"mean_token_accuracy": 0.07613298408687115,
|
|
"num_tokens": 641176.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.677836608886719,
|
|
"epoch": 0.01982750074353128,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.1341,
|
|
"mean_token_accuracy": 0.0822694730013609,
|
|
"num_tokens": 652144.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.5951337814331055,
|
|
"epoch": 0.0201579590892568,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.1135,
|
|
"mean_token_accuracy": 0.07866659983992577,
|
|
"num_tokens": 663040.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.540827846527099,
|
|
"epoch": 0.02048841743498232,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.1588,
|
|
"mean_token_accuracy": 0.08099368289113044,
|
|
"num_tokens": 674407.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.606076622009278,
|
|
"epoch": 0.020818875780707843,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.1684,
|
|
"mean_token_accuracy": 0.07814301066100597,
|
|
"num_tokens": 686575.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.528010988235474,
|
|
"epoch": 0.021149334126433363,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.1044,
|
|
"mean_token_accuracy": 0.08741299659013749,
|
|
"num_tokens": 697404.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.685597991943359,
|
|
"epoch": 0.021479792472158886,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.2433,
|
|
"mean_token_accuracy": 0.07762166783213616,
|
|
"num_tokens": 708462.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.546236228942871,
|
|
"epoch": 0.021810250817884406,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1297,
|
|
"mean_token_accuracy": 0.08397110104560852,
|
|
"num_tokens": 719485.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.449752140045166,
|
|
"epoch": 0.022140709163609926,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1155,
|
|
"mean_token_accuracy": 0.08090364784002305,
|
|
"num_tokens": 730417.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.544217443466186,
|
|
"epoch": 0.02247116750933545,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0674,
|
|
"mean_token_accuracy": 0.08530745506286622,
|
|
"num_tokens": 740644.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.413099479675293,
|
|
"epoch": 0.02280162585506097,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.0263,
|
|
"mean_token_accuracy": 0.0855514369904995,
|
|
"num_tokens": 751843.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.519398832321167,
|
|
"epoch": 0.023132084200786492,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.0071,
|
|
"mean_token_accuracy": 0.08101744726300239,
|
|
"num_tokens": 763193.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.416529798507691,
|
|
"epoch": 0.023462542546512012,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.088,
|
|
"mean_token_accuracy": 0.07980120629072189,
|
|
"num_tokens": 775393.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.473480892181397,
|
|
"epoch": 0.023793000892237532,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9638,
|
|
"mean_token_accuracy": 0.08316674306988717,
|
|
"num_tokens": 786292.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.4191234588623045,
|
|
"epoch": 0.024123459237963055,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0293,
|
|
"mean_token_accuracy": 0.07673819884657859,
|
|
"num_tokens": 798344.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.390763473510742,
|
|
"epoch": 0.024453917583688575,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.0266,
|
|
"mean_token_accuracy": 0.08823682218790055,
|
|
"num_tokens": 809058.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.3451203346252445,
|
|
"epoch": 0.0247843759294141,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.9562,
|
|
"mean_token_accuracy": 0.08663681447505951,
|
|
"num_tokens": 820848.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.416850471496582,
|
|
"epoch": 0.02511483427513962,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0447,
|
|
"mean_token_accuracy": 0.08994424119591712,
|
|
"num_tokens": 833361.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.370834159851074,
|
|
"epoch": 0.02544529262086514,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9881,
|
|
"mean_token_accuracy": 0.08523149564862251,
|
|
"num_tokens": 843801.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.4036249160766605,
|
|
"epoch": 0.02577575096659066,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.9772,
|
|
"mean_token_accuracy": 0.08082735612988472,
|
|
"num_tokens": 854477.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.330833148956299,
|
|
"epoch": 0.02610620931231618,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0413,
|
|
"mean_token_accuracy": 0.08312294557690621,
|
|
"num_tokens": 863380.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.37300477027893,
|
|
"epoch": 0.026436667658041705,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.9635,
|
|
"mean_token_accuracy": 0.08273139744997024,
|
|
"num_tokens": 874215.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.265693616867066,
|
|
"epoch": 0.026767126003767225,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0215,
|
|
"mean_token_accuracy": 0.09335545673966408,
|
|
"num_tokens": 885928.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.470329475402832,
|
|
"epoch": 0.027097584349492745,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.036,
|
|
"mean_token_accuracy": 0.08615454584360123,
|
|
"num_tokens": 896349.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.282833099365234,
|
|
"epoch": 0.027428042695218268,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.9012,
|
|
"mean_token_accuracy": 0.0931041918694973,
|
|
"num_tokens": 907126.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.305163431167602,
|
|
"epoch": 0.027758501040943788,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0214,
|
|
"mean_token_accuracy": 0.08308781683444977,
|
|
"num_tokens": 918021.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.268058013916016,
|
|
"epoch": 0.02808895938666931,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8958,
|
|
"mean_token_accuracy": 0.08962870463728904,
|
|
"num_tokens": 929321.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.27590012550354,
|
|
"epoch": 0.02841941773239483,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.9289,
|
|
"mean_token_accuracy": 0.08679753243923187,
|
|
"num_tokens": 940196.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.284714984893799,
|
|
"epoch": 0.028749876078120355,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.9195,
|
|
"mean_token_accuracy": 0.09327666312456132,
|
|
"num_tokens": 950600.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.263758373260498,
|
|
"epoch": 0.029080334423845874,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.9498,
|
|
"mean_token_accuracy": 0.08754595220088959,
|
|
"num_tokens": 961447.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.194220161437988,
|
|
"epoch": 0.029410792769571394,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8597,
|
|
"mean_token_accuracy": 0.08954929485917092,
|
|
"num_tokens": 973264.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.355491304397583,
|
|
"epoch": 0.029741251115296918,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.0987,
|
|
"mean_token_accuracy": 0.07984722442924977,
|
|
"num_tokens": 985123.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.258900594711304,
|
|
"epoch": 0.030071709461022438,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.935,
|
|
"mean_token_accuracy": 0.08211459219455719,
|
|
"num_tokens": 996037.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.292104005813599,
|
|
"epoch": 0.03040216780674796,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9936,
|
|
"mean_token_accuracy": 0.0812192477285862,
|
|
"num_tokens": 1006940.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.212660837173462,
|
|
"epoch": 0.03073262615247348,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8055,
|
|
"mean_token_accuracy": 0.08365916460752487,
|
|
"num_tokens": 1017259.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.161711931228638,
|
|
"epoch": 0.031063084498199,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.8381,
|
|
"mean_token_accuracy": 0.08933527991175652,
|
|
"num_tokens": 1028477.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.197433996200561,
|
|
"epoch": 0.031393542843924524,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.8515,
|
|
"mean_token_accuracy": 0.0891186311841011,
|
|
"num_tokens": 1039812.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.1398866176605225,
|
|
"epoch": 0.03172400118965005,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8207,
|
|
"mean_token_accuracy": 0.09132780060172081,
|
|
"num_tokens": 1050856.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.054386281967163,
|
|
"epoch": 0.032054459535375564,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6937,
|
|
"mean_token_accuracy": 0.09542593285441399,
|
|
"num_tokens": 1061491.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.081818532943726,
|
|
"epoch": 0.03238491788110109,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7775,
|
|
"mean_token_accuracy": 0.08975680768489838,
|
|
"num_tokens": 1071986.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.154221773147583,
|
|
"epoch": 0.03271537622682661,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8332,
|
|
"mean_token_accuracy": 0.08331749886274338,
|
|
"num_tokens": 1084174.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.191654109954834,
|
|
"epoch": 0.03304583457255213,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8399,
|
|
"mean_token_accuracy": 0.08998439535498619,
|
|
"num_tokens": 1094267.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.129807043075561,
|
|
"epoch": 0.03337629291827765,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.8272,
|
|
"mean_token_accuracy": 0.09337387830018998,
|
|
"num_tokens": 1105421.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.136587476730346,
|
|
"epoch": 0.033706751264003174,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.9057,
|
|
"mean_token_accuracy": 0.08978310525417328,
|
|
"num_tokens": 1117574.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.068832921981811,
|
|
"epoch": 0.0340372096097287,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7841,
|
|
"mean_token_accuracy": 0.09897544533014298,
|
|
"num_tokens": 1129666.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.095613145828247,
|
|
"epoch": 0.03436766795545421,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.8098,
|
|
"mean_token_accuracy": 0.09167956337332725,
|
|
"num_tokens": 1140936.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.066299915313721,
|
|
"epoch": 0.03469812630117974,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.8261,
|
|
"mean_token_accuracy": 0.09235902950167656,
|
|
"num_tokens": 1152239.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.073812246322632,
|
|
"epoch": 0.03502858464690526,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7493,
|
|
"mean_token_accuracy": 0.09044400379061698,
|
|
"num_tokens": 1162680.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.036542844772339,
|
|
"epoch": 0.035359042992630776,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8029,
|
|
"mean_token_accuracy": 0.09098404496908188,
|
|
"num_tokens": 1173155.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.014681577682495,
|
|
"epoch": 0.0356895013383563,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7485,
|
|
"mean_token_accuracy": 0.09319498017430305,
|
|
"num_tokens": 1183042.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.069463157653809,
|
|
"epoch": 0.03601995968408182,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8447,
|
|
"mean_token_accuracy": 0.08397522196173668,
|
|
"num_tokens": 1194512.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.06383261680603,
|
|
"epoch": 0.03635041802980734,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7105,
|
|
"mean_token_accuracy": 0.09679808020591736,
|
|
"num_tokens": 1205063.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.9953466892242435,
|
|
"epoch": 0.03668087637553286,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.7437,
|
|
"mean_token_accuracy": 0.0971693828701973,
|
|
"num_tokens": 1215967.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.036475419998169,
|
|
"epoch": 0.037011334721258386,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.7225,
|
|
"mean_token_accuracy": 0.09566427320241928,
|
|
"num_tokens": 1226712.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.018493223190307,
|
|
"epoch": 0.03734179306698391,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7553,
|
|
"mean_token_accuracy": 0.09875442162156105,
|
|
"num_tokens": 1238058.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.883628988265992,
|
|
"epoch": 0.037672251412709426,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.6518,
|
|
"mean_token_accuracy": 0.09676829054951668,
|
|
"num_tokens": 1248199.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.979853677749634,
|
|
"epoch": 0.03800270975843495,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7898,
|
|
"mean_token_accuracy": 0.08489116355776787,
|
|
"num_tokens": 1260755.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.026479578018188,
|
|
"epoch": 0.03833316810416047,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7464,
|
|
"mean_token_accuracy": 0.0949298694729805,
|
|
"num_tokens": 1270395.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.907937526702881,
|
|
"epoch": 0.03866362644988599,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.7393,
|
|
"mean_token_accuracy": 0.09430743306875229,
|
|
"num_tokens": 1282735.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.929727029800415,
|
|
"epoch": 0.03899408479561151,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6946,
|
|
"mean_token_accuracy": 0.09773564413189888,
|
|
"num_tokens": 1292991.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.953424596786499,
|
|
"epoch": 0.039324543141337036,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7259,
|
|
"mean_token_accuracy": 0.09127573445439338,
|
|
"num_tokens": 1303361.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.939087438583374,
|
|
"epoch": 0.03965500148706256,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.6299,
|
|
"mean_token_accuracy": 0.09687030985951424,
|
|
"num_tokens": 1314099.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.892501449584961,
|
|
"epoch": 0.039985459832788076,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6377,
|
|
"mean_token_accuracy": 0.09915548786520959,
|
|
"num_tokens": 1324059.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.982302951812744,
|
|
"epoch": 0.0403159181785136,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.7706,
|
|
"mean_token_accuracy": 0.09124096035957337,
|
|
"num_tokens": 1334208.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.0098623752594,
|
|
"epoch": 0.04064637652423912,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7337,
|
|
"mean_token_accuracy": 0.09305815249681473,
|
|
"num_tokens": 1345633.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.876529264450073,
|
|
"epoch": 0.04097683486996464,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5852,
|
|
"mean_token_accuracy": 0.09691078439354897,
|
|
"num_tokens": 1356478.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.807394647598267,
|
|
"epoch": 0.04130729321569016,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6882,
|
|
"mean_token_accuracy": 0.09402298852801323,
|
|
"num_tokens": 1368925.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.940099287033081,
|
|
"epoch": 0.041637751561415685,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7044,
|
|
"mean_token_accuracy": 0.09032920226454735,
|
|
"num_tokens": 1379150.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.922402429580688,
|
|
"epoch": 0.0419682099071412,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6115,
|
|
"mean_token_accuracy": 0.10022713616490364,
|
|
"num_tokens": 1388993.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.887875175476074,
|
|
"epoch": 0.042298668252866725,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.6978,
|
|
"mean_token_accuracy": 0.09888584762811661,
|
|
"num_tokens": 1399906.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.799718141555786,
|
|
"epoch": 0.04262912659859225,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6084,
|
|
"mean_token_accuracy": 0.09428260773420334,
|
|
"num_tokens": 1411494.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.889655542373657,
|
|
"epoch": 0.04295958494431777,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.6969,
|
|
"mean_token_accuracy": 0.0979191929101944,
|
|
"num_tokens": 1423298.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.769257879257202,
|
|
"epoch": 0.04329004329004329,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5803,
|
|
"mean_token_accuracy": 0.09798824638128281,
|
|
"num_tokens": 1433892.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.887624311447143,
|
|
"epoch": 0.04362050163576881,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.6894,
|
|
"mean_token_accuracy": 0.09577311277389526,
|
|
"num_tokens": 1443778.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.862514400482178,
|
|
"epoch": 0.043950959981494335,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.6303,
|
|
"mean_token_accuracy": 0.09138552024960518,
|
|
"num_tokens": 1454990.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.814253664016723,
|
|
"epoch": 0.04428141832721985,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6097,
|
|
"mean_token_accuracy": 0.09931056052446366,
|
|
"num_tokens": 1466678.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.9184776782989506,
|
|
"epoch": 0.044611876672945375,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6926,
|
|
"mean_token_accuracy": 0.10091055706143379,
|
|
"num_tokens": 1477011.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.720648336410522,
|
|
"epoch": 0.0449423350186709,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5383,
|
|
"mean_token_accuracy": 0.0934677354991436,
|
|
"num_tokens": 1487519.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.770115756988526,
|
|
"epoch": 0.045272793364396414,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6625,
|
|
"mean_token_accuracy": 0.09488041177392006,
|
|
"num_tokens": 1499831.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.773311138153076,
|
|
"epoch": 0.04560325171012194,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.5623,
|
|
"mean_token_accuracy": 0.09586277455091477,
|
|
"num_tokens": 1509884.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.797134351730347,
|
|
"epoch": 0.04593371005584746,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.566,
|
|
"mean_token_accuracy": 0.09943261742591858,
|
|
"num_tokens": 1519693.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.701273775100708,
|
|
"epoch": 0.046264168401572985,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5545,
|
|
"mean_token_accuracy": 0.09949971362948418,
|
|
"num_tokens": 1530649.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.817782831192017,
|
|
"epoch": 0.0465946267472985,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.6569,
|
|
"mean_token_accuracy": 0.1008459135890007,
|
|
"num_tokens": 1541285.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.7360601902008055,
|
|
"epoch": 0.046925085093024024,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5629,
|
|
"mean_token_accuracy": 0.10334613248705864,
|
|
"num_tokens": 1552405.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.708620166778564,
|
|
"epoch": 0.04725554343874955,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5703,
|
|
"mean_token_accuracy": 0.09885218217968941,
|
|
"num_tokens": 1563471.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.694813919067383,
|
|
"epoch": 0.047586001784475064,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6496,
|
|
"mean_token_accuracy": 0.09043260738253593,
|
|
"num_tokens": 1573499.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.73417477607727,
|
|
"epoch": 0.04791646013020059,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4942,
|
|
"mean_token_accuracy": 0.09848063513636589,
|
|
"num_tokens": 1583827.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.88015398979187,
|
|
"epoch": 0.04824691847592611,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6481,
|
|
"mean_token_accuracy": 0.0926512323319912,
|
|
"num_tokens": 1594525.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.7281975746154785,
|
|
"epoch": 0.048577376821651634,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6375,
|
|
"mean_token_accuracy": 0.09198551625013351,
|
|
"num_tokens": 1605263.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.725015354156494,
|
|
"epoch": 0.04890783516737715,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5674,
|
|
"mean_token_accuracy": 0.09970696419477462,
|
|
"num_tokens": 1616232.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.655939340591431,
|
|
"epoch": 0.049238293513102674,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.5418,
|
|
"mean_token_accuracy": 0.10249985605478287,
|
|
"num_tokens": 1627370.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.780833673477173,
|
|
"epoch": 0.0495687518588282,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.6559,
|
|
"mean_token_accuracy": 0.09198240488767624,
|
|
"num_tokens": 1639328.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.692653226852417,
|
|
"epoch": 0.049899210204553714,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5224,
|
|
"mean_token_accuracy": 0.09956051632761956,
|
|
"num_tokens": 1649910.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.702142667770386,
|
|
"epoch": 0.05022966855027924,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.5548,
|
|
"mean_token_accuracy": 0.09980311691761017,
|
|
"num_tokens": 1661429.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.625659704208374,
|
|
"epoch": 0.05056012689600476,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4848,
|
|
"mean_token_accuracy": 0.10349942594766617,
|
|
"num_tokens": 1672169.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.666405963897705,
|
|
"epoch": 0.05089058524173028,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5629,
|
|
"mean_token_accuracy": 0.09543908089399337,
|
|
"num_tokens": 1683860.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.645580005645752,
|
|
"epoch": 0.0512210435874558,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4625,
|
|
"mean_token_accuracy": 0.10984127521514893,
|
|
"num_tokens": 1694348.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.645284843444824,
|
|
"epoch": 0.05155150193318132,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5667,
|
|
"mean_token_accuracy": 0.09349910616874695,
|
|
"num_tokens": 1704673.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.770039892196655,
|
|
"epoch": 0.05188196027890685,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.6103,
|
|
"mean_token_accuracy": 0.09760363474488258,
|
|
"num_tokens": 1714887.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.663259124755859,
|
|
"epoch": 0.05221241862463236,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5466,
|
|
"mean_token_accuracy": 0.09423963278532028,
|
|
"num_tokens": 1725630.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.584082794189453,
|
|
"epoch": 0.05254287697035789,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5141,
|
|
"mean_token_accuracy": 0.10037482157349586,
|
|
"num_tokens": 1736526.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.771090745925903,
|
|
"epoch": 0.05287333531608341,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.6545,
|
|
"mean_token_accuracy": 0.09194476455450058,
|
|
"num_tokens": 1748138.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.5120241165161135,
|
|
"epoch": 0.053203793661808926,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.442,
|
|
"mean_token_accuracy": 0.10061683282256126,
|
|
"num_tokens": 1758691.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.637072944641114,
|
|
"epoch": 0.05353425200753445,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5277,
|
|
"mean_token_accuracy": 0.0915706716477871,
|
|
"num_tokens": 1770004.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.73297963142395,
|
|
"epoch": 0.05386471035325997,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5924,
|
|
"mean_token_accuracy": 0.09404471218585968,
|
|
"num_tokens": 1780269.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.589795207977295,
|
|
"epoch": 0.05419516869898549,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5231,
|
|
"mean_token_accuracy": 0.10108774527907372,
|
|
"num_tokens": 1790581.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.64861421585083,
|
|
"epoch": 0.05452562704471101,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4919,
|
|
"mean_token_accuracy": 0.10306402817368507,
|
|
"num_tokens": 1801930.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.569200468063355,
|
|
"epoch": 0.054856085390436536,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4747,
|
|
"mean_token_accuracy": 0.10015810206532479,
|
|
"num_tokens": 1812876.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.593674230575561,
|
|
"epoch": 0.05518654373616206,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.5424,
|
|
"mean_token_accuracy": 0.09639053866267204,
|
|
"num_tokens": 1825521.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.617445182800293,
|
|
"epoch": 0.055517002081887576,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5164,
|
|
"mean_token_accuracy": 0.09912106320261956,
|
|
"num_tokens": 1835268.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.559435272216797,
|
|
"epoch": 0.0558474604276131,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4468,
|
|
"mean_token_accuracy": 0.10214125663042069,
|
|
"num_tokens": 1846269.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.57283730506897,
|
|
"epoch": 0.05617791877333862,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5153,
|
|
"mean_token_accuracy": 0.10395664349198341,
|
|
"num_tokens": 1857710.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.483006477355957,
|
|
"epoch": 0.05650837711906414,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5148,
|
|
"mean_token_accuracy": 0.10404823273420334,
|
|
"num_tokens": 1868569.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.575480699539185,
|
|
"epoch": 0.05683883546478966,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4747,
|
|
"mean_token_accuracy": 0.10771524235606193,
|
|
"num_tokens": 1878580.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.6114942073822025,
|
|
"epoch": 0.057169293810515186,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4172,
|
|
"mean_token_accuracy": 0.10425916984677315,
|
|
"num_tokens": 1888963.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.547268295288086,
|
|
"epoch": 0.05749975215624071,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.598,
|
|
"mean_token_accuracy": 0.09941022247076034,
|
|
"num_tokens": 1900085.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.600188589096069,
|
|
"epoch": 0.057830210501966225,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4176,
|
|
"mean_token_accuracy": 0.10359677523374558,
|
|
"num_tokens": 1911447.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.464424467086792,
|
|
"epoch": 0.05816066884769175,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.448,
|
|
"mean_token_accuracy": 0.09791496470570564,
|
|
"num_tokens": 1922194.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.561442136764526,
|
|
"epoch": 0.05849112719341727,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.481,
|
|
"mean_token_accuracy": 0.10386462211608886,
|
|
"num_tokens": 1932829.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.573152256011963,
|
|
"epoch": 0.05882158553914279,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.5259,
|
|
"mean_token_accuracy": 0.10657603517174721,
|
|
"num_tokens": 1942504.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.475641679763794,
|
|
"epoch": 0.05915204388486831,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.5123,
|
|
"mean_token_accuracy": 0.10546292960643769,
|
|
"num_tokens": 1954934.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.5593757152557375,
|
|
"epoch": 0.059482502230593835,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4601,
|
|
"mean_token_accuracy": 0.10120314806699753,
|
|
"num_tokens": 1968005.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.476503992080689,
|
|
"epoch": 0.05981296057631935,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4575,
|
|
"mean_token_accuracy": 0.1004819169640541,
|
|
"num_tokens": 1978532.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.53993706703186,
|
|
"epoch": 0.060143418922044875,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5338,
|
|
"mean_token_accuracy": 0.10164398774504661,
|
|
"num_tokens": 1989051.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.523714780807495,
|
|
"epoch": 0.0604738772677704,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.406,
|
|
"mean_token_accuracy": 0.1086203046143055,
|
|
"num_tokens": 2001245.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.457377147674561,
|
|
"epoch": 0.06080433561349592,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4188,
|
|
"mean_token_accuracy": 0.10521991178393364,
|
|
"num_tokens": 2011840.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.440218067169189,
|
|
"epoch": 0.06113479395922144,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.369,
|
|
"mean_token_accuracy": 0.10432081371545791,
|
|
"num_tokens": 2022454.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.510140705108642,
|
|
"epoch": 0.06146525230494696,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.5518,
|
|
"mean_token_accuracy": 0.0980154186487198,
|
|
"num_tokens": 2032745.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.4728662967681885,
|
|
"epoch": 0.061795710650672485,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4647,
|
|
"mean_token_accuracy": 0.09817253425717354,
|
|
"num_tokens": 2043020.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.616012287139893,
|
|
"epoch": 0.062126168996398,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.5055,
|
|
"mean_token_accuracy": 0.09617855548858642,
|
|
"num_tokens": 2055288.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.403015899658203,
|
|
"epoch": 0.062456627342123525,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4289,
|
|
"mean_token_accuracy": 0.10303806513547897,
|
|
"num_tokens": 2066252.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.583564329147339,
|
|
"epoch": 0.06278708568784905,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.5182,
|
|
"mean_token_accuracy": 0.09759321138262748,
|
|
"num_tokens": 2079404.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.481687831878662,
|
|
"epoch": 0.06311754403357457,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3989,
|
|
"mean_token_accuracy": 0.1046754002571106,
|
|
"num_tokens": 2091420.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.471569585800171,
|
|
"epoch": 0.0634480023793001,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4209,
|
|
"mean_token_accuracy": 0.09959583207964898,
|
|
"num_tokens": 2101825.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.467436027526856,
|
|
"epoch": 0.0637784607250256,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3872,
|
|
"mean_token_accuracy": 0.1006256528198719,
|
|
"num_tokens": 2113847.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.3760229587554935,
|
|
"epoch": 0.06410891907075113,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3764,
|
|
"mean_token_accuracy": 0.10424125492572785,
|
|
"num_tokens": 2124855.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.363870906829834,
|
|
"epoch": 0.06443937741647665,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3566,
|
|
"mean_token_accuracy": 0.10852005258202553,
|
|
"num_tokens": 2135164.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.540998125076294,
|
|
"epoch": 0.06476983576220217,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.471,
|
|
"mean_token_accuracy": 0.09465332999825478,
|
|
"num_tokens": 2146458.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.4124726295471195,
|
|
"epoch": 0.0651002941079277,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.4226,
|
|
"mean_token_accuracy": 0.10414307788014412,
|
|
"num_tokens": 2159184.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.3850232601165775,
|
|
"epoch": 0.06543075245365322,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3661,
|
|
"mean_token_accuracy": 0.10461859554052352,
|
|
"num_tokens": 2170424.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.399285221099854,
|
|
"epoch": 0.06576121079937874,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3683,
|
|
"mean_token_accuracy": 0.1082889124751091,
|
|
"num_tokens": 2181193.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.315083360671997,
|
|
"epoch": 0.06609166914510425,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3219,
|
|
"mean_token_accuracy": 0.10567663609981537,
|
|
"num_tokens": 2191963.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.397964954376221,
|
|
"epoch": 0.06642212749082978,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.3973,
|
|
"mean_token_accuracy": 0.10832930952310563,
|
|
"num_tokens": 2202620.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.533894824981689,
|
|
"epoch": 0.0667525858365553,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.4228,
|
|
"mean_token_accuracy": 0.10299602895975113,
|
|
"num_tokens": 2213271.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.368393087387085,
|
|
"epoch": 0.06708304418228082,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.4129,
|
|
"mean_token_accuracy": 0.09798410907387733,
|
|
"num_tokens": 2223556.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.429405403137207,
|
|
"epoch": 0.06741350252800635,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.4296,
|
|
"mean_token_accuracy": 0.10144790187478066,
|
|
"num_tokens": 2233866.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.420916748046875,
|
|
"epoch": 0.06774396087373187,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.442,
|
|
"mean_token_accuracy": 0.10680803060531616,
|
|
"num_tokens": 2246388.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.485070371627808,
|
|
"epoch": 0.0680744192194574,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.4173,
|
|
"mean_token_accuracy": 0.11337929293513298,
|
|
"num_tokens": 2257799.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.429631853103638,
|
|
"epoch": 0.0684048775651829,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.4513,
|
|
"mean_token_accuracy": 0.09858574867248535,
|
|
"num_tokens": 2269320.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.250541496276855,
|
|
"epoch": 0.06873533591090843,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.299,
|
|
"mean_token_accuracy": 0.10207094028592109,
|
|
"num_tokens": 2280002.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.494193410873413,
|
|
"epoch": 0.06906579425663395,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.491,
|
|
"mean_token_accuracy": 0.1080724485218525,
|
|
"num_tokens": 2291698.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.4392985820770265,
|
|
"epoch": 0.06939625260235947,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.4857,
|
|
"mean_token_accuracy": 0.10086370706558227,
|
|
"num_tokens": 2302946.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.394264602661133,
|
|
"epoch": 0.069726710948085,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.4254,
|
|
"mean_token_accuracy": 0.10714520588517189,
|
|
"num_tokens": 2312556.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.5946221351623535,
|
|
"epoch": 0.07005716929381052,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.5625,
|
|
"mean_token_accuracy": 0.10093791782855988,
|
|
"num_tokens": 2325099.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.45420503616333,
|
|
"epoch": 0.07038762763953604,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.451,
|
|
"mean_token_accuracy": 0.10265908166766166,
|
|
"num_tokens": 2336561.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.294920635223389,
|
|
"epoch": 0.07071808598526155,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.3643,
|
|
"mean_token_accuracy": 0.1072157271206379,
|
|
"num_tokens": 2346264.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.44938645362854,
|
|
"epoch": 0.07104854433098708,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.3667,
|
|
"mean_token_accuracy": 0.1099700540304184,
|
|
"num_tokens": 2355627.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.320340871810913,
|
|
"epoch": 0.0713790026767126,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.3115,
|
|
"mean_token_accuracy": 0.10740844681859016,
|
|
"num_tokens": 2366693.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.422510385513306,
|
|
"epoch": 0.07170946102243812,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.3416,
|
|
"mean_token_accuracy": 0.10623999536037446,
|
|
"num_tokens": 2376954.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.327064418792725,
|
|
"epoch": 0.07203991936816365,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.4336,
|
|
"mean_token_accuracy": 0.09785263687372207,
|
|
"num_tokens": 2388023.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.400377702713013,
|
|
"epoch": 0.07237037771388917,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.3651,
|
|
"mean_token_accuracy": 0.1035020537674427,
|
|
"num_tokens": 2397683.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.3421402931213375,
|
|
"epoch": 0.07270083605961468,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.3374,
|
|
"mean_token_accuracy": 0.10992366895079612,
|
|
"num_tokens": 2408077.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.337559461593628,
|
|
"epoch": 0.0730312944053402,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.3712,
|
|
"mean_token_accuracy": 0.10661208927631378,
|
|
"num_tokens": 2418938.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.382309675216675,
|
|
"epoch": 0.07336175275106573,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.3715,
|
|
"mean_token_accuracy": 0.1041027471423149,
|
|
"num_tokens": 2430562.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.327745485305786,
|
|
"epoch": 0.07369221109679125,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.4583,
|
|
"mean_token_accuracy": 0.10101124793291091,
|
|
"num_tokens": 2442734.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.42706789970398,
|
|
"epoch": 0.07402266944251677,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.3024,
|
|
"mean_token_accuracy": 0.11131583824753762,
|
|
"num_tokens": 2454265.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.202260732650757,
|
|
"epoch": 0.0743531277882423,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.2635,
|
|
"mean_token_accuracy": 0.10871926620602608,
|
|
"num_tokens": 2464545.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.418315887451172,
|
|
"epoch": 0.07468358613396782,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.5126,
|
|
"mean_token_accuracy": 0.09977985322475433,
|
|
"num_tokens": 2476342.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.423433876037597,
|
|
"epoch": 0.07501404447969333,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.4156,
|
|
"mean_token_accuracy": 0.10508456379175186,
|
|
"num_tokens": 2487279.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.342500114440918,
|
|
"epoch": 0.07534450282541885,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.2813,
|
|
"mean_token_accuracy": 0.10875830128788948,
|
|
"num_tokens": 2497503.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.346517181396484,
|
|
"epoch": 0.07567496117114438,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.3795,
|
|
"mean_token_accuracy": 0.10393059179186821,
|
|
"num_tokens": 2507990.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.343197631835937,
|
|
"epoch": 0.0760054195168699,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.3247,
|
|
"mean_token_accuracy": 0.10598115846514702,
|
|
"num_tokens": 2520699.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.326574993133545,
|
|
"epoch": 0.07633587786259542,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.3057,
|
|
"mean_token_accuracy": 0.11388499960303307,
|
|
"num_tokens": 2531027.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.389388084411621,
|
|
"epoch": 0.07666633620832095,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.4589,
|
|
"mean_token_accuracy": 0.10438870191574097,
|
|
"num_tokens": 2543939.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.4396069049835205,
|
|
"epoch": 0.07699679455404647,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.4023,
|
|
"mean_token_accuracy": 0.1009863406419754,
|
|
"num_tokens": 2555721.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.384159183502197,
|
|
"epoch": 0.07732725289977198,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.4926,
|
|
"mean_token_accuracy": 0.10290684588253499,
|
|
"num_tokens": 2567312.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.239294576644897,
|
|
"epoch": 0.0776577112454975,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.238,
|
|
"mean_token_accuracy": 0.10982146114110947,
|
|
"num_tokens": 2577992.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.390829038619995,
|
|
"epoch": 0.07798816959122302,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.4177,
|
|
"mean_token_accuracy": 0.10007556304335594,
|
|
"num_tokens": 2588866.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.311258459091187,
|
|
"epoch": 0.07831862793694855,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.4175,
|
|
"mean_token_accuracy": 0.10478579625487328,
|
|
"num_tokens": 2600053.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.347922992706299,
|
|
"epoch": 0.07864908628267407,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.2668,
|
|
"mean_token_accuracy": 0.10886808261275291,
|
|
"num_tokens": 2610251.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.273433685302734,
|
|
"epoch": 0.0789795446283996,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.3048,
|
|
"mean_token_accuracy": 0.10753131210803986,
|
|
"num_tokens": 2621316.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.365851497650146,
|
|
"epoch": 0.07931000297412512,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.3617,
|
|
"mean_token_accuracy": 0.11127811819314956,
|
|
"num_tokens": 2632439.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.370825147628784,
|
|
"epoch": 0.07964046131985063,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.3267,
|
|
"mean_token_accuracy": 0.10939299240708351,
|
|
"num_tokens": 2644006.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.228281831741333,
|
|
"epoch": 0.07997091966557615,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.304,
|
|
"mean_token_accuracy": 0.10606826171278953,
|
|
"num_tokens": 2655434.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.371251440048217,
|
|
"epoch": 0.08030137801130167,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.3195,
|
|
"mean_token_accuracy": 0.11309513822197914,
|
|
"num_tokens": 2666521.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.23853235244751,
|
|
"epoch": 0.0806318363570272,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.3228,
|
|
"mean_token_accuracy": 0.10499642714858055,
|
|
"num_tokens": 2678334.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.331731510162354,
|
|
"epoch": 0.08096229470275272,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.361,
|
|
"mean_token_accuracy": 0.10780482292175293,
|
|
"num_tokens": 2690261.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.356092739105224,
|
|
"epoch": 0.08129275304847824,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.2904,
|
|
"mean_token_accuracy": 0.11208009794354439,
|
|
"num_tokens": 2702579.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.319092178344727,
|
|
"epoch": 0.08162321139420375,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.2844,
|
|
"mean_token_accuracy": 0.11137590855360031,
|
|
"num_tokens": 2713336.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.241940927505493,
|
|
"epoch": 0.08195366973992928,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.2338,
|
|
"mean_token_accuracy": 0.11121944114565849,
|
|
"num_tokens": 2724578.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.193857049942016,
|
|
"epoch": 0.0822841280856548,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.2239,
|
|
"mean_token_accuracy": 0.11443476602435113,
|
|
"num_tokens": 2736160.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.269001722335815,
|
|
"epoch": 0.08261458643138032,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.2598,
|
|
"mean_token_accuracy": 0.1093744695186615,
|
|
"num_tokens": 2748370.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.245749282836914,
|
|
"epoch": 0.08294504477710585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.2235,
|
|
"mean_token_accuracy": 0.11153279915452004,
|
|
"num_tokens": 2758556.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.228459167480469,
|
|
"epoch": 0.08327550312283137,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.2554,
|
|
"mean_token_accuracy": 0.10969504565000535,
|
|
"num_tokens": 2770110.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.314351272583008,
|
|
"epoch": 0.0836059614685569,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.3025,
|
|
"mean_token_accuracy": 0.10905899181962013,
|
|
"num_tokens": 2782226.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.236157655715942,
|
|
"epoch": 0.0839364198142824,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.1767,
|
|
"mean_token_accuracy": 0.11525296643376351,
|
|
"num_tokens": 2791815.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.145106315612793,
|
|
"epoch": 0.08426687816000793,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.2014,
|
|
"mean_token_accuracy": 0.11409328505396843,
|
|
"num_tokens": 2802472.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.254790163040161,
|
|
"epoch": 0.08459733650573345,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.1966,
|
|
"mean_token_accuracy": 0.1131843589246273,
|
|
"num_tokens": 2812262.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.218746042251587,
|
|
"epoch": 0.08492779485145897,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.1711,
|
|
"mean_token_accuracy": 0.11364943832159043,
|
|
"num_tokens": 2821806.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.2752289295196535,
|
|
"epoch": 0.0852582531971845,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.292,
|
|
"mean_token_accuracy": 0.10570647865533829,
|
|
"num_tokens": 2832544.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.2815282344818115,
|
|
"epoch": 0.08558871154291002,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.3097,
|
|
"mean_token_accuracy": 0.10787594094872474,
|
|
"num_tokens": 2843347.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.27464919090271,
|
|
"epoch": 0.08591916988863554,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.2873,
|
|
"mean_token_accuracy": 0.11007290184497834,
|
|
"num_tokens": 2854361.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.28824667930603,
|
|
"epoch": 0.08624962823436105,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.2942,
|
|
"mean_token_accuracy": 0.11148815751075744,
|
|
"num_tokens": 2865716.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.313663482666016,
|
|
"epoch": 0.08658008658008658,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.2973,
|
|
"mean_token_accuracy": 0.11001308336853981,
|
|
"num_tokens": 2876409.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.236247873306274,
|
|
"epoch": 0.0869105449258121,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.1624,
|
|
"mean_token_accuracy": 0.11299015209078789,
|
|
"num_tokens": 2885964.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.212308168411255,
|
|
"epoch": 0.08724100327153762,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.2009,
|
|
"mean_token_accuracy": 0.11274494603276253,
|
|
"num_tokens": 2896486.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.227869939804077,
|
|
"epoch": 0.08757146161726315,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.3043,
|
|
"mean_token_accuracy": 0.10914059206843377,
|
|
"num_tokens": 2907216.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.188024950027466,
|
|
"epoch": 0.08790191996298867,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.223,
|
|
"mean_token_accuracy": 0.10957143902778625,
|
|
"num_tokens": 2916975.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.274955558776855,
|
|
"epoch": 0.0882323783087142,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.2648,
|
|
"mean_token_accuracy": 0.1095338560640812,
|
|
"num_tokens": 2927824.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.2339982986450195,
|
|
"epoch": 0.0885628366544397,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.2159,
|
|
"mean_token_accuracy": 0.11319034472107888,
|
|
"num_tokens": 2938311.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.248281240463257,
|
|
"epoch": 0.08889329500016523,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.2644,
|
|
"mean_token_accuracy": 0.11580364927649497,
|
|
"num_tokens": 2948504.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.243304252624512,
|
|
"epoch": 0.08922375334589075,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.2667,
|
|
"mean_token_accuracy": 0.10695719793438911,
|
|
"num_tokens": 2959414.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.278574228286743,
|
|
"epoch": 0.08955421169161627,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.2457,
|
|
"mean_token_accuracy": 0.11053696125745774,
|
|
"num_tokens": 2969739.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.252624797821045,
|
|
"epoch": 0.0898846700373418,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.2745,
|
|
"mean_token_accuracy": 0.10981930121779442,
|
|
"num_tokens": 2980618.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.179713249206543,
|
|
"epoch": 0.09021512838306732,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.2588,
|
|
"mean_token_accuracy": 0.11370926424860954,
|
|
"num_tokens": 2993005.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.325649785995483,
|
|
"epoch": 0.09054558672879283,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.2234,
|
|
"mean_token_accuracy": 0.1116867408156395,
|
|
"num_tokens": 3003951.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.122556161880493,
|
|
"epoch": 0.09087604507451835,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.1785,
|
|
"mean_token_accuracy": 0.1196735866367817,
|
|
"num_tokens": 3014347.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.27095890045166,
|
|
"epoch": 0.09120650342024388,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.2813,
|
|
"mean_token_accuracy": 0.11019450649619103,
|
|
"num_tokens": 3025938.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.216333627700806,
|
|
"epoch": 0.0915369617659694,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.23,
|
|
"mean_token_accuracy": 0.11627952679991722,
|
|
"num_tokens": 3036774.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.2260870933532715,
|
|
"epoch": 0.09186742011169492,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.2357,
|
|
"mean_token_accuracy": 0.1122400440275669,
|
|
"num_tokens": 3048767.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.2184850692749025,
|
|
"epoch": 0.09219787845742045,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.1944,
|
|
"mean_token_accuracy": 0.10921861827373505,
|
|
"num_tokens": 3058632.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.121122741699219,
|
|
"epoch": 0.09252833680314597,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.0642,
|
|
"mean_token_accuracy": 0.12121142446994781,
|
|
"num_tokens": 3069288.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.113697719573975,
|
|
"epoch": 0.09285879514887148,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.1353,
|
|
"mean_token_accuracy": 0.10769173875451088,
|
|
"num_tokens": 3079060.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.21345329284668,
|
|
"epoch": 0.093189253494597,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.2299,
|
|
"mean_token_accuracy": 0.10818874016404152,
|
|
"num_tokens": 3089864.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.25667405128479,
|
|
"epoch": 0.09351971184032253,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.2142,
|
|
"mean_token_accuracy": 0.10830713286995888,
|
|
"num_tokens": 3099968.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.209473514556885,
|
|
"epoch": 0.09385017018604805,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.1982,
|
|
"mean_token_accuracy": 0.10743900611996651,
|
|
"num_tokens": 3110354.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.180089139938355,
|
|
"epoch": 0.09418062853177357,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.1477,
|
|
"mean_token_accuracy": 0.11282439082860947,
|
|
"num_tokens": 3121638.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.140780687332153,
|
|
"epoch": 0.0945110868774991,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.1543,
|
|
"mean_token_accuracy": 0.11213898211717606,
|
|
"num_tokens": 3132609.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.257613754272461,
|
|
"epoch": 0.09484154522322462,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.2194,
|
|
"mean_token_accuracy": 0.11128115206956864,
|
|
"num_tokens": 3144445.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.14837908744812,
|
|
"epoch": 0.09517200356895013,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.1785,
|
|
"mean_token_accuracy": 0.10812792181968689,
|
|
"num_tokens": 3154942.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.268062877655029,
|
|
"epoch": 0.09550246191467565,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.1964,
|
|
"mean_token_accuracy": 0.11239229813218117,
|
|
"num_tokens": 3165444.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.080770492553711,
|
|
"epoch": 0.09583292026040117,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.1447,
|
|
"mean_token_accuracy": 0.10863324329257011,
|
|
"num_tokens": 3175228.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.161494970321655,
|
|
"epoch": 0.0961633786061267,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.1406,
|
|
"mean_token_accuracy": 0.11648099794983864,
|
|
"num_tokens": 3186028.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.114691972732544,
|
|
"epoch": 0.09649383695185222,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.1006,
|
|
"mean_token_accuracy": 0.12160674259066581,
|
|
"num_tokens": 3196351.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.209639406204223,
|
|
"epoch": 0.09682429529757774,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.1886,
|
|
"mean_token_accuracy": 0.11488557681441307,
|
|
"num_tokens": 3207788.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.155808544158935,
|
|
"epoch": 0.09715475364330327,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.2009,
|
|
"mean_token_accuracy": 0.11256440505385398,
|
|
"num_tokens": 3219941.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.30653486251831,
|
|
"epoch": 0.09748521198902878,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.1727,
|
|
"mean_token_accuracy": 0.11391912326216698,
|
|
"num_tokens": 3229199.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.140384721755981,
|
|
"epoch": 0.0978156703347543,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.267,
|
|
"mean_token_accuracy": 0.11459414213895798,
|
|
"num_tokens": 3239414.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.296479082107544,
|
|
"epoch": 0.09814612868047982,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.2115,
|
|
"mean_token_accuracy": 0.10864720344543458,
|
|
"num_tokens": 3250403.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.155523252487183,
|
|
"epoch": 0.09847658702620535,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.2497,
|
|
"mean_token_accuracy": 0.11157322898507119,
|
|
"num_tokens": 3262579.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.1816212177276615,
|
|
"epoch": 0.09880704537193087,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.1422,
|
|
"mean_token_accuracy": 0.11603116244077682,
|
|
"num_tokens": 3273750.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.107717561721802,
|
|
"epoch": 0.0991375037176564,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.1289,
|
|
"mean_token_accuracy": 0.12265687435865402,
|
|
"num_tokens": 3283996.0,
|
|
"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": 5497780322304000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|