4035 lines
109 KiB
JSON
4035 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.11706175007316359,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742143154144287,
|
|
"epoch": 0.000292654375182909,
|
|
"grad_norm": 5.34375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7851,
|
|
"mean_token_accuracy": 7.092198356986045e-05,
|
|
"num_tokens": 10253.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742099285125732,
|
|
"epoch": 0.000585308750365818,
|
|
"grad_norm": 5.09375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7585,
|
|
"mean_token_accuracy": 7.072135922499001e-05,
|
|
"num_tokens": 19586.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742091464996339,
|
|
"epoch": 0.000877963125548727,
|
|
"grad_norm": 5.71875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7338,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 28743.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742110919952392,
|
|
"epoch": 0.001170617500731636,
|
|
"grad_norm": 4.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.702,
|
|
"mean_token_accuracy": 7.518797065131366e-05,
|
|
"num_tokens": 39263.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.7420015335083,
|
|
"epoch": 0.0014632718759145448,
|
|
"grad_norm": 5.0625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.6176,
|
|
"mean_token_accuracy": 0.00650847667711787,
|
|
"num_tokens": 50195.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.741524696350098,
|
|
"epoch": 0.001755926251097454,
|
|
"grad_norm": 4.5,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4398,
|
|
"mean_token_accuracy": 0.04643158838152885,
|
|
"num_tokens": 58982.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.739701747894287,
|
|
"epoch": 0.002048580626280363,
|
|
"grad_norm": 3.375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3385,
|
|
"mean_token_accuracy": 0.04858696423470974,
|
|
"num_tokens": 68619.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.733176803588867,
|
|
"epoch": 0.002341235001463272,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1755,
|
|
"mean_token_accuracy": 0.054676258191466334,
|
|
"num_tokens": 78276.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.719074249267578,
|
|
"epoch": 0.0026338893766461808,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.085,
|
|
"mean_token_accuracy": 0.05106211043894291,
|
|
"num_tokens": 87977.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.702382278442382,
|
|
"epoch": 0.0029265437518290896,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.007,
|
|
"mean_token_accuracy": 0.05418388731777668,
|
|
"num_tokens": 97168.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.688697719573975,
|
|
"epoch": 0.003219198127011999,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.978,
|
|
"mean_token_accuracy": 0.0560054711997509,
|
|
"num_tokens": 108162.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.677631187438966,
|
|
"epoch": 0.003511852502194908,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8907,
|
|
"mean_token_accuracy": 0.05185089595615864,
|
|
"num_tokens": 117963.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.672104358673096,
|
|
"epoch": 0.0038045068773778167,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8844,
|
|
"mean_token_accuracy": 0.05049301944673061,
|
|
"num_tokens": 128067.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.665418434143067,
|
|
"epoch": 0.004097161252560726,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8046,
|
|
"mean_token_accuracy": 0.05261277854442596,
|
|
"num_tokens": 138347.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.655702877044678,
|
|
"epoch": 0.004389815627743635,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7713,
|
|
"mean_token_accuracy": 0.05200894549489021,
|
|
"num_tokens": 148883.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.648012065887452,
|
|
"epoch": 0.004682470002926544,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.7304,
|
|
"mean_token_accuracy": 0.053603590652346614,
|
|
"num_tokens": 158416.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.638986778259277,
|
|
"epoch": 0.004975124378109453,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.624,
|
|
"mean_token_accuracy": 0.056516367569565774,
|
|
"num_tokens": 168380.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.622650527954102,
|
|
"epoch": 0.0052677787532923615,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.5594,
|
|
"mean_token_accuracy": 0.058524899929761884,
|
|
"num_tokens": 178623.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.598152160644531,
|
|
"epoch": 0.00556043312847527,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.5209,
|
|
"mean_token_accuracy": 0.05661297850310802,
|
|
"num_tokens": 189058.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.577679920196534,
|
|
"epoch": 0.005853087503658179,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.4548,
|
|
"mean_token_accuracy": 0.056520416587591174,
|
|
"num_tokens": 199465.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.56387767791748,
|
|
"epoch": 0.006145741878841089,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.4,
|
|
"mean_token_accuracy": 0.054618718847632405,
|
|
"num_tokens": 208712.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.543546962738038,
|
|
"epoch": 0.006438396254023998,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.3224,
|
|
"mean_token_accuracy": 0.05899658054113388,
|
|
"num_tokens": 218557.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.48940305709839,
|
|
"epoch": 0.006731050629206907,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.2068,
|
|
"mean_token_accuracy": 0.06870373338460922,
|
|
"num_tokens": 228961.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.402587509155273,
|
|
"epoch": 0.007023705004389816,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.1227,
|
|
"mean_token_accuracy": 0.06810536533594132,
|
|
"num_tokens": 238987.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.348681640625,
|
|
"epoch": 0.0073163593795727245,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.0633,
|
|
"mean_token_accuracy": 0.0605758685618639,
|
|
"num_tokens": 249600.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.30824089050293,
|
|
"epoch": 0.007609013754755633,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.8758,
|
|
"mean_token_accuracy": 0.07409285828471183,
|
|
"num_tokens": 259215.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.212008476257324,
|
|
"epoch": 0.007901668129938543,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.8303,
|
|
"mean_token_accuracy": 0.06557713933289051,
|
|
"num_tokens": 268888.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.150605010986329,
|
|
"epoch": 0.008194322505121452,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.74,
|
|
"mean_token_accuracy": 0.06668962053954601,
|
|
"num_tokens": 278451.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.055028533935547,
|
|
"epoch": 0.008486976880304361,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.7012,
|
|
"mean_token_accuracy": 0.06709126047790051,
|
|
"num_tokens": 289577.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.99300413131714,
|
|
"epoch": 0.00877963125548727,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.5708,
|
|
"mean_token_accuracy": 0.06588217578828334,
|
|
"num_tokens": 299057.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.817628383636475,
|
|
"epoch": 0.009072285630670179,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.4885,
|
|
"mean_token_accuracy": 0.06879487000405789,
|
|
"num_tokens": 308802.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.719903469085693,
|
|
"epoch": 0.009364940005853088,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.3738,
|
|
"mean_token_accuracy": 0.07012738101184368,
|
|
"num_tokens": 317365.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.5343448638916,
|
|
"epoch": 0.009657594381035996,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.3035,
|
|
"mean_token_accuracy": 0.07032971270382404,
|
|
"num_tokens": 326927.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.404919242858886,
|
|
"epoch": 0.009950248756218905,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.317,
|
|
"mean_token_accuracy": 0.06643726415932179,
|
|
"num_tokens": 337408.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.28181962966919,
|
|
"epoch": 0.010242903131401814,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.2328,
|
|
"mean_token_accuracy": 0.06949336491525174,
|
|
"num_tokens": 348282.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.099757957458497,
|
|
"epoch": 0.010535557506584723,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.1847,
|
|
"mean_token_accuracy": 0.06538042239844799,
|
|
"num_tokens": 357767.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.947648048400879,
|
|
"epoch": 0.010828211881767632,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.1401,
|
|
"mean_token_accuracy": 0.07173748835921287,
|
|
"num_tokens": 368665.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.832690525054932,
|
|
"epoch": 0.01112086625695054,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.125,
|
|
"mean_token_accuracy": 0.0688393272459507,
|
|
"num_tokens": 378741.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.722904491424561,
|
|
"epoch": 0.01141352063213345,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.1228,
|
|
"mean_token_accuracy": 0.07389259040355682,
|
|
"num_tokens": 388324.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.738927555084228,
|
|
"epoch": 0.011706175007316359,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.1796,
|
|
"mean_token_accuracy": 0.06952995508909225,
|
|
"num_tokens": 398450.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.615283966064453,
|
|
"epoch": 0.01199882938249927,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.0899,
|
|
"mean_token_accuracy": 0.07332116328179836,
|
|
"num_tokens": 408961.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.581227874755859,
|
|
"epoch": 0.012291483757682178,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.1226,
|
|
"mean_token_accuracy": 0.0683289546519518,
|
|
"num_tokens": 418517.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.532517910003662,
|
|
"epoch": 0.012584138132865087,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.035,
|
|
"mean_token_accuracy": 0.07064232043921947,
|
|
"num_tokens": 428059.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.60868797302246,
|
|
"epoch": 0.012876792508047996,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.0513,
|
|
"mean_token_accuracy": 0.06911276690661908,
|
|
"num_tokens": 438499.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.50686674118042,
|
|
"epoch": 0.013169446883230905,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.0542,
|
|
"mean_token_accuracy": 0.06983353272080421,
|
|
"num_tokens": 448761.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.424305152893066,
|
|
"epoch": 0.013462101258413814,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.0471,
|
|
"mean_token_accuracy": 0.07090565152466297,
|
|
"num_tokens": 458386.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.447117614746094,
|
|
"epoch": 0.013754755633596722,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.991,
|
|
"mean_token_accuracy": 0.06929317899048329,
|
|
"num_tokens": 468198.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.46488447189331,
|
|
"epoch": 0.014047410008779631,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.0203,
|
|
"mean_token_accuracy": 0.07129090093076229,
|
|
"num_tokens": 480142.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.370715618133545,
|
|
"epoch": 0.01434006438396254,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.8998,
|
|
"mean_token_accuracy": 0.08097687140107154,
|
|
"num_tokens": 490686.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.408738231658935,
|
|
"epoch": 0.014632718759145449,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.9758,
|
|
"mean_token_accuracy": 0.07468959763646126,
|
|
"num_tokens": 500099.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.416247749328614,
|
|
"epoch": 0.014925373134328358,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.9005,
|
|
"mean_token_accuracy": 0.07769993916153908,
|
|
"num_tokens": 509525.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.347339820861816,
|
|
"epoch": 0.015218027509511267,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.8978,
|
|
"mean_token_accuracy": 0.07274740003049374,
|
|
"num_tokens": 520141.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.357079029083252,
|
|
"epoch": 0.015510681884694176,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.0384,
|
|
"mean_token_accuracy": 0.06597915887832642,
|
|
"num_tokens": 531645.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.3228253364563,
|
|
"epoch": 0.015803336259877086,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.9109,
|
|
"mean_token_accuracy": 0.07127318009734154,
|
|
"num_tokens": 541779.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.300227451324464,
|
|
"epoch": 0.016095990635059995,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.8871,
|
|
"mean_token_accuracy": 0.07985152080655097,
|
|
"num_tokens": 551770.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.360063171386718,
|
|
"epoch": 0.016388645010242904,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.8829,
|
|
"mean_token_accuracy": 0.0752606987953186,
|
|
"num_tokens": 561814.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.235567474365235,
|
|
"epoch": 0.016681299385425813,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.9551,
|
|
"mean_token_accuracy": 0.06852101050317287,
|
|
"num_tokens": 571988.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.42728681564331,
|
|
"epoch": 0.016973953760608722,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.9876,
|
|
"mean_token_accuracy": 0.07176770456135273,
|
|
"num_tokens": 582814.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.26517457962036,
|
|
"epoch": 0.01726660813579163,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.9178,
|
|
"mean_token_accuracy": 0.07443804480135441,
|
|
"num_tokens": 592301.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.199700450897216,
|
|
"epoch": 0.01755926251097454,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.8119,
|
|
"mean_token_accuracy": 0.077539586648345,
|
|
"num_tokens": 602898.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.289494514465332,
|
|
"epoch": 0.01785191688615745,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.8914,
|
|
"mean_token_accuracy": 0.07545375041663646,
|
|
"num_tokens": 612731.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.186264896392823,
|
|
"epoch": 0.018144571261340357,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.843,
|
|
"mean_token_accuracy": 0.07743252031505107,
|
|
"num_tokens": 623067.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.256502819061279,
|
|
"epoch": 0.018437225636523266,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.8317,
|
|
"mean_token_accuracy": 0.08097588121891022,
|
|
"num_tokens": 633538.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.319536399841308,
|
|
"epoch": 0.018729880011706175,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.8898,
|
|
"mean_token_accuracy": 0.07923161759972572,
|
|
"num_tokens": 643704.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.158168601989747,
|
|
"epoch": 0.019022534386889084,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.9204,
|
|
"mean_token_accuracy": 0.07442782260477543,
|
|
"num_tokens": 653766.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.432480907440185,
|
|
"epoch": 0.019315188762071993,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.9175,
|
|
"mean_token_accuracy": 0.07209107242524623,
|
|
"num_tokens": 663289.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.145325660705566,
|
|
"epoch": 0.0196078431372549,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.8214,
|
|
"mean_token_accuracy": 0.07566024884581565,
|
|
"num_tokens": 673196.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.328914070129395,
|
|
"epoch": 0.01990049751243781,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.8481,
|
|
"mean_token_accuracy": 0.0807725053280592,
|
|
"num_tokens": 682694.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.15658712387085,
|
|
"epoch": 0.02019315188762072,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.8414,
|
|
"mean_token_accuracy": 0.07744252718985081,
|
|
"num_tokens": 692880.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.265804481506347,
|
|
"epoch": 0.02048580626280363,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.9339,
|
|
"mean_token_accuracy": 0.07408964931964875,
|
|
"num_tokens": 703972.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.137424850463868,
|
|
"epoch": 0.020778460637986537,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.7467,
|
|
"mean_token_accuracy": 0.07930146306753158,
|
|
"num_tokens": 713571.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.209496688842773,
|
|
"epoch": 0.021071115013169446,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.8355,
|
|
"mean_token_accuracy": 0.08222270905971527,
|
|
"num_tokens": 723642.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.136897945404053,
|
|
"epoch": 0.021363769388352355,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.8793,
|
|
"mean_token_accuracy": 0.07310881391167641,
|
|
"num_tokens": 733848.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.176177358627319,
|
|
"epoch": 0.021656423763535264,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.6089,
|
|
"mean_token_accuracy": 0.10752813518047333,
|
|
"num_tokens": 743752.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.144378280639648,
|
|
"epoch": 0.021949078138718173,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.8031,
|
|
"mean_token_accuracy": 0.08254543766379356,
|
|
"num_tokens": 754145.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.118795394897461,
|
|
"epoch": 0.02224173251390108,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.766,
|
|
"mean_token_accuracy": 0.07592462375760078,
|
|
"num_tokens": 763930.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.178821086883545,
|
|
"epoch": 0.02253438688908399,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.7934,
|
|
"mean_token_accuracy": 0.08122270852327347,
|
|
"num_tokens": 774851.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.122542095184325,
|
|
"epoch": 0.0228270412642669,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.7915,
|
|
"mean_token_accuracy": 0.07377454079687595,
|
|
"num_tokens": 784251.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.100419902801514,
|
|
"epoch": 0.023119695639449808,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.8395,
|
|
"mean_token_accuracy": 0.0817381426692009,
|
|
"num_tokens": 794089.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.19840030670166,
|
|
"epoch": 0.023412350014632717,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.8242,
|
|
"mean_token_accuracy": 0.07439705729484558,
|
|
"num_tokens": 805274.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.17535228729248,
|
|
"epoch": 0.02370500438981563,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.7672,
|
|
"mean_token_accuracy": 0.07779609337449074,
|
|
"num_tokens": 816036.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.004063892364503,
|
|
"epoch": 0.02399765876499854,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.7613,
|
|
"mean_token_accuracy": 0.07725987881422043,
|
|
"num_tokens": 826393.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.159784364700318,
|
|
"epoch": 0.024290313140181447,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.7542,
|
|
"mean_token_accuracy": 0.0782277960330248,
|
|
"num_tokens": 836456.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.072831869125366,
|
|
"epoch": 0.024582967515364356,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.759,
|
|
"mean_token_accuracy": 0.07723330594599247,
|
|
"num_tokens": 846493.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.16247329711914,
|
|
"epoch": 0.024875621890547265,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.7584,
|
|
"mean_token_accuracy": 0.07505144067108631,
|
|
"num_tokens": 857041.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.9756200313568115,
|
|
"epoch": 0.025168276265730174,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.6788,
|
|
"mean_token_accuracy": 0.08394450098276138,
|
|
"num_tokens": 867053.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.086863994598389,
|
|
"epoch": 0.025460930640913083,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.7398,
|
|
"mean_token_accuracy": 0.0781655989587307,
|
|
"num_tokens": 876944.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.080809020996094,
|
|
"epoch": 0.02575358501609599,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.6902,
|
|
"mean_token_accuracy": 0.08105823285877704,
|
|
"num_tokens": 887257.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.091656923294067,
|
|
"epoch": 0.0260462393912789,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.7086,
|
|
"mean_token_accuracy": 0.0815083347260952,
|
|
"num_tokens": 896877.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.037679195404053,
|
|
"epoch": 0.02633889376646181,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.6096,
|
|
"mean_token_accuracy": 0.08536934405565262,
|
|
"num_tokens": 906930.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.04052267074585,
|
|
"epoch": 0.026631548141644718,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.7471,
|
|
"mean_token_accuracy": 0.07020874097943305,
|
|
"num_tokens": 916841.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.061063146591186,
|
|
"epoch": 0.026924202516827627,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.6914,
|
|
"mean_token_accuracy": 0.07996720150113105,
|
|
"num_tokens": 927024.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.085123777389526,
|
|
"epoch": 0.027216856892010536,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.7227,
|
|
"mean_token_accuracy": 0.07815472632646561,
|
|
"num_tokens": 935786.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.955375385284424,
|
|
"epoch": 0.027509511267193445,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.6349,
|
|
"mean_token_accuracy": 0.08305523693561553,
|
|
"num_tokens": 946819.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.01088285446167,
|
|
"epoch": 0.027802165642376354,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.6793,
|
|
"mean_token_accuracy": 0.08371292352676392,
|
|
"num_tokens": 957876.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.070115327835083,
|
|
"epoch": 0.028094820017559263,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.7627,
|
|
"mean_token_accuracy": 0.07754282280802727,
|
|
"num_tokens": 969211.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.086713933944703,
|
|
"epoch": 0.02838747439274217,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.7091,
|
|
"mean_token_accuracy": 0.07998017743229865,
|
|
"num_tokens": 979090.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.120463180541993,
|
|
"epoch": 0.02868012876792508,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.7649,
|
|
"mean_token_accuracy": 0.07725568227469921,
|
|
"num_tokens": 990510.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.065446138381958,
|
|
"epoch": 0.02897278314310799,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.6837,
|
|
"mean_token_accuracy": 0.07927028760313988,
|
|
"num_tokens": 1000890.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.89040241241455,
|
|
"epoch": 0.029265437518290898,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.633,
|
|
"mean_token_accuracy": 0.08539719134569168,
|
|
"num_tokens": 1011019.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.053290367126465,
|
|
"epoch": 0.029558091893473807,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.5976,
|
|
"mean_token_accuracy": 0.0852201983332634,
|
|
"num_tokens": 1020703.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.924099731445312,
|
|
"epoch": 0.029850746268656716,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.6144,
|
|
"mean_token_accuracy": 0.08518284186720848,
|
|
"num_tokens": 1030539.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.876818609237671,
|
|
"epoch": 0.030143400643839625,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.6173,
|
|
"mean_token_accuracy": 0.07879232838749886,
|
|
"num_tokens": 1040977.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.9826685905456545,
|
|
"epoch": 0.030436055019022534,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.5867,
|
|
"mean_token_accuracy": 0.08224296122789383,
|
|
"num_tokens": 1050382.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.914853572845459,
|
|
"epoch": 0.030728709394205442,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.5776,
|
|
"mean_token_accuracy": 0.08655883669853211,
|
|
"num_tokens": 1062287.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.873267793655396,
|
|
"epoch": 0.03102136376938835,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.5624,
|
|
"mean_token_accuracy": 0.0834833487868309,
|
|
"num_tokens": 1072138.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.053305435180665,
|
|
"epoch": 0.031314018144571264,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.5478,
|
|
"mean_token_accuracy": 0.08701685890555381,
|
|
"num_tokens": 1082730.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.881413888931275,
|
|
"epoch": 0.03160667251975417,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.5191,
|
|
"mean_token_accuracy": 0.08792314156889916,
|
|
"num_tokens": 1093409.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.940295267105102,
|
|
"epoch": 0.03189932689493708,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.6596,
|
|
"mean_token_accuracy": 0.08377403020858765,
|
|
"num_tokens": 1103817.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.939820003509522,
|
|
"epoch": 0.03219198127011999,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.6265,
|
|
"mean_token_accuracy": 0.07738926894962787,
|
|
"num_tokens": 1114392.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.96156415939331,
|
|
"epoch": 0.0324846356453029,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.5224,
|
|
"mean_token_accuracy": 0.08881102874875069,
|
|
"num_tokens": 1124539.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.832987546920776,
|
|
"epoch": 0.03277729002048581,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.5564,
|
|
"mean_token_accuracy": 0.08254684880375862,
|
|
"num_tokens": 1135110.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.855937910079956,
|
|
"epoch": 0.03306994439566872,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.5979,
|
|
"mean_token_accuracy": 0.08505113944411277,
|
|
"num_tokens": 1145458.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.951333904266358,
|
|
"epoch": 0.033362598770851626,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.5383,
|
|
"mean_token_accuracy": 0.08062238246202469,
|
|
"num_tokens": 1155754.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.8793701171875,
|
|
"epoch": 0.033655253146034535,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.5816,
|
|
"mean_token_accuracy": 0.082074723392725,
|
|
"num_tokens": 1165771.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.901887035369873,
|
|
"epoch": 0.033947907521217444,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.534,
|
|
"mean_token_accuracy": 0.08429761826992035,
|
|
"num_tokens": 1175785.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.887288761138916,
|
|
"epoch": 0.03424056189640035,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.5557,
|
|
"mean_token_accuracy": 0.08287644758820534,
|
|
"num_tokens": 1186252.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.767520141601563,
|
|
"epoch": 0.03453321627158326,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.5037,
|
|
"mean_token_accuracy": 0.08624404892325402,
|
|
"num_tokens": 1196800.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.92415132522583,
|
|
"epoch": 0.03482587064676617,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.4895,
|
|
"mean_token_accuracy": 0.08825775608420372,
|
|
"num_tokens": 1207794.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.7734299659729,
|
|
"epoch": 0.03511852502194908,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.4153,
|
|
"mean_token_accuracy": 0.08845552653074265,
|
|
"num_tokens": 1217029.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.832931613922119,
|
|
"epoch": 0.03541117939713199,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.3933,
|
|
"mean_token_accuracy": 0.09399922043085099,
|
|
"num_tokens": 1227917.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.771529722213745,
|
|
"epoch": 0.0357038337723149,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.4464,
|
|
"mean_token_accuracy": 0.08992072939872742,
|
|
"num_tokens": 1238495.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.831154489517212,
|
|
"epoch": 0.035996488147497806,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.4915,
|
|
"mean_token_accuracy": 0.08896522894501686,
|
|
"num_tokens": 1249649.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.815850448608399,
|
|
"epoch": 0.036289142522680715,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.5417,
|
|
"mean_token_accuracy": 0.08101370558142662,
|
|
"num_tokens": 1260937.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.813495063781739,
|
|
"epoch": 0.036581796897863623,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.3702,
|
|
"mean_token_accuracy": 0.09065580368041992,
|
|
"num_tokens": 1270751.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.818518209457397,
|
|
"epoch": 0.03687445127304653,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.4965,
|
|
"mean_token_accuracy": 0.0885044053196907,
|
|
"num_tokens": 1282383.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.83252854347229,
|
|
"epoch": 0.03716710564822944,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.3179,
|
|
"mean_token_accuracy": 0.1026477910578251,
|
|
"num_tokens": 1292320.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.6734686374664305,
|
|
"epoch": 0.03745976002341235,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.4792,
|
|
"mean_token_accuracy": 0.08322202041745186,
|
|
"num_tokens": 1303054.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.8633969783782955,
|
|
"epoch": 0.03775241439859526,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.4465,
|
|
"mean_token_accuracy": 0.08914806470274925,
|
|
"num_tokens": 1311607.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.689844703674316,
|
|
"epoch": 0.03804506877377817,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.4234,
|
|
"mean_token_accuracy": 0.08909434452652931,
|
|
"num_tokens": 1321930.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.8022431373596195,
|
|
"epoch": 0.03833772314896108,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.4465,
|
|
"mean_token_accuracy": 0.08396812826395035,
|
|
"num_tokens": 1331533.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.769197797775268,
|
|
"epoch": 0.038630377524143986,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.4086,
|
|
"mean_token_accuracy": 0.08934582769870758,
|
|
"num_tokens": 1341961.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.769333171844482,
|
|
"epoch": 0.038923031899326895,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.4375,
|
|
"mean_token_accuracy": 0.08935849741101265,
|
|
"num_tokens": 1352741.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.791517686843872,
|
|
"epoch": 0.0392156862745098,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.432,
|
|
"mean_token_accuracy": 0.08128280714154243,
|
|
"num_tokens": 1363467.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.691935110092163,
|
|
"epoch": 0.03950834064969271,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.3877,
|
|
"mean_token_accuracy": 0.09103676527738572,
|
|
"num_tokens": 1372759.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.6941760063171385,
|
|
"epoch": 0.03980099502487562,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.3855,
|
|
"mean_token_accuracy": 0.09363519176840782,
|
|
"num_tokens": 1382785.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.757591485977173,
|
|
"epoch": 0.04009364940005853,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.4335,
|
|
"mean_token_accuracy": 0.0862673670053482,
|
|
"num_tokens": 1392748.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.777906894683838,
|
|
"epoch": 0.04038630377524144,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.4677,
|
|
"mean_token_accuracy": 0.08610242903232575,
|
|
"num_tokens": 1402473.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.787166547775269,
|
|
"epoch": 0.04067895815042435,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.439,
|
|
"mean_token_accuracy": 0.08554250001907349,
|
|
"num_tokens": 1413921.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.736286449432373,
|
|
"epoch": 0.04097161252560726,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.4369,
|
|
"mean_token_accuracy": 0.0805517353117466,
|
|
"num_tokens": 1424687.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.767450380325317,
|
|
"epoch": 0.041264266900790166,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.4388,
|
|
"mean_token_accuracy": 0.08601183965802192,
|
|
"num_tokens": 1435153.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.794674873352051,
|
|
"epoch": 0.041556921275973074,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.4462,
|
|
"mean_token_accuracy": 0.0846464328467846,
|
|
"num_tokens": 1446187.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.755715560913086,
|
|
"epoch": 0.04184957565115598,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.396,
|
|
"mean_token_accuracy": 0.08824568465352059,
|
|
"num_tokens": 1456299.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.5965576171875,
|
|
"epoch": 0.04214223002633889,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.3017,
|
|
"mean_token_accuracy": 0.09606124758720398,
|
|
"num_tokens": 1465654.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.70619683265686,
|
|
"epoch": 0.0424348844015218,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.3563,
|
|
"mean_token_accuracy": 0.09394391924142838,
|
|
"num_tokens": 1475248.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.698731327056885,
|
|
"epoch": 0.04272753877670471,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.4247,
|
|
"mean_token_accuracy": 0.08578455671668053,
|
|
"num_tokens": 1485570.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.636994504928589,
|
|
"epoch": 0.04302019315188762,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.3294,
|
|
"mean_token_accuracy": 0.09238605126738549,
|
|
"num_tokens": 1495257.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.687747764587402,
|
|
"epoch": 0.04331284752707053,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.3467,
|
|
"mean_token_accuracy": 0.08985281139612197,
|
|
"num_tokens": 1504882.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.706828022003174,
|
|
"epoch": 0.043605501902253437,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.3918,
|
|
"mean_token_accuracy": 0.08691600039601326,
|
|
"num_tokens": 1515791.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.595960521697998,
|
|
"epoch": 0.043898156277436345,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.3055,
|
|
"mean_token_accuracy": 0.09265599772334099,
|
|
"num_tokens": 1527837.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.674931144714355,
|
|
"epoch": 0.044190810652619254,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.3197,
|
|
"mean_token_accuracy": 0.09090850502252579,
|
|
"num_tokens": 1538357.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.597537708282471,
|
|
"epoch": 0.04448346502780216,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.318,
|
|
"mean_token_accuracy": 0.09620778560638428,
|
|
"num_tokens": 1548140.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.655205726623535,
|
|
"epoch": 0.04477611940298507,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.3052,
|
|
"mean_token_accuracy": 0.09487425237894058,
|
|
"num_tokens": 1558723.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.604353332519532,
|
|
"epoch": 0.04506877377816798,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.3288,
|
|
"mean_token_accuracy": 0.08416497483849525,
|
|
"num_tokens": 1568519.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.608450269699096,
|
|
"epoch": 0.04536142815335089,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.2961,
|
|
"mean_token_accuracy": 0.09094355553388596,
|
|
"num_tokens": 1578788.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.620551347732544,
|
|
"epoch": 0.0456540825285338,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.2536,
|
|
"mean_token_accuracy": 0.09547285735607147,
|
|
"num_tokens": 1588593.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.534824514389038,
|
|
"epoch": 0.04594673690371671,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.2914,
|
|
"mean_token_accuracy": 0.08960797935724259,
|
|
"num_tokens": 1598657.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.504811525344849,
|
|
"epoch": 0.046239391278899616,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.2001,
|
|
"mean_token_accuracy": 0.09868146777153015,
|
|
"num_tokens": 1608825.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.5206996440887455,
|
|
"epoch": 0.046532045654082525,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.2045,
|
|
"mean_token_accuracy": 0.09371785670518876,
|
|
"num_tokens": 1618733.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.602302169799804,
|
|
"epoch": 0.046824700029265434,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.2273,
|
|
"mean_token_accuracy": 0.09687273427844048,
|
|
"num_tokens": 1628728.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.53294186592102,
|
|
"epoch": 0.04711735440444834,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.3064,
|
|
"mean_token_accuracy": 0.0901516005396843,
|
|
"num_tokens": 1639016.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.574016046524048,
|
|
"epoch": 0.04741000877963126,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.297,
|
|
"mean_token_accuracy": 0.0918840229511261,
|
|
"num_tokens": 1649386.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.666880130767822,
|
|
"epoch": 0.04770266315481417,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.2977,
|
|
"mean_token_accuracy": 0.09063222482800484,
|
|
"num_tokens": 1659949.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.555622291564942,
|
|
"epoch": 0.04799531752999708,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.2765,
|
|
"mean_token_accuracy": 0.09335207715630531,
|
|
"num_tokens": 1670035.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.611965894699097,
|
|
"epoch": 0.048287971905179985,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.2132,
|
|
"mean_token_accuracy": 0.08826020434498787,
|
|
"num_tokens": 1679579.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.441506242752075,
|
|
"epoch": 0.048580626280362894,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.196,
|
|
"mean_token_accuracy": 0.09741625487804413,
|
|
"num_tokens": 1689900.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.5691015243530275,
|
|
"epoch": 0.0488732806555458,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.2469,
|
|
"mean_token_accuracy": 0.09050429165363312,
|
|
"num_tokens": 1700534.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.547771167755127,
|
|
"epoch": 0.04916593503072871,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.1963,
|
|
"mean_token_accuracy": 0.09931469187140465,
|
|
"num_tokens": 1710969.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.454426383972168,
|
|
"epoch": 0.04945858940591162,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.2267,
|
|
"mean_token_accuracy": 0.09168547242879868,
|
|
"num_tokens": 1721024.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.430084466934204,
|
|
"epoch": 0.04975124378109453,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.1318,
|
|
"mean_token_accuracy": 0.09713169336318969,
|
|
"num_tokens": 1730197.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.452590370178223,
|
|
"epoch": 0.05004389815627744,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.1541,
|
|
"mean_token_accuracy": 0.09219018146395683,
|
|
"num_tokens": 1741132.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.521068334579468,
|
|
"epoch": 0.05033655253146035,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.2637,
|
|
"mean_token_accuracy": 0.08954514041543007,
|
|
"num_tokens": 1751757.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.495842170715332,
|
|
"epoch": 0.050629206906643257,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.1582,
|
|
"mean_token_accuracy": 0.09514842405915261,
|
|
"num_tokens": 1762779.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.475547981262207,
|
|
"epoch": 0.050921861281826165,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.1182,
|
|
"mean_token_accuracy": 0.09630650877952576,
|
|
"num_tokens": 1772353.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.352218437194824,
|
|
"epoch": 0.051214515657009074,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.145,
|
|
"mean_token_accuracy": 0.09995641633868217,
|
|
"num_tokens": 1782648.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.392671585083008,
|
|
"epoch": 0.05150717003219198,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.1484,
|
|
"mean_token_accuracy": 0.09772149994969367,
|
|
"num_tokens": 1792275.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.387517881393433,
|
|
"epoch": 0.05179982440737489,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.1486,
|
|
"mean_token_accuracy": 0.09331418126821518,
|
|
"num_tokens": 1803244.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.4635416030883786,
|
|
"epoch": 0.0520924787825578,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.1907,
|
|
"mean_token_accuracy": 0.09984328448772431,
|
|
"num_tokens": 1814331.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.442190837860108,
|
|
"epoch": 0.05238513315774071,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.1557,
|
|
"mean_token_accuracy": 0.09669465869665146,
|
|
"num_tokens": 1823022.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.4167015075683596,
|
|
"epoch": 0.05267778753292362,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.2017,
|
|
"mean_token_accuracy": 0.09253377318382264,
|
|
"num_tokens": 1834416.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.530139684677124,
|
|
"epoch": 0.05297044190810653,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.1951,
|
|
"mean_token_accuracy": 0.09366482645273208,
|
|
"num_tokens": 1843507.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.3420909404754635,
|
|
"epoch": 0.053263096283289436,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.0998,
|
|
"mean_token_accuracy": 0.09217809438705445,
|
|
"num_tokens": 1854149.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.416664695739746,
|
|
"epoch": 0.053555750658472345,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.1686,
|
|
"mean_token_accuracy": 0.09251093789935112,
|
|
"num_tokens": 1864257.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.437055015563965,
|
|
"epoch": 0.053848405033655254,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.1707,
|
|
"mean_token_accuracy": 0.09187378212809563,
|
|
"num_tokens": 1873353.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.3977635383605955,
|
|
"epoch": 0.05414105940883816,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.1196,
|
|
"mean_token_accuracy": 0.09563436955213547,
|
|
"num_tokens": 1884465.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.365432357788086,
|
|
"epoch": 0.05443371378402107,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.1528,
|
|
"mean_token_accuracy": 0.09838449805974961,
|
|
"num_tokens": 1894872.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.34838981628418,
|
|
"epoch": 0.05472636815920398,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.0706,
|
|
"mean_token_accuracy": 0.10180790275335312,
|
|
"num_tokens": 1905440.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.341986465454101,
|
|
"epoch": 0.05501902253438689,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.1291,
|
|
"mean_token_accuracy": 0.09645774215459824,
|
|
"num_tokens": 1916861.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.406238269805908,
|
|
"epoch": 0.0553116769095698,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.0654,
|
|
"mean_token_accuracy": 0.10491539835929871,
|
|
"num_tokens": 1927144.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.35713381767273,
|
|
"epoch": 0.05560433128475271,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.138,
|
|
"mean_token_accuracy": 0.0931523747742176,
|
|
"num_tokens": 1937174.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.403324699401855,
|
|
"epoch": 0.055896985659935616,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.0506,
|
|
"mean_token_accuracy": 0.11088272705674171,
|
|
"num_tokens": 1947510.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.305504369735718,
|
|
"epoch": 0.056189640035118525,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.1238,
|
|
"mean_token_accuracy": 0.09536927789449692,
|
|
"num_tokens": 1957894.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.445551156997681,
|
|
"epoch": 0.056482294410301434,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.1862,
|
|
"mean_token_accuracy": 0.09522910863161087,
|
|
"num_tokens": 1968670.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.351961517333985,
|
|
"epoch": 0.05677494878548434,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.1451,
|
|
"mean_token_accuracy": 0.09402441680431366,
|
|
"num_tokens": 1979678.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.329155683517456,
|
|
"epoch": 0.05706760316066725,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.0501,
|
|
"mean_token_accuracy": 0.10141415372490883,
|
|
"num_tokens": 1990161.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.408869028091431,
|
|
"epoch": 0.05736025753585016,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.1421,
|
|
"mean_token_accuracy": 0.09915417581796646,
|
|
"num_tokens": 2000298.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.386150693893432,
|
|
"epoch": 0.05765291191103307,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.101,
|
|
"mean_token_accuracy": 0.0987740397453308,
|
|
"num_tokens": 2009946.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.297941255569458,
|
|
"epoch": 0.05794556628621598,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.064,
|
|
"mean_token_accuracy": 0.09957554042339326,
|
|
"num_tokens": 2019192.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.343301391601562,
|
|
"epoch": 0.05823822066139889,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.1112,
|
|
"mean_token_accuracy": 0.09475254267454147,
|
|
"num_tokens": 2029066.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.356789922714233,
|
|
"epoch": 0.058530875036581796,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.1051,
|
|
"mean_token_accuracy": 0.09616579562425613,
|
|
"num_tokens": 2040610.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.330273389816284,
|
|
"epoch": 0.058823529411764705,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 7.105,
|
|
"mean_token_accuracy": 0.0943969689309597,
|
|
"num_tokens": 2049867.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.243909168243408,
|
|
"epoch": 0.059116183786947614,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 7.0949,
|
|
"mean_token_accuracy": 0.09849751815199852,
|
|
"num_tokens": 2059335.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.482809591293335,
|
|
"epoch": 0.05940883816213052,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 7.0991,
|
|
"mean_token_accuracy": 0.09777801483869553,
|
|
"num_tokens": 2069498.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.322924661636352,
|
|
"epoch": 0.05970149253731343,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 7.1122,
|
|
"mean_token_accuracy": 0.10163608714938163,
|
|
"num_tokens": 2080077.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.265040874481201,
|
|
"epoch": 0.05999414691249634,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 7.0334,
|
|
"mean_token_accuracy": 0.10056126862764359,
|
|
"num_tokens": 2089699.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.259469652175904,
|
|
"epoch": 0.06028680128767925,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 7.0509,
|
|
"mean_token_accuracy": 0.09879743158817292,
|
|
"num_tokens": 2099219.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.274491453170777,
|
|
"epoch": 0.06057945566286216,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 7.0268,
|
|
"mean_token_accuracy": 0.10193772837519646,
|
|
"num_tokens": 2109218.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.383116436004639,
|
|
"epoch": 0.06087211003804507,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 7.0793,
|
|
"mean_token_accuracy": 0.09875930175185203,
|
|
"num_tokens": 2118384.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.2731462001800535,
|
|
"epoch": 0.061164764413227976,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 7.0464,
|
|
"mean_token_accuracy": 0.10151882916688919,
|
|
"num_tokens": 2127017.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.3074544906616214,
|
|
"epoch": 0.061457418788410885,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 7.1195,
|
|
"mean_token_accuracy": 0.09616761356592178,
|
|
"num_tokens": 2137568.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.266034030914307,
|
|
"epoch": 0.061750073163593794,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 7.0254,
|
|
"mean_token_accuracy": 0.09843881577253341,
|
|
"num_tokens": 2147063.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.276236915588379,
|
|
"epoch": 0.0620427275387767,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 7.0445,
|
|
"mean_token_accuracy": 0.10075422823429107,
|
|
"num_tokens": 2156525.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.229153347015381,
|
|
"epoch": 0.06233538191395961,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 7.0588,
|
|
"mean_token_accuracy": 0.09751596227288246,
|
|
"num_tokens": 2167294.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.209919691085815,
|
|
"epoch": 0.06262803628914253,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.9627,
|
|
"mean_token_accuracy": 0.1035116158425808,
|
|
"num_tokens": 2177341.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.297941446304321,
|
|
"epoch": 0.06292069066432543,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 7.0098,
|
|
"mean_token_accuracy": 0.09901509582996368,
|
|
"num_tokens": 2188044.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.183123016357422,
|
|
"epoch": 0.06321334503950835,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 7.0293,
|
|
"mean_token_accuracy": 0.10036754310131073,
|
|
"num_tokens": 2197612.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.282407426834107,
|
|
"epoch": 0.06350599941469125,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 7.0087,
|
|
"mean_token_accuracy": 0.09982658997178077,
|
|
"num_tokens": 2207638.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.160094165802002,
|
|
"epoch": 0.06379865378987416,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.9333,
|
|
"mean_token_accuracy": 0.10349727869033813,
|
|
"num_tokens": 2216714.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.322865962982178,
|
|
"epoch": 0.06409130816505706,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 7.0544,
|
|
"mean_token_accuracy": 0.09982285872101784,
|
|
"num_tokens": 2225900.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.1447443008422855,
|
|
"epoch": 0.06438396254023998,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.9718,
|
|
"mean_token_accuracy": 0.10523544400930404,
|
|
"num_tokens": 2235896.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.2013531684875485,
|
|
"epoch": 0.06467661691542288,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.94,
|
|
"mean_token_accuracy": 0.10068488419055939,
|
|
"num_tokens": 2246083.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.246619081497192,
|
|
"epoch": 0.0649692712906058,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.9572,
|
|
"mean_token_accuracy": 0.10323750525712967,
|
|
"num_tokens": 2256941.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.160521936416626,
|
|
"epoch": 0.0652619256657887,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.9737,
|
|
"mean_token_accuracy": 0.10002864897251129,
|
|
"num_tokens": 2266694.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.155581760406494,
|
|
"epoch": 0.06555458004097162,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.9638,
|
|
"mean_token_accuracy": 0.09659243822097778,
|
|
"num_tokens": 2276826.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.222516393661499,
|
|
"epoch": 0.06584723441615452,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.9739,
|
|
"mean_token_accuracy": 0.10149327889084817,
|
|
"num_tokens": 2287895.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.083696699142456,
|
|
"epoch": 0.06613988879133743,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.9513,
|
|
"mean_token_accuracy": 0.10140683799982071,
|
|
"num_tokens": 2297501.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.255348253250122,
|
|
"epoch": 0.06643254316652034,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.8291,
|
|
"mean_token_accuracy": 0.10531161874532699,
|
|
"num_tokens": 2306851.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.172169923782349,
|
|
"epoch": 0.06672519754170325,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 7.006,
|
|
"mean_token_accuracy": 0.09792810827493667,
|
|
"num_tokens": 2315710.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.135481214523315,
|
|
"epoch": 0.06701785191688615,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.9146,
|
|
"mean_token_accuracy": 0.10404545590281486,
|
|
"num_tokens": 2326144.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.1948600769042965,
|
|
"epoch": 0.06731050629206907,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.9143,
|
|
"mean_token_accuracy": 0.10875528082251548,
|
|
"num_tokens": 2335756.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.106131458282471,
|
|
"epoch": 0.06760316066725197,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.9304,
|
|
"mean_token_accuracy": 0.10725896134972572,
|
|
"num_tokens": 2345660.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.166688919067383,
|
|
"epoch": 0.06789581504243489,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.9403,
|
|
"mean_token_accuracy": 0.10006137192249298,
|
|
"num_tokens": 2355252.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.153073406219482,
|
|
"epoch": 0.06818846941761779,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.9305,
|
|
"mean_token_accuracy": 0.10227609500288963,
|
|
"num_tokens": 2364983.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.1287861347198485,
|
|
"epoch": 0.0684811237928007,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.9048,
|
|
"mean_token_accuracy": 0.10142005532979965,
|
|
"num_tokens": 2375505.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.152293968200683,
|
|
"epoch": 0.0687737781679836,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.9587,
|
|
"mean_token_accuracy": 0.10323912650346756,
|
|
"num_tokens": 2385300.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.039838171005249,
|
|
"epoch": 0.06906643254316652,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.9443,
|
|
"mean_token_accuracy": 0.10556500256061555,
|
|
"num_tokens": 2395966.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.262147569656372,
|
|
"epoch": 0.06935908691834942,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.9281,
|
|
"mean_token_accuracy": 0.10152315646409989,
|
|
"num_tokens": 2405429.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.163443088531494,
|
|
"epoch": 0.06965174129353234,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.9202,
|
|
"mean_token_accuracy": 0.10446830168366432,
|
|
"num_tokens": 2415362.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.095344686508179,
|
|
"epoch": 0.06994439566871524,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.8784,
|
|
"mean_token_accuracy": 0.10187279507517814,
|
|
"num_tokens": 2425998.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.126546478271484,
|
|
"epoch": 0.07023705004389816,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.9491,
|
|
"mean_token_accuracy": 0.09673603773117065,
|
|
"num_tokens": 2436226.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.202165269851685,
|
|
"epoch": 0.07052970441908106,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.8387,
|
|
"mean_token_accuracy": 0.10797644704580307,
|
|
"num_tokens": 2445296.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.082312726974488,
|
|
"epoch": 0.07082235879426398,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.8589,
|
|
"mean_token_accuracy": 0.11134556159377099,
|
|
"num_tokens": 2455743.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.09007248878479,
|
|
"epoch": 0.07111501316944688,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.9405,
|
|
"mean_token_accuracy": 0.1020224578678608,
|
|
"num_tokens": 2465239.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.164793825149536,
|
|
"epoch": 0.0714076675446298,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.922,
|
|
"mean_token_accuracy": 0.10908990055322647,
|
|
"num_tokens": 2474472.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.9930988311767575,
|
|
"epoch": 0.0717003219198127,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.7943,
|
|
"mean_token_accuracy": 0.11290957406163216,
|
|
"num_tokens": 2483549.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.140143394470215,
|
|
"epoch": 0.07199297629499561,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.8657,
|
|
"mean_token_accuracy": 0.10449625924229622,
|
|
"num_tokens": 2495699.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.029150915145874,
|
|
"epoch": 0.07228563067017851,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.9746,
|
|
"mean_token_accuracy": 0.10025271028280258,
|
|
"num_tokens": 2507612.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.133029508590698,
|
|
"epoch": 0.07257828504536143,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.8438,
|
|
"mean_token_accuracy": 0.10679830834269524,
|
|
"num_tokens": 2517281.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.167627429962158,
|
|
"epoch": 0.07287093942054433,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.8234,
|
|
"mean_token_accuracy": 0.10795086920261383,
|
|
"num_tokens": 2526570.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.026937675476074,
|
|
"epoch": 0.07316359379572725,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.8455,
|
|
"mean_token_accuracy": 0.10278113707900047,
|
|
"num_tokens": 2535813.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.024099922180175,
|
|
"epoch": 0.07345624817091015,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.7853,
|
|
"mean_token_accuracy": 0.11688904613256454,
|
|
"num_tokens": 2544828.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.1756445407867435,
|
|
"epoch": 0.07374890254609306,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.9021,
|
|
"mean_token_accuracy": 0.10272038280963898,
|
|
"num_tokens": 2554612.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.887376928329468,
|
|
"epoch": 0.07404155692127597,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.8311,
|
|
"mean_token_accuracy": 0.11283649727702141,
|
|
"num_tokens": 2565395.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.175661230087281,
|
|
"epoch": 0.07433421129645888,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.966,
|
|
"mean_token_accuracy": 0.1015208400785923,
|
|
"num_tokens": 2576520.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.063946008682251,
|
|
"epoch": 0.07462686567164178,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.8779,
|
|
"mean_token_accuracy": 0.10767215564846992,
|
|
"num_tokens": 2586093.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.0348388195037845,
|
|
"epoch": 0.0749195200468247,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.8673,
|
|
"mean_token_accuracy": 0.11059194058179855,
|
|
"num_tokens": 2596165.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.263307380676269,
|
|
"epoch": 0.0752121744220076,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.981,
|
|
"mean_token_accuracy": 0.09260072112083435,
|
|
"num_tokens": 2607108.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.985970163345337,
|
|
"epoch": 0.07550482879719052,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.8579,
|
|
"mean_token_accuracy": 0.10567844063043594,
|
|
"num_tokens": 2617313.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.193155908584595,
|
|
"epoch": 0.07579748317237343,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.942,
|
|
"mean_token_accuracy": 0.1037025772035122,
|
|
"num_tokens": 2628065.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.068187236785889,
|
|
"epoch": 0.07609013754755634,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.9342,
|
|
"mean_token_accuracy": 0.09517190679907798,
|
|
"num_tokens": 2638052.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.106878709793091,
|
|
"epoch": 0.07638279192273925,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.8695,
|
|
"mean_token_accuracy": 0.10798671543598175,
|
|
"num_tokens": 2648242.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.1013659000396725,
|
|
"epoch": 0.07667544629792215,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.8159,
|
|
"mean_token_accuracy": 0.10844166055321694,
|
|
"num_tokens": 2657824.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.991985273361206,
|
|
"epoch": 0.07696810067310507,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.7952,
|
|
"mean_token_accuracy": 0.10927849039435386,
|
|
"num_tokens": 2667774.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.040319395065308,
|
|
"epoch": 0.07726075504828797,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.7604,
|
|
"mean_token_accuracy": 0.1062987856566906,
|
|
"num_tokens": 2677697.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.020246982574463,
|
|
"epoch": 0.07755340942347089,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.8529,
|
|
"mean_token_accuracy": 0.10597260296344757,
|
|
"num_tokens": 2688553.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.039525413513184,
|
|
"epoch": 0.07784606379865379,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.7872,
|
|
"mean_token_accuracy": 0.1075766459107399,
|
|
"num_tokens": 2698704.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.057392406463623,
|
|
"epoch": 0.0781387181738367,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.856,
|
|
"mean_token_accuracy": 0.10357561856508254,
|
|
"num_tokens": 2709257.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.129636335372925,
|
|
"epoch": 0.0784313725490196,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.8684,
|
|
"mean_token_accuracy": 0.10322815775871277,
|
|
"num_tokens": 2719860.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.990602445602417,
|
|
"epoch": 0.07872402692420252,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.8517,
|
|
"mean_token_accuracy": 0.1050374872982502,
|
|
"num_tokens": 2730134.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.978021669387817,
|
|
"epoch": 0.07901668129938542,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.6989,
|
|
"mean_token_accuracy": 0.1164980985224247,
|
|
"num_tokens": 2739661.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.976407098770141,
|
|
"epoch": 0.07930933567456834,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.8151,
|
|
"mean_token_accuracy": 0.10699095502495766,
|
|
"num_tokens": 2749532.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.110589075088501,
|
|
"epoch": 0.07960199004975124,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.8471,
|
|
"mean_token_accuracy": 0.10450146123766899,
|
|
"num_tokens": 2759652.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.953937816619873,
|
|
"epoch": 0.07989464442493416,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.7311,
|
|
"mean_token_accuracy": 0.11042979061603546,
|
|
"num_tokens": 2770122.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.970408010482788,
|
|
"epoch": 0.08018729880011706,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.8092,
|
|
"mean_token_accuracy": 0.10888475328683853,
|
|
"num_tokens": 2779788.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.0338794708251955,
|
|
"epoch": 0.08047995317529998,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.805,
|
|
"mean_token_accuracy": 0.10510503351688386,
|
|
"num_tokens": 2790149.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.031380653381348,
|
|
"epoch": 0.08077260755048288,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.7248,
|
|
"mean_token_accuracy": 0.11531992107629777,
|
|
"num_tokens": 2799378.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.92920618057251,
|
|
"epoch": 0.0810652619256658,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.8058,
|
|
"mean_token_accuracy": 0.11184522807598114,
|
|
"num_tokens": 2809573.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.951801586151123,
|
|
"epoch": 0.0813579163008487,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.7261,
|
|
"mean_token_accuracy": 0.11088547632098197,
|
|
"num_tokens": 2819816.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.114043664932251,
|
|
"epoch": 0.08165057067603161,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.8456,
|
|
"mean_token_accuracy": 0.1005843736231327,
|
|
"num_tokens": 2830606.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.9624059200286865,
|
|
"epoch": 0.08194322505121451,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.8289,
|
|
"mean_token_accuracy": 0.10612485110759735,
|
|
"num_tokens": 2840925.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.966361713409424,
|
|
"epoch": 0.08223587942639743,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.7518,
|
|
"mean_token_accuracy": 0.11064208373427391,
|
|
"num_tokens": 2850697.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.990543603897095,
|
|
"epoch": 0.08252853380158033,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.8402,
|
|
"mean_token_accuracy": 0.10857592597603798,
|
|
"num_tokens": 2861001.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.022043466567993,
|
|
"epoch": 0.08282118817676325,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.7625,
|
|
"mean_token_accuracy": 0.11597265005111694,
|
|
"num_tokens": 2869959.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.995840072631836,
|
|
"epoch": 0.08311384255194615,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.728,
|
|
"mean_token_accuracy": 0.11714338809251786,
|
|
"num_tokens": 2878998.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.972711563110352,
|
|
"epoch": 0.08340649692712906,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.7609,
|
|
"mean_token_accuracy": 0.10615370124578476,
|
|
"num_tokens": 2888870.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.923924589157105,
|
|
"epoch": 0.08369915130231197,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.6755,
|
|
"mean_token_accuracy": 0.11179826408624649,
|
|
"num_tokens": 2898549.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.9445459842681885,
|
|
"epoch": 0.08399180567749488,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.7235,
|
|
"mean_token_accuracy": 0.11873029172420502,
|
|
"num_tokens": 2908614.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.906632804870606,
|
|
"epoch": 0.08428446005267778,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.7154,
|
|
"mean_token_accuracy": 0.1131787732243538,
|
|
"num_tokens": 2917845.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.026270580291748,
|
|
"epoch": 0.0845771144278607,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.8334,
|
|
"mean_token_accuracy": 0.10297970846295357,
|
|
"num_tokens": 2929234.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.907474088668823,
|
|
"epoch": 0.0848697688030436,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.7675,
|
|
"mean_token_accuracy": 0.10896839275956154,
|
|
"num_tokens": 2939011.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.063915395736695,
|
|
"epoch": 0.08516242317822652,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.7726,
|
|
"mean_token_accuracy": 0.10877819880843162,
|
|
"num_tokens": 2948796.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.904459810256958,
|
|
"epoch": 0.08545507755340942,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.8066,
|
|
"mean_token_accuracy": 0.10255414545536042,
|
|
"num_tokens": 2960938.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.976855278015137,
|
|
"epoch": 0.08574773192859234,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.7148,
|
|
"mean_token_accuracy": 0.11121275797486305,
|
|
"num_tokens": 2969964.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.9436931133270265,
|
|
"epoch": 0.08604038630377524,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.7419,
|
|
"mean_token_accuracy": 0.10835531502962112,
|
|
"num_tokens": 2980562.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.976116418838501,
|
|
"epoch": 0.08633304067895815,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.6753,
|
|
"mean_token_accuracy": 0.11625185832381249,
|
|
"num_tokens": 2990759.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.905919122695923,
|
|
"epoch": 0.08662569505414106,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.7076,
|
|
"mean_token_accuracy": 0.11023089289665222,
|
|
"num_tokens": 3000719.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.928934144973755,
|
|
"epoch": 0.08691834942932397,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.7253,
|
|
"mean_token_accuracy": 0.11021335422992706,
|
|
"num_tokens": 3011584.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.850312614440918,
|
|
"epoch": 0.08721100380450687,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.6914,
|
|
"mean_token_accuracy": 0.11205545589327812,
|
|
"num_tokens": 3021933.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.0218939781188965,
|
|
"epoch": 0.08750365817968979,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.7541,
|
|
"mean_token_accuracy": 0.10549476444721222,
|
|
"num_tokens": 3032533.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.886258840560913,
|
|
"epoch": 0.08779631255487269,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.6657,
|
|
"mean_token_accuracy": 0.11410991847515106,
|
|
"num_tokens": 3041556.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.93623046875,
|
|
"epoch": 0.0880889669300556,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 6.7447,
|
|
"mean_token_accuracy": 0.10976714193820954,
|
|
"num_tokens": 3052030.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.86750373840332,
|
|
"epoch": 0.08838162130523851,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.6783,
|
|
"mean_token_accuracy": 0.10968335196375847,
|
|
"num_tokens": 3062267.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.955923652648925,
|
|
"epoch": 0.08867427568042142,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.643,
|
|
"mean_token_accuracy": 0.11466655507683754,
|
|
"num_tokens": 3073357.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.84696593284607,
|
|
"epoch": 0.08896693005560433,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.7323,
|
|
"mean_token_accuracy": 0.11122780367732048,
|
|
"num_tokens": 3083250.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.95814471244812,
|
|
"epoch": 0.08925958443078724,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.6353,
|
|
"mean_token_accuracy": 0.11299381777644157,
|
|
"num_tokens": 3093144.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.820744323730469,
|
|
"epoch": 0.08955223880597014,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.6688,
|
|
"mean_token_accuracy": 0.11378093957901,
|
|
"num_tokens": 3103731.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.915271520614624,
|
|
"epoch": 0.08984489318115306,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.7017,
|
|
"mean_token_accuracy": 0.11074108928442002,
|
|
"num_tokens": 3114113.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.886846303939819,
|
|
"epoch": 0.09013754755633596,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.653,
|
|
"mean_token_accuracy": 0.11144338846206665,
|
|
"num_tokens": 3124167.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.852968740463257,
|
|
"epoch": 0.09043020193151888,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.5614,
|
|
"mean_token_accuracy": 0.12301889657974244,
|
|
"num_tokens": 3134751.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.885099744796753,
|
|
"epoch": 0.09072285630670178,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.6554,
|
|
"mean_token_accuracy": 0.11772447004914284,
|
|
"num_tokens": 3144084.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.754053544998169,
|
|
"epoch": 0.0910155106818847,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.6136,
|
|
"mean_token_accuracy": 0.11398938149213791,
|
|
"num_tokens": 3153156.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.984991645812988,
|
|
"epoch": 0.0913081650570676,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.6408,
|
|
"mean_token_accuracy": 0.11836520209908485,
|
|
"num_tokens": 3163098.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.801419591903686,
|
|
"epoch": 0.09160081943225051,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 6.6727,
|
|
"mean_token_accuracy": 0.11261920407414436,
|
|
"num_tokens": 3172623.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.9145081520080565,
|
|
"epoch": 0.09189347380743342,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 6.6369,
|
|
"mean_token_accuracy": 0.11749240681529045,
|
|
"num_tokens": 3182610.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.80106897354126,
|
|
"epoch": 0.09218612818261633,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.6364,
|
|
"mean_token_accuracy": 0.108591927587986,
|
|
"num_tokens": 3192978.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.943644762039185,
|
|
"epoch": 0.09247878255779923,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 6.6731,
|
|
"mean_token_accuracy": 0.11072808504104614,
|
|
"num_tokens": 3202791.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.783168792724609,
|
|
"epoch": 0.09277143693298215,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 6.6412,
|
|
"mean_token_accuracy": 0.11365418583154678,
|
|
"num_tokens": 3212408.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.906521463394165,
|
|
"epoch": 0.09306409130816505,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.6733,
|
|
"mean_token_accuracy": 0.11501292213797569,
|
|
"num_tokens": 3222182.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.849865865707398,
|
|
"epoch": 0.09335674568334797,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.6344,
|
|
"mean_token_accuracy": 0.11746996641159058,
|
|
"num_tokens": 3233058.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.8446972370147705,
|
|
"epoch": 0.09364940005853087,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 6.6006,
|
|
"mean_token_accuracy": 0.11764349192380905,
|
|
"num_tokens": 3243412.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.802833318710327,
|
|
"epoch": 0.09394205443371378,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 6.7006,
|
|
"mean_token_accuracy": 0.11396326944231987,
|
|
"num_tokens": 3254511.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.9019934177398685,
|
|
"epoch": 0.09423470880889669,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 6.5915,
|
|
"mean_token_accuracy": 0.11526288017630577,
|
|
"num_tokens": 3264949.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.764813375473023,
|
|
"epoch": 0.0945273631840796,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.6396,
|
|
"mean_token_accuracy": 0.10926437377929688,
|
|
"num_tokens": 3275014.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.861340761184692,
|
|
"epoch": 0.09482001755926252,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.6634,
|
|
"mean_token_accuracy": 0.11290878355503083,
|
|
"num_tokens": 3285446.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.844233989715576,
|
|
"epoch": 0.09511267193444542,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 6.5346,
|
|
"mean_token_accuracy": 0.1263396829366684,
|
|
"num_tokens": 3295383.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.765133380889893,
|
|
"epoch": 0.09540532630962834,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.6189,
|
|
"mean_token_accuracy": 0.11274374797940254,
|
|
"num_tokens": 3305263.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.842666339874268,
|
|
"epoch": 0.09569798068481124,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 6.6083,
|
|
"mean_token_accuracy": 0.1211700364947319,
|
|
"num_tokens": 3315434.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.784665536880493,
|
|
"epoch": 0.09599063505999415,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 6.5348,
|
|
"mean_token_accuracy": 0.12217146530747414,
|
|
"num_tokens": 3324908.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.831562328338623,
|
|
"epoch": 0.09628328943517706,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 6.6054,
|
|
"mean_token_accuracy": 0.12192678451538086,
|
|
"num_tokens": 3335987.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.853379344940185,
|
|
"epoch": 0.09657594381035997,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.561,
|
|
"mean_token_accuracy": 0.1174039103090763,
|
|
"num_tokens": 3345502.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.778363132476807,
|
|
"epoch": 0.09686859818554287,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 6.5545,
|
|
"mean_token_accuracy": 0.11502353250980377,
|
|
"num_tokens": 3355237.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.812596559524536,
|
|
"epoch": 0.09716125256072579,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.6318,
|
|
"mean_token_accuracy": 0.11809700429439544,
|
|
"num_tokens": 3366166.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.858366012573242,
|
|
"epoch": 0.09745390693590869,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 6.5995,
|
|
"mean_token_accuracy": 0.11467723920941353,
|
|
"num_tokens": 3375931.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.782172203063965,
|
|
"epoch": 0.0977465613110916,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.5999,
|
|
"mean_token_accuracy": 0.12378341481089591,
|
|
"num_tokens": 3386728.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.759363842010498,
|
|
"epoch": 0.09803921568627451,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.5791,
|
|
"mean_token_accuracy": 0.1205291859805584,
|
|
"num_tokens": 3396795.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.835077142715454,
|
|
"epoch": 0.09833187006145742,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 6.5568,
|
|
"mean_token_accuracy": 0.1181432493031025,
|
|
"num_tokens": 3406310.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.776544427871704,
|
|
"epoch": 0.09862452443664033,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 6.6114,
|
|
"mean_token_accuracy": 0.11417201384902001,
|
|
"num_tokens": 3418110.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.792936897277832,
|
|
"epoch": 0.09891717881182324,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.5952,
|
|
"mean_token_accuracy": 0.11885511875152588,
|
|
"num_tokens": 3428122.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.801793384552002,
|
|
"epoch": 0.09920983318700614,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.6457,
|
|
"mean_token_accuracy": 0.11425484046339988,
|
|
"num_tokens": 3438514.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.8519859790802,
|
|
"epoch": 0.09950248756218906,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 6.6262,
|
|
"mean_token_accuracy": 0.1146598532795906,
|
|
"num_tokens": 3449040.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.8020600318908695,
|
|
"epoch": 0.09979514193737196,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 6.5487,
|
|
"mean_token_accuracy": 0.12066433876752854,
|
|
"num_tokens": 3459244.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.881866407394409,
|
|
"epoch": 0.10008779631255488,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 6.6118,
|
|
"mean_token_accuracy": 0.11424820944666862,
|
|
"num_tokens": 3469270.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.784549045562744,
|
|
"epoch": 0.10038045068773778,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.592,
|
|
"mean_token_accuracy": 0.11819150224328041,
|
|
"num_tokens": 3480113.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.802189111709595,
|
|
"epoch": 0.1006731050629207,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 6.5813,
|
|
"mean_token_accuracy": 0.11734010130167008,
|
|
"num_tokens": 3490451.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.809964227676391,
|
|
"epoch": 0.1009657594381036,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 6.5405,
|
|
"mean_token_accuracy": 0.11609668955206871,
|
|
"num_tokens": 3501107.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.71238145828247,
|
|
"epoch": 0.10125841381328651,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 6.5099,
|
|
"mean_token_accuracy": 0.13233423829078675,
|
|
"num_tokens": 3510731.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.739808130264282,
|
|
"epoch": 0.10155106818846941,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.4388,
|
|
"mean_token_accuracy": 0.12671404406428338,
|
|
"num_tokens": 3520843.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.837344312667847,
|
|
"epoch": 0.10184372256365233,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.6177,
|
|
"mean_token_accuracy": 0.11465101763606071,
|
|
"num_tokens": 3530668.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.749877786636352,
|
|
"epoch": 0.10213637693883523,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 6.5428,
|
|
"mean_token_accuracy": 0.12134164273738861,
|
|
"num_tokens": 3540252.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.7890314102172855,
|
|
"epoch": 0.10242903131401815,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.5481,
|
|
"mean_token_accuracy": 0.11811421886086464,
|
|
"num_tokens": 3549069.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.757550573348999,
|
|
"epoch": 0.10272168568920105,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.58,
|
|
"mean_token_accuracy": 0.11727841719985008,
|
|
"num_tokens": 3559988.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.787223291397095,
|
|
"epoch": 0.10301434006438397,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.4955,
|
|
"mean_token_accuracy": 0.1283750705420971,
|
|
"num_tokens": 3569829.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.687646579742432,
|
|
"epoch": 0.10330699443956687,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.4395,
|
|
"mean_token_accuracy": 0.1254482589662075,
|
|
"num_tokens": 3578958.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.786183023452759,
|
|
"epoch": 0.10359964881474978,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 6.5076,
|
|
"mean_token_accuracy": 0.12173920199275017,
|
|
"num_tokens": 3588844.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.71337513923645,
|
|
"epoch": 0.10389230318993269,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.5819,
|
|
"mean_token_accuracy": 0.11760919094085694,
|
|
"num_tokens": 3599047.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.7201086521148685,
|
|
"epoch": 0.1041849575651156,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 6.4735,
|
|
"mean_token_accuracy": 0.1287480928003788,
|
|
"num_tokens": 3608416.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.780730867385865,
|
|
"epoch": 0.1044776119402985,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 6.5059,
|
|
"mean_token_accuracy": 0.11816626787185669,
|
|
"num_tokens": 3618460.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.67847204208374,
|
|
"epoch": 0.10477026631548142,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.5474,
|
|
"mean_token_accuracy": 0.12268414348363876,
|
|
"num_tokens": 3628795.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.735287475585937,
|
|
"epoch": 0.10506292069066432,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 6.5558,
|
|
"mean_token_accuracy": 0.11583011224865913,
|
|
"num_tokens": 3640278.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.766513156890869,
|
|
"epoch": 0.10535557506584724,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.5482,
|
|
"mean_token_accuracy": 0.12088547199964524,
|
|
"num_tokens": 3650497.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.722351598739624,
|
|
"epoch": 0.10564822944103014,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.5984,
|
|
"mean_token_accuracy": 0.1228600986301899,
|
|
"num_tokens": 3661101.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.922651243209839,
|
|
"epoch": 0.10594088381621306,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 6.5561,
|
|
"mean_token_accuracy": 0.1192373014986515,
|
|
"num_tokens": 3670698.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.736356639862061,
|
|
"epoch": 0.10623353819139596,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 6.5585,
|
|
"mean_token_accuracy": 0.11428969204425812,
|
|
"num_tokens": 3679723.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.864965009689331,
|
|
"epoch": 0.10652619256657887,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 6.5512,
|
|
"mean_token_accuracy": 0.11349868997931481,
|
|
"num_tokens": 3690362.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.747930908203125,
|
|
"epoch": 0.10681884694176177,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 6.5085,
|
|
"mean_token_accuracy": 0.12559529766440392,
|
|
"num_tokens": 3700087.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.717715930938721,
|
|
"epoch": 0.10711150131694469,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 6.5049,
|
|
"mean_token_accuracy": 0.12355283051729202,
|
|
"num_tokens": 3710086.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.7144248485565186,
|
|
"epoch": 0.10740415569212759,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.4642,
|
|
"mean_token_accuracy": 0.12195742204785347,
|
|
"num_tokens": 3720292.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.729002285003662,
|
|
"epoch": 0.10769681006731051,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 6.5342,
|
|
"mean_token_accuracy": 0.11787921264767647,
|
|
"num_tokens": 3730980.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.749273777008057,
|
|
"epoch": 0.10798946444249341,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.5779,
|
|
"mean_token_accuracy": 0.11964235603809356,
|
|
"num_tokens": 3741244.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.732828664779663,
|
|
"epoch": 0.10828211881767633,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 6.5632,
|
|
"mean_token_accuracy": 0.11899603232741356,
|
|
"num_tokens": 3751841.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.836530923843384,
|
|
"epoch": 0.10857477319285923,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 6.5206,
|
|
"mean_token_accuracy": 0.12107773423194886,
|
|
"num_tokens": 3762571.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.6559099674224855,
|
|
"epoch": 0.10886742756804214,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 6.5158,
|
|
"mean_token_accuracy": 0.12190034389495849,
|
|
"num_tokens": 3771778.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.7532881736755375,
|
|
"epoch": 0.10916008194322505,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.4896,
|
|
"mean_token_accuracy": 0.12468724176287652,
|
|
"num_tokens": 3781712.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.647825574874878,
|
|
"epoch": 0.10945273631840796,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 6.3472,
|
|
"mean_token_accuracy": 0.13525793552398682,
|
|
"num_tokens": 3790711.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.663978242874146,
|
|
"epoch": 0.10974539069359086,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 6.473,
|
|
"mean_token_accuracy": 0.1203851766884327,
|
|
"num_tokens": 3801271.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.769151782989502,
|
|
"epoch": 0.11003804506877378,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.4856,
|
|
"mean_token_accuracy": 0.12278474420309067,
|
|
"num_tokens": 3811518.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.705088186264038,
|
|
"epoch": 0.11033069944395668,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 6.5151,
|
|
"mean_token_accuracy": 0.1212370365858078,
|
|
"num_tokens": 3821454.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.742790365219117,
|
|
"epoch": 0.1106233538191396,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.5606,
|
|
"mean_token_accuracy": 0.11982747986912727,
|
|
"num_tokens": 3831886.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.7166282653808596,
|
|
"epoch": 0.1109160081943225,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.4898,
|
|
"mean_token_accuracy": 0.1260741837322712,
|
|
"num_tokens": 3841723.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.7906341552734375,
|
|
"epoch": 0.11120866256950541,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 6.4533,
|
|
"mean_token_accuracy": 0.1258271798491478,
|
|
"num_tokens": 3851374.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.740607738494873,
|
|
"epoch": 0.11150131694468832,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 6.4903,
|
|
"mean_token_accuracy": 0.12274285033345222,
|
|
"num_tokens": 3862627.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.659914445877075,
|
|
"epoch": 0.11179397131987123,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.4784,
|
|
"mean_token_accuracy": 0.12452571094036102,
|
|
"num_tokens": 3872344.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.735513925552368,
|
|
"epoch": 0.11208662569505413,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 6.5453,
|
|
"mean_token_accuracy": 0.11647359281778336,
|
|
"num_tokens": 3882896.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.713681602478028,
|
|
"epoch": 0.11237928007023705,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.4595,
|
|
"mean_token_accuracy": 0.12801918759942055,
|
|
"num_tokens": 3893773.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.6865918159484865,
|
|
"epoch": 0.11267193444541995,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.4424,
|
|
"mean_token_accuracy": 0.12188205718994141,
|
|
"num_tokens": 3905057.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.726850271224976,
|
|
"epoch": 0.11296458882060287,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.4427,
|
|
"mean_token_accuracy": 0.12147087454795838,
|
|
"num_tokens": 3914306.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.700537204742432,
|
|
"epoch": 0.11325724319578578,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 6.4905,
|
|
"mean_token_accuracy": 0.12801681384444236,
|
|
"num_tokens": 3924168.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.674844408035279,
|
|
"epoch": 0.11354989757096869,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.4243,
|
|
"mean_token_accuracy": 0.12779392600059508,
|
|
"num_tokens": 3934603.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.662173557281494,
|
|
"epoch": 0.1138425519461516,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 6.575,
|
|
"mean_token_accuracy": 0.12289680615067482,
|
|
"num_tokens": 3946181.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.859195804595947,
|
|
"epoch": 0.1141352063213345,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 6.4832,
|
|
"mean_token_accuracy": 0.12466611191630364,
|
|
"num_tokens": 3955732.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.668773794174195,
|
|
"epoch": 0.11442786069651742,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 6.4042,
|
|
"mean_token_accuracy": 0.12757309898734093,
|
|
"num_tokens": 3965569.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.693961477279663,
|
|
"epoch": 0.11472051507170032,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.664,
|
|
"mean_token_accuracy": 0.10505481734871865,
|
|
"num_tokens": 3976020.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.842778825759888,
|
|
"epoch": 0.11501316944688324,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 6.4668,
|
|
"mean_token_accuracy": 0.12651116624474526,
|
|
"num_tokens": 3986642.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.6270751953125,
|
|
"epoch": 0.11530582382206614,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 6.4971,
|
|
"mean_token_accuracy": 0.12009134292602539,
|
|
"num_tokens": 3997343.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.695995426177978,
|
|
"epoch": 0.11559847819724905,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 6.417,
|
|
"mean_token_accuracy": 0.12686994075775146,
|
|
"num_tokens": 4007715.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.663046932220459,
|
|
"epoch": 0.11589113257243196,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 6.5211,
|
|
"mean_token_accuracy": 0.1214706502854824,
|
|
"num_tokens": 4018268.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.7626746654510494,
|
|
"epoch": 0.11618378694761487,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.4082,
|
|
"mean_token_accuracy": 0.12497928068041801,
|
|
"num_tokens": 4028539.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.682868528366089,
|
|
"epoch": 0.11647644132279777,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 6.4965,
|
|
"mean_token_accuracy": 0.12758915945887567,
|
|
"num_tokens": 4039300.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.666601848602295,
|
|
"epoch": 0.11676909569798069,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 6.5343,
|
|
"mean_token_accuracy": 0.11495309770107269,
|
|
"num_tokens": 4048471.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.751353979110718,
|
|
"epoch": 0.11706175007316359,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 6.3476,
|
|
"mean_token_accuracy": 0.1322360999882221,
|
|
"num_tokens": 4058221.0,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"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": 886039736647680.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|