2035 lines
54 KiB
JSON
2035 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.058530875036581796,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 447947671633920.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|