fpadovani/eus-latn-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
eus-latn-100mb-after-ppt-sh…/checkpoint-6000/trainer_state.json
ModelHub XC 9e20616cf9 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/eus-latn-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
Source: Original Platform
2026-06-28 04:48:20 +08:00

12057 lines
329 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.5298481102084069,
"eval_steps": 3000,
"global_step": 6000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 4.790674829483033,
"epoch": 0.0004415400918403391,
"grad_norm": 13.0625,
"learning_rate": 2e-06,
"loss": 14.4349,
"mean_token_accuracy": 0.0,
"num_tokens": 9390.0,
"step": 5
},
{
"entropy": 4.818728256225586,
"epoch": 0.0008830801836806782,
"grad_norm": 14.1875,
"learning_rate": 4.5e-06,
"loss": 14.4117,
"mean_token_accuracy": 0.0,
"num_tokens": 18671.0,
"step": 10
},
{
"entropy": 4.857943296432495,
"epoch": 0.0013246202755210173,
"grad_norm": 16.25,
"learning_rate": 7e-06,
"loss": 14.1693,
"mean_token_accuracy": 0.00014005602570250631,
"num_tokens": 27614.0,
"step": 15
},
{
"entropy": 5.027469444274902,
"epoch": 0.0017661603673613563,
"grad_norm": 25.875,
"learning_rate": 9.5e-06,
"loss": 13.7713,
"mean_token_accuracy": 8.547008619643747e-05,
"num_tokens": 37850.0,
"step": 20
},
{
"entropy": 6.5380439281463625,
"epoch": 0.0022077004592016957,
"grad_norm": 26.375,
"learning_rate": 1.2e-05,
"loss": 12.2509,
"mean_token_accuracy": 0.0,
"num_tokens": 47166.0,
"step": 25
},
{
"entropy": 9.970003509521485,
"epoch": 0.0026492405510420347,
"grad_norm": 3.4375,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.9371,
"mean_token_accuracy": 0.00022374301915988325,
"num_tokens": 55500.0,
"step": 30
},
{
"entropy": 10.680026626586914,
"epoch": 0.0030907806428823736,
"grad_norm": 3.15625,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.6238,
"mean_token_accuracy": 0.009453117521479726,
"num_tokens": 63851.0,
"step": 35
},
{
"entropy": 10.703511428833007,
"epoch": 0.0035323207347227126,
"grad_norm": 3.21875,
"learning_rate": 1.95e-05,
"loss": 10.3602,
"mean_token_accuracy": 0.029037438705563544,
"num_tokens": 73697.0,
"step": 40
},
{
"entropy": 10.651962184906006,
"epoch": 0.003973860826563052,
"grad_norm": 2.625,
"learning_rate": 2.2e-05,
"loss": 10.0115,
"mean_token_accuracy": 0.05894971713423729,
"num_tokens": 83000.0,
"step": 45
},
{
"entropy": 10.439279747009277,
"epoch": 0.004415400918403391,
"grad_norm": 2.03125,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.8132,
"mean_token_accuracy": 0.05815875120460987,
"num_tokens": 92982.0,
"step": 50
},
{
"entropy": 10.318083763122559,
"epoch": 0.00485694101024373,
"grad_norm": 1.8203125,
"learning_rate": 2.7e-05,
"loss": 9.6231,
"mean_token_accuracy": 0.05530005097389221,
"num_tokens": 101455.0,
"step": 55
},
{
"entropy": 10.381121063232422,
"epoch": 0.005298481102084069,
"grad_norm": 1.921875,
"learning_rate": 2.95e-05,
"loss": 9.5438,
"mean_token_accuracy": 0.05805549845099449,
"num_tokens": 110782.0,
"step": 60
},
{
"entropy": 10.360444736480712,
"epoch": 0.005740021193924408,
"grad_norm": 1.71875,
"learning_rate": 3.2e-05,
"loss": 9.4168,
"mean_token_accuracy": 0.060499183088541034,
"num_tokens": 119241.0,
"step": 65
},
{
"entropy": 10.300647163391114,
"epoch": 0.006181561285764747,
"grad_norm": 1.75,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.4178,
"mean_token_accuracy": 0.055320289358496665,
"num_tokens": 127903.0,
"step": 70
},
{
"entropy": 10.332123184204102,
"epoch": 0.006623101377605086,
"grad_norm": 1.6328125,
"learning_rate": 3.7e-05,
"loss": 9.3721,
"mean_token_accuracy": 0.05736841931939125,
"num_tokens": 137370.0,
"step": 75
},
{
"entropy": 10.290982055664063,
"epoch": 0.007064641469445425,
"grad_norm": 1.625,
"learning_rate": 3.95e-05,
"loss": 9.2214,
"mean_token_accuracy": 0.06618293710052967,
"num_tokens": 146582.0,
"step": 80
},
{
"entropy": 10.196907424926758,
"epoch": 0.007506181561285765,
"grad_norm": 1.5703125,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.1585,
"mean_token_accuracy": 0.05961471572518349,
"num_tokens": 154933.0,
"step": 85
},
{
"entropy": 10.205323791503906,
"epoch": 0.007947721653126103,
"grad_norm": 1.5078125,
"learning_rate": 4.45e-05,
"loss": 9.1026,
"mean_token_accuracy": 0.072137650847435,
"num_tokens": 165157.0,
"step": 90
},
{
"entropy": 10.10411615371704,
"epoch": 0.008389261744966443,
"grad_norm": 1.328125,
"learning_rate": 4.7000000000000004e-05,
"loss": 8.9848,
"mean_token_accuracy": 0.0728946004062891,
"num_tokens": 174958.0,
"step": 95
},
{
"entropy": 10.01873140335083,
"epoch": 0.008830801836806783,
"grad_norm": 1.34375,
"learning_rate": 4.9500000000000004e-05,
"loss": 8.8889,
"mean_token_accuracy": 0.07516518756747245,
"num_tokens": 184256.0,
"step": 100
},
{
"entropy": 9.956882572174072,
"epoch": 0.009272341928647121,
"grad_norm": 1.546875,
"learning_rate": 5.2e-05,
"loss": 8.7839,
"mean_token_accuracy": 0.067950439453125,
"num_tokens": 192894.0,
"step": 105
},
{
"entropy": 9.884513092041015,
"epoch": 0.00971388202048746,
"grad_norm": 1.375,
"learning_rate": 5.45e-05,
"loss": 8.6868,
"mean_token_accuracy": 0.07383731976151467,
"num_tokens": 202675.0,
"step": 110
},
{
"entropy": 9.810705184936523,
"epoch": 0.010155422112327799,
"grad_norm": 1.1953125,
"learning_rate": 5.7e-05,
"loss": 8.624,
"mean_token_accuracy": 0.07006355635821819,
"num_tokens": 212261.0,
"step": 115
},
{
"entropy": 9.743825721740723,
"epoch": 0.010596962204168139,
"grad_norm": 1.171875,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.599,
"mean_token_accuracy": 0.06874018795788288,
"num_tokens": 222329.0,
"step": 120
},
{
"entropy": 9.528209781646728,
"epoch": 0.011038502296008477,
"grad_norm": 1.109375,
"learning_rate": 6.2e-05,
"loss": 8.4477,
"mean_token_accuracy": 0.06682575456798076,
"num_tokens": 231247.0,
"step": 125
},
{
"entropy": 9.442446994781495,
"epoch": 0.011480042387848817,
"grad_norm": 1.15625,
"learning_rate": 6.450000000000001e-05,
"loss": 8.3412,
"mean_token_accuracy": 0.06921537183225154,
"num_tokens": 239978.0,
"step": 130
},
{
"entropy": 9.272939491271973,
"epoch": 0.011921582479689156,
"grad_norm": 0.90625,
"learning_rate": 6.7e-05,
"loss": 8.3,
"mean_token_accuracy": 0.06850462295114994,
"num_tokens": 249735.0,
"step": 135
},
{
"entropy": 9.222266483306885,
"epoch": 0.012363122571529495,
"grad_norm": 1.15625,
"learning_rate": 6.950000000000001e-05,
"loss": 8.2344,
"mean_token_accuracy": 0.06959122642874718,
"num_tokens": 259369.0,
"step": 140
},
{
"entropy": 8.956540203094482,
"epoch": 0.012804662663369834,
"grad_norm": 0.81640625,
"learning_rate": 7.2e-05,
"loss": 8.2305,
"mean_token_accuracy": 0.06539506763219834,
"num_tokens": 268645.0,
"step": 145
},
{
"entropy": 8.88605546951294,
"epoch": 0.013246202755210172,
"grad_norm": 1.1640625,
"learning_rate": 7.45e-05,
"loss": 8.0685,
"mean_token_accuracy": 0.07155903875827789,
"num_tokens": 276667.0,
"step": 150
},
{
"entropy": 8.623716259002686,
"epoch": 0.013687742847050512,
"grad_norm": 0.703125,
"learning_rate": 7.7e-05,
"loss": 8.0682,
"mean_token_accuracy": 0.07539896108210087,
"num_tokens": 286017.0,
"step": 155
},
{
"entropy": 8.590844440460206,
"epoch": 0.01412928293889085,
"grad_norm": 0.83984375,
"learning_rate": 7.950000000000001e-05,
"loss": 8.0571,
"mean_token_accuracy": 0.07278457470238209,
"num_tokens": 295631.0,
"step": 160
},
{
"entropy": 8.562520027160645,
"epoch": 0.01457082303073119,
"grad_norm": 0.86328125,
"learning_rate": 8.2e-05,
"loss": 8.0486,
"mean_token_accuracy": 0.06986252851784229,
"num_tokens": 304704.0,
"step": 165
},
{
"entropy": 8.490843200683594,
"epoch": 0.01501236312257153,
"grad_norm": 0.7734375,
"learning_rate": 8.450000000000001e-05,
"loss": 8.0665,
"mean_token_accuracy": 0.07160350978374481,
"num_tokens": 314195.0,
"step": 170
},
{
"entropy": 8.416227722167969,
"epoch": 0.015453903214411868,
"grad_norm": 1.0078125,
"learning_rate": 8.7e-05,
"loss": 8.0632,
"mean_token_accuracy": 0.07028085552155972,
"num_tokens": 323379.0,
"step": 175
},
{
"entropy": 8.398184299468994,
"epoch": 0.015895443306252206,
"grad_norm": 0.734375,
"learning_rate": 8.95e-05,
"loss": 7.9637,
"mean_token_accuracy": 0.08185541778802871,
"num_tokens": 332322.0,
"step": 180
},
{
"entropy": 8.336036014556885,
"epoch": 0.016336983398092548,
"grad_norm": 0.79296875,
"learning_rate": 9.2e-05,
"loss": 7.9427,
"mean_token_accuracy": 0.08073886930942535,
"num_tokens": 341735.0,
"step": 185
},
{
"entropy": 8.337114715576172,
"epoch": 0.016778523489932886,
"grad_norm": 0.92578125,
"learning_rate": 9.45e-05,
"loss": 8.0349,
"mean_token_accuracy": 0.06938613168895244,
"num_tokens": 351209.0,
"step": 190
},
{
"entropy": 8.39198350906372,
"epoch": 0.017220063581773224,
"grad_norm": 0.98046875,
"learning_rate": 9.7e-05,
"loss": 7.9792,
"mean_token_accuracy": 0.07559169828891754,
"num_tokens": 360467.0,
"step": 195
},
{
"entropy": 8.235328102111817,
"epoch": 0.017661603673613566,
"grad_norm": 0.875,
"learning_rate": 9.95e-05,
"loss": 7.9423,
"mean_token_accuracy": 0.07669526152312756,
"num_tokens": 370361.0,
"step": 200
},
{
"entropy": 8.374059581756592,
"epoch": 0.018103143765453904,
"grad_norm": 1.2109375,
"learning_rate": 0.000102,
"loss": 8.0075,
"mean_token_accuracy": 0.07567069008946418,
"num_tokens": 380366.0,
"step": 205
},
{
"entropy": 8.206629276275635,
"epoch": 0.018544683857294242,
"grad_norm": 1.0,
"learning_rate": 0.00010449999999999999,
"loss": 7.9185,
"mean_token_accuracy": 0.08159190192818641,
"num_tokens": 390690.0,
"step": 210
},
{
"entropy": 8.24603796005249,
"epoch": 0.01898622394913458,
"grad_norm": 0.87109375,
"learning_rate": 0.000107,
"loss": 7.9601,
"mean_token_accuracy": 0.0793293446302414,
"num_tokens": 400722.0,
"step": 215
},
{
"entropy": 8.157498931884765,
"epoch": 0.01942776404097492,
"grad_norm": 0.84375,
"learning_rate": 0.0001095,
"loss": 7.8501,
"mean_token_accuracy": 0.08334142193198205,
"num_tokens": 410223.0,
"step": 220
},
{
"entropy": 8.205572509765625,
"epoch": 0.01986930413281526,
"grad_norm": 1.0,
"learning_rate": 0.000112,
"loss": 7.9021,
"mean_token_accuracy": 0.07716193869709968,
"num_tokens": 420214.0,
"step": 225
},
{
"entropy": 8.172825717926026,
"epoch": 0.020310844224655598,
"grad_norm": 0.88671875,
"learning_rate": 0.0001145,
"loss": 7.8564,
"mean_token_accuracy": 0.08035471551120281,
"num_tokens": 429407.0,
"step": 230
},
{
"entropy": 8.162760925292968,
"epoch": 0.02075238431649594,
"grad_norm": 1.4609375,
"learning_rate": 0.00011700000000000001,
"loss": 7.8246,
"mean_token_accuracy": 0.07542734369635581,
"num_tokens": 438403.0,
"step": 235
},
{
"entropy": 8.177341651916503,
"epoch": 0.021193924408336277,
"grad_norm": 1.2109375,
"learning_rate": 0.00011949999999999999,
"loss": 7.8545,
"mean_token_accuracy": 0.08535856604576111,
"num_tokens": 447466.0,
"step": 240
},
{
"entropy": 8.069422197341918,
"epoch": 0.021635464500176615,
"grad_norm": 0.921875,
"learning_rate": 0.000122,
"loss": 7.9366,
"mean_token_accuracy": 0.07459555268287658,
"num_tokens": 457141.0,
"step": 245
},
{
"entropy": 8.160084056854249,
"epoch": 0.022077004592016954,
"grad_norm": 0.86328125,
"learning_rate": 0.0001245,
"loss": 7.812,
"mean_token_accuracy": 0.08223466873168946,
"num_tokens": 465708.0,
"step": 250
},
{
"entropy": 8.142998504638673,
"epoch": 0.022518544683857295,
"grad_norm": 1.0859375,
"learning_rate": 0.000127,
"loss": 7.8339,
"mean_token_accuracy": 0.07565066292881965,
"num_tokens": 475369.0,
"step": 255
},
{
"entropy": 8.075135421752929,
"epoch": 0.022960084775697633,
"grad_norm": 1.0859375,
"learning_rate": 0.0001295,
"loss": 7.7972,
"mean_token_accuracy": 0.08645984381437302,
"num_tokens": 484249.0,
"step": 260
},
{
"entropy": 8.122587871551513,
"epoch": 0.02340162486753797,
"grad_norm": 1.109375,
"learning_rate": 0.000132,
"loss": 7.8872,
"mean_token_accuracy": 0.07687325775623322,
"num_tokens": 493303.0,
"step": 265
},
{
"entropy": 8.101485538482667,
"epoch": 0.023843164959378313,
"grad_norm": 1.0859375,
"learning_rate": 0.00013450000000000002,
"loss": 7.8664,
"mean_token_accuracy": 0.0807331919670105,
"num_tokens": 501503.0,
"step": 270
},
{
"entropy": 8.036290693283082,
"epoch": 0.02428470505121865,
"grad_norm": 1.2109375,
"learning_rate": 0.00013700000000000002,
"loss": 7.8074,
"mean_token_accuracy": 0.08591768592596054,
"num_tokens": 509661.0,
"step": 275
},
{
"entropy": 8.045488977432251,
"epoch": 0.02472624514305899,
"grad_norm": 1.1875,
"learning_rate": 0.0001395,
"loss": 7.7904,
"mean_token_accuracy": 0.08441019728779793,
"num_tokens": 519464.0,
"step": 280
},
{
"entropy": 8.107398653030396,
"epoch": 0.025167785234899327,
"grad_norm": 1.0625,
"learning_rate": 0.00014199999999999998,
"loss": 7.7489,
"mean_token_accuracy": 0.08773190379142762,
"num_tokens": 527968.0,
"step": 285
},
{
"entropy": 8.081705808639526,
"epoch": 0.02560932532673967,
"grad_norm": 1.1875,
"learning_rate": 0.0001445,
"loss": 7.7768,
"mean_token_accuracy": 0.0868467777967453,
"num_tokens": 537234.0,
"step": 290
},
{
"entropy": 7.99565052986145,
"epoch": 0.026050865418580007,
"grad_norm": 1.0078125,
"learning_rate": 0.000147,
"loss": 7.7747,
"mean_token_accuracy": 0.08527034223079681,
"num_tokens": 546398.0,
"step": 295
},
{
"entropy": 8.011523675918578,
"epoch": 0.026492405510420345,
"grad_norm": 1.015625,
"learning_rate": 0.0001495,
"loss": 7.7616,
"mean_token_accuracy": 0.08982880860567093,
"num_tokens": 555362.0,
"step": 300
},
{
"entropy": 8.107937812805176,
"epoch": 0.026933945602260687,
"grad_norm": 1.0,
"learning_rate": 0.000152,
"loss": 7.8221,
"mean_token_accuracy": 0.07775180079042912,
"num_tokens": 564575.0,
"step": 305
},
{
"entropy": 8.133016395568848,
"epoch": 0.027375485694101025,
"grad_norm": 1.0390625,
"learning_rate": 0.00015450000000000001,
"loss": 7.8384,
"mean_token_accuracy": 0.08304800540208816,
"num_tokens": 573915.0,
"step": 310
},
{
"entropy": 8.016209363937378,
"epoch": 0.027817025785941363,
"grad_norm": 1.0390625,
"learning_rate": 0.000157,
"loss": 7.7322,
"mean_token_accuracy": 0.08581754639744758,
"num_tokens": 583216.0,
"step": 315
},
{
"entropy": 7.982406425476074,
"epoch": 0.0282585658777817,
"grad_norm": 1.09375,
"learning_rate": 0.0001595,
"loss": 7.7553,
"mean_token_accuracy": 0.08679840788245201,
"num_tokens": 591955.0,
"step": 320
},
{
"entropy": 7.9430736064910885,
"epoch": 0.028700105969622042,
"grad_norm": 1.2265625,
"learning_rate": 0.000162,
"loss": 7.7588,
"mean_token_accuracy": 0.08934888392686843,
"num_tokens": 600999.0,
"step": 325
},
{
"entropy": 8.070584297180176,
"epoch": 0.02914164606146238,
"grad_norm": 1.2109375,
"learning_rate": 0.00016450000000000001,
"loss": 7.6563,
"mean_token_accuracy": 0.09217674285173416,
"num_tokens": 609478.0,
"step": 330
},
{
"entropy": 7.987708568572998,
"epoch": 0.02958318615330272,
"grad_norm": 1.3125,
"learning_rate": 0.00016700000000000002,
"loss": 7.7225,
"mean_token_accuracy": 0.08663035854697228,
"num_tokens": 618348.0,
"step": 335
},
{
"entropy": 7.911137056350708,
"epoch": 0.03002472624514306,
"grad_norm": 1.1484375,
"learning_rate": 0.00016950000000000003,
"loss": 7.7691,
"mean_token_accuracy": 0.083287762850523,
"num_tokens": 628548.0,
"step": 340
},
{
"entropy": 8.057271575927734,
"epoch": 0.0304662663369834,
"grad_norm": 1.0234375,
"learning_rate": 0.00017199999999999998,
"loss": 7.7102,
"mean_token_accuracy": 0.08196588605642319,
"num_tokens": 637489.0,
"step": 345
},
{
"entropy": 7.939978122711182,
"epoch": 0.030907806428823736,
"grad_norm": 0.953125,
"learning_rate": 0.00017449999999999999,
"loss": 7.689,
"mean_token_accuracy": 0.0814521424472332,
"num_tokens": 646715.0,
"step": 350
},
{
"entropy": 7.897878551483155,
"epoch": 0.031349346520664075,
"grad_norm": 0.98046875,
"learning_rate": 0.000177,
"loss": 7.6894,
"mean_token_accuracy": 0.08998864293098449,
"num_tokens": 656858.0,
"step": 355
},
{
"entropy": 8.019395637512208,
"epoch": 0.03179088661250441,
"grad_norm": 0.98828125,
"learning_rate": 0.0001795,
"loss": 7.6755,
"mean_token_accuracy": 0.08710955381393433,
"num_tokens": 665968.0,
"step": 360
},
{
"entropy": 8.001319217681885,
"epoch": 0.03223242670434476,
"grad_norm": 1.6328125,
"learning_rate": 0.000182,
"loss": 7.6655,
"mean_token_accuracy": 0.08621685430407525,
"num_tokens": 674295.0,
"step": 365
},
{
"entropy": 7.810992002487183,
"epoch": 0.032673966796185096,
"grad_norm": 1.234375,
"learning_rate": 0.0001845,
"loss": 7.591,
"mean_token_accuracy": 0.08370614722371102,
"num_tokens": 683559.0,
"step": 370
},
{
"entropy": 7.816927337646485,
"epoch": 0.033115506888025434,
"grad_norm": 1.21875,
"learning_rate": 0.000187,
"loss": 7.6253,
"mean_token_accuracy": 0.08996079638600349,
"num_tokens": 692402.0,
"step": 375
},
{
"entropy": 7.967683601379394,
"epoch": 0.03355704697986577,
"grad_norm": 1.2890625,
"learning_rate": 0.0001895,
"loss": 7.7304,
"mean_token_accuracy": 0.08065761215984821,
"num_tokens": 702052.0,
"step": 380
},
{
"entropy": 8.058749055862426,
"epoch": 0.03399858707170611,
"grad_norm": 1.6328125,
"learning_rate": 0.000192,
"loss": 7.6528,
"mean_token_accuracy": 0.08705045655369759,
"num_tokens": 711926.0,
"step": 385
},
{
"entropy": 7.8771873950958256,
"epoch": 0.03444012716354645,
"grad_norm": 0.96875,
"learning_rate": 0.0001945,
"loss": 7.612,
"mean_token_accuracy": 0.08773906156420708,
"num_tokens": 720948.0,
"step": 390
},
{
"entropy": 7.893786334991455,
"epoch": 0.034881667255386786,
"grad_norm": 1.5,
"learning_rate": 0.00019700000000000002,
"loss": 7.6301,
"mean_token_accuracy": 0.09444142654538154,
"num_tokens": 729611.0,
"step": 395
},
{
"entropy": 7.892533588409424,
"epoch": 0.03532320734722713,
"grad_norm": 1.15625,
"learning_rate": 0.00019950000000000002,
"loss": 7.6187,
"mean_token_accuracy": 0.08193654045462609,
"num_tokens": 738433.0,
"step": 400
},
{
"entropy": 7.945340347290039,
"epoch": 0.03576474743906747,
"grad_norm": 1.0859375,
"learning_rate": 0.000202,
"loss": 7.6397,
"mean_token_accuracy": 0.08668759167194366,
"num_tokens": 747310.0,
"step": 405
},
{
"entropy": 7.854477500915527,
"epoch": 0.03620628753090781,
"grad_norm": 1.1328125,
"learning_rate": 0.00020449999999999998,
"loss": 7.5994,
"mean_token_accuracy": 0.09020926207304,
"num_tokens": 756362.0,
"step": 410
},
{
"entropy": 7.90778489112854,
"epoch": 0.036647827622748146,
"grad_norm": 0.984375,
"learning_rate": 0.000207,
"loss": 7.6034,
"mean_token_accuracy": 0.08586042672395706,
"num_tokens": 764978.0,
"step": 415
},
{
"entropy": 7.87300386428833,
"epoch": 0.037089367714588484,
"grad_norm": 1.03125,
"learning_rate": 0.0002095,
"loss": 7.5707,
"mean_token_accuracy": 0.09018185958266259,
"num_tokens": 774058.0,
"step": 420
},
{
"entropy": 7.795767593383789,
"epoch": 0.03753090780642882,
"grad_norm": 1.0390625,
"learning_rate": 0.000212,
"loss": 7.5492,
"mean_token_accuracy": 0.0897379383444786,
"num_tokens": 783332.0,
"step": 425
},
{
"entropy": 7.853004789352417,
"epoch": 0.03797244789826916,
"grad_norm": 1.3203125,
"learning_rate": 0.0002145,
"loss": 7.6298,
"mean_token_accuracy": 0.08684360906481743,
"num_tokens": 792481.0,
"step": 430
},
{
"entropy": 7.766995525360107,
"epoch": 0.038413987990109505,
"grad_norm": 1.1015625,
"learning_rate": 0.00021700000000000002,
"loss": 7.5212,
"mean_token_accuracy": 0.09301207512617111,
"num_tokens": 801396.0,
"step": 435
},
{
"entropy": 7.8428326606750485,
"epoch": 0.03885552808194984,
"grad_norm": 1.0078125,
"learning_rate": 0.0002195,
"loss": 7.5865,
"mean_token_accuracy": 0.08940735682845116,
"num_tokens": 809903.0,
"step": 440
},
{
"entropy": 7.828377294540405,
"epoch": 0.03929706817379018,
"grad_norm": 0.99609375,
"learning_rate": 0.000222,
"loss": 7.5389,
"mean_token_accuracy": 0.0962544821202755,
"num_tokens": 819144.0,
"step": 445
},
{
"entropy": 7.7183678150177,
"epoch": 0.03973860826563052,
"grad_norm": 1.1328125,
"learning_rate": 0.0002245,
"loss": 7.5608,
"mean_token_accuracy": 0.08849129751324654,
"num_tokens": 828881.0,
"step": 450
},
{
"entropy": 7.764478397369385,
"epoch": 0.04018014835747086,
"grad_norm": 1.109375,
"learning_rate": 0.00022700000000000002,
"loss": 7.4877,
"mean_token_accuracy": 0.08765893578529357,
"num_tokens": 837588.0,
"step": 455
},
{
"entropy": 7.767373847961426,
"epoch": 0.040621688449311195,
"grad_norm": 1.0,
"learning_rate": 0.00022950000000000002,
"loss": 7.4849,
"mean_token_accuracy": 0.09265839084982871,
"num_tokens": 847002.0,
"step": 460
},
{
"entropy": 7.726333475112915,
"epoch": 0.041063228541151534,
"grad_norm": 1.1015625,
"learning_rate": 0.00023200000000000003,
"loss": 7.4945,
"mean_token_accuracy": 0.09333177357912063,
"num_tokens": 855791.0,
"step": 465
},
{
"entropy": 7.7462080955505375,
"epoch": 0.04150476863299188,
"grad_norm": 1.1484375,
"learning_rate": 0.00023449999999999998,
"loss": 7.5242,
"mean_token_accuracy": 0.0911882683634758,
"num_tokens": 865392.0,
"step": 470
},
{
"entropy": 7.736569499969482,
"epoch": 0.04194630872483222,
"grad_norm": 1.2109375,
"learning_rate": 0.000237,
"loss": 7.5873,
"mean_token_accuracy": 0.08873779252171517,
"num_tokens": 874807.0,
"step": 475
},
{
"entropy": 7.758917284011841,
"epoch": 0.042387848816672555,
"grad_norm": 1.1875,
"learning_rate": 0.0002395,
"loss": 7.5409,
"mean_token_accuracy": 0.09495326653122901,
"num_tokens": 883928.0,
"step": 480
},
{
"entropy": 7.777913904190063,
"epoch": 0.04282938890851289,
"grad_norm": 1.140625,
"learning_rate": 0.000242,
"loss": 7.4436,
"mean_token_accuracy": 0.09124857932329178,
"num_tokens": 893047.0,
"step": 485
},
{
"entropy": 7.662859010696411,
"epoch": 0.04327092900035323,
"grad_norm": 1.2421875,
"learning_rate": 0.0002445,
"loss": 7.4593,
"mean_token_accuracy": 0.09315531030297279,
"num_tokens": 901645.0,
"step": 490
},
{
"entropy": 7.743328475952149,
"epoch": 0.04371246909219357,
"grad_norm": 1.109375,
"learning_rate": 0.000247,
"loss": 7.4727,
"mean_token_accuracy": 0.09244368895888329,
"num_tokens": 911169.0,
"step": 495
},
{
"entropy": 7.7239625453948975,
"epoch": 0.04415400918403391,
"grad_norm": 1.4140625,
"learning_rate": 0.0002495,
"loss": 7.4748,
"mean_token_accuracy": 0.08498905003070831,
"num_tokens": 921382.0,
"step": 500
},
{
"entropy": 7.544922304153443,
"epoch": 0.04459554927587425,
"grad_norm": 1.234375,
"learning_rate": 0.000252,
"loss": 7.4121,
"mean_token_accuracy": 0.09429771155118942,
"num_tokens": 930409.0,
"step": 505
},
{
"entropy": 7.67856912612915,
"epoch": 0.04503708936771459,
"grad_norm": 1.078125,
"learning_rate": 0.0002545,
"loss": 7.3879,
"mean_token_accuracy": 0.09879431128501892,
"num_tokens": 939049.0,
"step": 510
},
{
"entropy": 7.718625736236572,
"epoch": 0.04547862945955493,
"grad_norm": 1.25,
"learning_rate": 0.000257,
"loss": 7.395,
"mean_token_accuracy": 0.0960740551352501,
"num_tokens": 947575.0,
"step": 515
},
{
"entropy": 7.709804058074951,
"epoch": 0.045920169551395267,
"grad_norm": 1.09375,
"learning_rate": 0.0002595,
"loss": 7.5643,
"mean_token_accuracy": 0.09047991409897804,
"num_tokens": 957848.0,
"step": 520
},
{
"entropy": 7.655015087127685,
"epoch": 0.046361709643235605,
"grad_norm": 1.1484375,
"learning_rate": 0.000262,
"loss": 7.3857,
"mean_token_accuracy": 0.09998803585767746,
"num_tokens": 966521.0,
"step": 525
},
{
"entropy": 7.688518905639649,
"epoch": 0.04680324973507594,
"grad_norm": 1.3359375,
"learning_rate": 0.00026450000000000003,
"loss": 7.4461,
"mean_token_accuracy": 0.09324755370616913,
"num_tokens": 975827.0,
"step": 530
},
{
"entropy": 7.606715154647827,
"epoch": 0.04724478982691628,
"grad_norm": 1.1796875,
"learning_rate": 0.00026700000000000004,
"loss": 7.4075,
"mean_token_accuracy": 0.09566703587770461,
"num_tokens": 985292.0,
"step": 535
},
{
"entropy": 7.616068124771118,
"epoch": 0.047686329918756626,
"grad_norm": 1.09375,
"learning_rate": 0.00026950000000000005,
"loss": 7.3841,
"mean_token_accuracy": 0.09411159604787826,
"num_tokens": 994791.0,
"step": 540
},
{
"entropy": 7.51567816734314,
"epoch": 0.048127870010596964,
"grad_norm": 1.1015625,
"learning_rate": 0.00027200000000000005,
"loss": 7.3594,
"mean_token_accuracy": 0.1026044063270092,
"num_tokens": 1003700.0,
"step": 545
},
{
"entropy": 7.510391616821289,
"epoch": 0.0485694101024373,
"grad_norm": 1.0,
"learning_rate": 0.0002745,
"loss": 7.381,
"mean_token_accuracy": 0.09829011410474778,
"num_tokens": 1012682.0,
"step": 550
},
{
"entropy": 7.683912038803101,
"epoch": 0.04901095019427764,
"grad_norm": 1.28125,
"learning_rate": 0.000277,
"loss": 7.4075,
"mean_token_accuracy": 0.09179475829005242,
"num_tokens": 1021018.0,
"step": 555
},
{
"entropy": 7.570155191421509,
"epoch": 0.04945249028611798,
"grad_norm": 1.3046875,
"learning_rate": 0.0002795,
"loss": 7.2759,
"mean_token_accuracy": 0.09699172824621201,
"num_tokens": 1029744.0,
"step": 560
},
{
"entropy": 7.495694351196289,
"epoch": 0.049894030377958316,
"grad_norm": 1.1484375,
"learning_rate": 0.00028199999999999997,
"loss": 7.3605,
"mean_token_accuracy": 0.09879247918725013,
"num_tokens": 1038805.0,
"step": 565
},
{
"entropy": 7.5144976615905765,
"epoch": 0.050335570469798654,
"grad_norm": 1.328125,
"learning_rate": 0.0002845,
"loss": 7.4131,
"mean_token_accuracy": 0.0988279327750206,
"num_tokens": 1047656.0,
"step": 570
},
{
"entropy": 7.647522783279419,
"epoch": 0.050777110561639,
"grad_norm": 1.140625,
"learning_rate": 0.000287,
"loss": 7.4048,
"mean_token_accuracy": 0.09629088416695594,
"num_tokens": 1056598.0,
"step": 575
},
{
"entropy": 7.6095935821533205,
"epoch": 0.05121865065347934,
"grad_norm": 1.1015625,
"learning_rate": 0.0002895,
"loss": 7.3994,
"mean_token_accuracy": 0.09847217947244644,
"num_tokens": 1065226.0,
"step": 580
},
{
"entropy": 7.529495334625244,
"epoch": 0.051660190745319676,
"grad_norm": 1.4375,
"learning_rate": 0.000292,
"loss": 7.3208,
"mean_token_accuracy": 0.10098938867449761,
"num_tokens": 1074661.0,
"step": 585
},
{
"entropy": 7.503559398651123,
"epoch": 0.052101730837160014,
"grad_norm": 1.1484375,
"learning_rate": 0.0002945,
"loss": 7.4251,
"mean_token_accuracy": 0.09441772177815437,
"num_tokens": 1083921.0,
"step": 590
},
{
"entropy": 7.540312194824219,
"epoch": 0.05254327092900035,
"grad_norm": 1.1328125,
"learning_rate": 0.000297,
"loss": 7.3162,
"mean_token_accuracy": 0.104298634827137,
"num_tokens": 1093399.0,
"step": 595
},
{
"entropy": 7.5133528232574465,
"epoch": 0.05298481102084069,
"grad_norm": 1.2734375,
"learning_rate": 0.0002995,
"loss": 7.391,
"mean_token_accuracy": 0.09825902208685874,
"num_tokens": 1104065.0,
"step": 600
},
{
"entropy": 7.433008003234863,
"epoch": 0.05342635111268103,
"grad_norm": 1.125,
"learning_rate": 0.000302,
"loss": 7.2778,
"mean_token_accuracy": 0.1005440428853035,
"num_tokens": 1112995.0,
"step": 605
},
{
"entropy": 7.47243971824646,
"epoch": 0.05386789120452137,
"grad_norm": 1.0859375,
"learning_rate": 0.0003045,
"loss": 7.3103,
"mean_token_accuracy": 0.10175202563405036,
"num_tokens": 1121637.0,
"step": 610
},
{
"entropy": 7.455365610122681,
"epoch": 0.05430943129636171,
"grad_norm": 1.140625,
"learning_rate": 0.000307,
"loss": 7.2549,
"mean_token_accuracy": 0.09826337993144989,
"num_tokens": 1131166.0,
"step": 615
},
{
"entropy": 7.4712036609649655,
"epoch": 0.05475097138820205,
"grad_norm": 1.0859375,
"learning_rate": 0.0003095,
"loss": 7.2562,
"mean_token_accuracy": 0.10475531965494156,
"num_tokens": 1140888.0,
"step": 620
},
{
"entropy": 7.551609897613526,
"epoch": 0.05519251148004239,
"grad_norm": 1.1171875,
"learning_rate": 0.000312,
"loss": 7.4148,
"mean_token_accuracy": 0.0961816966533661,
"num_tokens": 1150278.0,
"step": 625
},
{
"entropy": 7.433546924591065,
"epoch": 0.055634051571882726,
"grad_norm": 1.4375,
"learning_rate": 0.0003145,
"loss": 7.3742,
"mean_token_accuracy": 0.0970606379210949,
"num_tokens": 1159348.0,
"step": 630
},
{
"entropy": 7.624134588241577,
"epoch": 0.056075591663723064,
"grad_norm": 1.046875,
"learning_rate": 0.000317,
"loss": 7.3756,
"mean_token_accuracy": 0.0949991799890995,
"num_tokens": 1168883.0,
"step": 635
},
{
"entropy": 7.48681526184082,
"epoch": 0.0565171317555634,
"grad_norm": 1.0,
"learning_rate": 0.0003195,
"loss": 7.269,
"mean_token_accuracy": 0.1064944364130497,
"num_tokens": 1178572.0,
"step": 640
},
{
"entropy": 7.44178466796875,
"epoch": 0.05695867184740375,
"grad_norm": 1.046875,
"learning_rate": 0.000322,
"loss": 7.3576,
"mean_token_accuracy": 0.0987204596400261,
"num_tokens": 1188909.0,
"step": 645
},
{
"entropy": 7.466546869277954,
"epoch": 0.057400211939244085,
"grad_norm": 1.28125,
"learning_rate": 0.00032450000000000003,
"loss": 7.2649,
"mean_token_accuracy": 0.09890259429812431,
"num_tokens": 1197705.0,
"step": 650
},
{
"entropy": 7.450878572463989,
"epoch": 0.05784175203108442,
"grad_norm": 1.1171875,
"learning_rate": 0.00032700000000000003,
"loss": 7.1061,
"mean_token_accuracy": 0.10522415414452553,
"num_tokens": 1206351.0,
"step": 655
},
{
"entropy": 7.340301847457885,
"epoch": 0.05828329212292476,
"grad_norm": 1.3359375,
"learning_rate": 0.00032950000000000004,
"loss": 7.2237,
"mean_token_accuracy": 0.09693196043372154,
"num_tokens": 1214984.0,
"step": 660
},
{
"entropy": 7.4402018070220945,
"epoch": 0.0587248322147651,
"grad_norm": 1.1796875,
"learning_rate": 0.00033200000000000005,
"loss": 7.2999,
"mean_token_accuracy": 0.09738482013344765,
"num_tokens": 1224485.0,
"step": 665
},
{
"entropy": 7.435847473144531,
"epoch": 0.05916637230660544,
"grad_norm": 1.1796875,
"learning_rate": 0.00033450000000000005,
"loss": 7.3266,
"mean_token_accuracy": 0.09173622950911522,
"num_tokens": 1233560.0,
"step": 670
},
{
"entropy": 7.428315305709839,
"epoch": 0.05960791239844578,
"grad_norm": 1.3515625,
"learning_rate": 0.000337,
"loss": 7.2436,
"mean_token_accuracy": 0.09967414885759354,
"num_tokens": 1242628.0,
"step": 675
},
{
"entropy": 7.388672494888306,
"epoch": 0.06004945249028612,
"grad_norm": 1.0703125,
"learning_rate": 0.0003395,
"loss": 7.1697,
"mean_token_accuracy": 0.10538085550069809,
"num_tokens": 1251004.0,
"step": 680
},
{
"entropy": 7.459445238113403,
"epoch": 0.06049099258212646,
"grad_norm": 1.0703125,
"learning_rate": 0.000342,
"loss": 7.3463,
"mean_token_accuracy": 0.09609238728880883,
"num_tokens": 1260344.0,
"step": 685
},
{
"entropy": 7.343485164642334,
"epoch": 0.0609325326739668,
"grad_norm": 1.0625,
"learning_rate": 0.00034449999999999997,
"loss": 7.2517,
"mean_token_accuracy": 0.09760257676243782,
"num_tokens": 1269988.0,
"step": 690
},
{
"entropy": 7.340139007568359,
"epoch": 0.061374072765807135,
"grad_norm": 1.0859375,
"learning_rate": 0.000347,
"loss": 7.2126,
"mean_token_accuracy": 0.10715288370847702,
"num_tokens": 1280912.0,
"step": 695
},
{
"entropy": 7.350299119949341,
"epoch": 0.06181561285764747,
"grad_norm": 0.98828125,
"learning_rate": 0.0003495,
"loss": 7.2246,
"mean_token_accuracy": 0.10604915320873261,
"num_tokens": 1289684.0,
"step": 700
},
{
"entropy": 7.427703905105591,
"epoch": 0.06225715294948781,
"grad_norm": 1.0390625,
"learning_rate": 0.000352,
"loss": 7.2534,
"mean_token_accuracy": 0.09802542477846146,
"num_tokens": 1298853.0,
"step": 705
},
{
"entropy": 7.3198949813842775,
"epoch": 0.06269869304132815,
"grad_norm": 1.0390625,
"learning_rate": 0.0003545,
"loss": 7.214,
"mean_token_accuracy": 0.10874532908201218,
"num_tokens": 1309112.0,
"step": 710
},
{
"entropy": 7.372763156890869,
"epoch": 0.0631402331331685,
"grad_norm": 1.296875,
"learning_rate": 0.000357,
"loss": 7.1981,
"mean_token_accuracy": 0.10583075731992722,
"num_tokens": 1319064.0,
"step": 715
},
{
"entropy": 7.269387340545654,
"epoch": 0.06358177322500883,
"grad_norm": 1.1015625,
"learning_rate": 0.0003595,
"loss": 7.1748,
"mean_token_accuracy": 0.1100200168788433,
"num_tokens": 1327889.0,
"step": 720
},
{
"entropy": 7.324726533889771,
"epoch": 0.06402331331684917,
"grad_norm": 1.28125,
"learning_rate": 0.000362,
"loss": 7.1938,
"mean_token_accuracy": 0.10258080512285232,
"num_tokens": 1337241.0,
"step": 725
},
{
"entropy": 7.319574499130249,
"epoch": 0.06446485340868952,
"grad_norm": 1.1015625,
"learning_rate": 0.0003645,
"loss": 7.2533,
"mean_token_accuracy": 0.10085726305842399,
"num_tokens": 1346527.0,
"step": 730
},
{
"entropy": 7.314885807037354,
"epoch": 0.06490639350052985,
"grad_norm": 1.125,
"learning_rate": 0.000367,
"loss": 7.2315,
"mean_token_accuracy": 0.10445504561066628,
"num_tokens": 1355677.0,
"step": 735
},
{
"entropy": 7.396700429916382,
"epoch": 0.06534793359237019,
"grad_norm": 1.03125,
"learning_rate": 0.0003695,
"loss": 7.2163,
"mean_token_accuracy": 0.10588330775499344,
"num_tokens": 1364874.0,
"step": 740
},
{
"entropy": 7.285468482971192,
"epoch": 0.06578947368421052,
"grad_norm": 1.1015625,
"learning_rate": 0.000372,
"loss": 7.1378,
"mean_token_accuracy": 0.1090671844780445,
"num_tokens": 1373717.0,
"step": 745
},
{
"entropy": 7.375531625747681,
"epoch": 0.06623101377605087,
"grad_norm": 0.984375,
"learning_rate": 0.0003745,
"loss": 7.0955,
"mean_token_accuracy": 0.10741576477885247,
"num_tokens": 1382767.0,
"step": 750
},
{
"entropy": 7.1357104778289795,
"epoch": 0.0666725538678912,
"grad_norm": 1.3515625,
"learning_rate": 0.000377,
"loss": 7.1389,
"mean_token_accuracy": 0.10613262876868249,
"num_tokens": 1391190.0,
"step": 755
},
{
"entropy": 7.234480524063111,
"epoch": 0.06711409395973154,
"grad_norm": 1.1796875,
"learning_rate": 0.0003795,
"loss": 7.1509,
"mean_token_accuracy": 0.10508784130215645,
"num_tokens": 1400722.0,
"step": 760
},
{
"entropy": 7.402392435073852,
"epoch": 0.06755563405157189,
"grad_norm": 1.03125,
"learning_rate": 0.000382,
"loss": 7.1779,
"mean_token_accuracy": 0.10437385067343712,
"num_tokens": 1409328.0,
"step": 765
},
{
"entropy": 7.06873927116394,
"epoch": 0.06799717414341222,
"grad_norm": 1.2109375,
"learning_rate": 0.0003845,
"loss": 7.0531,
"mean_token_accuracy": 0.11192933171987533,
"num_tokens": 1418504.0,
"step": 770
},
{
"entropy": 7.440014839172363,
"epoch": 0.06843871423525257,
"grad_norm": 1.0703125,
"learning_rate": 0.00038700000000000003,
"loss": 7.1989,
"mean_token_accuracy": 0.10317453742027283,
"num_tokens": 1427690.0,
"step": 775
},
{
"entropy": 7.181108903884888,
"epoch": 0.0688802543270929,
"grad_norm": 1.125,
"learning_rate": 0.00038950000000000003,
"loss": 7.1806,
"mean_token_accuracy": 0.10798285007476807,
"num_tokens": 1436798.0,
"step": 780
},
{
"entropy": 7.2046185493469235,
"epoch": 0.06932179441893324,
"grad_norm": 0.96875,
"learning_rate": 0.00039200000000000004,
"loss": 7.1646,
"mean_token_accuracy": 0.10358999595046044,
"num_tokens": 1446357.0,
"step": 785
},
{
"entropy": 7.2555629253387455,
"epoch": 0.06976333451077357,
"grad_norm": 1.0703125,
"learning_rate": 0.00039450000000000005,
"loss": 7.0882,
"mean_token_accuracy": 0.11000654250383377,
"num_tokens": 1455998.0,
"step": 790
},
{
"entropy": 7.207996559143067,
"epoch": 0.07020487460261392,
"grad_norm": 0.984375,
"learning_rate": 0.00039700000000000005,
"loss": 7.145,
"mean_token_accuracy": 0.09857687279582024,
"num_tokens": 1465237.0,
"step": 795
},
{
"entropy": 7.24621729850769,
"epoch": 0.07064641469445426,
"grad_norm": 1.1015625,
"learning_rate": 0.0003995,
"loss": 7.0958,
"mean_token_accuracy": 0.11087250858545303,
"num_tokens": 1474363.0,
"step": 800
},
{
"entropy": 7.272359037399292,
"epoch": 0.0710879547862946,
"grad_norm": 1.3359375,
"learning_rate": 0.000402,
"loss": 7.1713,
"mean_token_accuracy": 0.10843008160591125,
"num_tokens": 1483379.0,
"step": 805
},
{
"entropy": 7.28739447593689,
"epoch": 0.07152949487813494,
"grad_norm": 1.1640625,
"learning_rate": 0.0004045,
"loss": 7.1265,
"mean_token_accuracy": 0.10922098532319069,
"num_tokens": 1492507.0,
"step": 810
},
{
"entropy": 7.144436979293824,
"epoch": 0.07197103496997527,
"grad_norm": 1.2734375,
"learning_rate": 0.00040699999999999997,
"loss": 7.0154,
"mean_token_accuracy": 0.11775125116109848,
"num_tokens": 1500888.0,
"step": 815
},
{
"entropy": 7.11500997543335,
"epoch": 0.07241257506181561,
"grad_norm": 0.9921875,
"learning_rate": 0.0004095,
"loss": 7.0709,
"mean_token_accuracy": 0.10802061259746551,
"num_tokens": 1510310.0,
"step": 820
},
{
"entropy": 7.1448290824890135,
"epoch": 0.07285411515365595,
"grad_norm": 1.1796875,
"learning_rate": 0.000412,
"loss": 7.0494,
"mean_token_accuracy": 0.11422519460320472,
"num_tokens": 1519427.0,
"step": 825
},
{
"entropy": 7.2035074710845945,
"epoch": 0.07329565524549629,
"grad_norm": 1.1328125,
"learning_rate": 0.0004145,
"loss": 7.0679,
"mean_token_accuracy": 0.1063395880162716,
"num_tokens": 1529456.0,
"step": 830
},
{
"entropy": 7.131991720199585,
"epoch": 0.07373719533733664,
"grad_norm": 1.078125,
"learning_rate": 0.000417,
"loss": 7.0241,
"mean_token_accuracy": 0.11403456106781959,
"num_tokens": 1537695.0,
"step": 835
},
{
"entropy": 7.203299617767334,
"epoch": 0.07417873542917697,
"grad_norm": 1.046875,
"learning_rate": 0.0004195,
"loss": 7.1456,
"mean_token_accuracy": 0.10954299196600914,
"num_tokens": 1547511.0,
"step": 840
},
{
"entropy": 7.255322885513306,
"epoch": 0.07462027552101731,
"grad_norm": 1.078125,
"learning_rate": 0.000422,
"loss": 7.1315,
"mean_token_accuracy": 0.1110302060842514,
"num_tokens": 1557035.0,
"step": 845
},
{
"entropy": 7.1888104438781735,
"epoch": 0.07506181561285764,
"grad_norm": 0.90234375,
"learning_rate": 0.0004245,
"loss": 7.0906,
"mean_token_accuracy": 0.11411306262016296,
"num_tokens": 1566773.0,
"step": 850
},
{
"entropy": 7.094766998291016,
"epoch": 0.07550335570469799,
"grad_norm": 1.2421875,
"learning_rate": 0.000427,
"loss": 7.0787,
"mean_token_accuracy": 0.10884842053055763,
"num_tokens": 1576873.0,
"step": 855
},
{
"entropy": 7.178222894668579,
"epoch": 0.07594489579653832,
"grad_norm": 1.03125,
"learning_rate": 0.0004295,
"loss": 7.111,
"mean_token_accuracy": 0.10762306824326515,
"num_tokens": 1586170.0,
"step": 860
},
{
"entropy": 7.286298131942749,
"epoch": 0.07638643588837866,
"grad_norm": 1.171875,
"learning_rate": 0.000432,
"loss": 7.154,
"mean_token_accuracy": 0.10613771453499794,
"num_tokens": 1596054.0,
"step": 865
},
{
"entropy": 7.1001307487487795,
"epoch": 0.07682797598021901,
"grad_norm": 1.0546875,
"learning_rate": 0.0004345,
"loss": 7.0262,
"mean_token_accuracy": 0.11607334911823272,
"num_tokens": 1604544.0,
"step": 870
},
{
"entropy": 7.172781848907471,
"epoch": 0.07726951607205934,
"grad_norm": 1.1171875,
"learning_rate": 0.000437,
"loss": 7.0446,
"mean_token_accuracy": 0.11472792029380799,
"num_tokens": 1614580.0,
"step": 875
},
{
"entropy": 7.132223224639892,
"epoch": 0.07771105616389969,
"grad_norm": 1.046875,
"learning_rate": 0.0004395,
"loss": 7.1232,
"mean_token_accuracy": 0.1109985999763012,
"num_tokens": 1624701.0,
"step": 880
},
{
"entropy": 7.128903436660766,
"epoch": 0.07815259625574002,
"grad_norm": 1.015625,
"learning_rate": 0.000442,
"loss": 7.0573,
"mean_token_accuracy": 0.10825628340244293,
"num_tokens": 1634085.0,
"step": 885
},
{
"entropy": 7.123282432556152,
"epoch": 0.07859413634758036,
"grad_norm": 1.0859375,
"learning_rate": 0.0004445,
"loss": 7.0322,
"mean_token_accuracy": 0.11617021560668946,
"num_tokens": 1643190.0,
"step": 890
},
{
"entropy": 7.060208940505982,
"epoch": 0.0790356764394207,
"grad_norm": 1.1796875,
"learning_rate": 0.000447,
"loss": 7.07,
"mean_token_accuracy": 0.11254222765564918,
"num_tokens": 1652705.0,
"step": 895
},
{
"entropy": 7.132848882675171,
"epoch": 0.07947721653126104,
"grad_norm": 1.046875,
"learning_rate": 0.00044950000000000003,
"loss": 7.0536,
"mean_token_accuracy": 0.10692465007305145,
"num_tokens": 1662210.0,
"step": 900
},
{
"entropy": 7.192712593078613,
"epoch": 0.07991875662310138,
"grad_norm": 0.984375,
"learning_rate": 0.00045200000000000004,
"loss": 7.1014,
"mean_token_accuracy": 0.10652303621172905,
"num_tokens": 1671893.0,
"step": 905
},
{
"entropy": 7.059550428390503,
"epoch": 0.08036029671494171,
"grad_norm": 1.1015625,
"learning_rate": 0.00045450000000000004,
"loss": 7.0402,
"mean_token_accuracy": 0.11181816533207893,
"num_tokens": 1681217.0,
"step": 910
},
{
"entropy": 7.149940156936646,
"epoch": 0.08080183680678206,
"grad_norm": 1.0390625,
"learning_rate": 0.00045700000000000005,
"loss": 7.053,
"mean_token_accuracy": 0.11131602600216865,
"num_tokens": 1690447.0,
"step": 915
},
{
"entropy": 7.081046295166016,
"epoch": 0.08124337689862239,
"grad_norm": 1.0,
"learning_rate": 0.00045950000000000006,
"loss": 7.1332,
"mean_token_accuracy": 0.10568991601467133,
"num_tokens": 1700355.0,
"step": 920
},
{
"entropy": 7.16390905380249,
"epoch": 0.08168491699046274,
"grad_norm": 1.015625,
"learning_rate": 0.000462,
"loss": 7.028,
"mean_token_accuracy": 0.10254786685109138,
"num_tokens": 1709449.0,
"step": 925
},
{
"entropy": 7.091014242172241,
"epoch": 0.08212645708230307,
"grad_norm": 1.1171875,
"learning_rate": 0.0004645,
"loss": 7.058,
"mean_token_accuracy": 0.10658924430608749,
"num_tokens": 1718838.0,
"step": 930
},
{
"entropy": 7.023260927200317,
"epoch": 0.08256799717414341,
"grad_norm": 1.046875,
"learning_rate": 0.000467,
"loss": 7.0217,
"mean_token_accuracy": 0.11323517858982086,
"num_tokens": 1728594.0,
"step": 935
},
{
"entropy": 7.1843287467956545,
"epoch": 0.08300953726598376,
"grad_norm": 0.984375,
"learning_rate": 0.0004695,
"loss": 7.0731,
"mean_token_accuracy": 0.11138227805495263,
"num_tokens": 1738814.0,
"step": 940
},
{
"entropy": 7.071042823791504,
"epoch": 0.08345107735782409,
"grad_norm": 1.1171875,
"learning_rate": 0.000472,
"loss": 7.0089,
"mean_token_accuracy": 0.11532488241791725,
"num_tokens": 1747644.0,
"step": 945
},
{
"entropy": 7.104792213439941,
"epoch": 0.08389261744966443,
"grad_norm": 1.21875,
"learning_rate": 0.0004745,
"loss": 7.0338,
"mean_token_accuracy": 0.11352440416812896,
"num_tokens": 1757489.0,
"step": 950
},
{
"entropy": 6.995518827438355,
"epoch": 0.08433415754150476,
"grad_norm": 1.0390625,
"learning_rate": 0.000477,
"loss": 6.9488,
"mean_token_accuracy": 0.11878458335995674,
"num_tokens": 1767546.0,
"step": 955
},
{
"entropy": 7.094525289535523,
"epoch": 0.08477569763334511,
"grad_norm": 1.3984375,
"learning_rate": 0.0004795,
"loss": 7.0088,
"mean_token_accuracy": 0.10509251430630684,
"num_tokens": 1776035.0,
"step": 960
},
{
"entropy": 7.100050449371338,
"epoch": 0.08521723772518544,
"grad_norm": 0.8984375,
"learning_rate": 0.000482,
"loss": 7.0869,
"mean_token_accuracy": 0.10708501487970352,
"num_tokens": 1786161.0,
"step": 965
},
{
"entropy": 7.161181020736694,
"epoch": 0.08565877781702579,
"grad_norm": 1.0625,
"learning_rate": 0.0004845,
"loss": 7.1355,
"mean_token_accuracy": 0.10680384710431098,
"num_tokens": 1796093.0,
"step": 970
},
{
"entropy": 7.064108896255493,
"epoch": 0.08610031790886613,
"grad_norm": 1.0390625,
"learning_rate": 0.000487,
"loss": 6.9829,
"mean_token_accuracy": 0.1097193941473961,
"num_tokens": 1805574.0,
"step": 975
},
{
"entropy": 7.07778491973877,
"epoch": 0.08654185800070646,
"grad_norm": 1.0703125,
"learning_rate": 0.0004895,
"loss": 7.0171,
"mean_token_accuracy": 0.11008013710379601,
"num_tokens": 1815175.0,
"step": 980
},
{
"entropy": 7.017868852615356,
"epoch": 0.08698339809254681,
"grad_norm": 1.03125,
"learning_rate": 0.000492,
"loss": 6.9322,
"mean_token_accuracy": 0.11862852200865745,
"num_tokens": 1824683.0,
"step": 985
},
{
"entropy": 7.004701948165893,
"epoch": 0.08742493818438714,
"grad_norm": 1.0703125,
"learning_rate": 0.0004945,
"loss": 6.9091,
"mean_token_accuracy": 0.1145630083978176,
"num_tokens": 1833174.0,
"step": 990
},
{
"entropy": 7.008507776260376,
"epoch": 0.08786647827622748,
"grad_norm": 1.203125,
"learning_rate": 0.000497,
"loss": 7.0124,
"mean_token_accuracy": 0.1165225401520729,
"num_tokens": 1842409.0,
"step": 995
},
{
"entropy": 6.900066137313843,
"epoch": 0.08830801836806781,
"grad_norm": 0.99609375,
"learning_rate": 0.0004995,
"loss": 6.9172,
"mean_token_accuracy": 0.1189465768635273,
"num_tokens": 1851441.0,
"step": 1000
},
{
"entropy": 7.234589004516602,
"epoch": 0.08874955845990816,
"grad_norm": 1.0703125,
"learning_rate": 0.000499999998589561,
"loss": 7.0545,
"mean_token_accuracy": 0.1098681665956974,
"num_tokens": 1861188.0,
"step": 1005
},
{
"entropy": 6.916832828521729,
"epoch": 0.0891910985517485,
"grad_norm": 1.1875,
"learning_rate": 0.0004999999928596523,
"loss": 6.9934,
"mean_token_accuracy": 0.1134356640279293,
"num_tokens": 1870284.0,
"step": 1010
},
{
"entropy": 6.8979510307312015,
"epoch": 0.08963263864358884,
"grad_norm": 1.046875,
"learning_rate": 0.0004999999827221219,
"loss": 6.9508,
"mean_token_accuracy": 0.11564922854304313,
"num_tokens": 1879744.0,
"step": 1015
},
{
"entropy": 7.141992807388306,
"epoch": 0.09007417873542918,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999999681769696,
"loss": 6.9612,
"mean_token_accuracy": 0.1163177601993084,
"num_tokens": 1889241.0,
"step": 1020
},
{
"entropy": 6.971645736694336,
"epoch": 0.09051571882726951,
"grad_norm": 1.09375,
"learning_rate": 0.000499999949224196,
"loss": 6.969,
"mean_token_accuracy": 0.11504201143980027,
"num_tokens": 1898247.0,
"step": 1025
},
{
"entropy": 7.006195020675659,
"epoch": 0.09095725891910986,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999999258638013,
"loss": 6.9244,
"mean_token_accuracy": 0.11498644798994065,
"num_tokens": 1907559.0,
"step": 1030
},
{
"entropy": 6.903467321395874,
"epoch": 0.09139879901095019,
"grad_norm": 1.09375,
"learning_rate": 0.0004999998980957861,
"loss": 6.924,
"mean_token_accuracy": 0.11912157312035561,
"num_tokens": 1917072.0,
"step": 1035
},
{
"entropy": 6.983310127258301,
"epoch": 0.09184033910279053,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999998659201508,
"loss": 6.8753,
"mean_token_accuracy": 0.11874028518795968,
"num_tokens": 1926597.0,
"step": 1040
},
{
"entropy": 7.006656980514526,
"epoch": 0.09228187919463088,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999998293368961,
"loss": 6.918,
"mean_token_accuracy": 0.11443031057715417,
"num_tokens": 1935978.0,
"step": 1045
},
{
"entropy": 6.924140882492066,
"epoch": 0.09272341928647121,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999997883460227,
"loss": 6.864,
"mean_token_accuracy": 0.1171707384288311,
"num_tokens": 1944424.0,
"step": 1050
},
{
"entropy": 6.984135913848877,
"epoch": 0.09316495937831155,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999997429475314,
"loss": 6.8384,
"mean_token_accuracy": 0.12138021439313888,
"num_tokens": 1953844.0,
"step": 1055
},
{
"entropy": 6.947112941741944,
"epoch": 0.09360649947015189,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999996931414232,
"loss": 6.9207,
"mean_token_accuracy": 0.12120825350284577,
"num_tokens": 1963974.0,
"step": 1060
},
{
"entropy": 6.977735805511474,
"epoch": 0.09404803956199223,
"grad_norm": 1.03125,
"learning_rate": 0.0004999996389276988,
"loss": 6.8969,
"mean_token_accuracy": 0.12291403263807296,
"num_tokens": 1973466.0,
"step": 1065
},
{
"entropy": 6.803595685958863,
"epoch": 0.09448957965383256,
"grad_norm": 1.09375,
"learning_rate": 0.0004999995803063596,
"loss": 6.918,
"mean_token_accuracy": 0.12123456448316575,
"num_tokens": 1983478.0,
"step": 1070
},
{
"entropy": 7.003172111511231,
"epoch": 0.0949311197456729,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999995172774065,
"loss": 6.9879,
"mean_token_accuracy": 0.11715293675661087,
"num_tokens": 1992775.0,
"step": 1075
},
{
"entropy": 7.030760860443115,
"epoch": 0.09537265983751325,
"grad_norm": 1.0625,
"learning_rate": 0.0004999994498408408,
"loss": 6.9419,
"mean_token_accuracy": 0.11398354098200798,
"num_tokens": 2002526.0,
"step": 1080
},
{
"entropy": 6.936420059204101,
"epoch": 0.09581419992935358,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999993779966639,
"loss": 6.9592,
"mean_token_accuracy": 0.11412434950470925,
"num_tokens": 2012476.0,
"step": 1085
},
{
"entropy": 6.838730955123902,
"epoch": 0.09625574002119393,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999993017448771,
"loss": 6.7924,
"mean_token_accuracy": 0.13063410446047782,
"num_tokens": 2021252.0,
"step": 1090
},
{
"entropy": 7.061969757080078,
"epoch": 0.09669728011303426,
"grad_norm": 1.0,
"learning_rate": 0.0004999992210854821,
"loss": 6.9411,
"mean_token_accuracy": 0.11837697625160218,
"num_tokens": 2031438.0,
"step": 1095
},
{
"entropy": 6.9324125289917,
"epoch": 0.0971388202048746,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999991360184801,
"loss": 6.9789,
"mean_token_accuracy": 0.11443927884101868,
"num_tokens": 2041319.0,
"step": 1100
},
{
"entropy": 6.95938081741333,
"epoch": 0.09758036029671494,
"grad_norm": 0.859375,
"learning_rate": 0.0004999990465438731,
"loss": 6.9746,
"mean_token_accuracy": 0.11487890034914017,
"num_tokens": 2052060.0,
"step": 1105
},
{
"entropy": 6.891486358642578,
"epoch": 0.09802190038855528,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999989526616628,
"loss": 6.8643,
"mean_token_accuracy": 0.12554761841893197,
"num_tokens": 2061331.0,
"step": 1110
},
{
"entropy": 6.952145195007324,
"epoch": 0.09846344048039563,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999988543718509,
"loss": 6.8733,
"mean_token_accuracy": 0.11660940647125244,
"num_tokens": 2070006.0,
"step": 1115
},
{
"entropy": 7.005694484710693,
"epoch": 0.09890498057223596,
"grad_norm": 1.171875,
"learning_rate": 0.0004999987516744394,
"loss": 6.8403,
"mean_token_accuracy": 0.12942354679107665,
"num_tokens": 2079089.0,
"step": 1120
},
{
"entropy": 6.814391231536865,
"epoch": 0.0993465206640763,
"grad_norm": 1.109375,
"learning_rate": 0.0004999986445694303,
"loss": 6.8164,
"mean_token_accuracy": 0.12300374433398246,
"num_tokens": 2087237.0,
"step": 1125
},
{
"entropy": 6.877581930160522,
"epoch": 0.09978806075591663,
"grad_norm": 1.109375,
"learning_rate": 0.0004999985330568258,
"loss": 6.7838,
"mean_token_accuracy": 0.12274843603372573,
"num_tokens": 2096598.0,
"step": 1130
},
{
"entropy": 6.867826128005982,
"epoch": 0.10022960084775698,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999984171366278,
"loss": 6.8802,
"mean_token_accuracy": 0.11098882853984833,
"num_tokens": 2106106.0,
"step": 1135
},
{
"entropy": 6.928639554977417,
"epoch": 0.10067114093959731,
"grad_norm": 0.98828125,
"learning_rate": 0.000499998296808839,
"loss": 6.856,
"mean_token_accuracy": 0.11591004803776742,
"num_tokens": 2115478.0,
"step": 1140
},
{
"entropy": 6.942514848709107,
"epoch": 0.10111268103143765,
"grad_norm": 1.25,
"learning_rate": 0.0004999981720734615,
"loss": 6.9007,
"mean_token_accuracy": 0.11097749546170235,
"num_tokens": 2124439.0,
"step": 1145
},
{
"entropy": 6.928429985046387,
"epoch": 0.101554221123278,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999980429304977,
"loss": 6.8983,
"mean_token_accuracy": 0.11434343308210373,
"num_tokens": 2133718.0,
"step": 1150
},
{
"entropy": 6.835229969024658,
"epoch": 0.10199576121511833,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999979093799502,
"loss": 6.801,
"mean_token_accuracy": 0.12133207321166992,
"num_tokens": 2142978.0,
"step": 1155
},
{
"entropy": 6.8165655612945555,
"epoch": 0.10243730130695868,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999977714218217,
"loss": 6.863,
"mean_token_accuracy": 0.117961073666811,
"num_tokens": 2152250.0,
"step": 1160
},
{
"entropy": 6.972959232330322,
"epoch": 0.102878841398799,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999976290561147,
"loss": 6.8608,
"mean_token_accuracy": 0.11624824330210685,
"num_tokens": 2161620.0,
"step": 1165
},
{
"entropy": 6.921932792663574,
"epoch": 0.10332038149063935,
"grad_norm": 0.9375,
"learning_rate": 0.0004999974822828322,
"loss": 6.8726,
"mean_token_accuracy": 0.1173239678144455,
"num_tokens": 2170856.0,
"step": 1170
},
{
"entropy": 6.885772609710694,
"epoch": 0.10376192158247968,
"grad_norm": 0.98828125,
"learning_rate": 0.000499997331101977,
"loss": 6.8626,
"mean_token_accuracy": 0.11568826138973236,
"num_tokens": 2180926.0,
"step": 1175
},
{
"entropy": 6.911135244369507,
"epoch": 0.10420346167432003,
"grad_norm": 1.1171875,
"learning_rate": 0.000499997175513552,
"loss": 6.8419,
"mean_token_accuracy": 0.11412648186087608,
"num_tokens": 2190248.0,
"step": 1180
},
{
"entropy": 6.86896710395813,
"epoch": 0.10464500176616037,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999970155175603,
"loss": 6.855,
"mean_token_accuracy": 0.12222710996866226,
"num_tokens": 2199833.0,
"step": 1185
},
{
"entropy": 6.856761360168457,
"epoch": 0.1050865418580007,
"grad_norm": 0.98828125,
"learning_rate": 0.000499996851114005,
"loss": 6.8115,
"mean_token_accuracy": 0.12742498219013215,
"num_tokens": 2208240.0,
"step": 1190
},
{
"entropy": 6.837878751754761,
"epoch": 0.10552808194984105,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999966823028894,
"loss": 6.8776,
"mean_token_accuracy": 0.11117666661739349,
"num_tokens": 2218758.0,
"step": 1195
},
{
"entropy": 6.94004077911377,
"epoch": 0.10596962204168138,
"grad_norm": 0.984375,
"learning_rate": 0.0004999965090842168,
"loss": 6.8665,
"mean_token_accuracy": 0.12320348769426345,
"num_tokens": 2228443.0,
"step": 1200
},
{
"entropy": 6.74250750541687,
"epoch": 0.10641116213352173,
"grad_norm": 1.0625,
"learning_rate": 0.0004999963314579905,
"loss": 6.7084,
"mean_token_accuracy": 0.1319122113287449,
"num_tokens": 2236787.0,
"step": 1205
},
{
"entropy": 6.956686353683471,
"epoch": 0.10685270222536206,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999961494242139,
"loss": 6.8901,
"mean_token_accuracy": 0.11468368023633957,
"num_tokens": 2247089.0,
"step": 1210
},
{
"entropy": 6.789958381652832,
"epoch": 0.1072942423172024,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999959629828908,
"loss": 6.816,
"mean_token_accuracy": 0.11569953635334969,
"num_tokens": 2256045.0,
"step": 1215
},
{
"entropy": 6.929527616500854,
"epoch": 0.10773578240904275,
"grad_norm": 0.96875,
"learning_rate": 0.0004999957721340248,
"loss": 6.8769,
"mean_token_accuracy": 0.12202595993876457,
"num_tokens": 2265835.0,
"step": 1220
},
{
"entropy": 6.807823610305786,
"epoch": 0.10817732250088308,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999955768776195,
"loss": 6.8076,
"mean_token_accuracy": 0.11702087000012398,
"num_tokens": 2275318.0,
"step": 1225
},
{
"entropy": 6.869472360610962,
"epoch": 0.10861886259272342,
"grad_norm": 1.09375,
"learning_rate": 0.0004999953772136788,
"loss": 6.7978,
"mean_token_accuracy": 0.12102394551038742,
"num_tokens": 2284821.0,
"step": 1230
},
{
"entropy": 6.885239315032959,
"epoch": 0.10906040268456375,
"grad_norm": 1.015625,
"learning_rate": 0.0004999951731422068,
"loss": 6.7645,
"mean_token_accuracy": 0.12054353281855583,
"num_tokens": 2294013.0,
"step": 1235
},
{
"entropy": 6.847450494766235,
"epoch": 0.1095019427764041,
"grad_norm": 1.125,
"learning_rate": 0.0004999949646632072,
"loss": 6.7626,
"mean_token_accuracy": 0.12177001982927323,
"num_tokens": 2302727.0,
"step": 1240
},
{
"entropy": 6.8077342987060545,
"epoch": 0.10994348286824443,
"grad_norm": 1.078125,
"learning_rate": 0.0004999947517766842,
"loss": 6.8031,
"mean_token_accuracy": 0.12163913846015931,
"num_tokens": 2312032.0,
"step": 1245
},
{
"entropy": 6.986685180664063,
"epoch": 0.11038502296008477,
"grad_norm": 1.1640625,
"learning_rate": 0.000499994534482642,
"loss": 6.8748,
"mean_token_accuracy": 0.1190544456243515,
"num_tokens": 2321839.0,
"step": 1250
},
{
"entropy": 6.811014032363891,
"epoch": 0.11082656305192512,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999943127810847,
"loss": 6.8536,
"mean_token_accuracy": 0.1122577242553234,
"num_tokens": 2331255.0,
"step": 1255
},
{
"entropy": 6.794656276702881,
"epoch": 0.11126810314376545,
"grad_norm": 1.015625,
"learning_rate": 0.0004999940866720169,
"loss": 6.6705,
"mean_token_accuracy": 0.12881582453846932,
"num_tokens": 2340038.0,
"step": 1260
},
{
"entropy": 6.71243953704834,
"epoch": 0.1117096432356058,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999938561554429,
"loss": 6.7797,
"mean_token_accuracy": 0.12242485880851746,
"num_tokens": 2348901.0,
"step": 1265
},
{
"entropy": 6.858448314666748,
"epoch": 0.11215118332744613,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999936212313672,
"loss": 6.8659,
"mean_token_accuracy": 0.11461173072457313,
"num_tokens": 2358842.0,
"step": 1270
},
{
"entropy": 6.8239977836608885,
"epoch": 0.11259272341928647,
"grad_norm": 1.0625,
"learning_rate": 0.0004999933818997943,
"loss": 6.7596,
"mean_token_accuracy": 0.12424605414271354,
"num_tokens": 2368650.0,
"step": 1275
},
{
"entropy": 6.831825399398804,
"epoch": 0.1130342635111268,
"grad_norm": 0.96875,
"learning_rate": 0.0004999931381607292,
"loss": 6.8252,
"mean_token_accuracy": 0.12058763056993485,
"num_tokens": 2377916.0,
"step": 1280
},
{
"entropy": 6.818245553970337,
"epoch": 0.11347580360296715,
"grad_norm": 1.09375,
"learning_rate": 0.0004999928900141764,
"loss": 6.7698,
"mean_token_accuracy": 0.12198482304811478,
"num_tokens": 2387507.0,
"step": 1285
},
{
"entropy": 6.819052505493164,
"epoch": 0.1139173436948075,
"grad_norm": 1.109375,
"learning_rate": 0.000499992637460141,
"loss": 6.8052,
"mean_token_accuracy": 0.12523134648799897,
"num_tokens": 2396148.0,
"step": 1290
},
{
"entropy": 6.816552209854126,
"epoch": 0.11435888378664782,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999923804986275,
"loss": 6.693,
"mean_token_accuracy": 0.11803872361779214,
"num_tokens": 2404891.0,
"step": 1295
},
{
"entropy": 6.840794086456299,
"epoch": 0.11480042387848817,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999921191296415,
"loss": 6.7153,
"mean_token_accuracy": 0.12199744880199433,
"num_tokens": 2414406.0,
"step": 1300
},
{
"entropy": 6.7452630519866945,
"epoch": 0.1152419639703285,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999918533531877,
"loss": 6.8046,
"mean_token_accuracy": 0.1228412576019764,
"num_tokens": 2424363.0,
"step": 1305
},
{
"entropy": 6.852883148193359,
"epoch": 0.11568350406216885,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999915831692714,
"loss": 6.7419,
"mean_token_accuracy": 0.1251549780368805,
"num_tokens": 2433753.0,
"step": 1310
},
{
"entropy": 6.7218766689300535,
"epoch": 0.11612504415400918,
"grad_norm": 1.0,
"learning_rate": 0.0004999913085778981,
"loss": 6.7685,
"mean_token_accuracy": 0.1185051940381527,
"num_tokens": 2443275.0,
"step": 1315
},
{
"entropy": 6.873374080657959,
"epoch": 0.11656658424584952,
"grad_norm": 1.0625,
"learning_rate": 0.0004999910295790729,
"loss": 6.7937,
"mean_token_accuracy": 0.11835979968309403,
"num_tokens": 2452510.0,
"step": 1320
},
{
"entropy": 6.8684648990631105,
"epoch": 0.11700812433768987,
"grad_norm": 1.078125,
"learning_rate": 0.0004999907461728014,
"loss": 6.8746,
"mean_token_accuracy": 0.1169828750193119,
"num_tokens": 2462742.0,
"step": 1325
},
{
"entropy": 6.740426445007325,
"epoch": 0.1174496644295302,
"grad_norm": 0.96875,
"learning_rate": 0.0004999904583590893,
"loss": 6.7434,
"mean_token_accuracy": 0.12029099017381668,
"num_tokens": 2471409.0,
"step": 1330
},
{
"entropy": 6.839800691604614,
"epoch": 0.11789120452137054,
"grad_norm": 0.984375,
"learning_rate": 0.0004999901661379418,
"loss": 6.6931,
"mean_token_accuracy": 0.12773663252592088,
"num_tokens": 2481011.0,
"step": 1335
},
{
"entropy": 6.690527105331421,
"epoch": 0.11833274461321087,
"grad_norm": 0.953125,
"learning_rate": 0.0004999898695093652,
"loss": 6.7866,
"mean_token_accuracy": 0.12104339599609375,
"num_tokens": 2490664.0,
"step": 1340
},
{
"entropy": 6.818962049484253,
"epoch": 0.11877428470505122,
"grad_norm": 1.046875,
"learning_rate": 0.0004999895684733648,
"loss": 6.7279,
"mean_token_accuracy": 0.12824407517910003,
"num_tokens": 2499799.0,
"step": 1345
},
{
"entropy": 6.8539299964904785,
"epoch": 0.11921582479689156,
"grad_norm": 1.53125,
"learning_rate": 0.0004999892630299467,
"loss": 6.7045,
"mean_token_accuracy": 0.1257259279489517,
"num_tokens": 2508780.0,
"step": 1350
},
{
"entropy": 6.706318616867065,
"epoch": 0.1196573648887319,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999889531791171,
"loss": 6.7138,
"mean_token_accuracy": 0.12127138078212737,
"num_tokens": 2517741.0,
"step": 1355
},
{
"entropy": 6.766215896606445,
"epoch": 0.12009890498057224,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999886389208817,
"loss": 6.7972,
"mean_token_accuracy": 0.11826895922422409,
"num_tokens": 2528742.0,
"step": 1360
},
{
"entropy": 6.717579746246338,
"epoch": 0.12054044507241257,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999883202552468,
"loss": 6.7345,
"mean_token_accuracy": 0.12455343306064606,
"num_tokens": 2538609.0,
"step": 1365
},
{
"entropy": 6.882754898071289,
"epoch": 0.12098198516425292,
"grad_norm": 1.046875,
"learning_rate": 0.0004999879971822189,
"loss": 6.7157,
"mean_token_accuracy": 0.11966117843985558,
"num_tokens": 2547772.0,
"step": 1370
},
{
"entropy": 6.69037971496582,
"epoch": 0.12142352525609325,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999876697018038,
"loss": 6.6897,
"mean_token_accuracy": 0.12769502475857736,
"num_tokens": 2556114.0,
"step": 1375
},
{
"entropy": 6.8372406482696535,
"epoch": 0.1218650653479336,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999873378140085,
"loss": 6.7182,
"mean_token_accuracy": 0.12253274098038673,
"num_tokens": 2566814.0,
"step": 1380
},
{
"entropy": 6.714360284805298,
"epoch": 0.12230660543977394,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999870015188389,
"loss": 6.6914,
"mean_token_accuracy": 0.12260655164718628,
"num_tokens": 2576030.0,
"step": 1385
},
{
"entropy": 6.743939208984375,
"epoch": 0.12274814553161427,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999866608163021,
"loss": 6.7176,
"mean_token_accuracy": 0.1260794699192047,
"num_tokens": 2585756.0,
"step": 1390
},
{
"entropy": 6.800289154052734,
"epoch": 0.12318968562345461,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999863157064045,
"loss": 6.7797,
"mean_token_accuracy": 0.1238692507147789,
"num_tokens": 2595676.0,
"step": 1395
},
{
"entropy": 6.8109955310821535,
"epoch": 0.12363122571529495,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999859661891529,
"loss": 6.7624,
"mean_token_accuracy": 0.1246532566845417,
"num_tokens": 2606197.0,
"step": 1400
},
{
"entropy": 6.76338677406311,
"epoch": 0.12407276580713529,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999856122645543,
"loss": 6.6854,
"mean_token_accuracy": 0.12518818601965903,
"num_tokens": 2615311.0,
"step": 1405
},
{
"entropy": 6.725164937973022,
"epoch": 0.12451430589897562,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999852539326154,
"loss": 6.5931,
"mean_token_accuracy": 0.12619537115097046,
"num_tokens": 2624074.0,
"step": 1410
},
{
"entropy": 6.681712102890015,
"epoch": 0.12495584599081597,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999848911933434,
"loss": 6.7411,
"mean_token_accuracy": 0.13295672461390495,
"num_tokens": 2633877.0,
"step": 1415
},
{
"entropy": 6.7498420715332035,
"epoch": 0.1253973860826563,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999845240467453,
"loss": 6.6702,
"mean_token_accuracy": 0.12471742331981658,
"num_tokens": 2643330.0,
"step": 1420
},
{
"entropy": 6.74535722732544,
"epoch": 0.12583892617449666,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999841524928282,
"loss": 6.6543,
"mean_token_accuracy": 0.12766205966472627,
"num_tokens": 2652070.0,
"step": 1425
},
{
"entropy": 6.722480249404907,
"epoch": 0.126280466266337,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999837765315997,
"loss": 6.697,
"mean_token_accuracy": 0.1277957484126091,
"num_tokens": 2660546.0,
"step": 1430
},
{
"entropy": 6.598982477188111,
"epoch": 0.12672200635817732,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999833961630669,
"loss": 6.5999,
"mean_token_accuracy": 0.1297621488571167,
"num_tokens": 2669938.0,
"step": 1435
},
{
"entropy": 6.852873420715332,
"epoch": 0.12716354645001765,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999830113872374,
"loss": 6.7248,
"mean_token_accuracy": 0.12213384285569191,
"num_tokens": 2679814.0,
"step": 1440
},
{
"entropy": 6.585267829895019,
"epoch": 0.127605086541858,
"grad_norm": 1.046875,
"learning_rate": 0.0004999826222041186,
"loss": 6.6355,
"mean_token_accuracy": 0.13383011817932128,
"num_tokens": 2688733.0,
"step": 1445
},
{
"entropy": 6.706393527984619,
"epoch": 0.12804662663369834,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999822286137182,
"loss": 6.6188,
"mean_token_accuracy": 0.12774292454123498,
"num_tokens": 2697744.0,
"step": 1450
},
{
"entropy": 6.633390617370606,
"epoch": 0.12848816672553867,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999818306160439,
"loss": 6.5827,
"mean_token_accuracy": 0.13278514444828032,
"num_tokens": 2707037.0,
"step": 1455
},
{
"entropy": 6.808015394210815,
"epoch": 0.12892970681737903,
"grad_norm": 0.90625,
"learning_rate": 0.0004999814282111034,
"loss": 6.7453,
"mean_token_accuracy": 0.12486135885119438,
"num_tokens": 2717345.0,
"step": 1460
},
{
"entropy": 6.7304778575897215,
"epoch": 0.12937124690921936,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999810213989047,
"loss": 6.7017,
"mean_token_accuracy": 0.12036777138710023,
"num_tokens": 2726892.0,
"step": 1465
},
{
"entropy": 6.667453098297119,
"epoch": 0.1298127870010597,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999806101794558,
"loss": 6.6615,
"mean_token_accuracy": 0.12705308422446251,
"num_tokens": 2736479.0,
"step": 1470
},
{
"entropy": 6.809973049163818,
"epoch": 0.13025432709290002,
"grad_norm": 1.09375,
"learning_rate": 0.0004999801945527648,
"loss": 6.7078,
"mean_token_accuracy": 0.12117967531085014,
"num_tokens": 2745998.0,
"step": 1475
},
{
"entropy": 6.725075721740723,
"epoch": 0.13069586718474038,
"grad_norm": 1.03125,
"learning_rate": 0.0004999797745188395,
"loss": 6.6906,
"mean_token_accuracy": 0.12821464985609055,
"num_tokens": 2754346.0,
"step": 1480
},
{
"entropy": 6.686868619918823,
"epoch": 0.13113740727658071,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999793500776886,
"loss": 6.6285,
"mean_token_accuracy": 0.1294437274336815,
"num_tokens": 2763391.0,
"step": 1485
},
{
"entropy": 6.674964761734008,
"epoch": 0.13157894736842105,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999789212293201,
"loss": 6.6827,
"mean_token_accuracy": 0.12898893728852273,
"num_tokens": 2772764.0,
"step": 1490
},
{
"entropy": 6.775622892379761,
"epoch": 0.1320204874602614,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999784879737423,
"loss": 6.78,
"mean_token_accuracy": 0.12160285860300064,
"num_tokens": 2782312.0,
"step": 1495
},
{
"entropy": 6.745266675949097,
"epoch": 0.13246202755210174,
"grad_norm": 1.0625,
"learning_rate": 0.0004999780503109642,
"loss": 6.6798,
"mean_token_accuracy": 0.1227384127676487,
"num_tokens": 2791159.0,
"step": 1500
},
{
"entropy": 6.650141859054566,
"epoch": 0.13290356764394207,
"grad_norm": 1.046875,
"learning_rate": 0.0004999776082409939,
"loss": 6.5068,
"mean_token_accuracy": 0.13456878885626794,
"num_tokens": 2799319.0,
"step": 1505
},
{
"entropy": 6.6320771217346195,
"epoch": 0.1333451077357824,
"grad_norm": 1.078125,
"learning_rate": 0.0004999771617638401,
"loss": 6.6316,
"mean_token_accuracy": 0.12712259590625763,
"num_tokens": 2807401.0,
"step": 1510
},
{
"entropy": 6.758873414993286,
"epoch": 0.13378664782762276,
"grad_norm": 1.078125,
"learning_rate": 0.0004999767108795118,
"loss": 6.6961,
"mean_token_accuracy": 0.12330949455499648,
"num_tokens": 2817734.0,
"step": 1515
},
{
"entropy": 6.644004774093628,
"epoch": 0.1342281879194631,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999762555880176,
"loss": 6.6783,
"mean_token_accuracy": 0.12152940481901169,
"num_tokens": 2828235.0,
"step": 1520
},
{
"entropy": 6.743390846252441,
"epoch": 0.13466972801130342,
"grad_norm": 1.046875,
"learning_rate": 0.0004999757958893666,
"loss": 6.7124,
"mean_token_accuracy": 0.1237283930182457,
"num_tokens": 2837453.0,
"step": 1525
},
{
"entropy": 6.60786280632019,
"epoch": 0.13511126810314378,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999753317835677,
"loss": 6.6795,
"mean_token_accuracy": 0.12087962031364441,
"num_tokens": 2847055.0,
"step": 1530
},
{
"entropy": 6.7429241180419925,
"epoch": 0.1355528081949841,
"grad_norm": 1.0,
"learning_rate": 0.0004999748632706299,
"loss": 6.6568,
"mean_token_accuracy": 0.13167096227407454,
"num_tokens": 2857101.0,
"step": 1535
},
{
"entropy": 6.673015403747558,
"epoch": 0.13599434828682444,
"grad_norm": 1.140625,
"learning_rate": 0.0004999743903505626,
"loss": 6.553,
"mean_token_accuracy": 0.1336723633110523,
"num_tokens": 2866685.0,
"step": 1540
},
{
"entropy": 6.676457214355469,
"epoch": 0.13643588837866477,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999739130233749,
"loss": 6.6731,
"mean_token_accuracy": 0.12713466510176658,
"num_tokens": 2876022.0,
"step": 1545
},
{
"entropy": 6.783061361312866,
"epoch": 0.13687742847050513,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999734312890761,
"loss": 6.6062,
"mean_token_accuracy": 0.12626957073807715,
"num_tokens": 2885560.0,
"step": 1550
},
{
"entropy": 6.59228024482727,
"epoch": 0.13731896856234546,
"grad_norm": 0.9375,
"learning_rate": 0.0004999729451476757,
"loss": 6.6439,
"mean_token_accuracy": 0.12623701319098474,
"num_tokens": 2894686.0,
"step": 1555
},
{
"entropy": 6.730476140975952,
"epoch": 0.1377605086541858,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999724545991835,
"loss": 6.6588,
"mean_token_accuracy": 0.13341889455914496,
"num_tokens": 2904390.0,
"step": 1560
},
{
"entropy": 6.662082195281982,
"epoch": 0.13820204874602615,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999719596436086,
"loss": 6.6982,
"mean_token_accuracy": 0.12678939029574393,
"num_tokens": 2913311.0,
"step": 1565
},
{
"entropy": 6.736054372787476,
"epoch": 0.13864358883786648,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999714602809611,
"loss": 6.578,
"mean_token_accuracy": 0.13448369055986403,
"num_tokens": 2923196.0,
"step": 1570
},
{
"entropy": 6.637621927261352,
"epoch": 0.1390851289297068,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999709565112506,
"loss": 6.6428,
"mean_token_accuracy": 0.12785085365176202,
"num_tokens": 2932813.0,
"step": 1575
},
{
"entropy": 6.658017587661743,
"epoch": 0.13952666902154715,
"grad_norm": 0.95703125,
"learning_rate": 0.000499970448334487,
"loss": 6.5634,
"mean_token_accuracy": 0.119098000228405,
"num_tokens": 2942694.0,
"step": 1580
},
{
"entropy": 6.5801304340362545,
"epoch": 0.1399682091133875,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999699357506803,
"loss": 6.5185,
"mean_token_accuracy": 0.1330326519906521,
"num_tokens": 2950932.0,
"step": 1585
},
{
"entropy": 6.701116371154785,
"epoch": 0.14040974920522784,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999694187598406,
"loss": 6.6085,
"mean_token_accuracy": 0.1314128704369068,
"num_tokens": 2960102.0,
"step": 1590
},
{
"entropy": 6.627189350128174,
"epoch": 0.14085128929706817,
"grad_norm": 1.578125,
"learning_rate": 0.0004999688973619777,
"loss": 6.6088,
"mean_token_accuracy": 0.12604895159602164,
"num_tokens": 2969968.0,
"step": 1595
},
{
"entropy": 6.588514184951782,
"epoch": 0.14129282938890853,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999683715571022,
"loss": 6.5542,
"mean_token_accuracy": 0.1347218669950962,
"num_tokens": 2978880.0,
"step": 1600
},
{
"entropy": 6.644829893112183,
"epoch": 0.14173436948074886,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999678413452242,
"loss": 6.5863,
"mean_token_accuracy": 0.12890450209379195,
"num_tokens": 2988369.0,
"step": 1605
},
{
"entropy": 6.667511320114135,
"epoch": 0.1421759095725892,
"grad_norm": 0.953125,
"learning_rate": 0.0004999673067263542,
"loss": 6.6373,
"mean_token_accuracy": 0.12620072290301323,
"num_tokens": 2997070.0,
"step": 1610
},
{
"entropy": 6.682367372512817,
"epoch": 0.14261744966442952,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999667677005026,
"loss": 6.5749,
"mean_token_accuracy": 0.13310741856694222,
"num_tokens": 3006547.0,
"step": 1615
},
{
"entropy": 6.584122562408448,
"epoch": 0.14305898975626988,
"grad_norm": 0.953125,
"learning_rate": 0.0004999662242676799,
"loss": 6.5986,
"mean_token_accuracy": 0.1310425490140915,
"num_tokens": 3015821.0,
"step": 1620
},
{
"entropy": 6.665965223312378,
"epoch": 0.1435005298481102,
"grad_norm": 0.96875,
"learning_rate": 0.0004999656764278968,
"loss": 6.5655,
"mean_token_accuracy": 0.1309148021042347,
"num_tokens": 3024750.0,
"step": 1625
},
{
"entropy": 6.623423147201538,
"epoch": 0.14394206993995054,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999651241811642,
"loss": 6.5389,
"mean_token_accuracy": 0.1308048278093338,
"num_tokens": 3033345.0,
"step": 1630
},
{
"entropy": 6.59770941734314,
"epoch": 0.1443836100317909,
"grad_norm": 1.03125,
"learning_rate": 0.0004999645675274925,
"loss": 6.5209,
"mean_token_accuracy": 0.13443350791931152,
"num_tokens": 3042060.0,
"step": 1635
},
{
"entropy": 6.660798358917236,
"epoch": 0.14482515012363123,
"grad_norm": 1.0625,
"learning_rate": 0.0004999640064668931,
"loss": 6.6684,
"mean_token_accuracy": 0.12563745751976968,
"num_tokens": 3052490.0,
"step": 1640
},
{
"entropy": 6.726674222946167,
"epoch": 0.14526669021547156,
"grad_norm": 1.046875,
"learning_rate": 0.0004999634409993766,
"loss": 6.6441,
"mean_token_accuracy": 0.12772516757249833,
"num_tokens": 3061934.0,
"step": 1645
},
{
"entropy": 6.556687307357788,
"epoch": 0.1457082303073119,
"grad_norm": 1.046875,
"learning_rate": 0.0004999628711249544,
"loss": 6.5591,
"mean_token_accuracy": 0.13611432090401648,
"num_tokens": 3070890.0,
"step": 1650
},
{
"entropy": 6.703302097320557,
"epoch": 0.14614977039915225,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999622968436373,
"loss": 6.5614,
"mean_token_accuracy": 0.12631918862462044,
"num_tokens": 3079933.0,
"step": 1655
},
{
"entropy": 6.5565461158752445,
"epoch": 0.14659131049099258,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999617181554369,
"loss": 6.6078,
"mean_token_accuracy": 0.13254086449742317,
"num_tokens": 3089910.0,
"step": 1660
},
{
"entropy": 6.718549633026123,
"epoch": 0.1470328505828329,
"grad_norm": 1.1875,
"learning_rate": 0.0004999611350603643,
"loss": 6.5916,
"mean_token_accuracy": 0.13437702059745787,
"num_tokens": 3098676.0,
"step": 1665
},
{
"entropy": 6.595283174514771,
"epoch": 0.14747439067467327,
"grad_norm": 1.0078125,
"learning_rate": 0.000499960547558431,
"loss": 6.7159,
"mean_token_accuracy": 0.1256335400044918,
"num_tokens": 3109055.0,
"step": 1670
},
{
"entropy": 6.624332904815674,
"epoch": 0.1479159307665136,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999599556496486,
"loss": 6.5472,
"mean_token_accuracy": 0.13295547068119049,
"num_tokens": 3117517.0,
"step": 1675
},
{
"entropy": 6.573185825347901,
"epoch": 0.14835747085835393,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999593593340286,
"loss": 6.5544,
"mean_token_accuracy": 0.13249536529183387,
"num_tokens": 3126606.0,
"step": 1680
},
{
"entropy": 6.704092359542846,
"epoch": 0.14879901095019427,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999587586115826,
"loss": 6.5733,
"mean_token_accuracy": 0.12993996366858482,
"num_tokens": 3135444.0,
"step": 1685
},
{
"entropy": 6.57456955909729,
"epoch": 0.14924055104203462,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999581534823226,
"loss": 6.4927,
"mean_token_accuracy": 0.14061653688549997,
"num_tokens": 3144967.0,
"step": 1690
},
{
"entropy": 6.607442092895508,
"epoch": 0.14968209113387496,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999575439462601,
"loss": 6.522,
"mean_token_accuracy": 0.12690635845065118,
"num_tokens": 3153898.0,
"step": 1695
},
{
"entropy": 6.536995601654053,
"epoch": 0.1501236312257153,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999569300034075,
"loss": 6.5921,
"mean_token_accuracy": 0.12649724259972572,
"num_tokens": 3162341.0,
"step": 1700
},
{
"entropy": 6.60729718208313,
"epoch": 0.15056517131755565,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999563116537764,
"loss": 6.3826,
"mean_token_accuracy": 0.14100785404443741,
"num_tokens": 3171420.0,
"step": 1705
},
{
"entropy": 6.54091591835022,
"epoch": 0.15100671140939598,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999556888973792,
"loss": 6.4366,
"mean_token_accuracy": 0.13033056780695915,
"num_tokens": 3180221.0,
"step": 1710
},
{
"entropy": 6.6625391960144045,
"epoch": 0.1514482515012363,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999550617342279,
"loss": 6.587,
"mean_token_accuracy": 0.12798949852585792,
"num_tokens": 3189156.0,
"step": 1715
},
{
"entropy": 6.539492607116699,
"epoch": 0.15188979159307664,
"grad_norm": 1.0390625,
"learning_rate": 0.000499954430164335,
"loss": 6.5149,
"mean_token_accuracy": 0.13213447630405425,
"num_tokens": 3199870.0,
"step": 1720
},
{
"entropy": 6.559759998321534,
"epoch": 0.152331331684917,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999537941877127,
"loss": 6.4883,
"mean_token_accuracy": 0.13440720960497857,
"num_tokens": 3208815.0,
"step": 1725
},
{
"entropy": 6.604632616043091,
"epoch": 0.15277287177675733,
"grad_norm": 0.921875,
"learning_rate": 0.0004999531538043735,
"loss": 6.5717,
"mean_token_accuracy": 0.13279605731368066,
"num_tokens": 3218692.0,
"step": 1730
},
{
"entropy": 6.584544372558594,
"epoch": 0.15321441186859766,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999525090143298,
"loss": 6.5367,
"mean_token_accuracy": 0.14099612906575204,
"num_tokens": 3227604.0,
"step": 1735
},
{
"entropy": 6.629879665374756,
"epoch": 0.15365595196043802,
"grad_norm": 1.03125,
"learning_rate": 0.0004999518598175946,
"loss": 6.5446,
"mean_token_accuracy": 0.14111834466457368,
"num_tokens": 3237912.0,
"step": 1740
},
{
"entropy": 6.453984022140503,
"epoch": 0.15409749205227835,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999512062141805,
"loss": 6.4239,
"mean_token_accuracy": 0.13701695203781128,
"num_tokens": 3246003.0,
"step": 1745
},
{
"entropy": 6.6436468124389645,
"epoch": 0.15453903214411868,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999505482040999,
"loss": 6.5558,
"mean_token_accuracy": 0.13121686428785323,
"num_tokens": 3256363.0,
"step": 1750
},
{
"entropy": 6.583334445953369,
"epoch": 0.154980572235959,
"grad_norm": 1.046875,
"learning_rate": 0.0004999498857873662,
"loss": 6.5518,
"mean_token_accuracy": 0.13320463374257088,
"num_tokens": 3265822.0,
"step": 1755
},
{
"entropy": 6.478305387496948,
"epoch": 0.15542211232779937,
"grad_norm": 1.03125,
"learning_rate": 0.0004999492189639921,
"loss": 6.4224,
"mean_token_accuracy": 0.13528963476419448,
"num_tokens": 3274614.0,
"step": 1760
},
{
"entropy": 6.61256685256958,
"epoch": 0.1558636524196397,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999485477339907,
"loss": 6.5201,
"mean_token_accuracy": 0.13317029252648355,
"num_tokens": 3283800.0,
"step": 1765
},
{
"entropy": 6.526872968673706,
"epoch": 0.15630519251148003,
"grad_norm": 1.078125,
"learning_rate": 0.0004999478720973753,
"loss": 6.4319,
"mean_token_accuracy": 0.14009243845939637,
"num_tokens": 3293221.0,
"step": 1770
},
{
"entropy": 6.5570995807647705,
"epoch": 0.1567467326033204,
"grad_norm": 0.97265625,
"learning_rate": 0.000499947192054159,
"loss": 6.6029,
"mean_token_accuracy": 0.1324251540005207,
"num_tokens": 3302852.0,
"step": 1775
},
{
"entropy": 6.650819587707519,
"epoch": 0.15718827269516072,
"grad_norm": 1.1484375,
"learning_rate": 0.000499946507604355,
"loss": 6.4641,
"mean_token_accuracy": 0.14009604677557946,
"num_tokens": 3311752.0,
"step": 1780
},
{
"entropy": 6.473453664779663,
"epoch": 0.15762981278700106,
"grad_norm": 1.078125,
"learning_rate": 0.000499945818747977,
"loss": 6.5492,
"mean_token_accuracy": 0.13722263872623444,
"num_tokens": 3321339.0,
"step": 1785
},
{
"entropy": 6.671073293685913,
"epoch": 0.1580713528788414,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999451254850383,
"loss": 6.5514,
"mean_token_accuracy": 0.1324629843235016,
"num_tokens": 3330269.0,
"step": 1790
},
{
"entropy": 6.550148677825928,
"epoch": 0.15851289297068175,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999444278155525,
"loss": 6.4576,
"mean_token_accuracy": 0.1376182422041893,
"num_tokens": 3340770.0,
"step": 1795
},
{
"entropy": 6.544644594192505,
"epoch": 0.15895443306252208,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999437257395333,
"loss": 6.5753,
"mean_token_accuracy": 0.13476464301347732,
"num_tokens": 3349976.0,
"step": 1800
},
{
"entropy": 6.597910165786743,
"epoch": 0.1593959731543624,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999430192569944,
"loss": 6.6158,
"mean_token_accuracy": 0.12608520165085793,
"num_tokens": 3359764.0,
"step": 1805
},
{
"entropy": 6.539244031906128,
"epoch": 0.15983751324620277,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999423083679498,
"loss": 6.4868,
"mean_token_accuracy": 0.13115186169743537,
"num_tokens": 3369939.0,
"step": 1810
},
{
"entropy": 6.639003801345825,
"epoch": 0.1602790533380431,
"grad_norm": 1.03125,
"learning_rate": 0.0004999415930724133,
"loss": 6.6346,
"mean_token_accuracy": 0.12994891554117202,
"num_tokens": 3381326.0,
"step": 1815
},
{
"entropy": 6.659589242935181,
"epoch": 0.16072059342988343,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999408733703988,
"loss": 6.5949,
"mean_token_accuracy": 0.12143847793340683,
"num_tokens": 3391016.0,
"step": 1820
},
{
"entropy": 6.533053731918335,
"epoch": 0.16116213352172376,
"grad_norm": 0.953125,
"learning_rate": 0.0004999401492619207,
"loss": 6.3366,
"mean_token_accuracy": 0.13945143148303032,
"num_tokens": 3400294.0,
"step": 1825
},
{
"entropy": 6.587634897232055,
"epoch": 0.16160367361356412,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999394207469928,
"loss": 6.5447,
"mean_token_accuracy": 0.12779992073774338,
"num_tokens": 3409685.0,
"step": 1830
},
{
"entropy": 6.636557579040527,
"epoch": 0.16204521370540445,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999386878256297,
"loss": 6.4611,
"mean_token_accuracy": 0.1320968374609947,
"num_tokens": 3418946.0,
"step": 1835
},
{
"entropy": 6.423011112213135,
"epoch": 0.16248675379724478,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999379504978457,
"loss": 6.468,
"mean_token_accuracy": 0.13338307663798332,
"num_tokens": 3428245.0,
"step": 1840
},
{
"entropy": 6.578200912475586,
"epoch": 0.16292829388908514,
"grad_norm": 1.046875,
"learning_rate": 0.000499937208763655,
"loss": 6.4741,
"mean_token_accuracy": 0.1402788795530796,
"num_tokens": 3437580.0,
"step": 1845
},
{
"entropy": 6.617180633544922,
"epoch": 0.16336983398092547,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999364626230724,
"loss": 6.4472,
"mean_token_accuracy": 0.14007550328969956,
"num_tokens": 3446393.0,
"step": 1850
},
{
"entropy": 6.451139450073242,
"epoch": 0.1638113740727658,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999357120761124,
"loss": 6.551,
"mean_token_accuracy": 0.13571736514568328,
"num_tokens": 3455956.0,
"step": 1855
},
{
"entropy": 6.6568724632263185,
"epoch": 0.16425291416460613,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999349571227898,
"loss": 6.5495,
"mean_token_accuracy": 0.13430711701512338,
"num_tokens": 3465722.0,
"step": 1860
},
{
"entropy": 6.55973629951477,
"epoch": 0.1646944542564465,
"grad_norm": 0.953125,
"learning_rate": 0.0004999341977631193,
"loss": 6.52,
"mean_token_accuracy": 0.12885117009282113,
"num_tokens": 3475467.0,
"step": 1865
},
{
"entropy": 6.513189268112183,
"epoch": 0.16513599434828682,
"grad_norm": 1.0,
"learning_rate": 0.0004999334339971157,
"loss": 6.4162,
"mean_token_accuracy": 0.13661258816719055,
"num_tokens": 3484931.0,
"step": 1870
},
{
"entropy": 6.572188520431519,
"epoch": 0.16557753444012716,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999326658247942,
"loss": 6.5161,
"mean_token_accuracy": 0.13407543525099755,
"num_tokens": 3494001.0,
"step": 1875
},
{
"entropy": 6.522925567626953,
"epoch": 0.16601907453196751,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999318932461696,
"loss": 6.4558,
"mean_token_accuracy": 0.1386028841137886,
"num_tokens": 3503021.0,
"step": 1880
},
{
"entropy": 6.46082649230957,
"epoch": 0.16646061462380785,
"grad_norm": 1.2421875,
"learning_rate": 0.0004999311162612571,
"loss": 6.4843,
"mean_token_accuracy": 0.13625017702579498,
"num_tokens": 3513005.0,
"step": 1885
},
{
"entropy": 6.661849021911621,
"epoch": 0.16690215471564818,
"grad_norm": 0.9609375,
"learning_rate": 0.000499930334870072,
"loss": 6.5629,
"mean_token_accuracy": 0.13853515014052392,
"num_tokens": 3523219.0,
"step": 1890
},
{
"entropy": 6.584684419631958,
"epoch": 0.1673436948074885,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999295490726296,
"loss": 6.51,
"mean_token_accuracy": 0.13359814062714576,
"num_tokens": 3532917.0,
"step": 1895
},
{
"entropy": 6.513212633132935,
"epoch": 0.16778523489932887,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999287588689453,
"loss": 6.5245,
"mean_token_accuracy": 0.1389150969684124,
"num_tokens": 3542845.0,
"step": 1900
},
{
"entropy": 6.599180459976196,
"epoch": 0.1682267749911692,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999279642590344,
"loss": 6.5438,
"mean_token_accuracy": 0.13201173320412635,
"num_tokens": 3552126.0,
"step": 1905
},
{
"entropy": 6.5960267066955565,
"epoch": 0.16866831508300953,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999271652429127,
"loss": 6.5561,
"mean_token_accuracy": 0.12684691920876504,
"num_tokens": 3561254.0,
"step": 1910
},
{
"entropy": 6.528163957595825,
"epoch": 0.1691098551748499,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999263618205958,
"loss": 6.3933,
"mean_token_accuracy": 0.1403527893126011,
"num_tokens": 3569781.0,
"step": 1915
},
{
"entropy": 6.555511140823365,
"epoch": 0.16955139526669022,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999255539920993,
"loss": 6.4866,
"mean_token_accuracy": 0.13409090787172318,
"num_tokens": 3579664.0,
"step": 1920
},
{
"entropy": 6.567546367645264,
"epoch": 0.16999293535853055,
"grad_norm": 1.015625,
"learning_rate": 0.0004999247417574391,
"loss": 6.5376,
"mean_token_accuracy": 0.13570395410060881,
"num_tokens": 3588671.0,
"step": 1925
},
{
"entropy": 6.54329285621643,
"epoch": 0.17043447545037088,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999239251166312,
"loss": 6.4281,
"mean_token_accuracy": 0.13799721151590347,
"num_tokens": 3597656.0,
"step": 1930
},
{
"entropy": 6.480014228820801,
"epoch": 0.17087601554221124,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999231040696914,
"loss": 6.4491,
"mean_token_accuracy": 0.13938046917319297,
"num_tokens": 3608017.0,
"step": 1935
},
{
"entropy": 6.529258060455322,
"epoch": 0.17131755563405157,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999222786166361,
"loss": 6.5236,
"mean_token_accuracy": 0.1355483777821064,
"num_tokens": 3618189.0,
"step": 1940
},
{
"entropy": 6.60916223526001,
"epoch": 0.1717590957258919,
"grad_norm": 1.125,
"learning_rate": 0.0004999214487574812,
"loss": 6.4772,
"mean_token_accuracy": 0.13526797890663148,
"num_tokens": 3627211.0,
"step": 1945
},
{
"entropy": 6.498088455200195,
"epoch": 0.17220063581773226,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999206144922431,
"loss": 6.4181,
"mean_token_accuracy": 0.13236208409070968,
"num_tokens": 3636781.0,
"step": 1950
},
{
"entropy": 6.4924522876739506,
"epoch": 0.1726421759095726,
"grad_norm": 1.0859375,
"learning_rate": 0.000499919775820938,
"loss": 6.4985,
"mean_token_accuracy": 0.13709900975227357,
"num_tokens": 3644891.0,
"step": 1955
},
{
"entropy": 6.591422748565674,
"epoch": 0.17308371600141292,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999189327435825,
"loss": 6.5295,
"mean_token_accuracy": 0.13574066162109374,
"num_tokens": 3655477.0,
"step": 1960
},
{
"entropy": 6.5104138374328615,
"epoch": 0.17352525609325326,
"grad_norm": 1.015625,
"learning_rate": 0.0004999180852601929,
"loss": 6.5386,
"mean_token_accuracy": 0.13653166219592094,
"num_tokens": 3664542.0,
"step": 1965
},
{
"entropy": 6.577597141265869,
"epoch": 0.17396679618509361,
"grad_norm": 1.2265625,
"learning_rate": 0.000499917233370786,
"loss": 6.4609,
"mean_token_accuracy": 0.1293606199324131,
"num_tokens": 3673806.0,
"step": 1970
},
{
"entropy": 6.484398126602173,
"epoch": 0.17440833627693395,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999163770753784,
"loss": 6.4524,
"mean_token_accuracy": 0.13555625528097154,
"num_tokens": 3683238.0,
"step": 1975
},
{
"entropy": 6.558987331390381,
"epoch": 0.17484987636877428,
"grad_norm": 1.2265625,
"learning_rate": 0.0004999155163739869,
"loss": 6.4372,
"mean_token_accuracy": 0.13345005139708518,
"num_tokens": 3692161.0,
"step": 1980
},
{
"entropy": 6.510820627212524,
"epoch": 0.17529141646061464,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999146512666284,
"loss": 6.4535,
"mean_token_accuracy": 0.13496344164013863,
"num_tokens": 3701431.0,
"step": 1985
},
{
"entropy": 6.530817365646362,
"epoch": 0.17573295655245497,
"grad_norm": 0.890625,
"learning_rate": 0.0004999137817533197,
"loss": 6.4293,
"mean_token_accuracy": 0.1400221474468708,
"num_tokens": 3710963.0,
"step": 1990
},
{
"entropy": 6.48544750213623,
"epoch": 0.1761744966442953,
"grad_norm": 1.125,
"learning_rate": 0.0004999129078340779,
"loss": 6.4221,
"mean_token_accuracy": 0.14005866199731826,
"num_tokens": 3720177.0,
"step": 1995
},
{
"entropy": 6.465721511840821,
"epoch": 0.17661603673613563,
"grad_norm": 1.09375,
"learning_rate": 0.0004999120295089202,
"loss": 6.3795,
"mean_token_accuracy": 0.14514245688915253,
"num_tokens": 3728123.0,
"step": 2000
},
{
"entropy": 6.571569442749023,
"epoch": 0.177057576827976,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999111467778639,
"loss": 6.4978,
"mean_token_accuracy": 0.1349009484052658,
"num_tokens": 3736869.0,
"step": 2005
},
{
"entropy": 6.4819518566131595,
"epoch": 0.17749911691981632,
"grad_norm": 1.109375,
"learning_rate": 0.000499910259640926,
"loss": 6.3989,
"mean_token_accuracy": 0.1346891440451145,
"num_tokens": 3745830.0,
"step": 2010
},
{
"entropy": 6.486936283111572,
"epoch": 0.17794065701165665,
"grad_norm": 0.9921875,
"learning_rate": 0.000499909368098124,
"loss": 6.4328,
"mean_token_accuracy": 0.13926308006048202,
"num_tokens": 3755019.0,
"step": 2015
},
{
"entropy": 6.517996597290039,
"epoch": 0.178382197103497,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999084721494754,
"loss": 6.4076,
"mean_token_accuracy": 0.13306454047560692,
"num_tokens": 3764814.0,
"step": 2020
},
{
"entropy": 6.465504217147827,
"epoch": 0.17882373719533734,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999075717949978,
"loss": 6.384,
"mean_token_accuracy": 0.14004691764712335,
"num_tokens": 3774258.0,
"step": 2025
},
{
"entropy": 6.422070741653442,
"epoch": 0.17926527728717767,
"grad_norm": 0.9375,
"learning_rate": 0.0004999066670347089,
"loss": 6.4155,
"mean_token_accuracy": 0.14247353076934816,
"num_tokens": 3783961.0,
"step": 2030
},
{
"entropy": 6.416653776168824,
"epoch": 0.179706817379018,
"grad_norm": 0.953125,
"learning_rate": 0.0004999057578686261,
"loss": 6.3804,
"mean_token_accuracy": 0.1390853337943554,
"num_tokens": 3792417.0,
"step": 2035
},
{
"entropy": 6.496200704574585,
"epoch": 0.18014835747085836,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999048442967675,
"loss": 6.3592,
"mean_token_accuracy": 0.13822411596775055,
"num_tokens": 3801479.0,
"step": 2040
},
{
"entropy": 6.456766891479492,
"epoch": 0.1805898975626987,
"grad_norm": 0.96875,
"learning_rate": 0.0004999039263191508,
"loss": 6.4305,
"mean_token_accuracy": 0.133939179033041,
"num_tokens": 3810799.0,
"step": 2045
},
{
"entropy": 6.5264753818511965,
"epoch": 0.18103143765453902,
"grad_norm": 1.03125,
"learning_rate": 0.0004999030039357943,
"loss": 6.48,
"mean_token_accuracy": 0.1298608623445034,
"num_tokens": 3820966.0,
"step": 2050
},
{
"entropy": 6.493787717819214,
"epoch": 0.18147297774637938,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999020771467158,
"loss": 6.4572,
"mean_token_accuracy": 0.1327923409640789,
"num_tokens": 3829247.0,
"step": 2055
},
{
"entropy": 6.596197843551636,
"epoch": 0.1819145178382197,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999011459519335,
"loss": 6.3948,
"mean_token_accuracy": 0.13827238082885743,
"num_tokens": 3838114.0,
"step": 2060
},
{
"entropy": 6.390679597854614,
"epoch": 0.18235605793006004,
"grad_norm": 0.96875,
"learning_rate": 0.0004999002103514655,
"loss": 6.4735,
"mean_token_accuracy": 0.13869670927524566,
"num_tokens": 3848075.0,
"step": 2065
},
{
"entropy": 6.585737419128418,
"epoch": 0.18279759802190038,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998992703453304,
"loss": 6.5106,
"mean_token_accuracy": 0.13122646436095237,
"num_tokens": 3857934.0,
"step": 2070
},
{
"entropy": 6.450649833679199,
"epoch": 0.18323913811374073,
"grad_norm": 0.953125,
"learning_rate": 0.0004998983259335466,
"loss": 6.3519,
"mean_token_accuracy": 0.14236594662070273,
"num_tokens": 3866707.0,
"step": 2075
},
{
"entropy": 6.58008508682251,
"epoch": 0.18368067820558107,
"grad_norm": 1.046875,
"learning_rate": 0.0004998973771161324,
"loss": 6.4244,
"mean_token_accuracy": 0.1374949462711811,
"num_tokens": 3875233.0,
"step": 2080
},
{
"entropy": 6.389654541015625,
"epoch": 0.1841222182974214,
"grad_norm": 1.015625,
"learning_rate": 0.0004998964238931065,
"loss": 6.4131,
"mean_token_accuracy": 0.1403422772884369,
"num_tokens": 3885173.0,
"step": 2085
},
{
"entropy": 6.564711236953736,
"epoch": 0.18456375838926176,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998954662644876,
"loss": 6.3803,
"mean_token_accuracy": 0.13195990920066833,
"num_tokens": 3894198.0,
"step": 2090
},
{
"entropy": 6.44653811454773,
"epoch": 0.1850052984811021,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998945042302943,
"loss": 6.382,
"mean_token_accuracy": 0.1373509407043457,
"num_tokens": 3904076.0,
"step": 2095
},
{
"entropy": 6.4586262702941895,
"epoch": 0.18544683857294242,
"grad_norm": 1.0625,
"learning_rate": 0.0004998935377905457,
"loss": 6.4943,
"mean_token_accuracy": 0.13231708630919456,
"num_tokens": 3913204.0,
"step": 2100
},
{
"entropy": 6.587808132171631,
"epoch": 0.18588837866478275,
"grad_norm": 0.984375,
"learning_rate": 0.0004998925669452605,
"loss": 6.4565,
"mean_token_accuracy": 0.1339179016649723,
"num_tokens": 3922148.0,
"step": 2105
},
{
"entropy": 6.401728963851928,
"epoch": 0.1863299187566231,
"grad_norm": 0.921875,
"learning_rate": 0.0004998915916944579,
"loss": 6.4234,
"mean_token_accuracy": 0.1407448723912239,
"num_tokens": 3931333.0,
"step": 2110
},
{
"entropy": 6.545709466934204,
"epoch": 0.18677145884846344,
"grad_norm": 1.078125,
"learning_rate": 0.0004998906120381568,
"loss": 6.3789,
"mean_token_accuracy": 0.1448886923491955,
"num_tokens": 3941061.0,
"step": 2115
},
{
"entropy": 6.505582046508789,
"epoch": 0.18721299894030377,
"grad_norm": 1.078125,
"learning_rate": 0.0004998896279763766,
"loss": 6.4761,
"mean_token_accuracy": 0.13257319629192352,
"num_tokens": 3950075.0,
"step": 2120
},
{
"entropy": 6.455861520767212,
"epoch": 0.18765453903214413,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998886395091365,
"loss": 6.3345,
"mean_token_accuracy": 0.1409289576113224,
"num_tokens": 3958885.0,
"step": 2125
},
{
"entropy": 6.417616128921509,
"epoch": 0.18809607912398446,
"grad_norm": 1.03125,
"learning_rate": 0.0004998876466364559,
"loss": 6.437,
"mean_token_accuracy": 0.13843559697270394,
"num_tokens": 3968218.0,
"step": 2130
},
{
"entropy": 6.400978994369507,
"epoch": 0.1885376192158248,
"grad_norm": 1.046875,
"learning_rate": 0.0004998866493583541,
"loss": 6.364,
"mean_token_accuracy": 0.14499804973602295,
"num_tokens": 3977435.0,
"step": 2135
},
{
"entropy": 6.388528203964233,
"epoch": 0.18897915930766512,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998856476748509,
"loss": 6.349,
"mean_token_accuracy": 0.14326094537973405,
"num_tokens": 3986608.0,
"step": 2140
},
{
"entropy": 6.440869951248169,
"epoch": 0.18942069939950548,
"grad_norm": 1.0625,
"learning_rate": 0.0004998846415859656,
"loss": 6.3602,
"mean_token_accuracy": 0.14130929261445999,
"num_tokens": 3996087.0,
"step": 2145
},
{
"entropy": 6.476541662216187,
"epoch": 0.1898622394913458,
"grad_norm": 1.0,
"learning_rate": 0.0004998836310917182,
"loss": 6.4058,
"mean_token_accuracy": 0.13654726892709732,
"num_tokens": 4006257.0,
"step": 2150
},
{
"entropy": 6.478487300872803,
"epoch": 0.19030377958318614,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998826161921282,
"loss": 6.4072,
"mean_token_accuracy": 0.14593140706419944,
"num_tokens": 4015904.0,
"step": 2155
},
{
"entropy": 6.427832221984863,
"epoch": 0.1907453196750265,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998815968872157,
"loss": 6.4181,
"mean_token_accuracy": 0.13653010204434396,
"num_tokens": 4025417.0,
"step": 2160
},
{
"entropy": 6.392728614807129,
"epoch": 0.19118685976686683,
"grad_norm": 0.984375,
"learning_rate": 0.0004998805731770007,
"loss": 6.2628,
"mean_token_accuracy": 0.15659967064857483,
"num_tokens": 4035181.0,
"step": 2165
},
{
"entropy": 6.4461814880371096,
"epoch": 0.19162839985870717,
"grad_norm": 0.9765625,
"learning_rate": 0.000499879545061503,
"loss": 6.4504,
"mean_token_accuracy": 0.13371687456965448,
"num_tokens": 4045112.0,
"step": 2170
},
{
"entropy": 6.5883321285247805,
"epoch": 0.1920699399505475,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998785125407432,
"loss": 6.5425,
"mean_token_accuracy": 0.12125966772437095,
"num_tokens": 4054566.0,
"step": 2175
},
{
"entropy": 6.552111434936523,
"epoch": 0.19251148004238786,
"grad_norm": 1.0078125,
"learning_rate": 0.000499877475614741,
"loss": 6.3758,
"mean_token_accuracy": 0.1340583384037018,
"num_tokens": 4063960.0,
"step": 2180
},
{
"entropy": 6.3515486240386965,
"epoch": 0.1929530201342282,
"grad_norm": 1.015625,
"learning_rate": 0.0004998764342835169,
"loss": 6.3157,
"mean_token_accuracy": 0.14617449343204497,
"num_tokens": 4072620.0,
"step": 2185
},
{
"entropy": 6.406283712387085,
"epoch": 0.19339456022606852,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998753885470915,
"loss": 6.2789,
"mean_token_accuracy": 0.14062159806489943,
"num_tokens": 4081590.0,
"step": 2190
},
{
"entropy": 6.425132322311401,
"epoch": 0.19383610031790888,
"grad_norm": 1.03125,
"learning_rate": 0.0004998743384054851,
"loss": 6.4202,
"mean_token_accuracy": 0.14044143706560136,
"num_tokens": 4090758.0,
"step": 2195
},
{
"entropy": 6.3962568759918215,
"epoch": 0.1942776404097492,
"grad_norm": 1.015625,
"learning_rate": 0.0004998732838587183,
"loss": 6.2458,
"mean_token_accuracy": 0.149021477997303,
"num_tokens": 4099281.0,
"step": 2200
},
{
"entropy": 6.40768404006958,
"epoch": 0.19471918050158954,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998722249068118,
"loss": 6.3451,
"mean_token_accuracy": 0.1404854990541935,
"num_tokens": 4108953.0,
"step": 2205
},
{
"entropy": 6.396564531326294,
"epoch": 0.19516072059342987,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998711615497863,
"loss": 6.4535,
"mean_token_accuracy": 0.13740591406822206,
"num_tokens": 4118799.0,
"step": 2210
},
{
"entropy": 6.579476261138916,
"epoch": 0.19560226068527023,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998700937876626,
"loss": 6.4122,
"mean_token_accuracy": 0.13571444600820542,
"num_tokens": 4127862.0,
"step": 2215
},
{
"entropy": 6.4028857231140135,
"epoch": 0.19604380077711056,
"grad_norm": 1.078125,
"learning_rate": 0.0004998690216204615,
"loss": 6.5068,
"mean_token_accuracy": 0.12704429849982263,
"num_tokens": 4139029.0,
"step": 2220
},
{
"entropy": 6.5047478675842285,
"epoch": 0.1964853408689509,
"grad_norm": 1.078125,
"learning_rate": 0.0004998679450482043,
"loss": 6.3028,
"mean_token_accuracy": 0.14782762974500657,
"num_tokens": 4148257.0,
"step": 2225
},
{
"entropy": 6.341253662109375,
"epoch": 0.19692688096079125,
"grad_norm": 0.9921875,
"learning_rate": 0.000499866864070912,
"loss": 6.2627,
"mean_token_accuracy": 0.14017492160201073,
"num_tokens": 4157626.0,
"step": 2230
},
{
"entropy": 6.429310417175293,
"epoch": 0.19736842105263158,
"grad_norm": 1.09375,
"learning_rate": 0.0004998657786886056,
"loss": 6.4713,
"mean_token_accuracy": 0.13993202298879623,
"num_tokens": 4166804.0,
"step": 2235
},
{
"entropy": 6.554042720794678,
"epoch": 0.1978099611444719,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998646889013066,
"loss": 6.3607,
"mean_token_accuracy": 0.14240839183330536,
"num_tokens": 4175701.0,
"step": 2240
},
{
"entropy": 6.4098762512207035,
"epoch": 0.19825150123631224,
"grad_norm": 0.96875,
"learning_rate": 0.0004998635947090362,
"loss": 6.4425,
"mean_token_accuracy": 0.14030273035168647,
"num_tokens": 4184711.0,
"step": 2245
},
{
"entropy": 6.475190782546997,
"epoch": 0.1986930413281526,
"grad_norm": 1.2421875,
"learning_rate": 0.0004998624961118158,
"loss": 6.4017,
"mean_token_accuracy": 0.14099944159388542,
"num_tokens": 4193931.0,
"step": 2250
},
{
"entropy": 6.44319109916687,
"epoch": 0.19913458141999293,
"grad_norm": 1.0390625,
"learning_rate": 0.000499861393109667,
"loss": 6.2899,
"mean_token_accuracy": 0.14828752726316452,
"num_tokens": 4203000.0,
"step": 2255
},
{
"entropy": 6.274944400787353,
"epoch": 0.19957612151183327,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998602857026114,
"loss": 6.2991,
"mean_token_accuracy": 0.1458234503865242,
"num_tokens": 4211977.0,
"step": 2260
},
{
"entropy": 6.472422647476196,
"epoch": 0.20001766160367362,
"grad_norm": 1.0625,
"learning_rate": 0.0004998591738906708,
"loss": 6.3814,
"mean_token_accuracy": 0.15182094275951385,
"num_tokens": 4220375.0,
"step": 2265
},
{
"entropy": 6.494894886016846,
"epoch": 0.20045920169551396,
"grad_norm": 1.03125,
"learning_rate": 0.0004998580576738668,
"loss": 6.3793,
"mean_token_accuracy": 0.1367909237742424,
"num_tokens": 4230506.0,
"step": 2270
},
{
"entropy": 6.404332447052002,
"epoch": 0.2009007417873543,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998569370522213,
"loss": 6.3524,
"mean_token_accuracy": 0.13977290093898773,
"num_tokens": 4240235.0,
"step": 2275
},
{
"entropy": 6.448247480392456,
"epoch": 0.20134228187919462,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998558120257563,
"loss": 6.4189,
"mean_token_accuracy": 0.14339498728513717,
"num_tokens": 4249775.0,
"step": 2280
},
{
"entropy": 6.419510173797607,
"epoch": 0.20178382197103498,
"grad_norm": 1.0625,
"learning_rate": 0.0004998546825944938,
"loss": 6.3504,
"mean_token_accuracy": 0.141066338121891,
"num_tokens": 4258523.0,
"step": 2285
},
{
"entropy": 6.419132709503174,
"epoch": 0.2022253620628753,
"grad_norm": 1.046875,
"learning_rate": 0.000499853548758456,
"loss": 6.358,
"mean_token_accuracy": 0.13891511633992196,
"num_tokens": 4267683.0,
"step": 2290
},
{
"entropy": 6.495290946960449,
"epoch": 0.20266690215471564,
"grad_norm": 1.3828125,
"learning_rate": 0.000499852410517665,
"loss": 6.417,
"mean_token_accuracy": 0.13713881745934486,
"num_tokens": 4276903.0,
"step": 2295
},
{
"entropy": 6.426017999649048,
"epoch": 0.203108442246556,
"grad_norm": 1.4296875,
"learning_rate": 0.0004998512678721431,
"loss": 6.3661,
"mean_token_accuracy": 0.135909353941679,
"num_tokens": 4287257.0,
"step": 2300
},
{
"entropy": 6.447468948364258,
"epoch": 0.20354998233839633,
"grad_norm": 1.0625,
"learning_rate": 0.000499850120821913,
"loss": 6.3875,
"mean_token_accuracy": 0.13645304143428802,
"num_tokens": 4297345.0,
"step": 2305
},
{
"entropy": 6.528428220748902,
"epoch": 0.20399152243023666,
"grad_norm": 1.09375,
"learning_rate": 0.0004998489693669967,
"loss": 6.3613,
"mean_token_accuracy": 0.13837311565876007,
"num_tokens": 4306533.0,
"step": 2310
},
{
"entropy": 6.392482471466065,
"epoch": 0.204433062522077,
"grad_norm": 1.046875,
"learning_rate": 0.000499847813507417,
"loss": 6.4023,
"mean_token_accuracy": 0.14094773977994918,
"num_tokens": 4316338.0,
"step": 2315
},
{
"entropy": 6.399602174758911,
"epoch": 0.20487460261391735,
"grad_norm": 1.0,
"learning_rate": 0.0004998466532431966,
"loss": 6.3549,
"mean_token_accuracy": 0.13948202207684518,
"num_tokens": 4326585.0,
"step": 2320
},
{
"entropy": 6.544098567962647,
"epoch": 0.20531614270575768,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998454885743581,
"loss": 6.4795,
"mean_token_accuracy": 0.1351695440709591,
"num_tokens": 4336490.0,
"step": 2325
},
{
"entropy": 6.363640975952149,
"epoch": 0.205757682797598,
"grad_norm": 1.046875,
"learning_rate": 0.0004998443195009242,
"loss": 6.3589,
"mean_token_accuracy": 0.1348782531917095,
"num_tokens": 4346264.0,
"step": 2330
},
{
"entropy": 6.461185693740845,
"epoch": 0.20619922288943837,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998431460229182,
"loss": 6.3318,
"mean_token_accuracy": 0.1448797807097435,
"num_tokens": 4355102.0,
"step": 2335
},
{
"entropy": 6.482104396820068,
"epoch": 0.2066407629812787,
"grad_norm": 1.125,
"learning_rate": 0.0004998419681403627,
"loss": 6.5133,
"mean_token_accuracy": 0.13086750581860543,
"num_tokens": 4365569.0,
"step": 2340
},
{
"entropy": 6.507464361190796,
"epoch": 0.20708230307311903,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998407858532809,
"loss": 6.36,
"mean_token_accuracy": 0.14296017587184906,
"num_tokens": 4375437.0,
"step": 2345
},
{
"entropy": 6.37006025314331,
"epoch": 0.20752384316495937,
"grad_norm": 1.0546875,
"learning_rate": 0.000499839599161696,
"loss": 6.3389,
"mean_token_accuracy": 0.1450774312019348,
"num_tokens": 4384420.0,
"step": 2350
},
{
"entropy": 6.460700035095215,
"epoch": 0.20796538325679972,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998384080656314,
"loss": 6.3106,
"mean_token_accuracy": 0.14050144031643869,
"num_tokens": 4393730.0,
"step": 2355
},
{
"entropy": 6.332716035842895,
"epoch": 0.20840692334864006,
"grad_norm": 1.0625,
"learning_rate": 0.00049983721256511,
"loss": 6.2798,
"mean_token_accuracy": 0.14351205080747603,
"num_tokens": 4402731.0,
"step": 2360
},
{
"entropy": 6.3846518993377686,
"epoch": 0.2088484634404804,
"grad_norm": 1.078125,
"learning_rate": 0.0004998360126601556,
"loss": 6.3369,
"mean_token_accuracy": 0.14289727210998535,
"num_tokens": 4411606.0,
"step": 2365
},
{
"entropy": 6.3337644100189205,
"epoch": 0.20929000353232075,
"grad_norm": 1.0625,
"learning_rate": 0.0004998348083507916,
"loss": 6.4062,
"mean_token_accuracy": 0.14081210866570473,
"num_tokens": 4421685.0,
"step": 2370
},
{
"entropy": 6.558507633209229,
"epoch": 0.20973154362416108,
"grad_norm": 1.078125,
"learning_rate": 0.0004998335996370416,
"loss": 6.3782,
"mean_token_accuracy": 0.14399294778704644,
"num_tokens": 4431765.0,
"step": 2375
},
{
"entropy": 6.321137380599976,
"epoch": 0.2101730837160014,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998323865189291,
"loss": 6.3523,
"mean_token_accuracy": 0.13661579713225364,
"num_tokens": 4441191.0,
"step": 2380
},
{
"entropy": 6.481552457809448,
"epoch": 0.21061462380784174,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998311689964781,
"loss": 6.4322,
"mean_token_accuracy": 0.13680859059095382,
"num_tokens": 4450156.0,
"step": 2385
},
{
"entropy": 6.4807047843933105,
"epoch": 0.2110561638996821,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998299470697125,
"loss": 6.4163,
"mean_token_accuracy": 0.14260546639561653,
"num_tokens": 4459466.0,
"step": 2390
},
{
"entropy": 6.461113500595093,
"epoch": 0.21149770399152243,
"grad_norm": 0.984375,
"learning_rate": 0.0004998287207386559,
"loss": 6.4156,
"mean_token_accuracy": 0.14247968047857285,
"num_tokens": 4468539.0,
"step": 2395
},
{
"entropy": 6.525749444961548,
"epoch": 0.21193924408336276,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998274900033326,
"loss": 6.315,
"mean_token_accuracy": 0.14835015684366226,
"num_tokens": 4477579.0,
"step": 2400
},
{
"entropy": 6.236664247512818,
"epoch": 0.21238078417520312,
"grad_norm": 0.984375,
"learning_rate": 0.0004998262548637667,
"loss": 6.2842,
"mean_token_accuracy": 0.14991160482168198,
"num_tokens": 4486800.0,
"step": 2405
},
{
"entropy": 6.444223546981812,
"epoch": 0.21282232426704345,
"grad_norm": 1.03125,
"learning_rate": 0.0004998250153199822,
"loss": 6.2465,
"mean_token_accuracy": 0.14562757611274718,
"num_tokens": 4495985.0,
"step": 2410
},
{
"entropy": 6.417825555801391,
"epoch": 0.21326386435888378,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998237713720036,
"loss": 6.4031,
"mean_token_accuracy": 0.14113787487149237,
"num_tokens": 4504944.0,
"step": 2415
},
{
"entropy": 6.344033908843994,
"epoch": 0.2137054044507241,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998225230198552,
"loss": 6.2875,
"mean_token_accuracy": 0.14928205609321593,
"num_tokens": 4515402.0,
"step": 2420
},
{
"entropy": 6.412223052978516,
"epoch": 0.21414694454256447,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998212702635614,
"loss": 6.355,
"mean_token_accuracy": 0.14224686175584794,
"num_tokens": 4525009.0,
"step": 2425
},
{
"entropy": 6.537710332870484,
"epoch": 0.2145884846344048,
"grad_norm": 1.1484375,
"learning_rate": 0.0004998200131031469,
"loss": 6.4066,
"mean_token_accuracy": 0.142412006855011,
"num_tokens": 4534460.0,
"step": 2430
},
{
"entropy": 6.355252885818482,
"epoch": 0.21503002472624513,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998187515386361,
"loss": 6.2239,
"mean_token_accuracy": 0.15292632952332497,
"num_tokens": 4543748.0,
"step": 2435
},
{
"entropy": 6.32159743309021,
"epoch": 0.2154715648180855,
"grad_norm": 1.046875,
"learning_rate": 0.0004998174855700538,
"loss": 6.3515,
"mean_token_accuracy": 0.14536840543150903,
"num_tokens": 4552722.0,
"step": 2440
},
{
"entropy": 6.455835342407227,
"epoch": 0.21591310490992582,
"grad_norm": 1.1953125,
"learning_rate": 0.0004998162151974248,
"loss": 6.2244,
"mean_token_accuracy": 0.1433302193880081,
"num_tokens": 4561607.0,
"step": 2445
},
{
"entropy": 6.376346635818481,
"epoch": 0.21635464500176615,
"grad_norm": 0.89453125,
"learning_rate": 0.000499814940420774,
"loss": 6.4926,
"mean_token_accuracy": 0.13043315410614015,
"num_tokens": 4572524.0,
"step": 2450
},
{
"entropy": 6.473890399932861,
"epoch": 0.21679618509360649,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998136612401266,
"loss": 6.306,
"mean_token_accuracy": 0.13808697760105132,
"num_tokens": 4581601.0,
"step": 2455
},
{
"entropy": 6.362298917770386,
"epoch": 0.21723772518544684,
"grad_norm": 1.0625,
"learning_rate": 0.0004998123776555071,
"loss": 6.355,
"mean_token_accuracy": 0.13434374257922171,
"num_tokens": 4591795.0,
"step": 2460
},
{
"entropy": 6.4316198348999025,
"epoch": 0.21767926527728718,
"grad_norm": 1.046875,
"learning_rate": 0.0004998110896669412,
"loss": 6.3127,
"mean_token_accuracy": 0.14232389852404595,
"num_tokens": 4600745.0,
"step": 2465
},
{
"entropy": 6.403714847564697,
"epoch": 0.2181208053691275,
"grad_norm": 0.953125,
"learning_rate": 0.0004998097972744539,
"loss": 6.3668,
"mean_token_accuracy": 0.13958390951156616,
"num_tokens": 4610490.0,
"step": 2470
},
{
"entropy": 6.421542739868164,
"epoch": 0.21856234546096787,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998085004780705,
"loss": 6.3027,
"mean_token_accuracy": 0.14644000679254532,
"num_tokens": 4619511.0,
"step": 2475
},
{
"entropy": 6.411759996414185,
"epoch": 0.2190038855528082,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998071992778164,
"loss": 6.3536,
"mean_token_accuracy": 0.13926490917801856,
"num_tokens": 4628186.0,
"step": 2480
},
{
"entropy": 6.416973400115967,
"epoch": 0.21944542564464853,
"grad_norm": 1.0390625,
"learning_rate": 0.000499805893673717,
"loss": 6.3574,
"mean_token_accuracy": 0.14341954439878463,
"num_tokens": 4637431.0,
"step": 2485
},
{
"entropy": 6.347478723526001,
"epoch": 0.21988696573648886,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998045836657982,
"loss": 6.2093,
"mean_token_accuracy": 0.14213306605815887,
"num_tokens": 4646627.0,
"step": 2490
},
{
"entropy": 6.368200349807739,
"epoch": 0.22032850582832922,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998032692540853,
"loss": 6.3012,
"mean_token_accuracy": 0.14045739471912383,
"num_tokens": 4656095.0,
"step": 2495
},
{
"entropy": 6.476345205307007,
"epoch": 0.22077004592016955,
"grad_norm": 1.03125,
"learning_rate": 0.0004998019504386044,
"loss": 6.3876,
"mean_token_accuracy": 0.13640450164675713,
"num_tokens": 4665807.0,
"step": 2500
},
{
"entropy": 6.444914436340332,
"epoch": 0.22121158601200988,
"grad_norm": 1.09375,
"learning_rate": 0.0004998006272193809,
"loss": 6.3493,
"mean_token_accuracy": 0.13855071663856505,
"num_tokens": 4674459.0,
"step": 2505
},
{
"entropy": 6.361084604263306,
"epoch": 0.22165312610385024,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997992995964412,
"loss": 6.4325,
"mean_token_accuracy": 0.13779560700058938,
"num_tokens": 4684063.0,
"step": 2510
},
{
"entropy": 6.45024299621582,
"epoch": 0.22209466619569057,
"grad_norm": 1.15625,
"learning_rate": 0.0004997979675698109,
"loss": 6.3029,
"mean_token_accuracy": 0.14212062656879426,
"num_tokens": 4692807.0,
"step": 2515
},
{
"entropy": 6.438331031799317,
"epoch": 0.2225362062875309,
"grad_norm": 1.0625,
"learning_rate": 0.0004997966311395164,
"loss": 6.2422,
"mean_token_accuracy": 0.14746622294187545,
"num_tokens": 4701100.0,
"step": 2520
},
{
"entropy": 6.284497547149658,
"epoch": 0.22297774637937123,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997952903055836,
"loss": 6.3071,
"mean_token_accuracy": 0.13971827551722527,
"num_tokens": 4710697.0,
"step": 2525
},
{
"entropy": 6.4242840766906735,
"epoch": 0.2234192864712116,
"grad_norm": 1.125,
"learning_rate": 0.000499793945068039,
"loss": 6.2875,
"mean_token_accuracy": 0.14202770590782166,
"num_tokens": 4718745.0,
"step": 2530
},
{
"entropy": 6.358328342437744,
"epoch": 0.22386082656305192,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997925954269088,
"loss": 6.2493,
"mean_token_accuracy": 0.15010830983519555,
"num_tokens": 4728056.0,
"step": 2535
},
{
"entropy": 6.390234851837159,
"epoch": 0.22430236665489225,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997912413822196,
"loss": 6.3892,
"mean_token_accuracy": 0.14226726815104485,
"num_tokens": 4737605.0,
"step": 2540
},
{
"entropy": 6.333422613143921,
"epoch": 0.2247439067467326,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997898829339979,
"loss": 6.216,
"mean_token_accuracy": 0.15019772350788116,
"num_tokens": 4746168.0,
"step": 2545
},
{
"entropy": 6.414393615722656,
"epoch": 0.22518544683857294,
"grad_norm": 1.015625,
"learning_rate": 0.00049978852008227,
"loss": 6.2827,
"mean_token_accuracy": 0.14390757903456688,
"num_tokens": 4755072.0,
"step": 2550
},
{
"entropy": 6.292112064361572,
"epoch": 0.22562698693041328,
"grad_norm": 0.96484375,
"learning_rate": 0.000499787152827063,
"loss": 6.2887,
"mean_token_accuracy": 0.14927180036902427,
"num_tokens": 4764580.0,
"step": 2555
},
{
"entropy": 6.362084197998047,
"epoch": 0.2260685270222536,
"grad_norm": 1.03125,
"learning_rate": 0.0004997857811684035,
"loss": 6.3109,
"mean_token_accuracy": 0.14619807451963424,
"num_tokens": 4774135.0,
"step": 2560
},
{
"entropy": 6.443705081939697,
"epoch": 0.22651006711409397,
"grad_norm": 1.03125,
"learning_rate": 0.0004997844051063183,
"loss": 6.3931,
"mean_token_accuracy": 0.1407366193830967,
"num_tokens": 4784733.0,
"step": 2565
},
{
"entropy": 6.393408250808716,
"epoch": 0.2269516072059343,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997830246408346,
"loss": 6.3304,
"mean_token_accuracy": 0.14791636019945145,
"num_tokens": 4795327.0,
"step": 2570
},
{
"entropy": 6.342170095443725,
"epoch": 0.22739314729777463,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997816397719791,
"loss": 6.3028,
"mean_token_accuracy": 0.14114121049642564,
"num_tokens": 4804314.0,
"step": 2575
},
{
"entropy": 6.458865118026734,
"epoch": 0.227834687389615,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997802504997792,
"loss": 6.3913,
"mean_token_accuracy": 0.13652418628335,
"num_tokens": 4813637.0,
"step": 2580
},
{
"entropy": 6.382553291320801,
"epoch": 0.22827622748145532,
"grad_norm": 1.015625,
"learning_rate": 0.0004997788568242621,
"loss": 6.2591,
"mean_token_accuracy": 0.14338775500655174,
"num_tokens": 4823094.0,
"step": 2585
},
{
"entropy": 6.327674865722656,
"epoch": 0.22871776757329565,
"grad_norm": 0.8984375,
"learning_rate": 0.000499777458745455,
"loss": 6.1969,
"mean_token_accuracy": 0.1477431207895279,
"num_tokens": 4833199.0,
"step": 2590
},
{
"entropy": 6.392719554901123,
"epoch": 0.22915930766513598,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997760562633853,
"loss": 6.2909,
"mean_token_accuracy": 0.14219107255339622,
"num_tokens": 4842970.0,
"step": 2595
},
{
"entropy": 6.394578695297241,
"epoch": 0.22960084775697634,
"grad_norm": 1.015625,
"learning_rate": 0.0004997746493780804,
"loss": 6.3788,
"mean_token_accuracy": 0.13738251477479935,
"num_tokens": 4852043.0,
"step": 2600
},
{
"entropy": 6.376160097122193,
"epoch": 0.23004238784881667,
"grad_norm": 1.0390625,
"learning_rate": 0.000499773238089568,
"loss": 6.2693,
"mean_token_accuracy": 0.13766007199883462,
"num_tokens": 4862232.0,
"step": 2605
},
{
"entropy": 6.345012950897217,
"epoch": 0.230483927940657,
"grad_norm": 1.125,
"learning_rate": 0.0004997718223978758,
"loss": 6.2081,
"mean_token_accuracy": 0.1466532751917839,
"num_tokens": 4871186.0,
"step": 2610
},
{
"entropy": 6.343753385543823,
"epoch": 0.23092546803249736,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997704023030315,
"loss": 6.3059,
"mean_token_accuracy": 0.150559451431036,
"num_tokens": 4879974.0,
"step": 2615
},
{
"entropy": 6.393126726150513,
"epoch": 0.2313670081243377,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997689778050627,
"loss": 6.3617,
"mean_token_accuracy": 0.14292784333229064,
"num_tokens": 4890300.0,
"step": 2620
},
{
"entropy": 6.369229030609131,
"epoch": 0.23180854821617802,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997675489039975,
"loss": 6.3301,
"mean_token_accuracy": 0.1426799289882183,
"num_tokens": 4900428.0,
"step": 2625
},
{
"entropy": 6.3974145412445065,
"epoch": 0.23225008830801835,
"grad_norm": 1.03125,
"learning_rate": 0.0004997661155998638,
"loss": 6.3245,
"mean_token_accuracy": 0.1442883849143982,
"num_tokens": 4910092.0,
"step": 2630
},
{
"entropy": 6.384046411514282,
"epoch": 0.2326916283998587,
"grad_norm": 1.03125,
"learning_rate": 0.0004997646778926898,
"loss": 6.3247,
"mean_token_accuracy": 0.13738622814416884,
"num_tokens": 4919593.0,
"step": 2635
},
{
"entropy": 6.373585891723633,
"epoch": 0.23313316849169904,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997632357825035,
"loss": 6.3001,
"mean_token_accuracy": 0.14202155098319053,
"num_tokens": 4929098.0,
"step": 2640
},
{
"entropy": 6.43012547492981,
"epoch": 0.23357470858353938,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997617892693333,
"loss": 6.3657,
"mean_token_accuracy": 0.1421157017350197,
"num_tokens": 4938265.0,
"step": 2645
},
{
"entropy": 6.394748878479004,
"epoch": 0.23401624867537973,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997603383532075,
"loss": 6.3016,
"mean_token_accuracy": 0.14679210409522056,
"num_tokens": 4946694.0,
"step": 2650
},
{
"entropy": 6.368328046798706,
"epoch": 0.23445778876722007,
"grad_norm": 1.078125,
"learning_rate": 0.0004997588830341545,
"loss": 6.3132,
"mean_token_accuracy": 0.1434150867164135,
"num_tokens": 4955296.0,
"step": 2655
},
{
"entropy": 6.32787938117981,
"epoch": 0.2348993288590604,
"grad_norm": 1.015625,
"learning_rate": 0.0004997574233122028,
"loss": 6.2759,
"mean_token_accuracy": 0.14597226828336715,
"num_tokens": 4964409.0,
"step": 2660
},
{
"entropy": 6.3884584426879885,
"epoch": 0.23534086895090076,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997559591873809,
"loss": 6.284,
"mean_token_accuracy": 0.1481903851032257,
"num_tokens": 4973449.0,
"step": 2665
},
{
"entropy": 6.345009279251099,
"epoch": 0.2357824090427411,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997544906597178,
"loss": 6.2779,
"mean_token_accuracy": 0.1470661997795105,
"num_tokens": 4983057.0,
"step": 2670
},
{
"entropy": 6.2613893985748295,
"epoch": 0.23622394913458142,
"grad_norm": 2.03125,
"learning_rate": 0.0004997530177292418,
"loss": 6.3532,
"mean_token_accuracy": 0.13861697241663934,
"num_tokens": 4991950.0,
"step": 2675
},
{
"entropy": 6.461032247543335,
"epoch": 0.23666548922642175,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997515403959823,
"loss": 6.2857,
"mean_token_accuracy": 0.14603266417980193,
"num_tokens": 5001042.0,
"step": 2680
},
{
"entropy": 6.356680679321289,
"epoch": 0.2371070293182621,
"grad_norm": 1.0625,
"learning_rate": 0.0004997500586599677,
"loss": 6.2198,
"mean_token_accuracy": 0.15022996366024016,
"num_tokens": 5009827.0,
"step": 2685
},
{
"entropy": 6.292784547805786,
"epoch": 0.23754856941010244,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997485725212274,
"loss": 6.2662,
"mean_token_accuracy": 0.1465997129678726,
"num_tokens": 5018708.0,
"step": 2690
},
{
"entropy": 6.334398937225342,
"epoch": 0.23799010950194277,
"grad_norm": 1.046875,
"learning_rate": 0.0004997470819797903,
"loss": 6.1678,
"mean_token_accuracy": 0.149826068431139,
"num_tokens": 5027522.0,
"step": 2695
},
{
"entropy": 6.312096786499024,
"epoch": 0.23843164959378313,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997455870356857,
"loss": 6.2858,
"mean_token_accuracy": 0.14754335582256317,
"num_tokens": 5035755.0,
"step": 2700
},
{
"entropy": 6.38718318939209,
"epoch": 0.23887318968562346,
"grad_norm": 1.078125,
"learning_rate": 0.0004997440876889429,
"loss": 6.2373,
"mean_token_accuracy": 0.14902258217334746,
"num_tokens": 5045289.0,
"step": 2705
},
{
"entropy": 6.2602025985717775,
"epoch": 0.2393147297774638,
"grad_norm": 1.03125,
"learning_rate": 0.0004997425839395913,
"loss": 6.2623,
"mean_token_accuracy": 0.14851141721010208,
"num_tokens": 5053774.0,
"step": 2710
},
{
"entropy": 6.4446056365966795,
"epoch": 0.23975626986930412,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997410757876602,
"loss": 6.3368,
"mean_token_accuracy": 0.13768139705061913,
"num_tokens": 5062911.0,
"step": 2715
},
{
"entropy": 6.34313178062439,
"epoch": 0.24019780996114448,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997395632331793,
"loss": 6.1974,
"mean_token_accuracy": 0.15056394785642624,
"num_tokens": 5072107.0,
"step": 2720
},
{
"entropy": 6.232982730865478,
"epoch": 0.2406393500529848,
"grad_norm": 1.078125,
"learning_rate": 0.0004997380462761781,
"loss": 6.1744,
"mean_token_accuracy": 0.15013156086206436,
"num_tokens": 5080588.0,
"step": 2725
},
{
"entropy": 6.360136985778809,
"epoch": 0.24108089014482514,
"grad_norm": 1.046875,
"learning_rate": 0.0004997365249166864,
"loss": 6.3571,
"mean_token_accuracy": 0.1455472856760025,
"num_tokens": 5090262.0,
"step": 2730
},
{
"entropy": 6.398047304153442,
"epoch": 0.2415224302366655,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997349991547342,
"loss": 6.2776,
"mean_token_accuracy": 0.15021264627575875,
"num_tokens": 5099285.0,
"step": 2735
},
{
"entropy": 6.356108903884888,
"epoch": 0.24196397032850583,
"grad_norm": 1.0625,
"learning_rate": 0.0004997334689903509,
"loss": 6.3226,
"mean_token_accuracy": 0.14855852872133254,
"num_tokens": 5109115.0,
"step": 2740
},
{
"entropy": 6.388672256469727,
"epoch": 0.24240551042034617,
"grad_norm": 1.2265625,
"learning_rate": 0.0004997319344235668,
"loss": 6.3429,
"mean_token_accuracy": 0.14180680066347123,
"num_tokens": 5117977.0,
"step": 2745
},
{
"entropy": 6.393019914627075,
"epoch": 0.2428470505121865,
"grad_norm": 1.015625,
"learning_rate": 0.000499730395454412,
"loss": 6.3025,
"mean_token_accuracy": 0.1451731264591217,
"num_tokens": 5127457.0,
"step": 2750
},
{
"entropy": 6.300089502334595,
"epoch": 0.24328859060402686,
"grad_norm": 1.0625,
"learning_rate": 0.0004997288520829166,
"loss": 6.3466,
"mean_token_accuracy": 0.14065672382712363,
"num_tokens": 5137310.0,
"step": 2755
},
{
"entropy": 6.394161605834961,
"epoch": 0.2437301306958672,
"grad_norm": 1.0625,
"learning_rate": 0.0004997273043091107,
"loss": 6.2725,
"mean_token_accuracy": 0.14155926927924156,
"num_tokens": 5146963.0,
"step": 2760
},
{
"entropy": 6.31079797744751,
"epoch": 0.24417167078770752,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997257521330248,
"loss": 6.2601,
"mean_token_accuracy": 0.14022860154509545,
"num_tokens": 5155521.0,
"step": 2765
},
{
"entropy": 6.40408935546875,
"epoch": 0.24461321087954788,
"grad_norm": 1.0625,
"learning_rate": 0.0004997241955546892,
"loss": 6.237,
"mean_token_accuracy": 0.14580907300114632,
"num_tokens": 5165182.0,
"step": 2770
},
{
"entropy": 6.289978647232056,
"epoch": 0.2450547509713882,
"grad_norm": 1.0,
"learning_rate": 0.0004997226345741343,
"loss": 6.2649,
"mean_token_accuracy": 0.13975519686937332,
"num_tokens": 5175511.0,
"step": 2775
},
{
"entropy": 6.354159593582153,
"epoch": 0.24549629106322854,
"grad_norm": 1.1015625,
"learning_rate": 0.000499721069191391,
"loss": 6.2126,
"mean_token_accuracy": 0.14699607565999032,
"num_tokens": 5184772.0,
"step": 2780
},
{
"entropy": 6.288126516342163,
"epoch": 0.24593783115506887,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997194994064896,
"loss": 6.3014,
"mean_token_accuracy": 0.14003223031759263,
"num_tokens": 5195136.0,
"step": 2785
},
{
"entropy": 6.391946315765381,
"epoch": 0.24637937124690923,
"grad_norm": 1.140625,
"learning_rate": 0.000499717925219461,
"loss": 6.2279,
"mean_token_accuracy": 0.15093553364276885,
"num_tokens": 5203163.0,
"step": 2790
},
{
"entropy": 6.346267318725586,
"epoch": 0.24682091133874956,
"grad_norm": 1.015625,
"learning_rate": 0.0004997163466303362,
"loss": 6.3224,
"mean_token_accuracy": 0.14383373707532882,
"num_tokens": 5213233.0,
"step": 2795
},
{
"entropy": 6.329722881317139,
"epoch": 0.2472624514305899,
"grad_norm": 0.96875,
"learning_rate": 0.000499714763639146,
"loss": 6.2346,
"mean_token_accuracy": 0.14386766105890275,
"num_tokens": 5222940.0,
"step": 2800
},
{
"entropy": 6.36411566734314,
"epoch": 0.24770399152243025,
"grad_norm": 1.125,
"learning_rate": 0.0004997131762459211,
"loss": 6.2596,
"mean_token_accuracy": 0.14432956129312516,
"num_tokens": 5232263.0,
"step": 2805
},
{
"entropy": 6.369926023483276,
"epoch": 0.24814553161427058,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997115844506932,
"loss": 6.2334,
"mean_token_accuracy": 0.14295720756053926,
"num_tokens": 5241536.0,
"step": 2810
},
{
"entropy": 6.348195028305054,
"epoch": 0.2485870717061109,
"grad_norm": 1.078125,
"learning_rate": 0.0004997099882534929,
"loss": 6.2732,
"mean_token_accuracy": 0.14211497604846954,
"num_tokens": 5250702.0,
"step": 2815
},
{
"entropy": 6.359058141708374,
"epoch": 0.24902861179795124,
"grad_norm": 1.015625,
"learning_rate": 0.0004997083876543519,
"loss": 6.2763,
"mean_token_accuracy": 0.14498503208160402,
"num_tokens": 5259811.0,
"step": 2820
},
{
"entropy": 6.397128582000732,
"epoch": 0.2494701518897916,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997067826533014,
"loss": 6.3615,
"mean_token_accuracy": 0.13723283037543296,
"num_tokens": 5270518.0,
"step": 2825
},
{
"entropy": 6.375804328918457,
"epoch": 0.24991169198163193,
"grad_norm": 1.421875,
"learning_rate": 0.0004997051732503726,
"loss": 6.2458,
"mean_token_accuracy": 0.14747673273086548,
"num_tokens": 5279538.0,
"step": 2830
},
{
"entropy": 6.313772678375244,
"epoch": 0.25035323207347226,
"grad_norm": 1.1015625,
"learning_rate": 0.0004997035594455975,
"loss": 6.2702,
"mean_token_accuracy": 0.13872402533888817,
"num_tokens": 5289633.0,
"step": 2835
},
{
"entropy": 6.370833015441894,
"epoch": 0.2507947721653126,
"grad_norm": 1.109375,
"learning_rate": 0.0004997019412390074,
"loss": 6.3603,
"mean_token_accuracy": 0.1444901555776596,
"num_tokens": 5299148.0,
"step": 2840
},
{
"entropy": 6.36065092086792,
"epoch": 0.2512363122571529,
"grad_norm": 1.015625,
"learning_rate": 0.000499700318630634,
"loss": 6.2527,
"mean_token_accuracy": 0.1457889422774315,
"num_tokens": 5309090.0,
"step": 2845
},
{
"entropy": 6.384716939926148,
"epoch": 0.2516778523489933,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996986916205092,
"loss": 6.3297,
"mean_token_accuracy": 0.14129810705780982,
"num_tokens": 5318798.0,
"step": 2850
},
{
"entropy": 6.313976621627807,
"epoch": 0.25211939244083365,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996970602086648,
"loss": 6.1848,
"mean_token_accuracy": 0.15023760497570038,
"num_tokens": 5327915.0,
"step": 2855
},
{
"entropy": 6.251888847351074,
"epoch": 0.252560932532674,
"grad_norm": 1.078125,
"learning_rate": 0.0004996954243951327,
"loss": 6.2192,
"mean_token_accuracy": 0.15385923832654952,
"num_tokens": 5336970.0,
"step": 2860
},
{
"entropy": 6.299870014190674,
"epoch": 0.2530024726245143,
"grad_norm": 1.0625,
"learning_rate": 0.0004996937841799451,
"loss": 6.1821,
"mean_token_accuracy": 0.15226729065179825,
"num_tokens": 5345167.0,
"step": 2865
},
{
"entropy": 6.1627833366394045,
"epoch": 0.25344401271635464,
"grad_norm": 1.28125,
"learning_rate": 0.0004996921395631342,
"loss": 6.1711,
"mean_token_accuracy": 0.14804726019501685,
"num_tokens": 5353399.0,
"step": 2870
},
{
"entropy": 6.380198335647583,
"epoch": 0.25388555280819497,
"grad_norm": 1.046875,
"learning_rate": 0.000499690490544732,
"loss": 6.285,
"mean_token_accuracy": 0.146748573333025,
"num_tokens": 5363224.0,
"step": 2875
},
{
"entropy": 6.372404766082764,
"epoch": 0.2543270929000353,
"grad_norm": 1.015625,
"learning_rate": 0.0004996888371247707,
"loss": 6.2862,
"mean_token_accuracy": 0.14108646661043167,
"num_tokens": 5372274.0,
"step": 2880
},
{
"entropy": 6.302175998687744,
"epoch": 0.2547686329918757,
"grad_norm": 1.0390625,
"learning_rate": 0.000499687179303283,
"loss": 6.2746,
"mean_token_accuracy": 0.15104661732912064,
"num_tokens": 5380240.0,
"step": 2885
},
{
"entropy": 6.269204902648926,
"epoch": 0.255210173083716,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996855170803012,
"loss": 6.138,
"mean_token_accuracy": 0.15041064321994782,
"num_tokens": 5389390.0,
"step": 2890
},
{
"entropy": 6.343308639526367,
"epoch": 0.25565171317555635,
"grad_norm": 1.03125,
"learning_rate": 0.0004996838504558581,
"loss": 6.2986,
"mean_token_accuracy": 0.14492825120687486,
"num_tokens": 5399425.0,
"step": 2895
},
{
"entropy": 6.379653215408325,
"epoch": 0.2560932532673967,
"grad_norm": 1.1640625,
"learning_rate": 0.000499682179429986,
"loss": 6.3089,
"mean_token_accuracy": 0.14139388352632523,
"num_tokens": 5408717.0,
"step": 2900
},
{
"entropy": 6.292103576660156,
"epoch": 0.256534793359237,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996805040027178,
"loss": 6.2399,
"mean_token_accuracy": 0.1403766691684723,
"num_tokens": 5418475.0,
"step": 2905
},
{
"entropy": 6.395513296127319,
"epoch": 0.25697633345107734,
"grad_norm": 1.0625,
"learning_rate": 0.0004996788241740863,
"loss": 6.2884,
"mean_token_accuracy": 0.143946073949337,
"num_tokens": 5428403.0,
"step": 2910
},
{
"entropy": 6.366812467575073,
"epoch": 0.2574178735429177,
"grad_norm": 1.0625,
"learning_rate": 0.0004996771399441243,
"loss": 6.3188,
"mean_token_accuracy": 0.14125285297632217,
"num_tokens": 5437347.0,
"step": 2915
},
{
"entropy": 6.395107555389404,
"epoch": 0.25785941363475806,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996754513128652,
"loss": 6.2216,
"mean_token_accuracy": 0.1553879424929619,
"num_tokens": 5446804.0,
"step": 2920
},
{
"entropy": 6.245992279052734,
"epoch": 0.2583009537265984,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996737582803416,
"loss": 6.1701,
"mean_token_accuracy": 0.14774591475725174,
"num_tokens": 5455888.0,
"step": 2925
},
{
"entropy": 6.349690961837768,
"epoch": 0.2587424938184387,
"grad_norm": 1.0859375,
"learning_rate": 0.0004996720608465868,
"loss": 6.1785,
"mean_token_accuracy": 0.14454589933156967,
"num_tokens": 5463977.0,
"step": 2930
},
{
"entropy": 6.251695680618286,
"epoch": 0.25918403391027905,
"grad_norm": 0.953125,
"learning_rate": 0.0004996703590116342,
"loss": 6.2901,
"mean_token_accuracy": 0.1413638859987259,
"num_tokens": 5473780.0,
"step": 2935
},
{
"entropy": 6.343139219284057,
"epoch": 0.2596255740021194,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996686527755171,
"loss": 6.1747,
"mean_token_accuracy": 0.15054369121789932,
"num_tokens": 5482151.0,
"step": 2940
},
{
"entropy": 6.287330961227417,
"epoch": 0.2600671140939597,
"grad_norm": 1.015625,
"learning_rate": 0.0004996669421382687,
"loss": 6.181,
"mean_token_accuracy": 0.15408090725541115,
"num_tokens": 5491103.0,
"step": 2945
},
{
"entropy": 6.23843822479248,
"epoch": 0.26050865418580005,
"grad_norm": 1.25,
"learning_rate": 0.0004996652270999228,
"loss": 6.2051,
"mean_token_accuracy": 0.1455566719174385,
"num_tokens": 5500367.0,
"step": 2950
},
{
"entropy": 6.401996898651123,
"epoch": 0.26095019427764043,
"grad_norm": 1.125,
"learning_rate": 0.0004996635076605128,
"loss": 6.2392,
"mean_token_accuracy": 0.1509515941143036,
"num_tokens": 5509631.0,
"step": 2955
},
{
"entropy": 6.3384003162384035,
"epoch": 0.26139173436948077,
"grad_norm": 1.15625,
"learning_rate": 0.0004996617838200725,
"loss": 6.2572,
"mean_token_accuracy": 0.14331620335578918,
"num_tokens": 5518635.0,
"step": 2960
},
{
"entropy": 6.241027069091797,
"epoch": 0.2618332744613211,
"grad_norm": 1.125,
"learning_rate": 0.0004996600555786357,
"loss": 6.2142,
"mean_token_accuracy": 0.1464727446436882,
"num_tokens": 5527696.0,
"step": 2965
},
{
"entropy": 6.348132085800171,
"epoch": 0.26227481455316143,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996583229362362,
"loss": 6.1834,
"mean_token_accuracy": 0.14780823439359664,
"num_tokens": 5536632.0,
"step": 2970
},
{
"entropy": 6.378821849822998,
"epoch": 0.26271635464500176,
"grad_norm": 1.1171875,
"learning_rate": 0.0004996565858929078,
"loss": 6.2627,
"mean_token_accuracy": 0.14528179541230202,
"num_tokens": 5545825.0,
"step": 2975
},
{
"entropy": 6.259585618972778,
"epoch": 0.2631578947368421,
"grad_norm": 1.03125,
"learning_rate": 0.0004996548444486847,
"loss": 6.1389,
"mean_token_accuracy": 0.15060140788555146,
"num_tokens": 5555158.0,
"step": 2980
},
{
"entropy": 6.116889953613281,
"epoch": 0.2635994348286824,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996530986036008,
"loss": 6.0795,
"mean_token_accuracy": 0.15272270664572715,
"num_tokens": 5564218.0,
"step": 2985
},
{
"entropy": 6.255494451522827,
"epoch": 0.2640409749205228,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996513483576907,
"loss": 6.2219,
"mean_token_accuracy": 0.14951637461781503,
"num_tokens": 5572760.0,
"step": 2990
},
{
"entropy": 6.423755645751953,
"epoch": 0.26448251501236314,
"grad_norm": 1.09375,
"learning_rate": 0.0004996495937109884,
"loss": 6.2825,
"mean_token_accuracy": 0.14191085398197173,
"num_tokens": 5581660.0,
"step": 2995
},
{
"entropy": 6.250067615509034,
"epoch": 0.26492405510420347,
"grad_norm": 1.046875,
"learning_rate": 0.0004996478346635283,
"loss": 6.1968,
"mean_token_accuracy": 0.1436440147459507,
"num_tokens": 5590664.0,
"step": 3000
},
{
"epoch": 0.26492405510420347,
"eval_entropy": 6.077911184598204,
"eval_loss": 6.2711029052734375,
"eval_mean_token_accuracy": 0.15016848111384373,
"eval_num_tokens": 5590664.0,
"eval_runtime": 26.2453,
"eval_samples_per_second": 1345.574,
"eval_steps_per_second": 168.221,
"step": 3000
},
{
"entropy": 6.211904573440552,
"epoch": 0.2653655951960438,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996460712153448,
"loss": 6.0603,
"mean_token_accuracy": 0.1621384307742119,
"num_tokens": 5598727.0,
"step": 3005
},
{
"entropy": 6.220605993270874,
"epoch": 0.26580713528788413,
"grad_norm": 1.3984375,
"learning_rate": 0.0004996443033664726,
"loss": 6.2746,
"mean_token_accuracy": 0.1402523137629032,
"num_tokens": 5608549.0,
"step": 3010
},
{
"entropy": 6.372675228118896,
"epoch": 0.26624867537972446,
"grad_norm": 1.125,
"learning_rate": 0.0004996425311169463,
"loss": 6.1935,
"mean_token_accuracy": 0.15231838524341584,
"num_tokens": 5617628.0,
"step": 3015
},
{
"entropy": 6.362924861907959,
"epoch": 0.2666902154715648,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996407544668005,
"loss": 6.2649,
"mean_token_accuracy": 0.14236303716897963,
"num_tokens": 5627182.0,
"step": 3020
},
{
"entropy": 6.325074291229248,
"epoch": 0.2671317555634052,
"grad_norm": 1.109375,
"learning_rate": 0.0004996389734160701,
"loss": 6.293,
"mean_token_accuracy": 0.14348414838314055,
"num_tokens": 5636413.0,
"step": 3025
},
{
"entropy": 6.389656829833984,
"epoch": 0.2675732956552455,
"grad_norm": 1.078125,
"learning_rate": 0.00049963718796479,
"loss": 6.2355,
"mean_token_accuracy": 0.14448917284607887,
"num_tokens": 5646025.0,
"step": 3030
},
{
"entropy": 6.272923135757447,
"epoch": 0.26801483574708584,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996353981129952,
"loss": 6.2316,
"mean_token_accuracy": 0.15121424347162246,
"num_tokens": 5656577.0,
"step": 3035
},
{
"entropy": 6.341066169738769,
"epoch": 0.2684563758389262,
"grad_norm": 1.09375,
"learning_rate": 0.0004996336038607206,
"loss": 6.2194,
"mean_token_accuracy": 0.1483485922217369,
"num_tokens": 5665163.0,
"step": 3040
},
{
"entropy": 6.198571634292603,
"epoch": 0.2688979159307665,
"grad_norm": 1.09375,
"learning_rate": 0.0004996318052080015,
"loss": 6.2208,
"mean_token_accuracy": 0.14988780170679092,
"num_tokens": 5675196.0,
"step": 3045
},
{
"entropy": 6.395411729812622,
"epoch": 0.26933945602260684,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996300021548731,
"loss": 6.163,
"mean_token_accuracy": 0.14734217673540115,
"num_tokens": 5684361.0,
"step": 3050
},
{
"entropy": 6.142444658279419,
"epoch": 0.26978099611444717,
"grad_norm": 1.03125,
"learning_rate": 0.0004996281947013707,
"loss": 6.1584,
"mean_token_accuracy": 0.1533804029226303,
"num_tokens": 5694187.0,
"step": 3055
},
{
"entropy": 6.307489728927612,
"epoch": 0.27022253620628756,
"grad_norm": 1.2265625,
"learning_rate": 0.0004996263828475298,
"loss": 6.2235,
"mean_token_accuracy": 0.1459271177649498,
"num_tokens": 5702858.0,
"step": 3060
},
{
"entropy": 6.287201023101806,
"epoch": 0.2706640762981279,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996245665933857,
"loss": 6.124,
"mean_token_accuracy": 0.1559446483850479,
"num_tokens": 5712727.0,
"step": 3065
},
{
"entropy": 6.197343635559082,
"epoch": 0.2711056163899682,
"grad_norm": 1.1484375,
"learning_rate": 0.0004996227459389741,
"loss": 6.2573,
"mean_token_accuracy": 0.14863042607903482,
"num_tokens": 5722904.0,
"step": 3070
},
{
"entropy": 6.400126695632935,
"epoch": 0.27154715648180855,
"grad_norm": 1.0859375,
"learning_rate": 0.0004996209208843307,
"loss": 6.3578,
"mean_token_accuracy": 0.14644555673003196,
"num_tokens": 5732228.0,
"step": 3075
},
{
"entropy": 6.322457504272461,
"epoch": 0.2719886965736489,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996190914294912,
"loss": 6.2574,
"mean_token_accuracy": 0.1481010966002941,
"num_tokens": 5743548.0,
"step": 3080
},
{
"entropy": 6.271878862380982,
"epoch": 0.2724302366654892,
"grad_norm": 1.015625,
"learning_rate": 0.0004996172575744914,
"loss": 6.2867,
"mean_token_accuracy": 0.13876855000853539,
"num_tokens": 5754112.0,
"step": 3085
},
{
"entropy": 6.337986612319947,
"epoch": 0.27287177675732954,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996154193193673,
"loss": 6.0993,
"mean_token_accuracy": 0.15423450618982315,
"num_tokens": 5763377.0,
"step": 3090
},
{
"entropy": 6.137074375152588,
"epoch": 0.27331331684916993,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996135766641549,
"loss": 6.0806,
"mean_token_accuracy": 0.15594624429941178,
"num_tokens": 5772000.0,
"step": 3095
},
{
"entropy": 6.32679500579834,
"epoch": 0.27375485694101026,
"grad_norm": 1.078125,
"learning_rate": 0.0004996117296088903,
"loss": 6.2131,
"mean_token_accuracy": 0.14269956201314926,
"num_tokens": 5782192.0,
"step": 3100
},
{
"entropy": 6.341950845718384,
"epoch": 0.2741963970328506,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996098781536095,
"loss": 6.2614,
"mean_token_accuracy": 0.1522984981536865,
"num_tokens": 5791163.0,
"step": 3105
},
{
"entropy": 6.3308337211608885,
"epoch": 0.2746379371246909,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996080222983492,
"loss": 6.259,
"mean_token_accuracy": 0.14087159857153891,
"num_tokens": 5801815.0,
"step": 3110
},
{
"entropy": 6.410363006591797,
"epoch": 0.27507947721653125,
"grad_norm": 1.078125,
"learning_rate": 0.0004996061620431453,
"loss": 6.2805,
"mean_token_accuracy": 0.1418459102511406,
"num_tokens": 5811545.0,
"step": 3115
},
{
"entropy": 6.3260101795196535,
"epoch": 0.2755210173083716,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996042973880344,
"loss": 6.2846,
"mean_token_accuracy": 0.14676327556371688,
"num_tokens": 5820285.0,
"step": 3120
},
{
"entropy": 6.260831451416015,
"epoch": 0.2759625574002119,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996024283330532,
"loss": 6.1242,
"mean_token_accuracy": 0.14824864715337754,
"num_tokens": 5829284.0,
"step": 3125
},
{
"entropy": 6.268897342681885,
"epoch": 0.2764040974920523,
"grad_norm": 1.03125,
"learning_rate": 0.000499600554878238,
"loss": 6.2951,
"mean_token_accuracy": 0.1431872047483921,
"num_tokens": 5838973.0,
"step": 3130
},
{
"entropy": 6.278724241256714,
"epoch": 0.27684563758389263,
"grad_norm": 1.2578125,
"learning_rate": 0.0004995986770236258,
"loss": 6.1837,
"mean_token_accuracy": 0.15050409361720085,
"num_tokens": 5847434.0,
"step": 3135
},
{
"entropy": 6.304096031188965,
"epoch": 0.27728717767573297,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995967947692533,
"loss": 6.1808,
"mean_token_accuracy": 0.14390211701393127,
"num_tokens": 5856464.0,
"step": 3140
},
{
"entropy": 6.311007452011109,
"epoch": 0.2777287177675733,
"grad_norm": 1.1171875,
"learning_rate": 0.0004995949081151571,
"loss": 6.2402,
"mean_token_accuracy": 0.14435049369931222,
"num_tokens": 5865892.0,
"step": 3145
},
{
"entropy": 6.353293609619141,
"epoch": 0.2781702578594136,
"grad_norm": 1.0625,
"learning_rate": 0.0004995930170613746,
"loss": 6.2075,
"mean_token_accuracy": 0.1576365649700165,
"num_tokens": 5874606.0,
"step": 3150
},
{
"entropy": 6.275713014602661,
"epoch": 0.27861179795125396,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995911216079425,
"loss": 6.2116,
"mean_token_accuracy": 0.15098029375076294,
"num_tokens": 5883853.0,
"step": 3155
},
{
"entropy": 6.34206337928772,
"epoch": 0.2790533380430943,
"grad_norm": 1.1484375,
"learning_rate": 0.0004995892217548981,
"loss": 6.2474,
"mean_token_accuracy": 0.1448906570672989,
"num_tokens": 5892841.0,
"step": 3160
},
{
"entropy": 6.324929285049438,
"epoch": 0.2794948781349347,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995873175022786,
"loss": 6.2295,
"mean_token_accuracy": 0.15006719902157784,
"num_tokens": 5901966.0,
"step": 3165
},
{
"entropy": 6.260264158248901,
"epoch": 0.279936418226775,
"grad_norm": 1.109375,
"learning_rate": 0.0004995854088501213,
"loss": 6.1281,
"mean_token_accuracy": 0.14603292495012282,
"num_tokens": 5910003.0,
"step": 3170
},
{
"entropy": 6.189469957351685,
"epoch": 0.28037795831861534,
"grad_norm": 1.0,
"learning_rate": 0.0004995834957984634,
"loss": 6.2606,
"mean_token_accuracy": 0.14485765993595123,
"num_tokens": 5920307.0,
"step": 3175
},
{
"entropy": 6.290289306640625,
"epoch": 0.28081949841045567,
"grad_norm": 1.015625,
"learning_rate": 0.0004995815783473428,
"loss": 6.1153,
"mean_token_accuracy": 0.15536403357982637,
"num_tokens": 5929875.0,
"step": 3180
},
{
"entropy": 6.247414398193359,
"epoch": 0.281261038502296,
"grad_norm": 1.1953125,
"learning_rate": 0.0004995796564967967,
"loss": 6.1021,
"mean_token_accuracy": 0.14264860302209853,
"num_tokens": 5938570.0,
"step": 3185
},
{
"entropy": 6.209772968292237,
"epoch": 0.28170257859413633,
"grad_norm": 1.078125,
"learning_rate": 0.0004995777302468628,
"loss": 6.2353,
"mean_token_accuracy": 0.14704401940107345,
"num_tokens": 5947693.0,
"step": 3190
},
{
"entropy": 6.332046413421631,
"epoch": 0.28214411868597666,
"grad_norm": 1.1171875,
"learning_rate": 0.0004995757995975789,
"loss": 6.3286,
"mean_token_accuracy": 0.14671236276626587,
"num_tokens": 5957377.0,
"step": 3195
},
{
"entropy": 6.426817464828491,
"epoch": 0.28258565877781705,
"grad_norm": 1.1171875,
"learning_rate": 0.0004995738645489828,
"loss": 6.2285,
"mean_token_accuracy": 0.14824536591768264,
"num_tokens": 5966443.0,
"step": 3200
},
{
"entropy": 6.258016681671142,
"epoch": 0.2830271988696574,
"grad_norm": 1.1875,
"learning_rate": 0.0004995719251011124,
"loss": 6.1471,
"mean_token_accuracy": 0.14848615527153014,
"num_tokens": 5975027.0,
"step": 3205
},
{
"entropy": 6.288780975341797,
"epoch": 0.2834687389614977,
"grad_norm": 1.1171875,
"learning_rate": 0.0004995699812540058,
"loss": 6.2244,
"mean_token_accuracy": 0.15162927508354188,
"num_tokens": 5983722.0,
"step": 3210
},
{
"entropy": 6.282262182235717,
"epoch": 0.28391027905333804,
"grad_norm": 0.9921875,
"learning_rate": 0.000499568033007701,
"loss": 6.2324,
"mean_token_accuracy": 0.1441471680998802,
"num_tokens": 5993358.0,
"step": 3215
},
{
"entropy": 6.377847146987915,
"epoch": 0.2843518191451784,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995660803622361,
"loss": 6.161,
"mean_token_accuracy": 0.1480330415070057,
"num_tokens": 6002743.0,
"step": 3220
},
{
"entropy": 6.323432493209839,
"epoch": 0.2847933592370187,
"grad_norm": 1.09375,
"learning_rate": 0.0004995641233176494,
"loss": 6.3205,
"mean_token_accuracy": 0.14193187803030013,
"num_tokens": 6013727.0,
"step": 3225
},
{
"entropy": 6.404472589492798,
"epoch": 0.28523489932885904,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995621618739792,
"loss": 6.2652,
"mean_token_accuracy": 0.14536072462797164,
"num_tokens": 6023397.0,
"step": 3230
},
{
"entropy": 6.271181678771972,
"epoch": 0.2856764394206994,
"grad_norm": 1.0625,
"learning_rate": 0.000499560196031264,
"loss": 6.1638,
"mean_token_accuracy": 0.15034203678369523,
"num_tokens": 6032453.0,
"step": 3235
},
{
"entropy": 6.19854884147644,
"epoch": 0.28611797951253976,
"grad_norm": 0.96875,
"learning_rate": 0.0004995582257895423,
"loss": 6.1258,
"mean_token_accuracy": 0.14395371079444885,
"num_tokens": 6042441.0,
"step": 3240
},
{
"entropy": 6.274320507049561,
"epoch": 0.2865595196043801,
"grad_norm": 1.140625,
"learning_rate": 0.0004995562511488528,
"loss": 6.1197,
"mean_token_accuracy": 0.15483347177505494,
"num_tokens": 6050881.0,
"step": 3245
},
{
"entropy": 6.220722341537476,
"epoch": 0.2870010596962204,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995542721092337,
"loss": 6.1872,
"mean_token_accuracy": 0.15040701180696486,
"num_tokens": 6060804.0,
"step": 3250
},
{
"entropy": 6.368199014663697,
"epoch": 0.28744259978806075,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995522886707244,
"loss": 6.2847,
"mean_token_accuracy": 0.14250023737549783,
"num_tokens": 6070774.0,
"step": 3255
},
{
"entropy": 6.330855846405029,
"epoch": 0.2878841398799011,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995503008333634,
"loss": 6.2491,
"mean_token_accuracy": 0.14379776269197464,
"num_tokens": 6080158.0,
"step": 3260
},
{
"entropy": 6.32779221534729,
"epoch": 0.2883256799717414,
"grad_norm": 1.09375,
"learning_rate": 0.0004995483085971897,
"loss": 6.16,
"mean_token_accuracy": 0.15112278908491134,
"num_tokens": 6089183.0,
"step": 3265
},
{
"entropy": 6.200159549713135,
"epoch": 0.2887672200635818,
"grad_norm": 1.078125,
"learning_rate": 0.0004995463119622424,
"loss": 6.2524,
"mean_token_accuracy": 0.14461245387792587,
"num_tokens": 6098536.0,
"step": 3270
},
{
"entropy": 6.196738433837891,
"epoch": 0.28920876015542213,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995443109285604,
"loss": 6.0112,
"mean_token_accuracy": 0.1629092276096344,
"num_tokens": 6107745.0,
"step": 3275
},
{
"entropy": 6.289895725250244,
"epoch": 0.28965030024726246,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995423054961832,
"loss": 6.2158,
"mean_token_accuracy": 0.15729496926069259,
"num_tokens": 6117512.0,
"step": 3280
},
{
"entropy": 6.227946424484253,
"epoch": 0.2900918403391028,
"grad_norm": 1.078125,
"learning_rate": 0.00049954029566515,
"loss": 6.2178,
"mean_token_accuracy": 0.15285916179418563,
"num_tokens": 6126030.0,
"step": 3285
},
{
"entropy": 6.346371364593506,
"epoch": 0.2905333804309431,
"grad_norm": 0.98828125,
"learning_rate": 0.0004995382814355,
"loss": 6.2625,
"mean_token_accuracy": 0.14053603783249854,
"num_tokens": 6134888.0,
"step": 3290
},
{
"entropy": 6.351688432693481,
"epoch": 0.29097492052278345,
"grad_norm": 1.09375,
"learning_rate": 0.0004995362628072728,
"loss": 6.2117,
"mean_token_accuracy": 0.1513692669570446,
"num_tokens": 6144274.0,
"step": 3295
},
{
"entropy": 6.218261671066284,
"epoch": 0.2914164606146238,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995342397805078,
"loss": 6.2217,
"mean_token_accuracy": 0.1503726065158844,
"num_tokens": 6153406.0,
"step": 3300
},
{
"entropy": 6.262264680862427,
"epoch": 0.29185800070646417,
"grad_norm": 1.0859375,
"learning_rate": 0.0004995322123552448,
"loss": 6.1233,
"mean_token_accuracy": 0.1533094823360443,
"num_tokens": 6162743.0,
"step": 3305
},
{
"entropy": 6.2753763675689695,
"epoch": 0.2922995407983045,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995301805315235,
"loss": 6.1201,
"mean_token_accuracy": 0.15415377020835877,
"num_tokens": 6171997.0,
"step": 3310
},
{
"entropy": 6.181772375106812,
"epoch": 0.29274108089014483,
"grad_norm": 1.1328125,
"learning_rate": 0.0004995281443093837,
"loss": 6.1422,
"mean_token_accuracy": 0.15062253326177596,
"num_tokens": 6181275.0,
"step": 3315
},
{
"entropy": 6.289641571044922,
"epoch": 0.29318262098198516,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995261036888652,
"loss": 6.2135,
"mean_token_accuracy": 0.14647497087717057,
"num_tokens": 6191640.0,
"step": 3320
},
{
"entropy": 6.357338809967041,
"epoch": 0.2936241610738255,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995240586700081,
"loss": 6.2137,
"mean_token_accuracy": 0.14479927867650985,
"num_tokens": 6201508.0,
"step": 3325
},
{
"entropy": 6.220925617218017,
"epoch": 0.2940657011656658,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995220092528522,
"loss": 6.1079,
"mean_token_accuracy": 0.15776659697294235,
"num_tokens": 6209902.0,
"step": 3330
},
{
"entropy": 6.257762432098389,
"epoch": 0.29450724125750616,
"grad_norm": 1.0546875,
"learning_rate": 0.000499519955437438,
"loss": 6.2514,
"mean_token_accuracy": 0.14023935049772263,
"num_tokens": 6219760.0,
"step": 3335
},
{
"entropy": 6.248654699325561,
"epoch": 0.29494878134934654,
"grad_norm": 1.109375,
"learning_rate": 0.0004995178972238054,
"loss": 6.2308,
"mean_token_accuracy": 0.14278148710727692,
"num_tokens": 6228721.0,
"step": 3340
},
{
"entropy": 6.228511571884155,
"epoch": 0.2953903214411869,
"grad_norm": 1.1171875,
"learning_rate": 0.000499515834611995,
"loss": 6.0645,
"mean_token_accuracy": 0.16055997535586358,
"num_tokens": 6237070.0,
"step": 3345
},
{
"entropy": 6.2931300640106205,
"epoch": 0.2958318615330272,
"grad_norm": 1.046875,
"learning_rate": 0.0004995137676020472,
"loss": 6.2051,
"mean_token_accuracy": 0.1466339647769928,
"num_tokens": 6245659.0,
"step": 3350
},
{
"entropy": 6.231089639663696,
"epoch": 0.29627340162486754,
"grad_norm": 1.09375,
"learning_rate": 0.0004995116961940023,
"loss": 6.1736,
"mean_token_accuracy": 0.15321153849363328,
"num_tokens": 6255175.0,
"step": 3355
},
{
"entropy": 6.241381883621216,
"epoch": 0.29671494171670787,
"grad_norm": 0.96875,
"learning_rate": 0.0004995096203879009,
"loss": 6.1761,
"mean_token_accuracy": 0.1443149983882904,
"num_tokens": 6264962.0,
"step": 3360
},
{
"entropy": 6.29106593132019,
"epoch": 0.2971564818085482,
"grad_norm": 1.09375,
"learning_rate": 0.0004995075401837837,
"loss": 6.0989,
"mean_token_accuracy": 0.15684471875429154,
"num_tokens": 6273411.0,
"step": 3365
},
{
"entropy": 6.135526657104492,
"epoch": 0.29759802190038853,
"grad_norm": 1.078125,
"learning_rate": 0.0004995054555816915,
"loss": 6.0795,
"mean_token_accuracy": 0.15481019616127015,
"num_tokens": 6282618.0,
"step": 3370
},
{
"entropy": 6.155969190597534,
"epoch": 0.2980395619922289,
"grad_norm": 1.0625,
"learning_rate": 0.0004995033665816651,
"loss": 6.0726,
"mean_token_accuracy": 0.15434197783470155,
"num_tokens": 6292008.0,
"step": 3375
},
{
"entropy": 6.147941923141479,
"epoch": 0.29848110208406925,
"grad_norm": 1.1328125,
"learning_rate": 0.0004995012731837454,
"loss": 6.0441,
"mean_token_accuracy": 0.15588683634996414,
"num_tokens": 6301079.0,
"step": 3380
},
{
"entropy": 6.323877429962158,
"epoch": 0.2989226421759096,
"grad_norm": 1.046875,
"learning_rate": 0.0004994991753879736,
"loss": 6.2273,
"mean_token_accuracy": 0.14481185078620912,
"num_tokens": 6310543.0,
"step": 3385
},
{
"entropy": 6.276103544235229,
"epoch": 0.2993641822677499,
"grad_norm": 1.296875,
"learning_rate": 0.0004994970731943904,
"loss": 6.1692,
"mean_token_accuracy": 0.1476268857717514,
"num_tokens": 6320037.0,
"step": 3390
},
{
"entropy": 6.177714109420776,
"epoch": 0.29980572235959024,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994949666030374,
"loss": 6.1795,
"mean_token_accuracy": 0.15479264855384828,
"num_tokens": 6328948.0,
"step": 3395
},
{
"entropy": 6.212888717651367,
"epoch": 0.3002472624514306,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994928556139557,
"loss": 6.0425,
"mean_token_accuracy": 0.1581359773874283,
"num_tokens": 6337449.0,
"step": 3400
},
{
"entropy": 6.162185001373291,
"epoch": 0.3006888025432709,
"grad_norm": 1.109375,
"learning_rate": 0.0004994907402271865,
"loss": 6.1028,
"mean_token_accuracy": 0.15499115511775016,
"num_tokens": 6346130.0,
"step": 3405
},
{
"entropy": 6.15036244392395,
"epoch": 0.3011303426351113,
"grad_norm": 1.1484375,
"learning_rate": 0.0004994886204427715,
"loss": 6.1344,
"mean_token_accuracy": 0.15851569175720215,
"num_tokens": 6354973.0,
"step": 3410
},
{
"entropy": 6.328536748886108,
"epoch": 0.3015718827269516,
"grad_norm": 1.1328125,
"learning_rate": 0.0004994864962607519,
"loss": 6.2148,
"mean_token_accuracy": 0.14824963808059693,
"num_tokens": 6364107.0,
"step": 3415
},
{
"entropy": 6.26984658241272,
"epoch": 0.30201342281879195,
"grad_norm": 1.1484375,
"learning_rate": 0.0004994843676811697,
"loss": 6.1438,
"mean_token_accuracy": 0.14578014612197876,
"num_tokens": 6372859.0,
"step": 3420
},
{
"entropy": 6.184666395187378,
"epoch": 0.3024549629106323,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994822347040664,
"loss": 6.0282,
"mean_token_accuracy": 0.1636571153998375,
"num_tokens": 6381818.0,
"step": 3425
},
{
"entropy": 6.257387351989746,
"epoch": 0.3028965030024726,
"grad_norm": 1.09375,
"learning_rate": 0.0004994800973294837,
"loss": 6.1797,
"mean_token_accuracy": 0.1482721135020256,
"num_tokens": 6391460.0,
"step": 3430
},
{
"entropy": 6.25637993812561,
"epoch": 0.30333804309431295,
"grad_norm": 1.0625,
"learning_rate": 0.0004994779555574636,
"loss": 6.2086,
"mean_token_accuracy": 0.15168215036392213,
"num_tokens": 6401461.0,
"step": 3435
},
{
"entropy": 6.262447023391724,
"epoch": 0.3037795831861533,
"grad_norm": 1.2265625,
"learning_rate": 0.000499475809388048,
"loss": 6.0842,
"mean_token_accuracy": 0.14643362984061242,
"num_tokens": 6410435.0,
"step": 3440
},
{
"entropy": 6.279139566421509,
"epoch": 0.30422112327799367,
"grad_norm": 1.125,
"learning_rate": 0.000499473658821279,
"loss": 6.2844,
"mean_token_accuracy": 0.14141586795449257,
"num_tokens": 6420710.0,
"step": 3445
},
{
"entropy": 6.305064296722412,
"epoch": 0.304662663369834,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994715038571986,
"loss": 6.1276,
"mean_token_accuracy": 0.149975299090147,
"num_tokens": 6429882.0,
"step": 3450
},
{
"entropy": 6.167029666900635,
"epoch": 0.30510420346167433,
"grad_norm": 1.1171875,
"learning_rate": 0.0004994693444958493,
"loss": 6.1142,
"mean_token_accuracy": 0.15782218649983407,
"num_tokens": 6439183.0,
"step": 3455
},
{
"entropy": 6.307245588302612,
"epoch": 0.30554574355351466,
"grad_norm": 1.0390625,
"learning_rate": 0.000499467180737273,
"loss": 6.288,
"mean_token_accuracy": 0.14309904649853705,
"num_tokens": 6448460.0,
"step": 3460
},
{
"entropy": 6.270366525650024,
"epoch": 0.305987283645355,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994650125815124,
"loss": 6.1656,
"mean_token_accuracy": 0.14977657794952393,
"num_tokens": 6457687.0,
"step": 3465
},
{
"entropy": 6.241862058639526,
"epoch": 0.3064288237371953,
"grad_norm": 1.1484375,
"learning_rate": 0.0004994628400286097,
"loss": 6.1063,
"mean_token_accuracy": 0.15189075246453285,
"num_tokens": 6467057.0,
"step": 3470
},
{
"entropy": 6.209890508651734,
"epoch": 0.30687036382903565,
"grad_norm": 1.1328125,
"learning_rate": 0.0004994606630786078,
"loss": 6.1745,
"mean_token_accuracy": 0.15533267706632614,
"num_tokens": 6477484.0,
"step": 3475
},
{
"entropy": 6.248384809494018,
"epoch": 0.30731190392087604,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994584817315492,
"loss": 6.233,
"mean_token_accuracy": 0.14967372938990592,
"num_tokens": 6488381.0,
"step": 3480
},
{
"entropy": 6.30739917755127,
"epoch": 0.30775344401271637,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994562959874765,
"loss": 6.1277,
"mean_token_accuracy": 0.1523176297545433,
"num_tokens": 6498279.0,
"step": 3485
},
{
"entropy": 6.166545104980469,
"epoch": 0.3081949841045567,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994541058464326,
"loss": 6.1476,
"mean_token_accuracy": 0.14837286472320557,
"num_tokens": 6508008.0,
"step": 3490
},
{
"entropy": 6.321815013885498,
"epoch": 0.30863652419639703,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994519113084605,
"loss": 6.1947,
"mean_token_accuracy": 0.1496584579348564,
"num_tokens": 6517263.0,
"step": 3495
},
{
"entropy": 6.262754631042481,
"epoch": 0.30907806428823736,
"grad_norm": 1.171875,
"learning_rate": 0.0004994497123736029,
"loss": 6.2631,
"mean_token_accuracy": 0.1426304429769516,
"num_tokens": 6527682.0,
"step": 3500
},
{
"entropy": 6.267001819610596,
"epoch": 0.3095196043800777,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994475090419034,
"loss": 6.1298,
"mean_token_accuracy": 0.15242742300033568,
"num_tokens": 6537143.0,
"step": 3505
},
{
"entropy": 6.250511837005615,
"epoch": 0.309961144471918,
"grad_norm": 1.09375,
"learning_rate": 0.0004994453013134047,
"loss": 6.1352,
"mean_token_accuracy": 0.14925057888031007,
"num_tokens": 6546561.0,
"step": 3510
},
{
"entropy": 6.185484886169434,
"epoch": 0.3104026845637584,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994430891881502,
"loss": 6.124,
"mean_token_accuracy": 0.15090959072113036,
"num_tokens": 6555806.0,
"step": 3515
},
{
"entropy": 6.202518653869629,
"epoch": 0.31084422465559874,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994408726661832,
"loss": 6.166,
"mean_token_accuracy": 0.1501057654619217,
"num_tokens": 6564836.0,
"step": 3520
},
{
"entropy": 6.245743179321289,
"epoch": 0.3112857647474391,
"grad_norm": 1.046875,
"learning_rate": 0.0004994386517475472,
"loss": 6.1713,
"mean_token_accuracy": 0.1496300369501114,
"num_tokens": 6574997.0,
"step": 3525
},
{
"entropy": 6.246868562698364,
"epoch": 0.3117273048392794,
"grad_norm": 1.15625,
"learning_rate": 0.0004994364264322856,
"loss": 6.1715,
"mean_token_accuracy": 0.14895060658454895,
"num_tokens": 6584589.0,
"step": 3530
},
{
"entropy": 6.162112808227539,
"epoch": 0.31216884493111974,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994341967204421,
"loss": 6.1199,
"mean_token_accuracy": 0.15284974724054337,
"num_tokens": 6592898.0,
"step": 3535
},
{
"entropy": 6.326304626464844,
"epoch": 0.31261038502296007,
"grad_norm": 0.984375,
"learning_rate": 0.0004994319626120603,
"loss": 6.1268,
"mean_token_accuracy": 0.16072259843349457,
"num_tokens": 6602454.0,
"step": 3540
},
{
"entropy": 6.1308966159820555,
"epoch": 0.3130519251148004,
"grad_norm": 1.1328125,
"learning_rate": 0.0004994297241071841,
"loss": 6.1467,
"mean_token_accuracy": 0.1551404133439064,
"num_tokens": 6612454.0,
"step": 3545
},
{
"entropy": 6.29251446723938,
"epoch": 0.3134934652066408,
"grad_norm": 1.0703125,
"learning_rate": 0.000499427481205857,
"loss": 6.1423,
"mean_token_accuracy": 0.15539032369852065,
"num_tokens": 6621286.0,
"step": 3550
},
{
"entropy": 6.253758335113526,
"epoch": 0.3139350052984811,
"grad_norm": 1.265625,
"learning_rate": 0.0004994252339081234,
"loss": 6.0651,
"mean_token_accuracy": 0.15716515630483627,
"num_tokens": 6629684.0,
"step": 3555
},
{
"entropy": 6.167656517028808,
"epoch": 0.31437654539032145,
"grad_norm": 1.0703125,
"learning_rate": 0.000499422982214027,
"loss": 6.1775,
"mean_token_accuracy": 0.15429823398590087,
"num_tokens": 6638836.0,
"step": 3560
},
{
"entropy": 6.254745101928711,
"epoch": 0.3148180854821618,
"grad_norm": 1.09375,
"learning_rate": 0.0004994207261236121,
"loss": 6.1189,
"mean_token_accuracy": 0.14906142503023148,
"num_tokens": 6647337.0,
"step": 3565
},
{
"entropy": 6.145832633972168,
"epoch": 0.3152596255740021,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994184656369227,
"loss": 6.1026,
"mean_token_accuracy": 0.145336801558733,
"num_tokens": 6657246.0,
"step": 3570
},
{
"entropy": 6.310924911499024,
"epoch": 0.31570116566584244,
"grad_norm": 1.1328125,
"learning_rate": 0.0004994162007540033,
"loss": 6.2505,
"mean_token_accuracy": 0.14196238815784454,
"num_tokens": 6667981.0,
"step": 3575
},
{
"entropy": 6.2675220489501955,
"epoch": 0.3161427057576828,
"grad_norm": 1.140625,
"learning_rate": 0.0004994139314748981,
"loss": 6.0974,
"mean_token_accuracy": 0.15776362121105195,
"num_tokens": 6677462.0,
"step": 3580
},
{
"entropy": 6.107239675521851,
"epoch": 0.31658424584952316,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994116577996517,
"loss": 6.1152,
"mean_token_accuracy": 0.15776521414518357,
"num_tokens": 6687225.0,
"step": 3585
},
{
"entropy": 6.2434509754180905,
"epoch": 0.3170257859413635,
"grad_norm": 1.1171875,
"learning_rate": 0.0004994093797283084,
"loss": 6.2122,
"mean_token_accuracy": 0.1487804666161537,
"num_tokens": 6696345.0,
"step": 3590
},
{
"entropy": 6.313143348693847,
"epoch": 0.3174673260332038,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994070972609132,
"loss": 6.1867,
"mean_token_accuracy": 0.14744829311966895,
"num_tokens": 6706207.0,
"step": 3595
},
{
"entropy": 6.24591326713562,
"epoch": 0.31790886612504415,
"grad_norm": 1.1953125,
"learning_rate": 0.0004994048103975103,
"loss": 6.0904,
"mean_token_accuracy": 0.15303746610879898,
"num_tokens": 6714430.0,
"step": 3600
},
{
"entropy": 6.204165840148926,
"epoch": 0.3183504062168845,
"grad_norm": 1.0703125,
"learning_rate": 0.000499402519138145,
"loss": 6.1197,
"mean_token_accuracy": 0.1557806834578514,
"num_tokens": 6724139.0,
"step": 3605
},
{
"entropy": 6.151504611968994,
"epoch": 0.3187919463087248,
"grad_norm": 1.078125,
"learning_rate": 0.0004994002234828619,
"loss": 6.1566,
"mean_token_accuracy": 0.1508852459490299,
"num_tokens": 6733525.0,
"step": 3610
},
{
"entropy": 6.321462392807007,
"epoch": 0.31923348640056515,
"grad_norm": 1.0859375,
"learning_rate": 0.000499397923431706,
"loss": 6.2666,
"mean_token_accuracy": 0.1465996690094471,
"num_tokens": 6743738.0,
"step": 3615
},
{
"entropy": 6.337309741973877,
"epoch": 0.31967502649240553,
"grad_norm": 1.21875,
"learning_rate": 0.0004993956189847226,
"loss": 6.1012,
"mean_token_accuracy": 0.15459158420562744,
"num_tokens": 6752900.0,
"step": 3620
},
{
"entropy": 6.150611686706543,
"epoch": 0.32011656658424587,
"grad_norm": 1.140625,
"learning_rate": 0.0004993933101419565,
"loss": 6.106,
"mean_token_accuracy": 0.1578878253698349,
"num_tokens": 6761776.0,
"step": 3625
},
{
"entropy": 6.23021969795227,
"epoch": 0.3205581066760862,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993909969034531,
"loss": 6.2486,
"mean_token_accuracy": 0.1432959534227848,
"num_tokens": 6771543.0,
"step": 3630
},
{
"entropy": 6.232846355438232,
"epoch": 0.3209996467679265,
"grad_norm": 1.1796875,
"learning_rate": 0.0004993886792692576,
"loss": 6.1261,
"mean_token_accuracy": 0.15022226572036743,
"num_tokens": 6780010.0,
"step": 3635
},
{
"entropy": 6.273227500915527,
"epoch": 0.32144118685976686,
"grad_norm": 1.09375,
"learning_rate": 0.0004993863572394156,
"loss": 6.186,
"mean_token_accuracy": 0.14826097190380097,
"num_tokens": 6788830.0,
"step": 3640
},
{
"entropy": 6.266057395935059,
"epoch": 0.3218827269516072,
"grad_norm": 1.03125,
"learning_rate": 0.0004993840308139724,
"loss": 6.163,
"mean_token_accuracy": 0.15327871441841126,
"num_tokens": 6799121.0,
"step": 3645
},
{
"entropy": 6.232186365127563,
"epoch": 0.3223242670434475,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993816999929738,
"loss": 6.1253,
"mean_token_accuracy": 0.15162717401981354,
"num_tokens": 6808519.0,
"step": 3650
},
{
"entropy": 6.129374217987061,
"epoch": 0.3227658071352879,
"grad_norm": 1.015625,
"learning_rate": 0.0004993793647764651,
"loss": 6.1462,
"mean_token_accuracy": 0.16090914756059646,
"num_tokens": 6817583.0,
"step": 3655
},
{
"entropy": 6.254270458221436,
"epoch": 0.32320734722712824,
"grad_norm": 1.078125,
"learning_rate": 0.0004993770251644923,
"loss": 6.1315,
"mean_token_accuracy": 0.15279360860586166,
"num_tokens": 6827732.0,
"step": 3660
},
{
"entropy": 6.188901424407959,
"epoch": 0.32364888731896857,
"grad_norm": 1.1953125,
"learning_rate": 0.0004993746811571013,
"loss": 6.1101,
"mean_token_accuracy": 0.15757904648780824,
"num_tokens": 6836903.0,
"step": 3665
},
{
"entropy": 6.073100471496582,
"epoch": 0.3240904274108089,
"grad_norm": 1.0625,
"learning_rate": 0.0004993723327543379,
"loss": 6.119,
"mean_token_accuracy": 0.15876784324645996,
"num_tokens": 6845684.0,
"step": 3670
},
{
"entropy": 6.200423860549927,
"epoch": 0.32453196750264923,
"grad_norm": 1.0703125,
"learning_rate": 0.000499369979956248,
"loss": 6.1035,
"mean_token_accuracy": 0.14955462887883186,
"num_tokens": 6854650.0,
"step": 3675
},
{
"entropy": 6.267183446884156,
"epoch": 0.32497350759448956,
"grad_norm": 1.125,
"learning_rate": 0.0004993676227628779,
"loss": 6.0977,
"mean_token_accuracy": 0.1460177183151245,
"num_tokens": 6864340.0,
"step": 3680
},
{
"entropy": 6.208888530731201,
"epoch": 0.3254150476863299,
"grad_norm": 1.171875,
"learning_rate": 0.0004993652611742736,
"loss": 6.1343,
"mean_token_accuracy": 0.15321332961320877,
"num_tokens": 6873244.0,
"step": 3685
},
{
"entropy": 6.15152006149292,
"epoch": 0.3258565877781703,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993628951904815,
"loss": 6.0176,
"mean_token_accuracy": 0.15752123296260834,
"num_tokens": 6882392.0,
"step": 3690
},
{
"entropy": 6.2300208568572994,
"epoch": 0.3262981278700106,
"grad_norm": 1.15625,
"learning_rate": 0.0004993605248115479,
"loss": 6.2446,
"mean_token_accuracy": 0.1435020685195923,
"num_tokens": 6891515.0,
"step": 3695
},
{
"entropy": 6.330176734924317,
"epoch": 0.32673966796185094,
"grad_norm": 1.0859375,
"learning_rate": 0.0004993581500375191,
"loss": 6.1177,
"mean_token_accuracy": 0.14757276177406312,
"num_tokens": 6900104.0,
"step": 3700
},
{
"entropy": 6.2001354694366455,
"epoch": 0.3271812080536913,
"grad_norm": 1.3203125,
"learning_rate": 0.0004993557708684417,
"loss": 6.2227,
"mean_token_accuracy": 0.14519331306219102,
"num_tokens": 6910443.0,
"step": 3705
},
{
"entropy": 6.208211469650268,
"epoch": 0.3276227481455316,
"grad_norm": 1.2109375,
"learning_rate": 0.0004993533873043625,
"loss": 6.1062,
"mean_token_accuracy": 0.15348025262355805,
"num_tokens": 6919882.0,
"step": 3710
},
{
"entropy": 6.281739521026611,
"epoch": 0.32806428823737194,
"grad_norm": 1.1015625,
"learning_rate": 0.000499350999345328,
"loss": 6.2142,
"mean_token_accuracy": 0.14149869233369827,
"num_tokens": 6929757.0,
"step": 3715
},
{
"entropy": 6.222444868087768,
"epoch": 0.32850582832921227,
"grad_norm": 1.1171875,
"learning_rate": 0.000499348606991385,
"loss": 6.201,
"mean_token_accuracy": 0.1466899633407593,
"num_tokens": 6938578.0,
"step": 3720
},
{
"entropy": 6.181999588012696,
"epoch": 0.32894736842105265,
"grad_norm": 1.1953125,
"learning_rate": 0.0004993462102425805,
"loss": 6.0957,
"mean_token_accuracy": 0.14744215086102486,
"num_tokens": 6947920.0,
"step": 3725
},
{
"entropy": 6.170785808563233,
"epoch": 0.329388908512893,
"grad_norm": 1.125,
"learning_rate": 0.0004993438090989612,
"loss": 6.0325,
"mean_token_accuracy": 0.15373560339212416,
"num_tokens": 6957291.0,
"step": 3730
},
{
"entropy": 6.204143905639649,
"epoch": 0.3298304486047333,
"grad_norm": 1.1875,
"learning_rate": 0.0004993414035605743,
"loss": 6.0944,
"mean_token_accuracy": 0.16135938167572023,
"num_tokens": 6966504.0,
"step": 3735
},
{
"entropy": 6.26701283454895,
"epoch": 0.33027198869657365,
"grad_norm": 1.1015625,
"learning_rate": 0.0004993389936274669,
"loss": 6.1381,
"mean_token_accuracy": 0.15607366263866423,
"num_tokens": 6976234.0,
"step": 3740
},
{
"entropy": 6.179308748245239,
"epoch": 0.330713528788414,
"grad_norm": 1.0859375,
"learning_rate": 0.0004993365792996862,
"loss": 6.1882,
"mean_token_accuracy": 0.1475161299109459,
"num_tokens": 6986372.0,
"step": 3745
},
{
"entropy": 6.258656454086304,
"epoch": 0.3311550688802543,
"grad_norm": 1.046875,
"learning_rate": 0.0004993341605772795,
"loss": 6.0768,
"mean_token_accuracy": 0.155860435962677,
"num_tokens": 6995553.0,
"step": 3750
},
{
"entropy": 6.148722696304321,
"epoch": 0.33159660897209464,
"grad_norm": 1.1015625,
"learning_rate": 0.0004993317374602941,
"loss": 5.9902,
"mean_token_accuracy": 0.16303742080926895,
"num_tokens": 7005675.0,
"step": 3755
},
{
"entropy": 6.081928777694702,
"epoch": 0.33203814906393503,
"grad_norm": 1.2734375,
"learning_rate": 0.0004993293099487777,
"loss": 6.1198,
"mean_token_accuracy": 0.14885973036289216,
"num_tokens": 7014953.0,
"step": 3760
},
{
"entropy": 6.263726997375488,
"epoch": 0.33247968915577536,
"grad_norm": 1.1328125,
"learning_rate": 0.0004993268780427776,
"loss": 6.1445,
"mean_token_accuracy": 0.1564795732498169,
"num_tokens": 7025075.0,
"step": 3765
},
{
"entropy": 6.201370096206665,
"epoch": 0.3329212292476157,
"grad_norm": 1.28125,
"learning_rate": 0.0004993244417423416,
"loss": 6.1224,
"mean_token_accuracy": 0.15349680185317993,
"num_tokens": 7034286.0,
"step": 3770
},
{
"entropy": 6.201808977127075,
"epoch": 0.333362769339456,
"grad_norm": 1.1796875,
"learning_rate": 0.0004993220010475174,
"loss": 6.1767,
"mean_token_accuracy": 0.14727241545915604,
"num_tokens": 7043957.0,
"step": 3775
},
{
"entropy": 6.2423442840576175,
"epoch": 0.33380430943129635,
"grad_norm": 1.109375,
"learning_rate": 0.0004993195559583526,
"loss": 6.091,
"mean_token_accuracy": 0.15680659636855127,
"num_tokens": 7053170.0,
"step": 3780
},
{
"entropy": 6.163934850692749,
"epoch": 0.3342458495231367,
"grad_norm": 1.0859375,
"learning_rate": 0.0004993171064748954,
"loss": 6.0909,
"mean_token_accuracy": 0.15337430387735368,
"num_tokens": 7062469.0,
"step": 3785
},
{
"entropy": 6.245288276672364,
"epoch": 0.334687389614977,
"grad_norm": 1.109375,
"learning_rate": 0.0004993146525971937,
"loss": 6.0751,
"mean_token_accuracy": 0.15703266561031343,
"num_tokens": 7071463.0,
"step": 3790
},
{
"entropy": 6.30460000038147,
"epoch": 0.3351289297068174,
"grad_norm": 1.03125,
"learning_rate": 0.0004993121943252955,
"loss": 6.1685,
"mean_token_accuracy": 0.1484996944665909,
"num_tokens": 7081574.0,
"step": 3795
},
{
"entropy": 6.146984243392945,
"epoch": 0.33557046979865773,
"grad_norm": 1.1171875,
"learning_rate": 0.0004993097316592489,
"loss": 6.0826,
"mean_token_accuracy": 0.15749146193265914,
"num_tokens": 7090835.0,
"step": 3800
},
{
"entropy": 6.072944211959839,
"epoch": 0.33601200989049806,
"grad_norm": 1.5859375,
"learning_rate": 0.0004993072645991023,
"loss": 6.0092,
"mean_token_accuracy": 0.14467609971761702,
"num_tokens": 7100521.0,
"step": 3805
},
{
"entropy": 6.172339153289795,
"epoch": 0.3364535499823384,
"grad_norm": 1.2421875,
"learning_rate": 0.000499304793144904,
"loss": 6.108,
"mean_token_accuracy": 0.15104963332414628,
"num_tokens": 7109011.0,
"step": 3810
},
{
"entropy": 6.258471536636352,
"epoch": 0.3368950900741787,
"grad_norm": 1.109375,
"learning_rate": 0.0004993023172967022,
"loss": 6.1451,
"mean_token_accuracy": 0.14512295573949813,
"num_tokens": 7119128.0,
"step": 3815
},
{
"entropy": 6.266326761245727,
"epoch": 0.33733663016601906,
"grad_norm": 1.1328125,
"learning_rate": 0.0004992998370545458,
"loss": 6.2073,
"mean_token_accuracy": 0.14363647550344466,
"num_tokens": 7128313.0,
"step": 3820
},
{
"entropy": 6.234633827209473,
"epoch": 0.3377781702578594,
"grad_norm": 1.21875,
"learning_rate": 0.0004992973524184831,
"loss": 6.1686,
"mean_token_accuracy": 0.14797609224915503,
"num_tokens": 7137567.0,
"step": 3825
},
{
"entropy": 6.209169197082519,
"epoch": 0.3382197103496998,
"grad_norm": 1.265625,
"learning_rate": 0.0004992948633885627,
"loss": 6.1355,
"mean_token_accuracy": 0.15170362889766692,
"num_tokens": 7147254.0,
"step": 3830
},
{
"entropy": 6.123815822601318,
"epoch": 0.3386612504415401,
"grad_norm": 1.125,
"learning_rate": 0.0004992923699648335,
"loss": 6.055,
"mean_token_accuracy": 0.15891509354114533,
"num_tokens": 7156215.0,
"step": 3835
},
{
"entropy": 6.286255836486816,
"epoch": 0.33910279053338044,
"grad_norm": 1.2265625,
"learning_rate": 0.0004992898721473445,
"loss": 6.1341,
"mean_token_accuracy": 0.14742862731218337,
"num_tokens": 7165653.0,
"step": 3840
},
{
"entropy": 6.107093811035156,
"epoch": 0.33954433062522077,
"grad_norm": 1.2265625,
"learning_rate": 0.0004992873699361444,
"loss": 6.0116,
"mean_token_accuracy": 0.15364348739385605,
"num_tokens": 7174797.0,
"step": 3845
},
{
"entropy": 6.231772947311401,
"epoch": 0.3399858707170611,
"grad_norm": 1.1015625,
"learning_rate": 0.0004992848633312822,
"loss": 6.076,
"mean_token_accuracy": 0.15550567209720612,
"num_tokens": 7184040.0,
"step": 3850
},
{
"entropy": 6.206028461456299,
"epoch": 0.34042741080890143,
"grad_norm": 1.21875,
"learning_rate": 0.0004992823523328071,
"loss": 6.0162,
"mean_token_accuracy": 0.156949782371521,
"num_tokens": 7193533.0,
"step": 3855
},
{
"entropy": 6.040725946426392,
"epoch": 0.34086895090074176,
"grad_norm": 1.4296875,
"learning_rate": 0.0004992798369407684,
"loss": 6.1271,
"mean_token_accuracy": 0.1540011927485466,
"num_tokens": 7203355.0,
"step": 3860
},
{
"entropy": 6.231026697158813,
"epoch": 0.34131049099258215,
"grad_norm": 1.140625,
"learning_rate": 0.0004992773171552152,
"loss": 6.0662,
"mean_token_accuracy": 0.15278246700763704,
"num_tokens": 7212569.0,
"step": 3865
},
{
"entropy": 6.1147346019744875,
"epoch": 0.3417520310844225,
"grad_norm": 1.1015625,
"learning_rate": 0.0004992747929761968,
"loss": 6.0862,
"mean_token_accuracy": 0.1568584769964218,
"num_tokens": 7221886.0,
"step": 3870
},
{
"entropy": 6.20992078781128,
"epoch": 0.3421935711762628,
"grad_norm": 1.234375,
"learning_rate": 0.0004992722644037628,
"loss": 6.1399,
"mean_token_accuracy": 0.14878576919436454,
"num_tokens": 7232065.0,
"step": 3875
},
{
"entropy": 6.2117961883544925,
"epoch": 0.34263511126810314,
"grad_norm": 0.98828125,
"learning_rate": 0.0004992697314379628,
"loss": 6.0802,
"mean_token_accuracy": 0.1546872690320015,
"num_tokens": 7242287.0,
"step": 3880
},
{
"entropy": 6.163637351989746,
"epoch": 0.3430766513599435,
"grad_norm": 1.203125,
"learning_rate": 0.0004992671940788462,
"loss": 6.0744,
"mean_token_accuracy": 0.15107770562171935,
"num_tokens": 7250638.0,
"step": 3885
},
{
"entropy": 6.146958065032959,
"epoch": 0.3435181914517838,
"grad_norm": 1.1171875,
"learning_rate": 0.0004992646523264628,
"loss": 6.0492,
"mean_token_accuracy": 0.1611901268362999,
"num_tokens": 7259526.0,
"step": 3890
},
{
"entropy": 6.119075059890747,
"epoch": 0.34395973154362414,
"grad_norm": 1.125,
"learning_rate": 0.0004992621061808625,
"loss": 6.0604,
"mean_token_accuracy": 0.15892861932516097,
"num_tokens": 7269260.0,
"step": 3895
},
{
"entropy": 6.161959505081176,
"epoch": 0.3444012716354645,
"grad_norm": 1.1484375,
"learning_rate": 0.000499259555642095,
"loss": 6.0786,
"mean_token_accuracy": 0.15186367481946944,
"num_tokens": 7279061.0,
"step": 3900
},
{
"entropy": 6.21967043876648,
"epoch": 0.34484281172730485,
"grad_norm": 1.1328125,
"learning_rate": 0.0004992570007102104,
"loss": 6.0831,
"mean_token_accuracy": 0.15683933049440385,
"num_tokens": 7289721.0,
"step": 3905
},
{
"entropy": 6.214530658721924,
"epoch": 0.3452843518191452,
"grad_norm": 1.265625,
"learning_rate": 0.0004992544413852587,
"loss": 6.2136,
"mean_token_accuracy": 0.14554179906845094,
"num_tokens": 7299836.0,
"step": 3910
},
{
"entropy": 6.225806951522827,
"epoch": 0.3457258919109855,
"grad_norm": 1.125,
"learning_rate": 0.00049925187766729,
"loss": 6.2249,
"mean_token_accuracy": 0.14765020608901977,
"num_tokens": 7309306.0,
"step": 3915
},
{
"entropy": 6.249141550064087,
"epoch": 0.34616743200282585,
"grad_norm": 1.203125,
"learning_rate": 0.0004992493095563545,
"loss": 6.1238,
"mean_token_accuracy": 0.16257761269807816,
"num_tokens": 7318486.0,
"step": 3920
},
{
"entropy": 6.111485767364502,
"epoch": 0.3466089720946662,
"grad_norm": 1.1484375,
"learning_rate": 0.0004992467370525026,
"loss": 6.083,
"mean_token_accuracy": 0.1557182028889656,
"num_tokens": 7327764.0,
"step": 3925
},
{
"entropy": 6.248839044570923,
"epoch": 0.3470505121865065,
"grad_norm": 1.140625,
"learning_rate": 0.0004992441601557848,
"loss": 6.168,
"mean_token_accuracy": 0.15208059847354888,
"num_tokens": 7337118.0,
"step": 3930
},
{
"entropy": 6.197002506256103,
"epoch": 0.3474920522783469,
"grad_norm": 1.203125,
"learning_rate": 0.0004992415788662514,
"loss": 6.0958,
"mean_token_accuracy": 0.1525207430124283,
"num_tokens": 7345973.0,
"step": 3935
},
{
"entropy": 6.194469833374024,
"epoch": 0.34793359237018723,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992389931839529,
"loss": 6.0928,
"mean_token_accuracy": 0.1534672871232033,
"num_tokens": 7355776.0,
"step": 3940
},
{
"entropy": 6.168269205093384,
"epoch": 0.34837513246202756,
"grad_norm": 1.1640625,
"learning_rate": 0.0004992364031089401,
"loss": 6.1161,
"mean_token_accuracy": 0.1555505856871605,
"num_tokens": 7364820.0,
"step": 3945
},
{
"entropy": 6.2373566150665285,
"epoch": 0.3488166725538679,
"grad_norm": 1.234375,
"learning_rate": 0.0004992338086412636,
"loss": 6.1218,
"mean_token_accuracy": 0.14930969327688218,
"num_tokens": 7373881.0,
"step": 3950
},
{
"entropy": 6.135814476013183,
"epoch": 0.3492582126457082,
"grad_norm": 1.2109375,
"learning_rate": 0.0004992312097809744,
"loss": 6.0892,
"mean_token_accuracy": 0.15562115162611007,
"num_tokens": 7383095.0,
"step": 3955
},
{
"entropy": 6.195784330368042,
"epoch": 0.34969975273754855,
"grad_norm": 1.125,
"learning_rate": 0.0004992286065281234,
"loss": 6.1216,
"mean_token_accuracy": 0.15000923871994018,
"num_tokens": 7392702.0,
"step": 3960
},
{
"entropy": 6.247405576705932,
"epoch": 0.3501412928293889,
"grad_norm": 1.1015625,
"learning_rate": 0.0004992259988827614,
"loss": 6.2769,
"mean_token_accuracy": 0.14470015615224838,
"num_tokens": 7402560.0,
"step": 3965
},
{
"entropy": 6.284795522689819,
"epoch": 0.35058283292122927,
"grad_norm": 1.125,
"learning_rate": 0.0004992233868449397,
"loss": 6.051,
"mean_token_accuracy": 0.15953975468873977,
"num_tokens": 7412375.0,
"step": 3970
},
{
"entropy": 6.195076513290405,
"epoch": 0.3510243730130696,
"grad_norm": 1.0625,
"learning_rate": 0.0004992207704147093,
"loss": 6.0928,
"mean_token_accuracy": 0.15187776535749437,
"num_tokens": 7420734.0,
"step": 3975
},
{
"entropy": 6.101401853561401,
"epoch": 0.35146591310490993,
"grad_norm": 1.109375,
"learning_rate": 0.0004992181495921216,
"loss": 6.0919,
"mean_token_accuracy": 0.15785269439220428,
"num_tokens": 7430156.0,
"step": 3980
},
{
"entropy": 6.222190999984742,
"epoch": 0.35190745319675026,
"grad_norm": 1.15625,
"learning_rate": 0.0004992155243772277,
"loss": 6.1007,
"mean_token_accuracy": 0.1574430137872696,
"num_tokens": 7439189.0,
"step": 3985
},
{
"entropy": 6.2206672668457035,
"epoch": 0.3523489932885906,
"grad_norm": 1.0703125,
"learning_rate": 0.0004992128947700795,
"loss": 6.1249,
"mean_token_accuracy": 0.1455397441983223,
"num_tokens": 7448768.0,
"step": 3990
},
{
"entropy": 6.180650806427002,
"epoch": 0.3527905333804309,
"grad_norm": 1.1171875,
"learning_rate": 0.000499210260770728,
"loss": 6.064,
"mean_token_accuracy": 0.15497558265924455,
"num_tokens": 7458468.0,
"step": 3995
},
{
"entropy": 6.150880622863769,
"epoch": 0.35323207347227126,
"grad_norm": 1.1328125,
"learning_rate": 0.000499207622379225,
"loss": 6.1475,
"mean_token_accuracy": 0.1480465464293957,
"num_tokens": 7467351.0,
"step": 4000
},
{
"entropy": 6.2564455509185795,
"epoch": 0.35367361356411164,
"grad_norm": 1.0703125,
"learning_rate": 0.0004992049795956222,
"loss": 6.1012,
"mean_token_accuracy": 0.14711347222328186,
"num_tokens": 7477116.0,
"step": 4005
},
{
"entropy": 6.234343814849853,
"epoch": 0.354115153655952,
"grad_norm": 1.2734375,
"learning_rate": 0.0004992023324199715,
"loss": 6.0249,
"mean_token_accuracy": 0.15656134784221648,
"num_tokens": 7485216.0,
"step": 4010
},
{
"entropy": 6.075209522247315,
"epoch": 0.3545566937477923,
"grad_norm": 1.03125,
"learning_rate": 0.0004991996808523245,
"loss": 6.1798,
"mean_token_accuracy": 0.1469147637486458,
"num_tokens": 7495084.0,
"step": 4015
},
{
"entropy": 6.306225442886353,
"epoch": 0.35499823383963264,
"grad_norm": 1.09375,
"learning_rate": 0.0004991970248927332,
"loss": 6.1973,
"mean_token_accuracy": 0.14604026302695275,
"num_tokens": 7503219.0,
"step": 4020
},
{
"entropy": 6.196821165084839,
"epoch": 0.35543977393147297,
"grad_norm": 1.234375,
"learning_rate": 0.0004991943645412498,
"loss": 5.9766,
"mean_token_accuracy": 0.1642112761735916,
"num_tokens": 7511961.0,
"step": 4025
},
{
"entropy": 5.985270690917969,
"epoch": 0.3558813140233133,
"grad_norm": 1.1015625,
"learning_rate": 0.0004991916997979263,
"loss": 6.1096,
"mean_token_accuracy": 0.1546097069978714,
"num_tokens": 7521927.0,
"step": 4030
},
{
"entropy": 6.103000640869141,
"epoch": 0.35632285411515363,
"grad_norm": 1.171875,
"learning_rate": 0.0004991890306628149,
"loss": 6.0451,
"mean_token_accuracy": 0.15554122179746627,
"num_tokens": 7531093.0,
"step": 4035
},
{
"entropy": 6.198269557952881,
"epoch": 0.356764394206994,
"grad_norm": 1.125,
"learning_rate": 0.0004991863571359678,
"loss": 6.0078,
"mean_token_accuracy": 0.15710966363549234,
"num_tokens": 7539813.0,
"step": 4040
},
{
"entropy": 6.244335699081421,
"epoch": 0.35720593429883435,
"grad_norm": 1.3125,
"learning_rate": 0.0004991836792174376,
"loss": 6.1875,
"mean_token_accuracy": 0.1476308137178421,
"num_tokens": 7548954.0,
"step": 4045
},
{
"entropy": 6.100079536437988,
"epoch": 0.3576474743906747,
"grad_norm": 1.09375,
"learning_rate": 0.0004991809969072765,
"loss": 6.0322,
"mean_token_accuracy": 0.1548540085554123,
"num_tokens": 7558275.0,
"step": 4050
},
{
"entropy": 6.13811764717102,
"epoch": 0.358089014482515,
"grad_norm": 1.234375,
"learning_rate": 0.0004991783102055371,
"loss": 6.0688,
"mean_token_accuracy": 0.15873494520783424,
"num_tokens": 7567870.0,
"step": 4055
},
{
"entropy": 6.249936914443969,
"epoch": 0.35853055457435534,
"grad_norm": 1.1015625,
"learning_rate": 0.0004991756191122723,
"loss": 6.0385,
"mean_token_accuracy": 0.15568251237273217,
"num_tokens": 7577485.0,
"step": 4060
},
{
"entropy": 6.142808294296264,
"epoch": 0.3589720946661957,
"grad_norm": 1.109375,
"learning_rate": 0.0004991729236275346,
"loss": 6.0888,
"mean_token_accuracy": 0.150615693628788,
"num_tokens": 7587398.0,
"step": 4065
},
{
"entropy": 6.10182294845581,
"epoch": 0.359413634758036,
"grad_norm": 1.125,
"learning_rate": 0.0004991702237513768,
"loss": 6.0896,
"mean_token_accuracy": 0.1515856146812439,
"num_tokens": 7596310.0,
"step": 4070
},
{
"entropy": 6.236705541610718,
"epoch": 0.3598551748498764,
"grad_norm": 1.078125,
"learning_rate": 0.0004991675194838517,
"loss": 6.1214,
"mean_token_accuracy": 0.14487750828266144,
"num_tokens": 7606971.0,
"step": 4075
},
{
"entropy": 6.157118654251098,
"epoch": 0.3602967149417167,
"grad_norm": 1.265625,
"learning_rate": 0.0004991648108250125,
"loss": 6.1073,
"mean_token_accuracy": 0.15116591304540633,
"num_tokens": 7616636.0,
"step": 4080
},
{
"entropy": 6.23088059425354,
"epoch": 0.36073825503355705,
"grad_norm": 1.1953125,
"learning_rate": 0.000499162097774912,
"loss": 6.1325,
"mean_token_accuracy": 0.15135489255189896,
"num_tokens": 7625128.0,
"step": 4085
},
{
"entropy": 6.168088626861572,
"epoch": 0.3611797951253974,
"grad_norm": 1.09375,
"learning_rate": 0.0004991593803336037,
"loss": 6.0464,
"mean_token_accuracy": 0.15914286002516748,
"num_tokens": 7634224.0,
"step": 4090
},
{
"entropy": 6.109174299240112,
"epoch": 0.3616213352172377,
"grad_norm": 1.1015625,
"learning_rate": 0.0004991566585011405,
"loss": 6.0204,
"mean_token_accuracy": 0.1594786301255226,
"num_tokens": 7643392.0,
"step": 4095
},
{
"entropy": 6.046089458465576,
"epoch": 0.36206287530907805,
"grad_norm": 1.265625,
"learning_rate": 0.0004991539322775758,
"loss": 6.0402,
"mean_token_accuracy": 0.15390734672546386,
"num_tokens": 7652491.0,
"step": 4100
},
{
"entropy": 6.333108568191529,
"epoch": 0.3625044154009184,
"grad_norm": 1.140625,
"learning_rate": 0.0004991512016629632,
"loss": 6.2631,
"mean_token_accuracy": 0.1374726377427578,
"num_tokens": 7663192.0,
"step": 4105
},
{
"entropy": 6.233335781097412,
"epoch": 0.36294595549275877,
"grad_norm": 1.140625,
"learning_rate": 0.0004991484666573558,
"loss": 6.0951,
"mean_token_accuracy": 0.1491971492767334,
"num_tokens": 7672801.0,
"step": 4110
},
{
"entropy": 6.150677967071533,
"epoch": 0.3633874955845991,
"grad_norm": 1.0859375,
"learning_rate": 0.0004991457272608077,
"loss": 6.1107,
"mean_token_accuracy": 0.1492188058793545,
"num_tokens": 7683013.0,
"step": 4115
},
{
"entropy": 6.215946054458618,
"epoch": 0.3638290356764394,
"grad_norm": 1.1484375,
"learning_rate": 0.0004991429834733721,
"loss": 6.1377,
"mean_token_accuracy": 0.14726671427488328,
"num_tokens": 7692760.0,
"step": 4120
},
{
"entropy": 6.212368583679199,
"epoch": 0.36427057576827976,
"grad_norm": 1.390625,
"learning_rate": 0.000499140235295103,
"loss": 6.0806,
"mean_token_accuracy": 0.15191175639629365,
"num_tokens": 7702005.0,
"step": 4125
},
{
"entropy": 6.164025735855103,
"epoch": 0.3647121158601201,
"grad_norm": 1.296875,
"learning_rate": 0.0004991374827260542,
"loss": 6.129,
"mean_token_accuracy": 0.1487259477376938,
"num_tokens": 7711604.0,
"step": 4130
},
{
"entropy": 6.113088750839234,
"epoch": 0.3651536559519604,
"grad_norm": 1.296875,
"learning_rate": 0.0004991347257662795,
"loss": 6.0471,
"mean_token_accuracy": 0.153825144469738,
"num_tokens": 7720434.0,
"step": 4135
},
{
"entropy": 6.218751907348633,
"epoch": 0.36559519604380075,
"grad_norm": 1.109375,
"learning_rate": 0.000499131964415833,
"loss": 6.0877,
"mean_token_accuracy": 0.1532296895980835,
"num_tokens": 7730733.0,
"step": 4140
},
{
"entropy": 6.158688402175903,
"epoch": 0.36603673613564114,
"grad_norm": 1.3828125,
"learning_rate": 0.0004991291986747689,
"loss": 6.1581,
"mean_token_accuracy": 0.153475009649992,
"num_tokens": 7740394.0,
"step": 4145
},
{
"entropy": 6.191565799713135,
"epoch": 0.36647827622748147,
"grad_norm": 1.15625,
"learning_rate": 0.0004991264285431412,
"loss": 6.1045,
"mean_token_accuracy": 0.15052587389945984,
"num_tokens": 7748919.0,
"step": 4150
},
{
"entropy": 6.293946027755737,
"epoch": 0.3669198163193218,
"grad_norm": 1.171875,
"learning_rate": 0.0004991236540210041,
"loss": 6.1009,
"mean_token_accuracy": 0.1537343256175518,
"num_tokens": 7758591.0,
"step": 4155
},
{
"entropy": 6.244499015808105,
"epoch": 0.36736135641116213,
"grad_norm": 1.15625,
"learning_rate": 0.0004991208751084122,
"loss": 6.1485,
"mean_token_accuracy": 0.1491132453083992,
"num_tokens": 7768269.0,
"step": 4160
},
{
"entropy": 6.249349975585938,
"epoch": 0.36780289650300246,
"grad_norm": 1.203125,
"learning_rate": 0.0004991180918054199,
"loss": 6.1767,
"mean_token_accuracy": 0.1463681861758232,
"num_tokens": 7776980.0,
"step": 4165
},
{
"entropy": 6.171654605865479,
"epoch": 0.3682444365948428,
"grad_norm": 1.1953125,
"learning_rate": 0.0004991153041120815,
"loss": 6.0811,
"mean_token_accuracy": 0.15733396261930466,
"num_tokens": 7786664.0,
"step": 4170
},
{
"entropy": 6.12093939781189,
"epoch": 0.3686859766866831,
"grad_norm": 1.171875,
"learning_rate": 0.0004991125120284519,
"loss": 5.9525,
"mean_token_accuracy": 0.1650825932621956,
"num_tokens": 7794981.0,
"step": 4175
},
{
"entropy": 6.110823106765747,
"epoch": 0.3691275167785235,
"grad_norm": 1.0859375,
"learning_rate": 0.0004991097155545856,
"loss": 6.0387,
"mean_token_accuracy": 0.1560913234949112,
"num_tokens": 7804850.0,
"step": 4180
},
{
"entropy": 6.106446075439453,
"epoch": 0.36956905687036384,
"grad_norm": 1.28125,
"learning_rate": 0.0004991069146905374,
"loss": 6.052,
"mean_token_accuracy": 0.15863640755414962,
"num_tokens": 7814117.0,
"step": 4185
},
{
"entropy": 6.183217334747314,
"epoch": 0.3700105969622042,
"grad_norm": 1.140625,
"learning_rate": 0.0004991041094363621,
"loss": 6.0928,
"mean_token_accuracy": 0.15649753212928771,
"num_tokens": 7823384.0,
"step": 4190
},
{
"entropy": 6.2000589847564695,
"epoch": 0.3704521370540445,
"grad_norm": 1.0703125,
"learning_rate": 0.0004991012997921149,
"loss": 6.0863,
"mean_token_accuracy": 0.154317145049572,
"num_tokens": 7834072.0,
"step": 4195
},
{
"entropy": 6.225698471069336,
"epoch": 0.37089367714588484,
"grad_norm": 1.1953125,
"learning_rate": 0.0004990984857578506,
"loss": 6.0718,
"mean_token_accuracy": 0.1523341119289398,
"num_tokens": 7843334.0,
"step": 4200
},
{
"entropy": 6.040940427780152,
"epoch": 0.37133521723772517,
"grad_norm": 1.5234375,
"learning_rate": 0.0004990956673336245,
"loss": 6.0132,
"mean_token_accuracy": 0.15985633432865143,
"num_tokens": 7853187.0,
"step": 4205
},
{
"entropy": 6.121350336074829,
"epoch": 0.3717767573295655,
"grad_norm": 1.1484375,
"learning_rate": 0.0004990928445194917,
"loss": 6.0836,
"mean_token_accuracy": 0.1519768550992012,
"num_tokens": 7862262.0,
"step": 4210
},
{
"entropy": 6.15352463722229,
"epoch": 0.3722182974214059,
"grad_norm": 1.1640625,
"learning_rate": 0.0004990900173155074,
"loss": 6.0184,
"mean_token_accuracy": 0.15704632550477982,
"num_tokens": 7870725.0,
"step": 4215
},
{
"entropy": 6.124208784103393,
"epoch": 0.3726598375132462,
"grad_norm": 1.2109375,
"learning_rate": 0.0004990871857217273,
"loss": 5.9447,
"mean_token_accuracy": 0.15963577330112458,
"num_tokens": 7879492.0,
"step": 4220
},
{
"entropy": 6.087597894668579,
"epoch": 0.37310137760508655,
"grad_norm": 1.375,
"learning_rate": 0.0004990843497382066,
"loss": 6.0649,
"mean_token_accuracy": 0.15454517751932145,
"num_tokens": 7888710.0,
"step": 4225
},
{
"entropy": 6.171204519271851,
"epoch": 0.3735429176969269,
"grad_norm": 1.109375,
"learning_rate": 0.0004990815093650009,
"loss": 6.0846,
"mean_token_accuracy": 0.15838453769683838,
"num_tokens": 7898313.0,
"step": 4230
},
{
"entropy": 6.163542985916138,
"epoch": 0.3739844577887672,
"grad_norm": 1.3203125,
"learning_rate": 0.0004990786646021659,
"loss": 6.0972,
"mean_token_accuracy": 0.15132492929697036,
"num_tokens": 7906726.0,
"step": 4235
},
{
"entropy": 6.143102693557739,
"epoch": 0.37442599788060754,
"grad_norm": 1.0859375,
"learning_rate": 0.0004990758154497573,
"loss": 6.0083,
"mean_token_accuracy": 0.16228035539388658,
"num_tokens": 7915554.0,
"step": 4240
},
{
"entropy": 6.224862670898437,
"epoch": 0.3748675379724479,
"grad_norm": 1.1875,
"learning_rate": 0.0004990729619078309,
"loss": 6.1357,
"mean_token_accuracy": 0.15337878912687303,
"num_tokens": 7924732.0,
"step": 4245
},
{
"entropy": 6.230613327026367,
"epoch": 0.37530907806428826,
"grad_norm": 1.0703125,
"learning_rate": 0.0004990701039764427,
"loss": 6.061,
"mean_token_accuracy": 0.1594015821814537,
"num_tokens": 7933808.0,
"step": 4250
},
{
"entropy": 6.101580095291138,
"epoch": 0.3757506181561286,
"grad_norm": 1.1171875,
"learning_rate": 0.0004990672416556485,
"loss": 6.0319,
"mean_token_accuracy": 0.15845490843057633,
"num_tokens": 7942558.0,
"step": 4255
},
{
"entropy": 6.129454803466797,
"epoch": 0.3761921582479689,
"grad_norm": 1.2265625,
"learning_rate": 0.0004990643749455045,
"loss": 6.1321,
"mean_token_accuracy": 0.14685340449213982,
"num_tokens": 7951951.0,
"step": 4260
},
{
"entropy": 6.2848834037780765,
"epoch": 0.37663369833980925,
"grad_norm": 1.1484375,
"learning_rate": 0.0004990615038460667,
"loss": 6.0247,
"mean_token_accuracy": 0.1576920345425606,
"num_tokens": 7960543.0,
"step": 4265
},
{
"entropy": 6.126399230957031,
"epoch": 0.3770752384316496,
"grad_norm": 1.109375,
"learning_rate": 0.0004990586283573916,
"loss": 6.007,
"mean_token_accuracy": 0.155757275223732,
"num_tokens": 7970273.0,
"step": 4270
},
{
"entropy": 6.158202123641968,
"epoch": 0.3775167785234899,
"grad_norm": 1.2109375,
"learning_rate": 0.0004990557484795355,
"loss": 6.1038,
"mean_token_accuracy": 0.1517634019255638,
"num_tokens": 7978888.0,
"step": 4275
},
{
"entropy": 6.186517333984375,
"epoch": 0.37795831861533025,
"grad_norm": 1.1953125,
"learning_rate": 0.0004990528642125545,
"loss": 6.0938,
"mean_token_accuracy": 0.15008659660816193,
"num_tokens": 7988843.0,
"step": 4280
},
{
"entropy": 6.1609944820404055,
"epoch": 0.37839985870717063,
"grad_norm": 1.171875,
"learning_rate": 0.0004990499755565055,
"loss": 6.0944,
"mean_token_accuracy": 0.15281789302825927,
"num_tokens": 7998403.0,
"step": 4285
},
{
"entropy": 6.190595436096191,
"epoch": 0.37884139879901096,
"grad_norm": 1.1015625,
"learning_rate": 0.0004990470825114448,
"loss": 6.1108,
"mean_token_accuracy": 0.15413309782743453,
"num_tokens": 8007493.0,
"step": 4290
},
{
"entropy": 6.112000417709351,
"epoch": 0.3792829388908513,
"grad_norm": 1.109375,
"learning_rate": 0.0004990441850774292,
"loss": 6.0581,
"mean_token_accuracy": 0.1584509640932083,
"num_tokens": 8016399.0,
"step": 4295
},
{
"entropy": 6.215829849243164,
"epoch": 0.3797244789826916,
"grad_norm": 1.09375,
"learning_rate": 0.0004990412832545155,
"loss": 6.0591,
"mean_token_accuracy": 0.1544952630996704,
"num_tokens": 8025891.0,
"step": 4300
},
{
"entropy": 6.077403736114502,
"epoch": 0.38016601907453196,
"grad_norm": 1.21875,
"learning_rate": 0.0004990383770427603,
"loss": 6.0097,
"mean_token_accuracy": 0.15796091556549072,
"num_tokens": 8034733.0,
"step": 4305
},
{
"entropy": 6.135886240005493,
"epoch": 0.3806075591663723,
"grad_norm": 1.109375,
"learning_rate": 0.0004990354664422209,
"loss": 6.0527,
"mean_token_accuracy": 0.15982279628515245,
"num_tokens": 8044443.0,
"step": 4310
},
{
"entropy": 6.160210800170899,
"epoch": 0.3810490992582126,
"grad_norm": 1.3125,
"learning_rate": 0.0004990325514529541,
"loss": 6.0403,
"mean_token_accuracy": 0.16555903106927872,
"num_tokens": 8053920.0,
"step": 4315
},
{
"entropy": 6.1733404159545895,
"epoch": 0.381490639350053,
"grad_norm": 1.1484375,
"learning_rate": 0.0004990296320750169,
"loss": 6.0242,
"mean_token_accuracy": 0.15823598951101303,
"num_tokens": 8063455.0,
"step": 4320
},
{
"entropy": 6.164728498458862,
"epoch": 0.38193217944189334,
"grad_norm": 1.1640625,
"learning_rate": 0.0004990267083084667,
"loss": 6.1096,
"mean_token_accuracy": 0.15378031879663467,
"num_tokens": 8073353.0,
"step": 4325
},
{
"entropy": 6.216868829727173,
"epoch": 0.38237371953373367,
"grad_norm": 1.171875,
"learning_rate": 0.0004990237801533607,
"loss": 6.1198,
"mean_token_accuracy": 0.1568579077720642,
"num_tokens": 8082164.0,
"step": 4330
},
{
"entropy": 6.222427320480347,
"epoch": 0.382815259625574,
"grad_norm": 1.21875,
"learning_rate": 0.0004990208476097562,
"loss": 6.126,
"mean_token_accuracy": 0.15226055085659027,
"num_tokens": 8090790.0,
"step": 4335
},
{
"entropy": 6.293479156494141,
"epoch": 0.38325679971741433,
"grad_norm": 1.15625,
"learning_rate": 0.0004990179106777109,
"loss": 6.1346,
"mean_token_accuracy": 0.14660235047340392,
"num_tokens": 8100412.0,
"step": 4340
},
{
"entropy": 6.121231889724731,
"epoch": 0.38369833980925466,
"grad_norm": 1.2265625,
"learning_rate": 0.0004990149693572819,
"loss": 6.1328,
"mean_token_accuracy": 0.14986878782510757,
"num_tokens": 8110002.0,
"step": 4345
},
{
"entropy": 6.1944701194763185,
"epoch": 0.384139879901095,
"grad_norm": 1.46875,
"learning_rate": 0.0004990120236485271,
"loss": 6.0262,
"mean_token_accuracy": 0.15945157259702683,
"num_tokens": 8118898.0,
"step": 4350
},
{
"entropy": 6.093376398086548,
"epoch": 0.3845814199929354,
"grad_norm": 1.1953125,
"learning_rate": 0.0004990090735515043,
"loss": 6.0696,
"mean_token_accuracy": 0.1521653488278389,
"num_tokens": 8128207.0,
"step": 4355
},
{
"entropy": 6.162809753417969,
"epoch": 0.3850229600847757,
"grad_norm": 1.21875,
"learning_rate": 0.000499006119066271,
"loss": 6.0659,
"mean_token_accuracy": 0.15488530248403548,
"num_tokens": 8137317.0,
"step": 4360
},
{
"entropy": 6.234450483322144,
"epoch": 0.38546450017661604,
"grad_norm": 1.125,
"learning_rate": 0.0004990031601928854,
"loss": 6.0901,
"mean_token_accuracy": 0.15392402857542037,
"num_tokens": 8146519.0,
"step": 4365
},
{
"entropy": 6.081244421005249,
"epoch": 0.3859060402684564,
"grad_norm": 1.1640625,
"learning_rate": 0.0004990001969314051,
"loss": 6.1156,
"mean_token_accuracy": 0.150378455221653,
"num_tokens": 8155916.0,
"step": 4370
},
{
"entropy": 6.188456583023071,
"epoch": 0.3863475803602967,
"grad_norm": 1.125,
"learning_rate": 0.0004989972292818884,
"loss": 6.1306,
"mean_token_accuracy": 0.15578001141548156,
"num_tokens": 8165251.0,
"step": 4375
},
{
"entropy": 6.154488277435303,
"epoch": 0.38678912045213704,
"grad_norm": 1.1328125,
"learning_rate": 0.0004989942572443934,
"loss": 6.1046,
"mean_token_accuracy": 0.154154072701931,
"num_tokens": 8174761.0,
"step": 4380
},
{
"entropy": 6.2399732112884525,
"epoch": 0.38723066054397737,
"grad_norm": 1.1328125,
"learning_rate": 0.0004989912808189784,
"loss": 6.0624,
"mean_token_accuracy": 0.1586390733718872,
"num_tokens": 8182980.0,
"step": 4385
},
{
"entropy": 6.286262083053589,
"epoch": 0.38767220063581775,
"grad_norm": 1.1015625,
"learning_rate": 0.0004989883000057013,
"loss": 6.1013,
"mean_token_accuracy": 0.15117012858390808,
"num_tokens": 8193306.0,
"step": 4390
},
{
"entropy": 6.023512887954712,
"epoch": 0.3881137407276581,
"grad_norm": 1.203125,
"learning_rate": 0.000498985314804621,
"loss": 5.9827,
"mean_token_accuracy": 0.16275162994861603,
"num_tokens": 8203446.0,
"step": 4395
},
{
"entropy": 6.129458618164063,
"epoch": 0.3885552808194984,
"grad_norm": 1.1015625,
"learning_rate": 0.0004989823252157958,
"loss": 6.0411,
"mean_token_accuracy": 0.14927417337894439,
"num_tokens": 8212414.0,
"step": 4400
},
{
"entropy": 6.107169580459595,
"epoch": 0.38899682091133875,
"grad_norm": 1.125,
"learning_rate": 0.0004989793312392841,
"loss": 5.9633,
"mean_token_accuracy": 0.16139311194419861,
"num_tokens": 8221509.0,
"step": 4405
},
{
"entropy": 6.121396112442016,
"epoch": 0.3894383610031791,
"grad_norm": 1.46875,
"learning_rate": 0.0004989763328751448,
"loss": 6.0825,
"mean_token_accuracy": 0.15815991312265396,
"num_tokens": 8230017.0,
"step": 4410
},
{
"entropy": 6.1887977600097654,
"epoch": 0.3898799010950194,
"grad_norm": 1.265625,
"learning_rate": 0.0004989733301234365,
"loss": 6.0249,
"mean_token_accuracy": 0.15967779457569123,
"num_tokens": 8238853.0,
"step": 4415
},
{
"entropy": 6.224899578094482,
"epoch": 0.39032144118685974,
"grad_norm": 1.078125,
"learning_rate": 0.000498970322984218,
"loss": 6.1023,
"mean_token_accuracy": 0.1519481733441353,
"num_tokens": 8248112.0,
"step": 4420
},
{
"entropy": 6.203225469589233,
"epoch": 0.39076298127870013,
"grad_norm": 1.3125,
"learning_rate": 0.0004989673114575483,
"loss": 6.0589,
"mean_token_accuracy": 0.1536906696856022,
"num_tokens": 8257166.0,
"step": 4425
},
{
"entropy": 6.128660726547241,
"epoch": 0.39120452137054046,
"grad_norm": 1.2265625,
"learning_rate": 0.0004989642955434863,
"loss": 6.0714,
"mean_token_accuracy": 0.1515656217932701,
"num_tokens": 8266229.0,
"step": 4430
},
{
"entropy": 6.163424253463745,
"epoch": 0.3916460614623808,
"grad_norm": 1.171875,
"learning_rate": 0.0004989612752420912,
"loss": 6.039,
"mean_token_accuracy": 0.15465213656425475,
"num_tokens": 8275102.0,
"step": 4435
},
{
"entropy": 6.022084140777588,
"epoch": 0.3920876015542211,
"grad_norm": 1.203125,
"learning_rate": 0.000498958250553422,
"loss": 5.9631,
"mean_token_accuracy": 0.16303362101316451,
"num_tokens": 8283847.0,
"step": 4440
},
{
"entropy": 6.193567132949829,
"epoch": 0.39252914164606145,
"grad_norm": 1.1640625,
"learning_rate": 0.0004989552214775381,
"loss": 6.0871,
"mean_token_accuracy": 0.15131851583719252,
"num_tokens": 8292622.0,
"step": 4445
},
{
"entropy": 6.194511032104492,
"epoch": 0.3929706817379018,
"grad_norm": 1.2734375,
"learning_rate": 0.0004989521880144988,
"loss": 5.9982,
"mean_token_accuracy": 0.17004551142454147,
"num_tokens": 8301026.0,
"step": 4450
},
{
"entropy": 6.184074640274048,
"epoch": 0.3934122218297421,
"grad_norm": 1.359375,
"learning_rate": 0.0004989491501643635,
"loss": 6.2578,
"mean_token_accuracy": 0.14575279951095582,
"num_tokens": 8310977.0,
"step": 4455
},
{
"entropy": 6.20936131477356,
"epoch": 0.3938537619215825,
"grad_norm": 1.0859375,
"learning_rate": 0.0004989461079271916,
"loss": 6.0296,
"mean_token_accuracy": 0.16169211864471436,
"num_tokens": 8319391.0,
"step": 4460
},
{
"entropy": 6.169527339935303,
"epoch": 0.39429530201342283,
"grad_norm": 1.1796875,
"learning_rate": 0.0004989430613030429,
"loss": 5.9922,
"mean_token_accuracy": 0.15554805994033813,
"num_tokens": 8328639.0,
"step": 4465
},
{
"entropy": 6.117289400100708,
"epoch": 0.39473684210526316,
"grad_norm": 1.2890625,
"learning_rate": 0.000498940010291977,
"loss": 6.096,
"mean_token_accuracy": 0.15155849754810333,
"num_tokens": 8338190.0,
"step": 4470
},
{
"entropy": 6.100512361526489,
"epoch": 0.3951783821971035,
"grad_norm": 1.171875,
"learning_rate": 0.0004989369548940536,
"loss": 6.0395,
"mean_token_accuracy": 0.15572706907987593,
"num_tokens": 8346874.0,
"step": 4475
},
{
"entropy": 6.190318775177002,
"epoch": 0.3956199222889438,
"grad_norm": 1.2265625,
"learning_rate": 0.0004989338951093327,
"loss": 6.0178,
"mean_token_accuracy": 0.16222874522209169,
"num_tokens": 8356446.0,
"step": 4480
},
{
"entropy": 6.1148622035980225,
"epoch": 0.39606146238078416,
"grad_norm": 1.140625,
"learning_rate": 0.0004989308309378741,
"loss": 6.0199,
"mean_token_accuracy": 0.15970418155193328,
"num_tokens": 8365694.0,
"step": 4485
},
{
"entropy": 6.157066297531128,
"epoch": 0.3965030024726245,
"grad_norm": 1.296875,
"learning_rate": 0.0004989277623797379,
"loss": 6.0754,
"mean_token_accuracy": 0.1534503474831581,
"num_tokens": 8374282.0,
"step": 4490
},
{
"entropy": 6.160027027130127,
"epoch": 0.3969445425644649,
"grad_norm": 1.15625,
"learning_rate": 0.0004989246894349841,
"loss": 6.0372,
"mean_token_accuracy": 0.1563117504119873,
"num_tokens": 8383315.0,
"step": 4495
},
{
"entropy": 6.214316082000733,
"epoch": 0.3973860826563052,
"grad_norm": 1.1796875,
"learning_rate": 0.0004989216121036732,
"loss": 6.0236,
"mean_token_accuracy": 0.15833714008331298,
"num_tokens": 8392263.0,
"step": 4500
},
{
"entropy": 6.106395578384399,
"epoch": 0.39782762274814554,
"grad_norm": 1.3671875,
"learning_rate": 0.0004989185303858651,
"loss": 6.0696,
"mean_token_accuracy": 0.14981550127267837,
"num_tokens": 8400734.0,
"step": 4505
},
{
"entropy": 6.156090450286865,
"epoch": 0.39826916283998587,
"grad_norm": 1.21875,
"learning_rate": 0.0004989154442816203,
"loss": 6.0793,
"mean_token_accuracy": 0.15283239632844925,
"num_tokens": 8410635.0,
"step": 4510
},
{
"entropy": 6.205539274215698,
"epoch": 0.3987107029318262,
"grad_norm": 1.3046875,
"learning_rate": 0.0004989123537909994,
"loss": 6.0664,
"mean_token_accuracy": 0.15485348254442216,
"num_tokens": 8420111.0,
"step": 4515
},
{
"entropy": 6.103611946105957,
"epoch": 0.39915224302366653,
"grad_norm": 1.2421875,
"learning_rate": 0.0004989092589140629,
"loss": 6.0177,
"mean_token_accuracy": 0.1484901040792465,
"num_tokens": 8429459.0,
"step": 4520
},
{
"entropy": 6.117227792739868,
"epoch": 0.39959378311550686,
"grad_norm": 1.3671875,
"learning_rate": 0.0004989061596508712,
"loss": 6.0403,
"mean_token_accuracy": 0.16086821481585503,
"num_tokens": 8438083.0,
"step": 4525
},
{
"entropy": 6.07325234413147,
"epoch": 0.40003532320734725,
"grad_norm": 1.1484375,
"learning_rate": 0.0004989030560014853,
"loss": 6.0506,
"mean_token_accuracy": 0.15744656324386597,
"num_tokens": 8447713.0,
"step": 4530
},
{
"entropy": 6.1523223400115965,
"epoch": 0.4004768632991876,
"grad_norm": 1.2421875,
"learning_rate": 0.0004988999479659657,
"loss": 6.0226,
"mean_token_accuracy": 0.158653824031353,
"num_tokens": 8457394.0,
"step": 4535
},
{
"entropy": 6.117883396148682,
"epoch": 0.4009184033910279,
"grad_norm": 1.296875,
"learning_rate": 0.0004988968355443737,
"loss": 5.9913,
"mean_token_accuracy": 0.16181344538927078,
"num_tokens": 8467333.0,
"step": 4540
},
{
"entropy": 6.124732971191406,
"epoch": 0.40135994348286824,
"grad_norm": 1.2421875,
"learning_rate": 0.0004988937187367699,
"loss": 6.1032,
"mean_token_accuracy": 0.15110900700092317,
"num_tokens": 8477530.0,
"step": 4545
},
{
"entropy": 6.217570829391479,
"epoch": 0.4018014835747086,
"grad_norm": 1.1328125,
"learning_rate": 0.0004988905975432154,
"loss": 6.1802,
"mean_token_accuracy": 0.15269524306058885,
"num_tokens": 8486861.0,
"step": 4550
},
{
"entropy": 6.170765733718872,
"epoch": 0.4022430236665489,
"grad_norm": 1.171875,
"learning_rate": 0.0004988874719637715,
"loss": 5.9867,
"mean_token_accuracy": 0.158355513215065,
"num_tokens": 8496541.0,
"step": 4555
},
{
"entropy": 6.052766561508179,
"epoch": 0.40268456375838924,
"grad_norm": 1.5234375,
"learning_rate": 0.0004988843419984994,
"loss": 6.064,
"mean_token_accuracy": 0.15494307354092599,
"num_tokens": 8505667.0,
"step": 4560
},
{
"entropy": 6.24229063987732,
"epoch": 0.4031261038502296,
"grad_norm": 1.1484375,
"learning_rate": 0.0004988812076474604,
"loss": 6.052,
"mean_token_accuracy": 0.15319542214274406,
"num_tokens": 8515133.0,
"step": 4565
},
{
"entropy": 6.170625066757202,
"epoch": 0.40356764394206995,
"grad_norm": 1.4140625,
"learning_rate": 0.0004988780689107158,
"loss": 6.007,
"mean_token_accuracy": 0.1620650038123131,
"num_tokens": 8524797.0,
"step": 4570
},
{
"entropy": 6.0161412239074705,
"epoch": 0.4040091840339103,
"grad_norm": 1.1015625,
"learning_rate": 0.0004988749257883271,
"loss": 6.0288,
"mean_token_accuracy": 0.15016194060444832,
"num_tokens": 8534667.0,
"step": 4575
},
{
"entropy": 6.143600559234619,
"epoch": 0.4044507241257506,
"grad_norm": 1.09375,
"learning_rate": 0.000498871778280356,
"loss": 6.0166,
"mean_token_accuracy": 0.15943924337625504,
"num_tokens": 8543874.0,
"step": 4580
},
{
"entropy": 6.1106805324554445,
"epoch": 0.40489226421759095,
"grad_norm": 1.1796875,
"learning_rate": 0.0004988686263868641,
"loss": 6.0353,
"mean_token_accuracy": 0.1529652863740921,
"num_tokens": 8553620.0,
"step": 4585
},
{
"entropy": 6.17731556892395,
"epoch": 0.4053338043094313,
"grad_norm": 1.2265625,
"learning_rate": 0.0004988654701079131,
"loss": 6.1113,
"mean_token_accuracy": 0.15474483817815782,
"num_tokens": 8563857.0,
"step": 4590
},
{
"entropy": 6.2202249526977536,
"epoch": 0.4057753444012716,
"grad_norm": 1.140625,
"learning_rate": 0.0004988623094435649,
"loss": 6.0898,
"mean_token_accuracy": 0.1531184583902359,
"num_tokens": 8572677.0,
"step": 4595
},
{
"entropy": 6.140727758407593,
"epoch": 0.406216884493112,
"grad_norm": 1.1796875,
"learning_rate": 0.0004988591443938813,
"loss": 6.016,
"mean_token_accuracy": 0.15535678565502167,
"num_tokens": 8581907.0,
"step": 4600
},
{
"entropy": 6.138199281692505,
"epoch": 0.4066584245849523,
"grad_norm": 1.1875,
"learning_rate": 0.0004988559749589244,
"loss": 6.0609,
"mean_token_accuracy": 0.15539143681526185,
"num_tokens": 8591701.0,
"step": 4605
},
{
"entropy": 6.19168872833252,
"epoch": 0.40709996467679266,
"grad_norm": 1.1328125,
"learning_rate": 0.0004988528011387563,
"loss": 6.1128,
"mean_token_accuracy": 0.15955741629004477,
"num_tokens": 8601054.0,
"step": 4610
},
{
"entropy": 6.115007019042968,
"epoch": 0.407541504768633,
"grad_norm": 1.2109375,
"learning_rate": 0.0004988496229334392,
"loss": 6.0089,
"mean_token_accuracy": 0.16271338164806365,
"num_tokens": 8610346.0,
"step": 4615
},
{
"entropy": 6.082312822341919,
"epoch": 0.4079830448604733,
"grad_norm": 1.3828125,
"learning_rate": 0.0004988464403430352,
"loss": 6.0114,
"mean_token_accuracy": 0.1530932977795601,
"num_tokens": 8620823.0,
"step": 4620
},
{
"entropy": 6.228973913192749,
"epoch": 0.40842458495231365,
"grad_norm": 1.109375,
"learning_rate": 0.0004988432533676067,
"loss": 6.1457,
"mean_token_accuracy": 0.14871701523661612,
"num_tokens": 8630184.0,
"step": 4625
},
{
"entropy": 6.265304517745972,
"epoch": 0.408866125044154,
"grad_norm": 1.390625,
"learning_rate": 0.0004988400620072163,
"loss": 6.1541,
"mean_token_accuracy": 0.14276653826236724,
"num_tokens": 8640064.0,
"step": 4630
},
{
"entropy": 6.173982954025268,
"epoch": 0.40930766513599437,
"grad_norm": 1.2109375,
"learning_rate": 0.0004988368662619263,
"loss": 6.0413,
"mean_token_accuracy": 0.15715423077344895,
"num_tokens": 8650503.0,
"step": 4635
},
{
"entropy": 6.191125011444091,
"epoch": 0.4097492052278347,
"grad_norm": 1.4296875,
"learning_rate": 0.0004988336661317994,
"loss": 6.0652,
"mean_token_accuracy": 0.15463789254426957,
"num_tokens": 8659125.0,
"step": 4640
},
{
"entropy": 6.157907199859619,
"epoch": 0.41019074531967503,
"grad_norm": 1.109375,
"learning_rate": 0.0004988304616168984,
"loss": 6.0841,
"mean_token_accuracy": 0.15441161543130874,
"num_tokens": 8668193.0,
"step": 4645
},
{
"entropy": 6.116786527633667,
"epoch": 0.41063228541151536,
"grad_norm": 1.21875,
"learning_rate": 0.0004988272527172858,
"loss": 6.0384,
"mean_token_accuracy": 0.15524010509252548,
"num_tokens": 8677515.0,
"step": 4650
},
{
"entropy": 6.21679277420044,
"epoch": 0.4110738255033557,
"grad_norm": 1.3515625,
"learning_rate": 0.0004988240394330246,
"loss": 6.0764,
"mean_token_accuracy": 0.14973534047603607,
"num_tokens": 8687549.0,
"step": 4655
},
{
"entropy": 6.157021522521973,
"epoch": 0.411515365595196,
"grad_norm": 1.375,
"learning_rate": 0.0004988208217641778,
"loss": 6.0639,
"mean_token_accuracy": 0.15594548732042313,
"num_tokens": 8697371.0,
"step": 4660
},
{
"entropy": 6.13969612121582,
"epoch": 0.41195690568703636,
"grad_norm": 1.5,
"learning_rate": 0.0004988175997108086,
"loss": 6.0735,
"mean_token_accuracy": 0.15609194859862327,
"num_tokens": 8707193.0,
"step": 4665
},
{
"entropy": 6.1602945804595945,
"epoch": 0.41239844577887674,
"grad_norm": 1.1953125,
"learning_rate": 0.0004988143732729797,
"loss": 6.0644,
"mean_token_accuracy": 0.15450926274061202,
"num_tokens": 8716052.0,
"step": 4670
},
{
"entropy": 6.081143188476562,
"epoch": 0.4128399858707171,
"grad_norm": 1.3671875,
"learning_rate": 0.0004988111424507546,
"loss": 6.0209,
"mean_token_accuracy": 0.1590244859457016,
"num_tokens": 8726140.0,
"step": 4675
},
{
"entropy": 6.134521245956421,
"epoch": 0.4132815259625574,
"grad_norm": 1.3515625,
"learning_rate": 0.0004988079072441964,
"loss": 6.0028,
"mean_token_accuracy": 0.16453344523906707,
"num_tokens": 8735299.0,
"step": 4680
},
{
"entropy": 6.157465314865112,
"epoch": 0.41372306605439774,
"grad_norm": 1.3046875,
"learning_rate": 0.0004988046676533687,
"loss": 6.0908,
"mean_token_accuracy": 0.15370513945817948,
"num_tokens": 8744686.0,
"step": 4685
},
{
"entropy": 6.140842294692993,
"epoch": 0.41416460614623807,
"grad_norm": 1.203125,
"learning_rate": 0.0004988014236783347,
"loss": 6.063,
"mean_token_accuracy": 0.1632717102766037,
"num_tokens": 8754942.0,
"step": 4690
},
{
"entropy": 6.232805824279785,
"epoch": 0.4146061462380784,
"grad_norm": 1.3203125,
"learning_rate": 0.0004987981753191582,
"loss": 6.0523,
"mean_token_accuracy": 0.15246548503637314,
"num_tokens": 8764054.0,
"step": 4695
},
{
"entropy": 6.0288361549377445,
"epoch": 0.41504768632991873,
"grad_norm": 1.296875,
"learning_rate": 0.0004987949225759027,
"loss": 5.9405,
"mean_token_accuracy": 0.15906094312667846,
"num_tokens": 8773050.0,
"step": 4700
},
{
"entropy": 6.1974766731262205,
"epoch": 0.4154892264217591,
"grad_norm": 1.3359375,
"learning_rate": 0.0004987916654486321,
"loss": 6.0544,
"mean_token_accuracy": 0.1537775442004204,
"num_tokens": 8782476.0,
"step": 4705
},
{
"entropy": 6.254334783554077,
"epoch": 0.41593076651359945,
"grad_norm": 1.21875,
"learning_rate": 0.0004987884039374099,
"loss": 6.1212,
"mean_token_accuracy": 0.1502728283405304,
"num_tokens": 8791147.0,
"step": 4710
},
{
"entropy": 6.237736749649048,
"epoch": 0.4163723066054398,
"grad_norm": 1.3203125,
"learning_rate": 0.0004987851380423001,
"loss": 6.1069,
"mean_token_accuracy": 0.1571350358426571,
"num_tokens": 8801151.0,
"step": 4715
},
{
"entropy": 6.067666149139404,
"epoch": 0.4168138466972801,
"grad_norm": 1.40625,
"learning_rate": 0.0004987818677633668,
"loss": 6.0577,
"mean_token_accuracy": 0.14970119222998618,
"num_tokens": 8809587.0,
"step": 4720
},
{
"entropy": 6.103376960754394,
"epoch": 0.41725538678912044,
"grad_norm": 1.3359375,
"learning_rate": 0.000498778593100674,
"loss": 5.94,
"mean_token_accuracy": 0.15202507078647615,
"num_tokens": 8818168.0,
"step": 4725
},
{
"entropy": 6.131261110305786,
"epoch": 0.4176969268809608,
"grad_norm": 1.171875,
"learning_rate": 0.0004987753140542857,
"loss": 6.0105,
"mean_token_accuracy": 0.16188293397426606,
"num_tokens": 8827477.0,
"step": 4730
},
{
"entropy": 6.104963493347168,
"epoch": 0.4181384669728011,
"grad_norm": 1.28125,
"learning_rate": 0.0004987720306242664,
"loss": 5.9847,
"mean_token_accuracy": 0.16286925673484803,
"num_tokens": 8837067.0,
"step": 4735
},
{
"entropy": 6.133073472976685,
"epoch": 0.4185800070646415,
"grad_norm": 1.3046875,
"learning_rate": 0.0004987687428106803,
"loss": 6.0872,
"mean_token_accuracy": 0.16172488033771515,
"num_tokens": 8845790.0,
"step": 4740
},
{
"entropy": 6.185379123687744,
"epoch": 0.4190215471564818,
"grad_norm": 1.5703125,
"learning_rate": 0.0004987654506135917,
"loss": 6.0226,
"mean_token_accuracy": 0.15944662541151047,
"num_tokens": 8855242.0,
"step": 4745
},
{
"entropy": 6.192673587799073,
"epoch": 0.41946308724832215,
"grad_norm": 1.265625,
"learning_rate": 0.0004987621540330652,
"loss": 6.0723,
"mean_token_accuracy": 0.15567026063799858,
"num_tokens": 8864459.0,
"step": 4750
},
{
"entropy": 6.19939341545105,
"epoch": 0.4199046273401625,
"grad_norm": 1.4296875,
"learning_rate": 0.0004987588530691653,
"loss": 6.1385,
"mean_token_accuracy": 0.14711003005504608,
"num_tokens": 8875028.0,
"step": 4755
},
{
"entropy": 6.135413789749146,
"epoch": 0.4203461674320028,
"grad_norm": 1.6640625,
"learning_rate": 0.0004987555477219569,
"loss": 5.9786,
"mean_token_accuracy": 0.15931818783283233,
"num_tokens": 8883857.0,
"step": 4760
},
{
"entropy": 6.110503196716309,
"epoch": 0.42078770752384315,
"grad_norm": 1.3125,
"learning_rate": 0.0004987522379915045,
"loss": 5.9705,
"mean_token_accuracy": 0.15814343243837356,
"num_tokens": 8893499.0,
"step": 4765
},
{
"entropy": 6.095895624160766,
"epoch": 0.4212292476156835,
"grad_norm": 1.25,
"learning_rate": 0.000498748923877873,
"loss": 5.9891,
"mean_token_accuracy": 0.1591852620244026,
"num_tokens": 8903368.0,
"step": 4770
},
{
"entropy": 6.126567029953003,
"epoch": 0.42167078770752386,
"grad_norm": 1.3984375,
"learning_rate": 0.0004987456053811273,
"loss": 6.0966,
"mean_token_accuracy": 0.1539273589849472,
"num_tokens": 8912701.0,
"step": 4775
},
{
"entropy": 6.1158490657806395,
"epoch": 0.4221123277993642,
"grad_norm": 1.2578125,
"learning_rate": 0.0004987422825013325,
"loss": 6.0444,
"mean_token_accuracy": 0.15109995752573013,
"num_tokens": 8921962.0,
"step": 4780
},
{
"entropy": 6.1833864688873295,
"epoch": 0.4225538678912045,
"grad_norm": 1.3515625,
"learning_rate": 0.0004987389552385536,
"loss": 6.0307,
"mean_token_accuracy": 0.15406385958194732,
"num_tokens": 8931923.0,
"step": 4785
},
{
"entropy": 6.131430625915527,
"epoch": 0.42299540798304486,
"grad_norm": 1.71875,
"learning_rate": 0.0004987356235928558,
"loss": 6.0635,
"mean_token_accuracy": 0.1470622941851616,
"num_tokens": 8940403.0,
"step": 4790
},
{
"entropy": 6.131550025939942,
"epoch": 0.4234369480748852,
"grad_norm": 1.1953125,
"learning_rate": 0.0004987322875643044,
"loss": 5.9887,
"mean_token_accuracy": 0.16142310500144957,
"num_tokens": 8949377.0,
"step": 4795
},
{
"entropy": 6.2069591045379635,
"epoch": 0.4238784881667255,
"grad_norm": 1.46875,
"learning_rate": 0.0004987289471529647,
"loss": 6.1304,
"mean_token_accuracy": 0.1479768604040146,
"num_tokens": 8958719.0,
"step": 4800
},
{
"entropy": 6.165348720550537,
"epoch": 0.42432002825856585,
"grad_norm": 1.234375,
"learning_rate": 0.0004987256023589022,
"loss": 6.1048,
"mean_token_accuracy": 0.14835046380758285,
"num_tokens": 8968226.0,
"step": 4805
},
{
"entropy": 6.13749942779541,
"epoch": 0.42476156835040624,
"grad_norm": 1.296875,
"learning_rate": 0.0004987222531821824,
"loss": 6.0528,
"mean_token_accuracy": 0.14842675924301146,
"num_tokens": 8976670.0,
"step": 4810
},
{
"entropy": 6.09095401763916,
"epoch": 0.42520310844224657,
"grad_norm": 1.171875,
"learning_rate": 0.0004987188996228709,
"loss": 5.9901,
"mean_token_accuracy": 0.16065036058425902,
"num_tokens": 8986185.0,
"step": 4815
},
{
"entropy": 6.155765628814697,
"epoch": 0.4256446485340869,
"grad_norm": 1.2578125,
"learning_rate": 0.0004987155416810334,
"loss": 5.9644,
"mean_token_accuracy": 0.16488435715436936,
"num_tokens": 8995124.0,
"step": 4820
},
{
"entropy": 6.094462633132935,
"epoch": 0.42608618862592723,
"grad_norm": 1.1796875,
"learning_rate": 0.0004987121793567356,
"loss": 6.0704,
"mean_token_accuracy": 0.15586088821291924,
"num_tokens": 9004380.0,
"step": 4825
},
{
"entropy": 6.142639970779419,
"epoch": 0.42652772871776756,
"grad_norm": 1.25,
"learning_rate": 0.0004987088126500436,
"loss": 5.963,
"mean_token_accuracy": 0.16008084118366242,
"num_tokens": 9013791.0,
"step": 4830
},
{
"entropy": 6.069281530380249,
"epoch": 0.4269692688096079,
"grad_norm": 1.234375,
"learning_rate": 0.000498705441561023,
"loss": 5.9998,
"mean_token_accuracy": 0.16714757829904556,
"num_tokens": 9023076.0,
"step": 4835
},
{
"entropy": 6.315524005889893,
"epoch": 0.4274108089014482,
"grad_norm": 1.21875,
"learning_rate": 0.0004987020660897401,
"loss": 6.1089,
"mean_token_accuracy": 0.14946657419204712,
"num_tokens": 9032720.0,
"step": 4840
},
{
"entropy": 6.199988222122192,
"epoch": 0.4278523489932886,
"grad_norm": 1.1953125,
"learning_rate": 0.000498698686236261,
"loss": 6.0577,
"mean_token_accuracy": 0.1480626255273819,
"num_tokens": 9042652.0,
"step": 4845
},
{
"entropy": 6.085298490524292,
"epoch": 0.42829388908512894,
"grad_norm": 1.3203125,
"learning_rate": 0.0004986953020006519,
"loss": 6.1376,
"mean_token_accuracy": 0.14969860166311263,
"num_tokens": 9052172.0,
"step": 4850
},
{
"entropy": 6.134743070602417,
"epoch": 0.4287354291769693,
"grad_norm": 1.140625,
"learning_rate": 0.0004986919133829788,
"loss": 5.9956,
"mean_token_accuracy": 0.15985623747110367,
"num_tokens": 9061798.0,
"step": 4855
},
{
"entropy": 6.040613985061645,
"epoch": 0.4291769692688096,
"grad_norm": 1.2109375,
"learning_rate": 0.0004986885203833086,
"loss": 5.9499,
"mean_token_accuracy": 0.15626863837242128,
"num_tokens": 9070429.0,
"step": 4860
},
{
"entropy": 6.098531293869018,
"epoch": 0.42961850936064994,
"grad_norm": 1.1953125,
"learning_rate": 0.0004986851230017075,
"loss": 6.0492,
"mean_token_accuracy": 0.16342882812023163,
"num_tokens": 9080200.0,
"step": 4865
},
{
"entropy": 6.163259077072143,
"epoch": 0.43006004945249027,
"grad_norm": 1.21875,
"learning_rate": 0.0004986817212382419,
"loss": 6.0382,
"mean_token_accuracy": 0.15508455336093901,
"num_tokens": 9089555.0,
"step": 4870
},
{
"entropy": 6.126989889144897,
"epoch": 0.4305015895443306,
"grad_norm": 1.234375,
"learning_rate": 0.0004986783150929786,
"loss": 6.0263,
"mean_token_accuracy": 0.15248029232025145,
"num_tokens": 9099091.0,
"step": 4875
},
{
"entropy": 6.172487115859985,
"epoch": 0.430943129636171,
"grad_norm": 1.234375,
"learning_rate": 0.0004986749045659845,
"loss": 6.1075,
"mean_token_accuracy": 0.14817112535238267,
"num_tokens": 9109008.0,
"step": 4880
},
{
"entropy": 6.152895927429199,
"epoch": 0.4313846697280113,
"grad_norm": 1.4921875,
"learning_rate": 0.0004986714896573261,
"loss": 5.9513,
"mean_token_accuracy": 0.16084639877080917,
"num_tokens": 9117882.0,
"step": 4885
},
{
"entropy": 6.139704322814941,
"epoch": 0.43182620981985165,
"grad_norm": 1.3125,
"learning_rate": 0.0004986680703670704,
"loss": 6.1215,
"mean_token_accuracy": 0.15487379878759383,
"num_tokens": 9126860.0,
"step": 4890
},
{
"entropy": 6.141199541091919,
"epoch": 0.432267749911692,
"grad_norm": 1.3828125,
"learning_rate": 0.0004986646466952845,
"loss": 6.0119,
"mean_token_accuracy": 0.15635189563035964,
"num_tokens": 9135819.0,
"step": 4895
},
{
"entropy": 6.10592885017395,
"epoch": 0.4327092900035323,
"grad_norm": 1.1640625,
"learning_rate": 0.0004986612186420353,
"loss": 5.9934,
"mean_token_accuracy": 0.15510470867156984,
"num_tokens": 9145302.0,
"step": 4900
},
{
"entropy": 6.044864749908447,
"epoch": 0.43315083009537264,
"grad_norm": 1.234375,
"learning_rate": 0.0004986577862073901,
"loss": 6.0575,
"mean_token_accuracy": 0.15394357293844224,
"num_tokens": 9154667.0,
"step": 4905
},
{
"entropy": 6.2363903522491455,
"epoch": 0.43359237018721297,
"grad_norm": 1.2578125,
"learning_rate": 0.0004986543493914159,
"loss": 6.098,
"mean_token_accuracy": 0.14959411323070526,
"num_tokens": 9164562.0,
"step": 4910
},
{
"entropy": 6.208088731765747,
"epoch": 0.43403391027905336,
"grad_norm": 1.296875,
"learning_rate": 0.0004986509081941805,
"loss": 6.1071,
"mean_token_accuracy": 0.15276289731264114,
"num_tokens": 9174872.0,
"step": 4915
},
{
"entropy": 6.079789733886718,
"epoch": 0.4344754503708937,
"grad_norm": 1.28125,
"learning_rate": 0.0004986474626157507,
"loss": 5.862,
"mean_token_accuracy": 0.17531196177005767,
"num_tokens": 9184322.0,
"step": 4920
},
{
"entropy": 6.058652734756469,
"epoch": 0.434916990462734,
"grad_norm": 1.25,
"learning_rate": 0.0004986440126561945,
"loss": 5.9894,
"mean_token_accuracy": 0.15689075142145156,
"num_tokens": 9194450.0,
"step": 4925
},
{
"entropy": 6.151858711242676,
"epoch": 0.43535853055457435,
"grad_norm": 1.4140625,
"learning_rate": 0.0004986405583155792,
"loss": 6.0305,
"mean_token_accuracy": 0.15203123837709426,
"num_tokens": 9203658.0,
"step": 4930
},
{
"entropy": 6.100604057312012,
"epoch": 0.4358000706464147,
"grad_norm": 1.1328125,
"learning_rate": 0.0004986370995939725,
"loss": 6.0239,
"mean_token_accuracy": 0.15403898507356645,
"num_tokens": 9213609.0,
"step": 4935
},
{
"entropy": 6.112157011032105,
"epoch": 0.436241610738255,
"grad_norm": 1.171875,
"learning_rate": 0.0004986336364914423,
"loss": 6.0414,
"mean_token_accuracy": 0.15200137123465537,
"num_tokens": 9222704.0,
"step": 4940
},
{
"entropy": 6.147404766082763,
"epoch": 0.43668315083009535,
"grad_norm": 1.484375,
"learning_rate": 0.0004986301690080564,
"loss": 6.0472,
"mean_token_accuracy": 0.1494756668806076,
"num_tokens": 9231599.0,
"step": 4945
},
{
"entropy": 6.176404428482056,
"epoch": 0.43712469092193573,
"grad_norm": 1.234375,
"learning_rate": 0.0004986266971438826,
"loss": 6.0761,
"mean_token_accuracy": 0.15215079635381698,
"num_tokens": 9241886.0,
"step": 4950
},
{
"entropy": 6.142639207839966,
"epoch": 0.43756623101377606,
"grad_norm": 1.34375,
"learning_rate": 0.000498623220898989,
"loss": 6.0651,
"mean_token_accuracy": 0.1492701292037964,
"num_tokens": 9251085.0,
"step": 4955
},
{
"entropy": 6.155983543395996,
"epoch": 0.4380077711056164,
"grad_norm": 1.4609375,
"learning_rate": 0.0004986197402734436,
"loss": 6.0272,
"mean_token_accuracy": 0.15382544845342636,
"num_tokens": 9259601.0,
"step": 4960
},
{
"entropy": 6.18567385673523,
"epoch": 0.4384493111974567,
"grad_norm": 1.2578125,
"learning_rate": 0.0004986162552673148,
"loss": 6.0699,
"mean_token_accuracy": 0.15705521255731583,
"num_tokens": 9268935.0,
"step": 4965
},
{
"entropy": 6.133043241500855,
"epoch": 0.43889085128929706,
"grad_norm": 1.4921875,
"learning_rate": 0.0004986127658806706,
"loss": 6.0813,
"mean_token_accuracy": 0.15293170362710953,
"num_tokens": 9277647.0,
"step": 4970
},
{
"entropy": 6.120587873458862,
"epoch": 0.4393323913811374,
"grad_norm": 1.2109375,
"learning_rate": 0.0004986092721135796,
"loss": 6.0199,
"mean_token_accuracy": 0.15239207521080972,
"num_tokens": 9286610.0,
"step": 4975
},
{
"entropy": 6.135066556930542,
"epoch": 0.4397739314729777,
"grad_norm": 1.2421875,
"learning_rate": 0.0004986057739661101,
"loss": 6.1032,
"mean_token_accuracy": 0.1508208692073822,
"num_tokens": 9295946.0,
"step": 4980
},
{
"entropy": 6.1115028858184814,
"epoch": 0.4402154715648181,
"grad_norm": 1.2109375,
"learning_rate": 0.0004986022714383307,
"loss": 6.0042,
"mean_token_accuracy": 0.1543491631746292,
"num_tokens": 9304903.0,
"step": 4985
},
{
"entropy": 6.15944766998291,
"epoch": 0.44065701165665844,
"grad_norm": 1.265625,
"learning_rate": 0.0004985987645303099,
"loss": 5.9185,
"mean_token_accuracy": 0.16130532771348954,
"num_tokens": 9313606.0,
"step": 4990
},
{
"entropy": 6.016908359527588,
"epoch": 0.44109855174849877,
"grad_norm": 1.296875,
"learning_rate": 0.0004985952532421164,
"loss": 6.0275,
"mean_token_accuracy": 0.15621849447488784,
"num_tokens": 9322815.0,
"step": 4995
},
{
"entropy": 6.17736701965332,
"epoch": 0.4415400918403391,
"grad_norm": 1.2109375,
"learning_rate": 0.0004985917375738193,
"loss": 5.9593,
"mean_token_accuracy": 0.15536014586687089,
"num_tokens": 9332630.0,
"step": 5000
},
{
"entropy": 6.067361211776733,
"epoch": 0.44198163193217943,
"grad_norm": 1.2109375,
"learning_rate": 0.0004985882175254871,
"loss": 5.9846,
"mean_token_accuracy": 0.15410226881504058,
"num_tokens": 9342216.0,
"step": 5005
},
{
"entropy": 6.146738815307617,
"epoch": 0.44242317202401976,
"grad_norm": 1.1484375,
"learning_rate": 0.0004985846930971887,
"loss": 6.0431,
"mean_token_accuracy": 0.1559235379099846,
"num_tokens": 9352295.0,
"step": 5010
},
{
"entropy": 6.107598447799683,
"epoch": 0.4428647121158601,
"grad_norm": 1.328125,
"learning_rate": 0.0004985811642889937,
"loss": 6.0348,
"mean_token_accuracy": 0.15718846172094345,
"num_tokens": 9361462.0,
"step": 5015
},
{
"entropy": 6.144611740112305,
"epoch": 0.4433062522077005,
"grad_norm": 1.234375,
"learning_rate": 0.0004985776311009705,
"loss": 5.9684,
"mean_token_accuracy": 0.15345567613840103,
"num_tokens": 9370379.0,
"step": 5020
},
{
"entropy": 6.103248453140258,
"epoch": 0.4437477922995408,
"grad_norm": 1.3046875,
"learning_rate": 0.0004985740935331888,
"loss": 6.0098,
"mean_token_accuracy": 0.15823583900928498,
"num_tokens": 9379255.0,
"step": 5025
},
{
"entropy": 6.085464525222778,
"epoch": 0.44418933239138114,
"grad_norm": 1.2421875,
"learning_rate": 0.0004985705515857177,
"loss": 5.99,
"mean_token_accuracy": 0.14899933189153672,
"num_tokens": 9389313.0,
"step": 5030
},
{
"entropy": 6.020923233032226,
"epoch": 0.4446308724832215,
"grad_norm": 1.203125,
"learning_rate": 0.0004985670052586268,
"loss": 5.7916,
"mean_token_accuracy": 0.17029385417699813,
"num_tokens": 9397778.0,
"step": 5035
},
{
"entropy": 6.082318449020386,
"epoch": 0.4450724125750618,
"grad_norm": 1.234375,
"learning_rate": 0.0004985634545519853,
"loss": 6.0589,
"mean_token_accuracy": 0.15298160612583162,
"num_tokens": 9407831.0,
"step": 5040
},
{
"entropy": 6.0772803783416744,
"epoch": 0.44551395266690214,
"grad_norm": 1.53125,
"learning_rate": 0.0004985598994658629,
"loss": 5.9669,
"mean_token_accuracy": 0.15956881046295165,
"num_tokens": 9418458.0,
"step": 5045
},
{
"entropy": 6.016619396209717,
"epoch": 0.44595549275874247,
"grad_norm": 1.21875,
"learning_rate": 0.0004985563400003291,
"loss": 5.8707,
"mean_token_accuracy": 0.17422997653484346,
"num_tokens": 9426911.0,
"step": 5050
},
{
"entropy": 6.14707236289978,
"epoch": 0.44639703285058285,
"grad_norm": 1.2421875,
"learning_rate": 0.0004985527761554539,
"loss": 6.0541,
"mean_token_accuracy": 0.1595864400267601,
"num_tokens": 9435896.0,
"step": 5055
},
{
"entropy": 6.207449245452881,
"epoch": 0.4468385729424232,
"grad_norm": 1.296875,
"learning_rate": 0.000498549207931307,
"loss": 6.0612,
"mean_token_accuracy": 0.15011950582265854,
"num_tokens": 9445272.0,
"step": 5060
},
{
"entropy": 6.06727409362793,
"epoch": 0.4472801130342635,
"grad_norm": 1.3359375,
"learning_rate": 0.0004985456353279581,
"loss": 6.0228,
"mean_token_accuracy": 0.15027147233486177,
"num_tokens": 9455377.0,
"step": 5065
},
{
"entropy": 6.108924150466919,
"epoch": 0.44772165312610385,
"grad_norm": 1.2734375,
"learning_rate": 0.0004985420583454774,
"loss": 6.0477,
"mean_token_accuracy": 0.1528654247522354,
"num_tokens": 9464918.0,
"step": 5070
},
{
"entropy": 6.118135070800781,
"epoch": 0.4481631932179442,
"grad_norm": 1.3203125,
"learning_rate": 0.0004985384769839349,
"loss": 6.0416,
"mean_token_accuracy": 0.15041410326957702,
"num_tokens": 9473322.0,
"step": 5075
},
{
"entropy": 6.191469526290893,
"epoch": 0.4486047333097845,
"grad_norm": 1.203125,
"learning_rate": 0.0004985348912434008,
"loss": 5.969,
"mean_token_accuracy": 0.16189600080251693,
"num_tokens": 9482255.0,
"step": 5080
},
{
"entropy": 6.092324352264404,
"epoch": 0.44904627340162484,
"grad_norm": 1.328125,
"learning_rate": 0.0004985313011239452,
"loss": 5.9959,
"mean_token_accuracy": 0.1581245869398117,
"num_tokens": 9491709.0,
"step": 5085
},
{
"entropy": 6.024058151245117,
"epoch": 0.4494878134934652,
"grad_norm": 1.296875,
"learning_rate": 0.0004985277066256388,
"loss": 5.9807,
"mean_token_accuracy": 0.16034325063228608,
"num_tokens": 9500594.0,
"step": 5090
},
{
"entropy": 6.112436008453369,
"epoch": 0.44992935358530556,
"grad_norm": 1.296875,
"learning_rate": 0.0004985241077485515,
"loss": 6.0018,
"mean_token_accuracy": 0.15738717019557952,
"num_tokens": 9509088.0,
"step": 5095
},
{
"entropy": 6.149496126174927,
"epoch": 0.4503708936771459,
"grad_norm": 1.4609375,
"learning_rate": 0.0004985205044927541,
"loss": 5.9725,
"mean_token_accuracy": 0.15938366055488587,
"num_tokens": 9517776.0,
"step": 5100
},
{
"entropy": 6.069735097885132,
"epoch": 0.4508124337689862,
"grad_norm": 1.2578125,
"learning_rate": 0.0004985168968583173,
"loss": 6.0348,
"mean_token_accuracy": 0.15837667435407637,
"num_tokens": 9527080.0,
"step": 5105
},
{
"entropy": 6.106137084960937,
"epoch": 0.45125397386082655,
"grad_norm": 1.265625,
"learning_rate": 0.0004985132848453114,
"loss": 5.8952,
"mean_token_accuracy": 0.16738586127758026,
"num_tokens": 9536358.0,
"step": 5110
},
{
"entropy": 6.04980616569519,
"epoch": 0.4516955139526669,
"grad_norm": 1.296875,
"learning_rate": 0.0004985096684538075,
"loss": 5.9813,
"mean_token_accuracy": 0.15662433505058287,
"num_tokens": 9545528.0,
"step": 5115
},
{
"entropy": 6.060785722732544,
"epoch": 0.4521370540445072,
"grad_norm": 1.3515625,
"learning_rate": 0.0004985060476838763,
"loss": 6.0113,
"mean_token_accuracy": 0.15891691744327546,
"num_tokens": 9554433.0,
"step": 5120
},
{
"entropy": 6.073312520980835,
"epoch": 0.4525785941363476,
"grad_norm": 1.3125,
"learning_rate": 0.0004985024225355887,
"loss": 5.9656,
"mean_token_accuracy": 0.1543935567140579,
"num_tokens": 9563932.0,
"step": 5125
},
{
"entropy": 6.100546646118164,
"epoch": 0.45302013422818793,
"grad_norm": 1.1953125,
"learning_rate": 0.0004984987930090158,
"loss": 5.9552,
"mean_token_accuracy": 0.15897123962640763,
"num_tokens": 9572829.0,
"step": 5130
},
{
"entropy": 6.025969839096069,
"epoch": 0.45346167432002826,
"grad_norm": 1.203125,
"learning_rate": 0.0004984951591042285,
"loss": 6.0045,
"mean_token_accuracy": 0.16426790058612822,
"num_tokens": 9583597.0,
"step": 5135
},
{
"entropy": 6.24551477432251,
"epoch": 0.4539032144118686,
"grad_norm": 1.25,
"learning_rate": 0.0004984915208212983,
"loss": 6.0912,
"mean_token_accuracy": 0.15156230181455613,
"num_tokens": 9593114.0,
"step": 5140
},
{
"entropy": 6.167973184585572,
"epoch": 0.4543447545037089,
"grad_norm": 1.1953125,
"learning_rate": 0.0004984878781602964,
"loss": 5.9722,
"mean_token_accuracy": 0.1515617176890373,
"num_tokens": 9601454.0,
"step": 5145
},
{
"entropy": 6.117550992965699,
"epoch": 0.45478629459554926,
"grad_norm": 1.203125,
"learning_rate": 0.0004984842311212939,
"loss": 6.0266,
"mean_token_accuracy": 0.1547075927257538,
"num_tokens": 9611994.0,
"step": 5150
},
{
"entropy": 6.099160146713257,
"epoch": 0.4552278346873896,
"grad_norm": 1.2890625,
"learning_rate": 0.0004984805797043625,
"loss": 5.9819,
"mean_token_accuracy": 0.1554704263806343,
"num_tokens": 9621318.0,
"step": 5155
},
{
"entropy": 6.086650943756103,
"epoch": 0.45566937477923,
"grad_norm": 1.3359375,
"learning_rate": 0.0004984769239095736,
"loss": 5.9871,
"mean_token_accuracy": 0.1622908428311348,
"num_tokens": 9630270.0,
"step": 5160
},
{
"entropy": 6.161284017562866,
"epoch": 0.4561109148710703,
"grad_norm": 1.1953125,
"learning_rate": 0.0004984732637369989,
"loss": 6.0164,
"mean_token_accuracy": 0.15295830443501474,
"num_tokens": 9640391.0,
"step": 5165
},
{
"entropy": 6.111026573181152,
"epoch": 0.45655245496291064,
"grad_norm": 1.828125,
"learning_rate": 0.0004984695991867099,
"loss": 6.0302,
"mean_token_accuracy": 0.15423648655414582,
"num_tokens": 9648827.0,
"step": 5170
},
{
"entropy": 6.066336679458618,
"epoch": 0.45699399505475097,
"grad_norm": 1.6484375,
"learning_rate": 0.0004984659302587788,
"loss": 5.9651,
"mean_token_accuracy": 0.1540757015347481,
"num_tokens": 9657940.0,
"step": 5175
},
{
"entropy": 6.132708024978638,
"epoch": 0.4574355351465913,
"grad_norm": 1.25,
"learning_rate": 0.000498462256953277,
"loss": 6.0864,
"mean_token_accuracy": 0.15138714611530305,
"num_tokens": 9668296.0,
"step": 5180
},
{
"entropy": 6.187513446807861,
"epoch": 0.45787707523843163,
"grad_norm": 1.421875,
"learning_rate": 0.0004984585792702767,
"loss": 6.0401,
"mean_token_accuracy": 0.14838093519210815,
"num_tokens": 9677914.0,
"step": 5185
},
{
"entropy": 6.183677721023559,
"epoch": 0.45831861533027196,
"grad_norm": 1.34375,
"learning_rate": 0.0004984548972098501,
"loss": 6.174,
"mean_token_accuracy": 0.13970668166875838,
"num_tokens": 9687339.0,
"step": 5190
},
{
"entropy": 6.133589363098144,
"epoch": 0.45876015542211235,
"grad_norm": 1.546875,
"learning_rate": 0.000498451210772069,
"loss": 5.9454,
"mean_token_accuracy": 0.16496401354670526,
"num_tokens": 9696889.0,
"step": 5195
},
{
"entropy": 6.0567436695098875,
"epoch": 0.4592016955139527,
"grad_norm": 1.4375,
"learning_rate": 0.0004984475199570058,
"loss": 5.9136,
"mean_token_accuracy": 0.1610390767455101,
"num_tokens": 9705498.0,
"step": 5200
},
{
"entropy": 6.099144554138183,
"epoch": 0.459643235605793,
"grad_norm": 2.046875,
"learning_rate": 0.0004984438247647329,
"loss": 5.9902,
"mean_token_accuracy": 0.15816803127527237,
"num_tokens": 9713936.0,
"step": 5205
},
{
"entropy": 6.152586174011231,
"epoch": 0.46008477569763334,
"grad_norm": 1.3671875,
"learning_rate": 0.0004984401251953223,
"loss": 6.06,
"mean_token_accuracy": 0.15645882338285447,
"num_tokens": 9723924.0,
"step": 5210
},
{
"entropy": 6.179743766784668,
"epoch": 0.4605263157894737,
"grad_norm": 1.21875,
"learning_rate": 0.0004984364212488469,
"loss": 6.0873,
"mean_token_accuracy": 0.15282048285007477,
"num_tokens": 9733427.0,
"step": 5215
},
{
"entropy": 6.035888767242431,
"epoch": 0.460967855881314,
"grad_norm": 1.3828125,
"learning_rate": 0.0004984327129253789,
"loss": 5.8837,
"mean_token_accuracy": 0.16696648448705673,
"num_tokens": 9742054.0,
"step": 5220
},
{
"entropy": 6.049908781051636,
"epoch": 0.46140939597315433,
"grad_norm": 1.25,
"learning_rate": 0.0004984290002249914,
"loss": 6.063,
"mean_token_accuracy": 0.1488596171140671,
"num_tokens": 9752447.0,
"step": 5225
},
{
"entropy": 6.131543159484863,
"epoch": 0.4618509360649947,
"grad_norm": 1.5,
"learning_rate": 0.0004984252831477567,
"loss": 5.8886,
"mean_token_accuracy": 0.16800648123025894,
"num_tokens": 9760878.0,
"step": 5230
},
{
"entropy": 5.97862868309021,
"epoch": 0.46229247615683505,
"grad_norm": 1.2890625,
"learning_rate": 0.0004984215616937477,
"loss": 5.9523,
"mean_token_accuracy": 0.1603931352496147,
"num_tokens": 9770200.0,
"step": 5235
},
{
"entropy": 6.214956617355346,
"epoch": 0.4627340162486754,
"grad_norm": 1.2265625,
"learning_rate": 0.0004984178358630374,
"loss": 6.0469,
"mean_token_accuracy": 0.15920519679784775,
"num_tokens": 9780303.0,
"step": 5240
},
{
"entropy": 6.130382776260376,
"epoch": 0.4631755563405157,
"grad_norm": 1.296875,
"learning_rate": 0.0004984141056556989,
"loss": 5.9529,
"mean_token_accuracy": 0.16305534839630126,
"num_tokens": 9790248.0,
"step": 5245
},
{
"entropy": 6.11019229888916,
"epoch": 0.46361709643235605,
"grad_norm": 1.3046875,
"learning_rate": 0.0004984103710718051,
"loss": 6.0503,
"mean_token_accuracy": 0.1460764303803444,
"num_tokens": 9799345.0,
"step": 5250
},
{
"entropy": 6.091036796569824,
"epoch": 0.4640586365241964,
"grad_norm": 1.390625,
"learning_rate": 0.000498406632111429,
"loss": 5.9308,
"mean_token_accuracy": 0.15862552225589752,
"num_tokens": 9808561.0,
"step": 5255
},
{
"entropy": 6.106261253356934,
"epoch": 0.4645001766160367,
"grad_norm": 1.2421875,
"learning_rate": 0.0004984028887746443,
"loss": 5.9739,
"mean_token_accuracy": 0.16280067563056946,
"num_tokens": 9818324.0,
"step": 5260
},
{
"entropy": 5.984898710250855,
"epoch": 0.4649417167078771,
"grad_norm": 1.4296875,
"learning_rate": 0.0004983991410615239,
"loss": 5.9645,
"mean_token_accuracy": 0.16320008635520936,
"num_tokens": 9827935.0,
"step": 5265
},
{
"entropy": 6.089527177810669,
"epoch": 0.4653832567997174,
"grad_norm": 1.3359375,
"learning_rate": 0.0004983953889721414,
"loss": 5.9782,
"mean_token_accuracy": 0.15890799909830094,
"num_tokens": 9837118.0,
"step": 5270
},
{
"entropy": 6.149358510971069,
"epoch": 0.46582479689155776,
"grad_norm": 1.4375,
"learning_rate": 0.0004983916325065703,
"loss": 6.0455,
"mean_token_accuracy": 0.1533641681075096,
"num_tokens": 9846197.0,
"step": 5275
},
{
"entropy": 6.196761894226074,
"epoch": 0.4662663369833981,
"grad_norm": 1.2421875,
"learning_rate": 0.0004983878716648842,
"loss": 6.107,
"mean_token_accuracy": 0.14858163744211197,
"num_tokens": 9856120.0,
"step": 5280
},
{
"entropy": 6.119464254379272,
"epoch": 0.4667078770752384,
"grad_norm": 1.15625,
"learning_rate": 0.0004983841064471567,
"loss": 6.056,
"mean_token_accuracy": 0.1537718027830124,
"num_tokens": 9865599.0,
"step": 5285
},
{
"entropy": 6.096899557113647,
"epoch": 0.46714941716707875,
"grad_norm": 1.203125,
"learning_rate": 0.0004983803368534617,
"loss": 5.9688,
"mean_token_accuracy": 0.15628019720315933,
"num_tokens": 9876471.0,
"step": 5290
},
{
"entropy": 6.196185064315796,
"epoch": 0.4675909572589191,
"grad_norm": 1.4453125,
"learning_rate": 0.0004983765628838728,
"loss": 6.0124,
"mean_token_accuracy": 0.15632506608963012,
"num_tokens": 9887680.0,
"step": 5295
},
{
"entropy": 6.205327701568604,
"epoch": 0.46803249735075947,
"grad_norm": 1.59375,
"learning_rate": 0.0004983727845384641,
"loss": 6.0798,
"mean_token_accuracy": 0.15046066045761108,
"num_tokens": 9897366.0,
"step": 5300
},
{
"entropy": 6.112229824066162,
"epoch": 0.4684740374425998,
"grad_norm": 1.328125,
"learning_rate": 0.0004983690018173096,
"loss": 6.0473,
"mean_token_accuracy": 0.15282203108072281,
"num_tokens": 9907437.0,
"step": 5305
},
{
"entropy": 6.03530478477478,
"epoch": 0.46891557753444013,
"grad_norm": 1.3125,
"learning_rate": 0.0004983652147204834,
"loss": 5.8493,
"mean_token_accuracy": 0.16888897120952606,
"num_tokens": 9915766.0,
"step": 5310
},
{
"entropy": 6.039136505126953,
"epoch": 0.46935711762628046,
"grad_norm": 1.328125,
"learning_rate": 0.0004983614232480598,
"loss": 6.0557,
"mean_token_accuracy": 0.1539039731025696,
"num_tokens": 9925389.0,
"step": 5315
},
{
"entropy": 6.1289163589477536,
"epoch": 0.4697986577181208,
"grad_norm": 1.234375,
"learning_rate": 0.0004983576274001127,
"loss": 6.0973,
"mean_token_accuracy": 0.15069840773940085,
"num_tokens": 9935798.0,
"step": 5320
},
{
"entropy": 6.164807653427124,
"epoch": 0.4702401978099611,
"grad_norm": 1.2578125,
"learning_rate": 0.000498353827176717,
"loss": 5.9409,
"mean_token_accuracy": 0.16221913695335388,
"num_tokens": 9945579.0,
"step": 5325
},
{
"entropy": 6.1347403049469,
"epoch": 0.4706817379018015,
"grad_norm": 1.328125,
"learning_rate": 0.0004983500225779466,
"loss": 5.9683,
"mean_token_accuracy": 0.15579652935266494,
"num_tokens": 9955248.0,
"step": 5330
},
{
"entropy": 6.085616779327393,
"epoch": 0.47112327799364184,
"grad_norm": 1.2890625,
"learning_rate": 0.0004983462136038764,
"loss": 5.9757,
"mean_token_accuracy": 0.16684675961732864,
"num_tokens": 9965078.0,
"step": 5335
},
{
"entropy": 6.127761745452881,
"epoch": 0.4715648180854822,
"grad_norm": 1.6953125,
"learning_rate": 0.0004983424002545809,
"loss": 5.9417,
"mean_token_accuracy": 0.16493815779685975,
"num_tokens": 9973932.0,
"step": 5340
},
{
"entropy": 6.168215322494507,
"epoch": 0.4720063581773225,
"grad_norm": 1.328125,
"learning_rate": 0.0004983385825301348,
"loss": 6.0002,
"mean_token_accuracy": 0.1548996612429619,
"num_tokens": 9983315.0,
"step": 5345
},
{
"entropy": 6.058854579925537,
"epoch": 0.47244789826916284,
"grad_norm": 2.0625,
"learning_rate": 0.0004983347604306129,
"loss": 5.9641,
"mean_token_accuracy": 0.15613725483417512,
"num_tokens": 9992987.0,
"step": 5350
},
{
"entropy": 6.019079732894897,
"epoch": 0.47288943836100317,
"grad_norm": 1.453125,
"learning_rate": 0.0004983309339560899,
"loss": 6.0237,
"mean_token_accuracy": 0.15275818705558777,
"num_tokens": 10002268.0,
"step": 5355
},
{
"entropy": 6.169502782821655,
"epoch": 0.4733309784528435,
"grad_norm": 1.3359375,
"learning_rate": 0.0004983271031066412,
"loss": 6.0223,
"mean_token_accuracy": 0.14990446120500564,
"num_tokens": 10011772.0,
"step": 5360
},
{
"entropy": 6.224719524383545,
"epoch": 0.4737725185446839,
"grad_norm": 1.5234375,
"learning_rate": 0.0004983232678823414,
"loss": 6.0393,
"mean_token_accuracy": 0.15731042325496675,
"num_tokens": 10021069.0,
"step": 5365
},
{
"entropy": 6.063888311386108,
"epoch": 0.4742140586365242,
"grad_norm": 1.3125,
"learning_rate": 0.0004983194282832657,
"loss": 5.9557,
"mean_token_accuracy": 0.161125111579895,
"num_tokens": 10029706.0,
"step": 5370
},
{
"entropy": 5.960548067092896,
"epoch": 0.47465559872836455,
"grad_norm": 1.421875,
"learning_rate": 0.0004983155843094895,
"loss": 5.8997,
"mean_token_accuracy": 0.16574549674987793,
"num_tokens": 10039633.0,
"step": 5375
},
{
"entropy": 6.088515663146973,
"epoch": 0.4750971388202049,
"grad_norm": 1.8125,
"learning_rate": 0.0004983117359610881,
"loss": 5.9683,
"mean_token_accuracy": 0.1593565970659256,
"num_tokens": 10048675.0,
"step": 5380
},
{
"entropy": 6.142480039596558,
"epoch": 0.4755386789120452,
"grad_norm": 1.1640625,
"learning_rate": 0.0004983078832381367,
"loss": 5.9571,
"mean_token_accuracy": 0.16286074072122575,
"num_tokens": 10057447.0,
"step": 5385
},
{
"entropy": 6.040254259109497,
"epoch": 0.47598021900388554,
"grad_norm": 1.75,
"learning_rate": 0.0004983040261407109,
"loss": 6.0401,
"mean_token_accuracy": 0.15268651247024537,
"num_tokens": 10067020.0,
"step": 5390
},
{
"entropy": 6.183468341827393,
"epoch": 0.47642175909572587,
"grad_norm": 1.5078125,
"learning_rate": 0.0004983001646688863,
"loss": 6.0737,
"mean_token_accuracy": 0.15229557305574418,
"num_tokens": 10076466.0,
"step": 5395
},
{
"entropy": 6.156057071685791,
"epoch": 0.47686329918756626,
"grad_norm": 1.40625,
"learning_rate": 0.0004982962988227383,
"loss": 5.9383,
"mean_token_accuracy": 0.1562927931547165,
"num_tokens": 10085384.0,
"step": 5400
},
{
"entropy": 6.085453033447266,
"epoch": 0.4773048392794066,
"grad_norm": 1.3828125,
"learning_rate": 0.000498292428602343,
"loss": 6.0543,
"mean_token_accuracy": 0.15518272593617438,
"num_tokens": 10094015.0,
"step": 5405
},
{
"entropy": 6.131287050247193,
"epoch": 0.4777463793712469,
"grad_norm": 1.40625,
"learning_rate": 0.0004982885540077758,
"loss": 5.9489,
"mean_token_accuracy": 0.16021449863910675,
"num_tokens": 10103107.0,
"step": 5410
},
{
"entropy": 6.061293983459473,
"epoch": 0.47818791946308725,
"grad_norm": 1.21875,
"learning_rate": 0.0004982846750391129,
"loss": 5.883,
"mean_token_accuracy": 0.165081886947155,
"num_tokens": 10112408.0,
"step": 5415
},
{
"entropy": 6.103114128112793,
"epoch": 0.4786294595549276,
"grad_norm": 1.3125,
"learning_rate": 0.0004982807916964303,
"loss": 5.9587,
"mean_token_accuracy": 0.15787963271141053,
"num_tokens": 10121605.0,
"step": 5420
},
{
"entropy": 6.012549638748169,
"epoch": 0.4790709996467679,
"grad_norm": 1.4609375,
"learning_rate": 0.000498276903979804,
"loss": 5.9534,
"mean_token_accuracy": 0.1598804622888565,
"num_tokens": 10130484.0,
"step": 5425
},
{
"entropy": 5.984193134307861,
"epoch": 0.47951253973860825,
"grad_norm": 1.2890625,
"learning_rate": 0.00049827301188931,
"loss": 5.906,
"mean_token_accuracy": 0.16575224399566652,
"num_tokens": 10140002.0,
"step": 5430
},
{
"entropy": 6.203652048110962,
"epoch": 0.47995407983044863,
"grad_norm": 1.4140625,
"learning_rate": 0.0004982691154250247,
"loss": 6.0681,
"mean_token_accuracy": 0.1568043977022171,
"num_tokens": 10150287.0,
"step": 5435
},
{
"entropy": 6.098856639862061,
"epoch": 0.48039561992228896,
"grad_norm": 1.328125,
"learning_rate": 0.0004982652145870245,
"loss": 5.9027,
"mean_token_accuracy": 0.16656555682420732,
"num_tokens": 10160615.0,
"step": 5440
},
{
"entropy": 5.997969436645508,
"epoch": 0.4808371600141293,
"grad_norm": 1.2890625,
"learning_rate": 0.0004982613093753856,
"loss": 5.9701,
"mean_token_accuracy": 0.16098668649792672,
"num_tokens": 10170271.0,
"step": 5445
},
{
"entropy": 6.051625394821167,
"epoch": 0.4812787001059696,
"grad_norm": 1.34375,
"learning_rate": 0.0004982573997901847,
"loss": 5.9498,
"mean_token_accuracy": 0.16217143833637238,
"num_tokens": 10179663.0,
"step": 5450
},
{
"entropy": 6.167967081069946,
"epoch": 0.48172024019780996,
"grad_norm": 1.265625,
"learning_rate": 0.0004982534858314982,
"loss": 6.0302,
"mean_token_accuracy": 0.1533224031329155,
"num_tokens": 10188608.0,
"step": 5455
},
{
"entropy": 6.123821449279785,
"epoch": 0.4821617802896503,
"grad_norm": 1.34375,
"learning_rate": 0.0004982495674994031,
"loss": 5.9888,
"mean_token_accuracy": 0.1592209592461586,
"num_tokens": 10198462.0,
"step": 5460
},
{
"entropy": 5.982554721832275,
"epoch": 0.4826033203814906,
"grad_norm": 1.3359375,
"learning_rate": 0.0004982456447939758,
"loss": 5.9323,
"mean_token_accuracy": 0.16437650620937347,
"num_tokens": 10208362.0,
"step": 5465
},
{
"entropy": 6.075630807876587,
"epoch": 0.483044860473331,
"grad_norm": 1.328125,
"learning_rate": 0.0004982417177152933,
"loss": 5.955,
"mean_token_accuracy": 0.1612927421927452,
"num_tokens": 10217915.0,
"step": 5470
},
{
"entropy": 6.136414337158203,
"epoch": 0.48348640056517134,
"grad_norm": 1.484375,
"learning_rate": 0.0004982377862634325,
"loss": 6.0141,
"mean_token_accuracy": 0.16026580333709717,
"num_tokens": 10226913.0,
"step": 5475
},
{
"entropy": 6.0771276473999025,
"epoch": 0.48392794065701167,
"grad_norm": 1.28125,
"learning_rate": 0.0004982338504384705,
"loss": 5.9819,
"mean_token_accuracy": 0.15748471468687059,
"num_tokens": 10236516.0,
"step": 5480
},
{
"entropy": 6.1290271282196045,
"epoch": 0.484369480748852,
"grad_norm": 1.296875,
"learning_rate": 0.0004982299102404843,
"loss": 5.9977,
"mean_token_accuracy": 0.15856396406888962,
"num_tokens": 10245492.0,
"step": 5485
},
{
"entropy": 5.977679443359375,
"epoch": 0.48481102084069233,
"grad_norm": 1.734375,
"learning_rate": 0.000498225965669551,
"loss": 5.8679,
"mean_token_accuracy": 0.17316461503505706,
"num_tokens": 10254094.0,
"step": 5490
},
{
"entropy": 6.03317437171936,
"epoch": 0.48525256093253266,
"grad_norm": 1.4296875,
"learning_rate": 0.0004982220167257482,
"loss": 5.9069,
"mean_token_accuracy": 0.1637963816523552,
"num_tokens": 10263645.0,
"step": 5495
},
{
"entropy": 6.013968801498413,
"epoch": 0.485694101024373,
"grad_norm": 1.2578125,
"learning_rate": 0.0004982180634091529,
"loss": 5.8919,
"mean_token_accuracy": 0.16816270351409912,
"num_tokens": 10273416.0,
"step": 5500
},
{
"entropy": 6.0709892272949215,
"epoch": 0.4861356411162134,
"grad_norm": 1.25,
"learning_rate": 0.0004982141057198427,
"loss": 5.9987,
"mean_token_accuracy": 0.15827764123678206,
"num_tokens": 10283194.0,
"step": 5505
},
{
"entropy": 6.133906984329224,
"epoch": 0.4865771812080537,
"grad_norm": 1.4375,
"learning_rate": 0.0004982101436578952,
"loss": 5.9737,
"mean_token_accuracy": 0.15922945886850357,
"num_tokens": 10292188.0,
"step": 5510
},
{
"entropy": 6.0081017971038815,
"epoch": 0.48701872129989404,
"grad_norm": 1.65625,
"learning_rate": 0.0004982061772233878,
"loss": 5.9009,
"mean_token_accuracy": 0.17360175549983978,
"num_tokens": 10301539.0,
"step": 5515
},
{
"entropy": 6.000823640823365,
"epoch": 0.4874602613917344,
"grad_norm": 1.3203125,
"learning_rate": 0.0004982022064163984,
"loss": 5.9093,
"mean_token_accuracy": 0.16490670889616013,
"num_tokens": 10310379.0,
"step": 5520
},
{
"entropy": 6.19787015914917,
"epoch": 0.4879018014835747,
"grad_norm": 1.296875,
"learning_rate": 0.0004981982312370047,
"loss": 6.1108,
"mean_token_accuracy": 0.1488179437816143,
"num_tokens": 10320622.0,
"step": 5525
},
{
"entropy": 6.114509153366089,
"epoch": 0.48834334157541504,
"grad_norm": 1.171875,
"learning_rate": 0.0004981942516852847,
"loss": 5.9356,
"mean_token_accuracy": 0.15839738100767137,
"num_tokens": 10330525.0,
"step": 5530
},
{
"entropy": 6.041282749176025,
"epoch": 0.48878488166725537,
"grad_norm": 1.203125,
"learning_rate": 0.0004981902677613161,
"loss": 5.9671,
"mean_token_accuracy": 0.1586508110165596,
"num_tokens": 10340288.0,
"step": 5535
},
{
"entropy": 6.1574572086334225,
"epoch": 0.48922642175909575,
"grad_norm": 1.21875,
"learning_rate": 0.0004981862794651771,
"loss": 6.0482,
"mean_token_accuracy": 0.15438254177570343,
"num_tokens": 10349836.0,
"step": 5540
},
{
"entropy": 6.124531698226929,
"epoch": 0.4896679618509361,
"grad_norm": 1.5078125,
"learning_rate": 0.0004981822867969459,
"loss": 5.9835,
"mean_token_accuracy": 0.1561306193470955,
"num_tokens": 10359481.0,
"step": 5545
},
{
"entropy": 6.077211523056031,
"epoch": 0.4901095019427764,
"grad_norm": 1.3203125,
"learning_rate": 0.0004981782897567006,
"loss": 5.9606,
"mean_token_accuracy": 0.16075108498334884,
"num_tokens": 10369837.0,
"step": 5550
},
{
"entropy": 6.108371019363403,
"epoch": 0.49055104203461675,
"grad_norm": 1.671875,
"learning_rate": 0.0004981742883445195,
"loss": 5.927,
"mean_token_accuracy": 0.16440120637416838,
"num_tokens": 10379276.0,
"step": 5555
},
{
"entropy": 6.127094459533692,
"epoch": 0.4909925821264571,
"grad_norm": 1.4453125,
"learning_rate": 0.000498170282560481,
"loss": 6.0562,
"mean_token_accuracy": 0.15287164598703384,
"num_tokens": 10388878.0,
"step": 5560
},
{
"entropy": 6.158031272888183,
"epoch": 0.4914341222182974,
"grad_norm": 1.3359375,
"learning_rate": 0.0004981662724046637,
"loss": 6.0416,
"mean_token_accuracy": 0.15418365895748137,
"num_tokens": 10399148.0,
"step": 5565
},
{
"entropy": 6.072212409973145,
"epoch": 0.49187566231013774,
"grad_norm": 1.2265625,
"learning_rate": 0.000498162257877146,
"loss": 5.9705,
"mean_token_accuracy": 0.16276238560676576,
"num_tokens": 10409002.0,
"step": 5570
},
{
"entropy": 6.052406740188599,
"epoch": 0.4923172024019781,
"grad_norm": 1.4296875,
"learning_rate": 0.0004981582389780065,
"loss": 6.0182,
"mean_token_accuracy": 0.15543360412120819,
"num_tokens": 10418083.0,
"step": 5575
},
{
"entropy": 6.2516388416290285,
"epoch": 0.49275874249381846,
"grad_norm": 1.3671875,
"learning_rate": 0.0004981542157073241,
"loss": 6.0216,
"mean_token_accuracy": 0.15745319724082946,
"num_tokens": 10428134.0,
"step": 5580
},
{
"entropy": 6.102121114730835,
"epoch": 0.4932002825856588,
"grad_norm": 1.3984375,
"learning_rate": 0.0004981501880651775,
"loss": 6.0144,
"mean_token_accuracy": 0.15700092390179635,
"num_tokens": 10437436.0,
"step": 5585
},
{
"entropy": 6.011652183532715,
"epoch": 0.4936418226774991,
"grad_norm": 1.2734375,
"learning_rate": 0.0004981461560516457,
"loss": 5.9437,
"mean_token_accuracy": 0.15731440335512162,
"num_tokens": 10446176.0,
"step": 5590
},
{
"entropy": 6.185711860656738,
"epoch": 0.49408336276933945,
"grad_norm": 1.3125,
"learning_rate": 0.0004981421196668075,
"loss": 6.1086,
"mean_token_accuracy": 0.15340851247310638,
"num_tokens": 10455953.0,
"step": 5595
},
{
"entropy": 6.20690336227417,
"epoch": 0.4945249028611798,
"grad_norm": 1.546875,
"learning_rate": 0.0004981380789107422,
"loss": 5.9589,
"mean_token_accuracy": 0.15672969669103623,
"num_tokens": 10465129.0,
"step": 5600
},
{
"entropy": 6.003274393081665,
"epoch": 0.4949664429530201,
"grad_norm": 1.40625,
"learning_rate": 0.0004981340337835287,
"loss": 5.9608,
"mean_token_accuracy": 0.1587096706032753,
"num_tokens": 10474376.0,
"step": 5605
},
{
"entropy": 6.04667649269104,
"epoch": 0.4954079830448605,
"grad_norm": 1.78125,
"learning_rate": 0.0004981299842852464,
"loss": 6.02,
"mean_token_accuracy": 0.1601344585418701,
"num_tokens": 10484849.0,
"step": 5610
},
{
"entropy": 6.119048452377319,
"epoch": 0.49584952313670083,
"grad_norm": 1.3671875,
"learning_rate": 0.0004981259304159747,
"loss": 5.9104,
"mean_token_accuracy": 0.17004732936620712,
"num_tokens": 10493950.0,
"step": 5615
},
{
"entropy": 6.0551595211029055,
"epoch": 0.49629106322854116,
"grad_norm": 1.3671875,
"learning_rate": 0.0004981218721757929,
"loss": 5.877,
"mean_token_accuracy": 0.16873401552438735,
"num_tokens": 10502890.0,
"step": 5620
},
{
"entropy": 5.9873803615570065,
"epoch": 0.4967326033203815,
"grad_norm": 1.421875,
"learning_rate": 0.0004981178095647805,
"loss": 5.8707,
"mean_token_accuracy": 0.160433566570282,
"num_tokens": 10511702.0,
"step": 5625
},
{
"entropy": 6.003235149383545,
"epoch": 0.4971741434122218,
"grad_norm": 1.328125,
"learning_rate": 0.0004981137425830171,
"loss": 5.9481,
"mean_token_accuracy": 0.16074998080730438,
"num_tokens": 10520212.0,
"step": 5630
},
{
"entropy": 6.066579341888428,
"epoch": 0.49761568350406216,
"grad_norm": 1.359375,
"learning_rate": 0.0004981096712305825,
"loss": 5.971,
"mean_token_accuracy": 0.1613484501838684,
"num_tokens": 10529652.0,
"step": 5635
},
{
"entropy": 6.099881601333618,
"epoch": 0.4980572235959025,
"grad_norm": 1.296875,
"learning_rate": 0.0004981055955075561,
"loss": 6.0225,
"mean_token_accuracy": 0.158574703335762,
"num_tokens": 10537980.0,
"step": 5640
},
{
"entropy": 6.089739847183227,
"epoch": 0.4984987636877429,
"grad_norm": 1.3125,
"learning_rate": 0.0004981015154140181,
"loss": 5.8995,
"mean_token_accuracy": 0.16764541566371918,
"num_tokens": 10546854.0,
"step": 5645
},
{
"entropy": 5.998034429550171,
"epoch": 0.4989403037795832,
"grad_norm": 1.359375,
"learning_rate": 0.0004980974309500483,
"loss": 5.923,
"mean_token_accuracy": 0.16142944097518921,
"num_tokens": 10555691.0,
"step": 5650
},
{
"entropy": 6.080042123794556,
"epoch": 0.49938184387142354,
"grad_norm": 1.90625,
"learning_rate": 0.0004980933421157267,
"loss": 6.0191,
"mean_token_accuracy": 0.1503999724984169,
"num_tokens": 10564750.0,
"step": 5655
},
{
"entropy": 6.1378124237060545,
"epoch": 0.49982338396326387,
"grad_norm": 1.3203125,
"learning_rate": 0.0004980892489111334,
"loss": 6.0389,
"mean_token_accuracy": 0.15747978240251542,
"num_tokens": 10573515.0,
"step": 5660
},
{
"entropy": 6.055855131149292,
"epoch": 0.5002649240551043,
"grad_norm": 1.3515625,
"learning_rate": 0.0004980851513363486,
"loss": 5.9161,
"mean_token_accuracy": 0.1634524017572403,
"num_tokens": 10582730.0,
"step": 5665
},
{
"entropy": 5.963370323181152,
"epoch": 0.5007064641469445,
"grad_norm": 5.15625,
"learning_rate": 0.0004980810493914526,
"loss": 5.8141,
"mean_token_accuracy": 0.18174145370721817,
"num_tokens": 10591267.0,
"step": 5670
},
{
"entropy": 6.040016841888428,
"epoch": 0.5011480042387849,
"grad_norm": 1.3046875,
"learning_rate": 0.0004980769430765256,
"loss": 5.918,
"mean_token_accuracy": 0.15779468268156052,
"num_tokens": 10599776.0,
"step": 5675
},
{
"entropy": 6.018471956253052,
"epoch": 0.5015895443306252,
"grad_norm": 2.03125,
"learning_rate": 0.0004980728323916484,
"loss": 5.9294,
"mean_token_accuracy": 0.15198549777269363,
"num_tokens": 10609670.0,
"step": 5680
},
{
"entropy": 6.0253712177276615,
"epoch": 0.5020310844224656,
"grad_norm": 1.21875,
"learning_rate": 0.0004980687173369009,
"loss": 5.8857,
"mean_token_accuracy": 0.16054306030273438,
"num_tokens": 10619032.0,
"step": 5685
},
{
"entropy": 6.13008975982666,
"epoch": 0.5024726245143059,
"grad_norm": 1.5390625,
"learning_rate": 0.0004980645979123644,
"loss": 5.9696,
"mean_token_accuracy": 0.15761574804782869,
"num_tokens": 10628851.0,
"step": 5690
},
{
"entropy": 6.1254706382751465,
"epoch": 0.5029141646061462,
"grad_norm": 1.3671875,
"learning_rate": 0.0004980604741181192,
"loss": 5.9384,
"mean_token_accuracy": 0.1661163553595543,
"num_tokens": 10638335.0,
"step": 5695
},
{
"entropy": 6.104246044158936,
"epoch": 0.5033557046979866,
"grad_norm": 1.265625,
"learning_rate": 0.0004980563459542461,
"loss": 6.0288,
"mean_token_accuracy": 0.15877759456634521,
"num_tokens": 10648450.0,
"step": 5700
},
{
"entropy": 6.129976463317871,
"epoch": 0.5037972447898269,
"grad_norm": 1.203125,
"learning_rate": 0.0004980522134208261,
"loss": 6.0499,
"mean_token_accuracy": 0.15289226770401002,
"num_tokens": 10658700.0,
"step": 5705
},
{
"entropy": 6.125590181350708,
"epoch": 0.5042387848816673,
"grad_norm": 1.34375,
"learning_rate": 0.0004980480765179401,
"loss": 6.0498,
"mean_token_accuracy": 0.15848255753517151,
"num_tokens": 10667952.0,
"step": 5710
},
{
"entropy": 6.015667390823364,
"epoch": 0.5046803249735076,
"grad_norm": 1.46875,
"learning_rate": 0.0004980439352456692,
"loss": 5.8856,
"mean_token_accuracy": 0.16176477670669556,
"num_tokens": 10677228.0,
"step": 5715
},
{
"entropy": 6.1053516387939455,
"epoch": 0.505121865065348,
"grad_norm": 1.203125,
"learning_rate": 0.0004980397896040944,
"loss": 5.974,
"mean_token_accuracy": 0.16118671298027037,
"num_tokens": 10686183.0,
"step": 5720
},
{
"entropy": 6.190397262573242,
"epoch": 0.5055634051571882,
"grad_norm": 1.265625,
"learning_rate": 0.0004980356395932969,
"loss": 5.9997,
"mean_token_accuracy": 0.15813823491334916,
"num_tokens": 10695293.0,
"step": 5725
},
{
"entropy": 6.0877281665802006,
"epoch": 0.5060049452490286,
"grad_norm": 1.21875,
"learning_rate": 0.0004980314852133581,
"loss": 5.9647,
"mean_token_accuracy": 0.1628525137901306,
"num_tokens": 10704853.0,
"step": 5730
},
{
"entropy": 6.098628950119019,
"epoch": 0.506446485340869,
"grad_norm": 1.2578125,
"learning_rate": 0.0004980273264643594,
"loss": 5.9596,
"mean_token_accuracy": 0.15651024580001832,
"num_tokens": 10714307.0,
"step": 5735
},
{
"entropy": 5.91804838180542,
"epoch": 0.5068880254327093,
"grad_norm": 1.2109375,
"learning_rate": 0.0004980231633463822,
"loss": 5.8195,
"mean_token_accuracy": 0.16056904792785645,
"num_tokens": 10723513.0,
"step": 5740
},
{
"entropy": 6.041822719573974,
"epoch": 0.5073295655245497,
"grad_norm": 1.453125,
"learning_rate": 0.0004980189958595081,
"loss": 6.009,
"mean_token_accuracy": 0.15504895150661469,
"num_tokens": 10732809.0,
"step": 5745
},
{
"entropy": 6.180126619338989,
"epoch": 0.5077711056163899,
"grad_norm": 1.1796875,
"learning_rate": 0.0004980148240038186,
"loss": 6.0285,
"mean_token_accuracy": 0.16252532303333284,
"num_tokens": 10742960.0,
"step": 5750
},
{
"entropy": 6.167002391815186,
"epoch": 0.5082126457082303,
"grad_norm": 1.5625,
"learning_rate": 0.0004980106477793957,
"loss": 6.0962,
"mean_token_accuracy": 0.15922853201627732,
"num_tokens": 10752835.0,
"step": 5755
},
{
"entropy": 6.090708923339844,
"epoch": 0.5086541858000706,
"grad_norm": 1.40625,
"learning_rate": 0.0004980064671863209,
"loss": 5.9568,
"mean_token_accuracy": 0.16255878955125808,
"num_tokens": 10762139.0,
"step": 5760
},
{
"entropy": 6.121759986877441,
"epoch": 0.509095725891911,
"grad_norm": 1.4296875,
"learning_rate": 0.0004980022822246763,
"loss": 6.0284,
"mean_token_accuracy": 0.1527095004916191,
"num_tokens": 10771961.0,
"step": 5765
},
{
"entropy": 6.189744853973389,
"epoch": 0.5095372659837514,
"grad_norm": 1.2421875,
"learning_rate": 0.0004979980928945439,
"loss": 6.0799,
"mean_token_accuracy": 0.14973168522119523,
"num_tokens": 10780163.0,
"step": 5770
},
{
"entropy": 6.200748443603516,
"epoch": 0.5099788060755917,
"grad_norm": 1.4921875,
"learning_rate": 0.0004979938991960056,
"loss": 6.0327,
"mean_token_accuracy": 0.156923408806324,
"num_tokens": 10788339.0,
"step": 5775
},
{
"entropy": 6.119477415084839,
"epoch": 0.510420346167432,
"grad_norm": 1.4296875,
"learning_rate": 0.0004979897011291436,
"loss": 6.0257,
"mean_token_accuracy": 0.15160492956638336,
"num_tokens": 10797315.0,
"step": 5780
},
{
"entropy": 6.097667026519775,
"epoch": 0.5108618862592723,
"grad_norm": 1.3515625,
"learning_rate": 0.0004979854986940402,
"loss": 5.9344,
"mean_token_accuracy": 0.15801928341388702,
"num_tokens": 10806864.0,
"step": 5785
},
{
"entropy": 6.078370189666748,
"epoch": 0.5113034263511127,
"grad_norm": 1.4921875,
"learning_rate": 0.0004979812918907777,
"loss": 5.971,
"mean_token_accuracy": 0.15703734010457993,
"num_tokens": 10816417.0,
"step": 5790
},
{
"entropy": 6.111645221710205,
"epoch": 0.511744966442953,
"grad_norm": 1.3984375,
"learning_rate": 0.0004979770807194385,
"loss": 6.0491,
"mean_token_accuracy": 0.15200948417186738,
"num_tokens": 10825692.0,
"step": 5795
},
{
"entropy": 6.125649499893188,
"epoch": 0.5121865065347934,
"grad_norm": 1.171875,
"learning_rate": 0.0004979728651801051,
"loss": 6.0184,
"mean_token_accuracy": 0.1593027725815773,
"num_tokens": 10834652.0,
"step": 5800
},
{
"entropy": 6.142418766021729,
"epoch": 0.5126280466266337,
"grad_norm": 1.296875,
"learning_rate": 0.0004979686452728602,
"loss": 6.0045,
"mean_token_accuracy": 0.15610153079032899,
"num_tokens": 10844491.0,
"step": 5805
},
{
"entropy": 6.076530647277832,
"epoch": 0.513069586718474,
"grad_norm": 1.3828125,
"learning_rate": 0.0004979644209977863,
"loss": 5.9576,
"mean_token_accuracy": 0.16532048285007478,
"num_tokens": 10853948.0,
"step": 5810
},
{
"entropy": 6.112010383605957,
"epoch": 0.5135111268103144,
"grad_norm": 1.515625,
"learning_rate": 0.0004979601923549661,
"loss": 5.9547,
"mean_token_accuracy": 0.16036146879196167,
"num_tokens": 10861745.0,
"step": 5815
},
{
"entropy": 6.1195940494537355,
"epoch": 0.5139526669021547,
"grad_norm": 1.2734375,
"learning_rate": 0.0004979559593444826,
"loss": 6.0383,
"mean_token_accuracy": 0.1518269196152687,
"num_tokens": 10871239.0,
"step": 5820
},
{
"entropy": 6.060505533218384,
"epoch": 0.5143942069939951,
"grad_norm": 1.3359375,
"learning_rate": 0.0004979517219664187,
"loss": 5.8893,
"mean_token_accuracy": 0.16240942627191543,
"num_tokens": 10880883.0,
"step": 5825
},
{
"entropy": 6.087244081497192,
"epoch": 0.5148357470858353,
"grad_norm": 1.34375,
"learning_rate": 0.0004979474802208572,
"loss": 5.9569,
"mean_token_accuracy": 0.16193697676062585,
"num_tokens": 10890002.0,
"step": 5830
},
{
"entropy": 6.015149307250977,
"epoch": 0.5152772871776757,
"grad_norm": 1.71875,
"learning_rate": 0.0004979432341078816,
"loss": 5.8864,
"mean_token_accuracy": 0.15554073452949524,
"num_tokens": 10898299.0,
"step": 5835
},
{
"entropy": 6.089050912857056,
"epoch": 0.5157188272695161,
"grad_norm": 1.3359375,
"learning_rate": 0.0004979389836275746,
"loss": 6.0774,
"mean_token_accuracy": 0.14996080696582795,
"num_tokens": 10907918.0,
"step": 5840
},
{
"entropy": 6.133493900299072,
"epoch": 0.5161603673613564,
"grad_norm": 1.296875,
"learning_rate": 0.0004979347287800198,
"loss": 5.9596,
"mean_token_accuracy": 0.15501177459955215,
"num_tokens": 10917318.0,
"step": 5845
},
{
"entropy": 6.060728597640991,
"epoch": 0.5166019074531968,
"grad_norm": 1.484375,
"learning_rate": 0.0004979304695653005,
"loss": 5.9144,
"mean_token_accuracy": 0.16151465103030205,
"num_tokens": 10926338.0,
"step": 5850
},
{
"entropy": 5.939118814468384,
"epoch": 0.5170434475450371,
"grad_norm": 1.3828125,
"learning_rate": 0.0004979262059835001,
"loss": 5.9432,
"mean_token_accuracy": 0.16033163890242577,
"num_tokens": 10935244.0,
"step": 5855
},
{
"entropy": 6.143280649185181,
"epoch": 0.5174849876368774,
"grad_norm": 1.3828125,
"learning_rate": 0.0004979219380347021,
"loss": 5.9905,
"mean_token_accuracy": 0.15961788594722748,
"num_tokens": 10943764.0,
"step": 5860
},
{
"entropy": 6.160701942443848,
"epoch": 0.5179265277287177,
"grad_norm": 1.3203125,
"learning_rate": 0.0004979176657189901,
"loss": 5.9547,
"mean_token_accuracy": 0.16057219803333284,
"num_tokens": 10952580.0,
"step": 5865
},
{
"entropy": 5.9959807872772215,
"epoch": 0.5183680678205581,
"grad_norm": 1.171875,
"learning_rate": 0.0004979133890364477,
"loss": 5.9825,
"mean_token_accuracy": 0.1650010645389557,
"num_tokens": 10961745.0,
"step": 5870
},
{
"entropy": 5.959691524505615,
"epoch": 0.5188096079123985,
"grad_norm": 1.296875,
"learning_rate": 0.000497909107987159,
"loss": 5.9087,
"mean_token_accuracy": 0.15640757530927657,
"num_tokens": 10971018.0,
"step": 5875
},
{
"entropy": 6.110277318954468,
"epoch": 0.5192511480042388,
"grad_norm": 1.2734375,
"learning_rate": 0.0004979048225712074,
"loss": 5.9408,
"mean_token_accuracy": 0.1647150531411171,
"num_tokens": 10980070.0,
"step": 5880
},
{
"entropy": 6.112304973602295,
"epoch": 0.5196926880960792,
"grad_norm": 1.265625,
"learning_rate": 0.0004979005327886772,
"loss": 5.9115,
"mean_token_accuracy": 0.15777890086174012,
"num_tokens": 10989090.0,
"step": 5885
},
{
"entropy": 6.028247594833374,
"epoch": 0.5201342281879194,
"grad_norm": 1.3828125,
"learning_rate": 0.0004978962386396525,
"loss": 5.9676,
"mean_token_accuracy": 0.16010279804468155,
"num_tokens": 10998652.0,
"step": 5890
},
{
"entropy": 5.936372900009156,
"epoch": 0.5205757682797598,
"grad_norm": 1.2578125,
"learning_rate": 0.0004978919401242171,
"loss": 5.8702,
"mean_token_accuracy": 0.17004906684160231,
"num_tokens": 11006323.0,
"step": 5895
},
{
"entropy": 5.953699111938477,
"epoch": 0.5210173083716001,
"grad_norm": 1.390625,
"learning_rate": 0.0004978876372424553,
"loss": 5.8013,
"mean_token_accuracy": 0.17337664365768432,
"num_tokens": 11015054.0,
"step": 5900
},
{
"entropy": 6.08972134590149,
"epoch": 0.5214588484634405,
"grad_norm": 1.2265625,
"learning_rate": 0.0004978833299944515,
"loss": 5.9875,
"mean_token_accuracy": 0.15530108660459518,
"num_tokens": 11025787.0,
"step": 5905
},
{
"entropy": 6.201382303237915,
"epoch": 0.5219003885552809,
"grad_norm": 1.2421875,
"learning_rate": 0.0004978790183802901,
"loss": 6.0957,
"mean_token_accuracy": 0.15121813490986824,
"num_tokens": 11036091.0,
"step": 5910
},
{
"entropy": 6.078470230102539,
"epoch": 0.5223419286471211,
"grad_norm": 1.328125,
"learning_rate": 0.0004978747024000554,
"loss": 6.0798,
"mean_token_accuracy": 0.15142182558774947,
"num_tokens": 11046442.0,
"step": 5915
},
{
"entropy": 6.137741947174073,
"epoch": 0.5227834687389615,
"grad_norm": 1.1875,
"learning_rate": 0.0004978703820538321,
"loss": 6.0767,
"mean_token_accuracy": 0.15347778424620628,
"num_tokens": 11056907.0,
"step": 5920
},
{
"entropy": 6.088362789154052,
"epoch": 0.5232250088308018,
"grad_norm": 1.4140625,
"learning_rate": 0.0004978660573417048,
"loss": 5.9226,
"mean_token_accuracy": 0.165810264647007,
"num_tokens": 11065298.0,
"step": 5925
},
{
"entropy": 6.02788200378418,
"epoch": 0.5236665489226422,
"grad_norm": 1.4609375,
"learning_rate": 0.000497861728263758,
"loss": 5.9277,
"mean_token_accuracy": 0.16316543370485306,
"num_tokens": 11074509.0,
"step": 5930
},
{
"entropy": 6.140137195587158,
"epoch": 0.5241080890144825,
"grad_norm": 1.4375,
"learning_rate": 0.0004978573948200769,
"loss": 5.9728,
"mean_token_accuracy": 0.16266270875930786,
"num_tokens": 11082433.0,
"step": 5935
},
{
"entropy": 6.069718503952027,
"epoch": 0.5245496291063229,
"grad_norm": 1.296875,
"learning_rate": 0.0004978530570107461,
"loss": 5.9336,
"mean_token_accuracy": 0.16215700507164002,
"num_tokens": 11090838.0,
"step": 5940
},
{
"entropy": 6.054668140411377,
"epoch": 0.5249911691981632,
"grad_norm": 1.3125,
"learning_rate": 0.0004978487148358506,
"loss": 5.9288,
"mean_token_accuracy": 0.1637731358408928,
"num_tokens": 11100048.0,
"step": 5945
},
{
"entropy": 5.982475614547729,
"epoch": 0.5254327092900035,
"grad_norm": 1.21875,
"learning_rate": 0.0004978443682954756,
"loss": 5.9389,
"mean_token_accuracy": 0.16071433573961258,
"num_tokens": 11109871.0,
"step": 5950
},
{
"entropy": 6.093014669418335,
"epoch": 0.5258742493818439,
"grad_norm": 1.171875,
"learning_rate": 0.0004978400173897061,
"loss": 5.924,
"mean_token_accuracy": 0.1567403718829155,
"num_tokens": 11119450.0,
"step": 5955
},
{
"entropy": 6.094651889801026,
"epoch": 0.5263157894736842,
"grad_norm": 1.4609375,
"learning_rate": 0.0004978356621186275,
"loss": 5.8755,
"mean_token_accuracy": 0.17035853564739228,
"num_tokens": 11128321.0,
"step": 5960
},
{
"entropy": 6.019528484344482,
"epoch": 0.5267573295655246,
"grad_norm": 1.2734375,
"learning_rate": 0.0004978313024823249,
"loss": 5.9397,
"mean_token_accuracy": 0.15585954636335372,
"num_tokens": 11138657.0,
"step": 5965
},
{
"entropy": 6.009906387329101,
"epoch": 0.5271988696573648,
"grad_norm": 1.2734375,
"learning_rate": 0.0004978269384808839,
"loss": 5.9209,
"mean_token_accuracy": 0.16706853806972505,
"num_tokens": 11147704.0,
"step": 5970
},
{
"entropy": 6.089446401596069,
"epoch": 0.5276404097492052,
"grad_norm": 1.2578125,
"learning_rate": 0.0004978225701143898,
"loss": 6.0138,
"mean_token_accuracy": 0.1562186986207962,
"num_tokens": 11156979.0,
"step": 5975
},
{
"entropy": 6.104236221313476,
"epoch": 0.5280819498410456,
"grad_norm": 1.3046875,
"learning_rate": 0.0004978181973829284,
"loss": 5.8985,
"mean_token_accuracy": 0.15810604989528657,
"num_tokens": 11165539.0,
"step": 5980
},
{
"entropy": 6.11846137046814,
"epoch": 0.5285234899328859,
"grad_norm": 1.1796875,
"learning_rate": 0.0004978138202865851,
"loss": 5.9758,
"mean_token_accuracy": 0.1503559224307537,
"num_tokens": 11175911.0,
"step": 5985
},
{
"entropy": 6.050204277038574,
"epoch": 0.5289650300247263,
"grad_norm": 1.390625,
"learning_rate": 0.0004978094388254459,
"loss": 5.9883,
"mean_token_accuracy": 0.15720559507608414,
"num_tokens": 11185365.0,
"step": 5990
},
{
"entropy": 6.095414018630981,
"epoch": 0.5294065701165666,
"grad_norm": 1.3125,
"learning_rate": 0.0004978050529995965,
"loss": 5.8979,
"mean_token_accuracy": 0.16393667906522752,
"num_tokens": 11195318.0,
"step": 5995
},
{
"entropy": 6.074192810058594,
"epoch": 0.5298481102084069,
"grad_norm": 1.234375,
"learning_rate": 0.0004978006628091228,
"loss": 6.0079,
"mean_token_accuracy": 0.15748494416475295,
"num_tokens": 11205246.0,
"step": 6000
},
{
"epoch": 0.5298481102084069,
"eval_entropy": 5.929435700441406,
"eval_loss": 5.984138011932373,
"eval_mean_token_accuracy": 0.16603976909265278,
"eval_num_tokens": 11205246.0,
"eval_runtime": 26.1397,
"eval_samples_per_second": 1351.009,
"eval_steps_per_second": 168.9,
"step": 6000
}
],
"logging_steps": 5,
"max_steps": 113230,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 3000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 1.6542104592384e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink